JointUNDP/World Bank Energy Sector Management Assistance Program KEPTYA ENERGY EFFICIENCY IN THE TEA INDUSTRY V O L W I1 - APPENDICES A REPORT BY CONSULTANTS MAY 1987 Household Energy Division Energy Department The World Bank Washington, D.C. 20433 TABLE OF CONTENTS. APPENDICES Page ..................................A-1 Glossary of Terms ................B-1 ...C-1 Tea Industry Energy Usage Data Base Biomass as an Energy Source for the Tea Industry .......................D-1 Supplies and Prices of Fuels .....................E-1 Indigenous and Byproduct Fuels Initial Evaluation of Projects.....................F-1 ..........G-1 Biomass and Wood Firing Technology Review .......................H-1 Oil Firing Technology Review Operation and Maintenance of Steam Systems.........1-1 Tea Factory Audit Program..........................J-1-1 ........................5-2-1 ..........................5-3-1 Energy Survey Questionnaire Factory Officers' Seminar APPENDIX A Q I O C j S A R Y O F m Units UsedthmqhmtthisreportamfarmtotheSystanInternatimale (SI using the followirKJ stadard -ti= and conversicm factors. Whemotherunitsamusedtheyareexp~inthetext. ax Gwemment of Kenya 2wiB Australian Develapnent Assistance Bureau World Bank t Intematimal Bank for ~ t n l c t i c m and DeveloEKlent) IEA Inlxrnatianal Energy Agency IFC Intematid Finance Corporation IMF Inmticmal I'4m&aq F'urd Km?i Kerrya lba -1-t Authority Mt3l3lD Ministry of Eneryy ard Regimal Developnent UNDP Wted Nations Develapnent Frogramne Ma- Wagner Pty Ltd in associaticm w i t h l&rz & mllan & Partnem, mmnB!cDmmll and Car9arry, Cmpem&LybagIldW D Scott tf-lyWD Scott & Go), F a r e s t r y = c a l M c e e P t y L t d , ard I.A. Gilnwxlr, University of Can-, N.Z. RENJY savings target achievable, based on *ilmdiatelywithminimnn tio on measures whichcouldbetaken engineer- - inprt arri, in general, minimrm capital investment. Typically, such measures would showa sinple payback of under ane year and would include basic house-keep- and v t i n g and mintenance pxactices. EImergy savings target achievable based on consemationmeasures which wuld be taken, w i t h appmpriate planning and engineering input and the expectation of capital investment in new or inpmved plant or processes or a mre eccmanic energy source, and which wmld give an acceptable ecananic and financial rate of xWxrn, (i.e.gmater than 15% real). Intensity aLe ratio of meqy ccxmqtdon to pmduction, typically expressed in units of gigajoules per tume of pmduct (GJ/t). Also referred to as specific energy cmmmphon. Gross Specific m, maaswlsd in GJ/t. GSE is used thra;lghout this report. Net Specific Eherqy my be abtained by mbtracting the latent heat of vaporisation of mter f d in the canbustion maction, ~chisafmctionofthehydmgencontent of the llnhlmt fuel. Typically, the ratio Net/- Specific Jhergy is about 0.94 for RFD and 0.98 for coal. Financial Year (1 July 1984 - 30 June 1985) . IWisture amtent basis Any arrSxlstible woody material, e.g. fran trees, shrubs, coffee husks, etc. wnai Iad factor Ratio of average electricity &naxd to xfmxhmdarrand. Maximnn dlnand, elcpressed aa m,is the highest rate of electricity emxgy usage in any twenty minute i n s during the meter mading period, which is typically 30 days. Maji Abiria Ebisture entering w i t h dryer inlet air anl causing a loss of efficiency, litarally maning in Kiswahili "a passerqer". (Nornvil English usage - "tranp nmhtum". Hem Abiria Air leakage into a -08 which serves no plrpose ard causes a 106s of efficiency. Made tea, i.e. fiml pmduct. Green leaf, as plucked anl sold to the factory, Liquified Petroleum Gas Gasoline (petrol) Imkstrial Diesel ail Refinery Fuel Oil Gas Oil Fluidised bed dryer - units A lhps (electricalcurrent) v volts telectrical pmssure) KV W Watt (electricpower) units kil-tts = 10% kilovolt Pvnp kilovolt Pvnp (xeactive) kilcwatt haurs = 3.6 KJ Mwh bar bar (pressure) kilopascals (pmssure) Joule (energy) = 1W.s kilojoule = lo3 Joules Megajoule = lo6 ~oules Gigajde = 109 Joules = 278 kMh 12 -joule = 10 Joules kilogram (mass) to~e lo3 kg = Litre (velum) kilolitre = 103 L = Im3 Megalitre = 106 L = 1000 m3 centipoise (viscosity) seccad (time) min minute = 60 s h hour=6Omin annum (year)= 8760 h rtletre square- cubic metre units km km2 ha Oc Centigrade tomes of oil equivalent ( =41.868 GJ) IEA Standard ixxmes of axl equivalent ( = 23.446GJ) IEA standard l#m.Lm Kglya Shillings of ~ ~ l shillings = y a lo3 X S ~ Millions of Kenya Shillings = 106 XSh Wted States Dollars Australian Dollars Cowersicm Factors- Imperial to SI (metric) Imperial Conversion Qgantitv Meaninq Units limerial to SI Area Square f- f t 2 0.092 903 m2 Density poundspercu.ft. lb/ft3 16.0185ky.m-3 Energy British 'Ihermal Btu 1.055 06 )rJ unit F l w gallon per minute gpn 0.075 768 L.s-' Heat value Ertu per cu.ft. Btu/ft3 37.258 9 W.m-3 B t U PerP m d B t u h 2326 ~.kg-' Heat f l w Camrersim Factors - Inmerial to SI (metric1 (Cont'dl -%lth inch in 0.0254 m foot ft 0.304 8m mss ton T 1016.05 kg m s flw p o u n d s p e r ~ lb/h 0.000 126 kg..-' Pressure bar bar 100 kPa poundspersq. in. Ib/sq.in 6894.76 Pa inch of mter "V3 248.642 Pa inch of mercury "Hg 3386.39 Pa Refrigeration capacity degme Fahrenheit % C273 + 0.556 (F-3211% cubic foot ft3 0.028 316 8m3 cubic yard f13 0.764 555 m3 gallon @ 4.546 Iitres Velocity foatpersec ft/s 0.304 8 m.s-' miles per bur mph 1.609 34 lun.h-' mw Grc#isSpecific Dsnsity Tbmes of Oil m 3 Wvalent W h toe Codl 1130 Charcaal NA *Wood 20% mmb 496 (solid) 30% " 527 (solid) 40% " 565 (solid) 50% " 609 (solid) 60% " 661 (&dl 70% " 722 (solid) Animal waste NA Bagasse (30%moisture) 115 Coffee husk NA Cotton seed husk NA Flax NA Jute sticks NA Leather shavings Iw Maize cobs NA Municipal waste NA Nut shells m Palm shell NA Palm shell & fibre NA Palm stalk NA R i c e husks NA Spent tan bark NA straw NA s m f l husk ~ NA Liquified Petroleum Gas (rn)- Propane Petrol (a) -/jet fuel (DPK) Gas oil (0) Incfustrial diesel (ID01 Fuel oil (RFD) Electricity 1 MWh * Note 1 cubic metre stacked d = 0.6 cubic : solid velum. APPEM)M B TEA 'INDUSIRY ENQCX USAGE DATABASE 1. DESIGN OF DATABASE In order to establish the pattern of energy ammnption in the KlDA and to develop a predict energy use d l , for Kenyan Tea Industry Eneryy Usage Database was prepared and is repruduced in this Appndix. The Database q i l e s available FY 1985 data cn energy use in the Kenya Tba Develogmnt Authority. Data for the audited tea factories has been obtained fnrn the 19 Energy Audit Reports which have been bowci and issued irdividually. Data for nan audited factories has been obtained £ranthereturned E S t e r g y S u r v e y Q u e s t i ~ . The D a a was prepared -1- the internationally known mtus 1-2-3 spreadsheet pmgmtm, using the pmject microcanpttes. So£- for the Database has been handed over to the PDERD for this purpose and it is -ed that such surveys amtinue to be campleted annually by the PDERD. Sane wrk d d be needed on survey form design, but this again is sixply a matter of progressive developnent. In acktition, the Database was extended using the I&us capabilities to include energy usage projectims up to the year 1995, the tine frame of the present study. 'Ihe overall objective m s to detexnine not c d y the existing pattern of energy camqkion in the ICIDA factories, but also the likely outcuw of an effective eneqy camemzition pmgmme, based m the results obtain& fromthe 19 -audits carried out and extrapolating them intelligently to cover all KKA factories. Notes are included below to indicate the -on and specific plrpose of each of the individual qmdshet files which fran the Database. As nated in this A g p d h , the Databse pawides infomtion on: Factory Ram, Province, Plant Installed, Pmduction and -=sly-,& En=m' Source - Type, cuet, -1 cbnm+im. 'an, Target m d a M n d projections for=& energy -wen= basedon the 1985 Ehergy Survey ~ t i m t d m e .With this basis, the mrgy amservatim ptential can be factored in as appmpriate for each individual factory. In ackiitim, the amemation potential in industry can be ex~pessed,for shplicity, in term3 of achievable targets for energy use per d t of gmdwtion, typically G J / m , and irrrhex usually referredtoas energy intensity. The targets are defined a t two levels, thus: Target 1 (sh0rtTkn-n) savings target, achievable, based an comervation nreasures which d d be taken almost imnediately with minimnn engineer* input ad, in general, minimnn capital investment. Typically, such measures would show a sinple payback of under me year and d include basic and oormct operating ard ~~practices. Target 2 (Img Term) Energy savings target, achievable, based an cmlsawatian measures which could be taken, with appmpriate planning and enginedng i n p t and the exgectation of capital investment in new or imprwed plant or Epocesses or a mre ecananic energy caurse, and which mdd give an acceptable ecmanic ard financial rate of return, (i.e. greater than 15% real). Targets were set in acmrdance with: the h t s of the audLt programne as m t s d in the 19 Energy Audit Reports; intermationally accepted stardards for the tea industry t suitably mdified for IGzyan d t i m s1, ard the Consultants' experience in similar mk. In this way, the Lla- and taxget settings can be used to establish existing and future energy cmmmpticms and to cmpute energy comematian ard cost saving potential for the factories for a wide range of scenarios. TheKenya~~tryEheryyUsageDa~catprisesamnrS3erofdata files, each of vhich serves a distinct plrpose. The I.lotss in the sections following describe the cantent and pupose of each of the following files: File File Title Energy Intensityby Factory by Ehergy Source - FY 1985 Eneqy Intensity: Curmnt and Fa- Perfonnmc= Fneqy Costs: Cwrent and Future Factory P e r f 0 ~ c e This file aggregates energy m a n for FY 1985 for a l l IQWi factories. It sets out RIlX energy -on by energy sauroe, giving forecast -on predictions for 1990 and 1995. The growth rate applied to 1985 ammqtion is takenas 3% for existing factories. Overlying this basic growth rate, pmvision is mde for tm energy mmagemnt factors to es-te the effects of the application of energy m m a w t initiatives as reported in each individual Bwqy Audit Report. The rate of energy perfonwmoe hpmwmmt believed achievable by KlmA factories is a specific emqy (GJ/t) reductian of 3% p.a. werdLl for 5 years, i.e. f r m FY 1986 - FY 1990 inclusive, as the essentially Target 1 ptential is realised. A reduction of 2% p.a. is then a l l a d for the five years FY 1991 - FY 1995 as the residual Target 1 and new Target 2 initiates are taken up. Shaild experience show that these figures are either too ambitious or too cmsematiue, then the factors can readily be changed to create a new and mre u s t i c pmxiiction on the Database. 'IILis file includes a l l data available to the M t a n t s on FY 1985 energy mmmqtian for KTIlA factories as obtahed fran the Energy Survey -tiomaims mtumed by all factories. A l lother files extract their data f r m this file. This file includes factmy m d e tea w a n figures for FY 1985 and hence derived energy intensity, firstly for each energy source, them ove.ral1 for the factory. RLis enables hmdiate carparisan between factories having the same plant arranganent, for exa~tplein electricity '=age KlDZi 5 - Ene.rqy Intensity- Current and Future Factory Performance This file, wfiich uses o u w t data frun the previous files, sets out the actual FY 1985 energy intensity, ht now includes the target energy intensities (Target 1 and Target 2) arising £runthe Consultants' mrk as given in the -Audit Reports. From these figures, the target energy rductioc18 in each factory a m obtained. For the a u i i t d factories the taryet reductions are based pthe ohemations of the Audit lkams as reported. For the nm-auiited factories, the consultants have taken the assmption that oarparable target perfomances for arrparable plant outpts, arrfiguraticm axrl fuel type can be achieved. - TMs file is based upon the input data for file K2TR 5, canverted to a oost base. For simplicity of file d p l l a t i o n , average cost figures based an field advised prices for each energy source adopt&. These averages were: Electricity 0.72 K S h m 200 m/GJ - For the m-awlited factories, Target (TI) energy cost saving8 based an the average cost (see above) of the main fuel saurce and the energy intensity reductim in F i l e KIW4 5. Tarpet 2 (T2)savings based on the asmtpti~lsof File RKlX 5. This file aggmgates the anticipated financial savings for all KKlA factories as a camquence 0fthe~ManagfmentProgramrre. RBlYA : mRGY EFPICIgWCT IN THB TEA INBETRY MI% : 22-Jul-86 MBRCY CONSWTIOW (TERAJWB) BY BWEBCY SOURCE 1985 TO 1995 F I U : KTM2 ------------------------------I----------------------------------------------------------- Q(geGYSMK3 : 1985 1990 1 9 9 5 1 ~ I U ~ ~ g l l E B G Y ~ : TJ TJ TJ : (Z PA) :1986-90 1991-95 I I I I I I I I 1 I I I I I I ID0 ( 24.8 24.8 26.1 : 3.0 -3.0 -2.0 BFO 1 1057.0 1057.0 1110.9 1 3.0 1 -3.0 -2.0 I I I I I I EtECrBICITY : 12S.O 125.0 131.4 : 3.0 1 -3.0 -2.0 I I I I I I WOOD 1 234.1 234.1 246.1 : 3.0 1 -3.0 -2.0 I I I I I I ----------------------------- ----me--------------------:------------- ------------------ TOTAL : 1440.9 1440.9 1514.4 : I I I I I I I I MII: 22-Jul-86 FIU: KTDA3 R l D A FACTORY DISTRICT PLMI AUDIT I#) BPO aECl WOO TOTAL -------------------------------------------------------------------------------------------------- Yh TJ/a TJ/a TJ/a TJ/a TJIa MANIA neu 25.3 3.5 28.7 RWRIBI mu 11.9 30.2 SIC 42.2 RAPRm rnI(110 41.4 W/S 41.4 KmR KERIQW) 0.1 9 . 7 4.6 58.4 LITEII RWIm 0.1 32.3 3.8 36.2 KXmxIR WRIQlO 32.0 4.6 36.6 r n A T m 1 0 0.1 44.8 3.8 40.6 W E Ruml 0.1 20.2 3.2 23.4 KmM Km 0.1 20.8 2.0 22.8 MTMRA I I I m 0.1 43.7 7.2 51.0 lWTA Ruml 0.1 22.9 4.0 26.9 WAITA KIBIWYAGA 6.4 3.5 49.6 59.5 KIllUlYB KIRIWYAGA 0.2 20.2 4.0 24.4 M I T A HRIWYAGA 0.1 4.2 2.4 30.9 37.6 KBIRICO KISII 0.4 10.9 3.0 44.3 K I W KISII 0.1 29.8 2.8 32.7 MAlUQia KISII 24.8 2.8 27.6 rnmm KISII 0.3 29.3 2.1 31.7 nYAnsIOIIGO KISII 0.1 19.8 3.6 23.5 OGm KISII 17.1 4.3 21.4 SUIGANYI KIWI 27.5 3.8 31.3 KlnBE RISII 24.2 6.1 30.2 G I r m m 2.2 56.6 58.9 1 m rI mu 33.8 3.8 37.5 RIBalI mu 9.5 25.9 SIC 35.4 KIm mu 0.1 20.6 3.4 24.1 arwmu Mnw'A 0.1 37.2 5.1 42.1 CInwm W ' A 0.2 33.9 3.5 37.6 IKWI W ' A 0.2 40.5 3.1 51.7 r n A I I I 1 ' W ' A 0.2 31.0 4.4 35.6 M m I M n w'A 0.2 31.5 MIS 31.6 WJU)III I M n w'A 0.1 31.4 1.3 32.8 mBur NWDI 0.1 3.0 40.7 43.7 MIWCA WYERI 0.2 49.6 3.9 53.7 a m 1 mwI 0.1 2.2 28.9 31.3 CITUCI mERI 25.5 4.0 29.5 IBIAINI m 1 0.1 41.6 4.1 45.8 NDIM lMRI 0.2 35.1 3.7 39.0 MATI m 1 2.4 27.4 29.8 LEMD : Y = Y ~ B 0 = O i l Firing S = Stem Htrtu HIS = kta not supplied - energy W= No U = W Firing A = Ah Hcrter corumptionaredcrivedfm S/G = Self Generated ertivtal energy i n m i t i e s = ha1 fuel boiler8 (1986) RTDA FACTORY DISTRICT AUDIT OUTPUT IDO RFO ELECT YOOD TOTAL Y/11 tur GJ/t GJ/t GJ/t GJ/t GJ/t .................................................................................................... MRIGANIA m RUKURIRI MBU WROROS KERIcm KAPSET llERICH0 LITEIN KERICHO t!o@xOsIEK r n I W TEGAT KERICHO KACWE KIW KAmAA RImU HATAARA K I r n TKETA K I r n KANGAITA KIRINYAGA K I r n KIRINYAGA 'IHIMAITA KIRINYAGA KEBIRIGO RISII K I M KISII N Y M E KISII IIYANKOBA KISII NYANSIOWCO KISII OCMBO KISII SAllCANYI KISII TOneE KISII G I m HERU I r n I ERU KIEGOI HERU KINOR0 HERU GANHGURU IluRANG'A GIMAneO WRANG'A IKUHBI NURANG'A KAmMYAINI MRANC'A WOnBORI HLlRm'A Iim m ' A MEBUT NAND1 CHINCA NYERI GATHUTHI m 1 GITUGI mull IRIAINI NYERI NDIHA mER1 RACATI NYER I ..................................................................................................... TOTAL / WEIGHTED AVERAGE 63072 --- --- --- --- 22.8 ------------------------------------------------iw----------------------------------------------..----- LEGEM) : Y = Yes N/S = Data Not Supplied - Energy intensities are esthates only R = N o SIC = Self Cewrated DATE: 23-Jule FILE: KTMS ...................................................................... aEIbGY WTm EnsaCY IWTENSIrr KTDA FACTORY DISTRICT USAGE W TU 1985 T A N K 1 TARGKT 2 TJ usr GJ/m GJ/m GJ/m MRlCANIA MBU RUKURIRI B(BU KAPROR(XS m 1 w KAm m 1 w LITEIN m 1 w KniOCOSIER KgRICHO TEAT m 1 w KACWE K I A W RAneAA K I m MTMRA K I r n META K I A m KMCAITA KIRINYllCA KIRINYE KIRINYAGA THLMAITA KIRINYhGA KEBIRIGO KISII KIAl#mM KISII NYAMCHE KISII NrANKOBA KISII NYANSICmGO KISII a m KISII SAWCANYI KISII nmBE KISII GI- mu I r n I mu KIECOI I R U KIlloR0 mu GANWCUBU MJRANG'A GIniMm HURANG'A IKUneI MIRAKi'A K m m A I N I tAIRMW;'A n A K m 1 lwwlG'A MTUWU MlRAWG'A CHEBUT HAND1 W N G A NYERI GATHUM1 NYERI GINCI m m I IRIAINI mERI NDIW m m I RAGATI NYERI REmA : RmGY EFFICIMCY IN THE TEA INDMRY DAlE: 23-Ju1-86 ENERGY COSTS : CURBMT AND PUNBE FACTORY PEMWANCH PILE: KfM6 ___________________------------------------------------------------------------------------------------------ EWERGY OVIPUT EmGYCOST MEBGY COST SAVIrmS KIDA FACTORY DISTRICT UME MDl TEA CURRMT TARGET 1 TARC TARGET 1 TARGET 2 TJ t HIIT kRSlk Sl/lr MI SH/W Z RSHIkgHT Z WNGMIA BIBU RUKURIRI MBU IUPROROS RERICH0 WSET RERICH0 LITEIN RERICH0 NOCOGOSIW KERICHO T E A T RERICH0 KAGWE RImu KAlJBM KIMU m r m KIAlIBU THETA KIMBU KAWCAITA KIRINYACA R I r n KIUINYAGA nflNAITA RIRINYAGA REBIRIGO RISII K I W KISII NYAHACHE KISII NYANKOBA RISII NYANSIONCO KISII a m KISII SA)ICANYI KISII W E KISII GITHONGO KERU m 1 MRU KIEGOI mu KIWORO mu GATUNGURU MIRANG'A GITHAtlBO MJRANG'A IKunBI lAlIWlC'A KAmMYAINI H U r u G ' A M10n83KI nuRANG'A NJUW fmAM'A CnEBUT NAND1 CHIKA m 1 G A m 1 mERI GITUCI NYERI IRIAINI NHRI NDIM m 1 RACATI NERI Prepared by: Forestry !kchical Senrices Fty Lid BIOMASS AS AN ENERGY SOURCE FOR THE TEA INDUSTRY TABLE OF CONTENTS PAGE EXECUTIVE SUMMARY INTRODUCTION AND TERMS OF REFERENCE 2.1 Introduction 2.2 Acknowledgements 2.3 Terms o f Reference FLIELWOOD SUPPLY AND DEMAND OVERVIEW 3.1 Forest Resources 3.2 Fuelwood Supply and Demand PRESENT BIOMASS USE - KTDA FACTORIES 4.1 Green Leaf Supply and Production 4.2 Current Fuel Use by KTDA Factories PRESENT AND POTENTIAL FUELWOOD SUPPLY FROM FOREST DEPARTMENT 5.1 Supply t o Present Fuelwood Using Factories 5.2 Future KTDA Fuelwood Supply Perspective 5.3 Cost o f Fuelwood Production 5.4 I n s t i t u t i o n a l Matters PRESENT FUELWOOD USE AND SUPPLY -PRIVATE ESTATES 6.1 Fuelwood Resource 6.2 Cost o f Fuelwood Production - Standing Tree 6.3 Opportunity Cost o f Fuelwood Production POTENTIAL OF FUELWOOD SUPPLY FROM UNUSED PUBLIC LANDS 7.1 Opportunities 7.2 Roadside Pla n t ings 7.3 Other Pub1i c Lands 7.4 Cost o f Fuelwood Production 7.5 I n s t i t u t i o n a l Matters and Constraints POTENTIAL FOR FUELWOOD SUPPLY FROM SMALLHOLDERS 8.1 Smallholder Land use i n the Highlands 8.2 Present Biomass Resource 8.3 Potential Resources f o r KTDA 8.4 Cost and Returns of Fuelwood Production - Standing Tree 8.5 Opportunity Cost o f Fuelwood Production 8.6 I n s t i t u t i o n a l Matters and Constraints 9.0 BIOMASS FROM CROP RESIDUES 9.1 F o r e s t R e s i d u e s o f F o r e s t D e p a r t m e n t 9.2 O t h e r R e s i d u e s 9.3 C o s t o f C r o p R e s i d u e s 10.0 SUMMARY OF BIOMASS RESOURCE OPTIONS AND COSTS 10.1 P r i c i n g o f B i o m a s s E n e r g y 10.2 C o n c l u s i o n s on S u p p l y O p t i o n s f o r KTDA F a c t o r i e s ANNEXURE C1 REFERENCES ANNEXURE C2 MODELS FOR FUELWOOD PRODUCTION ANNEXURE C3 BIOMASS DEMONSTRATION PROJECT ANNEXURE C4 BIOMASS SPECIES AND PROVENANCES MAP 1 KENYA SMALLHOLDER TEA GROWING AREAS & KTDA FACTORIES C-57 MAP 2 FOREST RESOURCES OF KENYA C-58 MAP 3 A EAST KENYA ADMINISTRATIVE BOUNDARIES C-59 MAP 3 8 CENTRAL KENYA ADMINISTRATIVE BOUNDARIES C-60 MAP 3C WEST KENYA ADMINISTRATIVE BOUNDARIES C-61 1.O EXECUTIVE SUMMARY 1.1 Woody biomass supplies 95% a l l biomass energy i n Kenya and 71% (as of 1980) o f the t o t a l energy demand o f a l l sectors. The dominant fuelwood sector i s domestic. 1.2 About 47% o f the national fuelwood supply comes from farm trees and bushes, 28% from f o r e s t trees and 25% from rangelands. 