STATDOC No. 22

REPORT ON THE FRAME SURVEY CONDUCTED IN THE UGANDAN PART OF L. VICTORIA
(3rd September – 20th December 1990).

by

UGA/87/007 BIOSTAT GROUP

edited by

Tumwebaze R. & E.J. Coenen

February, 1991

ABBREVIATIONS USED

1. INTRODUCTION

According to its Project document (1988), one of the objectives of the FAO/UNDP Project UGA/87/007 comprises the rehabilitation and improvement of a permanent fisheries monitoring system, inclusive of an improved Catch Assessment Survey on lake Victoria. In order to do this, recent and detailed information on fishing units and landings in the Ugandan part of lake Victoria is essential.

Several Frame Surveys were carried out for the Ugandan part of Lake Victoria in the early seventies: in 1970 by the Wild Life Services Ltd. (Wild Life Services Ltd., 1970), in 1971 by EAFFRO/FAO (Wetherall, 1972) and in 1972 by the Fisheries Department (Dhatemwa and Walker, 1972; Walker, 1972). The latter two surveys were aerial surveys involving counting of planked and dugout canoes. For the 1971 survey, Wetherall reports a total of 3264 fishing boats and 197 fish landings counted for the Ugandan portion of Lake Victoria.

From then onwards and up to 1988, baseline information concerning fishing factors was partly updated using time series and monthly or annual reports from the fisheries regional offices (Nyeko and Acere, 1990).

In 1988, the Planning Unit in MAIF carried out a nation-wide fisheries survey, inclusive of Lake Victoria. For the latter lake, only a total of 291 fishlandings and 3,470 canoes were identified, the latter of which 3,359 were planked and 111 were dugout canoes (MAIF, 1989).
However, Okaronon and Kamanyi (1989) considered the results of this Frame Survey not very reliable and therefore even recommended another FS for Lake Victoria “to rectify the controversy over the number of operating canoes”.

Also Project UGA/87/007 had serious doubts about the design, methodology used, operational aspects and especially the results (serious underestimates of the number of fish landings and canoes) obtained during this 1988 Frame Survey on Lake Victoria. In close collaboration with the Uganda Fisheries Department (UFD), it therefore carried out a detailed Frame Survey (FS) of fishing factors from 3rd September to 20th December 1990, covering the Ugandan part of Lake Victoria.

A preliminary analysis of its survey data, as described and discussed in this report, has provided a sampling frame for the setting up of an improved Catch Assessment Survey for this part of Lake Victoria (Malvestuto, 1990).

2. FRAME SURVEY METHODOLOGY AND PROCEDURES

2.1. Coverage

Spatial

Lake Victoria was divided into 5 operational regions. These were Ssese, Tororo, Jinja, Entebbe and Masaka, more or less corresponding to the limnological zones (differences in limnological characteristics result in differences in aquatic productivity, fish density, catch rates, etc.) established for the Ugandan part of Lake Victoria by G.E.B. Kitaka, EAFFRO limnologist, on the basis of extensive studies on the lake (Wetherall, 1972). These limnological zones were defined as follows :

Temporal

It was originally planned that the FS would be carried out simultaneously in the 5 regions within a short period (21 days) because of the fast changing dynamics in the fishery. However, this proved to be impossible due to several factors: loss of vital equipment (out-board engines) and because the project fibre glass boats, originally intended for deployment in the survey operations, were not yet ready.

Therefore, the survey had to be conducted in phases (survey boats and engines had to be passed on from one region to another) and finally lasted for about 4 months. Maximum care was however taken to avoid duplication or omission of fishing units and landings.

2.2. Frame Survey Questionnaire

Preparations for the planned Frame Survey started off with the design of the questionnaire early 1989, based on earlier work by Bazigos (Bazigos, 1974) but also incorporating a lot of questions on socio-economic factors of the fishing industry.

2.3. Testing the Questionnaire

The draft questionnaire was tested for its adequacy on lakes George and Edward (Nyeko, 1989) and for Lake Victoria around Entebbe; Kome Islands (Nyeko, 1990; Odongkara, 1990) and the Ssese Islands (Nyeko and Tumwebaze, 1990).

