WIND RESOURCE MAPPING IN PAKISTAN 24 MONTH SITE RESOURCE REPORT December 2018 This report was prepared by 3E, under contract to the World Bank. It is one of several outputs from the wind Renewable Energy Resource Mapping and Spatial Planning - Pakistan [Project ID: P146140]. This activity is funded and supported by the Energy Sector Management Assistance Program (ESMAP), a multi-donor trust fund administered by the World Bank, under a global initiative on Renewable Energy Resource Mapping. Further details on the initiative can be obtained from the ESMAP website. The content of this document is the sole responsibility of the consultant authors. Any improved or validated wind resource data will be incorporated into the Global Wind Atlas. Copyright © 2018 THE WORLD BANK Washington DC 20433 Telephone: +1-202-473-1000 Internet: www.worldbank.org The World Bank does not guarantee the accuracy of the data included in this work and accept no responsibility for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because the World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for non-commercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: +1-202-522-2625; e-mail: pubrights@worldbank.org. Furthermore, the ESMAP Program Manager would appreciate receiving a copy of the publication that uses this publication for its source sent in care of the address above, or to esmap@worldbank.org. All images remain the sole property of their source and may not be used for any purpose without written permission from the source. Attribution Please cite the work as follows: World Bank. 2018. Wind Resource Mapping in Pakistan: 24 Month site resource report. Washington, DC: World Bank. World Bank/AEDB Wind Mapping Project – Pakistan 2-year Site Resource Report of Meteorological Masts 12 Sites 1. Peshawar, Khyber Pakhtunkhwa 2. Haripur, Khyber Pakhtunkhwa 3. Chakri, Punjab 4. Quaidabad, Punjab 5. Bahawalpur, Punjab 6. Sadiqabad, Punjab 7. Quetta, Balochistan 8. Sanghar, Sindh 9. Umerkot, Sindh 10. Tando Ghulam Ali, Sindh 11. Gwadar, Balochistan 12. Sujawal, Sindh Document date: December 13th 2018 Prepared by: Olgu Yildirimlar, Baris Adiloglu info@3E.eu 3E nv/sa T +32 2 217 58 68 www.3E.eu Kalkkaai 6 Quai à la Chaux F +32 2 219 79 89 B-1000 Brussels TABLE OF CONTENTS 1. Introduction .................................................................................................................................... 1 2. Site Overview................................................................................................................................... 2 3. Mast Characteristics ........................................................................................................................ 4 4. Configuration of mast ...................................................................................................................... 5 4.1 Mast instrumentation .................................................................................................................. 6 5. Site 1 – Peshawar, Khyber Pakhtunkhwa .......................................................................................... 7 5.1 Wind data processing................................................................................................................... 7 Short-term wind regime................................................................................................................... 7 Long-term wind regime.................................................................................................................... 7 5.2 Wind flow modelling .................................................................................................................... 8 Terrain model .................................................................................................................................. 8 Wind flow model ............................................................................................................................. 9 5.3 Energy production calculation .................................................................................................... 10 Gross energy production................................................................................................................ 10 Energy production losses ............................................................................................................... 10 Net energy production ................................................................................................................... 11 5.4 Uncertainty analysis ................................................................................................................... 11 5.5 Turbulence analysis.................................................................................................................... 12 6. Site 2 – Haripur, Khyber Pakhtunkhwa ........................................................................................... 13 6.1 Wind data processing................................................................................................................. 13 Short-term wind regime................................................................................................................. 13 Long-term wind regime.................................................................................................................. 13 6.2 Wind flow modelling .................................................................................................................. 14 Terrain model ................................................................................................................................ 14 Wind flow model ........................................................................................................................... 15 6.3 Energy production calculation .................................................................................................... 16 Gross energy production................................................................................................................ 16 Energy production losses ............................................................................................................... 16 Net energy production ................................................................................................................... 17 6.4 Uncertainty analysis ................................................................................................................... 17 6.5 Turbulence analysis.................................................................................................................... 18 7. Site 3 – Chakri, Punjab ................................................................................................................... 19 7.1 Wind data processing................................................................................................................. 19 Short-term wind regime................................................................................................................. 19 Long-term wind regime.................................................................................................................. 19 7.2 Wind flow modelling .................................................................................................................. 21 Terrain model ................................................................................................................................ 21 Wind flow model ........................................................................................................................... 22 7.3 Energy production calculation .................................................................................................... 23 Gross energy production................................................................................................................ 23 Energy production losses ............................................................................................................... 23 Net energy production ................................................................................................................... 24 7.4 Uncertainty analysis ................................................................................................................... 24 7.5 Turbulence analysis.................................................................................................................... 25 8. Site 4 – Quaidabad, Punjab ............................................................................................................ 26 8.1 Wind data processing................................................................................................................. 26 Short-term wind regime................................................................................................................. 26 Long-term wind regime.................................................................................................................. 26 8.2 Wind flow modelling .................................................................................................................. 28 Terrain model ................................................................................................................................ 28 Wind flow model ........................................................................................................................... 29 8.3 Energy production calculation .................................................................................................... 29 Gross energy production................................................................................................................ 30 Energy production losses ............................................................................................................... 30 Net energy production ................................................................................................................... 30 8.4 Uncertainty analysis ................................................................................................................... 31 8.5 Turbulence analysis.................................................................................................................... 31 9. Site 5 – Bahawalpur, Punjab........................................................................................................... 33 9.1 Wind data processing................................................................................................................. 33 Short-term wind regime................................................................................................................. 33 Long-term wind regime.................................................................................................................. 33 9.2 Wind flow modelling .................................................................................................................. 35 Terrain model ................................................................................................................................ 35 Wind flow model ........................................................................................................................... 36 9.3 Energy production calculation .................................................................................................... 36 Gross energy production................................................................................................................ 37 Energy production losses ............................................................................................................... 37 Net energy production ................................................................................................................... 37 9.4 Uncertainty analysis ................................................................................................................... 38 9.5 Turbulence analysis.................................................................................................................... 38 10. Site 6 – Sadiqabad, Punjab ......................................................................................................... 40 10.1 Wind data processing................................................................................................................. 40 Short-term wind regime................................................................................................................. 40 Long-term wind regime.................................................................................................................. 40 10.2 Wind flow modelling .................................................................................................................. 41 Terrain model ................................................................................................................................ 42 Wind flow model ........................................................................................................................... 42 10.3 Energy production calculation .................................................................................................... 43 Gross energy production................................................................................................................ 43 Energy production losses ............................................................................................................... 44 Net energy production ................................................................................................................... 44 10.4 Uncertainty analysis ................................................................................................................... 44 10.5 Turbulence analysis.................................................................................................................... 45 11. Site 7 – Quetta, Balochistan ....................................................................................................... 47 11.1 Wind data processing................................................................................................................. 47 Short-term wind regime................................................................................................................. 47 Long-term wind regime.................................................................................................................. 47 11.2 Wind flow modelling .................................................................................................................. 48 Terrain model ................................................................................................................................ 49 Wind flow model ........................................................................................................................... 49 11.3 Energy production calculation .................................................................................................... 50 Gross energy production................................................................................................................ 50 Energy production losses ............................................................................................................... 51 Net energy production ................................................................................................................... 51 11.4 Uncertainty analysis ................................................................................................................... 51 11.5 Turbulence analysis.................................................................................................................... 52 12. Site 8 – Sanghar, Sindh ............................................................................................................... 54 12.1 Wind data processing................................................................................................................. 54 Short-term wind regime................................................................................................................. 54 Long-term wind regime.................................................................................................................. 54 12.2 Wind flow modelling .................................................................................................................. 55 Terrain model ................................................................................................................................ 56 Wind flow model ........................................................................................................................... 57 12.3 Energy production calculation .................................................................................................... 57 Gross energy production................................................................................................................ 58 Energy production losses ............................................................................................................... 58 Net energy production ................................................................................................................... 58 12.4 Uncertainty analysis ................................................................................................................... 59 12.5 Turbulence analysis.................................................................................................................... 59 13. Site 9 – Umerkot, Sindh.............................................................................................................. 61 13.1 Wind data processing................................................................................................................. 61 Short-term wind regime................................................................................................................. 61 Long-term wind regime.................................................................................................................. 61 13.2 Wind flow modelling .................................................................................................................. 63 Terrain model ................................................................................................................................ 63 Wind flow model ........................................................................................................................... 64 13.3 Energy production calculation .................................................................................................... 65 Gross energy production................................................................................................................ 65 Energy production losses ............................................................................................................... 65 Net energy production ................................................................................................................... 66 13.4 Uncertainty analysis ................................................................................................................... 66 13.5 Turbulence analysis.................................................................................................................... 67 14. Site 10 – Tando Ghulam Ali, Sindh .............................................................................................. 68 14.1 Wind data processing................................................................................................................. 68 Short-term wind regime................................................................................................................. 68 Long-term wind regime.................................................................................................................. 68 14.2 Wind flow modelling .................................................................................................................. 69 Terrain model ................................................................................................................................ 70 Wind flow model ........................................................................................................................... 71 14.3 Energy production calculation .................................................................................................... 71 Gross energy production................................................................................................................ 71 Energy production losses ............................................................................................................... 72 Net energy production ................................................................................................................... 72 14.4 Uncertainty analysis ................................................................................................................... 73 14.5 Turbulence analysis.................................................................................................................... 73 15. Site 11 – Gwadar, Balochistan .................................................................................................... 75 15.1 Wind data processing................................................................................................................. 75 Short-term wind regime................................................................................................................. 75 Long-term wind regime.................................................................................................................. 75 15.2 Wind flow modelling .................................................................................................................. 76 Terrain model ................................................................................................................................ 77 Wind flow model ........................................................................................................................... 78 15.3 Energy production calculation .................................................................................................... 78 Gross energy production................................................................................................................ 78 Energy production losses ............................................................................................................... 79 Net energy production ................................................................................................................... 79 15.4 Uncertainty analysis ................................................................................................................... 