Fisheries studies on and : first quarter final report June/July 2001

Item Type monograph

Authors Kamanyi, J.R.; Mbabazi, D.; Muhumuza, E.

Publisher Fisheries Resources Research Institute

Download date 26/09/2021 20:52:09

Link to Item http://hdl.handle.net/1834/34141 Fi heries Studies on Lake George and L ,~r\~~.,. ,Kazi~ga Channel >rP'cS>·'· fIrst/Quarter Final Report June/July 2001 L 30° 10' L '-Y'--~ -... i UCiAH • '''7 // / L r' , ( L J L L , -I

I l. cio . O' -

- '. ~ - .";..-~ :­ .....'!

o 5km ~ - b!=~,==,"l ;·..i ."f.." Swamp , \ • Fllhlnlil vi 1I0lill8 I ~ --eJ--­ 30~fO· -

Preparell by J.R Kamany" 0 Mbaba2, & E Muhumuza M.". lntegraled Lake ManagemenlProieCt (ILMj Fisllfmes Resources Research lns/llule (FlRR1) - P.O. Bol 34J. J1fI13. Fov 25&-4.)..120192, Financed by" Department lor IntematiOl1al Development (DF1D) National Agncultural Research Orgar,ll.aoon Email' ~n@!n'occrn COuo

- rJT~,,gB'86'J'dfu~~ tl~UI( The VJ8,..S ellprei.5e-d Deparlme'l' for Inlernllhonal Development (OFIO) for bellefi'l of develop,n9 count'les Gl;O ,." " , -~~"-ar~ no! "ec:~ss"nly of ""'0­ 6 "?>

._------'­ ---~~---~--~------_.-.

Fisbmts SlrltlJts .D r.u. CHrg• .. lYzUJ,. CbDlUld

TABLE OF CONTENTS

I Section Item Paqe

I Table of Contents i Scientific, English & local names of the fishes and fishing ii methods Abbreviations iii Executive Summarv iv 1 Background 1 Mao showing Lake George and Kazinga Channel 3 f-=---­ 2 Methodology 4 -­ 2.1 Frame Survey 4 2.2 Catch Assessment Survey 5 Section A Results 7 3.1 Global (Lake George and Kazinga Channel) picture of the 7 fishery 3.2 Fishina effort (boats) in relation to the fishing practices 10 3.3 Averaqe number of qears per boat 10 3.4 Overall fish species composition in commercial catches 14 3.5 Global size distribution of the major commercial fish 14 species 3.5.1 Length frequency distribution 14 3.5.2 Average size (kg) of major commercial fish species 26 3.5.3 Catch rates 26 3.5.4 Annual catch estimates disaqgregated by fishing practices 26 Section B Picture of the fisherv broken down by landing sites 30 3.6.1 Number of boats (FS) 30 1­ 3.6.2 Boats organized by fishinq practice 30 3.6.3 Averaqe number of qears per boat 32 3.6.4 Distribution of boats using different gill net sizes and hook 32 sizes 3.6.5 Fish species composition in gill net and hook fishery 36 3.6.6 Fish species composition in all the 8 fish fishing gractices 36 3.6.7 Size of the maior commercial fish species 42 3.6.8 Catch rates 42 3.6.9 Annual catch estimates 42 4 Discussion 49 52 Recommendations . 5 References 54 Annexes (A,B,C) ~

'"" FJslmilS SIIIJII.s ~1I Uk. GHlp • IlubJg. CbillUJ.1

Scientific, English & local names of the fishes, fishing practices, types of boat licences

Scientific name Enqlish name/meaninq Vernacular name/meanin~ I. Oreochromis ni/oticus til apia Nqege -J Oreochromis leucostictus Tilapia Bambala Tilapia zi/lii Tilapia Kajjansi Bagrus docmak Cat fish Semutundu • ~rias qariepinus Cat fish/mud fish Male Protopterus aefhiopicus Lunq fish Mamba -- • Barbus altianaJis Barbels Kisinja Mormvrus kannume Elephant snaut Kasulu • Beatino (active fishing) Tvcoon PiercinQ Punda punda Nets doubled in length by Mukira • ioininq two of them Hook and line - tying Kikwakurizo - hooks on papyrus • 1S1 and 2"0 world war Kawonawo boats veteran boats Veteran boats Recent • demobilized/retrenched soldiers Experimental boats Boats commissioned by • former (1996/2001) State Minister for Fisheries to carry out fishing on Lake • George, Edward and • Kazinga Channel • •I • I I 11 -.I

-~-----~------.- -­ - -~ ------~- ~~---~--

FlsJJuf•• Stadia DII IHU. CoD". If JlazbJg. l:JJ...d

ABBREVIATIONS

< Less than B Bushenyi CARE Cooperative Assistance and Relief Everywhere CAS Catch Assessment Survey DFID Department for International Development FIRRI Fisheries Resources Research Institute FL Fork Length FS Frame Survey g Gram ILM Integrated Lake Management K Kg Kilogram MSY Maximum Sustainable Yield Mt Metric tones NARO National Agricultural Research Organisation TL Total length UFRD Fisheries Resources Department UK United Kingdom LG Lake George KC Kazinga Channel

111 I~ r-. EXECUTIVE SUMMARY r-. Background ,­ FIRRI surveyed the fisheries of Lake George and Kazinga Channel between 20'h June and 20'h July 2001. This was the second survey FIRRI has conducted for the ILM :-. project on the water system. The first survey was conducted during November 2000. These data, the analyses and accompanying reports contribute to baseline information for the fishery being collected with the support of ILM that is required for lakewide • planning and management. • Eight fish landing sites (6 on Lake George) namely; Kahendero, Hamukungu, Kasenyi, Kashaka, Mahyoro, Kayinja (2 on Kazinga Channel) namely; Katunguru - K and B fall • within the focus of ILM and were surveyed during November 2000 and June/July 2001 -. over a three day period at each landing site in 2001 (Mahyoro 2 days). In November 2000, each landing was sampled once. FIRRI conducted a rapid FS and concurrently a -. CAS. All results are reported by landing site and then summed up (Global) for 8 sites -. on Lake George and Kazinga Channel. --I Fishing effort

Collecting reliable data however, remains a challenge. In FS, the number of boats. -. gears per boat as observed by the survey team were recorded. Additional effort -. estimates were obtained from community members. A total of 426 boats were recorded during the present survey. This was 67 boats less than the estimate of 493 based on -. community reports. When the FS of June/July 2001 is compared with the FS of November 2000 (FS recorded 388) shows an increase of 48 boats while the FS -. estimates of 2000 are lower by 121 boats (November 2000 community estimated 499 boats). This shows clearly that the establishment of a transparent system of recording '-I the true picture of effort on the water system for all stakeholders is an outstanding task. '-I -. IV -. ------._------

FJsbodlS Sf.lIl.. DB r.uo GHrgo If IfUilJ,. '1I...0J

The second element to establishing effort was the number of gears used per boat. Most boats (except the newly introduced practices of joining 2 nets - Mukira and hook and line) exceed the legal limit for the water system fishery and the pattern varies from landing site to landing site, season to season, fishing practices and differs between survey results of November 2000.

The legal limit is 10 gill nets fished passively per boat, or 100 hooks per boat. However, up to an average of 53 nets (2001) and 64 nets fished passively (2000) and average of about 600 hooks per boat in longline were in use.

ILM has emphasized the need to consider fishing practices (the gear plus its use) rather than simply looking at the gear type. During the June/July 2001, nine fishing practices un evenly distributed on different landing sites were identified. Gill nets used passively (majority 4.5" mesh), passive mukira in which gill nets are doubled in depth (majority used 4" - 4.5" mesh size), active mukira tycoon (majority 4" mesh), boat seining (majority 3.5" mesh size), mukira boat seine (majority 4.5" mesh), longline (majority size 9), hook and line (size 7) and traps were being used. The practice of doubling the depth of gill nets was noted in both surveys but had not been recorded in any previous surveys. The doubling of gill nets was recognized as a distinct fishing practice in both surveys and will be in all ILM fisheries work on the water system. A switch towards more boats fishing actively (tycoon) was noted in 2001 survey with potential implications for the whole fishery. It is not yet clear if this is a long-term trend or a short-term response by fishers to seasonal activities or fish catches.

Fish catch characteristics

Of the estimated total catch (June/July 2001) the relative fish species composition was as follows: O. niloticus (39.3%), P. aethiopicus (30.4%), C. gariepinus (15.2%), B. docmak (13.8%), O. /eucostictus (1.2%), B. a/tiana/is and T. zillii contributed 0.1 %. This was a significant change since November 2000 when the catch was dominated by P. aethiopicus (47.4%), B. docmak (21.7%), O. ni/oticus (13.6%), O. /eucostictus (2.7%),

v I­ Fish.ri•• Stud/•• '11 Ub fiHrg. " 1IuiII,. ClJUIII1 I. B. altianalis and T. zillii (0.2%). The rise in contribution of O. niloticus reflects the greater effort of tycoon fishery. Subsequent surveys will establish whether this is a '. permanent trend. l Comparison of catch rates (kg/boat/day) for all landing sites with the November 2000 survey gives the following picture on the water system. The following catch rates have l increased since the November 2000 survey: boat seining (10.3 - 17.4), active gill net mukira (18.4 - 71.3), boat seine mukira (5.2 - 48.8) and longline (17.9 -19.8) while -. lowered catch rates were recorded for gill net passive (37.9 - 22.9), passive gill net mukira (35.4 - 30.6), active gill net tycoon (21.6 - 17.4), hook and line (36.3 - 9.7) and -.• basket traps (13.2 - 4.2). Annual catch estimates were made for each landing site and global by fishing practice. -. An over all estimate for lake George and Kazinga Channel over the year was obtained. The November 2000 annual estimate was 2, 512.5 metric tones while the June/July 2001 was 2, 072.8 mt reflecting a 17.5% drop. Given that this is only the second I estimate in what will be a series of annual catch estimates; it is too early to talk of any trend of declining catches. When viewed against historical catch estimates, this figure • is not considered to be a significant change in the fishery. • • I I I I I vi I -, • ~ FWlod.. SIrIdJ.. DJJ WO C#Drvo 6 1UziJJ,. CbOIJIJoJ • 1. BACKGROUND 2 I Lake George (250 km ) is connected to the 33 km long Kazinga Channel (less than one km wide) which flows into (Fig.1). The fisheries are under the control of I Uganda Fisheries Resources Department (UFRD) and the local authorities of the districts surrounding the lake and the channel namely; Kasese, Bushenyi and I (formerly part of Kabalore district). There are 6 gazetted fish landing sites on Lake George. These are Kahendero, Hamukungu, Kasenyi (), I Mahyoro and Kayinja (Kamwenge district). The two fish landing sites (Katunguru-B and I K) on Kazinga Channel belong to Bushenyi and Kasese districts respectively. Lake George and Kazinga channel are the major sources of fish for the heavily I populated districts surrounding them. However, the fisheries have been under threat due to mainly increasing fishing effort and use of destructive fishing gears and methods I (Gwahaba, 1973, Crespi, et ai, 1995, Dunn, 1989). UFRD records of 1950-1988 showed that the average total fish catch from Lake George was 3,141 ± 159 mt and I varied between 1487 and 5097 mt. The estimated annual catches in 1997 based on 547 boats on Lake George was 6800 mt (Ogutu-Ohwayo et ai, 1997) while in 2000 survey (Kamanyi, et ai, 2000) was 2512 mt based on 426 boats on both Lake George and Kazinga channel. Dunn (1989) estimated the theoretical MSY figure of 3000 on Lake George. During FIRRI survey on Lake George (Ogutu-Ohwayo, 1997). 68.5% of the boats sampled used gill net mesh size of nets below 4.5". During 2000 survey, 2.8% passive and 53.2% active tycoon of the boats on Lake George and Kazinga channel used mesh sizes below 4.5" (Kamanyi, et ai, 2000). The presence of ungazetted landings and possibly increased fishing effort and use of destructive fishing gears and methods have been going on unchecked for quite some time. As one of the measures being under taken, Integrated Lake Management (ILM) project on Lake George and Kazinga channel is aiming at promoting community-based co-management of the fishery resource by establishing a more integrated and participatory approach to the management that cuts across the existing administrative boundaries and different natural resources sectors of Government. If the approach is successful, it would