Adjustment and finalisation of the questionnaire, fraform A1, dealing with fish landing information, was done during Coppola's visit in July 1990 when the UGASTAT computer programme was partially developed. At the same time, fraform A2 was finalised for obtaining information on individual boat particulars.

A Frame Survey Manual (Annex I) was elaborated early August 1990 and was explained and distributed during the Frame Survey refresher course (16–17/8/90) to be used as a reference guide during the survey operations.

2.4. Operational aspects of the Frame Survey

Survey teams were selected by the Regional Fisheries Officers (RFO's) from the 5 lake regions and sent to the Project Headquarters Entebbe for the above mentioned training session. Each team consisted of two fisheries field staff and the RFO. One of the team members was supposed to be conversant with boat operation and outboard engine maintenance.
During the refresher course, each team was given topographic maps of the region to survey (scale 1:50,000). The RFO's made estimates for fuel, a provisional itinerary for the survey and also indicated refuelling stop-over points in survey areas. Each survey team was equipped with basic requirements like a boat, an out-board engine, fuel, Frame Survey forms (fraform A1 and A2, see Annex II), gumboots, lifejackets, clipboards, pencils, one pair of binoculars, polyethylene sheeting and a torch.

3. FRAME SURVEY IMPLEMENTATION

3.1. Operational procedures

The itinerary and timeschedule of the execution of the Frame Survey in the different areas and the names of fisheries staff involved are shown in Table 6.

The survey teams used a powered boat to move along the shoreline locating fish landings with the help of a pair of binoculars. At each located landing, fishermen or other respondents were interviewed. Each landing was awarded a code number on the questionnaire, corresponding to the map code number used on the topographic maps.

At each landing, fraform A1 for the landing characteristics was filled in and also fraform A2 for each individual boat. Due to time constraints or because some boats were not present at the time of the survey, some boat cards were sometimes left behind with the staff covering the landings in question. These cards were later returned to the RFO or sent directly to the Project Headquarters in Entebbe by the field staff. Where there were anticipated difficulties in retrieval of such forms, the team divided themselves into two : one interviewed the responsible person at the landing while the other collected information on all the fishing boats using the boat cards.

The regional surveys were co-ordinated by staff from the Statistics Unit at Entebbe Project Headquarters. Several inspection/verification visits were made to the survey areas during implementation of the FS. These visits made it also possible to supply and re-supply the survey teams with fuel and questionnaires and to maintain a link between the field and Project Headquarters. Moreover, through the monthly meetings at Headquarters, personnel involved with the planning and implementation of the on-going survey were able to meet and discuss operational aspects/constraints of the FS. All problems encountered were sorted out with the respective RFO's.

At the end of the survey in each region, both the Headquarters's co-ordinator and the Regional Fisheries Officer submitted a report on the Frame Survey conduct.

3.2. Operational constraints

The loss of vital equipment (out-board engines) and the non-availability of an adequate number of survey boats caused, as mentioned earlier, an unavoidable delay in the overall execution of the Frame Survey. There were time lags in the implementation of the survey in the different regions as it always involved transfer of equipment from one area to another.

When arriving at a landing to be surveyed, the team always had to explain in detail to possible interviewees the purpose and necessity of the Frame Survey because respondants initially connected the visit of fisheries staff with law enforcement operations.

All the survey teams, except for Jinja, received fuel and oil for the outboard engines right from Entebbe Project Headquarters. Whenever supplies ran out, the team in question sent a message to the Project in Entebbe or one of the team members went there physically to get more fuel. But even then, fuel could not always be obtained as the country, during the survey period, was struck by regular fuel shortages. Because of this fact and the other constraints mentioned above, the completion of the Frame Survey took longer than expected.

As the Survey progressed, the teams encountered less and less problems in filling in the questionnaires, although some difficulties to fill in certain questions lasted throughout the survey: information on the gears used and their numbers was in most cases totally dependant on the answers given by the persons interviewed because all or some of the gears were often left in the fishing grounds. And even when the team could physically inspect the gears (especially gillnets) in the boat at the time of landing, it was impossible to start checking the mesh sizes and counting all the nets. This was even worse for the illegal gears as the fishermen would naturally try to hide them.