1.3 There i s a fuelwood d e f i c i t i n the high population density areas o f high a g r i c u l t u r a l potential, notably i n Western Kenya where tea production occurs and where a number of KTDA's factories are located. 1.4 The KTDA acts as managing agent f o r 39 tea factories which produced nearly 64 000 tonnes o f made tea i n 1984/85 from about 55 000 ha of tea grown by 150 000 smal lholders. This represents around 45% of Kenya's t o t a l tea production. A separate company owns each factory w i t h small holders being the main sharehol ders. The KTDA smallholder scheme which has been operating for some 20 years i s regarded as being among the nlost successful i n the world. 1.5 Subject t o a v a i l a b i l i t y o f supply, KTDA p o l i c y i s progressively t o convert a number of factories t o fuelwood for the drying process, d i spl acing imported o i 1. Six factories presently use wood exclusively and one predominantly. 1.6 If a11 f a c t o r i e s were hypothetically t o convert t o fuelwood, fuelwood usage a t 1985 production levels would be about 100 000 tonnes/y assuming use o f wood o f 527 kg/cu.m density a t 30% mcwb a t 14.2 GJ/t gross s p e c i f i c energy (GSE). Current energy i n t e n s i t i e s o f around 1.5 t of wood per tonne of made tea (22-23 GJ/t) could however be reduced t o avoid 1.0 t o f wood (14-15 GJ/t) w i t h improved combustion processes w i t h a corresponding drop i n the potential fuelwood demand t o about 65 000 t/y. 1.7 I n 1979 the Forest Department i s understood t o have promised adequate supplies o f fuelwood for 22 KTDA factories. By 1985 fuelwood supply t o the present 7 fuelwood operated factories had become uncertain, and it i s believed no confirmation has been received since then from the Forest Department as t o the l i k e l i h o o d of meeting the 1979 supply commitment. KTDA management sees the Forest Department as i t s primary source o f supply, and i s thus concerned t o see that supply commitments are honoured. 1.8 The 1985 market prices o f fuelwood energy were estimated from v i s i t s t o four f a c t o r i e s t o be about 15.1 KSh/GJ from native forest species, t o about 20.5 KSh/GJ using plantation Eucalyptus sal igna. This compared with the 1985 recorded cost of energy from f u e l o i l a t 77.6 KSh/GJ and 82.8 KSh/GJ a t f a c t o r i e s east and west o f the R i f t Valley respectively. However it i s required that the economic price for fuelwood i s higher than the harket p r i c e while by 1986 the reverse was t r u e f o r f u e l o i l . 1.9 A series of five, l a t e r six, options were examined f o r enhancement o f fuelwood supply t o KTDA w i t h i n the l i m i t e d time constraints o f the b r i e f . 8 These were: . . Forest Department lands. Unused roadside areas and other public lands available f o r fuelwood. . Future surplus i n private tea estate fuelwood plantations with improved . .. factory energy e ff ic iency Small holder wood1o t potentia1 . Crop residues. The Consultant was subsequently advised t h a t sawdust was available as a potenti a1 biomass resource i n D i s t r i c t s which have substantia1 sawmi1l i n g operation. Sawdust i s most appropriate fuel, but requires appropriate combustion equipment o f a type not now installed. 1.10 Combining Forest Department, pub1i c lands and smal 'Iholder models, the cost o f fuelwood production (royalty) i s 29%, harvesting 25% and transport 46% a t not more than lOkm distance from factory, making delivered cost very sensitive t o transport cost. 1.11 The f i r s t and most a t t r a c t i v e fuelwood supply p r i o r i t y i s from Forest Department lands. If existing fuelwood plantations were f u l l y productive they could produce two t h i r d s (66%) o f a l l KTDA factory needs if a l l y i e l d was sold t o KTDA. The Consultant has assumed 50% a1location but confirmation i s needed. Yield status of plantations should also be confirmed by inventory. Energy cost i s about 24 KSh/GJ a t lOkm distance. 1.12 Underuti 1ised pub1i c 1ands, if avai1able, have the potentia1 t o provide KTDA with say one quarter (25%) o f t o t a l estimated factory fuelwood needs, possibly a t say 35 KSh/GJ (nearly twice the cost o f other options if fencing i s used) but s t i l l h a l f that of o i l energy at 1985 prices. Innovative community involvement could reduce costs. However, there i s some doubt as t o community acceptability o f t h i s option which i s used successfu l y i n some developing countries. 1.13 Private tea estate factories have the potenti a1 t o reduce fuelwood use by about one t h i r d by increasing energy efficiency by improved combustion techniques. A surplus o f fuelwood i s however unlikely t o occur from t h i s source as tea would probably be planted on the land freed by fuelwood as a more p r o f i t a b l e crop, unless it were decreed otherwise. 1.14 From theoretical considerations and from recent land use survey data o f Bei j e r I n s t i t u t e the smal 'Iholder model appears capable o f producing more fuelwood than the needs o f a11 the KTDA factories given market and price guarantees. Energy cost o f about 23 KSh/GJ are estimated, being comparabl e w i t h the Forest Department model. Again some reservations are held local 1.y concerning comrnunity acceptabi 1ity o f t h i s option and careful testing would be needed. Furthermore market prices would need t o increase t o make t h i s option a t t r a c t i v e t o growers. 1.15 Some 6 000 tonnes/y o f r i c e husks are estimated t o be available, and could be u t i l ised by KTDA with appropriate furnace modifications a t a delivered cost of about 15 KSh/GJ, the apparent cost o f loading and transport. The avai 1abi 1ity and cost should be confirmed. 1.16 Some 33 000 tonnes/y of sawdust were advised t o the Consultants as being produced by a l l sawrni11s i n Kenya (Refer Appendix El. Sawdust i s a premiurn fuel and could a1so be delivered a t about 15 KSh/GJ. The a v a i l a b i l i t y and cost should also be confirmed t o those few factories located such t h a t a supply might be appropriate. 1.17 The Forest Department apparently has a conifer pulpwood surplus which i n quantity could be adequate f o r a l l KTDA factory needs, but probably a t say 35 KSh/GJ delivered cost. The a v a i l a b i l i t y and cost should agai n be confirmed . 1.18 I n sumnary, a fuelwood supply strategy f o r KTDA i s suggested as: Supply Source Approx t/y % ( I ) Probability o f Supply Preferred Supply Forest Department - Hardwood 25 - 50 000 25 - 50 High - Conifer pulpwood t o 140 000 t o 100 Low - medium Opportunity Supply ( t o be further evaluated) R i ce husks Low Sawdust Low - Medium Non-Preferred Supply Public lands 10 - 25 000 10 25 High Sma11ho1der 50 -100 000 50 -- 100 Medium - high - Note (11 Percent o f estimated KTDA needs of 100 000 t / y fuelwood. If efficiency gains can lead t o a reduced demand o f 65 000 t/y, the percentages are increased accordingly. Furthermore it has been c l e a r l y stated by KTDA t h a t by no means a l l factories, and p a r t i c u l a r l y those i n the fuelwood d e f i c i t areas of Western Kenya, are proposed for conversion from fuel o i l . 1.19 A P i l o t Biomass Demonstration Project i s proposed, as discussed i n Annexure C3, t o confirm the biomass data base of the Forest Department and t o carry out p i l o t scale fuelwood establishment t r i a l s . The estimated cost of the programme would be about $US 1.2 mi 11ion, spread over say 3 years. 1.20 Eucalyptus species, especi a1l y E. sal igna, and more recently E. grandis ( f o r the Tea Zone), have been used t o produce high wood and net energy yields as fuelwood. For the Coffee Zone Eucalyptus species from d r i e r parts o f Australia are suggested such as E. camaldulensis from North Queens1and. Other species are discussed i n Annexure C4. Species/ provenance t r i a l s should be i n i t i a t e d by KTDA i n co-operation with overseas agencies, important among which are Australian Centre f o r International Agricultural Research (ACIAR) which has begun a tree species t r i a l project with the Kenya Agricultural Research Institute. 2.0 INTRODUCTION AND TERMS OF REFERENCE 2.1 Introduction The Kenya Tea Development Authority i s a parastatal body established t o manage the processing o f t e a from the smal l holder sector. The small holder scheme has been very successful, with 150 000 farmers who delivered some 287 000 t o f green l e a f i n 1984/85 t o 39 K'rDA factories i n Kenya's f e r t i l e highlands, producing 63 700 t o f made tea. The area under tea i s 56 000 ha with an average tea p l o t size o f 0.38 ha per farm. The private tea estates produced nearly 55% o f Kenya's tea from around 24 000 ha with made tea y i e l d s averaging 2 800 kg/ha compared t o about 900 kg/ha f o r the KTDA smallholder, with the l a t t e r producing very high tea qua1ity. 32 KTDA factories are f u e l l e d with furnace o i l f o r drying and the remaining 7 factories operate on wood, with one of these using some o i l i n addition. KTDA's policy i s t o change a l l the o i l using factories progressively t o fuelwood as and when assured supplies become available, whether that be now o r i n the future. The Consultant's B r i e f reinforces t h i s policy. Fuelwood supplies t o date have come almost exclusively from the Forest Department. The KTDA considers t h i s as i t s 1ogi cal source of supply. Naturally it has concern for the continuing security of that supply to i t s 7 wood using factories, as well f o r as i t s poli c y t o convert the remaining factories t o wood as soon as possible. The KTDA sees the change from o i 1 t o cheaper ( a t 1985 prices) wood energy as a means o f decreasing i t s costs and conferring higher monetary benefits t o i t s growers. This has significance as each of the factories, i n the form o f a private company, are owned by some, but not a11, green l e a f growers and the KTDA acts as a managing agent f o r these 39 companies. The national viewpoint, which may differ somewhat from that o f the KTDA, i s t h a t the reduction of o i l imports which use scarce foreign exchange i s a desirable objective, and one which may i n part be achieved by the substitution o f indigenous fuels such as b imass, especial l y fuelwood. The absolute price o f fuelwood derived energy may not therefore be the major issue i n the national context, provided t h a t it bears a reasonable relationship with o i l energy t o maintain made tea prices a t a t t r a c t i v e p r i c e levels f o r export markets. 2.2 Ac know1edaements The Forestry Consultant worked closely with other study team members i n the engineering and economics disciplines as well as with the World Bank (ESMAP) Peri-Urban Fuelwood Study, and obtained significant assistance from s t a f f o f the KTDA, Forest Department and the Beij e r I n s t i t u t e Kenya Wood Fuel Project. Several o f the private estate tea companies such as Brooke Bond Kenya, African High1ands Produce Co Ltd, George W i 11iamson Engineering and Kaisugu L t d generously provided data, as also d i d BAT Kenya Ltd. The assistance of a11 the foregoing, additional t o those . recorded i n Volume I o f the Consultant's report, i s g r a t e f u l l y acknowl edged. 2.3 Terms o f Reference The Scope o f Work f o r the forestry and biomass specialist was out1ined as Task I11 o f the World Bank's A c t i v i t y I n i t i a t i o n Brief o f July 1985, and i s summarised as f o l lows: - . Establ ish whether existing forest resources o f the Forest Department could meet the additional KTDA requirement without undue deforestation consequences, and a t an economically acceptable cost. If natural forest reserves are not adequate, eval uate the technical feasi b i 1it y and economic v i abi 1ity o f fuelwood plantations. . Determine the land areas required t o provide fuelwood f o r the factories involved; examining 1and avai l a b i 1ity; 1and tenure; responsibility f o r establishment and management; legal and i n s t i t u t i o n a l arrangements; cost o f establ ishment and production; financing; l o g i s t i c s and transport. . Following t h i s review, rank the available land i n order o f desired development using requirement o f productive potenti a1, soci a1 acceptance and 1east cost economic development as ranking c r i t e r i a . . Recommend appropriate species or provenance t r i a l s f o r a 4-5 year r o t a t i o n cycle and establish a time schedule f o r plantation development. . Complete financia1 and economic analyses of actions proposed including internal rate o f return (IRR) and present value u n i t cost estimates. . Recommend any manpower t r a i n i n g and development needs f o r changed ' systems o f production, and feed i n t o Peri-Urban Fuelwood P l antation Programme action p l an. :3.0 FUELWOOD SUPPLY AND DEMAND - OVERVIEW 3.1 Forest Resources The nation's forest resources have been summarised by Openshaw (1982) as o f 1980, by Province and forest type i n Table 3.1 following: TABLE 3.1 - FOREST AREA BY PROVINCE AND FOREST TYPE (ha) Planted Forest Natural Forest(2) A l l Forest Province/ Eucalyptus Other Comnercial Non- Bamboo Total Forest Forest & wattle s ~ e c i e s tree area comn & scrub cover% Nyanza Exp 42 680 . 960 1 865 290 38 371) 0.3 Pro - 80 160 30 2 701 - NPk -- - - - - - Western Exp 535 11 245 15 160 23 698 14 807 65 445) 8.3 Pro - -- 650 1 020 630 2 300) NPk - - - - - Central Exp 2 675 24 952 47 340 55 558 79 675 210 206) Pro - 5 610 6 590 9 450 21 650) 20.8) NPk -- - 10 910 12 810 18 380 42 100) 1 Nairobi Exp 1 502 373 70 1 253 - -0 2 198) 1 Pro - - - -1 31.1) NPk - - 10 20 - 30 Eastern Exp 1 034 8 734 1 710 34 050 85 679 131 407) Pro - 140 2 760 6 960 9 860) 0.9 NPk -- - 10 110 280 400) Coast Exp 243 2 205 40 2 6 0 ' ~ ) 70 581(4) 5 556 118 845) 1.5 Pro -- - 570 1 290 100 1 960) NPk - - - - - R i f t Exp 14 884') 87 261 246 370 275 909 96 336 720 765) Val l e y Pro - - 113 510 14 640 5 780 33 930) 4.6 NPk - - 7 550 8 180 3 230 18 960) North East 0 0 0 0 0 0 0 TOTAL Exp 20 920 135 456 351 870 461 914 282 543 1 252 703 Pro - 20 560 26 460 22 950 69 970 NPk (5) -- - 18 480 21 120 21 890 61 490 GRAND TOTAL 20 920 135 456 390 910 509 486 327 383 1 384 163 2.4 Legend: Exp -Exploitable, Pro - Protection, NPk - National Park Notes: (1) Eucalyptus 9674 ha, wattle 5215. (2) The commercial and non-commercial species are mixed, the areas o f these two groups o f species have been divided according t o the volume o f each group. (3) The commercial area includes 21 677 ha o f mangrove forest. Total mangrove area i s 52 980 ha. (4) The non-commercial area includes 31 303 ha o f mangrove forest. (5) The area o f natural parks i s much larger than 61 500 ha but t h i s i s the t o t a l forest area within the parks. Table 3.1 includes a l l forest occurring i n areas over 10 ha i n size but excludes non-forest area w i t h i n the forest such as grasslands, farms and areas o f water. It must be noted that substantial amounts of biomass occur on smallholder farmlands, where they are unrecorded because they are under 10 ha i n area. This biomass resource i s dealt within Section C8. Table 3.2 from FA0 (1982) shows t h a t nearly a l l (87%) o f the land classed as forest estate i s within gazetted forest. While t h i s table cannot be d i r e c t l y compared t o Table 3.1 due t o d i f f e r e n t methods o f vegetation classification, it does indicate that there i s very little forest resource which i s not under government control. TABLE 3.2 - KENYA'S FOREST ESTATE Type o f Forests Area (ha) i Gazettedforests Natural high forests Man-made forests Bush Bamboo Grass Mangrove Sub-total 1 572 459 ( i i1 Ungazetted forests Managed by the Forest Department 89 399 Total forest area under Forest Department management Sub-total 1 661 858 ( i i i) Private forests on large farms 141 000 Kenya's Forest Estate Total 1 802 858 Source: F A 0 I19821 c.h 3.2 Fuelwood Su~r11.y and Demand Much has been w r i t t e n on Kenya's apparent fuelwood supply problems. The o f f i c i a l viewpoint i s summarised i n t h e National Energy Policy (Ministry Energy & Regional Development 1985). Evidence i s presented t h a t there wi 11 be a r a p i d l y developing fuelwood shortage due t o the assumed continued heavy r e l i a n c e on wood as a f u e l f o r domestic use. The 1983 estimate was t h a t 62.8% o f Kenya's gross energy demand was being met from fuelwood. 80% o f a11 t h e fuelwood consumed i n Kenya comes from p r i v a t e farms, 'woodlots or t r u s t lands, while only 13% comes from government controlled f o r e s t s and 7% from Trust Lands administered by County Councils (FA0 1982). I n considering f u e l source options t o enable t h e KTDA t o move from o i l t o fuelwood for i t s tea factories, the macro supplyldemand p i c t u r e should be understood. There i s a range o f supply options open t o t h e KTDA f o r obtaining up t o 100 000 t l y o f fuelwood estimated f o r i t s present 39 f a c t o r i e s a t current energy e f f i c i e n c y levels. However, the macro p i c t u r e indicates t h i s supply wi 11 face competing demands i n the market place. Table 3.3 below shows a growing s h o r t f a l l between supply and demand, estimated t o reach 10 Mtly by the year 1990, and 31 Mtly by the year 2000. If allowed t o proceed unchecked, o r without planned replenishment, the e f f e c t on standing stock i s q u i t e clear. TABLE 3.3 - NATIONAL WOOD FUEL. RESOURCE SUPPLYIDEMAND RELATIONSHIP IN KENYA (MILLION TONNES) Year Demand 20.0 24.5 30.3 38.6 47.1 Supply ( t o t a l 20.0 19.2 20.5 20.6 16.5 From Yield (1) 13.8 12.6 10.7 7.8 5.2 From Stock (2) 6.2 6.6 9.8 18.8 11.3 Supply s h o r t f a l l - 5.3 9.8 12.0 30.6 Standing stock 980.6 974.0 932.0 864.0 800.0 Source: Beij e r I n s t i t u t e 1982 and M i n i s t r y o f Energy and Regional Development 1985. Notes : (1) Yields: Net annual production. Only accessible y i e l d s service demand. (2) Stocks: Net reduction i n accessible standing stocks service demand when demand exceeds accessible yields. Openshaw (1982) has constructed a biomass energy production and consumption model f o r Kenya which appears t o form the basis f o r accepted status on fuelwood i n the national energy policy (MOERD 1984 and Table 3.3). The underlying assumptions t o Table 3.3 are as follows:- 1. Natural forest - Data was taken from a p a r t i a l inventory carried out by a Canadian f i r m i n 1963 -1967 and updated by the author following discussions i n Kenya with various authorities. As a consequence stem volume was reduced by 17%, but a further 20% was added f o r branch vol ume. 2. Bamboo and scrub -Area was mapped but not inventoried. Biomass was estimated by assuming the proportion o f bamboo land as constant at 48 cu.m /ha. 3. Forest plantation - Source o f data assumed was taken from Forest Department records. However, there have been comments from various sources t h a t more recent eucalyptus plantations may have been unsuccessful i n some areas and require inventory t o confirm y i e l d potenti a1 . 4. Tree cover on high potential land - F i e l d inventoried by Beij e r I n s t i t u t e using samples selected from a i r photography i n many cases. I n f a c t a l l woody biomass was sampled. A l i t t l e over 9% o f land area had woody b imass canopy. 'rhi s i s consistent with recent Beij e r I n s t i t u t e data which indicates 16%t o 31% biomass f o r smallholder tea growing areas. See Tab1e 8.1. 