16 120,,'­

ALb.?/i .~ -] C61-6-) KAYn ~cc 3\31 " • improve and sustain the exploited resource thus improving the livelihood of poor people • in the lake side communities and would demonstrate the feasibility of the approach on a medium - size lake and the channel under threat, with the intention of transferring a • methodology to much larger and complex lakes in Uganda which are facing similar problems. In order to commence rational participatory planning, accurate baseline • information on the state of the fishery resources is essential. FIRRI is involved in generating baseline information on the current status of the commercial fish stocks on • Lake George and Kazinga channel. The last study was under taken on behalf of ILM, by FIRRI in November 2000, generated information on catch effort, preliminary • assessments of the state of the fisheries and made management recommendations. The purpose of the present study which is the first quarter of the four quarters work is to • continue catch assessment surveys on a quarterly basis to provide more accurate baseline information for planning and management. The terms of reference are in • annex A. • • • • • • • • • • 2 • •.. I I I I I

860,,!II. riU!'·._I.::I •

dWD..S :.....~.. :~ ..; i o I I I I

.0 .... op I D; ••••• ...."""'...... I I I

1-'---'-­ .... : .I '. :: Ii

j, ..."". • i III! rfsbuf•• SIDtII...s r.u. CHIJ... hziDg. C/wmtl • 2. METHODOLOGY The rapid Frame Survey (FS) and Catch Assessment Survey (CAS) on six fish landing • sites on Lake George (Hamukungu, Kahendero and Kasenyi (Kasese district) Kashaka (Bushenyi district), Mahyoro, Kayinja (Kamwenge district - formerly in Kabarole district). • two landing sites on Kazinga channel (Katunguru -K, Kasese district and Katunguru -8, Bushenyi district (Fig. 1), were surveyed between 20 th June and 20 lh July 2001. Three • days were spent at each fish landing site except Mahyoro (2 days) for FS and CAS, both done concurrently. The Kasese district fish landing sites and Katunguru landing • sites were approached by road using Kasese town as a base while Kashaka and Kamwenge landing sites were approached by water using a motorized boat with • as the base. The fish landing times at different landing sites differed. These various • times are indicated in Table 1 for guidance during subsequent quarterly sampling. Table 1. Commercial fish landing times on Lake George and Kazinga Channel fish • landing sites (FS & CAS June/July 2001)

- • I Water body Landing site Fish landing time Lake Georqe Kahendero 8.00 a.m - 4.00 p.m j Hamukunqu -- 8.15 - 5.00...P.:.'!!. • Kasenyi 7.30 - 3.00 p.m Kashaka 8.00 - 2.00 p.m Mahvoro 7.00 - 2.00 p.m • Kayinja 7.30 - 1.00 p.m Kazinqa Channel Katunguru - 8 7.00 - 10.00 p.m Katunquru - K 7.30 - 5.00 p.m • ~ • 2.1. Frame Survey On arrival at every landing site, the FS was quickly carried out on operating boats that • never went fishing the previous night. During FS assessment, the type of boat, boat • propulsion, number of crew per boat, gear type, size, numbers (a single mukira net considered as two single nets) and fishing methods in use were recorded (FS data • sheet Appendix 8). During this exercise, assistance was obtained from whoever was • 4 • • III Tlm.d•• SIatU•• D" W. &Hr,. "guill,_ Cb.....t/ present among the chairpersons of fish landing sites or Fisheries extension staff or Lake George/Kazinga Channel Integrated Lake Management personnel (ILM) or from owner of the boat or fisher using the boat.

Information on the number of operating boats at each of the landing sites was obtained from either the chairperson or Fisheries Extension worker at the landing site. Different categories of boats such as licensed, experimental, transfer, veteran, illegal, Kawonawo were provided by the same personnel. Similar FS information was gathered from the CAS boats including the information for the catches that are sent through other boats. This information was transferred to FS data sheets to complete the information for the boats operating at each of the fish landing sites.

2.2. Catch Assessment Survey

The catch assessment survey was conducted in two groups and assisted by hired personnel mainly fisheries, extension staff or ILM personnel or experienced fishers to quicken the exercise of weighing and measuring. The exercise was carried out on almost all the boats that landed each day. On a few boats CAS could not be done due to many boats landing at the same time but FS would be done. Occasionally, where a few unruly fishers could not agree to wait claiming their catch would get spoilt or they would loose the market for their fish, only a FS was conducted. This problem was encountered mainly at Hamukungu and to a lesser extent at Kahendero fish landing sites usually with the active gill net fishers using small mesh size nets targeting Oreochromis leucostictus (Bambala) and O. niloticus (Ngege). On each boat, the crew (usually 2 fishers) were asked the average days the boat goes out fishing in a week including the number of times the boat is used in case it changes hands during the week, to establish the exact number of days the boat goes out fishing per week. The other information was on type of boat, propulsion and number of crew. Gear type, size, number, and mode of operation were recorded. The catch was sorted into species, weighed and total length (TL) or fork length (FL) taken depending on the species. This

5 '­ procedure was done for different species from different fishing gears, gear sizes and • fishing practices for each fish landing site (CAS data sheet Appendix C). • The estimated fishing effort, different gears, and fishing practices, catch rates, species composition and relative abundance for different gears and fishing practices were • tabulated for individual fish landing sites and for Lake George and Kazinga Channel as • a unit. The global size distribution plots for various fish species from different gears and fishing • practices were done. The annual estimated catch was presented in graphic form for • different landing sites and global for Lake George and Kazinga Channel. Methodology for annual catch estimates (tonnes) (June/July sampling 2001) per • landing per fishing method

• 1. Number of fishing boats (from FS) at each landing site (Table 9) for each fishing method was multiplied by: • 2. Mean catch per boat per day per fishing practice per landing site. 3. Multiplied by average number of days fished in a week per fishing practice per • landing site to give; catch per week per fishing practice. 4. Multiplied by 52 weeks in a year to give; estimated annual catch per fishing • practice per landing site 5. Summed estimated annual fish catch for all the landing sites per fishing practice ~.• to get the annual catch estimates (tonnes) for the entire water system of Lake

• George and Kazinga Channel. • • • 6 • • • FisIJ.rfts Stodios 00 W. CHrg. , lluiJJ,. Cbusu

3. RESULTS

Section A.

3.1. Global (Lake George and Kazinga Channel) picture of the fishery

There is a total of 8-gazetted fish landing sites on Lake George (6) and Kazinga Channel (2) and other unknown illegal landing sites scattered along the shores of the lake and the channel. A total of 426 actively fishing boats (FS 2001) compared with 388 (FS 2000) were recorded on Lake George and Kazinga Channel (Fig. 2a). These boats increased by 10% from those recorded in 2000 FS. However, the information gathered from communities showed that the boats in 2000 and 2001 stayed relatively stable (493 in June/July 2001 and 499 in November 2000) - Fig. 2b but showed 67 boats higher than the FS figure obtained in June/July 2001 FS and 111 boats higher than the November 2000 FS.

The total number of boats was disaggregated by license types during 2001 FS. In 2000 SUNey, ILM was assigned the task, as FIRRI team could not handle the issue during the first sUNey of the kind because correct information would not be given. The community information (2001) for boats disaggregated by license types is shown in Table 2. Five categories of licenses were recorded. These were licensed (40.6%), veteran (7.5%), experimental (26.8%), Kawonawo (1.2%) and illegal (23.9%).

7 ~

-Kahendero DHamukungu IllIKasenyi DKashaka • 500 (a) lmMahyoro [JKayinja -Katunguru-BDKatunguru-K 450

• 400 • 350 -. 3(XJ 250

200

• 150

100 • ., 50 -..o .t:l ... () • ...o ... .t:l E -Kahendero o Harru.d

• Fig. 2. Number of boats (a) observed (FS) and (b) from community information during • FIRRl surveys of November 2000 and June/July 2001. • 8 • • ~~---·········~~LLa_J

Table 2. Distribution of fishing effort (boats) on Lake George and Kazinga Channel (from community information) disaggregated by licence types (June/July 2001) compared with community information Nov. 2000 and FIRI-FS 1997.