The question about the year when the landing was established was in most cases a rough estimate because it was rare to find someone at the landing who knew the exact year when activities had started there, especially for landings which were established before the 1980's (Tumwebaze, 1990).

At most landings, there was no resident fisheries staff present and then the team had to fill in all the individual boat cards despite the fact that not all owners were present and not all boats were at the landing at the time of the visit. In such cases, it was very difficult to find out the name of the owner of the boat or to get information on the length of the boat, types of gear, gear sizes and numbers, etc. (Ikwaput, 1990).

4. FRAME SURVEY ANALYSIS

The FS forms were checked by the project's biologists/ statisticians as soon as they were returned to the Project headquarters in Entebbe. Because the UGASTAT computer programme for the entry and analysis of Frame and Catch Assessment Survey data was not yet functional, a partial analysis of the Lake Victoria FS data started late November 1990 during the visit of the consultant assigned to assist the project in the design of the Lake Victoria CAS. For this purpose, part of the FS data was analysed using Lotus 1-2-3 software.

Only a limited number of variables were selected for this analysis :

These variables were retrieved from the FS questionnaires originating from the five surveyed regions and entered into Lotus spreadsheets. The percentages of fishing, dugout, planked and powered canoes compared to the total number of active canoes for each landing were calculated. Totals and percentages for each region and for the whole of the lake were also obtained.

The analysis of all the FS data will be done when the UGASTAT computer programme is operational. For this purpose, a second visit of Mr. Coppola, FAO computer consultant, is expected for mid-April 1991.

5. RESULTS

The results of this preliminary FS analysis are presented in the form of figures/tables (Annex III) and maps (Annex IV).

All the surveyed landings, marked with code numbers on the topographic survey maps (scale 1 : 50,000), were replotted per region on regional maps (scale 1 : 750,000). This was done to get a clear picture of the distribution of the landings per region for the final CAS design. The location and number of fisheries staff is also indicated on these regional maps (see Annex VII).

Tables 2 to 6 show, per region, a listing of all landings surveyed together with their location, boat particulars and calculated percentages.

Table 7 gives a summary of the latter five tables showing the totals of selected variables for each each region and the totals for the whole of Lake Victoria (Ugandan part).
For the Ugandan part of Lake Victoria, a total of 715 fish landings were identified and a total of 8,674 active canoes counted.
The active canoes consist of 8,000 (92%) fishing canoes and 674 (8%) transport canoes. The composition of the fishing canoes by boat type is the following : 2,242 (26%) are dugouts and 6,432 (74%) are planked, the latter of which 1,250 (14%) are powered with outboard engines (dugouts are never powered).

Table 8 presents some characteristics for the four major landings at the time of the FS in the Ugandan part of Lake Victoria in respect of the highest number of 1) active canoes 2) transport canoes 3) fishing canoes 4) dugout canoes 5) planked canoes and 6) powered canoes :

• the biggest landing is Bwondha in Tororo region. It has the highest number of active canoes (127), the highest number of fishing canoes (122) and the highest number of planked canoes (117);

• Bulingu landing in Masaka region has the highest number of dugout canoes (100);

• Gaba landing in Entebbe region has the highest number of transport canoes (33);

• Masulya landing in Tororo region has the highest number of canoes with outboard engines (55).

Table 9 gives a comparison between percentages of some FS characteristics in the surveyed regions.

• Masaka region has the highest percentage (54.5%) of dugout canoes and they are mostly concentrated in the upper part of the region (90%) where there are sheltered bays and the lake is shallow.

• Ssese region has the lowest percentage (10.5%) of dugout canoes and this might be because of the open and deep waters around the islands where small canoes can hardly be used in strong waves.

• Ssese region also has the highest percentage of powered canoes (21%) because it is necessary to have big, strong and powered canoes for fishing in deep waters and for quick transportation of fish, people and other goods among and between the islands and the mainland.