5. Consumption - Assumptions are not given i n t h i s paper. The fuelwood Supply and Demand Model i s shown i n Table 3.4 on the following page* a- C n a L .r 3 u . E . U 0- cY tOJ m * Y CL.r C ' e Q a o r m -E ,.YU 0 L 3 U n a l L > r u C U Y U S E Y L m O f -7 6 u u a C u a u n r c o 1 C u m u - 7 a - c -a C n U.r 1 u r n > U C U m t 9:: 3 2 S E 2 5 C m u m m u Z C U U u- C r U u U U U C u a r n a - 0 cU & , C C 0 U n o w r OI O r U t J Q L C Y L L .. C CI Q - r m C = = = A m - W S L W H O U YVIQ Y W U Q Despite the l i m i t a t i o n s o f the data underlying the Openshaw (1982) model, it has served as a valuable framework f o r examination o f Kenya's fuelwood supply problems. The basis o f the assumptions i s being improved by continuing investigations by the Beijer I n s t i t u t e s t a f f through the Kenya Fuelwood Development Project, and the World Bank Peri-Urban Fuelwood Project o f the World Bank. Openshaw's main conclusions (1982) o u t l i n e the Kenyan fuelwood problems. They are summari sed below t o assist i n a better understanding o f the problems t o be overcome i n meeting the f u t u r e fuelwood needs o f the 39 KTDA tea factories, the primary aim o f t h i s study: Woody biomass supplies 95% o f the t o t a l biomass energy demand, and 71% ( i n 1980 values) o f the t o t a l national energy demand. The economic range f o r fuelwood supply i s within walking distance f o r the r u r a l consumer, or within economic transport distance o f the urban/i ndustrial consumer, typical l y 100 km f o r fuelwood and 200- 300 km f o r charcoal. The rangelands are the wood surplus areas and could theoretically supply (as o f 1980) nearly 90% o f the fuelwood demand, but are out o f economic transport range f o r the high d e f i c i t areas and are ecologically f r a g i l e . Farm trees and bushes supply about 47% o f fuelwood energy demand, with 28% from f o r e s t trees and remaining 25% from rangelands. Openshaw's figures (Table 3.4) indicate an erosion o f standing stock or woody biomass capital as o f 1980. Regardless o f the absolute v a l i d i t y o f the figures, the model i t s e l f appears sound and i s supported by the observations o f the Forestry Consultant. The main conclusion i s t h a t there i s a d e f i c i t i n fuelwood i n the high population density areas o f high agricultural potential. The main recommendation o f t h i s study, therefore, i s t h a t the largest e f f o r t be i n t r e e growing which should be i n high potential agricultural areas where the greatest s h o r t f a l l occurs. I n some o f these areas KTDA i s already seeking future wood supplies. 4.0 PRESENT BIOMASS USE - KTDA FACTORIES 4.1 Green Leaf Supply and Production Small holder t e a production i n Kenya under the management o f the KTDA involved some 150 000 farmers delivering, i n 1983/84, over 210 000 tonnes of green l e a f t o 39 KTDA factories. The project, now some 20 years old, has been a major success i n i t s f i e l d by world standards. It had World Bank support i n the e a r l y 1970's. Smallholder green l e a f production has increased from about 160 000 tonnes i n 1981/82 by 32% t o j u s t over 210 000 tonnes i n 1983/84. Further increases can be expected by extending the area planted, and t o some extent by increasing green l e a f y i e l d s per hectare. As o f 1983 some 81 500 ha o f t e a had been planted by a11 growers made up o f : Kenya Tea Growers Associ a t ion (KTGA) 23 830 (47 t e a estates) Kenya Tea Development Authority 54 970 (150 000 smallholders and 39 factories) Others 2 684 Tota1 81 484 The a g r i c u l t u r a l land classed as suitable f o r tea growing i n Kenya i s some 287 500 ha (Jaetzold & Schmidt 1983), leaving a vast potential f o r expansion o f t e a planting by smallholders. At present the mean y i e l d o f made tea f o r small holders and non-KTGA rowers combined i s 936 kg/ha compared with 2 760 kg/ha f o r the estates 7KTGA). The difference i s due t o a combination o f lower inputs (eg. less f e r t i l i s e r ) and the KTDA p o l i c y o f aiming f o r higher value t e a by only harvesting the f i r s t "two leaves and a bud". There e x i s t s therefore substanti a1 potenti a1 f o r a continuing increase i n the quantity o f l e a f processed by KTDA with a corresponding potential increase i n energy needed f o r withering and drying. The number o f KTDA f a c t o r i e s has increased as f o l 1ows: The B r i e f foreshadows the addition o f a f d r t h e r 10 f a c t o r i e s i n the foreseeable future. 4.2 Current Fuel Use by KTDA Factories The quantities o f f u e l used by each factory ( o i l and/or fuelwood), the energy value and wood equivalent i s shown i n Table 4.1 following. The q u a n t i t i e s are from KTDA head o f f i c e records, except f o r the four wood using factories v i s i t e d by the Consultant. Comparison o f head o f f i c e and f a c t o r y records f o r those factories are as follows f o r 1984/85. Head Office Records Factorv Records -Wood O i 1 - Wood O i 1 cu. rn t K es cu.m t 5 K es Kangait a 14 707 Ni1 Thumaita 11 005 Ni 1 G i thongo 7 373 Ni 1 Imenti 7 746 Ni 1 Discrepancies were found between head office and factory fuelwood records a t Kangaita (H.O. 8% higher) and Thumaita (H.O. 60% higher while factory used 25% o i l unstated i n H.O. records). Githongo and Imenti head o f f i c e and factory f u e l records agreed. Due t o time constraints the Consultant was unable t o check the records o f the other three wood using factories a t Chebut, Gathuthi and Ragati. Head o f f i c e records were used i n Table 4.1 f o r the l a t t e r and a l l the o i l using factories. Based on the parameters detailed i n Table 4.1 footnotes, and energy consurr~ption records, the 39 KTDA factories would require about 100 000 t/y of Eucalyptus saligna fuelwood a t 30% mcwb, assuming 1985 tea drying technologies and e f f i c i e n c y levels. From 1983/84 data (K'TDA 1984) some 99 300 tonnes of wood (equivalent) o r 1 410 TJ o f energy wsre used i n drying 63 700 tonnes o f made tea i n 1984/85. This gives an energy i n t e n s i t y about 1.5 tonnes o f wood equivalent t o 1.0 tonne of tea, or approximately 22 GJ/tonne. The private tea estates have around 1:1 by weight as an efficiency target (approximately 15 GJ/t) which the Study Team also considered t o be achievable i n the more modern KTDA factories. Ifthis were so the target for fuelwood could reduce t o about 65 000 tonnes. TABLE 4.1 - FUEL USED BY KTDA FACTOR1ES 1984/85 PROVINCE Factory Fuel KSh/t GSE GJ/y KSh/GJ Wood D i s t r i c t (1) t/Y (2) (3) GJ/t Equiv CENTRAL Kirinyaga Ndirna 0 814.9 3 328 42.9 34 959 77.6 2 462 Kangaita W 5076.3 247 14.2 72083 17.6 5 0 7 6 Kimunye 0 582.6 3 328 42.9 240 834 77.6 1 760 Thumaita W 2 943.3 215 14.2 41 796 15.1 3 254 0 102.9 328 42.9 4 4 1 4 178 246 12 552 Nyeri Chinga 0 1305.7 3 3 2 8 42.9 56015 77.6 3 9 4 5 Iriaini 0 811.3 " " 34 805 " 2 451 G i tugi 0 553.1 " " 23 728 " 1 671 Gathuthi W 1 208.7 291 14.2 17 164 20.5 1 209 Ragati W 3194.0 247 " 45 355 17.5 3 194 177 067 16 430 Muranga' a Njunu 0 829.0 3 328 42.9 35 564 77.6 2 505 " " Makomboki 0 778.8 33411 " 2 353 " 1kumbi 0 1124.4 " 48 237 " 3 397 " " Githambo 0 993.7 42 630 " 3 002 Kanyenyain i 0 719.8 " " 30879 " 2 175 Gatunguru 0 992.4 " " 42 574 " 2 998 233 295 16 430 Kiarnbu Kambaa 0 526.9 3 328 42.9 22 604 77.6 1 592 Theta 0 548.8 " " 23 544 " 1 658 Mataara 0 1083.3 " " 46 474 " 3 273 Kagwe 0 583.1 " 25015 1 762 I( . " 117 637 8 285 TABLE 4.1 - FUEL USED BY KTDA FACTORIES 1984/85 PROVINCE Factory Fuel KSh/t GSE GJ/y KSh/GJ Wood D i s t r i c t (1) t/Y (2) (3) GJ/t Equiv RIFT VALLEY Nandi Chebut W 4173.5 247 14.2 59264 17.4 4 1 7 4 Kericho Tegat 0 4071.3 3 5 5 2 42.9 45959 82.8 3 2 3 7 " " " Kapset 0 1249.0 53 582 3 773 Mogogosiek 0 1 826.1 78 340 5 517 11 " " L i t e i n 0 712.7 " " 30 575 " 2 153 Kapkoros 0 665.2 " " 28 537 " 2 010 236 993 16 690 Nyansiongo 0 448.9 3 5 5 2 42.9 20974 82.8 1 4 7 7 Nyankoba 0 640.8 " " I 1 27 490 1 936 Kiamokama 0 691.3 " 29657 " 2 089 11 Nyamache 0 513.1 " 22 012 " 1 550 I 1 Ogembo 0 396.7 " 17 018 " 1 198 I1 Sanganyi 0 754.8 " 32381 " 2 280 11 Kebirigo 0 946.6 . " " 40609 " 2 860 Tombe 0 566.6 " 11 " 24 307 1 712 214 448 15 102 EASTERN Meru Githongo W 3 155.6 215 14.2 44 810 15.1 3 156 Imenti W 3 315.3 215 " 47077 15.1 3 3 1 5 K i egoi Embu Mungania 0 625.2 3 328 42.9 26 821 77.6 1 889 Rukuriri 0 701.1 " " 30 077 u 2 118 56 898 4 007 A11 Factories -4 1 409 819 99 285 b Notes: 1. Data from KTDA head o f f i c e records except f o r d i r e c t factory data from consultant v i s i t s f o r Kangaita, G ithongo, Imenti and Thumaita. 2. O i 1 prices a t factory supplied by KTDA Head Office 26/9/85. East o f R i f t = 3.11 KSh/L, West = 3.32 KSh/L, Kiegoi = 3.32 KSh/L. 3. Fuelwood prices from consultant v i s i t s t o Thumaita, Kangaita, G ithongo and Imenti. Costs f o r other wood using factories as average o f Kangaita, G i thongo and Imenti . 5.0 PRESENT AND POTENTIAL FUELWOOD SUPPLY FROM FOREST DEPARTMENT 5.1 Supply t o Present Fuelwood Using Factories A t the time o f the Consultant's v i s i t t o Kenya, fuelwood supply t o K'TDA was almost 100% from the Forest Department. The KTDA1s position regarding fuelwood can be summarised as f o l lows: . KTDA policy i s t o convert as many factories from o i l t o fuelwood as soon as possible, provided a r e l i a b l e supply can be made available. . The p r i o r i t y i s t o convert the following factories as soon as possible: Mataara, Ikumbi, Kanyenyaini, Gatunguru, Chinga, I r i a i n i , Kimunye, and Mungania. ( I t i s noted that by mid 1986, most o f these factories had been converted t o dual oi 1/fuel wood f i r i n g ) . . Fuelwood supply from the Forest Department (up t o 1984/85) t o present fuelwood using factories has mainly come from indigenous forest species. With a government decree essentia1l y prohibiting further clearing o f indigenous forests, t h i s supply source i s expected t o d i sappear. . Future fuelwood supplies are expected t o come mainly from Forest Department p l antations, almost exclusively Eucalyptus sal igna. . The KTDA look t o the Forest Department as one source o f supply. A l e t t e r o f 12 January 1979 from the Chief Conservator of Forests which includes the following, has raised some expectations from KTDA. "Forest Department i s planting eucalyptus t o meet t e a factory requirements w i t h a programne now 2-3 years old. Areas are available f o r exploitation i n v i c i n i t y o f factories already f u l l y or p a r t i a l l y operational on fuelwood (i.e. Ragati, Kangaita, Gathuthi ) . Forest Department gives assurances o f fuelwood f o r 13 factories (Mataara, Gatunguru, Chinga, I r i a i n i , Gathuthi, Ragati, Kangaita, Kimunye). For some o f the factories above - it may be possible to use fuelwood e a r l i e r by making use o f indigenous trees f e l l e d from areas t o be converted t o plantations by using t h i n n i ngs." . The KTDA General . Manager wrote t o Chief Conservator of Forests on 20 May 1985 expresssing concern t h a t there i s now doubt over fuelwood f o r 8 factories targeted f o r conversion t o fuelwood and even fear t h a t fuelwood supply may be at r i s k for the 7 factories already converted t o fuelwood from o i l . . KTDA has expressed concern that it i s required t o compete with other major users o f eucalypt plantation resources, especi a1l y pole users such as Posts and Telegraphs and the Kenya Power and Lighting Company. The Consultant was advised there were formal ised ongoing discussions between the Forest Department and each major user t o reach agreement on future allocations. KTDA also expressed concern t o the Consultant t h a t plantations ostensibly established t o supply KTDA factories had not been very successful and were below expectations. The Consultant was unable t o confirm the status o f such plantations i n the l i m i t e d time avai 1able. . KTDA advised the Consultant that the primary reason f o r changing factories t o wood was because it continued t o be a substantially lower cost energy source. I n 1985, a t the time o f data collection f o r t h i s study, the equivalent energy cost advantage was over 3: 1 i n favour of fuelwood. The Consultant has shown i n t h i s report how it i s feasible t o produce fuelwood, under various supply options, a t continuingly low energy cost t o KTDA. 5.2 Future KTDA Fuelwood Supply Perspective I n order t o assess the prospects o f the Forest Department as a source o f fuelwood supply t o K'TDA, the Consultant examined the available data on eucalyptus plantations o f that Department and compared supply potentia1 with factory requirements. Table 5.1 provides estimates of fuelwood requirements compiled by KTDA and by the Consultant. I n terms o f tonnes/year the two estimates are close (7% difference). The Consultant has used the KTDA fuelwood y i e l d estimate of 11.85t/ha/y as a reasonable expectation. 'This i s well below the potential of the better upland s o i l s but may exceed present production 1eve1s. The potentia1 fuelwood supply position compared with KTDA demand requirement on a d i s t r i c t basis has been developed from Table 5.1 and i s shown i n Table 5.2. TABLE 5.1 -KTOA FACTORY FUELWOO DEMAND AN0 FOREST OEPARTHENT EUCALYPTUS (E) WATTLE (U) PLANTATION RESOURCES C t Olstrlct Factory Fuealwood Demand Forest Department Pl an tation supply KTOA Estimate WB Estimate Promised to KTOA 1979 Recorded Plantation Areas by Sources ::r Area ha t / y Avai lable Reco~~~ut. rota1 t/y ha ha t/Y ha t / y (2) (3) plant ha ha ha (4) 1983 1984 1984 data 1983 1983 data CENTRAL (5) (5) (6) Kirinyaga Ndie~a 2 370 . 200 2 500 0 0 0 0 Kangalt a 5 100 0 200 E 200 E 2 400 K11wnye 1 800 0 160E 160E 1 900 I I Thumait a 3 300 0 160E 160E 1 900 I 9 4 8 0 800 12 600 240 261 3 100 240 2 9 0 0 Nycrl Chinga 2 370 . 200 3 900 0 200 E 200 E 1 400 I r l a l n l 200 E. 200 E I 2 500 0 2 400 ' Gltugl m 1 700 CI Gathuthl 1 200 148 200.E ,348E 4 100 I w Ragati 5 900 I #I 3 200 298 200 'E 498 E w Muranga'a Njunu 2370 200 2 5 0 0 0 0 0 . O Makonlbokl .. I 2 400 0 0 0 0 Ikunbl I 3 400 0 0 0 0 GIt h d o I 3 000 0 . 0 0 0 Kanyenyaln l I 2 200 0 0 250 W 3 000 Gatunguru 3 000 0 500 W 500 W 6 000 K1aubu Kinbaa 2 370 200 1 600 0 0 0 0 Theta I 1 700 0 0 0 0 Mataara I Y 2 300 0 300 U 300 W 3 600 Y Kagwe 1 800 No data available RIFT VALLEY Nandl Chebut 2 370 200 4 200 243 266 3 200 211 2 5 0 0 Kerlcho Tegat 2 370 200 3 200 No data available " Kapset . Y 3 800 I Y Mogogoslek 5 500 " L l t e i n .. 2 200 Y Y " Kapkoros I 2 0 0 0 I I850 1000 16 100 993 1040 12 400 549 6 500- TADLE 5.1 KTOA FACTORY FUELWOOD DEMAND L FOREST OEPARTMNT EUCALYPTUS (El UATTtE (U) PLANTATION RESOllRCES PROVINCE Factory: Fuelwood Ocmnd Forcst DcparWnt plantation Supply D i s t r i c t KTOA Estimate wo Estimate Promiscd t o KTOA - -1979 Recorded Plantation Areas 't/y ' Arca ha L/Y Avai lablc HCCQII~. Iota1 tlY ha ha t/Y ha 1903 1904 1904data 1903 (1) (2) plant ha ha ha (3) RIFT VNLEY (Cont'd) (4) (4) (5) Nyansiongo No data --, availablc . Nyankoba . . .. .. Kiamkalna I1 Nyamachc . I 1 Ogenbo I I ,I Sanganyi .. .. I, Kcbirigo I, Taube I . I I\ 10960 1 6 0 0 15100 0 0 0 na EASTERN Heru Githongo 2 370 200 3 200 40 400 520 E 6 200 lrnenti 3 300 No data available , Kicgoi 1 700 0 0 . 0 0 Kinoro 1 400 0 400 E 400 E 5 700 Hungania 2 370 200 1 900 0 160 E 160 E 1 900 ' ' Rukuriri 2 100 0 0 0 0 , 4 740 400 4 000 134 155 1 800 na A11 factories 92 430 7 000 99 300 57 800 Note (1) 1 cu.m stackcd = 0.316 tonncs 302 uuwb 1 tonne = 14.2 GJ GSE I (2) KTM assurcs 10 year rotation (3) WO = Energy Efficiency i n Tca lndustry Study Tcaa b a n Annual Incramnt (MA11 = 22.5 cu.u/ha/y 11.3 t/ha/y 30 2 lncwb (4) Forcst D c p a r b n t y i e l d estimate from rccords (5) Oata fran Forcst Dcparturnt via World Dank Pcri-Urban Fuclwood Projcct (61 Oata fraa Forest Dcpartmcnt rccords via Kahuki (1905) TABLE 5.2 - FUELWOOD SUPPLY - DEMAND BY DISTRICT (Tonnes/Year) PROVINCE KTDA Demand ( 1) FD Yield (2) Surplus Deficit D i s t r i c t Ratio fuelwood/tea Potentia1 Ratio fuelwoodltea CENTRAL K i rinyaga 12 600 8 100 3 100 - 9 5 0 0 - 5 0 0 0 Nyeri 12 500 8 000 14 800 + 2 3 0 0 + 6 8 0 0 Muranga' a 16 500 10 600 500 -16 000 -10 100 K i ambu 8 400 5 400 28 300 +19 900 +22 900 RIFT VALLEY Nandi 4 200 2 700 2 800 - 1 4 0 0 + 100 Kericho 16700 10700 7 100 - 9 600 - 3 600 Kis i i 15 100. 9 100 0 -15100 - 9 7 0 0 EASTERN Meru Embu -- A11 Factories 99 600 64 000 64 400 -35 200 + 400 Source: Table 5.1 Notes : (1 The present KTDA fuelwood energy r a t i o i s 1.56 tonnes fuelwood (energy equivalent) t o 1 tonne made tea. 'The target r a t i o i s 1 tonne o f fuelwood t o 1 tonne made tea. (2) Estimates o f potential y i e l d compiled from Kahuki (1984) data f o r 1983 (Table 5.11, updated by adding 1984 planting from World Bank Peri-Urban Fuelwood Project data, used f o r K i s i i and Embu Districts. Assuming a l l f a c t o r i e s are fuelwood operated, Table 5.2 i n d i c a t e s t h a t : For Central Province based on present 1.56:l fuelwood t o made t e a r a t i o - Nyeri and Kiambu D i s t r i c t s have p o t e n t i a l fuelwood y i e l d s higher - than factory needs Kirinyaga and Muranga'a D i s t r i c t s show a d e f i c i t . This r e l a t i v e p o s i t i o n w i l l remain even if e f f i c i e n c y improves t o 1:1, a1though t h e o v e r a l l Province bal ance wi 11 change from a small fuelwood d e f i c i t t o a substantial surplus. The above assumes: - A l l the present growth can be harvested yearly. This i s o p t i m i s t i c as t h e age class d i s t r i b u t i o n i s uneven and generally biased towards younger trees, g i v i ng a 1ower sustained supply y i e l d than i s i n d i c a t e d by growth o f plantations. - The areas stated are f u l l y stocked and productive. This i s o p t i m i s t i c bearing i n mind t h e view t h a t some Forest Department eucalypt plantations may not be as productive as they could be. - A l l t h e y i e l d w i l l come t o KTDA. This i s o p t i m i s t i c as there wi 11 be competing demands from other users o f telephone, power, b u i l d i n g poles, and fuelwood due t o proximity t o Nairobi. It should be economically possible t o transport eucalyptus fuelwood from areas o f surplus, i n c l u d i n g neighbouri ng fuelwood r i c h regions and u t i l i s i n g otherwise unused r e t u r n i n g road vehicle capacity, t o areas o f d e f i c i t i Central and Eastern Provinces assuming 2.0 KSh/t/km or 0.14 KSh/GJ/km. At t h e r e l a t i v e fuelwood/oil p r i c e s i n 1.985, on an equivalent thermal energy basis, there appears q u i t e substantial p o t e n t i a l f o r the transport o f fuelwood from areas o f surplus. I n t h e R i f t Valley Province there are fuelwood d e f i c i t s , e s p e c i a l l y i n Kericho and K i s i i D i s t r i c t s , even if energy e f f i c i e n c i e s reach 1:l. This i s o p t i m i s t i c and t h e d e f i c i t s could be larger. I n t h e Eastern Province t h e r e i s a s l i g h t d e f i c i t a t 1:Iusage r a t i o . 5.3 Cost o f Fuelwood Production The cost o f production f o r a eucalyptus model was examined by Forest Department f o r an 8 year r o t a t i o n . This was updated by t h e Consultant who a1so developed a model from Forest Department data f o r i n d u s t r i a l c o n i f e r establishment costs. These models which assume 15% i n t e r e s t are d e t a i l e d i n Annexure C2, Tables 2.1, 2.2 and 2.3 and summarised as follows: 8 year r o t a t i on 67.53 128.15 9.03 4 year r o t a t ion 71.ll 134.66 9.49 These costs assume good management and f u l l y stocked plantations which grow at predicted rates o f 28 cu.m/ha/y. The KTDA has advised the Consultant that, i n practice, the Forest Department plantations have so f a r performed below these expectations. That level, however, was not able t o be determined by the Consultant. 5.4 I n s t i t u t i o n a l Matters The Forest Department resource needs a more detailed and updated inventory t o estimate the amount o f fuelwood available. A new long term fuelwood supply commitment could then be made t o the various resource users, such as KTDA, based on the resources identified. Plans should be made by the Forest Department t o upgrade i t s hardwood plantations t o maximise output consistent with existing policies which do not allow further f e l l i n g o f indigenous forests t o expand forest plantations. U n t i l the above are attended to, the shortfalls i n fuelwood supply t o KTDA factories cannot be quanti fied exactly. However, from Tables 5.1 and 5.2 it i s clear there w i l l be supply difficulties to factories i n most if not all Districts, even i f KTDA i s assumed to receive 100% a11ocation of the Forest Department hardwood plantations. These s h o r t f a l l s w i 11 therefore need t o be met by supply options outside Forest Department. These options are examined l a t e r i n t h i s Appendix . 