Water body Landing Community information FS/FIRI FS FIRRI FS/FIRRI 1997 2000 Jun/Jul 2001 Licensed Veteran Exp Kawonawo Illegal Tot. Jun/Jul Tot Nov. ,2001 2000 L. Georae Kahendero 36 9 20 5 30 100 120 103 82 69 Hamukungu 40 11 20 30 101 90 179 82 83 Kasenvi 28 20 12 60 57 48 47 51 Kashaka 30 20 50 50 73 40 54 Mahvoro 22 11 22 1 20 76 60 113 50 62 Kavinia 14 6 20 6 46 42 31 38 51 Sub-total 170 37 102 6 118 433 424 547 339 370 Kazinga Katunguru - K 15 20 35 38 32 39 Channel Katunguru - B 15 10 25 37 17 17 Sub-total 30 30 60 75 49 56 Total 200 37 132 6 118 493 499 388 426

9 ------• Fbb.d.s SfrldJ.. lUI l.dt Gtorp " l£ubJg. CbUlltl -. 3.2. Fishing effort (boats) in relation to the fishing practices

Nine different fishing practices were identified during the surveys (Fig. 3)c The major • fishing practices by boats were active tycoon 33.4% (2001 FS) and 22.9% (2000FS): long line 24.4% (2001 FS) and 31% (2000 FS); passive gill net 27.4% (2001 FS) and • 31.2% (2000 FS). There was a reduction in use of 4.5" mesh size of nets in passive fishing boats from 97.2% to 87.4% of the boats between Nov. 2000 and June/July 2001 • FS (Fig. 4a) and increased use of 4" and 3.5" meshes during the same period (Fig. 4b). In active tycoon fishing, the percentage of boats using 3" and 4.5" decreased but those • using 3.5" and 4" increased (Fig. 4b).

• In other gill net fishing methods the following were major gill net mesh sizes in use c.• during 2001 FS Fishing method Mesh size (inches) Percentage boats 2001 FS 2000 FS Mukira passive 4",4.5" 47,47.5 10.9,83.6 • Mukira boat seine 4.5" 82.4 70.5 Boat seine 3.5", 4" & 4.5" 39.9,33.7,5.8 0, 31.6, 55.3 • I [5'1ukira tycoon 4",4.5" 54.5,45.5 0, 100 • I There was increased use of hook size 9 and reduced use of size 8 and 7 between • November 2000 and 2001 FS. However this was not a major change in hook sizes in • long line fishery during the two surveys (Fig. 5). 3.3. Average number of gears per boat (FS) June/July 2001 and November 2000 • for the main fishing practices. • The average number of gears (hook, gill nets) in active tycoon and long line remained relatively constant but there was a reduction (18%) in average number of passive gill • nets per boat between 2000 and 2001 FS period (Table 3). • 10 • • .. ~ Tim"i..Stsdi•• /Ill LU. "Dr,• .. /luis,_ ClJmn.J

lOll

90

80

70

.9=

o Passive Gill nets ~-= 'i: 60 • Active Gill nets -=CO (tycoon) OJ • Passive Gill nets ..OJ) (Mukira) -'"..= 50 o Boat seine OJ

... I!l!l Active Gill nets .. ~ (Mukira) O Boats seme 40 (Mukira) IIiI Longw1e o Hook and line 30 II Basket trap;

20

10

o I I Nov. 00 (LG + Kq Jlnt1 uly 01 (LG + KC)

FIRRI surveys for ILM

Fig. 3. Overall percentage contribution of boats by fishing practice in the fishery of Lake George and Kazinga Channel based on (FS).

II • Flsb,dn SI.dln 0" UM GHrp of 1IJuIIJ!1. CbillJlJtl

• 100 (a) 90

• BO • 70 l1li3.5" 60

50

40 1llI4.5' , • 30 c 20 .51 • 10 .Q-= ';:: 0 -=0 • '" ... OJ) '" llx) -...c (b) • ... '"... 90 "'­ • 80 70 • 60 SO IilJ 4. 5" 40

30

20 • 10 0 • Nw. m (LG + Kq JureA u1y 01 (LG + Kq • FIRRI surveys for IlM

Fig. 4. Comparison of percentage number ofboats using different gill net mesh sizes of nets in (a) passive gill nets and (b) active gill net (tycoon) on Lake George and Kazinga • ,h"lInel • 12 • •

• ----­ ----­

------~ FIsb.nts Smdi.s l1li Ldt G••". " JIubJ,. CbUIJ.i

I 100

I• ~

80 ~

I Iw_ ,W . OlD I = .S? 60 .c-= I 1- 1liI9 I .;: -= '"OJ I 08 OIl GIiIBImIIIIIMIiI I '"

-OJ= .0 ~ .. 1ift,1I=¥¥hlI1 117 OJ'" I I I I Cl.

20

0+1----' !'bY. 00 (LG + Kq J..-eAuly01 (LG+ Kq FIRRI surveys for ILM

Fig. 5. Comparison of percentage number of boats using different hook sizes in (Iongline fishery of Lake George and Kazinga Channel (CAS Nov. 00 and June/July 01).

13 1II1II F/sbufts Smilts 01/ LUr CH". , 6uIII,. ClJVlIIl1

Table 3. Average number of gears per boat (three major fishing gears) on Lake George :. I­ and Kazinga Channel FS June/July 2001 compared with November 2000. ~ear Passive gill nets IActive tycoon Long line (Hooks) -­ JunelJuly 2001 52.7 3.1 599 I • lNovember 2000 64.1 3.2 580

'. The average number of gill nets and hooks per boat was higher than the legal limit of 10 I and 100-gill nets and hooks per boat respectively. An average of 7,362 gill nets and 60,809 hooks operated on Lake George and Kazinga Channel June/July 2001 • compared to 8292 gill nets and 72,350 hooks in Nov 2000 FS. 3.4. Overall Fish species composition in commercial catches of Lake George and • Kazinga Channel.

• There was a noted decrease of P. aethiopicus (Mamba) and B. docmak (Semutundu) and an increase of 0 .niloticus (Ngege) in June/July 2001 compared to November 2000. • The major commercial fishery in 2000 was P. aethiopicus (47.4%) while in 2001 was O. niloticus (39.3%) (Fig.6). This was noted as a major change. C. gariepinus remained • relatively the same. • 3.5. Global size distribution of the major commercial fish species on Lake George • and Kazinga Channel. • 3.5.1 Length frequency distribution • There was reduction in O. niloticus modal size in commercial catches from 24 cm TL (2000) to 22 cm TL (2001) (Fig. 7 a i & ii). A slight reduction in mode for P. aethiopicus • (Fig. 7b i & ii) was noted. There was slight increase in the mode of B. docmak in commercial catches of 2001 compared to 2000 (Fig. 7c i & ii). Larger sizes of C. • gariepinus were caught in 2000 compared to 2001 (Fig. 7d i & ii). Use of 3" mesh size • 14 • •.. ------_.- - ~------­ I , FIsIJ.n.. SlrIdJ.. •" £aU G••". IllubJg. CbiJUJoJ

11 100 rl 90 Il (l 80 r 70

o T. zillii

= 60 r ~ • M ka flllWne .J:>= r 'i: DB. allianalis -=0 ..<.I 50 IiiIP. aeUl;opictis ..o.l r -= lIDO. leuwshchlS ..<.I ..... 40 00. lIi/oliclIS Q.. r • C. gariqil1llS l1li B. doclIUlk r 30 f 20 ( 10

0+1----'

• r-kN. OO(lC' Kq Jun'~ uly 01 (LC + Kq • FIRRI surveys for ILM Fig. 6. Overall percentage comribulion by weight of individual fish species from • Lake George and Kazinga Channel (Nov. 00 and June/July aI) • • 15 I I I I. FhlJuks SllIdi•• '11 W. CHrg. If 1IuiD,. CbiUJJJtl

Si,-€ at 50% maturi Iv ~ i""-- Sire at 100% rmturitv >J , P"-

Ac~ve Gill nets (N= 5669) 20

10 l1 ----L-l-L~-C]:::::::J====!===>o_-_-_-_ L_-==c=t,, , • i=1I=ne=lts-_~ :;1 dflffibt:::::::lc:::)::=1""A=c=lti_Ve_G...... luki_.r_a_(N_=_49_4_) • , , , , , , Boat seine (N- 278) 15 , • to j , 5 • o -.c==i-l_!--J.--!--I-+~--1--J-lI--.L--I::::====>-_= • l_~---1r=t=I11rhI==!==","" B_oa~tse~ine-_M_uki_.r_a_(_N_~~2_1_8~)

• Traps (N~ 72) ~~L • 1~ ._~~-i=d-l=t=4=l=LJ-l-LJ-l=I::::J=::):=+-

20 Overall (N~ 9757) • 15 • 10 l!J 17 18 19 20 21 22 23 24 25 2& 27 28 29 30 31 32 33 34 35 • T ota length (em) Fig.7a (i). The overall length frequency distribution ofOreochromis niloticus caught by the different fishing methods in commercial catches from Lake George • and Kazinga Channel (June/July 01) • 16 • I FJ<1Iufn Studln Oil W. 6.." ... IIaWI!la CbiUUltl

50'0 maturirv ~l00%lTIdturitv

r 20 .' , Passive Gill rels (N~ 1935) r 15 10 , r 5 ,

lJ 20 : : (N~ r , , Active Gill rels 5669) , '-­ 15 , .--­ , ....-­ 10 , f-­ r , co­ f-­ 5 I f--­ ;., I I I r ... 0 C

=.. a' 25 Boat 50ire (N~ 64) .. , .:: 20 , : r , , ..OJ) , , '" 15 , , c , -..... 10 , r ... , =-.. 5 ,

r 0 r==l 20 Trap; (N~ 134)

15 , , , , I , 10 , , 5 ,

I ,

[) ~~ I 20 Overall (N~ 2155)

15

I 10

I 5

[)

16 17 18 19 W n II B ~ ~ ~ V ~ ~ ~ ~ n n ~ 35

I Tot. length (em)

I Fig. 7 a (ii). The overall length frequency distribution of Oreochromis lIifo/;CIlS caught by the different fishing methods in commercial catches from Lake George and Kazinga I rhannel (Nov. om 17 I I • q

, maturity J5 50 % matur ity --: :-+-, 100'1'0 . (N~ I. , , Passive Gill rets 1419) , 10 , ., 25 , , , , 20 , 15 , , • 10 5 • 0 il SO Passive Gill nels-Mukira (N"'" 106) .0 - 30

I 20

lO

I ;., 0 "... Hook and line (No: 46) "::l H" <:' ... 12 I ..= ... 10 I - ..QJ) , 8 " ... 4 ..." 2 I I:l. 0 12 Longline (N~ 682) , , III , - ',-­ , R I ,-­ 6 - 1--'-­ 4 2 ri r l • 0 30 Overall (N~ 2353) 25 • 20 15 10 • 5 II ~ ~ ~ ~ ~ ~ g; :0; ,1; ~ ~ ~ g ~ ~ ~ ;l ~ g :J. .. ~ ~ ?i ~ ~ ~ ~ ~ ~ :?, ,£ IS &\ :A d '" l '" • ~ :Ii ~ '" ~ " • " 3 • Tolaltength (em) Fig. 7 b (i). The length frequency distribution of Protopterus aethiopicus caught by the different fishing methods from Lake George and Kazinga Channel (June/July 01 • FS\ • 18 • •.. ~ FlslJ.ri.s Studl6 DJJ Lak. CoDr,. " 1UziJJ'8 Cb8JUJ.1