• Jinja and Tororo regions have the lowest percentages of powered canoes, 10.2 % and 12.1 % respectively, despite the fact that big parts of these regions consist of islands. This might be because a substantial amount of fish caught around the islands of these regions is not transported to the mainland but is easily disposed of on water to transport canoes smuggling fish to Kenya. There they get better prices and also purchase cheap commodities which they bring back to Uganda to sell with profit (Reynolds & al., 1989; Reynolds & Ssali, 1990). According to reports of the RFO's from these regions, the amount of fish smuggled to Kenya would be substantial and could therefore explain the high catch statistics given for the Kenyan part of lake Victoria.

The size distribution for all the landings surveyed in terms of their number of active canoes, is presented in Figure 1. Nearly two thirds of the landings (451 or 63.1 %) possess 1 to 10 active canoes; 147 or 20.6 % have 11 to 20 canoes; 58 or 8.1 % have 21 to 30 canoes; 29 or 4.1 % have 31 to 40 canoes, etc. (see Fig. 1).
For the landings with less than 11 canoes, the percentages vary however for the different regions (Table 9). The highest percentage is found in the Ssese Islands (83 %). This might be because fishermen around these islands are highly mobile, using mostly planked and powered boats, moving easily from one landing place to another in search of better fishing grounds. The second highest percentage is in Entebbe region (65 %). Entebbe possesses the major market centres and is also better served with road transport. Fishermen can therefore easily dispose of their catch no matter where they land their catches.

The lowest percentage is found in Tororo Region (27 %). Most landings in Tororo are permanent : fishermen may change fishing grounds but mostly keep their same landing places.

6. DISCUSSION AND RECOMMENDATIONS

A comparison between the results of earlier Frame Surveys and the present one (see table below) for the Ugandan part of Lake Victoria shows several interesting evolutions.

A comparison between the 1971 aerial survey and the 1990 on-water FS, both surveys having done total physical counts of landings and canoes (as opposed to the 1988 FS), reveals a considerable increase in the number of fishlandings and fishing canoes over the last 19 years, namely with 263 and 145 percent, respectively. This would suggest that the nature of the Lake Victoria fishery in Uganda has evolved into a situation of increased scattering of the fishing operations into more dispersed and smaller landings: from an overall average of about 17 fishing canoes per landing in 1971, a landing now has only 11 canoes on the average.

Landings on the Ugandan part of lake Victoria are in most cases not gazetted and the fishing activities might become more and more dispersed in future. The Fisheries Department together with the relevant authorities, should therefore come up with a solution to stop the establishment of more and more landings.

Another phenomenon is the increase in the overall number of small dugout canoes over the last years, mainly in sheltered areas (bays and swamps) on the mainland. This might be due to the fact that in recent years, a number of people entered the lake Victoria fishery by investing a minimum amount of money for the purchase of cheap dugout canoes and a limited number of gill-nets (for the capture of Tilapia mainly).

Based on the present FS results, some interesting reflexions on past catch statistists can be made. Okaronon and Kamanyi (1989), based on the results of their sample CAS for Lake Victoria and taking into account a total number of 3,195 fishing canoes for the 1988 MAIF Frame Survey (although the MAIF, 1989 report quotes a total number of 3,470 fishing canoes, see also above), estimated a total catch of 60,600 tonnes for the Ugandan part of Lake Victoria in 1989. However, the Uganda Fisheries Department, based on an 1988 estimated number of 7,000 canoes, presented a 1989 total catch figure of 132,400 tonnes for the Ugandan part of Lake Victoria. Considering the present 1990 count of 8,000 fishing canoes for Lake Victoria, the 1989 UFD total catch estimate for Lake Victoria seems much more realistic than the figure of Okaronon and Kamani. The latter's total 1989 catch estimate for the whole of Uganda, namely 132,000 tonnes, corrected for the probable Lake Victoria underestimate of about 72,000 tonnes, would therefore amount to 204,000 tonnes which is very close to the UFD estimate of 211,000 tonnes for 1989.