6.0 PRESENT BIOMASS USE AND SUPPLY - PRIVATE ESTATES 6.1 Fuelwood Resource I n 1983 t h e r e were 23 830 ha under t e a production by members o f t h e Kenya Tea Growers' Association ( p r i v a t e estates), p l u s 2 680 ha o f "others" apart from t h e small holder sector. It i s known that the estates of Brooke Bond, East A f r i c a n Produce, George W i 11iamson and East African Acceptance (Kaisugu), w i t h a confirmed planted t e a area o f 13 900 ha, use p l a n t a t i o n grown fuelwoods as an energy source i n drying. It i s assumed the balance o f estates and "others" also use fuelwood exclusively f o r drying. Fuelwood plantations are composed mainly o f Eucalyptus sal igna, although one e s t a t e has r e c e n t l y introduced a provenance o f E. grandis from Austral ia whi ch has produced greater y i e l d s over t h e same period. There i s substantial p o t e n t i a l f o r surplus fuelwood production from t h e p r i v a t e estates as described hereunder. 6.1 .I Improved E f f i c i e n c y i n Drying Tea E f f i c i e n c y i n drying f o r t h e p r i v a t e estates v i s i t e d i s around 1.5t fuelwood (30% mcwb a t 14.2 GJ/t GSE) t o 1.0 tMT. The t a r g e t i s 1:1 a l e v e l which has been achieved using d i r e c t drying from wood g a s i f i c a t i o n , and can also be achieved by e f f i c i e n t i n d i r e c t wood combustion i n b o i l e r s . Table 6.1 indicates t h a t a surplus o f about 33 000 t / y o f fuelwood i s p o t e n t i a l l y a v a i l a b l e from t h e p r i v a t e estates through increased e f f i c i e n c y a t present production l e v e l s . However, given market opportunities f o r increasing t e a sales, t h e estates preferred option may be t o decrease fuelwood area and increase t e a area where fuelwood i s growing on s i t e s s u i t a b l e i n t e r e s t . f o r tea, even though t h i s may not necessarily be i n t h e national The Consultant understands t h a t on many estates fuelwood plantations have been established on s i t e s l e s s s u i t e d t o tea. I n these cases t h e r e may be more opportunity t o produce surplus fuelwood f o r t h e market as energy e f f i c i e n c y increases. I n t h e Kericho d i s t r i c t , however, t h e Consultant was o f t h e opinion t h a t t h e fuelwood plantations o f t h e p r i v a t e estates appeared i n t h e main t o be on land s u i t e d t o tea, and hence t h e opportunity f o r producing surplus fuelwood may be l i m i t e d . TABLE 6.1 - ESTIMATED FUELWOOD YIELD AND USE FROM TEA ESTATE PLANTATIONS ( t / y ) D i s t r i c t Made Tea Usage/ E f fic iency Level Possible (1) 1.5:l 1 : l Surplus Kericho 33 387 50 081 33 387 16 694 S o t ik 7 764 11 646 .7 764 3 882 Nandi 18 477 27 716 18 477 9 239 K iambu/Li muru 6 119 9 179 6 119 3 063 TOTAL 65 747 98 621 65 744 32 877 Note: (1) Mean Estate made t e a y i e l d = 2 759 kg/ha f o r 1983 (KTGA 1984) 6.1.2 Improved Fuelwood Plantation Yields (a) Genetic Materi a1 Fuelwood y i e l d s on the p r i v a t e plantations are already high by world standards. Two major estates have supplied data, summarised i n Table 6.2 below. The most conservative y i e l d i s 23 t l h a l y f o r an 8 year r o t a t i o n o f Eucalyptus saligna w i t h wood of 527 kg1cu.m density a t 30% mcwb giving 14.2 GJIt GSE. One estate reports increasing y i e l d by 56% from 48cu.mlhaly (25.3 t l h a l y ) using E. saligna, t o 75cu.m/ha/y (39.5 t l h a l y ) using selected seed o f E. grandis collected i n Australia. TABLE 6.2 - EUCALYPTUS PLANTATION YIELDS - KERICHO DISTRICT Tea Eucalyptus Yield cu.m/ha Age MAIIhaly Estate Species stacked Years Cu.m s o l i d ( i ) Tonnes(2) Sal igna Grandi s 1 000 7.5 75.0 39.5 (7 t o 8) B. Sal igna 750 8 56.3 29.6 Notes: (1) cu.m stacked = 0.6 cu.m s o l i d (over bark volume) (2) Density f o r E. saligna a t 50% mcwb = 738 kg1cu.m 25% mcwb = 492 kg1cu.m a t 0% mcwb = 369 kg1cu.m Wood a t 30% mcwb and 527 kg1cu.m density has been used as the standard for f a s t grown up1and eucalyptus. Brgzi 1ian experience has g i ven y i e l d s of up t o 85 'cu .m/ha/y f o r selected clones o f E. grandis and expectations from current developments are t o r a i s e Mean Annual Increment (MAI) t o 110 cu.m/ha/y (Mercado 1985). However these are exceptional yields, on good s i t e s and with high technology and large f e r t i l i s e r inputs. They do indicate t h a t the Kericho d i s t r i c t , with less technology and f e r t i l i s e r only i n Year 1, must rank w i t h the world's most natural ly productive s i t e s f o r eucalyptus (and other species). Higher y i e l d s are be1ieved possible u t i 1i s i ng Brazi1ian methods o f genetic selection through clonal development by vegetative propagation. This would be a useful means o f increasing fuelwood production where land i s very limited. A preferred social option would be t o encourage the small holders t o produce market fuelwood where possible as an added economic benefit t o farm income. This option i s addressed i n Section 8. (b) Changi qg Plantation Management Methods ' Present spacing o f trees varies from 2.75m x 2.75m t o 1.5m x 1.5m f o r the major estates. It i s possible t o increase y i e l d by decreasing spacing between trees provided t h e r o t a t i o n i s decreased. A y i e l d o f 29.6 t/ha/y w i t h E. grandis a t 2.7m x 2.7m spacing, t h a t i s 1325 trees/ha on 7 t o 8 year r o t a t i o n should be capable of significant increase by planting say 5000 trees/ha on a 4 year rotation. This would a1so produce a smaller diameter stem which i s f a r more suitable for use as fuelwood. Production o f trees o f small e r diameter on a shorter r o t a t i on a11ows the use o f smaller and less c o s t l y fuel preparation plant. The shorter r o t a t i o n does, however, reduce fuelwood production. The Consultants propose elsewhere i n the report t h a t f o r high e f f i c i e n c y combustion, fuelwood should be i n a p a r t i c l e form rather than the present large (say 5-10 kg) pieces. Reduction o f fuelwood by chipping or hogging, f o l 1owed by f i r i n g i n a cyclonic combustion chamber, a1lows greater combustion e f f i c i e n c y and hence reduced f u e l consumption t o be achieved. 6.2 Cost o f Fuelwood Production - Standing Tree Two major tea estates provided data on costs o f establishing eucalyptus plantations and producing fuelwood. Details are given i n Annexure C2 Tables 2.5, 2.6, and summarised i n Table 10.1. Costs f o r comparable models using an 8 year r o t a t i o n and 15% i n t e r e s t give s i m i l a r production costs f o r these estates shown i n Table 6.3. TABLE 6.3 - FUELWOOD COSTS Estate m3/ha/y ~ ~ h / m ~KSh/t KSh/GJ KSh/ha/y Mean 52.1 44.14 83.76 5.90 2 300 6.3 Opportunity Cost o f Fuelwood Production The gross r e t u r n from estate tea growing i s o f the order o f 50 000 KSh/ha/y w i t h green l e a f a t say 5 KSh/kg. This i s t o compared w i t h an estimated cost o f 2 300 KSh/ha/y f o r fuelwood i n standing tree, or 7 300 KSh/ha/y when harvesting and transport costs are added (see Table 10.1 for 8 year model 1. On t h i s basis t h e opportunity cost t o the estates for wood are l i k e l y t o be t o o high t o warrant continuing t o grow fuelwood where tea can replace it as a crop. Therefore the estates may not become a s i g n i f i c a n t source o f surplus fuelwood availab1e t o KTDA factories. 7.0 POTENTIAL OF FUELWOOD SUPPLY FROM UNDERUTILISED PUBLIC LANDS 7.1 Opportunities I n many countries p lantabl e 1and exi sts i n road reservations a1ong highways and, t o a lesser extent, along secondary roads. 'There are also small areas unutilised around public buildings such as schools, along railways, and i n stream reservations. India has pioneered the successful use o f such lands i n i t s extensive social forestry programnes f o r the production o f fuelwood, poles and other community wood products. The tobacco company, BAT, has already made some use o f such lands i n Kenya by planting along the roadside and other public areas (Kuloki 1982). KTDA has an opportunity t o develop a similar programme, p a r t i c u l a r l y along roads leading t o and through the highlands close t o tea growing areas where s o i l s are generally deep and of high f e r t i lity. Such programmes would require a survey of land a v a i l a b i l i t y and agreement by the various public sector authorities responsible f o r administering these lands. It could also involve local communities which may be invited t o share t o some extent i n the costs and benefits. It i s proposed the main beneficiaries should be the KTDA tea factories which should i n i t i a t e and manage the programmes. 7.2 Roadside P l anti ngs Preliminary estimates were made o f lengths o f main road reservations which could be candidate 1ocations f o r p l anti ng f a s t growing fuelwood trees. Distances were estimated from a 1:1 000 000 scale roadmap u t i l i s i n g 4 rows o f trees spaced 1 metre i n the row and 2 metres between rows a t a stocking o f 5000/ha. The spacing i s somewhat a r b i t r a r y and could be decreased t o give an increased biomass y i e l d per year. The estimated y i e l d f o r selected main roads i s up t o 25 000 t/y. Yields were conservatively estimated by the Consultant based on Kenya experience and guided by agro-climatic data o f Jaetszold & Schmidt (1983). 7.3 Other Public Lands I n addition t o the potential main roads yield, there are many secondary roads along which one or more rows o f trees could be planted i n an arrangement between adjacent farmers and an appropriate government agency. There are a1so other public lands under the control o f Provincial and/or D i s t r i c t Councils and other agencies. This area i s unknown, and should be inventoried, but i s arbit r a r i ly estimated f o r reporting purposes a t 500 ha i n and adjacent t o the tea growing areas. This could y i e l d a potential further 5 000 t / y which, when added t o the possible 25 000 t / y from roadside plantings, could provide up t o 30 000 t / y or nearly one t h i r d of the estimated fuelwood needs f o r a l l KTDA factories. ; TABLE 7.1 - POTENTIAL FUELWOOD YIELD FROM PLANTING I N ROADSIDES AND OTHER PUBLIC LANDS t/y 1. Roadside Plantings - Main Roads PROVINCE km ha/km ha m3/ha/y t/y 4 rows (1 4 rows CENTRAL Thi ka-Muranga' a (Rain fa11 900-1 200mm) 40 4 160 20 1 686 Muranga-Nyeri ( Rain fa11 900-1 200mm) 40 4 160 20 1 686 Muranga- Embu 30 4 120 25 1 581 (Rai n fa11 1200mm) Nakuru-Keri cho 100 4 400 30 6 324 EASTERN Embu- Meru ( R a i n f a l l 1200-1400mm) 80 4 320 30 5 059 Meru-Maua (Rai n f a l 1 1200-1400mm) 40 4 160 30 1 686 RIFT VALLEY Keri cho-Ki s i i TOTAL say = 25 000 2. Other Roads, Other Public Land Includes school s, other pub1i c buildings, town commons, railway 1ines, streambank reserves etc. The a v a i l a b l e area i s taken as 500 ha x 20cu.m/ha/y. say = 5 000 t/y Total preliminary estimate Source: Preliminary consultant estimate from 1:1 m i l 1i o n scale map and f i e l d impressions. Requires f i e l d survey. (1) Yield estimate o f consultant based on r a i n f a l l and agro- c l i m a t i c data o f Jaetzold v Schmidt (1983). 7.4 Cost o f Fuelwood Production The plantation should be fenced f o r say 4 years t o prevent animal damage. The fence would be t h e major plantation cost, but it could be relocated and used f o r a t least one and possibly two subsequent p l antings. The fencing cost i s estimated at: 500m/ha o f fencing @ h a l f o f 25 KSh/m f o r a 6 wire fence (2 barbed 4 p l a i n ) = 6 250/ha For 4 yr r o t a t i o n @ 15% = 91.09 KSh/cu.m a t 30 cu.m/ha/y Using the 4 year small holder cost model : wood production cost = 61.90 + 91.09 = 152.99 KSh/cu.m Delivered cost 111.95 + 152.99 = 264.94 KSh/cu.m = 502.73 KSh/t = 35.40 KSh/GJ Note: (1) Tables 10.1 and Annexure C2 Table 2.7. This i s an expensive option compared w i t h the alternatives, but a t 36 KSh/GJ delivered i s s t i l l only about half (on a f i n a n c i a l cost basis) o f 1985 energy costs from o i l . 7.5 I n s t i t u t i o n a l Matters and Constraints The use o f roadside land, i s practised i n other developing countries, would be a departure from current practice and would c l e a r l y pose complex social problems as t o land usage and p r i o r i t i e s which l i e beyond the scope o f t h i s report. Nevertheless the agricultural opportunity exists, and it i s therefore put forward as one o f the options for fuelwood supply enhancement t h a t should be taken i n t o overall consideration. Roadside planting would n a t u r a l l y need agreement by the M i n i s t r y o f Highways and Communications, as we11 as by a11 o f the appropriate local government agencies. It would a1so significantly affect 1ocal people whose views would need t o be canvassed, and who should be involved i n planning and execution, and should receive some of the benefits. The local people would be i n a p o s i t i o n t o guarantee the security o f the project, mend fences and see t o t h e welfare of the plantation. Plantings on roadsides and unused public land could also serve as a catalyst t o small-holder production o f fuelwood, especial l y after the i n i t i a1 cash f 1ow, provided market prices r i s e considerably above present levels. The BAT model along roadsides and underutilised public lands appears t o be successful and should be studied t o i d e n t i f y the obvious problems which i n e v i t a b l y arise i n regard t o ownership, management, protection. ' 8.0 POTENTIAL FOR FUELWOOD SUPPLY FROM SMALLHOLDERS 8.1 Smallholder Land Use i n the Hiqhlands The Highlands o f Kenya, classed as suitable f o r t e a and coffee growing, have an elevation around 2000 m with general l y f e r t i l e soi 1s o r i g i n . These areas have day temperatures i n t h e range 16 8-f 1 vo canic 26 C, an absence o f f r o s t s and an annual r a i n f a l l 1200-2000 mn. 'The Kericho tea growing areas have extremely high potential for plant growth. Because o f t h e favourable c l i m a t i c conditions and 1and capabi lity, the r u r a l population i s dense and farm sizes small i n t h e region where KTDA f a c t o r i e s are located. There i s a size v a r i a t i o n from 1.1 ha i n the K i s i i D i s t r i c t t o 11.5 ha i n the Nandi D i s t r i c t , as shown i n Table 8.1 below. TABLE 8.1 - SMALLHOLDER FARM SIZE AND TEA PLOT AREA BY DISTRICTS Mean Mean Farm Di s t r i c t Farm Size ha (1 Tea P l o t Size ha (2) Kericho Ki s i i Nandi Kiambu Muranga' a Nyeri Kirinyaga Embu Meru - Note: (1) Jaetzold & Schmidt 1983 (2) KTDA Annual Report 1983/84 (3) Assumed 8.2 . Present Biomass Resource Work c a r r i e d out by the B e i j e r I n s t i t u t e i n t h e Kenya Highlands indicates there is between 10% and 30% o f the smal 'Iholder farms with some form o f permanent biomass resource. Beij e r s t a f f suggested 10% biomass cover as a safe f i g u r e f o r t h e Consultant t o use, with between 2% and 8% o f the farm under permanent or c u l t i v a t e d biomass as hedge, f r u i t trees, o r woodlot. I n the tea areas it was found that the higher t h e t e a percentage, t h e higher was the woodlot percentage; t h i s may i n d i c a t e the farmers are reacting t o market conditions f o r fuelwood as a cash commodity. It i s said t h a t 2% as woodlot i s "dedicated" by market forces. Beij e r studies, as exampled by Table 8.2 indicate t h a t woodlots are kept i n an uneven aged condition by selective harvesting o f trees f o r building poles and fuelwood on a 4 year average c u t t i n g cycle o r coppice rotation. The y i e l d can be safely taken as 30 cu.m/ha/y i n t h e h i g h l y productive Tea and Tea-Coffee Zones (Jaetzold & Schmidt 1. Tree diameters o f 8 t o 10 cms (breast height, over bark) are currently harvested on farms and are suitable for e i t h e r poles or fuelwood. The size i s appropriate f o r the proposed method of chipping and hogging, f o l lowed by f l u i d i sed i n j e c t i o n i n t o a cyclonic combustion chamber, the method o f f i r i n g small boilers more e f f i c i e n t l y on woodfuel proposed i n t h i s report. TABLE 8.2 - WOODY BIOMASS (WB) EXISTING IN TEA GROWING AREAS Area Pop'n Farm Tea Woodlot WB WB WB Oppor- Density Size % % Planted Total t u n i t i e s ha % % % -- - -- -- North K i s i i 432 0.8 9.5 12.6 14.0 17.7 3.7 South K i s i i 366 1.0 5.7 9.5 11.1 16.0 4.9 Muranga' a 523 1.1 18.4 7.0 9.1 17.6 8.5 Kiambu 430 '1.9 15.3 7.8 15.2 30.7 15.5 Source: Personal communication with P. Bradley o f the Beij e r I n s t i t u t e - Kenya Woodfuel Project. From survey data 1985. 8.3 Potentia1 Resource for KTDA It has been assumed from data provided t o the consultants that, conservatively, between 2% and say up t o 5% o f typical farm area could be avail able for market fuelwood production, and which could be accessed by existing KTDA factories. These factories offer a market of up t o around 100 000 t/y o f fuelwood a t present fuel efficiency levels. It i s however well recognised t h a t t h i s i s an average figure, and w i l l not be equally t r u e f o r a l l d i s t r i c t s . Indeed i n some d i s t r i c t s o f very high populating density, land p r i o r i t i e s f o r subsistence forming and animal husbandry may be well exert p r i o r claim. Furthermore current (but not future) market prices for fuelwood may not be seen t o provide s u f f i c i e n t incentive t o plant wood as a cash crop. Nevertheless the opportunity has been i d e n t i f i e d by researchers and i s reported upon here. The Tea, Tea-Coffee and Coffee Zones (Jaetzold & Schmidt 1983) are considered suitab1e f o r supply of fuelwood t o KTDA factories. Relevant areas are presented i n Table 8.3. Potential land available f o r growing market fuelwood for KTDA factories was assumed at 2% and 5% levels o f farm area with growth a t 15.8t/ha/y for Tea and Tea-Coffee Zones and a t 13.2t/ha/y for the Coffee Zone. The t o t a l potential fuelwood y i e l d s were then compared with the KTDA factory needs from Table 4.1 with conclusions as f 01lows: 1. I n aggregate there i s a potential f o r a l l the zones t o oversupply a l l KTDA factories by a factor of: 2.4 times a t 2% level a t present energy efficiency potential " 3.7 " I, I, I 1 I8 I t 5.9 " " 5% " " present " m a 9.2 " 11 11 I 1 I 1 potential fi 11 This would require small holder participation of around 42% , 27%, 17% and 11%respectively. 2. The Tea and Tea-Coffee Zones i n the D i s t r i c t s i n Central Province would be unable t o meet present needs a t 2% level, even with 100% participation. However, adding the Coffee Zone would result i n substanti a1 oversupply. 3. The Tea Zone alone i n a l l D i s t r i c t s o f R i f t Valley Province could potentially oversupply KTDA needs a t the 2% level with 100% partic i pation. 4. The Tea and Tea-Coffee Zones i n D i s t r i c t s i n Eastern Province would undersupply a t 2% but substantially over supply by the addition o f the Coffee Zone. 5. If mean factory fuelwood requirements are say 100 000 t/y f o r 39 factories then typical ly a single factory might require up t o 2 500 t / y o f fuelwood. This could be provided from about 160 ha ( a t 30 cu.m/ha/y) i n Tea and Tea-Coffee zones, and from about 190 ha ( a t 25 cu.m/ha/y) i n a Coffee zone. TABLE 8.3 -POTENTIAL FOR FUELWOOD PRODUCTION BY AGRO-ECONOMIC ZONES t l y (1) Compared with KTDA Factory Needs PROVINCE Tea Zone Coffee-Tea Zone Coffee Zone Total Factory Fuelwood Needs D i s t r i c t ha(2) t l y ha t / y ha t l y ha t/Y t l y by Ratio fuelwood/MT Fuelwood/Made Tea ('00) 2% 5% ('00) 2% 5% ('00) 2% 5% ('00) 2% 5% (5) (3) [ 3) (4) 1:1.56 1:l CENTRAL Kiriyaga 38 1 200 3 000 124 3 900 9 800 225 5 900 14 900 387 11 000 27 000 12 600 8 100 Nyert 189 6 000 14 900 61 1 900 4 800 380 10 000 25 000 630 17 900 44 700 12 500 8 000 Muranga' a 344 10 900 27 200 261 8 200 20 600 548 14 500 36 200 1 153 33 600 84 000 16 500 10 600 Kiambu 160 5 100 12 600 120 3 800 9 500 483 12 800 31 900 763 21 700 54000 8 400 5 400 23 200 57 700 17 800 44 700- 43 200 108 000 84 200 210 400 50 000 32 100 RIFT VALLEY '7' Nandi 221 7 000 17 500 314 9 900 24 800 93 2 5 0 0 6 1 0 0 628 19 400 48 400 4 200 2 700 W Kericho 955 30 200 75 400 255 8 100 20 100 508 13 400 33 500 1 718 51 700 129 000 16 700 10 700 w K i s i i 623 19 700 49 200 995 31 400 78 600 22 700 1 500 1 640 - 51 800 129 300 15 100 9 700 56 900 142 100 49 400 123 500 16 600 41 100 122 900 306 700 36 000 23 100 EASTERN Meru 96 3 000 7 600 165 5 200 13 000 565 14 900 37 300. 