---.: 30 50% matu

, ,, Passive Gill nels (N~ 976) , , • 25 ,, , 20 • 15 10

• 5 o , I I , I I I I , t==1 ,_, , _ .. ~,

• 20 , Hook and line (N~ 42) , 18 , , 16 , , 14 , ,

12 ,

10 , I ,

;., Cj 8L~ ..= ~ ; "a o ---etIl, • .. , .:: ,

.. 14 , LongJine (N~ 219) 01 :, , .. 12 , , I , , l-­ • -= , , ..Cj 10 , ... , , .. , , Il. 8 , , 1--' • 6 , I---r-r--" L.., l-­ I 4 I­ 2 rr- -f l o 0+------.------­ , I , =

, 25 , Overall (N~ 1237)

, ~ I 20 I-­ 15 I , r--" I , 10 , , I 5 l­ 1b:D I

o i I I

~ ~ ~ ~ ~

;<, ~ ~ ~ ~

i': • iF. $ ~ R •

'f 0'

;J; .,\ ~ ~

:l: ~

g ~ ~ ~ :i: ':I :i ~ R ~ ~ ~ ~ :;; 6 :<, :;; " I • '" § ::: - ­ fc1 0' ~ Total length (em) I Fig, 7 b (ii). The overall length frequency distribution of Proloplerus

aelhiopicus caught by the different fishing methods in commercial catches from

I r ,~lce Geor"e ~nrl K~7in,,~ rh~nnel (Nov. om • 19 .' I ~ , Flsb,d.. Sflltll.. /lII r,u, CHry, .. 1IuJR,. CIJUJlloJ , , 50% maluri!)' , :.- 100% maturity 30 , ---­ : Passive Gill rels (N= 1188) , 20 , , , • 10 , , 0 • 40 Passi"" Gill rets-Mukira (N= 459) 30 20 • 10 0

­ 40 Active Gill rels (N= 397) .. ;., 30 "... = 20 ;:l ...a' 10 .. .:: 0 ... ell '" 40 Acti"" Gill rets-Mukira (N= 295) .. -= ... 30 ..." 20 ~ 10 • 0 40 Boat seires (N= 75) 30 • 20 10 ". 0

50 Boat seire;-Mukira (N= 99) 40 • 30 20 10 ~-~~--~-~------. • 0 ~- -. 40 Overall (N= 2513) 30 20 10 • 0 • Fork lelU!th feml Fig. 7 c (i). The overall length frequency distribution of Bagrus docmak caught by the different fishing methods in commercial catches from Lake George and Kazinga • Chennell.lllne /.Illlv 01 \ • 20 -. • III ~ FisIJ.drs StatIJ•• DJI Ultr (;HI" , Kum,. Cll.""r/

, , 30 50 'to maturity -:, ,.... 100')0 maturity , ,, , , 25 , Passive Gill nets (N= 1332) , { 20 , ,, , , ]5 , ,, { 10 , 5 , 0 .0 Active Gill nelS (N= 268) 35 I 30 25 20 I 15 >. '" 10

=... 5

~ I "... 0 I I I , , [ , I ~ . _, __ ..:: ... 35 , I 0.< t--­ Longline (N= 82) -=...'" 30 ..'" 25 ... ~ I:l. 20 r-­ I r--­ 15 I 10

5 ; ~ 0 I , I , 30 ,, Overall (N= 1682) I 25 11-­ 20 I-­ 15 .--­ I 10

5

I 0

~ ~ ~ ~ ~ ~ ~ ~

N ;1, ~ ~ R

J, ~ ~

:;;:

......

6 ~ ~

:x ~ ~ ~. ~ ~ ~ ~ ~ • - N ,:; '" ,. ,. I Fork length (em) Fig. 7 c (ii) The overall length frequency distribution of Bagrus docmak caught by the different fishing methods in commercial catches from Lake George and Kazinga I Channel (Nov, 00) I 21 I I • '~ Fisb.ri•• Stlld/e. 011 T.alre lIIo,,. If 8uiIl,. Ch."lJtl

40 Passive Gill nets (N~ 329) 3S • 30 25 20 15 • - ;­ 10 5 .-- r----L • 0 60 Passive Gill nets-Mukira (N= 69) 50 • 40 ;\0 • 20 10 ;., (j 0 =... • 40 Active Gill nets (N~ 41) = r:1'... :\5 -= 30 ... 25 • ~ eo: 20 -...= 15 (j • ...... 10 5 "" 0 • 14 Longline(N= 573) 12 10 8 • ;-­ 6 f-­ ;-c-­ 4 • 2 hL 0 • 25 Overall (N= 1012) 20

15 • III 5

U ~ ~ ~ ~ ~ ~ ~ ;; i" 'f ~ ~ if ~ • ~ ~ :); J, J, 0 J, :J: " ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 0>'" ~ ~ " " 0' """ " ~ • Total length fern) Fig. 7 d (i). The overall length frequency distribution of Clarias gariepinus caught by the different fishing methods in commercial catches from Lake George and Kazinga • f:h"nnel (.JlInel.l"lv 01 \ • 22 • •.. • Fisb,d.. SiudJ.. all L.k, Cur" " 1lHhJ,. CbilJllJ,i r 3D .--­ Hook and line (N= 39) r 25 20 ­ r 15 r-- I--­ 10 r-­ r .--­ 5 I r o .0.0_.__ 18 Longhne(N= HI) f 16 14 I 12 10

8 .... I Col ..= 6 =a 4 I ...... 2 Ol os o -..= I Col ..... 16 Overall (N· 180) Q., I 14 12 I 10 8 I 6 4 I 2

0

0 0 0 0 0 0 0

~ ~ ~ ~

i'l ;;, :.I :; 0 ~

'f l'

~ .;, 0 .;, ~ :): .;, ~ .;, .;,

~ ~ ~ ~ ~ ~. ~ ~ ~ N :il ~ :i\ '" 0 0

I " " " " ~ Tolal length (em) I Fig. 7 d (ii). The overall length frequency distribution of Clarias gariepinlls caught by the different fishing methods in commercial catches from Lake George and Kazinga I rhannel (Nov. om 23 I I • Fisbrd..Studi.s DII I.do CHrgo " Huill!!" Cb.""t/

50% maturity 30 Oreochromi.r; nilotic~.r; Active Gill nets (N= 152) 25 • 20 15 10 5 I I I I I • 0 , 40 Boat seine (N= 48) • 30 20 - • 10 I I (] I

30 , Overall (N= 200) • ..., 25 "C '" 20 =a 15 • r-­ ..::'" 10 '" 5 I I I '" 0 I • -C .."'" Oreochromis lel/costictus ~'" 40 Active Gill nets (N= 214) • 30 20 • 10 I I 0 • 50 Passive Gill nets (N= 3--1) 40 30 • 20 10 I I I I 0

• 40 Overall (N= 248) 30 • 20 10 I h o • 16 17 18 19 20 21 22 23 24 25 26 27 28 • Totallenah fern) Fig. 7 e (i). The length frequency distribution of Oreochromis nito/icl/s and Oreochromis lel/cos/ic/I/s caught in gill nets of mesh size 3" from Lake George and • Kazinga Channel (all 0. lel/cos/ictl/s were above 15 em TL (size at first maturity)­ June/July 01 FS. • 24 • • II Fi./",i" Sfrldi•• DB Lak. CHrv• .. KulIJgA CbAIJIJU

:­ 35 , 100% maturity

Il , Active Gill nets (N~ 112) Oreochromis lIilotic~s , ,....-----, r 30

.... 25 r 20 r 15 r 10

5 r "..'"= =r::t' ..::.. o -C:l-U-_-, ..OJ) r '" "0 Oreochromis leucosticbls Active Gill nets (N= 2-16) -..=

- '"... 35 .. . ~

30 I 25 I------, r 20 r 15 10 I--­

r 5 o.l===I---l----l--l----.j~--l--I==l==T__~-~~-~~

I 16 17 18 19 20 21 22 23 " 25 2b ~7 28 " Total length (em)

Fig. 7e (II). The length frequency distribution of Oreochromis lIilotiCliS and • Oreochromis leucosticills caught in gill nets of mesh size 3" from Lake George and Kazinga Channel (all 0. leucostictus above 15 cm TL (size at first maturity)- Nov. 00 • FS. • 25 • • • ~1111 Fisb.d..Slail.. 011 r.u. CH"• .. IIuiIIg. CbUlltl

of gill nets brought O. leucostictus in commercial fishery and O. niloticus landed had the •• same mode as O. leucostictus (Fig 7e i & ii). .-.•• 3.5.2. Average size (kg) of major commercial fish species There was a noted decline in average size of O. niloticus and P. aethiopicus from 0.325 and 2.04 kg respectively in 2000 CAS to 0.266 and 1.66 kg respectively in 2001. C. • gariepinus and B. docmak average size remained relatively constant (Fig. 8).

• 3.5.3. Catch rates

• There was a decrease of 39.6% in catch rates (kg/boat/day) in passive gill net catches between 2000 and 2001 and 19.4% in active tycoon, but 10.6%, increase in long line • during the same period (Fig. 9). Some of the newly introduced fishing practices (boat seine, active gill net Mukira, boat seine Mukira) showed an increase but there was a • major decrease (74.4%) in hook and line between Nov. 2000 and June/July 2001 CAS.