The 1972 catch statistics for the Ugandan part of Lake Victoria varied between 24,000 tonnes (Wetherall, 1972) and 33,900 tonnes (MAIF, 1983) for a fishing effort of 3,264 fishing canoes and an estimated average catch of 25.8 kg per canoe per day. Since then, the Lake Victoria fishery has undergone quite some changes through the introduction of the Nile perch and the tremendous increase in fishing effort : the present total catch estimates fluctuate around 130,000 tonnes for a fishing effort of about 8,000 fishing canoes and an average catch of almost 44 kg per canoe per day.

The results of the analysis of the FS data for the Ugandan part of Lake Victoria, as presented in this report, represent only a partial analysis of all data collected. This was done during the visit of the consultant assisting the project in the design work for the Lake Victoria CAS. As can be seen from the Frame Survey questionnaires attached, a lot of data still remains unanalysed because the UGASTAT programme for the analysis of FS and CAS data is not yet fully operational. This can hopefully be redressed during the second visit of the computer consultant, planned for mid-April 1991.

Based on the analysis of the essential FS variables, a framework/design for the CAS of Lake Victoria was established (Malvestuto, 1990). Five major and 50 minor strata were defined and the new CAS, based on random sampling of landings and on time sampling schemes for the collection of catch statistics at the chosen landings every second, third or fourth day, will be implemented as from 1/3/1991.

For the new CAS, also an assessment of the fisheries staff deployment in the field was made. It was observed that, in general, the staff for the mainland minor strata were badly distributed or even insufficient in number. For the islands minor strata, the situation was even worse : very few islands have resident fisheries staff and, in some cases, staff who are in charge of certain islands make their supervisory visits from the mainland. Even then, most of their work often concentrates on law enforcement operations instead of extension work and fisheries statistics collection. Mutual trust between the fishing community and the fisheries staff has not yet been established. Fishermen still run away on seeing an approaching boat, particularly one from the Fisheries Department (Wadanya, 1990).

In order to make the new CAS feasible in all the 50 minor strata to sample, extra field fisheries staff was appointed by the UFD while other staff were posted elsewhere within the regions to obtain a optimum distribution for the CAS operations. It proves indeed to be very important to have adequate staffing in all the minor strata on the lake in order to achieve a viable system of CAS data collection.

Both the Jinja and Tororo survey teams observed smuggling of both Tilapia and Lates to Kenya. The Jinja survey team reported 30 boats which take fish across the border daily. Most of this smuggled fish is purchased on the water so that boats in the affected areas land very little or no catch at all. Since catch/effort data are collected at the fish landings, it is essential that all the fresh catch is landed before it is marketed. Therefore, the Fisheries Department should come up with immediate solutions to reinforce the control of the already existing regulation prohibiting the trade of fish on water and should inform and contact all relevant authorities to take a combined action against the smuggling of fish.

The importance of complete and regular fishery censuses or frame surveys as a basic frame-work for mounting all kinds of surveys for obtaining data on the characteristics and trends in a fishery as the one on Lake Victoria has not to be stressed again (Nyeko & Acere, 1990; Okaronon & al., 1990). The main specific recommendation regarding fisheries censuses for Lake Victoria during the FAO/UNDP Regional Workshop on Fisheries Statistics and Information Systems for Lake Victoria in Kampala (Uganda) from 26th to 29th of June 1990 was therefore that each of the three riparian countries (Kenya, Tanzania, Uganda) undertakes a census at least every three years and that the censuses would be on the basis of complete enumeration, to be updated every year through yearly sample surveys. The minimum survey items to be considered were recommended to include landing sites; fishermen; fishing craft by type and type of gear (Ssentongo, 1990). The next complete FS for the Ugandan portion of Lake Victoria should therefore be planned for 1993.

REFERENCES CITED

Bazigos, G.P. - 1974 - “The design of fisheries statistical surveys - Inland waters.” FAO Fish. Techn. Pap., (133): 122p.

Dhatemwa, C.M. & R.S. Walker - 1972 - “The use of aerial surveys for fisheries management in Uganda with reference to the 1972 aerial survey of fishing canoes.” Fisheries Department, Entebbe: 7p. (typed draft).