826 23 100 57 900 9 600 . 6 200 W u 11 300 900 84 2 700 6 600 125 3 300 8 300 220 6 300 15 800 4 000 2 600 TOTAL 2 637 83 400 208 300 2 379 75 100 187 800 2 949 78 000 194 700 7 965 236 500 590 800 99 600 64 000 - Source: Agro-econmic areas fran Jaetzold iZ Schmidt (1983). - Note: (1 Personal camnunication fran Beij e r I n s t i t u t e s t a f f Nairobi (3) Assumed y i e l d 30 cu.m/haly or 15.8 t l h a l y (Kenya Woodfuel Development Project) indicates 10%t o 30% of 30%mcwb f o r tea and Tea-Cof fee Zones. upland smallholder farms are covered with biomass. With a figure of 10% and assming levels o f 2% and 5% o f farm biomass (4) Assumed y i e l d 25 cu.m/ha/y or 13.2 t/ha/y area presently underuti1ised a potenti a1 reservoir f o r market 30%mcwb f o r Coffee Zone. (KTDA) fuelwood could be made available. (2) Areas of tea estates have been deducted (see KTGA 1983 areas) (5) From Table 4.1. Assuming 100% farmer participation the quantity o f fuelwood required f o r each factory can therefore be produced t h e o r e t i c a l l y w i t h i n the Tea and Tea-Coffee Zones as follows: . Assuming 2% o f farm area under woodlot, the t o t a l farm area required equals 7900 ha or 79 square kiloinetres, located w i t h i n 5 km radius o f the factory. . Assuming 5% o f farm a r q under woodlot, 3 200 ha or 32 square kilometres o f farm area located within 3.2 km radius of the factory are required. . If farmer p a r t i c i p a t i o n i s reduced t o 50%, for 2% and 5% o f farm area under woodlot, the radius around the factory would be extended t o 7 km and 4.5 km respectively i n order t o meet factory fuelwood needs. From 5. above it i s reasonable to assume a mean fuelwood transport distance t o factory o f between 2 and 4 km, a very small added cost t o t h a t o f the estimated smallholder model production cost i n Table 8.4. Currently green l e a f i s purchased from farms w i t h i n 3 kms o f factory buying centres. Accordingly, it should be q u i t e possible f o r many o f the tea growing farms t o be encouraged t o produce fuelwood f o r t h e i r own factory. 8.4 Cost and Returns o f Fuelwood Production - Standing Tree Farmers have been growing fuelwood trees (Note: Fuelwood can be produced from other than trees), especi a1l y Eucalyptus, successful l y f o r many years according t o s t a f f o f Beijer I n s t i t u t e i n Nairobi. Good q u a l i t y seedlings raised under coffee which has been f e r t i lised, or on o l d house f l o o r s o f high f e r t i l i t y , are examples o f low cost farmer technology. The impression gained by the Consultant i s that: 1. Farmers are quite capable o f producing market fuelwood i n woodlots. 2. Farmers however need some advice and improved seed supply (especially more suitable species and varieties). 3. Farmers need assurance o f a market w i t h the local tea factory and o f an acceptable p r o f i t margin. I n the absence of real data from farmer records, the Consultant has taken selected elements from three models provided by major estates o f t h e i r costs o f Eucalyptus fuelwood production and KTDA data. Excluding costs o f f e r t i 1iser and weedi cide the actual cost i s almost e n t i r e l y 1abour which can be shadow-priced a t a Consultant estimate o f 60%. The model a1lows an i n t e r e s t charge on the imputed cost o f the farmer's labour as a return t o the farmer f o r growing the standing tree. Providing the farmer s t i l l has unutilised time w i t h i n the family he may carry out the harvesting and stacking a t farm gate or roadside. Delivery o f cut wood t o the factory by animal c a r t transport may provide an additional income option over short distances. Expected farmer income i s given i n Annexure C2 Table 2.8. TABLE 8.4 - FARMER INCOME FROM FUELWOOD AT FULL LABOUR COST 4 Year Rotation a t 15% Wood Production 61.90 1 857 Harvesting 44.65 1 340 TOTAL 106.55 3 197 8.5 Opportunity Cost of Fuelwood Production Based on present prices and opportunities it i s more profitable f o r smallholders t o grow green l'eaf than market fuelwood. A present green l e a f y i e l d o f 3 770 kg/ha f o r an annual cost o f say 5 300 KSh/ha/y, and a net return o f say 22 000 KSh i s t o be compared w i t h 2 900 KSh f o r fuelwood. 8.6 I n s t i t u t i o n a l Matters and Constraints If a farmer had a choice between growing tea and fuelwood, he would grow tea as it i s far more profitable. However, there are about 265 000 ha o f land i n Kenya classed as suitable f o r tea growing (Jaetzwold & Schmidt 1983) o f which over 200.000 ha i s s t i l l unplanted. Similarly large areas unplanted t o tea or coffee occur i n the Tea-Coffee and Coffee Zones, and opportunities f o r expanding tea or coffee p lantings are severely constrained by 1i m ited export opportunities. Because o f t h i s constraint it i s assumed t h a t there would be some farmer i n t e r e s t i n growing fuelwood if KTDA were t o promote t h i s actively as a smallholder prograne and o f f e r guaranteed market opportunities i n the same manner as f o r the green leaf. To proceed, the following matters would require t o be addressed: 1. KTDA should decide how much fuelwood i s needed from the smallholders, a f t e r reviewing a11 other options suggested i n t h i s report. 2. KTDA should carry out a f e a s i b i l i t y study i n t o smallholder fuelwood production. 3. Assuming the project appeared feasible, it would then be necessary f o r KTDA to: . Carry out a research and demonstration woodlot programne. . Establish an advisory service for farmers i n appropriate methods of developing woodlots. . To d i s t r i b u t e seed and seedlings of the most suitable species f o r factory use and f o r farming systems. . Formulate a marketing structure, as for green l e a f v i a buying centres. . 9.0 BIOMASS FROM CROP RESIDUES 9.1 Forest Residues o f Forest Department 9.1.1 Resource Overview The harvesting o f indigenous forests i s now very limited, and according t o Presidential decree w i 11 v i r t u a l l y cease. For the purposes o f t h i s report consideration o f forest residues must therefore be r e s t r i c t e d t o p l antation materia1 . Hardwood p l antations, mainly o f Eucalyptus, have been establ ished 1argely t o provide poles and fuelwood. This resource has been covered i n an e a r l i e r chapter. Accordingly, forest residues t o be considered can only be Forest Department p lantation materi al, mainly conifers (Pinus patula, Cupressus l u s itanica and some P. radi ata. 1. Sources include residues from harvesting l e f t within the forest, and sawmill or other processing residues general ly l e f t outside the forest. For practical purposes harvesting resources following removal o f sawlogs, p l y logs and pulp logs have been assigned zero value f o r conifer plantations. Tree tops can be u t i l i s e d by local people f o r fuelwood, either formally or informally. As much as possible should be retained i n the forest, however f o r environmental purposes, especi a1l y nutrient recycl ing . 9.1.2 Sawmill Residues The Forest Department indicated that there were 1i m ited unuti1ised sawmi11 residues as about 85% o f m i 11s are located outside the forests close t o townships, with the following pattern of residue use: Residue Composition Sales - M i l l s near Towns 12% bark 100% sold attached t o slabs 15% sawdust Around 80% sold i n areas close t o towns 33% offcuts 100% sold as construction panels; slabs f o r fencing bU% of i n i t i a l volume For the 15% o f remote sawmi 11s the Forest Department estimates (Kahuki 1984) indicate a current domestic requirement of about 600 000 cu.m/y plantation roundwood far sawn timber and wood based panels. Assuming only the sawdust therefore i s not currently completely used, the available sawdust would be as follows: Total sawdust make 15% x 600 000 cu.m -- 90 000 cu.m/y Sawdust avai 1able .. Close Remote m i 11s 15%x 90 000 x 100% - 13 500 m i l l s 85% x 90 000 x 20% - 15 300 Total avai 1able - 28 800 cu.m/y A t density of say 0.Wcu.m a t 30% mcwb = 14 000 t / y Because sawdust oxidises, there would be a limited long term accumulation and the t o t a l sawdust inventory i s estimated at say 30 - 50 000 tonnes, much o f whith may be unsuitable. Forest Department records indicated (Table 9.1) t h a t there are 100 sawmills i n the KTDA d i s t r i c t s but with under 200 000 cu.m/y log input the usable sawdust residue i s therefore assumed at say 100 000 t/y, and widely scattered. FOREST DEPARTMENT RECORDS OF SAWMILLS I N THE KTDA DISTRICTS TABLE 9.1 - SAWMILL LOG INTAKE IN KTDA DISTRICTS 1983/84 PROVINCE Log Vol ume No o f D i s t r i c t m3 /a M i l l s CENTRAL K i rinyaga 2 390 11 Nyeri 46 195 Muranga' a - -11 Kiambu 77 520 . 16 126 105 38 RIFT VALLEY Nandi 2 000 1 Kericho 35 260 K i s i i - -16 37 260 17 EASTERN Meru 16 822 41 Embu 1 060 4 A l l D i s t r i c t s 181 247 100 The potential use o f sawdust by KTDA, therefore, requires considerably more study and collection o f data. Nevertheless Table 9.1 does indicate that, f o r a number o f factories placed reasonably near t o sawmi 11s (notably i n the D i s t r i c t s o f Nyeri, Kiambu, Kericho, Meru), they may well be viable opportunities. This report therefore commands further s i t e and factory specific study i n t o t h i s interesting option. 9.1.3 Conifer Plantation Thinni ngs A major opportunity f o r KTDA exists, and which should be explored with unuti 1ised n o n - c o m r c i a1 t h i n n i ngs. Forest Department records given t o the Consultant indicated about 240 000 cu.m o f pulpwood was harvested i n 1983 while Kahuki (1984) suggests the following y i e l d potential from conifer p lantati ons. TABLE 9.2 - PROJECTED PULPWOOD YIELDS FROM THE PULPWOOD MANAGEMENT DISTRICTS (1) Year o f Pla n t ing Harvesting Yield m3 - Note: ( 1) D i s t r i c t s o f Kakamega, Nandi, Uasin G i shu Investigations have been made i n t o establishment of a chemical pulpmill a t Thi ka, however the does appear doubtful . Forest Department advised t h a t the economic viabi l i t y It i s possible that non-commercial thinnings could be made i n these pulpwood plantations, chipped i n the forest and transported t o KTDA factories, a t least i n cases where transport distances would not render the operation uneconomic. As o f 1985 there appeared t o be about 140 000 t / y o f unused .conifer wood being produced which i s considerably more than the t o t a l demand o f KTDA even if all factories were converted to fuelwood f i r i n g . 9.2 Other Residues Kahuki (1984) has reviewed Kenya's b imass resources and the avai 1 abi 1ity and use i s sumrnarised i n Table 9.3. Rice husks are immediately available a t the cost o f loading and transport. Modification would be needed t o factory furnace systems t o accept r i c e husks as a f l u i d i s e d particulate fuel. Bamboo i s avai 1able but i t s communit y use and cost of harvesting and transport would need t o be examined. TABLE 9.3 - BIOMASS RESOURCES FROM CROP RESIDUES Resource t/Y Avai 1abi 1ity and Use Location Rice husks 6 000 Available and being burnt Embu d i s t r i c t Bagasse 290 000 Used as fuel i n sugar m i l 1s Various Bamboo 113 000 Avai 1able if economic t o Central 52 000 ha harvest Western 11 500 ha R i f t 40 400 ha Eastern 9 000 ha Straw Not known Used as animal feed Various Sisal 50 000 Used f o r sack, ropes, twine Vari ous Source: Kahuki C. D 1984 "Raw materia1 resources f o r obtaining fibrous m s f o r production o f paper and board. Forest Department Nairobi". 9.3 Cost o f C r o ~Residues Sawdust Assume royalty Assume cost o f loading, say Transport per km Delivered cost 10 km 100 km 200 km Conifer Thinnings Assume cost as f o r present Forest Department p1antat ion eucalyptus Delivered cost 10 km 100 km 200 km ' Rice Husks As f o r sawdust. Bamboo Probably similar t o conifer thinnings but add say 20% t o costs. .' 10.0 SUMMARY OF BIOMASS RESOURCE OPTIONS AND COSTS 10.1 Pricing o f Biomass Energy The mid 1985 cost o f energy from f u e l o i l derived t o KTDA f a c t o r i e s was about 77.6 KSh/GJ f o r East o f R i f t and 82.8 KSh/GJ f o r West o f R i f t plus Kiegoi (Meru d i s t r i c t ) . Based on these o i l energy costs, the delivered costs o f fuelwood shown i n the summary Table 10.1 could be allowed t o r i s e by a f a c t o r o f 3.3 t o 3.6 f o r the smallholder model, 2.2 t o 2.3 f o r the public lands model and 3.2 t o 3.5 for the Forest Department model. However, i n considering an acceptable delivered cost f o r fuelwood KTDA must take i n t o account current 1985 market prices f o r fuelwood i n t h e main centres. These are: Nairobi Kisumu Nakuru Eldoret Source: MOERD Nairobi 1985 A l l t h e fuelwood models developed i n t h i s report give a delivered cost t o KTDA f a c t o r i e s higher than t h e highest market p r i c e (Nairobi ). It i s generally agreed t h a t present market prices are f a r t o o low, and must r i s e a t l e a s t t o meet production .costs. The low prices probably r e f l e c t the philosophy o f fuelwood being a commodity, u n t i l recently "free" except f o r the 1abour cost o f gathering. With tightening policies, which w i l l r e s t r i c t access t o natural forest resources, dependence o f supply would s h i f t t o plantation fuelwood crops which would have t o be sold a t least a t prices indicated i n the Consultant's models. It i s concluded that KTDA would need t o accept the indicated model costs i n Table 10.1. An economic case could be argued f o r g i v i n g smallholders a p r i c e subsidy above the indicated prices ifan additional incentive were required t o encourage farm-based comrnercia1 wood1o t production. The economic case could be based on say a 20% shadow cost added t o t h e K'rDA p r i c e o f o i l because o f foreign exchange shortage. A summary o f fuelwood quantities and costs by production models i s given i n Tables 10.1 and 10.2. 10.2 Conclusions on Supply Options f o r KTDA Factories 1. KTDA p o l i c y i s t o obtain the maximum supply o f fuelwood from the Forest Department. This would i n due course be e n t i r e l y from fuelwood plantations, mainly eucalyptus, as indigenous f o r e s t supplies become unavailable. Indications are t h a t KTDA would be able t o negotiate a share (assumed a t 50%) o f projected y i e l d from these Forest Department fuelwood p l antations, though the percentage should be confirmed and Forest Department y i e l d be subject t o inventory. 2. P r i v a t e estates could have a p o t e n t i a l fuelwood surplus as energy e f f i c i e n c y improves. However, not o n l y are t h e estates l j k e l y t o p r e f e r t o p l a n t t h e i r surplus fuelwood areas w i t h t e a b u t also t h e opportunity cost o f t h e surplus fuelwood i s l i k e l y t o make t h i s energy form too c o s t l y f o r KTDA. Underuti lised p u b l i c lands such as roadsides and other areas have been successfully used by BAT f o r fuelwood production. It i s possible they could provide 25% o r more o f fuelwood needs f o r a l l 39 KTCA f a c t o r i e s a t c u r r e n t e f f i c i e n c y levels. The p r i c e o f delivered energy could however be t w i c e as h i g h as f o r other models mainly due t o t h e high c o s t o f fencing t o p r o t e c t roadside plantings. However, t h i s supply model could d e l i v e r energy a t about half t h e 1985 cost of o i l , and t h e experience of BAT i n using such a model would make it worth d e t a i l e d exami nation by KTDA. 4. The smallholder model could t h e o r e t i c a l l y supply more fuelwood than a l l 39 KTDA f a c t o r i e s would r e q u i r e a t c u r r e n t e f f i c i e n c y levels. The KTDA f a c t o r y system supplied by 150 000 farmers i s by i n t e r n a t i o n a l standards one o f t h e most successful smallholder projects. There i s no t h e o r e t i c a l c o n s t r a i n t f o r t h e same smallholders t o produce fuelwood f o r KTDA f a c t o r i e s , providing t h e p r o f i t margin was adequate and t h e market assured. The smallholder model could produce fuelwood a t no greater cost than from t h e Forest Department and w i t h wider spread o f socio-economic b e n e f i t . Some reservations are held by KTDA concerning t h i s model which w i l l r e q u i r e very c a r e f u l evaluation. The model would need t e s t i n g and farmers' reactions determined. A programme f o r supplying say 25% t o 50% of KTDA needs should be investigated. 5. Crop residues and sawdust could provide some or, i n f a c t , a l l of KTDA needs. 6 000 t / y o f r i c e husks are c u r r e n t l y being burnt t o waste and could be used w i t h appropriate combustion p l a n t modification. Coffee husks are already used for combustion i n i n d u s t r i e s i n Nairobi, w h i l e sawdust also f i n d s some i n d u s t r i a l use. However w i t h sawdust there i s s t i l l q u i t e a substantial surplus which could be made available. There would appear t o be a surplus of about 140 000 t / y o f Forest Department coniferous roundwood avai l a b l e due t o t h e proposed pulpmi 11 p r o j e c t n o t proceeding. The economic availabi lit y of t h i s resource, as we11 as non- commerci a1 thinnings, should be explored by KTDA. 6. A number o f biomass and biomass residue supply options have been i d e n t i f i e d which i n t o t a l could supply energy t o a l l 39 KTDA f a c t o r i e s a t well below t h e 1985 cost o f o i l energy. Further i n v e s t i g a t i v e work would need t o be c a r r i e d out by KTDA on these options, and the Forest Department would need t o c a r r y out the f o l l o w i n g tasks. . Inventory a11 fuelwood p l a n t a t i o n s and review p l a n t a t i o n management programmes. . Make c l e a r decisions on f u t u r e a l l o c a t i o n s o f p l a n t a t i o n fuelwood t o users such as KTDA. . Assuming inventories of conifer p l a n t a t i o n s are o f adequate standard, make c l e a r decisions regarding a l l o c a t i o n s o f roundwood (other than f o r sawlog o r p l y l o g ) f o r uses such as pulp logs and f u e lwood. TABLE 10.1 - SUMMARY OF FUELWOOD COST IN KSh BY PRODUCTION MODELS FOR EUCALYPTUS AT 15% INTEREST BUT EXCLUDING PROFIT MARGIN Model Standing Tree Harvesting Transport ELStacking Total Delivered cost K S ~per: m3 t a GJ m3 t GJ m3 t GJ m3 t GJ Forest Department ( 1) 8 y rotation 67.53 128.15 9.03 44.65 84.72 5.97 67.30 127.70 8.99 179.48 340.57 23.99 4 y rotation 71.11 134.66 9.49 44.65 84.72 5.97 67.30 127.70 8.99 183.06 347.09 24.45 Estate A (2) 8 y forest s i t e 31.88 60.52 4.26 30.83 58.50 4.12 83.33 158.12 11.13 146.04 277.14 19.52 Estate El (3) 8 y forest s i t e 56.39 107.00 7.54 40.92 77.65 5.47 99.06 187.97 13.24 196.37 372-62 26-24 - KTDA Factory model (4) Eucalyptus 19.58 varies Mative 19.58 varies 50% of each 19.58 varies I Snallholder Model i Eucalyptus I 4 y f u l l price 61.90 117.46 8.27 44.65 84.72 5.97 67.30 127.70 8.99 173.85 329.89 23-23 Public Lands Model 4 y f u l l price 152.99 290.31 20.44 44.65 84.72 5.97 67.30 127.70 8.99 264-94 502.73 35-40 (1) For royalty 8 y see Annexure C2 Table 2.2 and f o r 4 y see Annexure C2 Tables 2.3 and 2.4. Uses E. saligna 28 cu.m/ha/y but uses KTDA transport costs as realistic. (2) Uses E. saligna 56.2 cu.m/ha/y. (3) Forest s i t e = clear forest c/w grassland s i t e 8 y uses E. saligna 48 cu.m/ha/y. 7.5 y rotation uses E. grandis = 80 cu.m/ha/y. (4) Derived from fuelwood using factories Kangaita, Githongo, Imenti . See Table 4.1. (5) Derived from Annexure C2 Table A2.7. Uses Eucalyptus 30 cu.m/ha/y. Labour shadow priced a t 60%. (6) Imenti factory harvesting cost. Equates with Forest Department costs (Kimini EL W o k 1983). TABLE 10.2 - BIOMASS SUPPLY POTENTIAL AND KTDA FACTORY NEED Source Potentia1 by Province t l y Cost Delivered (5) Central R i f t Eastern Total KSh/t KSh/GJ Forest Department (1) 8 y model 24 000 5 000 4 000 33 000 321 22.6 Private Tea Estates 4 y model Very l i m i t e d Unused Pub1i c Lands 4 y model 11 000 7 000 7 000 25 000 560 39.5 Small holder 4 y model 84 000 29 000 123 000 236 000 311 21.9 (2) (3) (4) (4) Crop Residues 140000 6 0 0 0 146000 491/210 34.6/14.8 Total Supply 119 000 181 000 140 000 440 000 KTDA Demand 50 000 36 000 14 000 100 000 Source: Previous tables - Note: (1) Assumes 50% current potential wi 11 be available t o KTDA. (2) Unal 1ocated round c o n i f e r smal lwood o f Forest Department. Assume present KTDA delivered cost eucalyptus 291 KSh/t plus transport 100 km @ 2 KSh/tkm = 491 KSh/t. Distances t o f a c t o r i e s t o be confirmed. (3) Rice husks. Assumed loading 10 KSh/t and 100 km transport @ 2 KSh/tkm. (4) Small wood/rice husks. (5) Excludes any p r o f i t margin. ANNEXURE C1 - REFERENCES Beij e r I n s t i t u t e (and Scandinavian I n s t i t u t e o f African Studies 1984. Wood, energy and households, (ed C Barnes, J Ensminger, P O'Keefe) Sweden. GTZ (updated) Special energy programe i n Kenya. Woodfuel plantations i n Kenya. Nairobi. Beij e r I n s t i t u t e 1982. The estimation of land use and fuelwood supplies i n Kenya. A methodological note. Holder R H. Stockholm. Beijer I n s t i t u t e 1981. Wood supply and demand i n Kenya. K Openshaw, Stockholm. FA0 1982. Report on a f o r e s t r y sub-sector review and programing mission K t h e Republic o f Kenya June 1980. C Chandrasekharen, Rome, Jaetzold R. & Schmidt H. 1983. Farm management handbook o f Kenya. Pol I 1 Part A. West Kenya (Nyanza & Western Provinces) Part B. Central Kenya ( R i f t Valley & Central Provinces) Part C. East Kenya. (Eastern 8 Coastal Provinces). M i n Agri c. Nairobi and GTZ Germany. Kahuki 1984. Estimated sawlog and roundwood supply and demand f o r Kenya. Forest Department Nairobi . Kenya Tea Development Authority (KTDA) 1984. Annual Report. Nairobi Kenya Tea Growers' Association (KTGA) 1983. Executive o f f i c e r ' s s t a t i s t i c a l report t o r 1983. Kericho Kimini G N T & Ornbok C 0 1983. Royalty requirements o f eucalyptus fuelwood plantations i n Karura and Ngong forests. Forest Deptartment Nairobi . Kuloki T 1982. Woodfuel requirements f o r B.A.T Kenya L t d and experience i n t r e e planting (from: Agro-forestry systems f o r small scale farms. Doc ICRAF/BAT workshop he1d Nairobi September 1982). ICRAF Nairobi Mercado R S 1985. Universidade Federal Rural Do R i o De Janiero, Brazi 1. Personal communicati on. M i n is t r y o f Energy and Regi onal Development 1985. National Energy Policy. ma1rob1 . 