• 3.5.4. Annual catch estimates disaggregated by fishing practices • There was 18% decline in estimated annual catch to 2073 in 2001 from 2513 tonnes of • 2000 CAS. The catches were mainly from passive gill nets, active gill nets and long line (Fig. 10). However, the estimated annual catch from passive gill nets and long line • decreased by about 43.1 % and 9.6% respectively between Nov. 2000 and June/July 2001 CAS. The introduced fishing practices (passive gill net mukira, boat seine, boat • seine mukira, active gill net mukira) showed an increase in estimated annual catch but hook and line was a decrease. In active gill net tycoon there was an increase of about • 19.2% of the estimated annual catch between Nov. 2000 and June/July 2001 CAS. • • • 26 • • III •­ rld.n.s SllIdi.s DB W. CNII' "WiJlga Cb.ilUJo/

I- 7

• 6 • • 5 I o T. 2lilii

• O. lellcostichlS ~ 4

.:.: ~ EJM. kllml/lHle • .!:l" '" liB, alhallali> OJ) I "os ... 3 o P. aethiopic//s ," < I 1m O. "doaclls Dc. gariepiJlu5

I 2 liB. docmak I I I 0+1----' Nov. 00 (LG" KC) JUlll'Auly 01 (LG .. KC) I F1RRI surveys for IL M

I Fig. 8. Average size (kg) of individual fish from Lake George and Kazinga Channel (Nov. 00 and June/July 01) I I I 27 I I, ~ FisIJorJo. Studio. DB r.m lI«J'go .. 8uiJJga fbautl

.. • 250 •

• 200 • • • Passive Gill nets 150 o Active Gill nets • ~~sfv~bll nets '0'" • ~ ~b~*\'1.e

- '"0 .Q ~ III Active Gill nets 01 .:.: IMukira) • ~ ., o Boats seine '" 100 iii (Muki.ra) -...'" Longlme .c •

• 0-1----- Nov. 00 (LG + KC) JuneAuJy 01 (LG + KC)

• FIRRI surveys for Fig. 9. Catch rates (kg/boatJday) for each fishing practice on Lake George and • Kazinga Channel (Nov.OO and JunelJuly 0 I) • • • 28 • •• • I FIsIJ.r/.. Sludi.s DB W. Cur,. 6 _" Cbun.1 Il Il 25DO Il r .- 2000

.­ .PdSSive GIll nets ,-.

I>Jl

1500 ~ o Active Gill nets (tycoon) 0 r 0 • PdSSive Gill nets 0 ~ (Mukira) .. EI Boat seine ... -'" .5 III Active Gill nets -..'" (Mukira) .c o Boats seIne (Mukira) ... '" 1000 -'" i!J Longline -;'" o Hook and line r = -(= III Basket traps r 500

..r

oJ I I I ~• I N...w. 00 (LC T Kq J~l1uJyOl (LG+ Kq

FIRRI surveys for ILM Fig. 10. Lake George and Kazinga Channel (Global) annual catch estimates for • di fferent fishing practices (Nov. 00 and JunelJ uly 0 I) • • 29 • • • r, -­ • SECTION B 3.6. Picture of the fishery broken down by the different landing sites on Lake • George and Kazinga Channel • 3.6.1. Number of boats (FSJ • The total number of boats per landing site (FS-June/July 2001) compared with (FS Nov. 2000) is shown in table 2. The highest number of boats was noted at Kahendero • and Hamukungu for both FS. Kashaka. Mahyoro and Kayinja recorded boat increase of about 24-35% since Nov. 2000 FS. The rest of the landings had more or less the same • number of boats. A decrease of about 16% of the total boats observed in 2000 FS was • recorded in the present survey. • 3.6.2. Boats organized by fishing practice Passive gill net fishing was popular on Kashaka, Mahyoro, Kayinja and Katunguru-B • during both surveys (Table 4) while active tycoon fishing was mainly at Hamukungu and Kasenyi. There was a reduction in active tycoon fishing boats at Katunguru-K in 2001 • FS compared with 2000 while increase was at Kahendero, Kashaka, Mahyoro and Kayinja. Increased use of new fishing practices between 2000 and 2001 was recorded • at all landing sites except Mahyoro, Kayinja and Kahendero (Table 4). Mukira fishing was introduced at Kasenyi and Kashaka since 2000 FS while boat seine was introduced • at Kasenyi and Kayinja. Hook and line declined in 2001 compared to 2000 FS except at Kahendero. Katunguru-B, Kasenyi, Mahyoro, and Kayinja were not using the fishing • practice. The decreased use of long line was noted for Kahendero, and Mahyoro in 2001 compared to 20000 FS. The rest of the landing sites remained relatively • unchanged during the two surveys. • • • 30 • • -. -. -. -. -. -. -. -...... - .IIIIIl .. ....... • Fisb.ri.. Sfrltli.. DIJ W. GMrg. If JluiIJya CbiUUJoi

Table 4. Breakdown (percentage boats) observed in FS June/July 2001 compared with November 2000 Fs (in brackets) organized by fishing practices on different landing sites on Lake George and Kazinga Channel

Waler Landing site FishinQ Qears/practices body Gill nets Hooks Traps Passive Active Mukira Boat Mukira Mukira Longline ! Hook & line tycoon passive seine tycoon boat I (Kikwakurizo) seine

Lake Kahendero 13.5 (15.9) 37.7(11.0) I 4.6 (6.5 31.9 11.6 (6.5) 0.5 (1.2) George (54.9) I Hamukun9u 16.5 (14.6) 41.5 (43.9) 7.3 (2A) 10.9 (3.7) • (7.3) 21.3 OA (1.2) 2.0 (4.9) (22.0) ! Kasenyi 16A (21.3) 42.6 (40.4) 5.2 (-) 0.7 (-) 07 (-) 46 (6A) 27.6 - (4.3) I (27.7) Kashaka 34.6 (50) 17.1 (5.0) 16 (-) 3.1 (2.5) 4.9 (-) 6.1 (-) 29.9 1.2 (2.5) 1.2 (5.0) 135.0) Mahvoro 46.6 (54) 42.3 (16.0) - (2.0) 8.9 (26.0) Kayinja 50.7(76.3) 27.6(-) 0.7 (-) 20.4 0.7 (-) (23.7) Kazinga Channel Kalun9uru - K 15.1 (15.6) 27.7 (43.8) 24.4 (6.3) 5.0 27.7 - (9.4) (3.1) (21.9) Kalunguru - B 47.1 (29.4) 7.6 (-) 21.6 23.5 147.1) (23.5) I Overall 27.4 (31.2) 33.4 (22.9) 5.7 (3.1) 3.6 (3.1) 1.2 1.4 (2.3) 24.4 2.2 (3.1) ~.7 (2.3) L (0.3) (31.7)

31 ~ - 3.6.3 Average number of gears per boat .. The highest average number of gill nets per boat was recorded at Kayinja and Hamukungu in passive gill nets during both surveys. The highest number of hooks per boat was at Kahendero for both surveys (Table 5). However, there was a general • decrease in the average number of gill nets (passive) at every landing except at Katunguru-K where there was an increase. Kahendero, Hamukungu, Kasenyi, • Kashaka showed an increase in mean number of hooks per boat while the rest showed a decrease resulting in a relatively stable average number of hooks per boat in both • surveys (Table 5).

• The estimated total number of gears (gill nets, hooks, traps) operating on Lake George and Kazinga Channel on a single day per landing is shown in Table 6. The highest • numbers of gill nets were noted at Hamukungu, Kayinja and Mahyoro while Kahendero • registered the highest number of hooks during both FS. However, there was a general reduction of hooks at Mahyoro, Kayinja and Kahendero. Use of basket traps also • declined at Katunguru- B. • 3.6.4. Distribution of boats using different gill net mesh sizes and hook sizes per • landing site per fishing practice Gill net mesh size nets of 3.5" and below were recorded at Kahendero, Hamukungu and • Kasenyi (passive and active tycoon) and in boat seines at Hamukungu and Kahendero during 2001 FS (Table 7). The last two landings recorded the highest increase in • percentage of boats using 3.5" mesh size of nets in active tycoon. There was also general increased use of 4" mesh size of nets in passive gill nets between the surveys. • Use of hook size 9 in long line fishing remained dominant at all the landings in both years with increased use of hook size 8 especially at Kayinja and Katunguru-B (Table •,. 8).

- 32 ..-.I III -. ...... -. .. -. -.

-.-.-.-. ~~ ~ j St ~D .- - ­ rn .. Ll orgr ...... fill.

Table 5. Average number of gears per boat for different fishing practices on Lake George and Kazinga Channel FS June/July 2001 compared to FS of November 2000 (in brackets)

Water Landing site Fishing gears/practices body Gill nets Hooks Traps Passive !ACtive Mukira Boat MUkira Mukira boat Longline Hook & line tycoon passive seine tvcoon seine IKikwakurizo Lake Kahendero 25.2 (30.4) 3 (3) 3.1 (3) 900 725 100 (93) 20 (5) George Hamukunau 83.4 173.6 3.313.2 9.4(11) 4.5 (3.7) - 19.7 581 461 50 (20) 9.2 19.5) Kasenyi 51 152 3.1 (3.5 10 (-) 5.7 (-) 61-) 10 (10 647 492 0(5) Kashaka 28153.8 3 3 121-) 7.2 (3) 9.71-) 10 (- 4091261 ) 120 (120) 10147.5) Mahyoro 64 (72 3 3) - (3) 525 (600) Kavinia 70192.1 3 - 12 (-) 517 (756) 1 (-) Kazinga Channel KatunQuru - K 38 (22) 3.3 (3.3) 6.5(18 6 (6) 395 (543) - (93) Katunouru B 21.5 /32.8) 3 (-) 2.316.5 458 (475) Overall 52.7 (64.1) 3.1 (3.2) 10.519.2 4.713.2) 8 (6) 10 (9.8) 599 (579.5) 102.6 (89.2) 11.0 (16.4) I

I Table 6. Estimated number of gears (Hooks. nets and traps) per landing per day On Lake George and Kazinga Channel June/July 2001 and I November 2000 FS (in brackets). I

Gear Fish landings I Hamukungu Kahendero Kasenyi :khaka I Katunguru Katunguru K Kayinja Mahyoro Overall I

Gill nets 11680 (905.0) 2270 (395.0) 515.7 (520.0) 552 (1075.0) B 255.0 291.0 (146) 1788.7 1956.0 6753.0 I passive (216) (2670) (1945) (8292) Gill nets active 152.3 (185.0) 91.0 (48.0) 907 (96.0) 99 (9) 4.0 (0) 470 (52) 460 (0) 780 (30) 609.0 (420) Total gill nets 1321 (1090.0) 318.0 (443.0) 606.3 (616.0) 651.0 (1084) 295.0 338.0 (198) 1834.7 2034.0 7352.5 (216) (2670) (1975) (8292) Hook and line 0(20) 800 (650) 0(0) 66.7 (120) 0(0) 0(2801 0(0) 0(0) 866.7 (1070) Longline 10260(8300) 19800(32630) 9053 (6400) 6813 (3650) 1833 4217 (3800) 5341 2625 59942 (1900) . (6800) (7800) (71280)

Total hOOkS-l 10260(832~0600(33280) 9053 (6400) 6880 (3770) 1833 4217 (4080) 5341 2625 60809 i (1900) (6800) (7800) (72350) Traps 32 (38) I 0(5) 0 3.3 (10) 0(95) 0(0) 0.3 (0) 0(0) 36 (148)