Ikwaput, J. - 1990 - “Report on the Frame Survey team from lake Victoria East Entebbe.” Fishin - UGA/87/007., Biostat Report 18, Biostat reports series: Volume 2, April 1991: 5p., 3 app., 2 ann.

Malvestuto, S.P. - 1990 - “Catch Assessment Survey Design for the Ugandan Waters of Lake Victoria.” Fishin - UGA/87/007: 18p. (draft)

MAIF - 1983 - “Blue Print for Fisheries Development in Uganda.” Ministry of Animal Industry and Fisheries, Fisheries Department : 16 p.

MAIF (Planning Department) - 1989 - “Fisheries Survey 1988.” MAIF/Plan.Unit UFD.

Nyeko, D. - 1989 - “Report on the Tour of South-western Lakes : George and Edward landings.” Fishin - UGA/87/007., Biostat Report 7, Biostat reports series: Volume 1, March 1991: 3p., 2 tabs.

Nyeko, D. - 1990 - “Report on the Trial Frame Survey conducted around the Northern Kome Islands Complex (14–16 February 1990).” Fishin - UGA/87/007., Biostat Report 9, Biostat report series: Volume 1, March 1991: 5p., 3 figs., 2 app.

Nyeko, D. & T.O. Acere - 1990 - “Inventory survey of fishing factors in lake Victoria Uganda.” Paper prepared for the FAO/UNDP Regional Workshop on Fisheries Statistics and Information Systems for Lake Victoria, Kampala, Uganda, 26–29/6/1990, FAO/UNDP/IFIP/ UGA/87/007: 7p.

Nyeko, D. & R. Tumwebaze - 1990 - “The Lulamba complex Fisheries: Report on the trial Frame Survey (10–12 April 1990).” Fishin - UGA/87/007., Biostat Report 11, Biostat report series: Volume 2, April 1991: 5p., 3 figs, 2 ann., 5 tabs.

Odongkara, O.K. - 1990 - “Socio-economic aspects of the Kome Island fisheries : report on a trial frame survey.” SEC Field Report No. 12, FISHIN Notes and Records, Fisheries Statistics and Information Systems, FAO/UNDP Project UGA/87/007, February 1990.

Okaronon, J. O. & J. Kamanyi - 1989 - “Catch Assesment Survey of Uganda waters.” AFRP/UFFRO Joint Fisheries Surveys Report No. 2, Sept. 1989: 10p., 3 app., 7 tabs.

Okaronon, J.O., S. Nkusi, E.J. Coenen & J. Ikwaput - 1990 - “Catch/effort Sampling System(s) in Uganda.” Paper prepared for the FAO/UNDP Regional Workshop on Fisheries Statistics and Information Systems for Lake Victoria, Kampala, Uganda, 26–29/6/1990, FAO/UNDP/IFIP/ UGA/87/007: 10 p., 1 ann.

Reynolds, J.E., J. Wadanya & D. Nyeko, eds. - 1989 - “Fisheries statistics and information management in Uganda : past approaches, current status and future prospects.” Field Document No. 1, FISHIN Notes and Records. Fisheries Statistics and Information Systems, FAO/UNDP Project UGA/87/007, November 1989 : 84.

Reynolds, J.E. & W.M. Ssali - 1990 - “Lake Victoria Fisheries Industralisation: Recent Developments in Uganda.” Socio-economic Field Report No. 13, FISHIN Notes and Records. Fisheries Statistics and Information Systems, FAO/UNDP Project UGA/87/007, March 1990 : 43.

Ssentongo G.W. - 1990 - “Travel Report and Aide-Memoire - Mission to Kampala (Uganda) and Nairobi (Kenya), 22 June - 6 July 1990.” IFIP/UNDP/FAO - RAF/87/099, TRAM/35, July 1990 : 7 p. (restr. distrib.).

Tumwebaze, R. - 1990 - “Report on Jinja Frame Survey.” Fishin - UGA/87/007., Biostat Report 20, Biostat report series: Volume 2, April 1991: 4p., 6 ann.