80 15.5 Turbulence analysis.................................................................................................................... 80 16. Site 12 – Sujawal, Sindh ............................................................................................................. 82 16.1 Wind data processing................................................................................................................. 82 Short-term wind regime................................................................................................................. 82 Long-term wind regime.................................................................................................................. 82 16.2 Wind flow modelling .................................................................................................................. 83 Terrain model ................................................................................................................................ 84 Wind flow model ........................................................................................................................... 85 16.3 Energy production calculation .................................................................................................... 85 Gross energy production................................................................................................................ 85 Energy production losses ............................................................................................................... 86 Net energy production ................................................................................................................... 86 16.4 Uncertainty analysis ................................................................................................................... 87 16.5 Turbulence analysis.................................................................................................................... 87 References ............................................................................................................................................ 89 Annex A Coordinates of the sites...................................................................................................... 90 Annex B Mast instrument serial number and calibration information ............................................... 91 Site 1: Peshawar, Khyber Pakhtunkhwa.......................................................................................... 91 Site 2: Haripur, Khyber Pakhtunkhwa ............................................................................................. 91 Site 3: Chakri, Punjab ..................................................................................................................... 91 Site 4: Quaidabad, Punjab .............................................................................................................. 92 Site 5: Bahawalpur, Punjab ............................................................................................................ 92 Site 6: Sadiqabad, Punjab ............................................................................................................... 92 Site 7: Quetta, Balochistan ............................................................................................................. 93 Site 8: Sanghar, Sindh .................................................................................................................... 93 Site 9: Umerkot, Sindh ................................................................................................................... 93 Site 10: Tando Ghulam Ali, Sindh ................................................................................................... 94 Site 11: Gwadar, Balochistan.......................................................................................................... 94 Site 12: Sujawal, Sindh ................................................................................................................... 94 Annex C Data recovery rates over the short-term ............................................................................ 95 Annex D The wind regime observed over the short-term.................................................................. 99 Site 1: Peshawar, Khyber Pakhtunkhwa.......................................................................................... 99 Site 2: Haripur, Khyber Pakhtunkhwa ............................................................................................. 99 Site 3: Chakri, Punjab ................................................................................................................... 100 Site 4: Quaidabad, Punjab ............................................................................................................ 100 Site 5: Bahawalpur, Punjab .......................................................................................................... 101 Site 6: Sadiqabad, Punjab ............................................................................................................. 101 Site 7: Quetta, Balochistan ........................................................................................................... 102 Site 8: Sanghar, Sindh .................................................................................................................. 103 Site 9: Umerkot, Sindh ................................................................................................................. 103 Site 10: Tando Ghulam Ali, Sindh ................................................................................................. 104 Site 11: Gwadar, Balochistan........................................................................................................ 104 Site 12: Sujawal, Sindh ................................................................................................................. 105 Annex E The wind regime estimated over the long-term ................................................................ 106 Site 6: Sadiqabad, Punjab ............................................................................................................. 106 Site 8: Sanghar, Sindh .................................................................................................................. 106 Site 9: Umerkot, Sindh ................................................................................................................. 107 Site 10: Tando Ghulam Ali, Sindh ................................................................................................. 107 Site 11: Gwadar, Balochistan........................................................................................................ 108 Site 12: Sujawal, Sindh ................................................................................................................. 108 Annex F Weibull parameters of the short-term wind regime.......................................................... 110 Annex G Weibull parameters of the long-term wind regime ........................................................... 111 Annex H Comparison of predicted wind regime and estimated wind regime .................................. 112 Annex I Short-term seasonal and diurnal variations in wind characteristics ................................... 113 Site 1: Peshawar, Khyber Pakhtunkhwa........................................................................................ 113 Site 2: Haripur, Khyber Pakhtunkhwa ........................................................................................... 115 Site 3: Chakri, Punjab ................................................................................................................... 117 Site 4: Quaidabad, Punjab ............................................................................................................ 119 Site 5: Bahawalpur, Punjab .......................................................................................................... 121 Site 6: Sadiqabad, Punjab ............................................................................................................. 123 Site 7: Quetta, Balochistan ........................................................................................................... 125 Site 8: Sanghar, Sindh .................................................................................................................. 127 Site 9: Umerkot, Sindh ................................................................................................................. 129 Site 10: Tando Ghulam Ali, Sindh ................................................................................................. 131 Site 11: Gwadar, Balochistan........................................................................................................ 133 Site 12: Sujawal, Sindh ................................................................................................................. 135 Annex J Long-term correlation coefficients.................................................................................... 137 Site 1: Peshawar, Khyber Pakhtunkhwa........................................................................................ 137 Site 2: Haripur, Khyber Pakhtunkhwa ........................................................................................... 138 Site 3: Chakri, Punjab ................................................................................................................... 139 Site 4: Quaidabad, Punjab ............................................................................................................ 140 Site 5: Bahawalpur, Punjab .......................................................................................................... 141 Site 6: Sadiqabad, Punjab ............................................................................................................. 142 Site 7: Quetta, Balochistan ........................................................................................................... 143 Site 8: Sanghar, Sindh .................................................................................................................. 144 Site 9: Umerkot, Sindh ................................................................................................................. 145 Site 10: Tando Ghulam Ali, Sindh ................................................................................................. 146 Site 11: Gwadar, Balochistan........................................................................................................ 147 Site 12: Sujawal, Sindh ................................................................................................................. 148 Annex K Estimates of equivalent mean and directional Weibull wind speed distributions .............. 149 Site 1: Peshawar, Khyber Pakhtunkhwa........................................................................................ 149 Site 2: Haripur, Khyber Pakhtunkhwa ........................................................................................... 151 Site 3: Chakri, Punjab ................................................................................................................... 152 Site 4: Quaidabad, Punjab ............................................................................................................ 154 Site 5: Bahawalpur, Punjab .......................................................................................................... 155 Site 6: Sadiqabad, Punjab ............................................................................................................. 157 Site 7: Quetta, Balochistan ........................................................................................................... 158 Site 8: Sanghar, Sindh .................................................................................................................. 160 Site 9: Umerkot, Sindh ................................................................................................................. 161 Site 10: Tando Ghulam Ali, Sindh ................................................................................................. 163 Site 11: Gwadar, Balochistan........................................................................................................ 164 Site 12: Sujawal, Sindh ................................................................................................................. 166 Annex L Uncertainties associated with AEP results – single turbine at mast location ...................... 168 Annex M Mean air density .............................................................................................................. 170 Annex N Potential Layout ............................................................................................................... 171 Annex O Calibration certificates ..................................................................................................... 173 1. INTRODUCTION The World Bank in collaboration with the Alternative Energy Development Board (AEDB) - Government of Pakistan is implementing Renewable Energy Mapping Project. In total 12 wind masts were installed at the sites identified in different areas in the province of Khyber Pakhtunkhwa, Punjab, Balochistan, and Sindh between December 2015 and November 2016. They have all completed two full years of measurements and will be dismantled. Of the 12 masts installed, 11 have a height of 80 m. One wind mast (in Quetta) has a height of only 67 m as this was the height authorized by the Civil Aviation Authority (CAA) for that location. The wind masts were designed according to TIA/EIA 222-G and the installation of the masts was carried out according to the Ed.1 version of the IEC Standard 61400-12-1. The design was verified with detailed FEA analysis for required design parameters. Mast characteristics are given in Section 3 but further detailed in the installation report [1]. The aim of this report is; to have a general idea about the wind regime of different parts of Pakistan. 1 2. SITE OVERVIEW Twelve wind masts are installed in the provinces of Khyber Pakhtunkhwa, Punjab, Balochistan, and Sindh. The sites are located either on private or on government land and they all have flat terrain characteristics. All wind mast sites are easily accessible by any type of vehicles. The following table present the overview of the 12 sites. Site # 1 2 3 4 5 6 Site name Peshawar Haripur Chakri Quaidabad Bahawalpur Sadiqabad Province Khyber Khyber Punjab Punjab Punjab Punjab Pakhtunkhwa Pakhtunkhwa Host institution UET Jalozai - - - QA Solar Park - Campus (Univ. (Quaid-e-Azam of Engineering Solar Park) & Technology) Land government private land private land - government private land land land, secured by QA Solar Park administration by fencing around site Site access 180 km (2.5- 90 km (2-hour 60 km (1-hour 280 km (4-hour 25 km (30- 25 km (30- hour drive) drive) from drive) from drive) from minute drive) minute drive) from Islamabad; Islamabad; Islamabad; Islamabad; from from Sheikh easily accessible easily accessible easily accessible easily accessible Bahawalpur Zayed Int’l by any type of by any type of by any type of by any type of Airport; Airport; vehicles vehicles vehicles vehicles easily accessible easily accessible by any type of by any type of vehicles vehicles Site description Land with flat Land with Land with flat Land with flat Land with flat Land with flat terrain with no relatively flat terrain with no terrain and is terrain and is terrain and is major terrain, mainly obstruction or wide open with wide open with wide open with obstruction or open farmland roughness in no obstruction no obstruction no obstruction roughness in with distributed the in the in the in the the rows of trees surroundings of surroundings of surroundings of surroundings of surroundings of (~5-7m high) the wind mast wind mast the wind mast the wind mast. the wind mast and low The surrounding buildings in area of the mast surroundings of is used for wind mast cultivation of crops 2 Site # 7 8 9 10 11 12 Site name Quetta Sanghar Umerkot Tando Ghulam Gwadar Sujawal Ali Province Balochistan Sindh Sindh Sindh Balochistan Sindh Host institution BUITEMS - - - GIT (Gwadar - (Balochistan Institute of Univ. of Technology) Information Technology, Engineering & Management Sciences) Land - private land private land private land government private land land Site access 8 km from the 250 km (4-hour 305 km (5-hour 160 km (4-hour 641 km (1.5- 160 km (4- hour Quetta Airport; drive) from drive) from drive) from hour flight) drive) from easily accessible Jinnah Int’l Karachi; Karachi; from Karachi; Karachi; by any type of Airport Karachi; easily accessible easily accessible easily accessible easily accessible vehicles easily accessible by any type of by any type of by any type of by any type of by any type of vehicles vehicles vehicles vehicles vehicles Site description Land with Land with flat Land with flat Land with flat Land with flat Land with flat terrain which is terrain and is terrain with no terrain with no terrain with no terrain with no relatively flat wide open with major obstruction or obstruction or obstruction or no obstruction obstruction or roughness in roughness in roughness in in the roughness in the the the surroundings of the surroundings of surroundings of surroundings of the wind mast surroundings of the wind mast the wind mast wind mast the wind mast 3 3. MAST CHARACTERISTICS All masts are made of triangular lattice tower secured with guy wires. Eleven of the 12 masts installed have a height of 80 m. One wind mast (in Site 7 in Quetta) has a height of only 67 m as this was the height authorized by the Civil Aviation Authority (CAA) for that location. The main characteristics of the masts are presented in the tables below. Site 1 Peshawar 2 Haripur 3 Chakri 4 Quaidabad 5 Bahawalpur 6 Sadiqabad Mast ID# SESI/WB/Jalozai SESI/WB/Donali SESI/WB/Chakri SESI/WB/Quaid SESI/WB/QA SESI/WB/Sadiqa Campus of UET /02/2016 /03/2016 abad/04/2016 Solar/05/2016 bad/06/2016 Peshawar/01/2 016 Area 2500 m2 8000 m2 9000 m2 181603 m2 809371 m2 80937 m2 Mast height 80 m 80 m 80 m 80 m 80 m 80 m Fencing - Around mast Around mast - Around mast Around mast Coordinates Longitude 1 71°47'44.25"E 73°01'59.30"E 72°44'17.71"E 71°53'44.30"E 71°48'56.03"E 70° 0'29.21"E Latitude 33°55'19.51"N 33°58'23.22"N 33°19'13.54"N 32°20'47.72"N 29°19'35.97"N 28°12'48.08"N UTM X 758445 318294 289462 772525 773426 598927 UTM Y 3757038 3760919 3689094 3582540 3247470 3121249 UTM zone 42 43 43 42 42 42 Magnetic +2.73° +2.6° +2.48° +2.32° +1.61° +1.48° declination2 Elevation 387 m 673 m 360 m 192 m 123 m 76 m Site 7 Quetta 8 Sanghar 9 Umerkot 10 Tando 11 Gwadar 12 Sujawal Mast ID# SESI/WB/BUITE SESI/WB/Kandia SESI/WB/Kunri/ SESI/WB/Tando SESI/WB/GIT SESI/WB/Shaha MS/07/2016 ri/08/2016 09/2016 Ghulam Gwadar/11/201 bad/12/2016 Ali/10/2016 6 Area 2500 m2 9712455 m2 2500 m2 404686 m2 2500 m2 809371 m2 Mast height 67 m 80 m 80 m 80 m 80 m 80 m Fencing - Around mast - Around mast Around mast - Coordinates Longitude1 66°56'12.37"E 69° 2'15.12"E 69°34'13.22"E 68°52'31.62"E 62°20'46.95"E 68°11'18.50"E Latitude 30°16'17.71"N 25°48'57.26"N 25° 5'1.75"N 25° 7'24.84"N 25°16'47.30"N 24°31'25.15"N UTM X 301529 503762 557514 487444 434193 417794 UTM Y 3350682 2855298 2774350 2778636 2796092 2712446 UTM zone 42 42 42 42 41 42 Magnetic +2.6° +1.07° +0.9° +0.93° +1.38° +0.85° declination2 Elevation 1582 m 20 m 17 m 25 m 13 m 17 m 1 All coordinates are using the WGS84 datum. 2 All orientations in this document are provided with respect to magnetic North, which is only a few degrees off true North, as mentioned. However, the wind direction data used for the calculations have all been corrected with respect to true north considering the specific declination on site. 4 4. CONFIGURATION OF MAST The following figure presents the overview of the meteo mast and the placement of the sensors and equipment on the mast. The numbers in the brackets are for the 67-m mast at Site 7 (Quetta). Aviation Light @ 82 m (Site 7: 67 m) Lightning Arrester @ 82 m (Site 7: 67 m) Lightning Arrester @ 81.5 m (Site 7: 66.5 m) Lightning Arrester @ 81.5 m (Site 7: 66.5 m) Instrument: Thies Anemometer Instrument: Vector Anemometer Model: Thies S11100 Model: Vector S14100 Height: 80 m (Site 7: 64 m) Height: 80 m (Site 7: 64 m) Instrument: Thies Wind Vane TMR Inst: Thies S21110H Height: 78.5 m (Site 7: 62 m) Instrument: Temperature Sensor Model: S42100 Height: 76 m (Site 7: 61 m) Instrument: Thies Anemometer Inst: Thies S111000 Height: 60 m (Site 7: 60 m) Instrument: Thies Wind Vane TMR Inst: Thies S21110H Height: 58.5 m (Site 7: 58 m) Instrument: Thies Anemometer Inst: Thies S111000 Height: 40 m (Site 7: 40 m) Instrument: Thies Anemometer Inst: Thies S111000 Height: 20 m (Site 7: 20 m) Instrument: Humidity/Temperature Sensor Solar PV Module (Mono-60W) Model: S52100 Data logger Instrument: Air Pressure Sensor Model: Meteo-40M Inst: S31200M UMTS Modem # 359 998 044393504 Figure 1: Configuration of mast 5 4.1 MAST INSTRUMENTATION The make and model of the measurement instruments are given in the table below. The instrument serial number, calibration certificate reference, slope and offset, and boom orientation for each mast are given in Annex B. Instrument Make Model Anemometer Thies S11100 Anemometer Vector S14100 Anemometer Thies S11100 Anemometer Thies S11100 Anemometer Thies S11100 Wind Vane Thies S21110H Wind Vane Thies S21110H Temperature Sensor Galltec TPC1.S/6-ME Temperature & Humidity Sensor Galltec KPC1.S/6-ME Barometer Ammonit AB 100 The heights of the measurement instruments are as follows: Site 1 Peshawar 2 Haripur 3 Chakri 4 Quaidabad 5 Bahawalpur 6 Sadiqabad Anemometer 80 m 80 m 80 m 80 m 80 m 80 m Anemometer 80 m 80 m 80 m 80 m 80 m 80 m Anemometer 60 m 60 m 60 m 60 m 60 m 60 m Anemometer 40 m 40 m 40 m 40 m 40 m 40 m Anemometer 20 m 20 m 20 m 20 m 20 m 20 m Wind Vane 78.5 m 78.5 m 78.5 m 78.5 m 78.5 m 78.5 m Wind Vane 58.5 m 58.5 m 58.5 m 58.5 m 58.5 m 58.5 m Temp 76 m 76 m 76 m 76 m 76 m 76 m Temp & hum 5m 5m 5m 5m 5m 10 m Barometer 4m 4m 4m 4m 4m 7m Site 7 Quetta 8 Sanghar 9 Umerkot 10 Tando 11 Gwadar 12 Sujawal Anemometer 64 m 80 m 80 m 80 m 80 m 80 m Anemometer 64 m 80 m 80 m 80 m 80 m 80 m Anemometer 60 m 60 m 60 m 60 m 60 m 60 m Anemometer 40 m 40 m 40 m 40 m 40 m 40 m Anemometer 20 m 20 m 20 m 20 m 20 m 20 m Wind Vane 62 m 78.5 m 78.5 m 78.5 m 78.5 m 78.5 m Wind Vane 58 m 58.5 m 58.5 m 58.5 m 58.5 m 58.5 m Temp 61 m 76 m 76 m 76 m 76 m 76 m Temp & hum 5m 5m 5m 10 m 5m 5m Barometer 4m 4m 3m 6m 4m 4m 6 5. SITE 1 – PESHAWAR, KHYBER PAKHTUNKHWA 5.