33 • • • ••• •••••••••••••• 1

FisII.d.. SfIlii.. 011 W. CHrv' ~ Ildzi1l!T' CbUJJId

Table 7. Percentage distribution of boats uSing different gill net mesh size of nets per landing per fishing method (June/July 2001 FS) compared with November 2000 FS (in brackets)

Method of fishing Mesh size Fish landinos Kahendero Hamukungu Kasenyi Kashaka Mahyoro Kayinja Katunguru Katunguru B Overall K Passive oill nets 3" 1.0 (-) 0.2 (-) 3.5" 57.1 (-) 1.8 (-) 4.1 (-) 2.9 (-) 4" 38.5 (39.21 15.31-) 29.2 (-) 14.4 (5.6) 4.9 (-) 2.2 (-) 51.2 (-) 11.9 (2.8) 4.5" 4.4 (60.8) 81.9 (97.7) 66.7 81.6 (94.4) 95.1 (100) 97.8 48.8 (100) 94.8 (100) 84.4 (97) (100) (100) 5" - (2.3) - (0.2) 6" 4.0 (-I 1.4 (-\ 0.4 (­ 7" 3.9 (.l 0.1 1­ Active tycoon 3" 21.1 (11.11 11.2 (20.7) 1.5 (-) 7.2 (9.4 3.5" 70.7 (44.4) 54.3 (5.2) 46.7 (4.6) 36.3 (7.3 4" 8.3 (33.31 28.4 (69.0) 51.8(-) 89.2 (50) 88.5 (22.2) 100 (-) 82.1 (15.2) 50.9 (36.5 4.5" "(11.1) 0.9 (2.6 - (95.5) 10.8 (SOl 13.5 (77.81 17.9 (84.8\ 1001-) 5.1 (45.8 5" 3.4 (2.6 0.911.0 6" 1.7(­ 0.5 (- Mukira passive 3.5" 22.5 (­ 5.5 (­ 4" 77.6154.5 85.7 (-) 70.4 (-I 47 (10.9) 4.5" - (45.5 14.3 (-\ 100 (-I 29.6 (100) 100188.5) 47.5 (83.6) 5" - (11.5\ - (5.5) Boat seine 3" 9.7 (­ 5.4 (-I 4.3 (-1 3.5" 58.1 (­ 580 (-) 39.9 (-) 4" 32.3 (28.6 11.6 (54.5) 100(-) 83.3 (-I 33.7 (31.6) 4.5" 5.4 (-) 16.7 (100) - (100) 5.8 (55.3) 5" 14.4 (45.5) 7.7 (13.2) 6" 5.4 1-) 8.7 (.l Mukira tycoon 4" 100 (-) 31 I-I 100 (-) 54.5 (.l 4.5" 69 (-) - (100) 45.5 (100) Mukira boat seine 4" - (44.51 12 (-) 20 (-\ 17.6 (29.5) l 45" - 155.2\ 88 (1001 . 80 (-1 82.4 (70.5)

34 \-­ ----~-- _~' , - -- .. .. __ .....

lU1J~ri~ Stumn lUI LUt CHrgt & 8ubJgtl (build

Table 8. Percentage distribution of boats using different hook sizes per landing per fishing practice June/July 2001 FS compared with November 2000 FS (in brackets).

Fishing Hook oractice size Fish landinas Hook and Kahendero Hamukungu Kasenyi Kashaka I Mahyoro Kayinja Katunguru K Kalunguru B Overall line 6 - 147.11 - (100) - It9.6f 7 100/52.91 - (30) 93.1 121.6) 8 - (100) 50 (-I - 11001 - (701 - (100) 3.4 158.8 9 50 (-) 3.4 (­ Lanoline 7 - (39.7\ 1.6 (-) 13.7 (-) 2.2 {5.0 8 14.4113.81 12.1 (13.6) 11.9 (-) 48.7 (44.6) 31.3 (32) 54.0 (-) 39.5 (51.5) 91.7 (56.81 31.6131.2 9 85.6 (46.6) 86.3 (86.4) 83.9 51.3 (55.4) 68.8 (68) 32.3 (100) 60.5 (48.5) 8.3 (43.2) 65.7 (63.9) (100) 10 4.1 I-I 0.51-\

35 .. • 3.6.5. Fish species composition in gill net and hook fishery The gill net fishery was dominated by O. ni/oticus in June/July 2001 (Fig. 11 a). • However, Kayinja and Katunguru-B, P. aethiopicus was the most dominant species. Other species of imporlance were C. gariepinus at Kahendero and B. docmak at Kashaka and Katunguru-B. During November 2000 (Fig. 11 b) P. aethiopicus was • dominant species in gill net fishery except at Kasenyi and Hamukungu where B. docmak was dominant. C. gariepinus was important at Kahendero, Kasenyi and • Katunguru-K while of all the landings, Hamukungu contributed the highest percentage of 0. niloticus. Contribution of O. /eucostictus was recorded at • Kahendero in 2000 CAS (Fig. 11a & b).

• There was an increase in percentage of 0. niloticus between 2000 and 2001 CAS in the gill net fishery in commercial catches at all the landings except Katunguru-B. • (Fig. 11 a & b). P. aethiopicus percentage contribution decreased at all the landings between the two surveys. However, C. gariepinus remained relatively stable during • the two surveys.

• Hook fishery during both surveys was dominated by P. aethiopicus at all the landings except at Kasenyi where C. gariepinus was the dominant species (Fig. • 11a,b). • P. aethiopicus in both surveys was the dominant species in hook fishery with higher percentages at Kayinja, Mahyoro, Katunguru-B in both years. However there was • increased percentage of C. gariepinus at Hamukungu, Kahendero and Kasenyi between 2000 and 2001 CAS. The details of percentage contribution of different fish • species in individual fishing practices per landing are shown in (Fig. 12 a & b). • 3.6.6. Species composition in all the eight fishing practices • In boat seines, the dominant species was 0. ni/oticus (Fig.12 a & b) except at Hamukungu(2001) Fig. 12a) where B. docmak dominated and at Kahendero and • Kasenyi where the dominant species were O. leucostictus and B. docmak respectively (Fig. 12b). Passive gill net fishery was mainly dominated by P. • 36 • ..• FlJI",I.. SID4Jts .11 LU. 'to"•• KuhJ,. CJJDIIIIU aethiopicus in both surveys except at Kahendero where 0. niloticus was the dominant species. Other gill net fishing practices were dominated by O. niloticus with the exception of Katunguru-B and Kashaka (2001- Fig. 12a) where B. docmak (Mukira passive and active) dominated. O. leucostietus dominated in active tycoon fishing at Hamukungu and Kahendero during 2000 (Fig. 12b) while in Mukira passive, B. docmak was the dominant species.

Though in long line fishery P. aethiopicus was the dominant species. in both surveys, Hamukungu. Kahendero and Kasenyi (2001) had C. gariepinus as the major fish landed (Fig.12a). Basket trap catches were dominated by O. niloticus.

37 ~ FlslI..l .. Studl.. 0" wn Coo". " Hazi"ga flJaud

mB. docmak &I B. allianalis • C. gtlrlcpinus -M kannwne m0. nilohclis UD O. leucosticfus • Gill net fisher V lip. netllior;fus o T. zill;; -. 100%

• 75%

• 50%

• c .;::0 :::l 25% .c f • c ...0 '"OJ) 0% • '"c - Hook fisher V '"..... '" --• "" • • • • O°{, ~ .>. ro ro co e ~ .0- e U c 0 ~ rn c .>. >- 2 '" '"~ " .c .c ~ "; + ~ c '" ~ " ro '" ~ " '" .c '" ~'" '" ::E c c 0 • E " '" ,3'" ,3'" '" '" '" I'" '" '" '" '" • Landing sites '" '" Fig. Ila. Fish species composition of gill net and hook fishery from • commercial catches of Lake George and Kazinga Channel (CAS JuneIJuly 01) •

• 38

- •

--• I11III • Fls1l.d.. Sllldl...s uk. ' ..rg. of ZU/sg. Cb"",.1 r mB. dOCIIUlk 1m B. alhllllll[ij • C. 811ricptllus • Nt k'WHllllk' 1m 0. lli!ohclIS mO. lellcosht:h,s II P. tlt'tllioviclIs o T. zillii r. Gill net fishery 100% mm! t~Il!J :lojiHill!!!' ::lS!! fi [!h.lili;~! ~1!1'lr 111!!i'.Ilnnl r ~Ii @. !IUI:! l!l! ·11i!.•P..; 1""1'I'llll! 75% :H'IJU',jJ ••• niiHl jiliij

l!qi' HIHII' !iimi .. :yni~ mi!!! m~l~ :1.,-' • 501}'1)

c= .;:: 25% ,~I:

P­= • I.

-c= <.I ~I"" ~ .. . !!lfll bedl Mfr. Ii\;wi! lE'i!l 111 hdl • ..bJl

-..= Hook fishery <.I 100%

I. ,jUi .. • il- ~ II!gR. • 75% I

50%

• 25% •

0% ~ .>. 0> e c -"! .C- e 0

c .>.'" ~ '"2 '"2

~

~ .c: ~ ~ + " ~ c .c:

" '" 0> 0> '" "

~ .c: '" '" c c • '" ::;; ~ 0!­ E " '" '" '"

'" '" n; '" '" I J:'" '" "''" ~ '" '" "> landing sites 0

I Fig. llb. Fish species composition of gill net and hook fishery from catch~s I commercial of Lake George and Kazinga Channel (CAS JunelJuly 00) I 39 I I • 1111

III SB. docmak liB. alliana/is -C gariepinus -M kannume Ell O. niio tinLs on O. leucostictus III 1~~ tLI;i~'~m6 L~ ~IDI_I:'ae~!Ooicus DT. zillii Q"\, "•.':," . , , ' l;,". ',. Passive Gill nets 100% 7S• 50'\ • 25\ 0% -I-.....='-~ I. Active Gill nets 100% 7S"b 50' 25% I. n Passive Gill nets ­ Mukira 100'\ 75% II 50% e ,g 25% -= 0% Active Gill nets - Mukira I ""';: 100'" e 75'\ -0 '"' 50\, "b1 25"'0 I 0' '"e - Hook and line "'"'... • ...." ,qU,....---.JI~I~~------.-JL Lon~ line- hooks too?o 75"'. • 50'\, 25'\ 0' Basket traps

• 75'\, 50% 100"LI25~o I l 0% L...... --~--~----' L~ ~ • Overall fishin't practices

75"~ 50% • 25'\, 0"" KitlUllQuru B I(~lungu'u I( O"er,,11 (lG'KC) • Landine sites Fig, 12a. Fish species composition from commercial catches by fishing practice, • landing site and overall Lake George and Kazinga Channel (CAS JunelJuly 0 I) •

• 40 • ..~

, .wk~ C1J.nn~J • Fisbuit$ Studln OIJ eN"., &lluJng.