0 enshaw K 1982. Costs and benefits o f proposed t r e e planting programe h y i n g Kenya's wood energy requirements. Beij e r I n s t i t u t e . Stock holm. 0 enshaw K 1982. Inventory on biomass i n Kenya - conditionally renewable &he Beijer I n s t i t u t e for Energy and Human Ecology. Stockholm. ANNEXURE C2 - MODELS FOR FUELWOOD PRODUCTION 2.1 Forest Department 2.1.1 Eucalyptus Plantations The Forest Department (1983) carried out a detailed study o f costs of growing eucalyptus f o r fuelwood on an 8 year r o t a t i o n w i t h two following coppice rotations. The r o y a l t y suggested i s based on actual costs and 10% discount rates be1ow. TABLE 2.1 - ROYALTY FOR EUCALYPTUS FUELWOOD TO COVER COST OF PRODUCTION 8 YEAR ROTATION AT 10% DISCOUNT (2) Royalty (1) Ksh/m3 Ksh/m3 Ksh/m3 Ksh/m3 Ksh/m3 KShlt KSh/GJ With land r e n t 60.32 71.36 84.56 95.55 107.49 203.97 14.36 Without land r e n t 32.11 37.99 45.02 50.87 57.23 108.60 7.65 Source: Forest Department 1983 for prices t o 1982/83. I n f l a t e d using Study Team data t o 1984/85. 3 Note: (1) Volume ni (cu.m) i s solid. The sale p r i c e i n the forest, without land rent, has been derived adding labour costs and net p r o f i t as i n Annexure C2 Table 2.2 below. This i s the p r i c e a t roadside t o which transport and loading costs are added t o obtain delivered cost t o factory. (2) Forest Department data does not a1low recalculation a t 15% TABLE 2.2 -SALE PRICE AND NET PROFIT FOR EUCALYPTUS FllELWOOD 8 YEAR ROTATION AT 10%DISCOllNT (2) Price a t Roadside Delivered Price Factory (1 1980/81 1984/85 1984/85 Royalty 32.11 57.23 Labour Costs 12.08 21.53 Net P r o f i t 13.18 23.49 Sale Price Source: Forest Department 1983 for 1980/81 prices. Note: (1) Assume lOkm road transport a t 2.0 KSh/tkm plus 10.00 KSh/t f o r loading. (2) Forest Department data does not allow recalculation a t 15% 2.1.2 Plantations Establishment An internal Forest Department document (Sept 1985) gave the cost o f conifer plantation establishment from 14 stations. Costs were stated t o include overheads such as: . Supervision and admin i station. . Maintenance (roads, bui l d ings, water supply) . Transport . Protection ( f i r e , patrol 1ing, boundary cleaning, etc) . Replacement o f bui1dings, vehicles, other equipment The costs were f o r a 30 year rotation coniferous timber production plantation but were modified by the Consultant f o r a 4 year eucalyptus plantation. TABLE 2.3 - COSTS KSh/ha EUCALYPTUS FUELWOOD PRODUCTION 4 YEAR ROTATION Year 1 2 3 4 Nursery Plant r a i s i n g 844 Nursery maintenance 278 Plantation Establishment Land demarcation Land c learing Cutting stakes Staking out P i t t i n g Planting Survey Beating up Weeding 1 Weeding 2 Climber cutting Source: Forest Department Yield would be 28 cu.mlha/y f o r the best sites t o 23 cu.m/ha/y for medium qua1it y sites. With adequate f e r t i 1iser inputs more .than 30 cu.m/ha/y should be a achievable. Annexure C2 Table 2.4 shows the cost o f production or royalty f o r high and medium sites without land rent. : TABLE 2.4 - COST OF FUELWOOD PRODUCTION (ROYALTY ONLY) 4 YEAR ROTATION Ksh/m3 SOLID Yield 28m3/ha/y 23m3/h a/y Interest rates ~ ~ h / m ~ KSh/t KSh/GJ Ksh/m3 KSh/t KSh/GJ Source: Derived from Annexure C2 Table 2.3. - Note: (1) lm3 s o l i d = 0.527t 30% mcwb = 14.2 GJ/t. (2) No f e r t i l i s e r used. Using an 8 year r o t a t i o n a t 28m3/ha/y production costs i n Annexure C2 Using an 8 year r o t a t i o n a t 2&n /ha/y/ production costs i s Annexure C2 Table 2.4 are changed as follows: 2.2 Estate A Data was supplied by Estate A for cost of production f o r E.saligna f u e lwood on an 8 year rotation, and i s given be1ow. TABLE 2.5 - COST OF EUCALYPTUS FUELWOOD PRODUCTION 8 YR ROTATION (KSh/ha) 15% Discount Years 1 2 3 4 5 6 7 8 Remove trash l i n i n g 56.21n~/ha/~s o l i d hol ing, planting , spray 29.6 t/ha/y 2 times 1 400 Plants 4400/ha 800 Weedicide & fert. NPK 1 oz. x 4400/ha 200 Level o l d trash, put i n roads 500 Beating up, spray 2 x 710 Weedicide & NPK 360 Spray 1 x slash 460 1 x weedicide 80 Spray 1 x slash 330 I(sh/m3 stacked basis Harvesting F e l l & s p l i t 18.50 13 875 Transport t o factory 42.00 31 500 Stack a t factory 8.00 6 000 TOTAL 3 970 540 330 51 375 Royalty A t Roadside Del. & Stacked - Cost ~ ~ h / rs onl~i d 31.88 62.71 146.04 KSh/t 60.52 119.02 277.25 KSh/GJ 4.26 8.38 19.53 2.3 Estate B Data was supplied by Estate B f o r cost of production f o r E. sal igna fuelwood on an 8 year rotation, and is-given below. TABLE 2.6 - COST OF EUCALYPTUS FUELWOOD PRODUCTION 8 YR ROTATION KSh/ha 15%DISCOUNT) Year F e l l & Clear 2 560 48m3/haly s o l i d Dig & fork 788 25.3 t/ha/y Line, hole, plant 1 181 Weed 656 Terrace 394 Roads 500 P l ants 1600/ha 1 000 KSh/m3 stacked basis Harvesting 24.55 Transport 56.30 Stack a t factory 3.14 TOTAL 7 079 Royalt y A t Roadside Del. & Stacked - Cost ~ ~ h sol / mi d~ 56.39 97.31 196.37 KSh/t 107.00 184.65 372.63 KSh/GJ 7.54 13.01 26.24 2.4 Smallholder Fuelwood Production S t a f f of the Beij e r I n s t i t u t e i n Nairobi advised that the mean volume y i e l d measured on smallholder woodlots was 114 cu.m/ha f o r eucalyptus. Assuming a 4 year rotation as reasonable f o r production of poles/fuelwood f o r sale, t h i s gives about 30 cu.m/ha/y y i e l d i n the tea zone with low cost inputs. The Consultant has made assumptions i n producing the c m t model i n Annexure C2 Table 2.7 by taking his selected fuelwood growing costs from records of two major estates. The costs chosen are almost e n t i r e l y farmer's labour which has been shadow costed at 60%, a figure assumed by the Consultant a f t e r discussion with World Bank Peri-Urban Fuelwood Study Team i n Nairobi . TABLE 2.7 -COST OF FUELWOOD PRODUCTION -- - 4 YRS ROTATION KSh/ha (Smallholder1 Years 15% interest Plough o f grass1and F u l l Shadow 301n~/ha/~s o l i d 5 hrs @ 100 KSh 500 500 15.8t/ha/y Plants 5000/ha 900 540 Dig & f o r k 788 473 Line, hole p lant 1 1 8 1 709 Weed 656 394 Maintenance 68 TOTAL Roya1ty F u l l Shadow - Cost ~ ~ h / sr onl i~d 61.90 34.39 KSh/t 117.46 78.54 KSh/GJ 8.27 5.53 - Note: (11 Shadow p r i c e labour a t 60% 68 = f u l l priced labour - 41 = shadow priced labour ANNEXURE C3 - BIOMASS DEMONSTRATION PROJECT 1.O PROJECT SCOPE If KTDA proposes t o change i t s tea factories t o using biomass as a source o f energy for drying, there are a number o f questions t o be answered, constraints t o be overcome and policy matters t o be dealt with. The Consultant believes the easiest means t o effect the changeover i s through a small p i l o t project which would examine the resource questions i n detail and explore the ways o f obtaining and growing fuelwood. It would serve t o transfer the necessary technology, t o KTDA s t a f f and farmers i n particular, and help make the social and i n s t i t u t i o n a l changes needed t o establish KTDA as a viable fuelwood market. The project scope i s outlined as follows: 1.1 Forest Department Resources . Carry out Forest Department fuelwood plantation inventory. . Develop management plans for existing plantations, including. rehab- ilitation of substandard plantations. . I d e n t i f y other Forest Department sites which may be planted consistent with present government policy of no further clearing o f indigenous forests. . Planning o f fuelwood or multi-purpose plantations on sites. . Inventory Forest Department industri a1 conifer plantations, especially those from which small roundwood may be underuti lised or f o r which there i s no foreseeable market. Confirm opportunities f o r KTDA fuelwood supply. . Agree fuelwood and other product allocations from all hardwood plantations and appropri ate conifer plantations. . Agree royalty schedule f o r fuelwood which r e f l e c t s the real cost o f production, which can be used as a benchmark f o r general fuelwood pricing i n Kenya and which encourages the production o f fuelwood as a market crop f o r the private sector, especia1l y small holders. 1.2 KTDA Fuelwood Proarammes . Carry out a survey t o i d e n t i f y areas of public land which may be used f o r fuelwood production. Areas w i 11 include roadsides, schools and other pub1i c buildings, Provincial, D i s t r i c t or County Council managed areas, etc. . Confer with local authorities and develop viable programmes which w i l l involve community participation i n costs and benefits. . Consult with BAT regarding i t s successful fuelwood programme. . Plan and commence a p i 1o t programe o f say 200 ha f o r 1986 and 1987 with a f i n a l target which could be of the order o f 1000 t o 2000 ha or more. . Agree with smallholders on fuelwood prices a t stages of on-farm royalty, harvested and del ivered. . Determine market opportunities f o r specific factories and target dates for supply levels. . Determine fuelwood species and stem diameters t o s u i t recommended cyclonic combustion technology. . Develop an extension service t o supply good seed and advise farmers on t r e e growing. Spread the techno1ogy o f innovative farmers. . Develop a fuelwood marketing structure which may be along the 1ines of the present KTDA system f o r green leaf. . Decide on a target f o r fuelwood supply from smallholders which may be say 50 000 t / y coffee rotation. The programme could begin with a p i l o t phase involving say 5 factories i n 1987 and increasing t o 10 i n 1988 t o a l l 39 by 1989. b 1.3 Crop Residues KTDA should examine the f e a s i b i l i t y of using the 6 000 t/y o f r i c e husk ava.i 1able i n Embu d i s t r i c t . This would require modifications t o existing factory combustion systems taking i n t o considerati on the high s i 1i c a content o f t h i s biomass. . 1.4 Sawdust KTDA should further investigate the avail able sawdust potentia1 i d e n t i f i e d by the Consultants. 2 .O ESTIMATED PROJECT COSTS T The project would assist KTDA i n the p1,anning and conversion of a11 factories t o biomass f u e l and i n the establishment o f p i l o t scale fuelwood production systems. It should establ ish a number o f small p lantations along roadsides and on available public lands t o act as demonstrations t o encourage local involvement and t o act as p il o t scale technical and social t r i a l s . The same kind o f programme should be extended t o individual farms and both should be supported through an active extension service which would involve d i r e c t contact with local communities possibly using an audio-visual unit. The l i k e l y cost o f such programnes would be: . KSh ('000) Public lands 200 ha of demonstration @ 12 500 KSh/ha 2 500 . . Smallholders s 1000 farmer demonstration plantings 100 ha @ 5 000 KSh/ha 500 Demonstration nurseries 10 @120 000 KSh 1 200 . Audi o-Vi sual Training Unit Capital cost of van and equipment 250 . AdvS sory Services (1)' Forestry p lanner/soci a1 forester 24 months Extension/institutions speci a1is t 24 months Other short term 12 months . Transport/Equi pment 2 4wd vehicles Other TOTAL ESTIMATED COST Note (1) UNDP budget rate $US 15 000/month ANNEXURE C4 - BIOMASS SPECIES AND PROVENANCES 1.0 Biomass S ~ e c i e si n Current Use Various Eucalyptus species have been introduced t o Kenya and several have proven t o be highly successful. A primary objective was t o produce fuel f o r the railways, although t h i s was no longer required when diesel-electric engines were substituted f o r fuelwood f i r e d steam engines. O f the Eucalyptus species introduced, E. saligna was found t o be the most suitable for use as fuel wood and v i 11age bui1ding materia1s. The Forest Department has mainly used E. saligna i n i t s hardwood planting programmes along with some Acacias. I n the Tea Zone E. saligna has likewise been selected and yields produced have been very high by world standards. More recently one o f the private estates has introduced E. grandis, increasing yields i n volume, wood weight and net energy. 2.0 Biomass S ~ e ces i f o r Future Production 2.1 Forest Department Plantat ion The main constraint t o high wood y i e l d i n the Forest Department plantations appears t o be management rather than better species selection. However, improvement i n y i e l d . can be achieved by careful attention t o selection, especi a1l y o f appropri ate geographical 1ocations o f species c a l led provenances. The fuelwood production programne should continue w i t h the following basic Eucalyptus species w i t h suggested associated tria1s. These can be started as part o f the proposed KTDA/Forest Department Project as follows. Tea Zone Using E. saligna as a standard f o r comparison, establish t r i a l s of:- . E. grandis - North Queensland and North NSW provenance. . E. globulus - best provenances from Tasmania. . Acacia mangium and A. auriculiformis from North Queensland. While the Acacias are valuable nitrogen producers, t h e i r a b i l i t y t o coppice i s not high. Coffee Zone With lower r a i n f a l l , eucalypts from d r i e r parts of Northern Queensland are candidates. E. camaldulensis of Petford provenance i s suggested as a basic choice. T r i a l s should include E. tereticorni s from Northern Australian provenances and Leucaena 1eucocephal a (Hawaiian giant s t r a i ns) where soi 1 pH i s above 5.5. 2.2 Smallholder Plantations The smallholder model i d e a l l y should use tree species which have the f o l lowing features: \ a. Fast growth b. Coppici ng abi 1it y c. Nitrogen f i x i n g d. Fodder production from leaves e. Produce fuelwood and bui 1ding materials. Clearly a l l the features are d i f f i c u l t t o f i n d i n one species and eucalyptus are d i f f i c u l t t o exclude although they do not include features c and d. Some species o f Casuarina feature a, c and e but are deficient i n b and d. Some legumes are useful candidates, though the Acacias are general l y deficient i n some features. Species such as Leucaena leucocephala and i t s close relatives are candidates and may be used as a mixture with, say, eucalyptus, i n appropriate spacings, as we11 as i n agroforestry production systems. Suggested species f o r small holders i n the Tea Zone are as follows: E. saligna, E. grandis (North Queensland) E. globulus (best Tasmanian), Casuari na e q u i s e t i f o l i a, Leucaena leucocephala ( s o i l over 5.5 pH) Acacia mangium and A. auricul iformi s (North Queens1and). 2.3 Research Proaramme The KTDA should cooperate with other agencies i n developing a research programme t o establish the most appropriate species and provenances. These agencies are: .. Forest Department .. International Centre f o r Research i n Agroforestry (ICRAF) Australian Centre f o r International Agricultural Research. (ACIAR) Kenya Woodfuel Development Project (Beij e r I n s t i t u t e ) . T- t Q. Q E cn a W CT a acn z w 3z 00 ul- a 0 #< W k 6 Y ad El- ~2 94 -I 1 a t cn -- . e 0 .a a P E 3 8- a- 8- Q- C K R- 3 0- i: i B . O D.F.O. KAPENGURIA FOREST RESOURCES C o n s e r v a n c y C o n s e r v a n c y D.F.O. SAMBURU D.F.O. LAKPlA 05.0. MERU n --Lonservancr I Bcnmd4rles vl w ------ Dlstrkt Forest Olfke boundaries Forest statlorn o -1.S. stat- 8 Forest and RILLS. rtatlons L( N ~ ~ I ~ M I capnal Q Provhclal capltalr Provhclal Bcnmd4rles hterallmal bourdarles Nakuru o za 40 m m-wn I . L . 1 . I . I l I 1 I I I , 0-F.0. NAROK O O 2 0 5 0 U ) S Q M TANZANIA \ MAP 2 - EAST KENYA ' I Provlnclal Headquarter District Headquarter MAP 3A CENTRAL KENYA ADMINISTRATIVE BQUNDARIES a I ...' i\ \ .-. i II RIFT \ I '\ / WEST >\ /-I - i ,/I \ \\ * POKOT f' -\ '\ \ 1 '--+* \ / SAMBURU \ \ \ / '- \ Maralal \ \ \ f? C---, <------- /- ,/-· '+-A LAIKIPIA 0' i -*-- 0 es Narok i '\ L.. -- NAROK / / ' - / '-.'.. 2, - I' 1 \ I '.-'.. '\ Kallado '1 . '.'- 0 IOOkm '.* I C,L, I 20s -------International *\/. KAJlADO Boundary ' '.. 3 .------ '.. 8 Provincial Bouncrary \ ------- '. . District Boundary '.. '\, Provincial Headquarter '.. . District Heddquarter "*,MAP 38 '.. ' . I I I I I APPENDIX D SUPPLIES AND PRICES OF FV&S APPENDIX D - SUPPLIES AND PRICES OF FUELS 1. INTRODUCTION This Appendix describes the relwant fuels in Kenya and their price ranges fuel prices used to evaluate and test the sensitivity of the investment reccmnendations mde in the study. There are generally three prices used being base, financial and economic. The base prices were those collected at the individual plants and tea factories audited. The reader should refer to the individual Energy Audit Report for each plant or tea factory to obtain the base price used at that particular plant or factory. local variations were found to be substantial even between enterprises located quite close to each other. The economic and financial prices are figures developed from observation and analysis and used to assess the project proposals initially selected using the base prices. This Appendix is largely devoted to the derivation of the financial and economic price ranges. The ranges -re first developed before the oil price declines of 1986 but then adjusted during the sensitivity testing to provide a rigorous basis for testing the -dations. 