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with some of the other masts), the period covering 2 complete years (15/09/2016 to 14/09/2018) is selected. The mast shading effect is corrected by alternatively using the measurements of both top anemometers depending on the wind direction. Height Primary Secondary Wind directions where secondary anemometer anemometer anemometer is used 80 m Thies Vector 54 ° - 102 ° The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 5.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% 7 confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. Some of the considered long-term reference datasets have been discarded from further analysis despite good correlation due to their inconsistent behavior over time. Although the data availability is high for some other datasets, for none of them, a high enough correlation has been obtained with the measurements (much lower than 0.7 both for r (all data) and for r (monthly averages)). Therefore, no long-term extrapolation has been applied. The short-term data has been used for the further purpose of the study. This has been taken into account in the uncertainty analysis (Section 5.4). 5.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5]) based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 5.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. 8 Figure 2: Elevation map 15x15km (with mast in center) with Figure 3: Ground roughness map 20x20km (with mast in 10m elevation difference between lines. Altitudes in map center). Background roughness length is 0.07m, range from 286m to 1169m (warmer colors indicate higher corresponding to distributed rows of trees and low buildings. altitudes). RIX3 value at mast is 0.0% using radius of 3,500m, Roughness length for specific areas are 0.50m for towns (rose steepness threshold of 30% (17°) and frequency distributed color), 0m for rivers (yellow color) and 0.03m for open field directional weight with distributed rows of trees and low buildings (purple color) Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, WAsP parameters are adapted so that the calculated wind shear agrees with the short- term measured wind shear. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 5.1) and the vertical wind speed profile calculated by WAsP from the measurements at 80 m AGL and from the adapted model parameters are given in the following graph. 3 The ruggedness index (RIX) at a specific location is the percentage of the ground surface that has a slope above a given threshold (here 30%) within a certain distance (here 3.5 km). 9 Figure 4: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 80 m AGL 5.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated for a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6]. For this project, air density at hub height is estimated to be 1.081 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. 10 • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.7 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.6 Net energy production The expected wind farm energy production figures at the mast location are as follows: Gross energy production [MWh/y] 1,118 Total energy production losses [%] 5,6 Net energy production (AEP) [MWh/y] 1,055 Net full load equivalent hours [h/y] 528 Net capacity factor [%] 6.0 5.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent, and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) 11 The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 5.3 Long-term extrapolation [% AEP] 0.0 Vertical extrapolation [% AEP] 0.0 Future wind variability [20 years] [% AEP] 17.3 Spatial variation [% AEP] 0.0 Power curve [% AEP] 10.0 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 21.3 5.5 TURBULENCE ANALYSIS The measured turbulence at 80 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 80 m AGL is above the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3) for wind speeds between 8.5-9.5 m/s, 10.5-11.5 m/s, 12.5- 16.5 m/s and above 18.5 m/s. Figure 5: Turbulence at 80 m AGL compared to the IEC curves 12 6. SITE 2 – HARIPUR, KHYBER PAKHTUNKHWA 6.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with some of the other masts), the period covering 2 complete years (15/09/2016 to 14/09/2018) is selected. The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 6.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] in order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. 13 Some of the considered long-term reference datasets have been discarded from further analysis despite good correlation due to their inconsistent behavior over time. Although the data availability is high for some other datasets, for none of them, a high enough correlation has been obtained with the measurements (much lower than 0.7 both for r (all data) and for r (monthly averages)). Therefore, no long-term extrapolation has been applied. The short-term data has been used for the further purpose of the study. This has been taken into account in the uncertainty analysis (Section 6.4). 6.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5] based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 6.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. 14 Figure 7: Ground roughness map 20x20km (with mast in center). Background roughness length is 0.1m, corresponding to open field with distributed rows of trees and low buildings. Roughness length for specific areas are 0.4m for towns and Figure 6: Elevation map 15x15km (with mast in center) with forests (rose color), 0m for lakes (yellow color) and 0.3m for 10m elevation difference between lines. Altitudes in map one specific area close to mast covered with fruit trees (black range from 460m to 1600m (warmer colors indicate higher color) altitudes). RIX3 value at mast is 1.5% using radius of 3,500m, steepness threshold of 30% (17°) and frequency distributed directional weight Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, WAsP parameters are adapted so that the calculated wind shear agrees with the short- term measured wind shear. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 6.1) and the vertical wind speed profile calculated by WAsP from the measurements at 80 m AGL and based on the adapted model parameters are given in the following graph. 15 Figure 8: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 80 m AGL 6.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated with a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6]. For this project, air density at hub height is estimated to be 1.124 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. 16 • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.5 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.4 Net energy production The expected wind farm energy production figures at the mast location are as follows: Gross energy production [MWh/y] 1,623 Total energy production losses [%] 5.4 Net energy production (AEP) [MWh/y] 1,536 Net full load equivalent hours [h/y] 768 Net capacity factor [%] 8.8 6.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) 17 The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 8.8 Long-term extrapolation [% AEP] 0.0 Vertical extrapolation [% AEP] 0.0 Future wind variability [20 years] [% AEP] 19.3 Spatial variation [% AEP] 0.0 Power curve [% AEP] 11.0 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 24.6 6.5 TURBULENCE ANALYSIS The measured turbulence at 80 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 80 m AGL is above the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3) for wind speeds above 9.5 m/s. Figure 9: Turbulence at 80 m AGL compared to the IEC curves 18 7. SITE 3 – CHAKRI, PUNJAB 7.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with some of the other masts), the period covering 2 complete years (01/10/2016 to 30/09/2018) is selected. The mast shading effect is corrected by alternatively using the measurements of both top anemometers depending on the wind direction. Height Primary Secondary Wind directions where secondary anemometer anemometer anemometer is used 80 m Thies Vector 335 ° - 20 ° The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 7.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] in order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% 19 confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. Some of the considered long-term reference datasets have been discarded from further analysis despite good correlation due to their inconsistent behavior over time. 3E considers 3 state-of-the-art long-term extrapolation methods: Linear regression MCP, Matrix MCP and Wind Index. 3E only considers MCP methods if r (all data) exceeds a threshold of 0.7. For the Wind Index method, 3E considers that the same threshold applies, but this time using the monthly averaged r-value. For each selected reference dataset, 3E applies the applicable extrapolation method(s), depending on r (all data) and r (monthly averages). The least uncertainty is obtained from ERA-Interim N32.6 E73.1 data using the Wind Index method, which is therefore the selected combination of reference data and extrapolation method. The result of the Wind Index method is a long-term correction factor, as presented in the following table. This correction factor is applied to the energy production calculated from the short-term wind regime. It should be noted that an additional uncertainty contribution is introduced in the uncertainty assessment to account for the fact that this method relies on the assumption that the short-term wind rose is representative of the long-term. Height AGL [m] 80 Long-term period [-] 15 years (1/8/2003-31/7/2018) Long-term correction factor [-] 0.85 Figure 10: Annual windiness relative to last concurrent year 20 7.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5]) based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 7.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. 21 Figure 11: Elevation map 15x15km (with mast in center) with Figure 12: Ground roughness map 20x20km (with mast in 10m elevation difference between lines. Altitudes in map center). Background roughness length is 0.03m, range from 330m to 550m (warmer colors indicate higher corresponding to open field with distributed rows of trees and altitudes). RIX3 value at mast is 0.2% using radius of 3,500m, buildings. Roughness length for specific areas are 0.53m for steepness threshold of 30% (17°) and frequency distributed towns and forests (rose color), 0.50m for towns (rose color) directional weight and 0m for rivers (yellow color) Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, the calculated wind shear agrees with the short-term measured wind shear. Therefore, no specific model calibration is necessary. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 7.1) and the vertical wind speed profile calculated by WAsP from the measurements at 80 m AGL are given in the following graph. 22 Figure 13: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 80 m AGL 7.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated with a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6]. For this project, air density at hub height is estimated to be 1.120 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. 23 • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.7 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.6 Net energy production The expected wind farm energy production figures at the mast location are as follows: Gross energy production [MWh/y] 1,339 Total energy production losses [%] 5.6 Net energy production (AEP) [MWh/y] 1,263 Net full load equivalent hours [h/y] 632 Net capacity factor [%] 7.2 7.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) 24 The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 6.3 Long-term extrapolation [% AEP] 9.9 Vertical extrapolation [% AEP] 0.0 Future wind variability [20 years] [% AEP] 5.6 Spatial variation [% AEP] 0.0 Power curve [% AEP] 9.5 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 16.2 7.5 TURBULENCE ANALYSIS The measured turbulence at 80 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 80 m AGL is above the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3) for wind speeds above 8.0 m/s. Figure 14: Turbulence at 80 m AGL compared to the IEC curves 25 8. SITE 4 – QUAIDABAD, PUNJAB 8.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with some of the other masts), the period covering 2 complete years (15/09/2016 to 14/09/2018) is selected. The mast shading effect is corrected by alternatively using the measurements of both top anemometers depending on the wind direction. Height Primary Secondary Wind directions where secondary anemometer anemometer anemometer is used 80 m Vector Thies 342 ° - 15 ° The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 8.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] in order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% 26 confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. Some of the considered long-term reference datasets have been discarded from further analysis despite good correlation due to their inconsistent behavior over time. 3E considers 3 state-of-the-art long-term extrapolation methods: Linear regression MCP, Matrix MCP and Wind Index. 3E only considers MCP methods if r (all data) exceeds a threshold of 0.7. For the Wind Index method, 3E considers that the same threshold applies, but this time using the monthly averaged r-value. For each selected reference dataset, 3E applies the applicable extrapolation method(s), depending on r (all data) and r (monthly averages). The least uncertainty is obtained from ERA-5 N32.2 E72.0 data using the Wind Index method, which is therefore the selected combination of reference data and extrapolation method. The result of the Wind Index method is a long-term correction factor, as presented in the following table. This correction factor is applied to the energy production calculated from the short-term wind regime. It should be noted that an additional uncertainty contribution is introduced in the uncertainty assessment to account for the fact that this method relies on the assumption that the short-term wind rose is representative of the long-term. Height AGL [m] 80 Long-term period [-] 18 years (1/7/2000-30/6/2018) Long-term correction factor [-] 1.03 Figure 15: Annual windiness relative to last concurrent year 27 8.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5]) based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 8.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. Figure 16: Elevation map 15x15km (with mast in center) with Figure 17: Ground roughness map 20x20km (with mast in 10m elevation difference between lines. Altitudes in map center). Background roughness length is 0.07m, 28 range from 187m to 816m (warmer colors indicate higher corresponding to distributed rows of trees and low buildings. altitudes). RIX3 value at mast is 0% using radius of 3,500m, Roughness length for specific areas are 0.50m for towns (rose steepness threshold of 30% (17°) and frequency distributed color), 0m for rivers (yellow color) and 0.03m for open field directional weight with distributed rows of trees and low buildings (purple color) Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, the calculated wind shear agrees with the short-term measured wind shear. Therefore, no specific model calibration is necessary. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 8.1) and the vertical wind speed profile calculated by WAsP from the measurements at 80 m AGL are given in the following graph. Figure 18: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 80 m AGL 8.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated with a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. 29 Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6]. For this project, air density at hub height is estimated to be 1.151 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.5 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.4 Net energy production The expected wind farm energy production figures at the mast location are as follows: Gross energy production [MWh/y] 2,920 Total energy production losses [%] 5.4 Net energy production (AEP) [MWh/y] 2,761 Net full load equivalent hours [h/y] 1,381 Net capacity factor [%] 15.7 30 8.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 3.9 Long-term extrapolation [% AEP] 8.6 Vertical extrapolation [% AEP] 0.0 Future wind variability [20 years] [% AEP] 4.3 Spatial variation [% AEP] 0.0 Power curve [% AEP] 9.8 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 14.6 8.5 TURBULENCE ANALYSIS The measured turbulence at 80 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 80 m AGL is above the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3) for wind speeds between 12.5 m/s. 31 Figure 19: Turbulence at 80 m AGL compared to the IEC curves 32 9. SITE 5 – BAHAWALPUR, PUNJAB 9.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with some of the other masts), the period covering 2 complete years (01/10/2016 to 30/09/2018) is selected. The mast shading effect is corrected by alternatively using the measurements of both top anemometers depending on the wind direction. Height Primary Secondary Wind directions where secondary anemometer anemometer anemometer is used 80 m Thies Vector 264 ° - 295 ° The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 9.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] in order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% 33 confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. Some of the considered long-term reference datasets have been discarded from further analysis despite good correlation due to their inconsistent behavior over time. 3E considers 3 state-of-the-art long-term extrapolation methods: Linear regression MCP, Matrix MCP and Wind Index. 3E only considers MCP methods if r (all data) exceeds a threshold of 0.7. For the Wind Index method, 3E considers that the same threshold applies, but this time using the monthly averaged r-value. For each selected reference dataset, 3E applies the applicable extrapolation method(s), depending on r (all data) and r (monthly averages). The least uncertainty is obtained from ERA-5 N29.4 E71.9 data using the Wind Index method, which is therefore the selected combination of reference data and extrapolation method. The result of the Wind Index method is a long-term correction factor, as presented in the following table. This correction factor is applied to the energy production calculated from the short-term wind regime. It should be noted that an additional uncertainty contribution is introduced in the uncertainty assessment to account for the fact that this method relies on the assumption that the short-term wind rose is representative of the long-term. Height AGL [m] 80 Long-term period [-] 16 years (1/8/2002-31/7/2018) Long-term correction factor [-] 1.02 Figure 20: Annual windiness relative to last concurrent year 34 9.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5]) based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 9.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. Figure 22: Ground roughness map 20x20km (with mast in center). Background roughness length is 0.06m, corresponding to open field with distributed rows of trees and Figure 21: Elevation map 15x15km (with mast in center) with low buildings. Roughness length for specific areas are 0.03m 10m elevation difference between lines. Altitudes in map range from 110m to 135m (warmer colors indicate higher 35 altitudes). RIX3 value at mast is 0% using radius of 3,500m, for arid areas (purple color), 0.50m for towns (pink color) and steepness threshold of 30% (17°) and frequency distributed 0.13m for agricultural natural vegetation (mustard color) directional weight Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, the calculated wind shear agrees with the short-term measured wind shear. Therefore, no specific model calibration is necessary. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 9.1) and the vertical wind speed profile calculated by WAsP from the measurements at 80 m AGL are given in the following graph. Figure 23: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 80 m AGL 9.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated with a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. 36 Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6]. For this project, air density at hub height is estimated to be 1.157 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.7 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.6 Net energy production The expected wind farm energy production figures at the mast location are as follows: Gross energy production [MWh/y] 4,243 Total energy production losses [%] 5.6 Net energy production (AEP) [MWh/y] 4,005 Net full load equivalent hours [h/y] 2,002 Net capacity factor [%] 22.8 37 9.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 5.5 Long-term extrapolation [% AEP] 8.6 Vertical extrapolation [% AEP] 0.0 Future wind variability [20 years] [% AEP] 4.7 Spatial variation [% AEP] 0.0 Power curve [% AEP] 8.2 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 14.0 9.5 TURBULENCE ANALYSIS The measured turbulence at 80 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 80 m AGL is above the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3) for wind speeds between 15.5-18.5 m/s and 20.5-21.5 m/s. 38 Figure 24: Turbulence at 80 m AGL compared to the IEC curves 39 10. SITE 6 – SADIQABAD, PUNJAB 10.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with some of the other masts), the period covering 2 complete years (01/10/2016 to 30/09/2018) is selected. The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 10.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] in order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. Some of the considered long-term reference datasets have been discarded from further analysis despite good correlation due to their inconsistent behavior over time. 40 3E considers 3 state-of-the-art long-term extrapolation methods: Linear regression MCP, Matrix MCP and Wind Index. 3E only considers MCP methods if r (all data) exceeds a threshold of 0.7. For the Wind Index method, 3E considers that the same threshold applies, but this time using the monthly averaged r-value. For each selected reference dataset, 3E applies the applicable extrapolation method(s), depending on r (all data) and r (monthly averages). The least uncertainty is obtained from ERA-5 N28.2 E70.0 data using the Matrix MCP method, which is therefore the selected combination of reference data and extrapolation method. The results of the long-term extrapolation based on the MCP method is a new time series of expected wind speeds and directions, over the long-term period. The mean Weibull parameters of this new time series are as follows. The Weibull parameters per sector are presented in Annex G. Figure 25: Annual windiness relative to last concurrent year Annex H presents the seasonal and diurnal variations observed over the long-term. 10.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. 41 Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5]) based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 10.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. Figure 27: Ground roughness map 20x20km (with mast in Figure 26: Elevation map 15x15km (with mast in center) with center). Background roughness length is 0.07m, 10m elevation difference between lines. Altitudes in map corresponding to distributed rows of trees and low buildings. range from 69m to 95m (warmer colors indicate higher Roughness length for specific areas are 0.03m for arid areas altitudes). RIX3 value at mast is 0% using radius of 3,500m, (purple color), 0.50m for towns (pink color) steepness threshold of 30% (17°) and frequency distributed directional weight Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. 42 a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, the calculated wind shear agrees with the short-term measured wind shear. Therefore, no specific model calibration is necessary. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 10.1) and the vertical wind speed profile calculated by WAsP from the measurements at 80 m AGL are given in the following graph. Figure 28: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 80 m AGL 10.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated with a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. 43 Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6] For this project, air density at hub height is estimated to be 1.163 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.7 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.6 Net energy production The expected wind farm energy production figures at the mast location are as follows: Mean wind speed [m/s] 5.1 Gross energy production [MWh/y] 3,876 Total energy production losses [%] 5.6 Net energy production (AEP) [MWh/y] 3,658 Net full load equivalent hours [h/y] 1,829 Net capacity factor [%] 20.9 10.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. 44 The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 5.5 Long-term extrapolation [% AEP] 7.9 Vertical extrapolation [% AEP] 0.0 Future wind variability [20 years] [% AEP] 4.6 Spatial variation [% AEP] 0.0 Power curve [% AEP] 8.5 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 13.7 10.5 TURBULENCE ANALYSIS The measured turbulence at 80 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 80 m AGL is above the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3) for wind speeds above 17.5 m/s. 45 Figure 29: Turbulence at 80 m AGL compared to the IEC curves 46 11. SITE 7 – QUETTA, BALOCHISTAN 11.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with some of the other masts), the period covering 2 complete years (01/10/2016 to 30/09/2018) is selected. The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 11.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] in order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. 47 3E considers 3 state-of-the-art long-term extrapolation methods: Linear regression MCP, Matrix MCP and Wind Index. 3E only considers MCP methods if r (all data) exceeds a threshold of 0.7. For the Wind Index method, 3E considers that the same threshold applies, but this time using the monthly averaged r-value. For each selected reference dataset, 3E applies the applicable extrapolation method(s), depending on r (all data) and r (monthly averages). The least uncertainty is obtained from ERA-5 N30.5 E66.9 data using the Wind Index method, which is therefore the selected combination of reference data and extrapolation method. The result of the Wind Index method is a long-term correction factor, as presented in the following table. This correction factor is applied to the energy production calculated from the short-term wind regime. It should be noted that an additional uncertainty contribution is introduced in the uncertainty assessment to account for the fact that this method relies on the assumption that the short-term wind rose is representative of the long-term. Height AGL [m] 64 Long-term period [-] 18 years (1/8/2000-31/7/2018) Long-term correction factor [-] 0.98 Figure 30: Annual windiness relative to last concurrent year 11.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. 48 Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5]) based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 11.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. Figure 31: Elevation map 15x15km (with mast in center) with Figure 32: Ground roughness map 20x20km (with mast in 10m elevation difference between lines. Altitudes in map center). Background roughness length is 0.03m, range from 1530m to 2760m (warmer colors indicate higher corresponding to arid areas. Roughness length for specific altitudes). RIX3 value at mast is 0.7% using radius of 3,500m, areas are 0.07m for distributed rows of trees and low steepness threshold of 30% (17°) and frequency distributed buildings (purple color), 0.5m for towns (rose color) and 0m directional weight for lakes (yellow color) Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. 49 a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, WAsP parameters are adapted so that the calculated wind shear agrees with the short- term measured wind shear. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 11.1) and the vertical wind speed profile calculated by WAsP from the measurements at 64 m AGL and based on the adapted model parameters are given in the following graph. Figure 33: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 64 m AGL 11.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated with a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. 50 Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6]. For this project, air density at hub height is estimated to be 1.007 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.7 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.6 Net energy production The expected wind farm energy production figures at the mast location are as follows: Gross energy production [MWh/y] 2,309 Total energy production losses [%] 5.6 Net energy production (AEP) [MWh/y] 2,179 Net full load equivalent hours [h/y] 1,089 Net capacity factor [%] 12.4 11.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. 51 The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 4.5 Long-term extrapolation [% AEP] 8.8 Vertical extrapolation [% AEP] 1.8 Future wind variability [20 years] [% AEP] 4.8 Spatial variation [% AEP] 0.0 Power curve [% AEP] 10.3 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 15.5 11.5 TURBULENCE ANALYSIS The measured turbulence at 64 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 64 m AGL is above the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3) for wind speed at 13.5 m/s. 52 Figure 34: Turbulence at 64 m AGL compared to the IEC curves 53 12. SITE 8 – SANGHAR, SINDH 12.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with the other masts), the period covering 2 complete years (11/11/2016 to 10/11/2018) is selected. The mast shading effect is corrected by alternatively using the measurements of both top anemometers depending on the wind direction. Height Primary Secondary Wind directions where secondary anemometer anemometer anemometer is used 80 m Thies Vector 278 ° - 314 ° The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 12.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] in order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% 54 confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. Some of the considered long-term reference datasets have been discarded from further analysis despite good correlation due to their inconsistent behavior over time. 3E considers 3 state-of-the-art long-term extrapolation methods: Linear regression MCP, Matrix MCP and Wind Index. 3E only considers MCP methods if r (all data) exceeds a threshold of 0.7. For the Wind Index method, 3E considers that the same threshold applies, but this time using the monthly averaged r-value. For each selected reference dataset, 3E applies the applicable extrapolation method(s), depending on r (all data) and r (monthly averages). The least uncertainty is obtained from ERA5 N25.7 E69.0 data using the MCP Matrix method, which is therefore the selected combination of reference data and extrapolation method. The results of the long-term extrapolation based on the MCP method is a new time series of expected wind speeds and directions, over the long-term period. The mean Weibull parameters of this new time series are as follows. The Weibull parameters per sector are presented in Annex G. Figure 35: Annual windiness relative to last concurrent year Annex H presents the seasonal and diurnal variations observed over the long-term. 12.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the 55 present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5]) based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 12.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. Figure 36: Elevation map 15x15km (with mast in center) with Figure 37: Ground roughness map 20x20km (with mast in 10m elevation difference between lines. Altitudes in map center). Background roughness length is 0.07m, range from 13m to 25m (warmer colors indicate higher corresponding to open field with distributed rows of trees and altitudes). RIX3 value at mast is 0% using radius of 3,500m, low buildings. Roughness length for specific areas are 0.5m for towns (rose color) and 0m for rivers (yellow color) 56 steepness threshold of 30% (17°) and frequency distributed directional weight Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, the calculated wind shear agrees with the short-term measured wind shear. Therefore, no specific model calibration is necessary. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 12) and the vertical wind speed profile calculated by WAsP from the measurements at 80 m AGL are given in the following graph. Figure 38: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 80 m AGL 12.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated with a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. 57 Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6]. For this project, air density at hub height is estimated to be 1.160 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.7 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.6 Net energy production The expected wind farm energy production figures at the mast location are as follows: Mean wind speed [m/s] 6.80 Gross energy production [MWh/y] 7,316 Total energy production losses [%] 5.6 Net energy production (AEP) [MWh/y] 6,905 Net full load equivalent hours [h/y] 3,452 Net capacity factor [%] 39.4 58 12.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 3.1 Long-term extrapolation [% AEP] 3.2 Vertical extrapolation [% AEP] 0.0 Future wind variability [20 years] [% AEP] 3.4 Spatial variation [% AEP] 0.0 Power curve [% AEP] 5.7 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 8.1 12.5 TURBULENCE ANALYSIS The measured turbulence at 80 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 80 m AGL is above the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3) for wind speeds above 18.5m/s. 59 Figure 39: Turbulence at 80 m AGL compared to the IEC curves 60 13. SITE 9 – UMERKOT, SINDH 13.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with the other masts), the period covering 2 complete years (11/11/2016 to 10/11/2018) is selected. The mast shading effect is corrected by alternatively using the measurements of both top anemometers depending on the wind direction. Height Primary Secondary Wind directions where secondary anemometer anemometer anemometer is used 80 m Vector Thies 111 ° - 159 ° The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 13.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] in order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% 61 confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. Some of the considered long-term reference datasets have been discarded from further analysis despite good correlation due to their inconsistent behavior over time. 3E considers 3 state-of-the-art long-term extrapolation methods: Linear regression MCP, Matrix MCP and Wind Index. 3E only considers MCP methods if r (all data) exceeds a threshold of 0.7. For the Wind Index method, 3E considers that the same threshold applies, but this time using the monthly averaged r-value. For each selected reference dataset, 3E applies the applicable extrapolation method(s), depending on r (all data) and r (monthly averages). The least uncertainty is obtained from ERA5 N24.9 E69.6 data using the MCP-Regression method, which is therefore the selected combination of reference data and extrapolation method. The results of the long-term extrapolation based on the MCP method is a new time series of expected wind speeds and directions, over the long-term period. The mean Weibull parameters of this new time series are as follows. The Weibull parameters per sector are presented in Annex G. Figure 40: Annual windiness relative to last concurrent year Annex H presents the seasonal and diurnal variations observed over the long-term. 62 13.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5]) based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 13.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. 63 Figure 41: Elevation map 15x15km (with mast in center) with Figure 42: Ground roughness map 20x20km (with mast in 10m elevation difference between lines. Altitudes in map center). Background roughness length is 0.07m, range from 0m to 84m (warmer colors indicate higher corresponding to open field with distributed rows of trees and altitudes). RIX3 value at mast is 0% using radius of 3,500m, low buildings. Roughness length for specific areas are 0.003m steepness threshold of 30% (17°) and frequency distributed for desert (blue color) directional weight Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, WAsP parameters are adapted so that the calculated wind shear agrees with the short- term measured wind shear. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 13.1) and the vertical wind speed profile calculated by WAsP from the measurements at 80 m AGL and based on the adapted model parameters are given in the following graph. 64 Figure 43: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 80 m AGL 13.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated with a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6] For this project, air density at hub height is estimated to be 1.154 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. 65 • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.7 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.6 Net energy production The expected wind farm energy production figures at the mast location are as follows: Mean wind speed [m/s] 6.50 Gross energy production [MWh/y] 6.765 Total energy production losses [%] 5.6 Net energy production (AEP) [MWh/y] 6.385 Net full load equivalent hours [h/y] 3.192 Net capacity factor [%] 36.4 13.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) 66 The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 3.4 Long-term extrapolation [% AEP] 3.5 Vertical extrapolation [% AEP] 0.0 Future wind variability [20 years] [% AEP] 3.7 Spatial variation [% AEP] 0.0 Power curve [% AEP] 6.0 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 8.6 13.5 TURBULENCE ANALYSIS The measured turbulence at 80 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 80 m AGL is above the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3) for wind speeds above 18.5-19.5 m/s. Figure 44: Turbulence at 80 m AGL compared to the IEC curves 67 14. SITE 10 – TANDO GHULAM ALI, SINDH 14.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with some of the other masts), the period covering 2 complete years (01/10/2016 to 30/09/2018) is selected. The mast shading effect is corrected by alternatively using the measurements of both top anemometers depending on the wind direction. Height Primary Secondary Wind directions where secondary anemometer anemometer anemometer is used 80 m Thies Vector 142 ° - 186 ° The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 14.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] in order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% 68 confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. Some of the considered long-term reference datasets have been discarded from further analysis despite good correlation due to their inconsistent behavior over time. 3E considers 3 state-of-the-art long-term extrapolation methods: Linear regression MCP, Matrix MCP and Wind Index. 3E only considers MCP methods if r (all data) exceeds a threshold of 0.7. For the Wind Index method, 3E considers that the same threshold applies, but this time using the monthly averaged r-value. For each selected reference dataset, 3E applies the applicable extrapolation method(s), depending on r (all data) and r (monthly averages). The least uncertainty is obtained from ERA-5 N25.2 E68.7 data using the MatrixMCP method, which is therefore the selected combination of reference data and extrapolation method. The results of the long-term extrapolation based on the MCP method is a new time series of expected wind speeds and directions, over the long-term period. The mean Weibull parameters of this new time series are as follows. The Weibull parameters per sector are presented in Annex G. Figure 45: Annual windiness relative to last concurrent year Annex H presents the seasonal and diurnal variations observed over the long-term. 14.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. 69 The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5]) based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 14.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. Figure 47: Ground roughness map 20x20km (with mast in center). Background roughness length is 0.05m, corresponding to open field with distributed rows of trees and Figure 46: Elevation map 15x15km (with mast in center) with low buildings. Roughness length for specific areas are 0.20m 10m elevation difference between lines. Altitudes in map for towns and distributed rows of trees (black color), 0m for range from 10m to 20m (warmer colors indicate higher lakes (yellow color) altitudes). RIX3 value at mast is 0% using radius of 3,500m, steepness threshold of 30% (17°) and frequency distributed directional weight 70 Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, WAsP parameters are adapted so that the calculated wind shear agrees with the short- term measured wind shear. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 14.1) and the vertical wind speed profile calculated by WAsP from the measurements at 80 m AGL and based on the adapted model parameters are given in the following graph. Figure 48: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 80 m AGL 14.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated with a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and 71 hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6] For this project, air density at hub height is estimated to be 1.167 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.7 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.6 Net energy production The expected wind farm energy production figures at the mast location are as follows: Mean wind speed [m/s] 7.10 Gross energy production [MWh/y] 7,932 Total energy production losses [%] 5.6 Net energy production (AEP) [MWh/y] 7,486 Net full load equivalent hours [h/y] 3,743 Net capacity factor [%] 42.7 72 14.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 3.1 Long-term extrapolation [% AEP] 2.9 Vertical extrapolation [% AEP] 0.0 Future wind variability [20 years] [% AEP] 3.5 Spatial variation [% AEP] 0.0 Power curve [% AEP] 5.3 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 7.8 14.5 TURBULENCE ANALYSIS The measured turbulence at 80 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 80 m AGL is above the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3) for wind speeds between 17.5-18.5 m/s. 73 Figure 49: Turbulence at 80 m AGL compared to the IEC curves 74 15. SITE 11 – GWADAR, BALOCHISTAN 15.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with the other masts), the period covering 2 complete years (11/11/2016 to 10/11/2018) is selected. The mast shading effect is corrected by alternatively using the measurements of both top anemometers depending on the wind direction. Height Primary Secondary Wind directions where secondary anemometer anemometer anemometer is used 80 m Thies Vector 84 ° - 117 ° The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 15.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] in order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% 75 confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. 3E considers 3 state-of-the-art long-term extrapolation methods: Linear regression MCP, Matrix MCP and Wind Index. 3E only considers MCP methods if r (all data) exceeds a threshold of 0.7. For the Wind Index method, 3E considers that the same threshold applies, but this time using the monthly averaged r-value. For each selected reference dataset, 3E applies the applicable extrapolation method(s), depending on r (all data) and r (monthly averages). The least uncertainty is obtained from ERA5 N25.2 E62.4 data using the MCP Matrix method, which is therefore the selected combination of reference data and extrapolation method. The results of the long-term extrapolation based on the MCP method is a new time series of expected wind speeds and directions, over the long-term period. The mean Weibull parameters of this new time series are as follows. The Weibull parameters per sector are presented in Annex G. Figure 50: Annual windiness relative to last concurrent year Annex H presents the seasonal and diurnal variations observed over the long-term. 15.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. 76 The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5]) based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 15.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. Figure 51: Elevation map 15x15km (with mast in center) Figure 52: Ground roughness map 20x20km (with mast in with 10m elevation difference between lines. Altitudes in center). Background roughness length is 0.03m, corresponding map range from 0m to 365m (warmer colors indicate higher to open field with distributed rows of trees and low buildings. altitudes). RIX3 value at mast is 0.4% using radius of 3,500m, Roughness length for specific areas are 0.50m for towns (rose steepness threshold of 30% (17°) and frequency distributed color) and 0m for sea and lake (yellow color) directional weight 77 Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, WAsP parameters are adapted so that the calculated wind shear agrees with the short- term measured wind shear. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 15.1) and the vertical wind speed profile calculated by WAsP from the measurements at 80 m AGL and based on the adapted model parameters are given in the following graph. Figure 53: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 80 m AGL 15.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated with a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and 78 hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6] For this project, air density at hub height is estimated to be 1.154 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.7 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.6 Net energy production The expected wind farm energy production figures at the mast location are as follows: Mean wind speed [m/s] 4.70 Gross energy production [MWh/y] 3,191 Total energy production losses [%] 5.6 Net energy production (AEP) [MWh/y] 3,012 Net full load equivalent hours [h/y] 1,506 Net capacity factor [%] 17.2 79 15.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 4.5 Long-term extrapolation [% AEP] 7.5 Vertical extrapolation [% AEP] 0.0 Future wind variability [20 years] [% AEP] 4.7 Spatial variation [% AEP] 0.0 Power curve [% AEP] 9.8 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 14.2 15.5 TURBULENCE ANALYSIS The measured turbulence at 80 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 80 m AGL is below the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3). 80 Figure 54: Turbulence at 80 m AGL compared to the IEC curves 81 16. SITE 12 – SUJAWAL, SINDH 16.1 WIND DATA PROCESSING Short-term wind regime The anemometer calibration parameters found in the calibration reports are applied to the raw data. The data are then cleaned. In order to provide consistent results (and comparable with the other masts), the period covering 2 complete years (01/10/2016 to 30/09/2018) is selected. The mast shading effect is corrected by alternatively using the measurements of both top anemometers depending on the wind direction. Height Primary Secondary Wind directions where secondary anemometer anemometer anemometer is used 80 m Thies Vector 108 ° - 147 ° The Weibull parameters of the short-term wind regime over this period per sector are presented in Annex D. Annex I presents the seasonal and diurnal variations observed over the short-term. Long-term wind regime The long-term extrapolation is performed in three steps: first, the most reliable reference datasets are identified, then the best combination of reference data and extrapolation method is selected. Eventually, the combination of dataset and method resulting in the lowest uncertainty (cf. section 16.4) is selected. 3E selects reference dataset from the following sources: • MERRA-2, ERA5 and post-processed ERA-Interim reanalysis data from WindPRO (4 closest grid points), • Meteorological station data from WindPRO, The following criteria are used to select reference datasets from these sources: • Agreement: the reference dataset should agree with the measurements in terms of wind speed variations over time. This agreement is quantified by the Pearson correlation coefficient “r”. 3E considers a Pearson coefficient of 0.7 (all data or monthly averages) as a minimum prerequisite for a reference dataset to be considered for long-term extrapolation. • Time resolution: the time resolution of the reference dataset should be constant over time. In case time resolution varies, 3E resamples data to a constant time resolution. • Data availability: missing periods should be limited and evenly distributed over time. 3E considers data availability above 80 % as a minimum prerequisite for a reference dataset to be used for long-term extrapolation. • Consistency: the reference dataset should not reveal any abrupt change or unrealistic trend. 3E applies a SNHT test [2] in order to identify discontinuities. If this happens, then the available period is limited to ensure homogeneity. 3E then also applies a Mann-Kendall test [3][4] (90% 82 confidence interval) in order to identify possible trends. Again, the available period is limited to ensure the absence of a trend. When several reference datasets from the same reanalysis project are considered, 3E only selects the one providing the best r (all data) and the one providing the best r (monthly averages). The correlation between the on-site measurements and the reference datasets is presented in Annex J. 3E considers 3 state-of-the-art long-term extrapolation methods: Linear regression MCP, Matrix MCP and Wind Index. 3E only considers MCP methods if r (all data) exceeds a threshold of 0.7. For the Wind Index method, 3E considers that the same threshold applies, but this time using the monthly averaged r-value. For each selected reference dataset, 3E applies the applicable extrapolation method(s), depending on r (all data) and r (monthly averages). The least uncertainty is obtained from ERA-5 N24.6 E68.1 data using the MCP Matrix method, which is therefore the selected combination of reference data and extrapolation method. The results of the long-term extrapolation based on the MCP method is a new time series of expected wind speeds and directions, over the long-term period. The mean Weibull parameters of this new time series are as follows. The Weibull parameters per sector are presented in Annex G. Figure 55: Annual windiness relative to last concurrent year Annex H presents the seasonal and diurnal variations observed over the long-term. 16.2 WIND FLOW MODELLING Terrain features influence the wind flow and thus play a significant role in the spatial extrapolation of the wind regime. The software package WindPRO and the WAsP wind flow model are used in the present study. WAsP requires a terrain model describing elevation, roughness and other relevant obstacles to the wind flow that are not modelled as roughness. 83 The terrain model used in this study represents the current conditions, which are assumed to remain the same over the wind farm lifetime. Terrain model a. Elevation The wind regime can be highly influenced by elevation differences across the site. For this study, terrain elevation is modelled within a radius of 15 km (in line with WAsP recommendations [5]) based on SRTM data. Height contour lines are then generated with an elevation difference of 10 m between two successive lines. It should be noted that SRTM is a digital surface model (DSM), which includes features such as forests and buildings. This is accounted for in the uncertainty assessment (cf. section 16.4). b. Roughness length Roughness length is a key parameter of the equation that governs wind shear. Changes in roughness length cause variations of wind shear, which propagate vertically as the air flows over the site. The impact at measurement or hub height therefore varies with distance to roughness changes, but is also related to atmospheric conditions. Given that roughness length is closely related to land use, terrain roughness is typically modelled using a land-use database. However, no such suitable database with the required quality and resolution is available for the project sites considered in this study. Therefore, the roughness maps have been created manually based on aerial photos and covering an area with a radius of 20km around each site. The roughness length values considered are presented in the caption of the following figures. Figure 57: Ground roughness map 20x20km (with mast in center). Background roughness length is 0.054m, corresponding to open field with distributed rows of trees and low buildings. Roughness length for specific areas are 0.03 m Figure 56: Elevation map 15x15km (with mast in center) with for inland humid zone (purple color) and 0.0005m for water 10m elevation difference between lines. Altitudes in map bodies (red color) range from 5m to 15m (warmer colors indicate higher altitudes). RIX3 value at mast is 0% using radius of 3,500m, steepness threshold of 30% (17°) and frequency distributed directional weight 84 Wind flow model WAsP is used to extrapolate the wind regime to the mast location. It involves two steps: a vertical extrapolation of the wind regime and a horizontal extrapolation of the wind regime. a. Horizontal extrapolation of the wind regime In this study, wind measurements are only available at a single location, which does not allow any validation of the horizontal extrapolation of the wind regime. b. Vertical extrapolation of the wind regime By default, WAsP is configured for atmospheric conditions typical of North-Western Europe. Therefore, parameters sometimes need to be adapted. In particular, some parameters strongly affect the vertical extrapolation of the wind regime and can be validated and calibrated if necessary by comparison of the measured and calculated wind shears. In this study, WAsP parameters are adapted so that the calculated wind shear agrees with the short- term measured wind shear. The mean wind speeds measured at the various heights over the short-term period limited to 2 complete years (cf. Section 16.1) and the vertical wind speed profile calculated by WAsP from the measurements at 80 m AGL and based on the adapted model parameters are given in the following graph. Figure 58: Mean wind speeds measured over the short-term period limited to 2 complete years and vertical wind speed profile calculated by WAsP using measurements at 80 m AGL 16.3 ENERGY PRODUCTION CALCULATION In agreement with the World Bank, the energy production is calculated with a single turbine at the mast location. A generic 2MW turbine with a hub height of 80 m is selected for the purpose of this study. Gross energy production A gross energy production refers to the theoretical energy production that would be achieved if there was no operational loss. It is calculated by combining the wind regime at a wind turbine location and 85 hub height to the power curve specific to the considered wind turbine type and corrected for local hub height air density. This is done using the software WindPRO. Since the energy content of the wind varies proportionally to air density, power curves are adapted accordingly before being used in calculations. The adaptation is done using the new recommended WindPRO method (adjusted IEC 61400-12 method, improved to match turbine control) [6] For this project, air density at hub height is estimated to be 1.167 kg/ m³. Air density is calculated by WindPRO based on the temperature, pressure and humidity measurements from the mast. Energy production losses In addition to energy conversion losses taken into account in the power curve, other losses affect the electrical power expected to be delivered to the grid. The following losses are taken into account in this study and are summarised below. • Wake losses: Wake losses are due to the mutual influence of the wind turbines and are calculated using the N.O. Jensen (EMD) : 2005 wake model implemented in WindPRO. • Unavailability losses: Unavailability losses are due to downtime of the wind turbines or balance of plant (maintenance or technical incidents) as well as downtime of the power grid. • Performance losses: Turbine performance losses are typically due to high wind hysteresis, yaw misalignment, wind flow inclination, turbulence, wind shear and other differences between turbine power curve test conditions. • Electrical losses: Electrical losses occur in cables and transformers ensuring electrical transmission to the wind farm substation. • Environmental losses: Environmental losses account for the performance degradation of the wind turbines due to environmental conditions. The energy production losses defined in the preceding section are summarized below. Wake losses [%] 0.0 Unavailability losses [%] 3.7 Performance losses [%] 0.0 Electrical losses [%] 1.5 Environmental losses [%] 0.5 Total losses [%] 5.6 Net energy production The expected wind farm energy production figures at the mast location are as follows: Mean wind speed [m/s] 7.30 Gross energy production [MWh/y] 8,321 Total energy production losses [%] 5.6 Net energy production (AEP) [MWh/y] 7,853 Net full load equivalent hours [h/y] 3,927 Net capacity factor [%] 44.8 86 16.4 UNCERTAINTY ANALYSIS Some uncertainty components are directly quantified in terms of energy production, whereas some other uncertainty components are first quantified in terms of wind speed, then translated into uncertainties in terms of energy production, by applying a sensitivity factor. The sensitivity factor relates energy production change to wind speed change. The global uncertainty is then calculated from the individual uncertainty components by assuming that they are independent and that the resulting uncertainty follows a normal distribution (central-limit theorem). They can therefore be combined by calculating the square root of the sum of the squares of each uncertainty. In this study, the following sources of uncertainty are considered. • Wind measurements (wind speed) • Long-term extrapolation (wind speed) • Vertical extrapolation (wind speed) • Future wind variability (wind speed) • Spatial variation (wind speed) • Power curve (production) • Energy production losses (production) The table below presents the breakdown of uncertainty analysis in terms of annual energy production. Wind measurements [% AEP] 3.1 Long-term extrapolation [% AEP] 2.9 Vertical extrapolation [% AEP] 0.0 Future wind variability [20 years] [% AEP] 3.2 Spatial variation [% AEP] 0.0 Power curve [% AEP] 5.2 Production losses [% AEP] 1.1 Combined uncertainty [20 years] [% AEP] 7.5 16.5 TURBULENCE ANALYSIS The measured turbulence at 80 m AGL compared to the IEC curves is given below. The effective turbulence intensity measured at 80 m AGL is above the characteristic turbulence intensity of Class A wind turbines (IEC 61400-1, ed. 3) for wind speeds above 19.5 m/s. 87 Figure 59: Turbulence at 80 m AGL compared to the IEC curves 88 REFERENCES [1] Installation report of meteorological masts, 3E, SESI, February 2017 [2] T. Burton, D. Sharpe, N. Jenkins, E. Boussanyi. Wind Energy Handbook. [3] H. Alexandersson, A homogeneity test applied to precipitation data. J. Climatol, 1986 [4] H.B. Mann, Non-parametric tests against trend, Econometrica, 1945 [5] The WAsP team, "WAsP best practices and checklist", DTU, June 2013. [6] WindPro user manual 89 ANNEX A COORDINATES OF THE SITES Site 1 Peshawar 2 Haripur 3 Chakri 4 Quaidabad 5 Bahawalpur 6 Sadiqabad Mast ID# SESI/WB/Jalozai SESI/WB/Donali SESI/WB/Chakri SESI/WB/Quaid SESI/WB/QA SESI/WB/Sadiqa Campus of UET /02/2016 /03/2016 abad/04/2016 Solar/05/2016 bad/06/2016 Peshawar/01/2 016 Area 2500 m2 8000 m2 9000 m2 181603 m2 809371 m2 80937 m2 Mast height 80 m 80 m 80 m 80 m 80 m 80 m Coordinates Latitude (y) 33.922086° 33.973117° 33.320428° 32.346589° 29.326658° 28.213356° Longtitude (x) 71.795625° 73.033139° 72.738253° 71.895639° 71.815564° 70.008114° Elevation 387 m 673 m 360 m 192 m 123 m 76 m Site 7 Quetta 8 Sanghar 9 Umerkot 10 Tando 11 Gwadar 12 Sujawal Mast ID# SESI/WB/BUITE SESI/WB/Kandia SESI/WB/Kunri/ SESI/WB/Tando SESI/WB/GIT SESI/WB/Shaha MS/07/2016 ri/08/2016 09/2016 Ghulam Gwadar/11/201 bad/12/2016 Ali/10/2016 6 Area 2500 m2 9712455 m2 2500 m2 404686 m2 2500 m2 809371 m2 Mast height 67 m 80 m 80 m 80 m 80 m 80 m Coordinates Latitude (y) / 30.271586° 25.815906° 25.083819° 25.123567° 25.279806° 24.523653° Longtitude (x) 66.936769° 69.037532° 69.570339° 68.875450° 62.346375° 68.188472° Elevation 1582 m 20 m 17 m 25 m 13 m 17 m 90 ANNEX B MAST INSTRUMENT SERIAL NUMBER AND CALIBRATION INFORMATION Site 1: Peshawar, Khyber Pakhtunkhwa CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 08 157890 1522609 0.04608 0.2407 80 110° Anemometer Vector S14100 16801 1514328 0.09750 0.1439 80 290° Anemometer Thies S11100 08 157889 1522610 0.04614 0.2141 60 110° Anemometer Thies S11100 08 157888 1522611 0.04597 0.2580 40 110° Anemometer Thies S11100 08 157887 1522612 0.04611 0.2312 20 110° Wind Vane Thies S21110H 05 150047 1521952 1.00106 0.9088 78.5 110° Wind Vane Thies S21110H 05 150046 1521953 1.00044 1.6304 58.5 110° Temperature Galltec TPC1.S/6-ME 154299 A66353A030 100 -30 76 - Sensor Temperature & 100 (T) -30 (T) Humidity Galltec KPC1.S/6-ME 154284 A65630A110 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B14-0489 E 00/03/01 100 600 4 - Site 2: Haripur, Khyber Pakhtunkhwa CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 07 157638 1533648 0.04599 0.2371 80 30° Anemometer Vector S14100 16845 1514943 0.09775 0.2012 80 210° Anemometer Thies S11100 07 157637 1533649 0.04604 0.2238 60 30° Anemometer Thies S11100 07 157636 1533650 0.04594 0.2418 40 30° Anemometer Thies S11100 07 157635 1533651 0.04604 0.2386 20 30° Wind Vane Thies S21110H 05 150044 1521955 1.00037 2.5092 78.5 30° Wind Vane Thies S21110H 05 150043 1521956 1.00043 0.5650 58.5 30° Temperature Galltec TPC1.S/6-ME 154311 A66356A030 100 -30 76 - Sensor Temperature & 100 (T) -30 (T) Humidity Galltec KPC1.S/6-ME 154319 A65630A120 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B14-0480 E 00/03/01 100 600 4 - Site 3: Chakri, Punjab CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 07 157646 1533640 0.04596 0.2528 80 342ᵒ Anemometer Vector S14100 16799 1514326 0.09726 0.1726 80 162ᵒ Anemometer Thies S11100 07 157645 1533641 0.04591 0.2641 60 342ᵒ Anemometer Thies S11100 07 157644 1533642 0.04599 0.2279 40 342ᵒ Anemometer Thies S11100 07 157643 1533643 0.04600 0.2399 20 342ᵒ Wind Vane Thies S21110H 05 150037 1521962 0.00107 1.3334 78.5 342ᵒ Wind Vane Thies S21110H 05 150036 1521963 1.00047 0.9840 58.5 342ᵒ Temperature Galltec TPC1.S/6-ME 154309 A66353A030 100 -30 76 - Sensor Temperature 100 (T) -30 (T) & Humidity Galltec KPC1.S/6-ME 154329 A65630A120 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B15-0242 E 00/03/01 100 600 4 - 91 Site 4: Quaidabad, Punjab CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 07 157678 1513898 0.04600 0.2515 80 -126® Anemometer Vector S14100 16844 1514942 0.09746 0.1985 80 54® Anemometer Thies S11100 07 157677 1513899 0.04600 0.2640 60 -126® Anemometer Thies S11100 07 157676 1513900 0.04600 0.2595 40 -126® Anemometer Thies S11100 07 157675 1513901 0.04599 0.2578 20 -126® Wind Vane Thies S21110H 05 150035 1521964 1.00052 0.9263 78.5 -126® Wind Vane Thies S21110H 05 150034 1521965 1.00099 1.3276 58.5 -126® Temperature Galltec TPC1.S/6-ME 154298 A66353A030 100 -30 76 - Sensor Temperature & 100 (T) -30 (T) Humidity Galltec KPC1.S/6-ME 154283 A65630A110 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B14-0491 E 00/03/01 100 600 4 - Site 5: Bahawalpur, Punjab CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 07 157654 1533626 0.04601 0.2342 80 -70ᵒ Anemometer Vector S14100 16793 1514320 0.09760 0.1517 80 -250ᵒ Anemometer Thies S11100 07 157653 1533627 0.04595 0.2407 60 -70ᵒ Anemometer Thies S11100 07 157652 1533628 0.04600 0.2315 40 -70ᵒ Anemometer Thies S11100 07 157651 1533629 0.04604 0.2168 20 -70ᵒ Wind Vane Thies S21110H 05 150050 1521949 1.00100 1.8315 78.5 -70ᵒ Wind Vane Thies S21110H 05 150049 1521950 1.00042 0.6864 58.5 -70ᵒ Temperature Galltec TPC1.S/6-ME 154303 A66353A030 100 -30 76 - Sensor Temperature & 100 (T) -30 (T) Humidity Galltec KPC1.S/6-ME 154285 A65630A110 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B15-0237 E 00/03/01 100 600 5 - Site 6: Sadiqabad, Punjab CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 07 157658 1533621 0.04606 0.2136 80 -70ᵒ Anemometer Vector S14100 16794 1514321 0.09745 0.1759 80 -250ᵒ Anemometer Thies S11100 07 157657 1533623 0.04603 0.2239 60 -70ᵒ Anemometer Thies S11100 07 157656 1533624 0.04600 0.2333 40 -70ᵒ Anemometer Thies S11100 07 157655 1533625 0.04598 0.2305 20 -70ᵒ Wind Vane Thies S21110H 05 150052 1521947 1.00078 1.6328 78.5 -70ᵒ Wind Vane Thies S21110H 05 150051 1521948 1.00043 2.2360 58.5 -70ᵒ Temperature Galltec TPC1.S/6-ME 154304 A66353A030 100 -30 76 - Sensor Temperature & 100 (T) -30 (T) Humidity Galltec KPC1.S/6-ME 154286 A65630A110 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B15-0238 E 00/03/01 100 600 4 - 92 Site 7: Quetta, Balochistan CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 07 157662 1533617 0.04601 0.2429 64 10° Anemometer Vector S14100 16795 1514329 0.09778 0.1331 64 190° Anemometer Thies S11100 07 157661 1533618 0.04610 0.2119 60 10° Anemometer Thies S11100 07 157660 1533619 0.04606 0.2262 40 10° Anemometer Thies S11100 07 157659 1533620 0.04591 0.2549 20 10° Wind Vane Thies S21110H 05 150054 1521945 1.00030 2.4564 62 10° Wind Vane Thies S21110H 05 150053 1521946 1.00077 2.1104 58 10° Temperature Galltec TPC1.S/6-ME 154305 A66353A030 100 -30 61 - Sensor Temperature & 100 (T) -30 (T) Humidity Galltec KPC1.S/6-ME 154287 A65630A110 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B15-0235 E 00/03/01 100 600 4 - Site 8: Sanghar, Sindh CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 07 157670 1513906 0.04604 0.2504 80 234® Anemometer Vector S14100 16796 1514323 0.09783 0.1206 80 54® Anemometer Thies S11100 07 157669 1513907 0.04598 0.2479 60 234® Anemometer Thies S11100 07 157668 1513908 0.04607 0.2413 40 234® Anemometer Thies S11100 07 157667 1513909 0.04599 0.2544 20 234® Wind Vane Thies S21110H 04 150020 1522274 1.00048 -0.5266 78.5 234® Wind Vane Thies S21110H 04 150019 1522275 1.00065 -0.0707 58.5 234® Temperature Galltec TPC1.S/6-ME 154296 A66353A030 100 -30 76 - Sensor Temperature & 100 (T) -30 (T) Humidity Galltec KPC1.S/6-ME 154282 A65630A110 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B14-0492 E 00/03/01 100 600 4 - Site 9: Umerkot, Sindh CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 07 157666 1513910 0.04599 0.2556 80 240° Anemometer Vector S14100 16798 1514325 0.09763 0.1450 80 60° Anemometer Thies S11100 07 157665 1513911 0.04594 0.2675 60 240° Anemometer Thies S11100 07 157664 1513913 0.04597 0.2631 40 240° Anemometer Thies S11100 07 157663 1533616 0.04598 0.2409 20 240° Wind Vane Thies S21110H 05 150056 1521943 1.00107 1.9539 78.5 240° Wind Vane Thies S21110H 05 150055 1521944 1.00043 2.1661 58.5 240° Temperature Galltec TPC1.S/6-ME 154297 A66353A030 100 -30 76 - Sensor Temperature & 100 (T) -30 (T) Humidity Galltec KPC1.S/6-ME 154291 A65630A110 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B14-0490 E 00/03/01 100 600 4 - 93 Site 10: Tando Ghulam Ali, Sindh CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 07 157650 1533630 0.04600 0.2263 80 193° Anemometer Vector S14100 16792 1514319 0.09761 0.1690 80 13° Anemometer Thies S11100 07 157649 1533631 0.04590 0.2530 60 193° Anemometer Thies S11100 07 157648 1533632 0.04597 0.2359 40 193° Anemometer Thies S11100 07 157647 1533633 0.04596 0.2388 20 193° Wind Vane Thies S21110H 05 150041 1521958 1.00249 0.8963 78.5 13° Wind Vane Thies S21110H 05 150040 1521959 1.00044 1.2569 58.5 13° Temperature Galltec TPC1.S/6-ME 154295 A66353A030 100 -30 76 - Sensor Temperature & 100 (T) -30 (T) Humidity Galltec KPC1.S/6-ME 154281 A65630A110 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B15-0236 E 00/03/01 100 600 4 - Site 11: Gwadar, Balochistan CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 07 157674 1513902 0.04602 0.2433 80 80° Anemometer Vector S14100 16800 1514327 0.09757 0.1461 80 260° Anemometer Thies S11100 07 157673 1513903 0.04603 0.2473 60 80° Anemometer Thies S11100 07 157672 1513904 0.04606 0.2449 40 800° Anemometer Thies S11100 07 157671 1513905 0.04596 0.2599 20 80° Wind Vane Thies S21110H 05 150048 1521951 1.00304 0.4691 78.5 80° Wind Vane Thies S21110H 05 150042 1521957 1.00058 1.6060 58.5 80° Temperature Galltec TPC1.S/6-ME 154300 A66353A030 100 -30 76 - Sensor Temperature & 100 (T) -30 (T) Humidity Galltec KPC1.S/6-ME 154288 A65630A110 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B14-0487 E 00/03/01 100 600 4 - Site 12: Sujawal, Sindh CALIBRATION HEIGHT BOOM Instrument MAKE MODEL SERIAL NO. SLOPE OFFSET CERTIFICATE [m] ORIENTATION Anemometer Thies S11100 07 157642 1533644 0.04600 0.2352 80 50° Anemometer Vector S14100 16790 1514317 0.09719 0.1637 80 230° Anemometer Thies S11100 07 157641 1533645 0.04602 0.2334 60 50° Anemometer Thies S11100 07 157640 1533646 0.04596 0.2379 40 50° Anemometer Thies S11100 07 157639 1533647 0.04600 0.2308 20 50° Wind Vane Thies S21110H 05 150039 1521960 1.00091 1.6293 78.5 50° Wind Vane Thies S21110H 05 150038 1521961 1.00061 0.9398 58.5 50° Temperature Galltec TPC1.S/6-ME 154294 A66353A030 100 -30 76 - Sensor Temperature & 100 (T) -30 (T) Humidity Galltec KPC1.S/6-ME 154280 A65630A110 5 - 100 (RH) 0 (RH) Sensor Barometer Ammonit AB 100 B14-0486 E 00/03/01 100 600 4 - 94 ANNEX C DATA RECOVERY RATES OVER THE SHORT-TERM Site 1 Peshawar 2 Haripur 3 Chakri Months Availability [%] Availability [%] Availability [%] Thies Vector Thies Vector Thies Vector Anemometer Anemometer Anemometer Anemometer Anemometer Anemometer @80m @80m @80m @80m @80m @80m 09/2016 100.0* 100.0* 87.4* 100.0* N/A N/A 10/2016 100.0 100.0 100.0 100.0 100.0 99.8 11/2016 100.0 100.0 100.0 100.0 100.0 99.6 12/2016 100.0 100.0 99.9 99.8 99.9 99.7 01/2017 100.0 100.0 100.0 96.5 100.0 99.7 02/2017 99.2 99.2 99.2 97.7 99.2 99.0 03/2017 100.0 100.0 100.0 98.7 100.0 99.8 04/2017 100.0 100.0 100.0 99.8 100.0 99.7 05/2017 100.0 100.0 100.0 99.7 100.0 99.7 06/2017 100.0 100.0 100.0 99.9 100.0 99.5 07/2017 100.0 100.0 100.0 99.6 100.0 99.9 08/2017 100.0 100.0 100.0 99.3 100.0 100.0 09/2017 100.0 100.0 100.0 100.0 100.0 99.9 10/2017 100.0 100.0 98.4 98.4 100.0 99.1 11/2017 78.3 99.9 99.8 98.8 99.7 95.0 12/2017 99.9 99.9 100.0 99.8 100.0 98.2 01/2018 100.0 100.0 100.0 99.6 100.0 97.5 02/2018 99.8 99.5 100.0 99.5 100.0 99.5 03/2018 100.0 100.0 100.0 99.8 100.0 98.6 04/2018 100.0 100.0 100.0 73.5 100.0 99.1 05/2018 100.0 100.0 0.0 99.6 100.0 99.1 06/2018 99.4 99.4 0.0 99.7 100.0 99.2 07/2018 100.0 100.0 0.0 99.7 100.0 99.7 08/2018 100.0 99.9 0.0 98.3 100.0 98.4 09/2018 100.0* 100.0* 0.0* 98.1* 99.9 95.7 Mean 99.0 99.9 82.5 98.2 99.9 99.0 *Not covering the full month 95 Site 4 Quaidabad 5 Bahawalpur 6 Sadiqabad Months Availability [%] Availability [%] Availability [%] Thies Vector Thies Vector Thies Vector Anemometer Anemometer Anemometer Anemometer Anemometer Anemometer @80m @80m @80m @80m @80m @80m 09/2016 100.0* 100.0* N/A N/A N/A N/A 10/2016 100.0 100.0 99.9 99.9 99.9 99.9 11/2016 100.0 100.0 100.0 100.0 99.9 99.9 12/2016 100.0 100.0 100.0 100.0 100.0 100.0 01/2017 100.0 100.0 100.0 100.0 100.0 100.0 02/2017 99.2 99.2 99.2 99.2 98.5 98.5 03/2017 100.0 100.0 100.0 100.0 99.9 99.9 04/2017 100.0 100.0 100.0 100.0 100.0 100.0 05/2017 100.0 100.0 100.0 100.0 100.0 100.0 06/2017 100.0 100.0 100.0 100.0 100.0 100.0 07/2017 100.0 100.0 100.0 100.0 100.0 100.0 08/2017 100.0 100.0 100.0 99.7 100.0 100.0 09/2017 100.0 99.8 97.0 96.9 99.9 99.9 10/2017 99.9 99.9 86.8 86.6 100.0 99.6 11/2017 100.0 100.0 99.9 98.4 99.8 99.6 12/2017 99.9 99.9 100.0 99.1 100.0 99.6 01/2018 100.0 99.9 100.0 100.0 100.0 99.9 02/2018 100.0 100.0 100.0 100.0 100.0 99.8 03/2018 100.0 100.0 100.0 100.0 63.4 100.0 04/2018 100.0 100.0 100.0 99.8 90.9 94.9 05/2018 100.0 100.0 100.0 99.8 93.7 93.7 06/2018 74.5 74.5 100.0 99.8 100.0 100.0 07/2018 100.0 100.0 100.0 99.9 100.0 100.0 08/2018 100.0 100.0 100.0 99.8 100.0 100.0 09/2018 100.0* 100.0* 100.0 99.3 100.0 100.0 Mean 99.0 99.0 99.3 99.1 97.7 99.4 *Not covering the full month 96 Site 7 Quetta 8 Sanghar 9 Umerkot Months Availability [%] Availability [%] Availability [%] Thies Vector Thies Vector Thies Vector Anemometer Anemometer Anemometer Anemometer Anemometer Anemometer @64m @64m @80m @80m @80m @80m 10/2016 100.0 100.0 N/A N/A N/A N/A 11/2016 100.0 100.0 99.9* 100.0* 100.0* 100.0* 12/2016 100.0 100.0 100.0 100.0 100.0 100.0 01/2017 100.0 100.0 100.0 100.0 100.0 100.0 02/2017 99.2 99.2 99.2 99.2 99.2 99.2 03/2017 100.0 100.0 100.0 100.0 100.0 100.0 04/2017 100.0 100.0 100.0 100.0 100.0 100.0 05/2017 100.0 100.0 100.0 100.0 100.0 100.0 06/2017 100.0 100.0 100.0 100.0 100.0 100.0 07/2017 100.0 100.0 100.0 100.0 100.0 100.0 08/2017 96.5 96.5 100.0 100.0 100.0 100.0 09/2017 100.0 100.0 100.0 100.0 100.0 100.0 10/2017 100.0 100.0 98.2 98.2 97.8 97.8 11/2017 100.0 100.0 99.8 99.8 100.0 100.0 12/2017 100.0 100.0 100.0 100.0 100.0 100.0 01/2018 100.0 100.0 100.0 100.0 100.0 100.0 02/2018 100.0 100.0 100.0 100.0 100.0 100.0 03/2018 100.0 100.0 100.0 100.0 100.0 100.0 04/2018 100.0 100.0 100.0 100.0 100.0 100.0 05/2018 100.0 100.0 100.0 100.0 92.2 92.2 06/2018 100.0 100.0 100.0 100.0 100.0 100.0 07/2018 100.0 100.0 100.0 100.0 100.0 100.0 08/2018 100.0 100.0 100.0 100.0 100.0 100.0 09/2018 100.0 100.0 100.0 99.9 100.0 100.0 10/2018 N/A N/A 100.0 100.0 100.0 100.0 11/2018 N/A N/A 100.0* 99.9* 100.0* 100.0* Mean 99.8 99.8 99.9 99.9 99.5 99.5 *Not covering the full month 97 Site 10 Tando 11 Gwadar 12 Sujawal Months Availability [%] Availability [%] Availability [%] Thies Vector Thies Vector Thies Vector Anemometer Anemometer Anemometer Anemometer Anemometer Anemometer @80m @80m @80m @80m @80m @80m 10/2016 100.0 100.0 N/A N/A 99.9 99.9 11/2016 100.0 100.0 100.0* 100.0* 99.8 100.0 12/2016 100.0 100.0 100.0 100.0 99.9 100.0 01/2017 100.0 100.0 99.9 100.0 99.9 100.0 02/2017 99.3 99.3 99.3 99.3 98.4 98.4 03/2017 100.0 100.0 99.9 100.0 100.0 99.9 04/2017 100.0 100.0 100.0 100.0 100.0 100.0 05/2017 100.0 100.0 100.0 100.0 100.0 100.0 06/2017 100.0 100.0 100.0 100.0 100.0 100.0 07/2017 100.0 100.0 100.0 100.0 100.0 100.0 08/2017 100.0 100.0 100.0 100.0 99.8 100.0 09/2017 100.0 100.0 100.0 100.0 100.0 100.0 10/2017 99.9 100.0 100.0 100.0 100.0 100.0 11/2017 100.0 100.0 99.9 99.9 100.0 100.0 12/2017 100.0 100.0 100.0 100.0 100.0 100.0 01/2018 100.0 99.8 100.0 100.0 97.3 97.3 02/2018 100.0 99.9 100.0 100.0 32.6 32.6 03/2018 100.0 100.0 100.0 100.0 73.2 73.2 04/2018 100.0 100.0 100.0 100.0 100.0 100.0 05/2018 100.0 100.0 89.7 89.7 100.0 100.0 06/2018 100.0 100.0 100.0 100.0 100.0 100.0 07/2018 100.0 100.0 100.0 100.0 100.0 100.0 08/2018 100.0 100.0 100.0 100.0 100.0 100.0 09/2018 100.0 100.0 100.0 100.0 100.0 100.0 10/2018 N/A N/A 100.0 100.0 N/A N/A 11/2018 N/A N/A 100.0* 100.0* N/A N/A Mean 100.0 100.0 99.5 99.5 96.1 96.1 *Not covering the full month 98 ANNEX D THE WIND REGIME OBSERVED OVER THE SHORT-TERM Site 1: Peshawar, Khyber Pakhtunkhwa Energy Rose Wind Frequency Rose Mean Wind Speed Rose Site 2: Haripur, Khyber Pakhtunkhwa Energy Rose Wind Frequency Rose 99 Mean Wind Speed Rose Site 3: Chakri, Punjab Energy Rose Wind Frequency Rose Mean Wind Speed Rose Site 4: Quaidabad, Punjab Energy Rose Wind Frequency Rose 100 Mean Wind Speed Rose Site 5: Bahawalpur, Punjab Energy Rose Wind Frequency Rose Mean Wind Speed Rose Site 6: Sadiqabad, Punjab Energy Rose Wind Frequency Rose 101 Mean Wind Speed Rose Site 7: Quetta, Balochistan Energy Rose Wind Frequency Rose Mean Wind Speed Rose 102 Site 8: Sanghar, Sindh Energy Rose Wind Frequency Rose Mean Wind Speed Rose Site 9: Umerkot, Sindh Energy Rose Wind Frequency Rose Mean Wind Speed Rose 103 Site 10: Tando Ghulam Ali, Sindh Energy Rose Wind Frequency Rose Mean Wind Speed Rose Site 11: Gwadar, Balochistan Energy Rose Wind Frequency Rose Mean Wind Speed Rose 104 Site 12: Sujawal, Sindh Energy Rose Wind Frequency Rose Mean Wind Speed Rose 105 ANNEX E THE WIND REGIME ESTIMATED OVER THE LONG-TERM Site 6: Sadiqabad, Punjab Energy Rose Wind Frequency Rose Mean Wind Speed Rose Site 8: Sanghar, Sindh Energy Rose Wind Frequency Rose 106 Mean Wind Speed Rose Site 9: Umerkot, Sindh Energy Rose Wind Frequency Rose Mean Wind Speed Rose Site 10: Tando Ghulam Ali, Sindh Energy Rose Wind Frequency Rose 107 Mean Wind Speed Rose Site 11: Gwadar, Balochistan Energy Rose Wind Frequency Rose Mean Wind Speed Rose Site 12: Sujawal, Sindh Energy Rose Wind Frequency Rose 108 Mean Wind Speed Rose 109 ANNEX F WEIBULL PARAMETERS OF THE SHORT-TERM WIND REGIME Site 1 Peshawar 2 Haripur 3 Chakri 4 Quaidabad 5 Bahawalpur 6 Sadiqabad Height AGL 80 80 80 80 80 80 [m] Selected period 15/09/2016 - 15/09/2016 - 01/10/2016 – 15/09/2016 - 01/10/2016 – 01/10/2016 – [-] 14/09/2018 14/09/2018 30/09/2018 14/09/2018 30/09/2018 30/09/2018 Arithmetic 3.09 3.61 3.28 4.20 5.18 4.98 mean wind speed [m/s] Weibull mean 3.08 3.69 3.30 4.20 5.21 4.94 wind speed [m/s] Weibull A 3.45 4.16 3.65 4.70 5.88 5.58 [m/s] Weibull k 1.636 1.923 1.500 1.627 2.199 2.057 [m/s] Prevailing wind WSW, SSW ENE ENE, NNE, E S, N N, SSW directions WNW [-] Wind directions WSW, SSW ENE WNW, NNE, E, WNW, NNW S, N N, SSW, W with most ENE energy content [-] Site 7 Quetta 8 Sanghar 9 Umerkot 10 Tando 11 Gwadar 12 Sujawal Height AGL 64 80 80 80 80 80 [m] Selected period 01/10/2016 – 11/11/2016 – 11/11/2016 – 01/10/2016 – 11/11/2016 – 01/10/2016 – [-] 30/09/2018 10/11/2018 10/11/2018 30/09/2018 10/11/2018 30/09/2018 Arithmetic 4.03 6.59 6.51 7.05 4.66 7.38 mean wind speed [m/s] Weibull mean 4.02 6.68 6.55 7.08 4.70 7.36 wind speed [m/s] Weibull A 4.51 7.54 7.38 7.97 5.31 8.25 [m/s] Weibull k 1.707 2.270 2.449 2.626 1.946 2.914 [m/s] Prevailing wind WNW, E SSW SSW, WSW WSW SSW WSW, W directions [-] Wind directions E, ESE, WNW, SSW SSW, WSW WSW SSW WSW, W with most NNW energy content [-] 110 ANNEX G WEIBULL PARAMETERS OF THE LONG-TERM WIND REGIME Site 6 Sadiqabad 8 Sanghar 9 Umerkot Height AGL 80 80 80 [m] Long-term data set ERA5_N28.2_E70.0 ERA5_N25.7_E69.0 ERA5_N25.2_E69. [-] 6 Long-term period 17 years 17 years 17 years [-] Arithmetic mean wind speed 5.05 6.62 6.46 [m/s] Weibull mean wind speed 5.05 6.72 6.53 [m/s] Weibull A 5.70 7.58 7.36 [m/s] Weibull k 2.123 2.308 2.454 [m/s] Prevailing wind directions N, SSW, S, NNW SSW SSW, WSW [-] Wind directions with most N, S, NNW, SSW SSW SSW, WSW energy content [-] Site 10 Tando 11 Gwadar 12 Sujawal Height AGL 80 80 80 [m] Long-term data set ERA5_N25.2_E68.7 ERA5_N25.2_E62.4 ERA5_N24.6_E68. [-] 1 Long-term period 16 years 18 years 18 years [-] Arithmetic mean wind speed 7.00 4.60 7.28 [m/s] Weibull mean wind speed 7.04 4.64 7.27 [m/s] Weibull A 7.93 5.24 8.16 [m/s] Weibull k 2.588 1.948 2.827 [m/s] Prevailing wind directions WSW SSW WSW, W [-] Wind directions with most WSW SSW WSW, W energy content [-] 111 ANNEX H COMPARISON OF PREDICTED WIND REGIME AND ESTIMATED WIND REGIME During the site selection phase, the “preliminary wind regime” was predicted at each site via the use of long-term datasets. After the measurement campaign, short-term wind regime is indicated as “modelled wind regime” at each site. Both preliminary and modelled wind regimes as well as the expected wind power density at each site are given in the tables below. Site 1 Peshawar 2 Haripur 3 Chakri 4 Quaidabad 5 Bahawalpur 6 Sadiqabad Preliminary wind regime Height AGL 80 80 80 80 80 80 [m] Expected mean 2.8 3.1 3.1 3.7 4.4 4.6 wind speed [m/s] Expected power 551 715 773 1,223 1,280 1,382 density [kWh/m²/year] Modelled wind regime Expected mean 3.1 3.7 3.3 4.2 5.2 5.1 wind speed [m/s] Expected power 376 483 509 960 1,301 1,242 density [kWh/m²/year] Site 7 Quetta 8 Sanghar 9 Umerkot 10 Tando 11 Gwadar 12 Sujawal Preliminary wind regime Height AGL 64 80 80 80 80 80 [m] Expected mean 3.2 6.0 6.2 6.6 4.7 6.9 wind speed [m/s] Expected power 702 2,611 2,953 3,121 1,296 3,373 density [kWh/m²/year] Modelled wind regime Expected mean 4.2 6.8 6.6 7.1 4.7 7.3 wind speed [m/s] Expected power 766 2,31 2,313 2,801 1,002 2,941 density [kWh/m²/year] 112 ANNEX I SHORT-TERM SEASONAL AND DIURNAL VARIATIONS IN WIND CHARACTERISTICS Hourly wind regime is calculated for each specific month based on the short-term data for each site. Thus, seasonal and diurnal variations in wind characteristics are given for each site as frequency in the table below. Site 1: Peshawar, Khyber Pakhtunkhwa HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 2.4% 2.4% 4.4% 3.5% 4.3% 4.1% 4.6% 4.0% 2.5% 3.2% 2.3% 1.7% 3.3% 01:00 2.1% 2.1% 4.6% 3.7% 4.0% 3.1% 4.3% 3.7% 2.8% 2.8% 1.8% 1.2% 3.0% 02:00 2.1% 2.0% 3.6% 2.5% 2.3% 2.6% 3.0% 2.4% 2.2% 3.0% 1.6% 1.2% 2.4% 03:00 2.4% 2.1% 2.7% 1.5% 1.8% 1.8% 2.4% 1.5% 1.4% 1.9% 1.2% 0.8% 1.8% 04:00 1.6% 1.7% 1.7% 1.1% 3.0% 1.2% 2.1% 1.2% 0.9% 1.5% 0.6% 0.6% 1.5% 05:00 1.3% 1.1% 1.7% 1.5% 3.6% 1.9% 1.8% 1.5% 0.9% 1.2% 0.3% 0.5% 1.5% 06:00 1.2% 1.5% 1.8% 2.2% 3.2% 3.0% 2.3% 1.6% 1.6% 1.4% 0.4% 0.5% 1.7% 07:00 1.0% 2.1% 2.3% 3.9% 4.5% 4.5% 3.7% 3.5% 3.7% 1.4% 0.5% 0.7% 2.7% 08:00 1.0% 2.8% 2.7% 4.3% 6.3% 7.1% 5.5% 6.9% 5.8% 2.0% 0.7% 0.7% 3.8% 09:00 0.8% 2.9% 3.4% 4.2% 6.2% 10.0% 7.8% 8.6% 6.2% 2.5% 1.8% 0.9% 4.6% 10:00 0.8% 3.0% 4.0% 5.1% 8.5% 13.5% 11.4% 7.8% 8.8% 2.9% 1.6% 0.7% 5.7% 11:00 0.8% 2.1% 5.4% 6.7% 13.2% 17.4% 11.5% 7.3% 9.4% 3.2% 1.2% 0.7% 6.6% 12:00 0.7% 2.0% 7.6% 8.6% 13.0% 16.5% 12.0% 9.1% 10.8% 3.0% 1.0% 0.7% 7.2% 13:00 0.9% 1.8% 5.9% 9.1% 6.7% 15.9% 12.3% 6.5% 8.2% 3.4% 0.8% 1.1% 6.1% 14:00 0.8% 2.5% 6.4% 8.9% 10.3% 14.2% 11.3% 5.0% 5.7% 2.6% 1.1% 0.9% 5.9% 15:00 1.4% 2.8% 6.6% 7.6% 7.0% 15.8% 7.6% 4.1% 6.6% 2.4% 1.4% 0.9% 5.4% 16:00 2.2% 3.0% 6.4% 8.4% 8.3% 14.6% 11.6% 3.6% 3.9% 2.2% 1.4% 1.1% 5.6% 17:00 2.0% 3.1% 6.2% 9.2% 10.5% 9.0% 7.7% 3.4% 3.5% 2.8% 1.6% 1.4% 5.1% 18:00 2.2% 3.4% 5.5% 5.5% 12.0% 5.0% 5.7% 2.1% 3.2% 4.0% 2.4% 1.2% 4.4% 19:00 2.1% 3.5% 3.8% 4.2% 10.8% 6.2% 7.1% 2.3% 5.8% 3.7% 3.2% 1.3% 4.5% 20:00 2.4% 3.5% 5.2% 4.0% 8.7% 7.2% 8.7% 3.0% 5.4% 3.7% 3.5% 1.5% 4.8% 21:00 2.3% 2.3% 5.1% 5.8% 5.1% 6.0% 11.0% 5.7% 4.0% 3.3% 3.2% 1.4% 4.6% 22:00 2.5% 2.3% 5.2% 4.6% 5.0% 5.4% 7.0% 5.1% 3.0% 3.6% 3.4% 1.6% 4.1% 23:00 2.7% 2.6% 4.5% 3.1% 6.0% 5.1% 6.5% 3.4% 2.8% 3.9% 2.7% 1.7% 3.8% TOTAL 3.3% 4.9% 8.9% 9.9% 13.7% 15.9% 14.1% 8.6% 9.1% 5.5% 3.3% 2.1% 113 Figure 60: Mean annual power density for short-term measured data between 15/09/2016-14/09/2018 114 Site 2: Haripur, Khyber Pakhtunkhwa HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 2.9% 3.9% 5.1% 9.0% 7.3% 5.3% 4.5% 6.9% 5.7% 9.7% 4.0% 3.2% 5.6% 01:00 3.1% 3.4% 5.4% 7.4% 4.8% 4.3% 3.8% 4.6% 4.7% 7.9% 3.5% 3.1% 4.7% 02:00 2.5% 3.4% 4.9% 5.3% 1.8% 2.2% 2.7% 3.1% 3.8% 6.2% 3.5% 2.6% 3.5% 03:00 1.7% 2.3% 2.1% 2.2% 0.9% 1.9% 2.0% 2.2% 1.2% 3.2% 1.7% 2.0% 1.9% 04:00 0.9% 0.8% 0.6% 3.5% 2.7% 2.8% 1.3% 1.4% 0.4% 2.0% 0.6% 0.6% 1.5% 05:00 0.8% 0.9% 1.1% 1.8% 2.5% 3.3% 1.0% 1.1% 0.8% 1.6% 0.5% 0.4% 1.3% 06:00 0.8% 1.7% 1.7% 1.9% 4.1% 3.6% 1.4% 1.3% 1.2% 1.3% 0.7% 0.6% 1.7% 07:00 1.2% 1.8% 1.9% 2.2% 4.4% 4.3% 2.0% 1.7% 2.3% 1.6% 1.1% 0.9% 2.1% 08:00 1.4% 1.9% 1.8% 1.8% 4.2% 4.4% 2.5% 1.9% 2.4% 2.3% 1.0% 1.0% 2.2% 09:00 1.1% 2.4% 1.8% 3.9% 4.6% 5.1% 3.6% 1.8% 1.8% 2.3% 1.2% 0.9% 2.5% 10:00 0.9% 2.2% 2.6% 4.7% 5.2% 9.1% 5.8% 3.5% 1.4% 1.9% 1.3% 0.9% 3.3% 11:00 0.9% 1.7% 3.0% 4.5% 5.1% 6.7% 5.1% 2.3% 1.1% 1.5% 0.6% 0.9% 2.8% 12:00 1.0% 2.4% 3.2% 4.1% 6.7% 9.0% 3.1% 2.0% 1.2% 0.7% 0.6% 0.6% 2.9% 13:00 1.4% 2.5% 1.9% 4.0% 8.8% 10.3% 3.1% 1.1% 2.3% 0.9% 2.2% 1.7% 3.4% 14:00 3.5% 2.7% 2.2% 5.5% 4.6% 7.8% 2.9% 1.5% 2.3% 6.2% 5.1% 3.5% 4.0% 15:00 3.6% 3.7% 5.3% 8.7% 6.3% 5.3% 2.5% 1.6% 6.2% 9.4% 6.3% 4.6% 5.2% 16:00 3.7% 4.6% 6.6% 8.5% 8.3% 5.5% 2.8% 3.8% 7.4% 11.0% 7.2% 5.1% 6.2% 17:00 3.9% 6.1% 6.6% 10.7% 9.1% 6.4% 3.6% 4.4% 8.5% 10.6% 6.4% 5.0% 6.7% 18:00 4.5% 6.6% 6.9% 12.5% 8.6% 7.8% 3.0% 5.6% 6.9% 11.4% 6.2% 4.8% 7.0% 19:00 5.1% 6.6% 7.9% 8.8% 8.3% 7.4% 2.9% 4.8% 6.6% 10.2% 5.5% 4.8% 6.5% 20:00 5.6% 5.7% 7.1% 8.9% 7.6% 5.9% 4.2% 4.4% 7.6% 10.1% 5.9% 4.8% 6.4% 21:00 4.8% 4.9% 6.3% 9.2% 10.5% 8.3% 4.6% 4.7% 6.3% 9.9% 5.4% 4.0% 6.5% 22:00 4.2% 3.8% 7.6% 9.5% 8.0% 5.8% 3.5% 5.3% 6.3% 9.0% 5.5% 3.7% 6.0% 23:00 3.7% 4.2% 7.0% 8.5% 8.2% 6.6% 3.5% 5.2% 6.8% 8.9% 5.5% 4.0% 6.0% TOTAL 5.3% 6.7% 8.4% 12.3% 11.9% 11.6% 6.3% 6.3% 7.9% 11.7% 6.8% 5.3% 115 Figure 61: Mean annual power density for short-term measured data between 15/09/2016-14/09/2018 116 Site 3: Chakri, Punjab HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 1.4% 2.1% 3.5% 5.4% 5.4% 4.1% 6.3% 5.4% 4.2% 1.6% 1.0% 1.3% 3.5% 01:00 1.0% 2.0% 4.1% 4.1% 3.8% 3.2% 4.6% 3.8% 3.6% 1.6% 0.8% 1.0% 2.8% 02:00 1.2% 3.7% 2.5% 4.3% 3.5% 2.7% 5.2% 4.6% 1.6% 1.8% 1.1% 1.6% 2.8% 03:00 1.1% 3.4% 2.3% 1.7% 4.8% 2.4% 2.7% 1.9% 0.6% 1.4% 0.9% 1.6% 2.1% 04:00 1.5% 4.4% 2.4% 3.6% 6.4% 2.5% 2.2% 1.2% 0.5% 1.3% 0.5% 1.5% 2.3% 05:00 1.8% 6.5% 5.6% 5.4% 7.5% 3.0% 2.3% 1.4% 0.7% 0.7% 1.1% 2.6% 3.2% 06:00 2.9% 8.6% 6.4% 5.7% 6.4% 7.1% 2.5% 1.5% 0.7% 0.8% 2.0% 3.8% 4.0% 07:00 3.8% 8.5% 6.9% 5.5% 5.2% 7.3% 3.6% 1.4% 1.1% 1.0% 3.1% 5.3% 4.4% 08:00 4.0% 7.7% 6.1% 6.0% 5.4% 8.2% 4.7% 2.2% 0.9% 1.2% 2.9% 5.5% 4.5% 09:00 3.9% 7.0% 5.9% 5.4% 4.8% 5.9% 5.1% 2.4% 2.1% 1.3% 2.6% 4.8% 4.3% 10:00 3.1% 5.7% 6.1% 5.2% 5.4% 5.3% 4.7% 3.1% 2.6% 1.5% 2.1% 3.5% 4.0% 11:00 2.9% 5.3% 4.6% 6.5% 6.9% 12.1% 4.3% 4.1% 4.1% 1.3% 2.1% 3.2% 4.8% 12:00 3.0% 5.0% 5.0% 6.8% 7.7% 11.1% 5.9% 3.9% 4.3% 2.7% 2.8% 2.6% 5.1% 13:00 3.1% 5.1% 7.1% 8.1% 12.2% 11.2% 5.3% 2.8% 3.6% 3.6% 2.9% 3.0% 5.7% 14:00 2.6% 4.3% 10.5% 7.5% 13.8% 12.5% 5.7% 3.5% 5.8% 2.7% 2.8% 3.5% 6.3% 15:00 2.6% 4.6% 10.6% 8.9% 12.7% 8.3% 6.5% 4.6% 6.0% 1.6% 2.7% 3.2% 6.0% 16:00 2.3% 3.8% 7.8% 10.5% 13.6% 5.7% 5.4% 4.1% 7.0% 1.2% 2.5% 2.5% 5.5% 17:00 2.2% 3.3% 6.0% 7.6% 10.3% 4.1% 3.6% 2.8% 3.2% 1.7% 2.0% 2.0% 4.1% 18:00 2.1% 3.6% 2.3% 9.2% 12.2% 3.9% 3.4% 2.7% 2.4% 1.8% 1.8% 2.1% 4.0% 19:00 2.1% 3.1% 3.5% 7.5% 13.1% 4.7% 3.3% 3.1% 3.4% 1.8% 1.9% 2.0% 4.1% 20:00 2.0% 2.9% 2.8% 8.5% 10.4% 3.7% 3.7% 2.9% 3.5% 1.6% 1.6% 2.3% 3.8% 21:00 1.5% 3.0% 4.1% 10.8% 8.0% 4.1% 3.7% 3.4% 6.0% 1.5% 1.4% 1.7% 4.1% 22:00 1.5% 1.9% 7.4% 6.3% 6.2% 4.2% 10.7% 6.2% 3.6% 1.5% 1.1% 1.8% 4.4% 23:00 1.7% 1.9% 5.1% 5.9% 5.9% 8.7% 7.5% 6.4% 3.6% 1.3% 1.1% 1.8% 4.3% TOTAL 4.6% 9.0% 10.7% 13.0% 15.9% 12.2% 9.4% 6.6% 6.3% 3.2% 3.7% 5.3% 117 Figure 62: Mean annual power density for short-term measured data between 01/10/2016-30/09/2018 118 Site 4: Quaidabad, Punjab HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 3.3% 4.0% 4.1% 7.8% 8.3% 7.3% 7.8% 6.8% 5.8% 2.6% 4.2% 4.5% 5.5% 01:00 3.8% 3.9% 3.9% 5.5% 6.2% 5.2% 7.7% 5.2% 4.6% 2.7% 4.7% 5.6% 4.9% 02:00 4.0% 4.1% 4.4% 4.0% 4.7% 4.3% 6.5% 3.7% 3.5% 3.1% 5.6% 5.5% 4.5% 03:00 3.7% 4.5% 3.4% 3.7% 4.0% 5.1% 6.2% 3.2% 1.8% 2.1% 4.9% 3.9% 3.9% 04:00 2.7% 3.4% 2.3% 2.1% 3.8% 3.7% 5.0% 2.6% 1.2% 0.9% 2.3% 2.1% 2.7% 05:00 1.5% 1.9% 1.7% 1.4% 2.8% 4.4% 5.2% 2.9% 0.9% 0.8% 1.2% 1.4% 2.2% 06:00 1.2% 1.9% 1.8% 1.6% 2.5% 3.5% 5.6% 3.4% 0.9% 0.8% 1.3% 1.1% 2.1% 07:00 1.3% 2.5% 1.6% 2.0% 2.5% 3.7% 5.4% 3.2% 0.9% 0.9% 1.5% 1.3% 2.2% 08:00 1.5% 2.2% 1.5% 2.1% 2.1% 3.6% 4.9% 5.5% 0.9% 1.0% 1.6% 1.3% 2.3% 09:00 1.5% 2.2% 1.7% 1.8% 2.3% 4.8% 5.1% 3.7% 1.2% 0.9% 1.5% 1.4% 2.3% 10:00 1.3% 2.4% 1.9% 1.8% 2.9% 6.8% 5.1% 4.7% 3.8% 1.0% 1.3% 1.3% 2.8% 11:00 1.6% 3.2% 2.1% 3.7% 3.3% 8.0% 8.6% 5.4% 2.4% 1.6% 1.6% 1.5% 3.5% 12:00 2.0% 3.2% 3.1% 5.6% 5.1% 10.4% 6.3% 8.5% 2.7% 2.6% 2.2% 1.8% 4.4% 13:00 2.1% 3.1% 5.0% 7.1% 7.2% 7.8% 6.3% 8.4% 5.1% 3.0% 2.2% 1.6% 4.9% 14:00 1.8% 4.1% 5.4% 8.1% 11.0% 9.1% 7.8% 9.0% 4.0% 2.6% 2.3% 1.8% 5.5% 15:00 2.0% 3.8% 5.1% 9.1% 9.1% 8.5% 9.1% 8.0% 3.8% 2.0% 2.0% 2.0% 5.3% 16:00 2.7% 2.8% 5.1% 7.2% 9.4% 7.0% 9.9% 4.9% 3.8% 1.6% 1.6% 1.8% 4.8% 17:00 3.0% 3.3% 4.0% 8.3% 13.8% 5.2% 7.3% 3.5% 3.7% 1.2% 1.5% 2.0% 4.7% 18:00 3.1% 3.2% 3.8% 8.5% 15.5% 4.9% 7.3% 3.9% 3.8% 1.0% 1.6% 2.4% 4.9% 19:00 3.0% 3.5% 3.6% 8.1% 13.1% 5.2% 7.3% 4.3% 3.8% 0.9% 1.8% 3.8% 4.9% 20:00 3.1% 3.7% 4.6% 7.7% 11.9% 5.8% 7.9% 3.9% 3.9% 1.3% 2.0% 4.3% 5.0% 21:00 3.3% 3.9% 5.4% 6.6% 11.3% 5.8% 9.9% 4.6% 5.2% 1.7% 3.2% 4.1% 5.4% 22:00 3.6% 3.0% 4.7% 6.6% 10.5% 6.2% 9.6% 6.0% 4.9% 1.7% 4.1% 4.8% 5.5% 23:00 3.5% 3.2% 4.8% 7.0% 9.6% 8.0% 8.2% 7.1% 4.8% 1.8% 4.3% 4.5% 5.6% TOTAL 5.0% 6.4% 7.1% 10.6% 14.4% 12.0% 14.2% 10.2% 6.4% 3.3% 5.0% 5.5% 119 Figure 63: Mean annual power density for short-term measured data between 15/09/2016-14/09/2018 120 Site 5: Bahawalpur, Punjab HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 3.7% 5.0% 6.6% 8.8% 7.2% 8.5% 4.9% 6.8% 5.5% 5.0% 2.0% 3.2% 5.6% 01:00 3.5% 5.1% 6.1% 8.4% 5.9% 7.8% 4.7% 6.5% 5.1% 5.2% 2.2% 3.2% 5.3% 02:00 3.4% 4.2% 6.0% 4.5% 4.0% 7.3% 5.2% 7.0% 4.3% 4.9% 2.1% 3.2% 4.7% 03:00 2.9% 3.1% 2.6% 2.4% 3.6% 8.2% 5.3% 7.3% 3.6% 2.1% 1.6% 2.6% 3.8% 04:00 1.4% 1.4% 2.0% 2.2% 3.3% 7.3% 4.8% 5.3% 3.1% 1.9% 0.6% 1.1% 2.9% 05:00 0.8% 1.4% 2.0% 2.4% 3.2% 5.4% 4.2% 4.6% 3.0% 2.6% 0.6% 0.7% 2.6% 06:00 0.9% 1.6% 1.9% 2.6% 3.6% 4.9% 3.7% 4.3% 3.2% 3.2% 0.9% 0.7% 2.6% 07:00 1.2% 1.7% 2.0% 2.7% 3.4% 5.1% 3.3% 4.5% 3.6% 3.1% 1.2% 1.0% 2.7% 08:00 1.3% 2.0% 1.9% 2.5% 3.2% 4.9% 3.4% 5.0% 3.5% 2.7% 1.2% 1.1% 2.7% 09:00 1.4% 2.7% 2.0% 2.4% 3.5% 4.6% 3.4% 4.9% 3.3% 2.4% 1.2% 1.0% 2.7% 10:00 1.5% 2.9% 2.4% 2.6% 4.6% 3.9% 3.5% 4.8% 3.2% 2.1% 1.2% 1.1% 2.8% 11:00 1.5% 2.6% 2.7% 2.8% 4.5% 4.1% 4.0% 3.9% 2.6% 1.7% 1.0% 1.1% 2.7% 12:00 1.6% 2.5% 2.8% 2.6% 4.8% 3.9% 4.4% 3.4% 1.8% 1.3% 1.3% 1.4% 2.7% 13:00 2.5% 3.8% 3.6% 2.8% 3.6% 6.1% 4.0% 2.5% 1.9% 1.8% 1.8% 1.8% 3.0% 14:00 3.2% 5.3% 5.5% 4.6% 4.8% 5.3% 3.1% 2.3% 2.2% 1.9% 2.2% 2.1% 3.5% 15:00 3.7% 6.2% 7.2% 8.0% 7.3% 5.9% 3.5% 2.9% 2.6% 2.0% 2.3% 2.3% 4.5% 16:00 4.0% 6.4% 8.7% 9.0% 8.2% 6.1% 4.5% 3.8% 3.6% 2.2% 2.5% 2.3% 5.1% 17:00 3.8% 5.8% 8.9% 9.1% 7.5% 8.3% 5.3% 4.6% 4.1% 2.5% 2.5% 2.5% 5.4% 18:00 4.0% 5.9% 9.0% 9.7% 7.6% 8.2% 4.9% 5.8% 4.7% 2.8% 2.3% 2.4% 5.6% 19:00 4.5% 5.8% 8.5% 11.2% 7.5% 8.9% 5.4% 6.3% 4.5% 3.0% 2.1% 2.5% 5.9% 20:00 4.5% 5.5% 8.7% 11.6% 7.2% 9.2% 5.1% 6.4% 4.8% 3.2% 1.9% 2.4% 5.9% 21:00 4.1% 5.7% 7.8% 10.6% 8.5% 7.8% 5.2% 6.4% 4.7% 3.5% 1.8% 2.8% 5.7% 22:00 4.2% 6.0% 7.5% 9.8% 7.9% 7.3% 6.2% 6.6% 5.2% 3.9% 1.8% 3.0% 5.8% 23:00 4.2% 5.7% 7.4% 8.9% 7.1% 8.0% 5.2% 6.7% 5.4% 4.4% 2.1% 3.1% 5.7% TOTAL 5.7% 8.2% 10.3% 11.9% 11.0% 13.1% 8.9% 10.2% 7.5% 5.8% 3.4% 4.0% 121 Figure 64: Mean annual power density for short-term measured data between 01/10/2016-30/09/2018 122 Site 6: Sadiqabad, Punjab HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 4.4% 6.4% 6.1% 6.6% 5.5% 5.9% 4.7% 4.2% 3.7% 3.1% 2.8% 3.3% 4.7% 01:00 4.4% 6.2% 5.8% 7.2% 5.4% 5.1% 3.9% 3.8% 3.2% 3.5% 2.8% 3.4% 4.5% 02:00 4.2% 6.4% 5.0% 5.5% 4.1% 5.3% 3.9% 4.6% 2.5% 3.0% 2.4% 2.9% 4.1% 03:00 3.1% 4.5% 3.4% 3.6% 3.1% 4.8% 3.1% 4.1% 2.0% 1.5% 2.0% 2.3% 3.1% 04:00 1.9% 2.6% 2.5% 3.2% 3.2% 4.3% 2.5% 3.2% 1.6% 0.8% 0.8% 1.4% 2.3% 05:00 1.1% 2.2% 2.4% 3.3% 3.3% 3.9% 2.4% 2.7% 1.3% 0.9% 0.7% 1.1% 2.1% 06:00 1.3% 2.3% 2.5% 2.7% 3.2% 4.3% 2.2% 2.7% 1.4% 1.2% 0.7% 1.0% 2.1% 07:00 1.3% 2.1% 2.2% 2.2% 2.8% 4.1% 2.0% 2.6% 1.5% 1.4% 0.8% 1.0% 2.0% 08:00 1.1% 2.3% 1.8% 1.9% 2.4% 4.2% 2.0% 2.5% 1.5% 1.4% 0.9% 1.1% 1.9% 09:00 1.2% 2.3% 1.8% 2.0% 2.2% 4.4% 2.2% 2.9% 1.4% 1.3% 0.9% 1.1% 2.0% 10:00 1.2% 2.4% 1.9% 1.9% 2.5% 4.5% 2.4% 2.9% 1.3% 1.3% 0.7% 1.0% 2.0% 11:00 1.2% 3.3% 2.1% 2.3% 2.9% 5.1% 3.0% 3.5% 1.4% 1.4% 0.8% 1.0% 2.3% 12:00 1.4% 3.4% 2.1% 2.9% 3.8% 8.0% 3.5% 3.5% 1.5% 1.6% 1.1% 1.6% 2.8% 13:00 2.2% 4.2% 2.7% 3.2% 5.3% 8.0% 3.9% 3.9% 1.7% 2.2% 1.7% 2.4% 3.5% 14:00 2.9% 5.0% 3.5% 5.2% 8.4% 8.4% 4.5% 5.3% 1.9% 2.3% 2.3% 2.8% 4.4% 15:00 4.2% 6.2% 4.3% 6.4% 7.1% 13.7% 6.9% 7.2% 2.7% 2.3% 2.6% 3.0% 5.6% 16:00 5.1% 6.9% 5.1% 7.4% 8.0% 13.9% 7.4% 10.2% 3.4% 2.5% 3.3% 3.1% 6.4% 17:00 4.9% 7.4% 5.8% 9.6% 8.5% 13.4% 8.9% 10.6% 4.9% 2.7% 3.5% 3.2% 6.9% 18:00 4.7% 7.5% 7.2% 10.1% 9.9% 11.8% 8.7% 9.4% 5.6% 3.1% 3.7% 4.0% 7.1% 19:00 4.8% 8.3% 7.9% 9.0% 9.5% 11.2% 9.5% 8.2% 5.7% 3.7% 3.7% 3.8% 7.1% 20:00 5.2% 8.9% 7.1% 7.6% 7.3% 10.5% 7.5% 7.5% 5.6% 3.9% 3.6% 3.7% 6.5% 21:00 4.9% 8.5% 6.3% 7.5% 7.2% 9.1% 6.5% 6.8% 5.4% 4.0% 3.1% 3.3% 6.0% 22:00 4.7% 8.4% 6.1% 8.3% 6.6% 8.2% 5.2% 5.8% 4.6% 3.6% 3.2% 3.2% 5.6% 23:00 4.7% 7.6% 5.8% 7.4% 6.1% 6.7% 4.5% 4.7% 3.7% 3.2% 3.0% 3.2% 5.0% TOTAL 6.3% 10.4% 8.4% 10.6% 10.7% 14.9% 9.3% 10.2% 5.8% 4.6% 4.3% 4.8% 123 Figure 65: Mean annual power density for short-term measured data between 01/10/2016-30/09/2018 124 Site 7: Quetta, Balochistan HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 2.5% 3.7% 2.7% 2.0% 1.5% 3.4% 3.3% 1.6% 1.3% 0.9% 0.6% 0.8% 2.0% 01:00 3.1% 3.7% 2.5% 2.2% 1.4% 2.8% 2.2% 0.9% 1.0% 0.5% 0.6% 1.2% 1.8% 02:00 2.7% 3.4% 1.8% 2.0% 1.3% 1.9% 1.5% 0.7% 0.6% 0.6% 0.5% 0.7% 1.5% 03:00 2.6% 4.5% 2.2% 1.8% 2.3% 1.5% 1.1% 0.7% 0.3% 0.3% 0.4% 0.6% 1.5% 04:00 2.5% 4.0% 1.9% 2.3% 2.5% 2.1% 1.4% 1.1% 0.4% 0.4% 0.4% 0.7% 1.6% 05:00 2.5% 5.1% 2.7% 2.8% 3.6% 3.3% 1.7% 1.3% 1.0% 0.8% 0.5% 0.7% 2.1% 06:00 3.3% 5.1% 4.0% 3.5% 5.2% 4.9% 2.1% 2.0% 1.6% 1.5% 1.1% 0.8% 2.9% 07:00 3.7% 7.4% 6.9% 6.4% 8.3% 7.9% 2.7% 2.5% 2.3% 3.0% 1.9% 1.5% 4.5% 08:00 4.5% 8.6% 9.6% 8.0% 9.0% 10.1% 3.4% 2.9% 2.8% 4.8% 2.6% 2.4% 5.7% 09:00 5.5% 9.8% 10.4% 8.7% 10.1% 13.2% 4.2% 3.2% 3.4% 5.7% 3.2% 2.6% 6.7% 10:00 6.5% 9.0% 11.3% 10.1% 11.2% 12.8% 5.7% 4.4% 4.1% 6.8% 3.8% 3.0% 7.4% 11:00 6.1% 9.2% 9.6% 10.4% 11.6% 13.9% 6.4% 5.8% 6.4% 7.8% 3.8% 3.4% 7.9% 12:00 5.7% 9.0% 10.1% 9.8% 13.2% 13.6% 7.5% 7.7% 8.3% 7.5% 3.2% 3.1% 8.2% 13:00 5.7% 7.8% 8.2% 8.0% 13.0% 14.7% 8.4% 8.3% 6.8% 5.0% 2.5% 2.5% 7.6% 14:00 4.4% 6.8% 5.6% 5.9% 9.7% 10.0% 9.1% 6.4% 3.0% 2.6% 1.8% 2.1% 5.6% 15:00 4.7% 5.6% 3.9% 4.3% 5.9% 7.1% 9.3% 5.6% 1.9% 2.0% 1.6% 1.7% 4.5% 16:00 3.7% 5.4% 4.4% 4.1% 3.8% 5.4% 10.6% 5.7% 1.4% 1.6% 1.1% 1.5% 4.1% 17:00 3.9% 5.3% 3.8% 4.6% 3.7% 6.5% 12.6% 5.4% 1.5% 1.7% 1.5% 1.5% 4.3% 18:00 3.2% 6.1% 4.7% 3.5% 3.0% 7.0% 11.9% 5.5% 1.9% 1.9% 1.2% 1.5% 4.3% 19:00 3.2% 6.1% 4.1% 2.5% 2.7% 6.5% 10.4% 4.9% 2.1% 1.7% 1.2% 1.4% 3.9% 20:00 3.6% 5.3% 3.8% 2.7% 2.1% 5.7% 9.8% 4.0% 2.0% 1.5% 1.2% 1.3% 3.6% 21:00 3.1% 5.7% 3.4% 2.6% 2.0% 4.9% 8.7% 3.3% 1.9% 1.3% 0.9% 1.3% 3.3% 22:00 2.9% 5.3% 2.9% 2.2% 1.9% 4.8% 6.3% 3.1% 1.7% 1.2% 0.7% 1.1% 2.8% 23:00 2.5% 4.6% 2.5% 2.2% 1.8% 4.1% 4.7% 2.1% 1.6% 1.0% 0.5% 1.0% 2.4% TOTAL 7.7% 12.2% 10.3% 9.4% 10.9% 14.0% 12.1% 7.4% 4.9% 5.2% 3.0% 3.2% 125 Figure 66: Mean annual power density for short-term measured data between 01/10/2016-30/09/2018 126 Site 8: Sanghar, Sindh HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 3.0% 4.5% 2.9% 4.9% 4.0% 7.4% 4.7% 4.8% 2.6% 2.8% 2.2% 2.0% 3.8% 01:00 2.5% 4.0% 2.6% 5.0% 4.0% 7.7% 4.7% 4.6% 2.2% 2.7% 2.1% 1.7% 3.6% 02:00 2.6% 3.6% 2.1% 3.9% 4.1% 8.3% 5.0% 4.4% 1.9% 2.4% 1.8% 1.8% 3.5% 03:00 1.9% 2.6% 1.3% 3.4% 4.1% 9.8% 5.6% 5.2% 2.2% 1.2% 1.2% 1.6% 3.3% 04:00 1.0% 1.4% 0.7% 3.1% 4.1% 9.6% 5.7% 5.3% 2.1% 0.6% 0.4% 0.7% 2.9% 05:00 1.0% 1.4% 0.6% 3.2% 3.8% 9.2% 5.6% 5.2% 1.8% 0.7% 0.3% 0.3% 2.8% 06:00 1.0% 1.4% 1.0% 3.4% 3.8% 9.0% 5.4% 4.6% 1.8% 0.9% 0.5% 0.4% 2.8% 07:00 1.3% 2.1% 1.1% 3.4% 3.8% 8.5% 5.7% 4.6% 1.7% 1.0% 0.7% 0.5% 2.9% 08:00 1.5% 2.2% 1.3% 3.6% 3.9% 8.6% 5.9% 4.9% 1.7% 1.0% 0.7% 0.6% 3.0% 09:00 1.5% 1.7% 1.4% 4.1% 3.9% 8.9% 6.7% 5.3% 1.7% 0.9% 0.7% 0.7% 3.1% 10:00 1.4% 1.4% 1.5% 5.0% 4.0% 9.6% 7.4% 6.2% 1.8% 0.8% 0.7% 0.5% 3.4% 11:00 1.3% 1.1% 1.5% 5.3% 4.2% 10.1% 8.6% 7.7% 1.9% 0.7% 0.6% 0.5% 3.7% 12:00 1.5% 0.8% 1.8% 5.3% 5.3% 10.2% 9.6% 8.2% 2.2% 1.0% 0.9% 1.0% 4.0% 13:00 2.1% 1.6% 2.4% 5.7% 6.1% 9.9% 9.7% 8.6% 2.8% 1.9% 1.6% 2.2% 4.6% 14:00 3.4% 3.6% 3.8% 7.3% 6.3% 9.8% 8.6% 8.6% 3.4% 3.1% 2.4% 3.3% 5.3% 15:00 4.0% 4.8% 4.6% 8.4% 6.9% 9.9% 8.3% 6.9% 4.3% 5.0% 3.1% 3.7% 5.8% 16:00 4.2% 5.3% 5.0% 8.2% 7.2% 10.5% 9.4% 7.1% 4.7% 5.9% 3.7% 3.6% 6.2% 17:00 4.0% 4.4% 5.0% 7.4% 7.2% 10.0% 7.3% 6.5% 4.7% 5.8% 3.9% 3.7% 5.8% 18:00 4.5% 4.5% 4.0% 7.4% 6.3% 9.7% 6.8% 5.9% 4.3% 6.4% 4.1% 3.9% 5.7% 19:00 4.1% 4.3% 3.6% 7.8% 5.4% 9.1% 6.0% 5.3% 3.9% 6.6% 4.1% 4.0% 5.3% 20:00 4.2% 4.4% 3.5% 7.4% 4.9% 8.3% 5.8% 5.4% 3.4% 6.5% 3.8% 3.2% 5.1% 21:00 3.7% 4.8% 3.7% 6.9% 4.9% 7.6% 5.6% 5.0% 3.1% 5.5% 3.3% 3.0% 4.8% 22:00 3.3% 4.4% 3.8% 6.4% 4.8% 7.5% 5.5% 4.6% 2.8% 4.3% 3.0% 2.7% 4.4% 23:00 3.3% 4.1% 3.2% 5.7% 4.4% 7.1% 5.0% 4.6% 2.8% 3.7% 2.5% 2.4% 4.1% TOTAL 5.2% 6.2% 5.2% 11.0% 9.8% 18.0% 13.2% 11.6% 5.5% 6.0% 4.0% 4.0% 127 Figure 67: Mean annual power density for short-term measured data between 11/11/2016-10/11/2018 128 Site 9: Umerkot, Sindh HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 3.2% 2.6% 2.9% 5.0% 5.3% 4.0% 4.2% 3.4% 1.9% 1.9% 1.5% 1.0% 3.1% 01:00 2.7% 2.3% 2.7% 4.6% 5.3% 4.6% 4.2% 3.5% 2.1% 1.6% 1.3% 0.9% 3.0% 02:00 2.5% 2.1% 2.0% 4.3% 6.4% 6.8% 4.9% 3.6% 1.7% 1.2% 1.1% 1.0% 3.2% 03:00 1.8% 1.7% 1.3% 4.6% 6.7% 8.3% 6.8% 4.6% 2.4% 0.8% 0.7% 0.8% 3.4% 04:00 0.7% 0.8% 0.9% 4.3% 6.8% 8.2% 8.1% 5.7% 2.8% 0.6% 0.4% 0.4% 3.3% 05:00 0.7% 0.8% 1.2% 4.6% 6.4% 7.6% 8.5% 6.3% 2.6% 0.6% 0.3% 0.3% 3.3% 06:00 0.8% 1.0% 1.4% 5.1% 6.6% 7.7% 8.4% 6.7% 2.4% 0.8% 0.5% 0.3% 3.5% 07:00 1.2% 1.3% 1.5% 5.7% 6.7% 8.2% 8.0% 6.1% 2.5% 1.0% 0.7% 0.5% 3.6% 08:00 1.6% 1.6% 1.7% 6.3% 7.0% 8.6% 7.8% 6.4% 2.4% 1.1% 0.8% 0.7% 3.9% 09:00 1.8% 2.1% 2.2% 6.3% 7.2% 8.4% 7.9% 6.4% 2.4% 1.2% 1.0% 0.8% 4.0% 10:00 1.8% 2.3% 2.5% 6.9% 7.5% 8.7% 8.7% 7.1% 2.5% 1.4% 1.1% 1.0% 4.3% 11:00 1.5% 1.9% 2.6% 7.8% 8.5% 9.6% 8.8% 6.6% 2.6% 1.8% 1.3% 1.2% 4.6% 12:00 1.9% 1.7% 2.6% 8.4% 9.5% 10.6% 9.3% 6.4% 2.7% 2.3% 2.0% 1.7% 5.0% 13:00 3.4% 2.3% 3.5% 8.4% 9.4% 11.6% 8.5% 6.1% 2.5% 3.3% 3.0% 2.8% 5.4% 14:00 4.4% 3.3% 4.5% 8.1% 8.5% 12.6% 8.1% 5.9% 2.6% 3.7% 4.2% 3.6% 5.8% 15:00 5.1% 5.0% 4.7% 8.2% 8.6% 9.7% 7.8% 5.2% 3.0% 4.2% 4.8% 4.1% 5.9% 16:00 5.3% 6.3% 5.0% 7.9% 7.4% 9.3% 6.4% 5.1% 3.1% 4.2% 4.7% 3.7% 5.7% 17:00 4.7% 5.8% 4.8% 7.4% 7.0% 7.8% 6.1% 4.8% 3.5% 4.0% 3.8% 3.2% 5.3% 18:00 4.1% 5.0% 3.8% 6.6% 6.3% 7.8% 5.5% 4.9% 3.2% 4.1% 3.5% 2.6% 4.8% 19:00 3.9% 4.4% 3.5% 6.2% 6.3% 6.7% 5.6% 4.5% 2.7% 4.1% 2.9% 2.1% 4.4% 20:00 3.6% 4.0% 3.6% 5.9% 5.3% 6.2% 5.4% 4.1% 2.4% 3.3% 2.6% 2.2% 4.1% 21:00 3.5% 3.4% 3.5% 5.8% 4.6% 6.6% 5.0% 3.8% 2.2% 2.8% 2.1% 1.8% 3.8% 22:00 3.2% 3.2% 3.2% 5.6% 4.7% 5.5% 4.7% 3.5% 2.1% 2.2% 1.7% 1.5% 3.4% 23:00 3.1% 3.1% 3.1% 5.5% 5.3% 4.5% 4.8% 3.5% 1.9% 1.9% 1.3% 1.1% 3.3% TOTAL 5.6% 5.7% 5.7% 12.5% 13.6% 15.8% 13.6% 10.3% 5.0% 4.5% 4.0% 3.3% 129 Figure 68: Mean annual power density for short-term measured data between 11/11/2016-10/11/2018 130 Site 10: Tando Ghulam Ali, Sindh HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 3.4% 4.0% 3.2% 4.8% 4.5% 5.9% 5.2% 4.5% 3.2% 2.5% 2.6% 3.7% 4.0% 01:00 3.2% 4.0% 3.2% 4.4% 4.4% 6.1% 4.9% 4.5% 3.3% 2.3% 2.4% 3.7% 3.9% 02:00 2.9% 3.5% 2.6% 3.5% 5.0% 7.4% 5.2% 4.7% 3.1% 1.9% 2.0% 3.4% 3.8% 03:00 2.0% 2.2% 1.6% 3.1% 5.5% 10.0% 6.8% 5.6% 3.5% 1.3% 1.2% 2.3% 3.8% 04:00 0.8% 1.0% 1.3% 3.0% 5.2% 10.2% 7.8% 6.5% 4.1% 1.1% 0.5% 1.0% 3.5% 05:00 0.6% 0.8% 1.3% 3.1% 5.0% 9.3% 8.0% 6.9% 3.9% 1.0% 0.4% 0.7% 3.4% 06:00 0.8% 0.9% 1.5% 3.3% 5.1% 8.7% 7.5% 7.0% 3.8% 1.0% 0.5% 0.8% 3.4% 07:00 1.0% 1.0% 1.5% 3.6% 5.0% 8.3% 7.4% 6.7% 3.5% 1.0% 0.7% 0.9% 3.4% 08:00 1.1% 1.1% 1.6% 3.7% 5.1% 8.2% 7.2% 6.9% 3.4% 1.1% 0.7% 1.1% 3.4% 09:00 1.2% 1.1% 1.8% 4.0% 5.1% 8.4% 7.3% 6.9% 3.6% 1.1% 0.8% 1.3% 3.6% 10:00 1.2% 1.2% 2.2% 4.3% 6.0% 8.7% 7.6% 7.1% 3.7% 1.1% 0.7% 1.4% 3.8% 11:00 1.2% 1.1% 2.6% 5.0% 7.1% 9.4% 8.0% 7.1% 3.8% 1.2% 0.8% 1.4% 4.1% 12:00 1.3% 1.2% 3.1% 5.9% 8.2% 10.0% 8.3% 6.9% 3.4% 1.5% 1.1% 1.8% 4.4% 13:00 2.0% 1.8% 3.2% 6.3% 8.0% 9.9% 8.0% 5.8% 3.1% 2.0% 2.0% 2.9% 4.6% 14:00 3.0% 3.0% 3.7% 6.3% 7.0% 9.1% 7.2% 5.3% 3.4% 2.6% 2.9% 3.9% 4.8% 15:00 3.4% 3.9% 4.2% 5.9% 6.5% 9.2% 6.9% 5.3% 3.5% 3.1% 3.7% 4.5% 5.0% 16:00 3.7% 4.2% 4.2% 5.7% 6.0% 8.2% 6.5% 5.6% 3.4% 3.3% 4.1% 4.9% 5.0% 17:00 4.0% 4.5% 4.4% 5.6% 5.7% 8.3% 6.0% 5.1% 3.4% 3.4% 4.5% 4.9% 5.0% 18:00 4.0% 4.8% 4.4% 5.2% 5.3% 7.9% 5.7% 5.0% 3.3% 3.5% 4.4% 4.9% 4.9% 19:00 3.9% 5.1% 4.2% 5.2% 5.0% 7.6% 5.4% 4.8% 3.3% 3.4% 4.4% 4.9% 4.8% 20:00 3.9% 4.9% 4.1% 5.1% 4.9% 7.2% 5.5% 4.8% 3.3% 3.4% 4.1% 4.8% 4.7% 21:00 4.1% 4.8% 4.0% 4.7% 4.8% 6.7% 5.2% 4.8% 3.3% 3.3% 3.8% 4.6% 4.5% 22:00 3.9% 4.4% 3.8% 4.8% 4.5% 6.5% 5.2% 4.8% 3.0% 3.0% 3.3% 4.2% 4.3% 23:00 3.7% 4.1% 3.6% 4.6% 4.5% 6.3% 5.2% 4.4% 3.1% 3.0% 2.9% 3.8% 4.1% TOTAL 5.0% 5.7% 5.9% 9.3% 11.1% 16.5% 13.2% 11.4% 6.9% 4.3% 4.5% 6.0% 131 Figure 69: Mean annual power density for short-term measured data between 01/10/2016-30/09/2018 132 Site 11: Gwadar, Balochistan HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 3.2% 3.4% 2.6% 3.7% 3.1% 1.4% 0.6% 0.6% 0.8% 2.0% 9.7% 1.7% 2.7% 01:00 3.1% 2.8% 2.0% 3.5% 2.9% 1.3% 0.5% 0.5% 0.9% 1.4% 8.8% 1.6% 2.4% 02:00 2.9% 2.3% 2.7% 2.9% 2.3% 1.8% 0.6% 0.5% 0.9% 1.0% 8.0% 2.1% 2.3% 03:00 3.0% 3.2% 1.9% 2.1% 2.5% 1.8% 0.8% 0.5% 0.7% 0.9% 7.2% 2.0% 2.2% 04:00 3.0% 3.0% 1.5% 1.8% 3.6% 1.8% 1.1% 0.7% 0.6% 0.6% 6.6% 1.8% 2.1% 05:00 3.6% 3.1% 1.8% 2.1% 3.7% 2.4% 1.5% 1.0% 0.6% 0.7% 6.0% 2.1% 2.4% 06:00 3.4% 3.5% 3.1% 2.9% 4.7% 2.5% 2.1% 1.9% 1.0% 1.1% 4.2% 2.4% 2.7% 07:00 3.6% 3.8% 5.5% 4.4% 6.6% 3.9% 3.1% 3.4% 2.0% 1.9% 3.3% 1.8% 3.6% 08:00 3.7% 3.3% 8.2% 7.7% 10.2% 5.8% 4.1% 5.8% 3.7% 3.5% 3.1% 2.0% 5.1% 09:00 4.3% 3.8% 12.3% 12.1% 15.0% 7.5% 5.5% 8.3% 6.3% 6.1% 3.9% 3.2% 7.4% 10:00 5.2% 4.3% 13.7% 16.2% 18.1% 9.3% 7.0% 10.5% 8.2% 8.8% 4.8% 4.8% 9.3% 11:00 5.6% 5.4% 14.3% 18.0% 18.3% 9.9% 7.4% 10.9% 9.0% 8.9% 5.7% 6.1% 10.0% 12:00 5.7% 4.7% 14.6% 15.7% 16.6% 8.0% 6.3% 9.5% 8.3% 7.7% 5.7% 6.0% 9.1% 13:00 4.7% 4.9% 11.4% 11.2% 14.7% 5.6% 5.1% 7.5% 6.6% 6.0% 5.3% 5.3% 7.4% 14:00 3.5% 5.9% 8.2% 7.5% 10.7% 3.8% 3.4% 5.0% 3.8% 3.6% 4.0% 3.5% 5.2% 15:00 2.8% 5.1% 5.9% 5.7% 6.9% 2.7% 2.3% 2.7% 2.3% 2.3% 5.8% 3.1% 3.9% 16:00 2.7% 4.0% 3.9% 4.7% 5.6% 1.9% 1.6% 2.0% 1.7% 2.4% 8.1% 2.8% 3.4% 17:00 2.5% 2.7% 3.6% 4.2% 4.2% 1.9% 1.2% 1.5% 1.9% 2.1% 8.2% 2.0% 3.0% 18:00 2.3% 2.5% 3.7% 3.8% 4.0% 1.7% 1.1% 1.1% 0.9% 2.1% 7.3% 1.6% 2.6% 19:00 2.6% 3.0% 3.1% 3.3% 3.9% 1.8% 1.1% 0.9% 0.6% 2.1% 8.3% 1.7% 2.7% 20:00 2.7% 3.2% 2.6% 2.7% 3.4% 1.5% 1.1% 0.9% 0.6% 1.5% 9.0% 2.1% 2.6% 21:00 2.3% 3.0% 3.3% 2.8% 3.7% 1.0% 1.1% 0.9% 0.7% 1.1% 8.7% 2.0% 2.5% 22:00 1.9% 3.9% 2.9% 2.8% 3.2% 1.4% 1.2% 0.8% 0.7% 1.2% 10.2% 1.8% 2.6% 23:00 2.5% 3.4% 3.3% 3.2% 3.2% 1.5% 1.0% 0.6% 0.9% 1.4% 10.5% 2.3% 2.7% TOTAL 6.7% 7.3% 11.3% 12.1% 14.2% 6.9% 5.1% 6.5% 5.3% 5.9% 13.5% 5.5% 133 Figure 70: Mean annual power density for short-term measured data between 11/11/2016-10/11/2018 134 Site 12: Sujawal, Sindh HOUR/MONTH JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC YEAR 00:00 3.6% 4.7% 4.1% 4.3% 4.5% 4.1% 4.1% 4.0% 2.3% 2.5% 3.1% 4.9% 3.8% 01:00 3.7% 4.5% 3.9% 4.6% 4.2% 3.9% 4.3% 3.7% 2.4% 2.4% 2.9% 4.7% 3.8% 02:00 3.9% 4.3% 3.0% 3.6% 4.8% 5.1% 4.6% 3.9% 2.1% 2.0% 2.6% 4.3% 3.7% 03:00 2.9% 2.9% 2.0% 3.5% 6.2% 6.8% 6.1% 4.9% 2.8% 1.5% 1.6% 3.1% 3.8% 04:00 1.7% 1.6% 2.0% 3.6% 6.4% 7.1% 7.3% 5.8% 3.6% 1.5% 0.7% 1.9% 3.7% 05:00 1.6% 1.6% 2.0% 4.1% 6.8% 7.2% 7.6% 6.5% 3.7% 1.5% 0.6% 1.6% 3.8% 06:00 1.6% 1.6% 2.3% 4.6% 7.3% 7.5% 7.9% 6.8% 3.9% 1.5% 0.6% 1.7% 4.0% 07:00 1.7% 1.5% 2.5% 5.2% 7.8% 7.6% 8.3% 6.5% 3.7% 1.6% 0.7% 1.8% 4.1% 08:00 1.7% 1.5% 2.8% 5.8% 8.5% 7.9% 8.2% 6.8% 3.8% 1.6% 0.7% 1.9% 4.4% 09:00 1.7% 1.5% 3.1% 6.1% 9.3% 7.9% 8.2% 6.5% 3.8% 1.8% 0.8% 2.0% 4.5% 10:00 1.7% 1.6% 3.4% 6.2% 9.9% 8.3% 8.2% 6.6% 3.7% 1.8% 0.9% 2.0% 4.6% 11:00 1.6% 1.7% 3.4% 7.0% 9.5% 8.5% 8.1% 6.6% 3.7% 2.1% 0.9% 1.9% 4.7% 12:00 1.8% 1.7% 3.8% 6.9% 9.0% 8.6% 7.9% 6.3% 3.8% 2.4% 1.4% 2.4% 4.8% 13:00 2.5% 2.2% 4.0% 6.3% 8.0% 8.5% 7.7% 6.0% 3.6% 2.6% 2.5% 3.5% 4.9% 14:00 3.4% 3.6% 4.2% 5.5% 6.5% 7.2% 6.7% 5.1% 3.4% 2.6% 3.2% 4.3% 4.7% 15:00 3.8% 4.5% 4.0% 4.8% 5.2% 6.2% 5.7% 4.8% 2.8% 2.5% 3.5% 5.0% 4.4% 16:00 4.0% 5.0% 3.7% 4.6% 4.8% 5.6% 4.7% 4.4% 2.6% 2.5% 3.5% 5.3% 4.2% 17:00 4.1% 4.7% 3.5% 4.5% 4.7% 5.1% 4.5% 4.0% 2.5% 2.5% 3.7% 5.0% 4.1% 18:00 4.2% 4.5% 3.8% 4.9% 4.8% 4.9% 4.4% 3.9% 2.4% 2.6% 3.8% 5.0% 4.1% 19:00 4.1% 4.5% 4.0% 4.8% 4.6% 4.9% 4.5% 3.9% 2.5% 2.7% 3.8% 5.2% 4.1% 20:00 4.1% 4.7% 4.2% 4.6% 4.8% 4.9% 4.5% 4.0% 2.4% 2.6% 3.8% 5.1% 4.1% 21:00 3.9% 4.7% 3.8% 4.5% 4.6% 4.5% 4.3% 4.0% 2.3% 2.5% 3.8% 5.0% 4.0% 22:00 3.9% 4.8% 4.0% 4.5% 4.6% 4.3% 4.1% 4.1% 2.3% 2.3% 3.5% 5.0% 3.9% 23:00 3.9% 4.7% 4.1% 4.4% 4.4% 4.3% 3.8% 4.0% 2.2% 2.5% 3.4% 5.0% 3.9% TOTAL 5.9% 6.5% 6.8% 9.9% 12.6% 12.6% 12.1% 10.3% 6.0% 4.3% 4.7% 7.3% 135 Figure 71: Mean annual power density for short-term measured data between 01/10/2016-30/09/2018 136 ANNEX J LONG-TERM CORRELATION COEFFICIENTS Site 1: Peshawar, Khyber Pakhtunkhwa r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N34.0_E71.9 2h 0.246 0.531 1/8/2000-31/7/2018 1.88 1.00 100 MERRA MERRA2_N33.5_E71.9 3h 0.221 0.357 1/8/2003-31/7/2018 1.88 1.00 100 MERRA MERRA2_N34.0_E71.3 1h 0.191 0.284 1/8/2000-31/7/2018 1.88 1.00 100 MERRA MERRA2_N34.5_E71.9 -1h 0.215 0.495 1/8/2003-31/7/2018 1.88 1.00 100 ERA-Interim EmdERA_N34.0_E71.7 1h 0.167 0.169 1/7/2000-30/6/2018 1.79 6.00 100 ERA-Interim EmdERA_N34.0_E72.4 1h 0.104 -0.221 1/7/2000-30/6/2018 1.79 6.00 100 ERA-Interim EmdERA_N33.3_E71.7 1h 0.176 0.253 1/7/2000-30/6/2018 1.79 6.00 100 ERA-Interim EmdERA_N34.0_E71.0 2h 0.180 0.201 1/7/2000-30/6/2018 1.79 6.00 100 *Met. Station Islamabad_Airport_METAR_N33. -2h 0.270 0.878 1/10/2006-30/9/2017 1.04 1.00 90 – METAR 6_E73.1 Met. Station - SAIDU_SHARIF_SYNOP_41- -2h 0.059 0.620 31/3/2007-31/3/2017 0.54 6.00 87 SYNOP 523_N34.7_E72.4 *Met. Station ISLAMABAD_(CIV_MIL)_SYNOP_4 -3h 0.130 0.812 31/3/2005-31/3/2017 0.54 3.00 85 - SYNOP 1-571_N33.6_E73.1 *Met. Station KAKUL_SYNOP_41- -1h 0.083 0.704 30/3/2006-30/3/2018 1.54 3.00 89 - SYNOP 535_N34.2_E73.3 ERA5 ERA5_N33.9_E71.7 0h 0.262 0.424 1/7/2000-30/6/2018 1.79 1.00 100 ERA5 ERA5_N33.9_E72.0 1h 0.261 0.524 1/7/2000-30/6/2018 1.79 1.00 100 ERA5 ERA5_N34.2_E71.7 0h 0.304 0.601 1/7/2000-30/6/2018 1.79 1.00 100 ERA5 ERA5_N34.2_E72.0 1h 0.272 0.608 1/7/2000-30/6/2018 1.79 1.00 100 * Discarded from further analysis despite good correlation due to their inconsistent behavior over time 137 Site 2: Haripur, Khyber Pakhtunkhwa r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N34.0_E73.1 1h 0.068 0.236 1/8/2008-31/7/2018 2.58 1.00 100 MERRA MERRA2_N34.0_E72.5 0h 0.108 -0.090 1/8/2008-31/7/2018 2.58 1.00 100 MERRA MERRA2_N33.5_E73.1 1h 0.101 0.084 1/8/2003-31/7/2018 2.58 1.00 100 MERRA MERRA2_N34.5_E73.1 0h 0.129 0.160 1/8/2008-31/7/2018 2.58 1.00 100 ERA-Interim EmdERA_N34.0_E73.1 1h 0.077 -0.167 1/7/2000-30/6/2018 2.49 6.00 100 ERA-Interim EmdERA_N34.0_E72.4 2h 0.175 -0.244 1/7/2000-30/6/2018 2.49 6.00 100 ERA-Interim EmdERA_N33.3_E73.1 2h 0.114 0.124 1/7/2001-30/6/2018 2.49 6.00 100 ERA-Interim EmdERA_N34.0_E73.8 1h 0.041 -0.105 1/7/2000-30/6/2018 2.49 6.00 100 Met. Station - Islamabad_Airport_METAR_N33. 0h 0.132 0.280 1/10/2006-30/9/2017 1.74 1.00 90 METAR 6_E73.1 Met. Station - METAR_N34.4_E70.5 -3h 0.062 0.633 12/5/2009-11/5/2016 0.36 1.00 84 METAR Met. Station - KAKUL_SYNOP_41- 1h 0.003 0.002 30/3/2006-30/3/2018 2.24 3.00 89 SYNOP 535_N34.2_E73.3 Met. Station - SYNOP_41-573_N33.9_E73.4 -1h 0.074 0.030 30/3/2008-30/3/2018 2.24 6.00 94 SYNOP Met. Station - ISLAMABAD_(CIV_MIL)_SYNOP_4 -1h 0.103 0.110 31/3/2005-31/3/2017 1.24 3.00 85 SYNOP 1-571_N33.6_E73.1 Met. Station - ERA5_N33.9_E72.9 1h 0.075 0.354 1/7/2000-30/6/2018 2.49 1.00 100 METAR ERA5 ERA5_N34.2_E72.9 1h 0.130 0.445 1/7/2000-30/6/2018 2.49 1.00 100 ERA5 ERA5_N33.9_E73.3 2h 0.053 0.123 1/7/2003-30/6/2018 2.49 1.00 100 ERA5 ERA5_N34.2_E73.3 1h 0.118 0.429 1/7/2000-30/6/2018 2.49 1.00 100 138 Site 3: Chakri, Punjab r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N33.5_E72.5 2h 0.330 0.745 1/9/2006-31/8/2018 2.53 1.00 100 MERRA MERRA2_N33.5_E73.1 2h 0.252 0.327 1/9/2003-31/8/2018 2.53 1.00 100 MERRA MERRA2_N33.0_E72.5 2h 0.329 0.767 1/9/2003-31/8/2018 2.53 1.00 100 MERRA MERRA2_N33.0_E73.1 3h 0.300 0.448 1/9/2000-31/8/2018 2.53 1.00 100 ERA-Interim EmdERA_N33.3_E72.4 1h 0.337 0.784 1/8/2000-31/7/2018 2.45 6.00 100 ERA-Interim EmdERA_N33.3_E73.1 3h 0.311 0.684 1/8/2001-31/7/2018 2.45 6.00 100 ERA-Interim EmdERA_N32.6_E72.4 3h 0.370 0.872 1/8/2001-31/7/2018 2.45 6.00 100 ERA-Interim EmdERA_N32.6_E73.1 3h 0.386 0.929 1/8/2003-31/7/2018 2.45 6.00 100 ERA5 ERA5_N33.3_E72.6 1h 0.433 0.682 1/8/2000-31/7/2018 2.45 1.00 100 ERA5 ERA5_N33.3_E72.9 1h 0.433 0.582 1/8/2000-31/7/2018 2.45 1.00 100 ERA5 ERA5_N33.6_E72.6 1h 0.428 0.691 1/8/2000-31/7/2018 2.45 1.00 100 ERA5 ERA5_N33.0_E72.6 0h 0.424 0.759 1/8/2002-31/7/2018 2.45 1.00 100 *Met. Station ISLAMABAD_(CIV_MIL)_SYNOP_4 1h 0.396 0.755 31/3/2005-31/3/2017 1.11 6.00 89 - SYNOP 1-571_N33.6_E73.1 *Met. Station Islamabad_Airport_METAR_N33. 0h 0.464 0.794 1/10/2006-30/9/2017 1.61 1.00 90 - METAR 6_E73.1 Met. Station - METAR_N34.4_E70.5 -2h 0.201 0.957 12/5/2007-11/5/2016 0.23 1.00 80 METAR * Discarded from further analysis despite good correlation due to their inconsistent behavior over time 139 Site 4: Quaidabad, Punjab r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N32.5_E71.9 0h 0.356 0.709 1/8/2005-31/7/2018 1.90 1.00 100 MERRA MERRA2_N32.0_E71.9 0h 0.354 0.704 1/8/2004-31/7/2018 1.90 1.00 100 MERRA MERRA2_N32.5_E72.5 0h 0.310 0.252 1/8/2003-31/7/2018 1.90 1.00 100 MERRA MERRA2_N32.5_E71.3 0h 0.289 0.751 1/8/2003-31/7/2018 1.90 1.00 100 ERA-Interim EmdERA_N32.6_E71.7 1h 0.318 0.586 1/7/2000-30/6/2018 1.82 6.00 100 ERA-Interim EmdERA_N31.9_E71.7 1h 0.351 0.675 1/7/2001-30/6/2018 1.82 6.00 100 ERA-Interim EmdERA_N32.6_E72.4 1h 0.348 0.678 1/7/2001-30/6/2018 1.82 6.00 100 ERA-Interim EmdERA_N31.9_E72.4 2h 0.395 0.829 1/7/2003-30/6/2018 1.82 6.00 100 *Met. Station Islamabad_Airport_METAR_N33. -3h 0.266 0.735 1/10/2006-30/9/2017 1.07 1.00 87 - METAR 6_E73.1 *Met. Station DERA_ISMAIL_KHAN_SYNOP_41- 0h 0.143 0.734 30/3/2003-30/3/2018 1.56 6.00 88 - SYNOP 624_N31.8_E70.9 Met. Station - SYNOP_41-630_N31.4_E73.1 3h 0.142 0.388 1/4/2007-31/3/2017 0.57 6.00 93 SYNOP Met. Station - ISLAMABAD_(CIV_MIL)_SYNOP_4 -2h 0.074 0.217 31/3/2005-31/3/2017 0.57 3.00 85 SYNOP 1-571_N33.6_E73.1 Met. Station - JHELUM_SYNOP_41- -1h 0.140 0.677 31/3/2001-30/3/2018 1.56 6.00 88 SYNOP 598_N32.9_E73.7 ERA5 ERA5_N32.5_E72.0 1h 0.467 0.886 1/7/2002-30/6/2018 1.82 1.00 100 ERA5 ERA5_N32.2_E71.7 1h 0.453 0.765 1/7/2000-30/6/2018 1.82 1.00 100 ERA5 ERA5_N32.2_E72.0 1h 0.519 0.916 1/7/2000-30/6/2018 1.82 1.00 100 ERA5 ERA5_N32.5_E71.7 1h 0.364 0.696 1/7/2000-30/6/2018 1.82 1.00 100 * Discarded from further analysis despite good correlation due to their inconsistent behavior over time 140 Site 5: Bahawalpur, Punjab r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N29.5_E71.9 -1h 0.470 0.712 1/9/2004-31/8/2018 2.38 1.00 100 MERRA MERRA2_N29.0_E71.9 0h 0.462 0.703 1/9/2001-31/8/2018 2.38 1.00 100 MERRA MERRA2_N29.5_E71.3 -1h 0.407 0.668 1/9/2004-31/8/2018 2.38 1.00 100 MERRA MERRA2_N29.0_E71.3 -1h 0.436 0.672 1/9/2002-31/8/2018 2.38 1.00 100 ERA-Interim EmdERA_N29.1_E71.7 0h 0.573 0.905 1/8/2002-31/7/2018 2.30 6.00 100 ERA-Interim EmdERA_N29.8_E71.7 0h 0.513 0.885 1/8/2003-31/7/2018 2.30 6.00 100 ERA-Interim EmdERA_N29.1_E72.4 0h 0.544 0.903 1/8/2002-31/7/2018 2.30 6.00 100 ERA-Interim EmdERA_N29.8_E72.4 0h 0.520 0.886 1/8/2002-31/7/2018 2.30 6.00 100 ERA5 ERA5_N29.4_E71.9 1h 0.642 0.923 1/8/2002-31/7/2018 2.30 1.00 100 ERA5 ERA5_N29.4_E71.6 1h 0.637 0.910 1/8/2003-31/7/2018 2.30 1.00 100 ERA5 ERA5_N29.1_E71.9 1h 0.627 0.902 1/8/2002-31/7/2018 2.30 1.00 100 ERA5 ERA5_N29.4_E72.2 1h 0.616 0.916 1/8/2003-31/7/2018 2.30 1.00 100 Met. Station - BAHAWALPUR_SYNOP_41- 2h 0.426 0.772 25/9/2003-24/9/2018 2.44 6.00 90 SYNOP 700_N29.4_E71.8 *Met. Station MULTAN_SYNOP_41- 2h 0.305 0.821 25/9/2003-24/9/2018 2.44 6.00 90 - SYNOP 675_N30.2_E71.4 Met. Station - KHANPUR_SYNOP_41- 3h 0.353 0.765 25/9/2008-24/9/2018 2.44 6.00 94 SYNOP 718_N28.7_E70.7 Met. Station - BAHAWALNAGAR_SYNOP_41- 3h 0.212 0.739 25/9/2007-24/9/2018 2.44 6.00 93 SYNOP 678_N29.9_E73.3 * Discarded from further analysis despite good correlation due to their inconsistent behavior over time 141 Site 6: Sadiqabad, Punjab r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N28.0_E70.0 0h 0.551 0.761 1/9/2000-31/8/2018 2.34 1.00 100 MERRA MERRA2_N28.5_E70.0 0h 0.551 0.834 1/9/2000-31/8/2018 2.34 1.00 100 MERRA MERRA2_N28.0_E70.6 0h 0.527 0.760 1/9/2000-31/8/2018 2.34 1.00 100 MERRA MERRA2_N28.0_E69.4 -1h 0.536 0.778 1/9/2000-31/8/2018 2.34 1.00 100 ERA-Interim EmdERA_N28.4_E70.3 1h 0.607 0.883 1/8/2000-31/7/2018 2.25 6.00 100 ERA-Interim EmdERA_N28.4_E69.6 1h 0.564 0.849 1/8/2000-31/7/2018 2.25 6.00 100 ERA-Interim EmdERA_N27.7_E70.3 1h 0.552 0.803 1/8/2002-31/7/2018 2.25 6.00 100 ERA-Interim EmdERA_N27.7_E69.6 1h 0.557 0.763 1/8/2000-31/7/2018 2.25 6.00 100 ERA5 ERA5_N28.2_E70.0 1h 0.710 0.922 1/8/2001-31/7/2018 2.25 1.00 100 ERA5 ERA5_N28.2_E69.7 1h 0.701 0.944 1/8/2001-31/7/2018 2.25 1.00 100 ERA5 ERA5_N28.2_E70.3 1h 0.687 0.892 1/8/2001-31/7/2018 2.25 1.00 100 *Met. Station KHANPUR_SYNOP_41- 3h 0.334 0.866 25/9/2008-24/9/2018 2.40 6.00 94 - SYNOP 718_N28.7_E70.7 Met. Station - ROHRI_SYNOP_41- 1h 0.253 0.565 25/9/2006-24/9/2018 2.40 6.00 94 SYNOP 725_N27.7_E68.9 Met. Station - JACOBABAD_(CIV_MIL)_SYNOP_4 3h 0.064 0.403 25/9/2008-24/9/2018 2.40 3.00 89 SYNOP 1-715_N28.3_E68.5 Met. Station - JAISALMER_SYNOP_42- 3h 0.269 0.614 1/4/2007-31/3/2017 0.91 3.00 97 SYNOP 328_N26.9_E70.9 * Discarded from further analysis despite good correlation due to their inconsistent behavior over time 142 Site 7: Quetta, Balochistan r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N30.5_E66.9 0h 0.544 0.647 1/9/2000-31/8/2018 2.24 1.00 100 MERRA MERRA2_N30.0_E66.9 -1h 0.435 0.210 1/9/2000-31/8/2018 2.24 1.00 100 MERRA MERRA2_N30.5_E67.5 -1h 0.266 0.063 1/9/2000-31/8/2018 2.24 1.00 100 MERRA MERRA2_N30.0_E67.5 0h 0.052 -0.138 1/9/2001-31/8/2018 2.24 1.00 100 ERA-Interim EmdERA_N30.5_E66.8 -1h 0.460 0.700 1/8/2000-31/7/2018 2.15 6.00 100 ERA-Interim EmdERA_N29.8_E66.8 1h 0.480 0.471 1/8/2000-31/7/2018 2.15 6.00 100 ERA-Interim EmdERA_N30.5_E67.5 2h 0.380 0.477 1/8/2000-31/7/2018 2.15 6.00 100 ERA-Interim EmdERA_N29.8_E67.5 -1h 0.299 0.592 1/8/2000-31/7/2018 2.15 6.00 100 ERA5 ERA5_N30.2_E66.9 0h 0.628 0.878 1/8/2000-31/7/2018 2.15 1.00 100 ERA5 ERA5_N30.2_E67.2 -1h 0.282 -0.078 1/8/2000-31/7/2018 2.15 1.00 100 ERA5 ERA5_N30.5_E66.9 0h 0.619 0.953 1/8/2000-31/7/2018 2.15 1.00 100 ERA5 ERA5_N30.5_E67.2 0h 0.614 0.877 1/8/2000-31/7/2018 2.15 1.00 100 *Met. Station QUETTA_(CIV_MIL)_SYNOP_41- 0h 0.604 0.901 25/9/2005-24/9/2018 2.30 3.00 86 - SYNOP 660_N30.3_E66.9 Met. Station - SIBI_SYNOP_41- 0h 0.175 0.521 25/9/2003-24/9/2018 2.30 6.00 89 SYNOP 697_N29.6_E67.9 Met. Station - KALAT_&_SYNOP_41- -2h 0.091 0.044 25/9/2013-24/9/2018 2.30 3.00 92 SYNOP 696_N29.0_E66.6 Met. Station - JACOBABAD_(CIV_MIL)_SYNOP_4 -3h 0.148 0.620 25/9/2008-24/9/2018 2.30 3.00 89 SYNOP 1-715_N28.3_E68.5 * Discarded from further analysis despite good correlation due to their inconsistent behavior over time 143 Site 8: Sanghar, Sindh r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N26.0_E68.8 0h 0.757 0.983 1/10/2000-30/9/2018 2.17 1.00 100 MERRA MERRA2_N26.0_E69.4 0h 0.754 0.982 1/10/2000-30/9/2018 2.17 1.00 100 MERRA MERRA2_N25.5_E68.8 -1h 0.781 0.990 1/10/2000-30/9/2018 2.17 1.00 100 MERRA MERRA2_N25.5_E69.4 0h 0.760 0.987 1/10/2000-30/9/2018 2.17 1.00 100 ERA-Interim EmdERA_N25.6_E68.9 1h 0.806 0.984 1/9/2004-31/8/2018 2.09 6.00 100 ERA-Interim EmdERA_N26.3_E68.9 2h 0.774 0.990 1/9/2003-31/8/2018 2.09 6.00 100 ERA-Interim EmdERA_N25.6_E69.6 2h 0.779 0.969 1/9/2004-31/8/2018 2.09 6.00 100 ERA-Interim EmdERA_N26.3_E69.6 2h 0.771 0.978 1/9/2003-31/8/2018 2.09 6.00 100 ERA5 ERA5_N25.7_E69.0 1h 0.854 0.987 1/9/2001-31/8/2018 2.09 1.00 100 ERA5 ERA5_N25.9_E69.0 1h 0.848 0.990 1/9/2001-31/8/2018 2.09 1.00 100 ERA5 ERA5_N25.7_E69.3 1h 0.833 0.986 1/9/2002-31/8/2018 2.09 1.00 100 ERA5 ERA5_N25.9_E69.3 1h 0.829 0.993 1/9/2002-31/8/2018 2.09 1.00 100 Met. Station - HYDERABAD_AIRPORT_SYNOP_4 -3h 0.434 0.884 1/4/2007-31/3/2017 0.67 3.00 91 SYNOP 1-764_N25.3_E68.5 *Met. Station CHHOR_SYNOP_41- -3h 0.533 0.968 25/9/2008-24/9/2018 2.15 3.00 91 - SYNOP 768_N25.5_E69.8 Met. Station - NAWABSHAH_SYNOP_41- -3h 0.359 0.974 25/9/2003-24/9/2018 2.15 6.00 89 SYNOP 749_N26.3_E68.4 Met. Station - BADIN_SYNOP_41- -3h 0.492 0.926 25/9/2007-24/9/2018 2.15 6.00 94 SYNOP 785_N24.6_E68.9 * Discarded from further analysis despite good correlation due to their inconsistent behavior over time 144 Site 9: Umerkot, Sindh r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N25.0_E69.4 1h 0.782 0.981 1/11/2000- 1.97 1.00 100 31/10/2018 MERRA MERRA2_N25.0_E70.0 2h 0.764 0.975 1/11/2000- 1.97 1.00 100 31/10/2018 MERRA MERRA2_N25.5_E69.4 1h 0.756 0.964 1/11/2000- 1.97 1.00 100 31/10/2018 MERRA MERRA2_N25.5_E70.0 2h 0.738 0.956 1/11/2000- 1.97 1.00 100 31/10/2018 ERA-Interim EmdERA_N24.9_E69.6 1h 0.806 0.982 1/9/2003-31/8/2018 1.80 6.00 100 ERA-Interim EmdERA_N25.6_E69.6 1h 0.782 0.979 1/9/2004-31/8/2018 1.80 6.00 100 ERA-Interim EmdERA_N24.9_E68.9 1h 0.811 0.983 1/9/2004-31/8/2018 1.80 6.00 100 ERA-Interim EmdERA_N24.9_E70.3 2h 0.763 0.981 1/9/2002-31/8/2018 1.80 6.00 100 ERA5 ERA5_N25.2_E69.6 1h 0.839 0.989 1/9/2001-31/8/2018 1.80 1.00 100 ERA5 ERA5_N24.9_E69.6 1h 0.847 0.990 1/9/2001-31/8/2018 1.80 1.00 100 ERA5 ERA5_N25.2_E69.3 1h 0.838 0.988 1/9/2001-31/8/2018 1.80 1.00 100 ERA5 ERA5_N25.2_E69.9 1h 0.823 0.989 1/9/2002-31/8/2018 1.80 1.00 100 *Met. Station CHHOR_SYNOP_41- -3h 0.614 0.962 25/9/2008-24/9/2018 1.87 3.00 91 - SYNOP 768_N25.5_E69.8 *Met. Station BADIN_SYNOP_41- -3h 0.598 0.962 25/9/2007-24/9/2018 1.87 3.00 92 - SYNOP 785_N24.6_E68.9 Met. Station - HYDERABAD_AIRPORT_SYNOP_4 -3h 0.423 0.693 1/4/2007-31/3/2017 0.39 3.00 91 SYNOP 1-764_N25.4_E68.6 Met. Station - NAWABSHAH_SYNOP_41- -3h 0.378 0.910 25/9/2003-24/9/2018 1.87 6.00 89 SYNOP 749_N26.3_E68.4 * Discarded from further analysis despite good correlation due to their inconsistent behavior over time 145 Site 10: Tando Ghulam Ali, Sindh r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N25.0_E68.8 -1h 0.789 0.985 1/9/2000-31/8/2018 2.46 1.00 100 MERRA MERRA2_N25.5_E68.8 0h 0.766 0.972 1/9/2000-31/8/2018 2.46 1.00 100 MERRA MERRA2_N25.0_E69.4 0h 0.763 0.986 1/9/2000-31/8/2018 2.46 1.00 100 MERRA MERRA2_N25.5_E69.4 0h 0.735 0.974 1/9/2000-31/8/2018 2.46 1.00 100 ERA-Interim EmdERA_N24.9_E68.9 1h 0.821 0.982 1/8/2004-31/7/2018 2.37 6.00 100 ERA-Interim EmdERA_N25.6_E68.9 1h 0.779 0.965 1/8/2004-31/7/2018 2.37 6.00 100 ERA-Interim EmdERA_N24.9_E68.2 1h 0.793 0.963 1/8/2003-31/7/2018 2.37 6.00 100 ERA-Interim EmdERA_N24.9_E69.6 1h 0.776 0.970 1/8/2003-31/7/2018 2.37 6.00 100 ERA5 ERA5_N25.2_E69.0 1h 0.848 0.988 1/8/2001-31/7/2018 2.38 1.00 100 ERA5 ERA5_N25.2_E68.7 1h 0.862 0.984 1/8/2002-31/7/2018 2.38 1.00 100 ERA5 ERA5_N24.9_E69.0 1h 0.835 0.987 1/8/2001-31/7/2018 2.38 1.00 100 ERA5 ERA5_N24.9_E68.7 0h 0.846 0.984 1/8/2002-31/7/2018 2.38 1.00 100 *Met. Station HYDERABAD_AIRPORT_SYNOP_4 -1h 0.653 0.965 1/4/2007-31/3/2017 1.04 3.00 91 - SYNOP 1-764_N25.4_E68.6 Met. Station - BADIN_SYNOP_41- 0h 0.569 0.942 25/9/2007-24/9/2018 2.52 6.00 94 SYNOP 785_N24.6_E68.9 Met. Station - CHHOR_SYNOP_41- -3h 0.561 0.964 25/9/2008-24/9/2018 2.52 6.00 92 SYNOP 768_N25.5_E69.8 Met. Station - NAWABSHAH_SYNOP_41- -3h 0.299 0.845 23/9/2002-24/9/2018 2.52 6.00 89 SYNOP 749_N26.3_E68.4 * Discarded from further analysis despite good correlation due to their inconsistent behavior over time 146 Site 11: Gwadar, Balochistan r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N25.5_E62.5 2h 0.575 0.750 1/10/2007-30/9/2018 1.89 1.00 100 MERRA MERRA2_N25.0_E62.5 1h 0.627 0.889 1/10/2004-30/9/2018 1.89 1.00 100 MERRA MERRA2_N25.5_E61.9 1h 0.617 0.827 1/10/2008-30/9/2018 1.89 1.00 100 MERRA MERRA2_N25.0_E61.9 1h 0.626 0.845 1/10/2008-30/9/2018 1.89 1.00 100 ERA-Interim EmdERA_N25.6_E62.6 2h 0.644 0.786 1/9/2008-31/8/2018 1.80 6.00 100 ERA-Interim EmdERA_N24.9_E62.6 2h 0.667 0.633 1/9/2000-31/8/2018 1.80 6.00 100 ERA-Interim EmdERA_N25.6_E61.9 2h 0.648 0.628 1/9/2008-31/8/2018 1.80 6.00 100 ERA-Interim EmdERA_N24.9_E61.9 2h 0.618 0.530 1/9/2008-31/8/2018 1.80 6.00 100 ERA5 ERA5_N25.2_E62.4 1h 0.738 0.920 1/9/2000-31/8/2018 1.80 1.00 100 ERA5 ERA5_N25.4_E62.4 2h 0.723 0.923 1/9/2000-31/8/2018 1.80 1.00 100 ERA5 ERA5_N25.2_E62.1 1h 0.723 0.866 1/9/2000-31/8/2018 1.80 1.00 100 ERA5 ERA5_N25.4_E62.1 2h 0.730 0.837 1/9/2000-31/8/2018 1.80 1.00 100 Met. Station - JIWANI_(CAPE)_SYNOP_41- 0h 0.526 0.477 25/9/2001-24/9/2018 1.87 3.00 84 SYNOP 756_N25.1_E61.8 Met. Station - PASNI_SYNOP_41- 0h 0.616 0.603 24/9/2005-24/9/2018 1.87 6.00 90 SYNOP 759_N25.3_E63.5 Met. Station - CHAH_BAHAR_(IR- -1h 0.343 -0.101 24/9/2007-24/9/2018 1.87 6.00 94 SYNOP AFB)_SYNOP_40- 898_N25.3_E60.6 Met. Station - SARAVAN_SYNOP_40- SYNOP 878_N27.3_E62.3 1h 0.322 0.502 24/9/2007-24/9/2018 1.87 3.00 93 Met. Station - 1/11/2010- METAR Iranshahr_METAR_N27.2_E60.7 1h 0.259 -0.137 31/10/2017 0.97 2.00 88 Met. Station - Seeb,_International_Airport_MET 1/11/2007- METAR AR_N23.6_E58.3 -2h 0.314 0.603 31/10/2017 0.97 1.00 85 147 Site 12: Sujawal, Sindh r (all r (monthly Concurrent Type Name Timeshift Selected LT period Time resolution Data availability data) averages) period MERRA MERRA2_N24.5_E68.1 -1h 0.789 0.960 1/9/2000-31/8/2018 2.47 1.00 100 MERRA MERRA2_N25.0_E68.1 0h 0.738 0.936 1/9/2000-31/8/2018 2.47 1.00 100 MERRA MERRA2_N24.5_E68.8 0h 0.742 0.957 1/9/2000-31/8/2018 2.47 1.00 100 MERRA MERRA2_N24.0_E68.1 -1h 0.770 0.957 1/9/2000-31/8/2018 2.47 1.00 100 ERA-Interim EmdERA_N24.2_E68.2 0h 0.786 0.952 1/8/2002-31/7/2018 2.39 6.00 100 ERA-Interim EmdERA_N24.9_E68.2 1h 0.753 0.931 1/8/2003-31/7/2018 2.39 6.00 100 ERA-Interim EmdERA_N24.2_E67.5 0h 0.729 0.936 1/8/2002-31/7/2018 2.39 6.00 100 ERA-Interim EmdERA_N24.2_E68.9 1h 0.772 0.959 1/8/2002-31/7/2018 2.39 6.00 100 ERA5 ERA5_N24.6_E68.1 0h 0.856 0.981 1/8/2000-31/7/2018 2.39 1.00 100 ERA5 ERA5_N24.6_E68.4 1h 0.833 0.981 1/8/2000-31/7/2018 2.39 1.00 100 ERA5 ERA5_N24.3_E68.1 0h 0.850 0.982 1/8/2000-31/7/2018 2.39 1.00 100 ERA5 ERA5_N24.3_E68.4 1h 0.832 0.979 1/8/2000-31/7/2018 2.39 1.00 100 Met. Station - BADIN_SYNOP_41- 2h 0.576 0.884 25/9/2007-24/9/2018 2.54 3.00 92 SYNOP 785_N24.6_E68.9 *Met. Station HYDERABAD_AIRPORT_SYNOP_4 -1h 0.600 0.958 1/4/2007-31/3/2017 1.06 3.00 91 - SYNOP 1-764_N25.4_E68.6 Met. Station - KARACHI_INTL_ARPT_SYNOP_41- -1h 0.459 0.831 25/9/2001-24/9/2018 2.54 3.00 87 SYNOP 780_N24.9_E67.1 Met. Station - NAWABSHAH_SYNOP_41- -3h 0.224 0.698 23/9/2002-24/9/2018 2.54 6.00 89 SYNOP 749_N26.3_E68.4 * Discarded from further analysis despite good correlation due to their inconsistent behavior over time 148 ANNEX K ESTIMATES OF EQUIVALENT MEAN AND DIRECTIONAL WEIBULL WIND SPEED DISTRIBUTIONS Weibull parameters (A, k) at different heights A.G.L. with a roughness value of 0.03m and directional wind frequency are given in the tables below for each site. Then wind speed and power density at each height are calculated and given as a graph below. Site 1: Peshawar, Khyber Pakhtunkhwa Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 8.2 8.5 5.2 3.5 3.2 6.9 7.0 12.8 20.5 8.8 7.9 7.7 50m A [m/s) 3.1 3.7 3.5 3.2 3.4 4.8 3.1 3.7 3.6 2.8 2.6 2.6 A.G.L. k 1.63 1.73 1.41 1.13 1.20 1.47 1.17 1.86 2.04 1.57 1.20 1.33 100m A [m/s) 3.5 4.2 3.9 3.6 3.8 5.4 3.5 4.2 4.1 3.2 2.9 3.0 A.G.L. k 1.62 1.71 1.40 1.12 1.19 1.46 1.16 1.84 2.01 1.55 1.19 1.32 150m A [m/s) 3.7 4.4 4.0 3.7 4.0 5.7 3.6 4.5 4.4 3.3 2.9 3.1 A.G.L. k 1.40 1.48 1.21 0.98 1.03 1.27 1.01 1.59 1.74 1.35 1.04 1.15 200m A [m/s) 3.8 4.6 4.1 3.7 4.0 5.8 3.6 4.6 4.5 3.4 3.0 3.1 A.G.L. k 1.26 1.33 1.09 0.89 0.94 1.14 0.92 1.42 1.56 1.21 0.94 1.03 149 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. 150 Site 2: Haripur, Khyber Pakhtunkhwa Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 7.3 31.3 12.6 3.8 2.1 1.9 3.9 10.9 14.3 7.5 2.7 1.8 50m A [m/s) 5.7 6.0 5.0 3.8 3.1 2.7 3.0 3.4 3.4 3.4 3.0 2.5 A.G.L. k 2.47 3.05 2.37 1.75 1.21 0.99 1.26 2.08 2.60 2.28 1.65 1.15 100m A [m/s) 6.0 6.3 5.2 3.9 3.2 2.7 3.1 3.5 3.6 3.5 3.1 2.5 A.G.L. k 2.33 2.88 2.24 1.65 1.14 0.95 1.19 1.96 2.45 2.15 1.56 1.09 150m A [m/s) 6.0 6.3 5.2 3.9 3.1 2.7 3.1 3.5 3.6 3.5 3.1 2.5 A.G.L. k 2.19 2.70 2.10 1.55 1.08 0.90 1.12 1.84 2.30 2.02 1.47 1.03 200m A [m/s) 5.9 6.3 5.1 3.9 3.0 2.6 3.0 3.5 3.5 3.5 3.0 2.4 A.G.L. k 2.07 2.55 1.99 1.47 1.02 0.85 1.06 1.74 2.17 1.90 1.39 0.97 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. 151 Site 3: Chakri, Punjab Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 5.3 15.9 16.2 9.7 6.0 5.0 5.9 4.8 3.7 5.5 12.6 9.3 50m A [m/s) 2.8 3.7 3.4 2.8 2.2 2.0 2.1 2.2 2.3 3.1 4.5 4.2 A.G.L. k 1.14 1.76 1.63 1.31 1.14 1.05 1.23 1.26 1.16 1.46 1.89 1.54 100m A [m/s) 3.3 4.4 4.0 3.4 2.6 2.4 2.6 2.6 2.7 3.6 5.3 5.0 A.G.L. k 1.21 1.88 1.74 1.40 1.21 1.12 1.31 1.34 1.24 1.56 2.02 1.65 150m A [m/s) 3.7 4.9 4.5 3.7 2.9 2.7 2.9 2.9 3.0 4.1 5.9 5.6 A.G.L. k 1.19 1.84 1.71 1.37 1.19 1.10 1.28 1.31 1.21 1.53 1.97 1.61 200m A [m/s) 4.0 5.4 4.9 4.1 3.2 2.9 3.1 3.2 3.3 4.4 6.5 6.1 A.G.L. k 1.17 1.81 1.67 1.35 1.17 1.08 1.26 1.29 1.19 1.50 1.94 1.58 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. 152 Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. 153 Site 4: Quaidabad, Punjab Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 5.1 4.6 4.5 21.2 14.6 5.9 5.5 3.4 3.0 5.9 13.1 13.2 50m A [m/s) 3.2 4.3 3.8 6.2 4.8 3.6 3.6 3.2 2.7 3.1 4.2 4.6 A.G.L. k 1.15 1.11 1.24 2.26 1.93 1.62 1.37 1.32 1.26 1.51 1.53 1.54 100m A [m/s) 3.9 5.1 4.6 7.3 5.7 4.3 4.3 3.8 3.3 3.7 5.0 5.5 A.G.L. k 1.22 1.18 1.33 2.42 2.06 1.73 1.46 1.41 1.34 1.61 1.63 1.65 150m A [m/s) 4.3 5.7 5.1 8.2 6.4 4.8 4.8 4.2 3.6 4.1 5.5 6.1 A.G.L. k 1.20 1.16 1.30 2.36 2.01 1.69 1.42 1.38 1.31 1.57 1.59 1.61 200m A [m/s) 4.7 6.2 5.6 8.9 7.0 5.3 5.2 4.6 4.0 4.5 6.0 6.7 A.G.L. k 1.18 1.14 1.28 2.32 1.98 1.66 1.40 1.36 1.29 1.54 1.56 1.58 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. 154 Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. Site 5: Bahawalpur, Punjab Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 13.0 11.1 8.7 7.3 6.2 5.9 10.9 10.8 7.3 4.6 4.7 9.3 50m A [m/s) 5.3 5.5 5.9 5.8 5.1 4.8 6.6 5.9 4.6 3.9 3.9 4.9 A.G.L. k 2.21 2.23 2.33 2.25 2.14 2.10 2.59 2.12 1.88 1.90 1.40 1.82 100m A [m/s) 6.3 6.6 7.0 6.9 6.1 5.7 7.8 7.0 5.5 4.6 4.6 5.8 A.G.L. k 2.36 2.38 2.49 2.40 2.28 2.24 2.77 2.26 2.01 2.04 1.49 1.94 150m A [m/s) 7.1 7.3 7.8 7.7 6.8 6.4 8.7 7.9 6.2 5.1 5.1 6.5 A.G.L. k 2.30 2.33 2.43 2.35 2.23 2.19 2.71 2.21 1.97 1.99 1.46 1.90 200m A [m/s) 7.7 8.0 8.6 8.4 7.4 7.0 9.5 8.6 6.7 5.6 5.6 7.1 A.G.L. k 2.26 2.29 2.38 2.31 2.19 2.15 2.66 2.17 1.93 1.96 1.43 1.87 155 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. 156 Site 6: Sadiqabad, Punjab Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 14.0 8.5 3.9 2.7 2.9 5.5 10.7 12.8 11.4 9.8 7.0 10.9 50m A [m/s) 5.7 5.2 4.2 3.6 3.7 4.9 5.8 5.4 5.1 5.7 5.0 5.7 A.G.L. k 2.61 2.28 1.72 1.36 1.53 2.00 2.44 2.11 1.65 1.60 1.44 2.15 100m A [m/s) 6.7 6.1 5.0 4.3 4.5 5.8 6.9 6.4 6.1 6.7 6.0 6.8 A.G.L. k 2.78 2.44 1.83 1.45 1.63 2.14 2.61 2.25 1.76 1.71 1.53 2.29 150m A [m/s) 7.5 6.9 5.6 4.8 5.0 6.4 7.7 7.2 6.8 7.5 6.7 7.6 A.G.L. k 2.72 2.38 1.79 1.42 1.60 2.09 2.55 2.20 1.72 1.67 1.50 2.24 200m A [m/s) 8.2 7.5 6.1 5.3 5.4 7.0 8.4 7.9 7.4 8.2 7.3 8.3 A.G.L. k 2.67 2.34 1.76 1.39 1.57 2.05 2.51 2.16 1.69 1.64 1.47 2.20 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. 157 Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. Site 7: Quetta, Balochistan Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 7.6 7.0 8.9 9.9 9.0 8.2 6.9 6.1 7.9 8.9 10.1 9.5 50m A [m/s) 5.0 4.7 4.8 4.6 4.8 4.6 4.5 4.7 5.0 4.4 4.2 4.6 A.G.L. k 1.69 1.71 1.80 1.67 1.50 1.71 1.67 1.68 1.66 1.54 1.54 1.75 100m A [m/s) 5.6 5.3 5.4 5.2 5.4 5.2 5.0 5.3 5.6 5.0 4.7 5.2 A.G.L. k 1.71 1.72 1.81 1.68 1.51 1.71 1.69 1.69 1.67 1.55 1.55 1.76 150m A [m/s) 6.0 5.7 5.8 5.5 5.7 5.5 5.3 5.7 5.9 5.3 5.0 5.5 A.G.L. k 1.59 1.60 1.69 1.57 1.40 1.60 1.57 1.58 1.56 1.45 1.45 1.65 200m A [m/s) 6.3 5.9 6.1 5.8 6.0 5.8 5.6 5.9 6.2 5.5 5.2 5.8 A.G.L. k 1.51 1.52 1.60 1.48 1.33 1.51 1.49 1.49 1.47 1.37 1.37 1.56 158 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. 159 Site 8: Sanghar, Sindh Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 9.9 8.6 4.2 2.0 1.5 2.4 7.4 37.0 11.7 4.4 4.2 6.6 50m A [m/s) 6.2 6.7 5.5 3.8 3.1 3.7 6.4 9.1 8.3 5.5 4.6 5.4 A.G.L. k 2.20 2.61 1.85 1.40 1.10 1.47 2.54 3.52 2.70 1.63 1.69 1.81 100m A [m/s) 7.3 7.9 6.5 4.6 3.8 4.4 7.6 10.8 9.8 6.5 5.5 6.4 A.G.L. k 2.35 2.79 1.97 1.49 1.17 1.57 2.71 3.76 2.88 1.75 1.81 1.93 150m A [m/s) 8.2 8.8 7.3 5.1 4.2 5.0 8.5 12.1 11.0 7.3 6.1 7.2 A.G.L. k 2.29 2.72 1.93 1.46 1.15 1.54 2.65 3.67 2.81 1.71 1.77 1.89 200m A [m/s) 8.9 9.6 7.9 5.6 4.5 5.4 9.3 13.2 12.0 8.0 6.7 7.9 A.G.L. k 2.25 2.67 1.89 1.44 1.13 1.51 2.60 3.61 2.76 1.68 1.74 1.86 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. 160 Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. Site 9: Umerkot, Sindh Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 50m A [m/s) 6.0 5.8 4.7 3.7 3.3 4.1 5.5 8.6 7.6 5.5 4.8 5.6 A.G.L. k 2.37 2.40 2.07 1.64 1.28 1.97 1.44 3.13 2.97 2.03 1.80 2.10 100m A [m/s) 7.2 6.9 5.8 4.5 4.1 5.0 6.6 10.3 9.2 6.6 5.8 6.8 A.G.L. k 2.58 2.63 2.24 1.79 1.40 2.15 1.57 3.42 3.24 2.22 1.96 2.29 150m A [m/s) 8.1 7.9 6.4 5.1 4.6 5.6 7.5 11.7 10.4 7.5 6.6 7.7 A.G.L. k 2.56 2.61 2.23 1.78 1.38 2.14 2.56 3.39 3.22 2.20 1.95 2.28 200m A [m/s) 9.0 8.7 7.1 5.6 5.1 6.2 8.3 13.0 11.5 8.3 7.3 8.5 A.G.L. k 2.55 2.59 2.22 1.76 1.38 2.13 1.55 3.37 3.20 2.19 1.94 2.26 161 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. 162 Site 10: Tando Ghulam Ali, Sindh Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 12.7 9.1 3.1 1.2 0.9 1.0 2.1 19.7 35.9 8.2 2.6 3.4 50m A [m/s) 7.1 5.8 4.3 3.0 2.5 3.2 4.4 7.5 8.7 7.3 5.5 5.6 A.G.L. k 2.72 1.98 1.53 1.31 1.16 1.53 1.83 3.52 3.21 2.92 1.98 2.23 100m A [m/s) 8.5 7.0 5.2 3.7 3.1 3.9 5.3 9.0 10.4 8.7 6.5 6.7 A.G.L. k 2.95 2.14 1.66 1.42 1.25 1.65 1.98 3.81 3.47 3.17 2.14 2.42 150m A [m/s) 9.6 7.9 5.8 4.2 3.5 4.4 6.0 10.2 11.8 9.8 7.4 7.6 A.G.L. k 2.91 2.12 1.64 1.40 1.24 1.63 1.96 3.77 3.43 3.13 2.12 2.39 200m A [m/s) 10.5 8.7 6.4 4.6 3.8 4.8 6.5 11.2 13.0 10.8 8.1 8.3 A.G.L. k 2.88 2.10 1.62 1.38 1.22 1.62 1.94 3.73 3.40 3.10 2.10 2.37 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. 163 Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. Site 11: Gwadar, Balochistan Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 3.7 5.0 4.4 8.2 7.4 5.9 12.3 20.1 13.1 11.5 5.5 2.8 50m A [m/s) 5.6 6.2 3.4 4.0 4.2 4.0 5.3 6.1 5.1 5.0 3.8 3.2 A.G.L. k 1.81 1.88 1.64 2.13 2.30 2.14 2.70 2.77 1.98 1.89 1.65 1.48 100m A [m/s) 6.0 6.5 3.6 4.3 4.4 4.2 5.6 6.5 5.4 5.3 4.0 3.4 A.G.L. k 1.75 1.82 1.58 2.06 2.22 2.07 2.61 2.68 1.92 1.83 1.60 1.43 150m A [m/s) 6.1 6.7 3.6 4.4 4.5 4.3 5.8 6.7 5.5 5.4 4.1 3.4 A.G.L. k 1.68 1.75 1.53 1.98 2.14 1.99 2.51 2.58 1.84 1.76 1.53 1.38 200m A [m/s) 6.1 6.7 3.7 4.4 4.6 4.4 5.8 6.8 5.6 5.4 4.1 3.4 A.G.L. k 1.63 1.69 1.47 1.91 2.07 1.93 2.42 2.49 1.78 1.70 1.48 1.33 164 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. 165 Site 12: Sujawal, Sindh Roughness length: 0.03m Sectors 0° 30° 60° 90° 120° 150° 180° 210° 240° 270° 300° 330° frequency [%] 3.9 15.4 3.2 0.7 0.3 0.4 0.6 4.8 40.2 21.8 6.2 2.6 50m A [m/s) 5.7 7.8 5.0 3.5 3.6 3.4 4.4 7.0 8.4 8.1 6.9 4.9 A.G.L. k 1.69 3.19 1.75 1.49 1.42 1.53 1.78 3.49 3.12 3.08 2.91 1.87 100m A [m/s) 6.9 9.3 6.0 4.2 4.3 4.2 5.3 8.3 10.0 9.7 8.2 5.9 A.G.L. k 1.83 3.45 1.90 1.61 1.53 1.65 1.92 3.78 3.38 3.34 3.15 2.03 150m A [m/s) 7.8 10.5 6.7 4.8 4.9 4.7 6.0 9.4 1.13 10.9 9.3 6.6 A.G.L. k 1.81 3.41 1.87 1.59 1.51 1.63 1.90 3.74 3.34 3.30 3.11 2.00 200m A [m/s) 8.5 11.6 7.4 5.2 5.4 5.2 6.6 10.4 12.4 12.0 10.2 7.3 A.G.L. k 1.79 3.37 1.85 1.57 1.50 1.62 1.88 3.70 3.31 3.27 3.08 1.98 Roughness length: 0.03m / z: 50m A.G.L. Roughness length: 0.03m / z: 100m A.G.L. 166 Roughness length: 0.03m / z: 150m A.G.L. Roughness length: 0.03m / z: 200m A.G.L. 167 ANNEX L UNCERTAINTIES ASSOCIATED WITH AEP RESULTS – SINGLE TURBINE AT MAST LOCATION Site 1 Peshawar 2 Haripur 3 Chakri Turbine sensitivity 3.0 3.2 2.8 [%AEP / %WS] [WS] [% AEP] [WS] [% AEP] [WS] [% AEP] Wind measurements 1.7 5.3 2.8 8.8 2.3 6.3 Long-term extrapolation 0.0 0.0 0.0 0.0 3.5 9.9 Vertical extrapolation 0.0 0.0 0.0 0.0 0.0 0.0 Future wind variability 5.7 17.3 6.1 19.3 2.0 5.6 (20 years) Spatial variation 0.0 0.0 0.0 0.0 0.0 0.0 Power curve /// 10.0 /// 11.0 /// 9.5 Production losses /// 1.1 /// 1.1 /// 1.1 Combined Uncertainty /// 21.3 /// 24.6 /// 16.2 (20 years) Site 4 Quaidabad 5 Bahawalpur 6 Sadiqabad Turbine sensitivity 2.4 2.5 2.4 [%AEP / %WS] [WS] [% AEP] [WS] [% AEP] [WS] [% AEP] Wind measurements 1.6 3.9 2.3 5.5 2.3 5.5 Long-term extrapolation 3.6 8.6 3.5 8.6 3.2 7.9 Vertical extrapolation 0.0 0.0 0.0 0.0 0.0 0.0 Future wind variability 1.8 4.3 1.9 4.7 1.9 4.6 (20 years) Spatial variation 0.0 0.0 0.0 0.0 0.0 0.0 Power curve /// 9.8 /// 8.2 /// 8.5 Production losses /// 1.1 /// 1.1 /// 1.1 Combined Uncertainty /// 14.6 /// 14.0 /// 13.7 (20 years) Site 7 Quetta 8 Sanghar 9 Umerkot Turbine sensitivity 2.6 1.8 2.0 [%AEP / %WS] [WS] [% AEP] [WS] [% AEP] [WS] [% AEP] Wind measurements 1.7 4.5 1.7 3.1 1.7 3.4 Long-term extrapolation 3.3 8.8 1.7 3.2 1.8 3.5 Vertical extrapolation 0.7 1.8 0.0 0.0 0.0 0.0 Future wind variability 1.8 4.8 1.9 3.4 1.9 3.7 (20 years) Spatial variation 0.0 0.0 0.0 0.0 0.0 0.0 Power curve /// 10.3 /// 5.7 /// 6.0 Production losses /// 1.1 /// 1.1 /// 1.1 Combined Uncertainty /// 15.5 /// 8.1 /// 8.6 (20 years) 168 Site 10 Tando 11 Gwadar 12 Sujawal Turbine sensitivity 1.8 2.6 1.8 [%AEP / %WS] [WS] [% AEP] [WS] [% AEP] [WS] [% AEP] Wind measurements 1.7 3.1 1.7 4.5 1.7 3.1 Long-term extrapolation 1.6 2.9 2.9 7.5 1.6 2.9 Vertical extrapolation 0.0 0.0 0.0 0.0 0.0 0.0 Future wind variability 1.9 3.5 1.8 4.7 1.8 3.2 (20 years) Spatial variation 0.0 0.0 0.0 0.0 0.0 0.0 Power curve /// 5.3 /// 9.8 /// 5.2 Production losses /// 1.1 /// 1.1 /// 1.1 Combined Uncertainty /// 7.8 /// 14.2 /// 7.5 (20 years) 169 ANNEX M MEAN AIR DENSITY Site 1 Peshawar 2 Haripur 3 Chakri 4 Quaidabad 5 Bahawalpur 6 Sadiqabad Height AGL 80 80 80 80 80 80 [m] Air density at 1.081 1.124 1.120 1.151 1.157 1.163 hub height [kg/m3] Site 7 Quetta 8 Sanghar 9 Umerkot 10 Tando 11 Gwadar 12 Sujawal Height AGL 64 80 80 80 80 80 [m] Air density at 1.007 1.160 1.154 1.167 1.154 1.167 hub height [kg/m3] 170 ANNEX N POTENTIAL LAYOUT It should be noted that a potential wind farm layout with a capacity of 30.75 MW has been studied for Site 12: Sujawal, Sindh considering 10 same type 3 MW generic turbines. Figure 72 illustrates the best practice with a wind turbine spacing of at least seven rotor diameters along the prevailing wind directions and four rotor diameters across. Figure 72: Wind farm layout, ellipses indicating the recommended spacing Uncertainties associated with energy production results were then evaluated. They equal 14.4 % in total for a period of 20 years, for this specific wind farm configuration and break down as follows: Configuration @80m Wind measurements 3.6 Long-term extrapolation 3.4 Vertical extrapolation 0.0 Future wind variability (20 years) 3.7 Spatial variation 11.7 Power curve 5.1 Production losses 2.4 Combined uncertainty (20 years) 14.4 The expected AEP and other energy production figures are presented below. The following results are provided: • Mean wind speed: corresponds to the lowest and highest mean wind speeds expected at the location and hub height of wind turbines. 171 • Gross energy production: corresponds to the theoretically recoverable annual energy production at the outlet side of the generator, without production losses. • Energy production losses: corresponds to wake losses, unavailability losses, performance losses, electrical losses, environmental losses and curtailment losses (if any). • Net energy production (AEP): corresponds to the annual energy production expected to be delivered to the grid (taking into account all energy production losses). • Net full load equivalent hours: is the amount of time it would take for the wind farm to yield its annual production if it was able to constantly produce at full load. • Net capacity factor: is the net full load equivalent hours divided by the total number of hours in a year. It represents the usage of the installed capacity. Configuration @80m Mean wind speed [m/s] 7.3 - 7.4 Gross energy production [MWh/y] 109,336 Wake losses [%] 5.6 Curtailment losses [%] 0.0 Other losses [%] 6.1 Total energy production losses [%] 11.4 Net energy production (AEP) [MWh/y] 96,892 Net full load equivalent hours [h/y] 3,151 Net capacity factor [%] 35.9 AEPs over 1, 10, 15 and 20 year periods exceeded with probabilities of 50% (P50) to 95% (P95) are provided below. Configuration @80m 1 year AEP (P50) [MWh/y] 96,892 AEP (P75) [MWh/y] 84,599 AEP (P90) [MWh/y] 73,535 AEP (P95) [MWh/y] 66,914 10 years AEP (P50) [MWh/y] 96,892 AEP (P75) [MWh/y] 87,246 AEP (P90) [MWh/y] 78,564 AEP (P95) [MWh/y] 73,368 15 years AEP (P50) [MWh/y] 96,892 AEP (P75) [MWh/y] 87,441 AEP (P90) [MWh/y] 78,936 AEP (P95) [MWh/y] 73,846 20 years AEP (P50) [MWh/y] 96,892 AEP (P75) [MWh/y] 87,490 AEP (P90) [MWh/y] 79,027 AEP (P95) [MWh/y] 73,963 172 ANNEX O CALIBRATION CERTIFICATES Individual calibration certificates for the sensors for all masts could be found in a separate document (annex file to the installation report), available for access on the World Bank/ESMAP online data repository energydata.info. 173