.­ El B. docmak BI B. altiallalis -c. oariepi~ut5 -M kannume OJ O. ni/oticlls lID O. ?eucoshchl5 Boat seines Dip. aetJliovicus o T. zillii 100% 75% .­ 50% 25% 0% I rmm.-. Passive Gill nets

, lOOOk 75~

50%

f 2~~ I f"Y~:sm Active Gill nets 100% 75% 50% , 25% 0% Passive Gm nets - Mukira 100% 75% 50% 25% 0% I IW9 VAl killl IWfl [\Ii>;! , • =0 II ;:: I Active Gill nets - Mukira .;:.&>= 75%

-0= I ..'" '0011~~~ 07'; U;t1 •• Oll 0% ' " -..= 100% ::;'" 75% Q., 50%

• 25% 0% L o~ line- hooks 100% 75%

• 50% 25~4 0%

• Basket traps '~~1I~5~. • 0% I l Overall fishine: practices 100% 75% • 50% 25% 0·"

H;llmukuollU Kanenlleru Kaseny, Kalnua Kay,nl:l Mahyoro Ka,,,ngu(u 8 l

• 4\ • • • 1II1II

III 3.6.7. Size ofthe major commercial fish species I. The major notable change during the two surveys was a decline in average size of 0. niloticus and P. aethiopicus at all the landing sites (Fig. 13). B. docmak though with variations in average size at different landing sites showed more or less the • same mean size. A major increase in average size between 2000 and 2001 was noted for C. gariepinus at Kahendero, otherwise, the rest of the landings showed '. size decrease. • 3.6.8. Catch rates There was an overall decline in catch rates (kg/boaVday) at all the landing siles • except Hamukungu and Kasenyi, which showed an increase between November 2000, and June/July2001 (Fig. 14). Active tycoon MUkira was generally introduced • during 2001 and had the highest catch rates at Kasenyi followed by Hamukungu compared to other fishing practices at other landing sites during 2001 and 2000 • CAS. There was also a general decline in catch rates in gill net active tycoon especially at Mahyoro and Kashaka landing sites. A general decline in catch rates In • passive gill net was observed almost at all the landings with highest decline at Mahyoro, Kashaka and Kahendero while in long line, catch rates were relatively • stable in both surveys except for an increase noted at Katunguru-B and Hamukungu • (Fig. 14). • • • • •

• 42 • • III I FlsbBlIs Stud/IS 1111 W, CNIf' " JIni1l911 CJJUIIII

I 7 Nov. 00 III B. docmak Dc. ga rie PI1~US en O. niLohc/ls Dp. aethiopicu5 liB. altianalls ElM. kamlllllle I , .0. leucoshctus o T. zillii I , I 3

I 2

I -

~ ~""' 0 OJ I .;;;N OJ J untiJ uly 01) ...""' 7 1 II B. docmak DC. gariepillus 111 O. nila hells "OJ o P. aethiopicus III B. altiaMlis EJ /vi ka III/HIlle ;. .0. leucostictus o T. zillii I 00( , I I , I 3 I 2 I o H"mw..UOKU Kolhi-ndero Kolseny. KdSh.l.kol KolYllljol Mahyoro K.lrunKuru B K"lunKuru K O..er,,11 (LL + KC) I Landing sites

Fig. 13. Average size (Kg) of individual fish by landing from Lake George and • Kazinga Channel (Nov. 00 and June/July 01) • • 43 • • • III III Nov. 00 300 III Baskellraps CI Hook and line IllI Longline CI Boals seine (Mukira) '-II III Active Gill nels (Mukira) EJ Boat seine • Passive Gill nels (Mukira) CI Active Gill nels (tycoon) !-II 250 • Passive Gill nels

200 II 150

~ ;>, .. 100 • "" '"..0 ~ 50 • '"~ "'" -..... o .c • u u-.. 300 • }until uly 01 • 250 200

• 150

• 100

• 50 • " HarllukllTWJ Kiilwl1l.\?ro KasE'fl\; K.uhaka Ka~iJ1a M

• Passive Gill rets

III 500 o Active Gill rets • W~'JZ~)Gill rets ...,(Mukir~) 'III 400 1;::1 Boat selre • Active Gill rets (Mukira) o (loots selre (Mukira) I. '00 IllJ Longlire

~ OIl 2lX) o Hook and lire 0 ""0 '. 0 ~ ­...'" 100 • .5" -...'" o .c u • -u TunliT ulv 011 .." 600 = • -0:= • 500 • 400 • 300 • 200 • 100

Ka,o;tuka K.t~ui~ Maho.'oro KahJWru B Ka~ K • Landing sites • Fig. 15. Lake George and Kazinga Channel annual catch estimates for different • fishing practices by landings (Nov. 00 and June/July 01)

• 46 • • II

1'islJ.rJ.s StrJdln Q.D r.u. eN". , KuiIJg~ CIJ~IID.J

3iXXJ

2500

200:> ~ OJ) o Kdtuguru- K ..:.: 0

0 • Katunt,'l1ru-B

~ ~ '" EI Kdyirjd -'" 1500 .5'" III Mdhyow -'" -='" DKdShakd

..'-'

..'-' III Kasenyi ::s <:: 100:> HarruJkungu <:: j o « ~ ~III.. . mKdhen,'ew

50]

o I I

t\bv. 00 (LG T KC) ]1h!,1u1y 01 (lG T Kq

FIRRI surveys for ILM

Fig 16. Lake George and Kazinga Channel annual catch estimates for different landings (Nov. 00 and June/July 01)

47 StrJdJ~ W~ CHrr~ FisII~d~ 011 " 1Uzi1l'.t'JIaJl6

Table 9. Total number of boats observed in FS per landing per fishing practice November 2000 compared to June/July 2001 FS (in brackets)

'fishing Dractice Fish landings I Gill nets Hamukunqu Kahendero Kasenyi Kashaka Katunquru B Katunquru K Kayinja Mahyoro Overall Active boat 3 (8.3) 8 (3.3) - (1.0) - (1.7) (0.3) 1 (-) 12.0 (14.7) seine -~ Baal seine 6 (-) 3 (1.7) - (4.0) 8.9 (5.7) Mukira Mukira tycoon - (0.3 - 2.0 1 (1.3 1.2(3.7) Gill net tycoon 36 (35.3 9 (26.7) 19(21.7 2 9.3 - 1.31 141117 - 114.01 9 (26.0) 88.9 (146.0 Mukira 2 (6.0 - (2.3 • 0.3 8 3.71 219.7 12.0 122.0 Passive 12(13.3 13 (9.01 10 (9.7 20119.31 5 8.Dl 5 (6.0 29 (25.7) 27 130.5} 121.1 (121.5 Hand line 1(-1 7 (8.0) 1 (0.3 3 (- 12.0 (8.3 Lanoline 18 (17.7 45 (22.0) 13114.0 14116.7 414.01 7110.7 9110.3 13 (5.01 123.0 (100.3 TraDs 4 (2.0 2 (. 210.3 - 10.3 8.9 (2.7 Total 82 (82.6 82 (69.0) 47150.7 40153.9 17117.01 32 (39.41 38 (50.6 50 (61.51 3881424.8

48 Fld,n,s Stud/'s 011 LB, �", • b:dIJ,,, O ..~l

4. DISCUSSION

The legal number of boats permitted by law to operate on Lake George is 144 and possibly 30 on the two landings on Kazinga Channel. Considering the November 2001 community information of 493 boats, there are 319 extra boats. Presently (2001), 200 boats (from community information) have been licensed (Table 2) and additional licenses included those which were given to the people who were retrenched or demobilized from the army- veterans (7.5% of boats from community information), those who participated in World war one and two (Kawonawo) usually very old men (1.2%), the experimental boats Which from gathered information were authorized to operate on Lake George and Kazinga Channel by the then Minister of State for FiSheries (1996-2000), 26.8% and the illegal ones (23.9%). Such licenses make it difficult to control the fishing effort on the waters. The categories of these boats were not recorded during the FS of November 2000. This was for ILM to handle because if the team got involved, response from the communities and fishers to the other questions regarding the fishery would be negative for fear of the outcome. It is not known therefore whether there was an increase or decrease of these boat types of license categories on the waters. However, the FS of 2001 showed a 9% increase in total boats from those recorded in November 2000.

Although the number of boats (community information) for the two surveys were more or less the same (493) and (499) in 2000 and 2001 respectively, these figures are far higher than 388 and 426 recorded in 2000 (21.0%) and 2001 (14.6%) FS respectively. This shows that collecting reliable effort data remains a big challenge.

The number of boats (FS) at Hamukungu, Kasenyi, Mahyoro and Katunguru - Band Katunguru - K remained relatively the same during the two surveys. A decrease since 2000 FS was noted at Kahendero. Other landings showed slight increases. This situation is not alarming but a challenge and problems of collecting the exact information.

The number of gear (gill nets, hooks) per boat is higher than the legal ones supposed to be used on the lake (10 nets per boat and 100 hooks per boat). Though these were established long time ago, the average number is very low. In both

49 I~

III surveys, the average per boat is very high (52 & 64 gill nets/boat; 599 & 580 hooks/boat in (2000) and (2001) respectively from community information data for III passive gill nets and long line. Variations from year to year and landing site to landing site were also noted. With subsequent sampling, an agreeable effort figure '. will be established. Of recent, the longline and gill nets set on a single day virtually cover the entire lake I. and the channel with more of the gill nets at Mahyoro and Kayinja and hooks at Kahendero (Table 6). These landings since 2000 FS have been recorded having the I. highest number of gears (gill nets or hooks) compared to the rest of the landings.

There was global increase in active tycoon fishing boats and passive gill net Mukira, '. and reduction in passive gill net fishing boats and hook and line between November I 2000 and June/July 2001 (Fig. 3). Gill net active tycoon increase between 2000 and 2001 was mainly noted at Kahendero, Kashaka, Mahyoro and Kayinja (Table 4). At I Kayinja the increase was due to about 10 fishers migrating from the neighbouring illegal landing of Nyakera looking for fish market at Kayinja. There were no tycoon boats at Kayinja during November 2000 and all fishers from Nyakera carried out • tycoon fishing.

• The increase in active tycoon fishing and reduction in hook and line boats could have been due to a response to reduction in catch rates in passive gill nets (Fig. 15) which • usually occurs during dry seasons and possibly also could be due to laxation in fisheries management. Hook and line targeting P. aethiopicus and C. gariepinnus • are also more popular during rainy season. The June/July 2001 was a dry seasons. Hook and line fishing was abandoned at Katunguru-K, reduced at Kashaka and • Hamukungu during 2001 (Table 3). Joining of two nets increased during 2001 compared to 2000 (Table 4) and targeted mainly B. docmak and O. nifoticus better • than in passive gill nets (Fig. 12 a & b). • However, the introduced fishing methods (passive MUkira, Mukira boat seine (7c i), could be aiming at catching a lot of fish. If larger mesh size of nets were used as in • passive gill nets, mature fish would be landed. Introduction of new fishing methods could be a response to targeting increased catch rates as shown in Fig. 9. The

• 50 • •..

Fldm~s StutlJ~s fJJI LUI GHrg, " /UzbJg. C1IAJUJ~J reduction in average size of the major commercial fish species (0. ni/oticus and P. aethiopicus) in the commercial fishery between November 2000 and June/July 2001 (Fig. 8) was due to reduced use of larger gill net mesh size and nets both in passive and active tycoon fishing and increased use of active tycoon fishing during the 2001 survey in case of 0. ni/oticus (Fig. 3 & 4).

The average number of gill nets per boat in active tycoon remained relatively the same during the two surveys while in passive gill nets there was a decline (Table 5) especially at Kashaka, Mahyoro, Kayinja and Katunguru-B (Table 5). This could have also been the cause of decline in catch rates In passive gill nets (Fig. 14) and subsequent decline in estimated annual catches in the fishing practice. The landings that were mostly affected had reduced catch rates in the fishing practice and SUbsequently decline in estimated annual catch.

Though the longline fishing boats remained relatively stable between 2000 and 2001. a major reduction was noted at Kahendero and Mahyoro (Table 5) resulting into reduced estimated annual catch (Table 4). The most notable change was the increase in catch rates of the introduced fishing practice (Mukira gill nets). The catch rates increased between the two surveys notably at Hamukungu, Kashaka and at two Katunguru landings (Fig. 14). Mukira fishing contributed significantly to the estimated annual catch (Fig. 15) at these landings and could soon become a popular fishing practice on the water system to harvest fish in deeper waters. Catch rates dropped at Kashaka and Mahyoro in active tycooning (Fig. 14) probably because most of the fishers mainly went out during the morning and returned at or before 2.00 p.m (Table 1) thus fishing for fewer hours.

Early return to base of tycooning boats was common at these two landings during June/July 2001. Global drop in tycoon catch rates could be seasonal. Possibly higher catches are obtained during rainy seasons when O. ni/oticus tends to be actively moving. Catch rates of different fishing practices over time provide useful in sight of the status of the fishery as a decline in catch rates when the effort is fairly constant indicates that the fishery is declining. However, catcn rates may cnange in different seasons or different patterns of fishing. Indeed, one of the reasons for the

51 1111 rt6ltnt•• SlrltU...11 W. GH'P , 1uiII,. 01/1/1/.1

III regular monitoring is to identify any seasonality in the fishery of Lake George and Kazinga Channel III RECOMMENDATIONS • • A socio-economic assessment of fishing in relation to other non-fishing activities be initiated; I. • There is need for a series of systematic sampling to establish the effect of I seasonal changes on the fishery; • Further surveys needed to establish the optimal fishing effort for the fishery and initiate program for standardizing the number of gears per boat across the I landing sites; • Tycoon fishing should be discouraged as the method disrupts the breeding I process of mouth brooders especially the tilapiines. • There is need to assess the long-term impact of the newly introduced fishing practices of joining the nets (Mukira); • • There is need to establish the size at first maturity for C. gariepinus; • FIRRI research team should be facilitated and availed time to collect long­ • term catch effort data from the riparian districts of Kasese. Bushenyi and Kamwenge; • • Use of the minimum mesh size of 4.5" gill nets and minimum hook size of 9 be • encouraged to protect the major fish species. • • • ..• • • 52 .... I FiUl";.. ShIll.. ..un GHrg. I KuiD,. Cb.....1 I 5. References I Crespi & GD. Ardizzone, 1995. Fishery resources and some economic aspects of four fishing villages on lakes George and Edward in the Queen Elizabeth National I Park, Uganda. Afr. J. Trop. Hydrobiol. Fish 6, 11-20.

Dunn. I.G., 1989. Fisheries management study in the Queen Elizabeth National Park. I Uganda. With Annex. Technical assistance to Uganda Institute of ECOlogy I conservation of natural resources. Project No. 4100.037.42.44. 36 pp

Gwahaba, J.J., 1973. Effects of fishing on the Tilapia nilotica (Linne 1757) population I in Lake George, Uganda over the past 20 years. E. Afr. Wildl. J. 11: 317-328.

I Kamanyi JR, D. Mbabazi and Muhumuza E. 2000. Fisheries studies on Lake George and Kazinga Channel. A report prepared for Integrated Lake Management I Project (ILM) on Lake George and Kazinga Channel in Uganda.

I Ogutu-Ohwayo, R, JR. Kamanyi, S.B. Wandera, R. Amiina, F. Mugume., 1997. The fisheries and fish stocks of Lake George. Their productivity, exploitation, I management and conservation. A report prepared for: The Queen Elizabeth National Park fishing village conservation project, CARE International in I Uganda. I I I I I I I 53 I I ANNEXA. TERMS OF REFERENCE FOR FIRRI

Preliminary fisheries study on Lake George and Kazinga Channel

[I Background

The Integrated Lake Management (ILM) project aims to improve livelihoods of poor people in 1 lakeside communities by establishing a more integrated and participatory approach to the management of lake resources that cuts across existing administrative boundaries and I different natural resource sectors of government. The project works in eight fishing villages of Lake George and the Kazinga Channel to promote community-based co-management of the lake resources. In order to commence rational participatory planning there must be I accurate baseline information on the current state of lake resources. The last study of the fishery was undertaken by FIRRI in 1996/97, on the basis of which several management I recommendations were proposed.

I The purpose of this preliminary study of the fishery is to assess the appropriateness of some of the principal findings of FIRRI in 1997 and the subsequent management I recommendations. This includes a rapid investigation of, among other things, the nature and magnitude of catch rates, gill net mesh sizes used by fishermen, average length of dominant species and their size at maturity. A rapid boat count and frame (census) survey will provide I up-to-date information on the nature and scale of both the legal and illegal fishery. this information will be applied in PRA work undertaken as part of the lakewide participatory I planning process relating to use of aquatic resources.

I Specific tasks I The consultant will undertake the following specific tasks:

1. Be briefed on all aspects of the work by the ILM team leader and examine all documents I relevant to this study. 2. In collaboration with District Government fisheries staff, compile all relevant background I information and data on fish catch and fishing effort (legal and illegal) on Lake George and the Kazinga Channel, particularly for the most recent period, 1995-2000. I 3. Assist in the design and implementation of a rapid count of fishing boats on Lake George and the Kazinga Channel. The count will be made by using water and land transport. I I 54 • I FlsIInI" Slutll" .11 W, CHrif' 6 1UzhJif' CbiUJII,J

I 4. Assist in the design and implementation of a frame (census) survey of fishing activities on Lake George and the Kazinga Channel. 5. Complete preliminary studies on fishing activities to check the appropriateness of fishery I management recommendations made by FIRRI in 1997. This will involve sampling fishermen's catches at selected landings in each of three riparian districts to determine I parameters such as catch rates by gear and species, lengths of dominant species in catches, their lengths at first maturity, and mesh sizes of legal and illegal gill nets and hook sizes used in hook and line fisheries. 6. Compare the findings of this study (tasks No. 2-5 above) with those of 1997 and use this comparison as a basis to review and, if necessary and feasible, modify existing fishery management recommendations. 7. On the basis of the findings of this and previous FIRRI studies on Lake George and the Kazinga Channel, identify gaps in knowledge of the current status of the fishery and make recommendations on areas of future research needed to fill such gaps.

Outputs and Reporting

Outputs will comprise: • Originals or photocopies of all documents reviewed as part of the consultancy • One copy of the report, plus an electronic version in MS Word compatible format. The electronic version should be submitted as one file, with an automatically generated table of contents. If necessary, annexes should be saved in a separate file.

All correspondences and reports should be submitted directly to the Team Leader of the ILM Project.

Time and Duration

A total of fifteen (15) working days have been assigned to FIRRI for completing the assignment during November 2000. This includes time for data analysis preparing a report and, if necessary, modifying it following comments from the ILM project (Table 1)

The report will be submitted on or before 15 December 2000

55 Table 1. SChedule of Work

November: Date Activity I Monday 13 Arrive Kasese, meet ILMP =1 Tuesday 14 Catch samples 1 landing Wednesday 15 Catch samples 1 landing Thursday 16 Rapid lakewide boat count Friday 17 Catch samples 1 landing Saturday 18 Catch samples 1 landing Sunday 19 Analysis of sample data Monday 20 Frame Survey ~ Tuesday 21 Frame Survey ~c Wednesday 22 Frame Survey Thursday 23 Frame Survey Friday 24 Return FIRRI ~28,29 Analysis and report preparation

I

56 , CbUUJ~ W~ Fisb~drs StDdirs DD korgt "8uiJJg.

FORMAl CATCH AND EFFORT DATA SHEET FOR LAKE GEORGE AND KAZINGA CHANNEL

Date _ Name of landing _ Sub Counly, _ Districl _ SheeINo. ___

No. of fishing boats by type at landing site .

OTHERS TIME BOAT DETAILS GEAR DETAILS CATCH DETAILS Boal Ave. DIN Boat Prop. No. of Gear Operation Gear No. Species No. Weight Total length (TL) I Folk length (FL) (em) SIN days type crewl type Active! size (kg) fished! boat Passive week

I

Comments:

57 kolJ~ ~ga C1JaJJ~ LU~ FJsb~drs Siadirs tiD. ,.

FRAME SURVEY DATA SHEET: LAKE GEORGE, 2001 Sheet No .

Landing .. Sub-County...... District.. .. Date:

I~--~ SIN Boat Prop. No. of GEARTYPE & NUMBER crew. o type I I GILLNETS SIZES (Inches II Hl BS1 BS2 TR T Others <25 2.5 3 35 4 4.5 5 5.5 6 65 7 B 9 10 No. Size No. Size , 1 2 3 4 15 6 7 8 9 10 11 12 13 14 15 16 17 18 I 19 120

LL- Longline, BS- Boal seine, T-Tycoon (piercing & bealing), HL- Hook & line. TR- Traps, P- Parachute, S5-Scsse.

58 ------~--