2. CQAL SUPPLIES Kenya's present coal supplies cane f m Mziland at a CIF (cost insurance and freight)Mmbasa price of $US 50 per tonne. A duty of 8 percent applied in 1985. This coal has 21 percent ash and 14 percent Volatiles. In subsequent calculations 23.5 GJ/to~ehas been used for the gross specific energy of this coal. The coal is unusual because of its low volatile content and this property together with high ash makes it difficult to burn in conventional boilers. It is suitable for cement making but not for boilers. Swazi reserves are thought to ccanprise 500 million tonnes which are considered adequate for existing uses but not for substantial substitution for oil. Kenya appears to have taken 85,000 tonnes of the 125,000 tonnes 1984 production f m the Mpaka mine. Supplies are prone to interruption by insurgency activities in Mozambique and alternative supplies of similar coals are not readily available froan neighbowing countries. It muld not be advisable for Kenya to embark on a mjor coal substitution programne on the basis of this coal In addition to the possible supply difficulties, boilers would have to be specially designed for it, at extra cost, and muld be incapable of burning more readily available substitute coals in the event of an interruption to supply Other coals are likely to be substantially more expensive, especially in the relatively small volums likely to be used in Kenya. Coal in 1985 is available £ran Australia and other countries at about $US 34 per tonne FOB, but rates for mall tonnages are likely to be higher than those paid for Swaziland coal. Available data for deliveries to Europe in July 1985 range frun $US 41 to $US 54 per tonne (CIF). It appears that coal for Kenya might be reliably obtained at $US 35 per tonne FOB, plus an allowance of $US 15 per tonne for freight and a $US 5 penalty for M m b s a and small shipnent sizes. The key reasons for the $5 penalty relate to the size of shipnent and the unloading facilities likely to be available. Shipnent sizes of 20,000 tonnes and less are not attractive to scmre coal exporters because they occupy facilities designed to load much larger volums (60,000 to 150,000 tomes). Small self-unloading vessels are costly to charter and find it difficult to obtain continuous loading unless they are part of a distribution trade. In 1986 Kenya can expect to pay $US 55 per tonne CIF for thermal coal for annual tonnages of less than 300,000 tonnes. Should annual sugply contracts reach 1,000,000 tomes to be delivered in ships of 40,000 LWT, then rates could fall below $US 55 per tonne CIF Mombasa, possibly reaching $US 50 per tonne. 2.1 Coal Prices in Kenm Coal is imported to Kenya and in 1985 was subject to import duty at the rate of 8 percent for coal sourced in the Preferential Trade Area (PTA) of Eastern and Southern African countries and 20 percent for other sources. Swaziland coal benefits f m n the PTA concession. Coal for use outside the cement industq would' be a coal with higher volatile content than Swaziland coal and is expected to have a price of around $US 55 per tonne CIF Manbasa. TABLE D.1 CIF PRICES OF COAL AT MCMB?SA -----,--,----,----,------------- Price basis _--_-----_-_-_--------------------------------------------------------- Swaziland Coal T h e m 1 Coal CIF ex duty 50 55 CIF with duty _---_-_--_---_--_------------------------------------------------------ 54 66 Note: CIF stands for Cost, Insurance and Freight. 2.2 Coal Handlinq Coal is at present received by Bamburi Portland Cement and handling figures are based on their estimates provided in May 1985. These have been confirmed by observation of handling methods. TABLE D.2 ESTIMATED HANDLING COSrS l3X CQAL (KSh/t ----------------------------------------------------------------------- Swaziland Thermal ----------------------------------------------------------------------- Demurrage 32 Wharfage 1.5% CIF L/ 13 Insurance, Brokerage, Interest 20 Import Licence Fee 16 Handling Cost 18 Handling lasses (2%) 19 Rail Imding 25 - 11 Converting $US at KSh 17 per $US. 2.3 Transpart Costs Kenya Railways has provided freight rates while road transport costs were collected from the Keyna Transport Association. The results are shown in Table D.3. TABLE D.3 ...................................................................... Rail ..................... Destination Coal Oil Road ...................................................................... KSh/t Nairobi 300 355 585 Eldoret 450 543 958 Athi River 290 326 500 Many of the plants audited receive their oil supplies by road even though rail terminals are available nearby The reason for this appeared to be unreliability of rail delivery. This suggests that the econanic costs of delivery of oil or coal by rail are understated if rail freight rates are used. Similarly, although road freight rates are significantly higher than rail freight, there is evidence fram 1981 (1) that road freight charges do not provide for full recovery of the costs of road damage and war. Subsequently road licence fees have increased and toll charges have been introduced to recwer same of the cost of road damage and wear. For the purposes of calculation, the costs of rail delivered coal were campared with the costs of rail delivered oil and no provision was mde for added rail costs to make oil delivery reliable. 2.4 Coal Unloadinu and Handlincr into Plant The cost of coal to a user will depend upon the cost of unloading and handling coal into plant. ....................................................................... (11 Kampsax International, Studv of Road User Charaes and Axle Laad Limits, Final Report (Nairobi: Govennent of Kenya, 19811. Discussions at Bamburi Portland Cement in 1985 indicated that handling coal out of a stockpile into trucks for on shipnent is about KSh 10 per tome (not including transport costs). This estimate is based on KSh 800 per hour including equipnent hire and a handling rate of 80 tomes per hour. Table D.3 already includes an allowance of KSh 25 per tonne for loading onto rail. A further KSh 25 per tonne is then required for unloading and handling into plant (exceptMcanbasa). 2.5 Coal Costs The cost of coal delivered by rail into the plants in the various centres can be mde up f m the previous data. The.resultsare shown in Tables D.4 (KSh/t)and D.5 (KSh/GJ). TABLE D.4 CQAT.ICOSTS DELIVERED INIV PLANT ...................................................................... Destination Swaziland Coal Them1 Coal ($US55 C.1.F) (withduty) (exduty) (withduty) (exduty) ...................................................................... Nairobi 1386 1318 1591 1404 Eldoret 1536 1468 1741 1554 Magadi 1406 1338 1611 1424 Athi River 1376 1308 1681 1394 TABLE D.5 COAL COSTS DELIVERED INIO PLANT (KSh per GJ> Destination Swaziland Coal Thermal Coal ($US55 C.1.F) ......................................................................(ex (with duty) (exduty) (with duty) duty) Nairobi 59.06 56.17 59.37 52.39 Nakuru 62.05 59.15 61 98 55.00 Eldoret 65.46 62 56 64.96 57.99 Magadi 59.92 57.02 60.11 53.12 Thika 59.92 57.02 60.11 53.13 Athi River 58.64 55.74 62.72 52.01 Note: Calorific Values Swaziland Coal 23.46 =/tonne T h e m 1 Coal 26.8 =/tonne 2.6 Financial and Econcanic Costs of Coal Applying exchange rate and labour cost adjustments to each of the coal delivery steps enables the calculation of economic costs. Table D.6 shows the subdivision of costs into foreign and local components and the resulting financial (ornrarket) prices and econcanic costs. 2.7 Forecast Coal Prices Future prices of coal are forecast to remain fixed at current levels in real tenns. KENYA: FINANCIAL AND ECONaMIC PRICES OF COAL 1985 (KSh/tonne) (DELIVEREDNAIROBI Swaziland Theml* Financial Economic (23 5 GJ/t) (26.8 GJ/t) Foreign &a1 Foreign Local Swazi Thermal Swazi Thermal ........................................................................... F.O.B. Port of 510 - 595 - 510 595 612 714 Origin Shipping 255 85 340 - 340 340 391 408 .......................................................................... C.I.F. SUB KSh/t 765 153 935 187 918 1122 1003 1122 WrAL KSh/t 32.60 6.52 34.89 6.98 39.12 41.87 42.74 41.87 .......................................................................... Demurrage 24 8 32 32 32 37 38 Wharfage 1.5% CIF 6 7 4 10 13 14 14 15 Insurance, Interest 20 Export Licence 16 16 16 16 16 16 Handling 18 18 18 18 9 9 Lasses 19 19 19 19 23 23 Rail loading 20 5 20 5 25 25 29 29 Freight NBI 240 60 240 60 300 300 348 348 Rail Unloading -------------------------------------------------------------------------- 20 5 20 5 25 25 29 29 KSh/t 1114 272 1290 301 1386 1591 1532 1653 WTAL KSh/GJ 47.47 11.59 48.13 11.33 59.06 59.37 65.29 61.68 ______-___-_-------------------------------------------------------------- Sources: See text. Shadow prices as below: Financial Economic Exchange Rate Labour (urban) * Thermal coal is assumed available at $35/tonne F 0 B. port of origin and shipping charges are set at $20/tonne 3. REFINED OIL PRODUCTS Kenya can obtain refined oil products by direct import or through the Kenya Petroleum Refineries Ltd refinery at Mombasa. To safeguard the interest of the refinery it is the government policy to discourage imports of refined product; while surplus refined product, mainly fuel oil of a generally lower quality than used in Kenyan industry, is re-exported into spot markets. 3.1 Current Prices Prices of refined product in Kenya differ from the cost of imported refined product. The pricing of refined product recovers the total cost of the crude, refinery fees and transport and handling charges up country. Allocation of these costs between individual refined products is affected by demand and policy factors. Refinery charges in Kenya average $US 15.5 to 16.0 per tonne of crude. Using an actual refinery out turn, it is possible to derive the effective cost of crude and the total cost of refinery derived fuels. The actual price of crude CIF Mcanbasa was estimated to have fallen from $US 228/tonne in 1983 to $US 207/tonne in mid 1985. The 1985 dcanestic prices of refined product are shown in Table D.7 in a form canparable with the CIF prices of imported refined product. These estimates are from the World Bank and appear to reflect contract prices in the Middle East. The Table also shows the estimated price of exported fuel oil FOB Mombasa obtained in discussion with MOERD. In addition, this Table s h m prices charged in-bond for various refined products in Kenya. These provide the basis for financial prices of residual fuel oil in Kenya. There are t w possible bases which could be considered for the economic pricing of fuel oil in Kenya, namely either the import parity price (being based on the CIF Mambasa price of imported refined fuel oil) or the export price FOB Mcfnbasa. of exported surplus refined fuel oil, adjusted for transport costs The Consultants have considered both approaches and believe the FOB m s a detexmined price to be the appropriate basis since the question of refinery policy lies outside the scope of this stucty and because it reflects the actual economic circumstances facing Kenya. TABLE D.7 REFINED PRODUCT PRICES IN KENYA, 1985 In Bond Price Import Price Export Price CIF Mmbasa FOB Mombasa KSh per KSh per $US per $US per tonne '000 litres tonne tonne LPG Premium Regular Kerosene ?Go* Jet Kero Marine G.O.* ID0 FO 125CST FO l8OCST FO 280CST Converted to $US 1 = KSh 17 * Density assumed identical Source: Ministry of Energy and Regional Developnent (Annex11, and World Bank The adoption of l m prices for fuel oil in Kenya reflects the situation in which fuel oil, as a surplus product f m refining in Kenya, must be exported and receive low prices. The product which is re-exported is heavy grade fuel oil; the lighter grades fuel oil are sold in Kenya and in neighbouring countries, with soarre sales being made in spot mrkets by each marketer, and the product shipped f m Kenya. Same fuel oil is sold as feedstock for refineries in other countries which can crack it to lighter fractions. In 1985 the price received for exports of fuel oil froan Kenya approximate $US 135 per tonne (FOBdestination), f m which shipping costs of $US 10 per tonne needed to be deducted. Contract prices were at the time above spot prices and marketers with contract sales were able to sell fuel oil and dual pupose kerosene overseas at higher prices than in Kenya. By 1986 the FOB Mcanbasa price had fallen to USS41-50 per tonne. 3.2 Oil Transport There are differences in transport costs to different parts of Kenya. Table D.8 gives the transport differentials needed for calculating fuel oil prices outside Nairobi using the delivery charge basis discussed under Oil Products in Section 6.1. OIL TRANSPORT DIFFEREERlIALS (KSh/GJ) Location Financial Economic KSh/GJ ........................................................... Nairobi Mombasa Nakuru Eldoret Kisumu Magadi Thika Athi River Source: See text. 3.3 Price Forecasts for Refined Products Over the period to 1995 there are expectations of possible short term lower crude oil prices, perhaps allowing fuel oil to fall relative to coal, (althoughany such move is not expected to result in lower coal prices). The cdination of growing non OPEC oil supply with a short tern fall and then a slow growth in d d , yields a forecast of a drop in oil prices followed by recwery to about 1985 levels by 1995. As this report was being finalised, spot prices for crude oil had fallen well below $US 20 per barrel to around $US 12-15. The level to a c h prices fall and the duration of low prices are the subject of conjecture and will depend upon the willingness of high cost oil producers to continue to discover, develop, store and deliver oil at low prices; and on the d d for oil at low prices. Meanwhile consumers in Kenya will not obtain all of the benefit of falling oil prices because there will be a need to increase refining margins. Nevertheless, lawer crude oil prices will reduce the oil import burden, freeing foreign exchange for other needed imports and reducing the pre-tax cost of refined products. The future patterns of fuel oil prices in relation to crude price will depend upon the match between mrld refinery capacity to upgrade residual products, coal prices and demand. World denand for residual fuel oil, apart f m refinery feedstock, has been relatively constant as coal and natural gas substitutes and impruving energy efficiency all have had their effect. Progressively smaller shares of crude have been saleable as fuel oil and so more and m r e has been refined to lighter product L/. In general, fuel oil is expected to remain constant in relation to crude in order to m e t growing canpetition f m coal, the prices of which are expected to remain constant in real terms. Gasoline, kerosene and gas oil prices are forecast to rise in relation to crude oil by 1990 as refiners attempt to recover maryins lost while crude oil rose and demand slackened 2/. Table D.9 repruduces these forecasts. World Bank 1985 forecasts were for declining crude prices recovering to current levels by 1995. The Bank's revisions to its forecasts after the oil pricc fall in 1986 indicated 1995 figures sane 25% - 30% lower H m r there is a diversity of views about future oil prices. In this report, therefore, the forecasts have been left as cmputed in 1985 while, for the purpose of evaluation, the sensitivity analyses described in Chapter 8 and this Appendix have been designed to allow the introduction of other prices if preferred. In the longer tern, against which investment decisions will need to be based, the Consultants expect oil prices to recover substantially frwn the 1986 levels. Prices for liquefied petroleum gas are not quoted in reports to hand This product is capable of being imported in the quantities in which it is used in Kenya, but at a cost penalty which relates to its volatile nature. It is forecast to vary with crude but to fall less rapidly betwen 1985 and 1990. -/ 1 T.R. Eketon "Further Declines in World Demand" Petroleum Econanist, June 1984. - 2/ The Econdst, 197: 7415, (October 12, 1985) p.15. TABLE D.9 RECENT AND FORECAST CIF IMPORT PRICES M(%rlBASA FOR REFINED AND CRUDE OIL PRODUCTS ($US per tonne) ....................................................................... Year Crude Gasoline Kerosene Gas Oil Fuel Oil LPG ....................................................................... Forecast (1985 dollars) Source: See text * LPG calculated as 135 percent of kerosene price for 1985. Future prices of fuelwood in Kenya are an important input to choice of alternative fuels, especially indigenous fuels. The principal issue is future wood price relative to future fuel oil price. The key point about future fuelwood prices is the ccanplete uncertainty as to what they will be. Fuel- supplies cgne fran many sources and the mrket is at present well supplied. Current financial prices are low (KSh 250 per tonne maximum) ccanpared with the proper ecunmic cost and fuel- users, with few exceptions, report little difficulty obtaining supply. Major sources include land clearance operations, forestry operations, carmrercial modlots (which often are caannitted to particular use), and individual farms . Land clearance operations supply wad for charcoal mking and are a major source. There is also emerging coamrarcial and camunity activity in forest plantations. Haever, wood fran plantations will be inherently more costly than wood fran land clearance which is likely to continue for scane time driven by rapid population growth. Plantations have to recover capitalised costs of planting, tending and holding land as well as costs of felling. Foresters and envimnmntalistsworry that when land clearance supplies of wood fall away, there will be insufficient forest yield to cope with demnd. This tends to ignore the effect on forest investment of an expectation of rising wood prices. The projections of future shortages are widely understood and at least sam investment in new forest resource is now taking place in expectation of future higher prices. The upper limit for future prices of fuel- muld occur if all supply were to c- froan fuelwood plantation cropping. This point seems cost unlikely to be reached, huwever, because land clearance and agro forestry activities can be expected to continue,and these sources should always be lower cost, at site, than mnocultural plantation forestry. This is because clearance and agro-forestry recover sane costs froan sales of cleared land and produce as well as mod. 4.1 Previous Studies The min infoxmation on this subject drives from the ongoing Peri-Urban F'uelwwd Study and a series of studies by the Beijer Institute L/. Most results are reported in Sections 2.3 and 2.4 of Appendix E with others in Appendix C of the report on the companion study Energy Efficiency in the Tea Industry Early results of the Peri-Urban Fuelwood Study were available for review by the Consultants. Previous studies, and particularly the Peri-Urban Fuelmod Study, place emphasis on the foreign exchange cost of alternative fuels, especially kerosene, and use discount rates of 10 percent which is low. They also shadow price labour in rural areas where labour is scarce. The Consultants have modified the results of the Peri-Urban Study to allow for 15 percent discount rate and to reduce the shadow pricing of scarce rural labxr. Different land costs have also been considered. The view formed is that the econanic price of fuelwood could be as high as KSh 650/tonne or about 45 KSh/GJ. 4.1.2 F'ueld Transport Fuelwocd which is low cost on the stump can rapidly became high cost once removed far froan forested zones. ....................................................................... - 1/ See especially Keith Openshaw, ,Costs and Benefits of Proposed Tree Planting Programne for Satisfying Kenya Wood Energy Requirements, Part 11, Appendices, Nairobi: The Beijer Institute, (1982); F'ueld Policy Pmgranme Cost Estates, A Supplement to Energy Developnent in Kenya: Problems and Opportunities, Nairobi: The Beijer Institute, February 1982, and Richard Hosier, "SocialCost Benefit Analysis of Fuelwood Developnent Projects" Nairobi' The Beijer Institute,April 1982. Table D.10 provides the basis for costing the transport of fuel- taking into account fuelwood bulk density and freight rates noted in Kenya in 1985. It should, however, be noted that, f m observations of fuel- supply, it is anticipated that an appropriate selection of wed species will be able to be gmun near the industrial sites or tea factories. 4.1.3 Re-estimates for Fuelwood The calculation procedure used by the Peri-Urban Fuelwwd Study estimates the year by year costs of the plantation and the cubic metre yields obtained. Both yields and costs are discounted to the start year. The cost unit is obtained by dividing the discounted cost by the discounted yield. Delivery costs would add between KSh 75 and KSh 100 per cubic metre to these costs. In general the future financial cost of wood, excluding land costs, could be in the range of KSh 150 to KSh 350 per cubic metre which is equivalent to around 20 - 47 KSh/GJ. If land costs are added, costs could rise by a further KSh 300 to reach KSh 450 600 per cubic metre, or 60 - 87 KSh/GJ. The actual figures will be strongly influenced by any supply/dmand imbalance; and the Consultants found no sound reasons why fuel- could not reach or exceed fuel oil prices if the shortages being claimed for this market were to mterialise. 4.2 Econdc and Financial Prices of Fuelwwd The future path of fuelmod prices is subject to great uncertainty. Field investigations by the Consultants indicated that industrial users were having no difficulty obtaining fuelwwd at very law prices. However, some evidence points to widespread clearance of forest and a future shortage at current market prices. Moreover, increase daMnd due to rapid urban growth plus fuel.wood substitution for oil seems likely to encourage a camnercialised fuelwwd m k e t with m c h higher prices which muld stimulate private supply f m yield as -11 as stock. It muld also extend the area of accessible fuelwood stock as higher prices recover the cost of longer transport distances and of access to more remote forested areas. Based on the econanic costs of producing short rotation fuelwood with land preparation by share croppers, plantation cost could be KSh 400 per tonne of fuelwood and land costs could add KSh 100 per tonne. Fuelwwd so produced muld have an ass- Gross Specific energy of 14.2 GJ per tonne and need to be transported to plant. With up to KSh 300/tonne allowed for transport over same distance the economic price could total near KSh 800/per tonne of 56 KSh/GJ. Various studies have indicated economic prices of fuelwwd ranging as high as KSh 600/tonne delivered while low esthtes fnm the mi-Urban Fuelwood Study are as little as 4 KSh/GJ on the stump. The range of prices selected for analysis in this study provides for the transport and handling of the fuelwood and has been largely based on the prices reported by the Peri-Urban Fuelwood Study. At present a m r c i a l mrket for fuel- is not fully established as it is for charcoal and canpmhensive data on fuelwood prices are not available. Nevertheless the Consultants believe it likely that financial prices will rise t d s econdc cost as the fuelwood beccanes camnercialised, possibly over a ten year horison. The range of financial prices of fuelwood includes at the low end, the average of observed costs obtained at audited plants and tea factories. This average of 18.31 KSh/GJ was seen to be affected by the sales of wood fran Government forests. However, studies by the Peri-Urban F'uelwood Group have shown that these sales do not recover the costs of growing the timber so that future supplies fran this source can be considered unreliable. The results in Table D.ll suggest that looking ahead a similar period the financial price of fuelwood, delivered, could reach 64 KSh/GJ canpared with themal coal 69 KSh/GJ and fuel oil of around 60 KSh/GJ also, depending on grade. This outccatle is dependent upon several issues, including: the rate of v e n c e of plantation supplies and from agm-forestry canpared with loclller cost supplies fran land clearance; the extent of coannercialisation of the fuel- market. At present, fuel- supply is not significantly comercialised and prices are low; the rate of emeryence of industrial demand for fuel- in the camnercial mrket. Where supplies can be obtained off mrket, particularly where land can be obtained at no cost, then supplies will be much cheaper. 5 . ELEXYRICITY TARIFFS The electricity supply authority in Kenya is Kenya Power and Lighting Company Limited (KPLC) under the Ministry of Ehergy and Regional Developnent . According to discussion w i t h K P X , tariffs are based on long run marginal costs calculated frun 5 year capital investment plans and demand projectians . However recent tariff increases have had more to do with changes in the exchange rate relative to the currencies in w h i c hdebts are denaninated and the effect of this on capital and interest repayments. Demand for electricity has not been sensitive to tariff but has been sensitive to industrial output. KPLC has forecast growth of total electrical energy damnd to be 2 percent faster than the growth of Gross Damestic Product (GDP). Industrial electrical energy demnd is forecast to g m at the same rate as GDP. These forecasts suggest a 6 percent per annum gruwth in electrical energy demnd. The Power System Efficiency Report dated March 1984, prepared jointly by UNDP and World Bank under ESMAP, discusses long run marginal costs The existing system as daninated by hydroelectric and geothermal capacity. Power is also imported froan Uganda. System capacity is therefore partly dependent on rainfall and the limited thermal capacity is intended to be used only when rainfall is inadequate. Existing thermal capacity is dilapidated and unreliable. In May 1985 following a visit to Kipevu m r Station it was estimated that of a name plate capacity of 114 MW the current capability was 13 MW. Routine maintenance was expected to return 39MW to service and a further 45 MW muld require at least 4 mths and long tern maintenance or replacement. The Kipevu Pmer Station is unsuited by site and location to any coal conversion. A study to produce a Master Power Developnent Plan was comnenced in February 1986, after the fieldmrk for this study ended. KPIC gives the following typical electricity generating capacity costs in dollars per kilowatt installed for mid 1983- Generator Type $US per kw Capacity Hydro - Kiambere Turkwell Gorge Other Geothermal Coal Thermal Oil Thermal Diesel Gas Turbine Generation costs for the various types of generators are given in Appendix G as follows: TYPe cents/kwfi $US/GJ Hydro electric 4 Geotechnical 4 Thermal (FuelOil) 3 Thermal (Coal) 5 In the period to 1987-90,when Kiambere hydro electric power station is expected to ccane on stream, KPLC expects to have an electrical energy shortage which will partly be made up froan thermal sources. The financial burden should also remain constant once Kiambere is in place and this should result in tariffs remaining constant in real terms for a few years after 1988. The financial and econcmic prices used in this report are derived respectively f m observed costs of audited plants and froan other hydroelectric project evaluations. These figures are presented in Table D.ll. The econ~nicprices of electricity reflect the use of thermal plant to supply during periods of hydro energy shortage. The costs of new capacity will rise and have a large cmpnent of foreign exchange. 6. S-Y OF ECONOMIC AND FINANCIAt PRICES OF FUELS 6.1. General Table D.ll gives the financial and economic prices assumed for relevant fuels. For ease of canparison results are given in KSh per GJ together with the assumed gross specific energy values. The concept of econcanic price is intended to reflect the opportunity cost to Kenya of using a particular material as fuel. The value is esthted as a border cost for internationally traded carnnodities,that is the cost to import the carmrxlity. For untraded and damestically produced goods, the econcanic price is an esthte of the resources required to replace the good in its next mst valued use or the value foregone in not having the resource available for that use. Values of untraded fuels are given at source. Allowance for delivery into plant will vary, being particularly significant for fuelwood. A summry of the econdc and price assumptions adopted for each relevant fuel is given in Table D.10 which describes the main features of cost, excluding foreign exchange and labour, which are used in esthting the values in Table D.11. - Coal: All coal prices and delivery costs are derived £ram table D.6. Oil Products: The economic price of oil products in Kenya is the F.O.B. price of consenred oil re-exported, plus the current average cost of transport to the KTDA factories. Loading charges at Mornbasa are ignored because that are small. Following recent fluctuations the F.O.B. price of fuel oil exported from M s a averaged 25 KSh/GJ between July 1985 and June 1986. Financial prices are based on the average of prices observed at plants (71.5 KSh/GJ) less delivery costs estimated as follows. Financial delivery charges are derived from -ex 1 taking account of pipeline and town delivery fees and adjusting for the exchange rate (17KSh per dollar), density (intonnes per 1000 litres) and Gross Specific Energy. The Economic delivery charge is derived £ram the market rate by adjusting for the foreign exchange cmpnent as given in Table D.lO. For all transport the import compnent is assumed to be is 80 percent of cost and the import canpnent is given a premium of 20 percent. This means that for all oil products the econcanic delivery charge is 16 percent above the financial delivery charge. These armunts are e l l and special conversion of pipeline costs was not considered warranted. Fuelwood: Table D.10 and Section 4 give the basis of the econcanic fuel& prices. Financial and econcanic delivery costs are also described with delivery being discussed further in Section 6.2.2. As noted earlier, it is expected that fuelwood, in line with practice observed during the audits, will be obtained fram nearby sources, often grown for the purpose. Under these circumstances the range of econdc and financial prices for fuelwood,given in table D.lO, is sufficiently wide for locational differences to be covered. In Nairobi the lmv range of fuelwood prices muld not be expected to obtain because of high land costs. Elsewhere the ranges provide sufficient guidance for prefeasibility evaluations. Sawdust and gvpruduct: The econamic and financial prices of sawdust and byproduct are taken to be identical with the derivation given in Table D.10 based on discussion in Appendix E. Electricity: The econamic price of electricity is based on results derived for the recent relwant hydroelectric studies The financial price of electricity is taken £ran the energy charges in the published 1985 tariff for 11000 volt supplies. Tariff details are provided in each audit report. Also used in setting this range are average electricity costs observed at audited plants. A high estirnate is derived fran the case in which energy d d managemnt leads to rising tariffs during the planning period. The Consultants understand that energy supply difficultiescould be experienced in the next few years until new capacity cams on stream. Since the need is only to establish range values for the sensitivity tests no distinction is made between demand and energy charges in these electricity tariff forecasts. The evaluation of each proposed investment has been based on the denand and energy tariffs applying at that plant in 1984 or 1985 with the evaluations taking account of load factor. SUMMARY OF EcoNauC PRICING ASSUMPTIONS ...................................................................... Foreign Foreign Basis of Exchange Basis of Exchange Delivery in Econanic Econanic Cost Delivery Fuel Price Cost into Plant Cost Fuel O i l Export earnings 100% Rail Delivery 80% on displaced consumption Electricity Turlcwell Gorge 80% n.a. Study Fuelwood KSh 4OO/tonne 15% 100 lan round 70% plantation cost trip 2.5 tonne plus KSh loo/ in 7 tome truck. tonne land cost KSh 1.4 per and Peri-Urban tonne km Fuelwood Study Sawdust Price to sawmill 0 100 )an round 70% KSh 100/tcmne trip 7 tonne truck delivers 4 cu metres in container. KSh 1 4 per tonne km Density 738 kg/m3 Coal C.I.F. price Mmbsa Rail Delivery 80% Grain As sawdust As for fuel- 70% Byproduct KSh lOO/tonne _______----------------------------------------------------------.------ Source: Seetext. Market Economic Exchange rate adjustment 17 20.4 KSh/$ Labour 100 50 Pe-t s a ~ ~ x d 30 a6mx aqe~xdaddem? xoj paqaldwn aq p ~ n m6yqsaq L q ~ ~ ~ q ~7-s u a s pa6Ps~~uaaxo3araW sm q~ squauqsa~q pa- Lm? 30 ssaupunos a amsua oq xapo UI *sq3a[a;~d6q6pn[ q asn oq s~seqquaTq3jns R qou a ~ sa3mos maua m q o pue -Tan3 a ~ ' l ~ o30 s w ~ x d986'1 a- 7- pas- xaqqznj sm q1 * s w ~ x Tans q sa6u~q3oq aATqTsuas aq d pTnm aauaq pue ' s q u a q s a ~ qpasodard xTaq unxrj 6qqlnsa.x s 6 u ~ ~ Tan3 e s 30 lam1 aW uodn puadap plnm sxmfard paqwyxou LUQxoj u o ~ q e a ~ j ~ q s n f a- q q a u u e x h ~ d a- 30 qasqno a- qe paqed~qque SQM 31 *saa-sTp quaxa3jTp xoj paqez-md aq m sTaw pap~xqun zoj sak5z-q~A x a ~ ~ ~ a *quauqsn[pe qsm a6u~q3xa u 6 ~ a r o j a s a p n p q alqm sm *sa6xq3 pue saTqnp sapnpq aa~xd TPT3ueU?d 'qsm L q v ~ o d d oxo a ~ p mpoq ST qso3 3ymw33 d :aqw u o ~ s s m s ~s ps m ~ pue waq aas q :a;3mos .......................................................................... The Consultants forecast fuel prices to rise. Indeed, as noted in this Appendix the main forecasts ccxnpleted in 1985 anticipated fuel oil returning to the same 1985 prices in real tern by 1995. With the collapse of oil prices in 1986 others (including the World Bank) developed mre conservative expectations of oil's future, suggesting 1995 prices down same 25% - 30% on those figures although stillwell up on 1986 levels. It is an area of considerable uncertainty with it becoming apparent that the investment decisions, when they are rrrade, could well face a very different range of energy prices. For these reasons a sensitivity analysis approach was designed and used to clarify the viability of the various projects and at the same time pmvide a basis for assessing their priorities in the light of preferred price scenarios for the various fuels. 7.2 Evaluation Procedure All projects identified as prospective opportunities during the Consultants' fieldmrk were subjected to a full financial analysis as a basis for checking their relative importance. This involved scheduling all relevant costs and benefits over the ten year time horizon and then the computating the internal rates of return, net present values at a 15% discount rate and simple pay-back periods for all projects. Froan this waluation, those which continued to look prcsnising were then grouped into categories of like characteristics. The projects froan each of these categories were then evaluated using three different econdc frammorks. These covered respectively an equity financed cash flow, a debt financed cash flow and a foreign debt financed econamic waluation. &pity financed cash flow debited the full capital cost of the project at the time of investment. The estimated fuel cost savings were credited after tax, together with depreciation tax benefits. Debt financed cash flow debited half the capital cost of the project, at the time of investment, with eight equal annual debt repayments for half the capital cost starting froan the second year after investment. The estimated fuel cost savings were credited after interest and tax, together with depreciation tax benefits. Foreign debt financed econdc evaluation debited the project with its full capital cost at the time of investment and applied a foreign exchange loading to the imported ccanponents. A credit and foreign exchange premium was provided for foreign debt drawdown of half the capital cost and debt repayment and interest were debited with a foreign exchange premium as they were paid. Interest was assumed to begin imnediately and debt repayment after a two year lag. Total savings of fuel were credited to the project. In addition, for each project using sets of fuel prices the breakeven price relationship was established for the principal fuel(s) involved i.e. establishing the relative prices of which a zero value for Nmr for that project would be obtained. Where two fuels were involved (suchas fuelwood substituting for fuel oil) the breakeven relationship between the t m was presented in the fom: -even price of Fuel 1 = Constant + Coefficient x Expected price of Fuel 2 (wherethe breakwen price results in NPV=O) This relationship provided a continuous basis for sensitivity testing, known as 'crossoveranalysis', enabling the viability of the project to be assessed at any nominated forecast prices for the fuels concerned. The Consultants subjected these results for all of the projects to sensitivity analyses - again ccanputing the breakeven relationships, confinning that these closely matched the economic case, and then assessing each project's viability in the light of forecast prices for the fuels. Frcan this process it was possible to identify which projects strongly justified investmnt, which would be unrealistic to implement, and which were marginal cases that could require more detailed analysis to detennine their suitability . In all instances the analysis was checked to see that the economic and financial results and breakeven relationships were closely related. Following that comelation, the project priorities were assessed on the basis of the financial results, these being a principal factor likely to influence the implementation of these projects. 7.3 Priorities for Projects The actual ranking of projects by Kenya will ultimately depend upon a number of factors including the da~ndsfor capital as well as the availability of skills and the impact of other government or lending authority priorities. However, experience has shown that the NPV (net present value) represents an effective criteria for early project evaluation. For each project a positive NW reflects an annual return of at leqst 15% real in this analysis, prwides a guide to the effectiveuse of resources, and attributes to it an absolute priority relativity (evenwhen grouped in different ways to pennit comparisons with other similar projects). In this analysis, therefore, projects were therefore listed in NPV order within categories,with the most attractive projects having the greatest NPV beginning each list. Each of the project categories was then considered in turn, with the analysis providing one or more price figures at which the NPV would be reduced to zero (those figures being designated as the "breakeven" prices for the project, i.e. the price(s) at which the project would yield a 15% p.a. real return). The projects in each list were r e v i d against the current and expected prices for fuel oil, fuel& and/or electricity as appropriate. The ranges of prices used are set out in Table D.12. In general it w s seen that those projects with l m r NPV's =re the mst susceptible to falling oil prices or rising fuel- prices, while the low electricity tariffs current in Kenya had more wide ranging impact. It will be appreciated that this analysis was designed to check the apparent viability of the projects. Other factors will cane into play determining whether or not to proceed, including the results of m r e detailed feasibility studies, and the practicability of implanenting a number of projects at the one time. TABLE D.12: KENYA FUEL PRICE RANGES ....................................................................... Fuel ....................................................................... Financial Price EconCmic Price KSh/GJ KSh/GJ ....................................................................... Fuel Oil**** - high 75 81 - base 72 51 - low 55 45 Fuelwood - high - base - low Electricity - high 206 - base 164 - 10w 147 coal - high 60 - base # 50 -law # 40 -- Source notes: All data derived in this Appendix w i t h the exceptions noted. * ESMAP, mi-Urban Charcoal/Fuelwwd Study, Phase 1 Report, Wrking Paper VII. ** ESMAP, op cit., medium value doubled to allaw for transport and same m t added to l