Wadanya, J. - 1990 - “Notes on the fisheries of Lake Victoria, Tororo region, made during the Frame Survey exercise.” Fishin - UGA/87/007., Biostat Report 15: Biostat report series: Volume 2, April 1991: 6p., 4 app., 1 tab., 1 ann.

Walker, R.S. - 1972 - “A statistical analysis of the aerial survey carried out by Uganda Fisheries Department, 6th June 1972 – 13th June 1972.” Fisheries Department, Entebbe, August 1972 (mimeo).

Wetherall, J.A. - 1972 - “Catch Assessment Survey (CAS) of Lake Victortia.” EAFFRO Occasional Paper No. 14: 57p.

Wildlife Services Ltd. - 1970 - “A census of fishing canoes in Uganda.” Wildlife Services Ltd.

ANNEX I. FRAME SURVEY 1990

GOVERNMENT OF UGANDA
MAIF
FISHERIES DEPT.

MANUAL FOR FIELD STAFF

1. INTRODUCTION

This manual is intended to be used by those field officers who will participate in the conduct of a frame survey on Lakes Victoria and Albert.

Participants from the six fisheries regions are to attend a refresher course on frame survey (16–17 August 1990) organised by the Project UGA/87/007.

It is hoped the manual will be invaluable as a reference guide during the conduct of the Frame Survey (FS).

2. BRIEF ON THE FRAME SURVEY

• Conducted periodically to obtain information on the magnitude and distribution of fishing factors (fish landings, Boats, gears, fishermen, facilities and amenities at fish landings etc.).

• This would be by interviewing respondents (responsible and knowledgeable persons) at all fish landings and where possible carry out gross inspection to verify accuracy of data.

The success of the Frame Survey (FS) depends on:

• ability of recorders to elicit acceptable responses.

• results obtained from the FS being uniform through out the survey area with respect to specified definitions.

• those selected are assumed capable of having full understanding of the survey, particularly the questionnaire format, procedures to be followed and uniformity in execution.

It is therefore the responsibility of the survey supervisors and all concerned with the survey conduct to ensure that the above guidelines are adhered to. This manual presents hints and guide lines that will be used to carry out the FS.

3. GENERAL

The FS has two forms, entitled QUESTIONNAIRE FOR FRAME SURVEY (FRAFORM.A1) and INDIVIDUAL BOAT PARTICULARS (FRAFORM.A2).

FRAFORM.A1

• This is filled for each fish landing visited by interviewing one or more knowledgeable respondents. An example of a completed form is attached.

• It is important that question (7) is cross checked by gross inspection and own judgement reached by way of probing. Number of Active boats must equal combined totals for Planked Powered canoes + Non Powered + Dug-out.

• Boats that are used for fishing only may not in any way be used for fish transport (FISHING ONLY).

• Boats may sometimes be used for fishing as well as transport of the fresh catch to a major landing market (FISHING/TRANSP.).

• Some boats are used only for transport (TRANSP. ONLY) and may carry fish (FISH TRANSP.) or other goods/passengers. (NON FISH. TRANSP.).

• Question (13) relationate boat types with letters P, D, F, T & N standing for Planked, Dug-out, Fibreglass, Trawler and Not classified boats, respectively.

FRAFORM.A2

• This is basically a boat card. It is to be filled in for each active boat at the visited landing during the survey period.

• Smaller landings with few boats and no resident fisheries staff are filled on the survey visit-day.

• The forms must be entrusted to a resident fisheries staff where it may not be feasible to fill in particulars for every boat at the time of the survey. It must be submitted to the RFO within one month from the date of visit. It must be collected by the survey team by the end of the survey period.

• If the boat owner is a female, clearly indicate it with (F) against the name. e.g. Mary Gimbo (F).

4. DEFINITIONS OF SURVEY ITEMS

In the questionnaire form (FRAFORM.A1) provided, you are expected to fill in all that is required of a respondent. This should be identified according to the following agreed definitions.

Fish landing

• A place on the shore where fresh catches are landed. It may have smaller satellite places where catches are also landed, still representing the same landing, on the condition that any fisheries officer surveying that landing is able to adequately sample all incoming boats.

Active boat

• One which is either operating or is expected to resume operation after a temporary ‘rest’ (loss of nets, disrepair of boats, non use of transport boats due to inadequate catches etc.)

Disused boat

• one which is abandoned or un-serviceable and not able to resume operation.

5. PROCEDURE

• Using a powered boat, you are required to cover all fish landings along the assigned shoreline.

• Plan the the overall operation beforehand and adjust daily depending on experience gained from the previous survey day.

• Contact the fisheries staff covering that part of the shoreline. He should be able to accompany the team for that particular stretch and assist in introducing them to local officials and help locate possible hidden landings.

• In order to detect hidden landings one member of the team must visually cover the shoreline using the field binoculars provided.

• Another member must mark the position of the spotted or visited landing precisely on a map (scale 1:50000) and give it a code. The same code should be entered in the survey form for that fish landing.

• Once the subject matter has been introduced and potential respondents chosen, the team should split into two : One member executing FRAFORM.A1 and another FRAFORM.A2.

6. HINTS

• Certain landings where the fishermen's dwellings are far away from the lake shore may present a problem of finding a respondent. Information on such landings may be gathered from knowledgeable fishermen in nearby landings.

• Do not leave any form unfilled because of non response/no respondent. You are expected to revisit any such landing in case the first attempt failed.

• Make it a point to establish an atmosphere of trust with the respondents in particular and the fishing community in general.

• Greet and briefly explain the purpose of your visit. Assure them that informations gathered will be utilised for the betterment of their community. Information given by individual units will be in absolute confidence.

• Ask exactly the questions on the questionnaire with wording and order as stated therein. This would reduce response errors.

• Try as much as is reasonable to probe further when you gauge the respondent may be giving incorrect answers. Where possible verify by a rough gross inspection or cross examination. Differential probing within survey team members and amongst team members of the different regions is a major source of error introducing a bias in the results of a FS.

• When you are satisfied with the response, record exactly as you were told.

• Write legibly in pencil all your elicited responses.

• Avoid cheating as this may be cross checked in a later Coverage Check Survey.

7. SUPERVISION

Supervisors should ensure that :

• payment of those on survey is effected in good time during the survey period.

• Simplify field movements of the team by ensuring petrol and engine oil are in sufficient quantity for the survey period.

• queries arising from time to time during the survey period are solved.

• team members actually made all the interviews claimed and that they are asking questions, interpreting and recording as instructed.

• team members do select the survey factors according to established definitions and all forms are completely filled.

8. SURVEY EQUIPMENTS

• Care must be taken to ensure proper maintenance and security of the survey equipment (canoe, outboard engine, binoculars).

Outboard Engines

• Consult the Operator's Manual regularly for proper engine operation.

• It is essential that the Break-In Procedure (Page H-5 in Owner's Handbook) must be strictly adhered to during initial operation.

Others

Provided are Binoculars, Boats, Topographical Maps, Staff list etc. all of which should be well cared for and maintained.

ANNEX II. FRAME SURVEY FORMS

* TRANSP. ONLY = FISH. TRANSP.                      + NON FISH. TRANSP.                     

ACTIVITY OF OWNER						1 = Part of Crew	2 = Supervisor
						3 = Other
NUMBER OF CREW						LENGTH OF BOAT (m)
TYPE OF BOAT :				1 = Planked		2 = Dug-out	3 = Fibreglass
				4 = Trawler		5 = Not Classified.
ENGINE (Y/N)				IF YES, WHICH HORSE POWER :                              HP
ACTIVITY				1 = FISHING ONLY			2 = TRANSPORT ONLY
				3 = FISHING and TRANSPORT
GEAR TYPE(S) USED
1 = GILL NETS	MESH SIZE				NO.
2. = HOOKLINES			SIZE		NO.
3. = CAST NETS
4. = BEACH SEINES
5. = TRAPS
6. = TRAWL NETS
9. = NOT CLASSIFIED.
REMARKS: