SWIOP/WP/18 - Results of the First Year of Implementation of the Kenyan Catch Assessment Survey

Table of Contents

By

G. CARRARA (SWIOP)
S.R. COPPOLA (FIDI)

MAY 1985
RAF/79/065/WP/18/85/E

REGIONAL PROJECT FOR THE DEVELOPMENT & MANAGEMENT OF FISHERIES IN THE SOUTHWEST INDIAN OCEAN

PROJET REGIONAL POUR LE DEVELOPPEMENT ET L'AMENAGEMENT DES PECHES DANS L'OCEAN INDIEN SUD-OCCIDENTAL

c/o UNITY HOUSE P.O. BOX 487, VICTORIA, MAHE SEYCHELLES

TELEPHONE 23773
TELEX 2254 SWIOP SZ

This electronic document has been scanned using optical character recognition (OCR) software and careful manual recorrection. Even if the quality of digitalisation is high, the FAO declines all responsibility for any discrepancies that may exist between the present document and its original printed version.

Table of Contents

1. INTRODUCTION

1.1. Estimates
1.2 Quality of the Collected Data

2. METHODOLOGY

2.1. Sample Survey Design
2.2. Selection Process
2.3. Selection of Sample Days
2.4. Estimating Process
2.5. The Estimations

3. NOTE ON THE IMPLEMENTATION OF THE CAS

3.1. Regional Reports

3.1.1. Lamu Stratum
3.1.2. Malindi Stratum
3.1.3. Mombasa Stratum
3.1.4. Kwale Stratum

4. COMPUTERISED DATA PROCESSING

5. DISCUSSION ON THE CATCH ASSESSMENT SURVEY

5.1. Stratum 01: Lamu
5.2. Stratum 02: Malindi
5.3. Stratum 03: Mombasa
5.4. Stratum 04: Kwale

6. CRITICAL ANALYSIS OF THE RESULTS

7. COMPARISON BETWEEN THE CATCH ASSESSMENT SURVEY AND NATIONAL STATISTICS

7.1. Comparison by Stratun and by Month of Final Results

8. CONCLUSIONS AND RECOMMENDATIONS

8.1. Source of Error
8.2. Recommendations

APPENDIX A - CAS MONTHLY ESTIMATES BY GEAR AND BY STRATUM

Monthly results by stratum and by gear
Graphs of monthly total catch estimates by strata
Graphs of monthly total catches by gear and by strata

APPENDIX B

KENYA ARTISANAL FISHERY SAMPLING FRAME
BIG GEAR TABLE

APPENDIX C - LIST OF FORMS

CAS-KAF 2.1 - Monthly Inventory Characteristics
CAS-KAF 2.2 - Daily Catch Report
CAS-KAF 2.3 - Monthly Catch Record
CAS-KAF 2.4B - Estimation Process
CAS-KAF 2.5 - Total Monthly Catch Estimates by fishing methods

APPENDIX D - METHODOLOGICAL NOTES ON THE IMPLEMENTATION OF THE CAS

APPENDIX E - AREA DISTRIBUTION

APPENDIX F - STAFF SUPERVISORY

APPENDIX G - CATCH COMPOSITION IN CATCH ASSESSMENT SURVEY

Percentage species in the catch of January 1984 - February 1985
Graph of monthly distribution of important species and species groups
Monthly distribution of catches of Lobster
Percentages species in catches of stratum

1. INTRODUCTION

1.1. Estimates
1.2 Quality of the Collected Data

This report has been elaborated a year after implementation of the newly-introduced catch and effort assessment survey for the artisanal sector of the Marine fishery in Kenya.

It was launched on line with the proposed Statistical Survey Programme to revise the National Fishery Statistical Survey System and represents the conclusion of the first phase of such a programme.*

* See reference 1

The programme was carried out as a joint effort of various offices, such as the Kenyan Department of Fisheries, the UNDP/FAO OFFSHORE TRAWLING SURVEY Project (KEN 74/023), the South West Indian Ocean Programme (SWIOP) and the FAO Department of Fisheries Headquarters (FIDI) Rome.

Although its implementation suffered various financial and logestic constraints, completion of this phase should be considered as a positive contribution to the improvement of the level of the statistical system in the Country.

Also to process the survey data, microcomputer based software was developed and introduced in this field for the first time by the Department of Fishery.

1. RESULTS OF THE CATCH ASSESSMENT SURVEY

1.1. Estimates

The Catch Assessment Survey (CAS) was designed and implemented to provide monthly catch estimates by gear, by species and by region for the marine artisanal fishery.

The yearly total catch for this sector, as estimated through the CAS 1984 is 9862.476 metric tonnes. This figure was obtained by a fishing industry of about 1,828 fishing boats and fishermen. (see Table 1.1.1 and 1.1.2)*

* See references 2/3

In table 1.1.1 the monthly catch estimates are presented by gear used for the country as a whole (marine waters), whereas regional estimates by gear used and the supporting graphs are given in appendix A.

The relative importance of each species group in the catches is presented in appendix G.

1.2 Quality of the Collected Data

1.2. Quality of the collected data Due to financial and technical problems the data collection systems in some sample fishing sites was in some cases discontinued and in others was started with some delay.

Also, many cases of no reports and late reports of the monthly submission of the fish catch record from the periphery was found. These irregular submissions of the reports caused some inconvenience and delay in the data processing at the central office and in decision making to monitor the survey. Specifically, in stratum 02 (Malindi district) and stratum 04 (Kwale district) the catch assessment survey was initiated with three months delay (April 1984). In stratum 01 (Lamu district) apart from the chronic difficulty to control the fishery and the working behaviour of the Baugue fishermen working in Kenya and Somalia simultaneously, two/three sample fishing sites, because of local problems, were abandoned by the fishermen community; unfortunately the situation was not brought to the attention of the survey engineer (FAO) nor the survey coordinator in Mombasa and therefore no replacements were introduced.

Stratum 04 (Kwale district) suffered from other kind of event which occurred in Vanga, the southern most fishing village of the coast, adjacent to the Tanzanian border. In fact with the opening of the border between Kenya and Tanzania (March/April 1984) a great number of unaccounted Tanzanian fishermen must have moored into Vanga. Tanzanian fishermen are known for their expertise, especially with beach seine nets, and this has created some inbalance in the statistics collected on this particular gear.

This discrepancy was detected only during the data processing phase at the end of the survey. The two latter cases together with some other factors have once more demonstrated the vital importance of a qualified field supervisor.

2. METHODOLOGY

2.1. Sample Survey Design
2.2. Selection Process
2.3. Selection of Sample Days
2.4. Estimating Process
2.5. The Estimations

2.1. Sample Survey Design

The sampling method used for the catch Assessment Survey 1984, can be described as sampling in space and time.

a. Sampling in space

As far as the space is concerned, the stratification used in the Frame Survey 1982/83 was maintained.*

* see reference 2/3;

The geographical/administrative units of the strata into which the coast was divided are given in figures 1 through 7 in Appendix E.

Within each stratum a number of fishing sites (Primary Sampling Units) were selected for further sub sampling (secondary sampling units). The Primary Sampling Unit (PSU) is defined as a permanent fishing village where fish landings take place regularly.

The Secondary Sampling Unit (SSU) is defined as the Fishing Economic Unit which is aggregation of fishing boats, fishing gear and fishermen to carry out fishing operations.

For sampling purposes the fishing Economic Units within the selected fishing sites are grouped into classes by taking into account types of boat and gear used.

TABLE 1.1.1 - 1984 CAS RESULTS - KENYA ARTISANAL FISHERIES: ALL KINDS OF BOAT. (UNITS IN KGS)

MONTH	CAST NET	GILL NET	BEACH SEINE	HAND LINE	OTHER GEAR	LOBSTER	TOTAL
JANUARY	119631	236640	153740	183823	182399	5467	881700
FEBRUARY	112988	238142	82004	203409	157727	8555	802825
MARCH	109619	293794	173163	175283	133876	17150	902885
APRIL	118920	166433	139867	181229	111848	5816	724113
MAY	124365	247809	90672	163482	212207	5558	844093
JUNE	102549	219548	100234	150476	118278	5419	696504
JULY	114601	211737	86698	159978	148697	13616	375327
AUGUST	83102	156500	117976	159705	108869	21449	647601
SEPTEMBER	93056	193823	152567	187117	146214	8184	780961
OCTOBER	94941	281756	263522	219997	164723	18950	1043889
NOVEMBER	66386	189465	137009	211315	200000	8134	812309
DECEMBER	96473	253927	320886	165920	144392	8671	990269
TOTALS	1236631	2689574	1818338	2161734	1829230	126969	9862476

TABLE 1.1.2 - NUMBER OF FISHING BOATS PER GEAR PER STRATUM

	Gear
Stratum	CAST NET	GILL NET	BEACH SEINE	HAND LINE	OTHER GEARS	LOBSTER	TOTAL
01 LAMU	51	164	33	111	37	112	508
02 MALINDI	34	120	16	127	58	6	361*
03 MOMBASA	94	42	5	121	139	-	401
04 KWALE	49	78	20	155	235	21	558
TOTAL	228	404	84	514	469	139	1828

* During the implementation of the CAS 16 Beach Seining Boats and 6 Lobster Fishing Boats not appearing in the Sampling Frame were found in Stratum 02 Malindi.

A random sample of landings (or the totality in the case of a limited number) is selected from each category of FEUs for objective measurement of the surveyed characteristics.

b. Sampling in time

As far as time is concerned, the Catch Assessment Survey used a flexible reference period (normally 10 days per month) to be determined from time to time according to the presence and importance of the boat/gear sub-population concerned.

The obtained sample data within the surveyed period are used to provide estimates on a monthly basis by gear/boat and by species. Annual estimates are calculated by adding up the monthly estimates.

2.2. Selection Process

The CAS is designed as a sample survey with two stage sampling as follows:

First stage Sampling Units (PSU) of landing sites
Second stage Sampling Units (SSU) of boat landings

At the first stage, sample landing sites have been selected from all landing sites within each administrative Divisions (strata).

In table 2.2.1 the sample of the Primary Sampling Units is linked together with the appropriate sampling fraction in boat number.

In order to give an approximate idea of the variability of the operational pattern of the fishing units within the Selected PSUs, in table 2.2.2 the number of operating boats per month is given. Primary Sampling Units were selected through the method of sampling with probability proportioned to size (PPS method).

However due to some human and technical limitations, some constraints were imposed to this selection. In other words a selected fishing site where there was no resident recorder was withdrawn.

At the second stage of selection* the recorders were instructed to use one of the two allowed alternatives given to them: to use census approach or sampling approach.

* see appendix D

The census approach had to be used in the case of a limited number of landings in the sample day (up to 10 landings), whereas a sampling approach had to be used for a larger number of landings.

In the latter case, the method of selection adopted was systematic sampling where the starting point and the space interval determine the whole sample.

TABLE 2.2.1 - Inventory of Artisanal Fishing Units: The Sample of Primary Sampling Units (PSU's).

STRATUM		Area stratification - serial number	Name of the PSU	Sample number	Probability of selection	Remarks
1. LAMU		1210	KIUNGA	1.1	21/508
		1330	KIZINGITINI	1.2	88/508
		1255	MOTANDONI	1.3	25/508
		1365	LAMU	1.4	57/508
		1277	NDUNUMI	1.5	11/508
		1385	KIPINI	1.6	22/508
	Reserve	1251	SHANGA	R1.	13/508
2. MALINDI		2310	NGOMENI	2.1	37/339
		2225	MALINDI	2.2	22/339
		2230	WATAMU	2.3	30/339
		2245	KILIFI	2.4	36/339
	Reserve	2160	KANAMOI	R2.	5/339
3. MOMBASA		3220	BAMBURI	3.1	35/401
		3225	OLD PORT	3.2	27/401
		3245	PORT REITZ	3.3	68/401
	Reserve	3115	UTANGE	R3.	23/401
4. KWALE		4105	TIWI	4.1	25/558
		4115	KINODO	4.2	40/558
		4255	SHIMONI	4.3	35/558
		4170	CHWAKA-KIBUJUNI	4.4	49/558
		4285	VANGA	4.5	76/558
	Reserve	4125	MSAMBWENI	R4.	29/558

2.3. Selection of Sample Days*

* see Appendix D

At the beginning of each month, the fish recorder provided on the "Monthly Inventory Characteristics" form (CAS-KAF 2.1) the following information on the structure of the fishery in the PSU where he was stationed.

- total number of boats and by type
- number of boats by gear

The supervisor then indicated on the same form the number of consecutive days the recorder had to sample each gear type; this period of time is called the "Reference Period" as earlier stated.

Following the supervisor's instructions, the fish recorders sampled landings throughout the month. To record daily catch by species and effort in fishing hours, the recorders used Form CAS-KAF/2.1. Also recorded was the total number of boats landing within the gear type being sampled and the number of boats sampled each day. The ratio between these two values (R1) is used to raise the value of the sampled catch to the total catch on that day for that gear type.

2.4. Estimating Process

At the end of the month the supervisor transferred sampling data of each PSU to Form CAS-KAF/2.3 and summed them up for each gear type. The sample totals for each gear type were then raised to reflect the number of days fished in the month by multiplying them by the ratio of days fishing to days sampled in the reference period (R2).

In this manner, the monthly total estimates of catch and effort by gear type for each PSU were obtained. In each stratum the figures for each PSU were transferred to Form CAS-KAF/2.4B. Monthly total stratum estimates of catch and effort by gear type were then obtained by raising the PSU totals to stratum totals. The raising factor used here was the ratio for each gear type of boats in the stratum to boats in the PSUs (R3). Each PSU therefore, produced a different estimate of effort and production of any given gear type within the stratum. A mean was extracted (the best value) from these values, and taken as the stratum estimate. As a final step, in order to obtain estimates for the whole coast, the stratum estimates of production and effort were transferred to Form CAS-KAF/2.5 where they were summed.

2.5. The Estimations

Detailed instructions on the steps to be taken in completing the forms mentioned above are given in Appendix D. The same process expressed in a mathematical way will be as follows:

assuming that,

h = stratum (1, 2, 3, 4)
i = selected PSU
j = boat sampled
k = gear used
M = number of days in the calendar month
D = number of days in the reference period
d = number of actual fishing days sampled
n = number of boat in the PSU
l = number of boat sampled
N = number of boat in the stratum
Y = catch (effort)
S = number of sampled village in the h stratum
L = number of boats landed

Gives the total catch (sample date) landed in PSU "i" in stratum "h" by boats using gear "k" during the days sampled.

(monthly)

Gives the monthly total catch (sample date) landed in PSU "i" in stratum "h" by boats using gear "k" during the month.

(stratum)

Gives the monthly total catch landed in stratum "h" by boats using gear "k" estimated using sample data from PSU "i". The final estimated monthly total catch landed in stratum "h" by boats using gear "k" is obtained by taking the average of the different stratum estimates calculated form the sample date of the PSUs weighted by the number of landings in each PSU.

3. NOTE ON THE IMPLEMENTATION OF THE CAS

3.1. Regional Reports

(Field Work)

The Catch Assessment Survey was implemented by personnel from the Kenyan Department of Fishery, Coastal Province Headquarters. [In appendix G the list of staff supervisors is given].

Data collection started in January 1984 in strata 01 and 03 and in April of the same year in strata 03 and 04. Table 3.1 shows the sampling efficiency and the number of boats operating from each PSU each month.

3.1. Regional Reports

3.1.1. Lamu Stratum
3.1.2. Malindi Stratum
3.1.3. Mombasa Stratum
3.1.4. Kwale Stratum

3.1.1. Lamu Stratum

From January to December 1984 only 40% of the available PSUs had been correctly sampled in each month.

Very poor supervision of the fish recorders resulted in entire sets of forms wrongly completed. These have not been processed. This situation is mostly due to transport problems and lack of personnel. Action has now been taken to improve supervision by assigning two Fisheries Assistants to Lamu stratum for the field supervision of the fish recorders.

Personnel

The personnel involved in the survey were:-

1 Fisheries Officer in Lamu (part time)
2 Fisheries Assistants in Lamu
1 Fisheries Assistant in Kipini
1 Scout in Kiunga
1 Scout in Kizingitini
1 Scout in Lamu
1 Scout in Matondoni
1 Scout in Shanga (Reserve PSU)

Transport

Lamu stratum has, as means of transport, two landrovers, one inboard engined boat and one speed boat.

Distances from the Lamu stratum to the different PSUs are as follow:

Kiunga:	169 km by road
Kizingitini:	4 hours with the inboard boat
Kipini:	86 km by road.

The fishing village of Matondoni was discarded as a sampling unit due to difficult communications. In the future the village of Faza could be a substitute PSU.

To enable a constant monthly supervision of the Fish Scouts in the PSUs in Lamu Stratum, funds have to be allocated for transport expenses and perdiem allowances.

Table 3.1 IMPLEMENTATION OF THE CATCH ASSESSMENT SURVEY:

Yearly Transport Costs

Gasoline and oil for landrovers:	KShs. 13,000/-
Diesel fuel for the inboard boat:	KShs. 2,400/-
Perdiem allowances for the supervisors:	KShs. 4,800/-
TOTAL:	KShs. 20,200/-

Equipment

During the survey it was noted that some equipment is needed to facilitate the work in the field and in the offices. Dial spring scales are actually used to weigh the fish in the sampling villages. Many of these weighing scales will have to be replaced wither with less expensive and more precise beam balances or with other dial spring scales at a cost of around 4,800/- KShs. a piece. Simple calculators with a print-out would also be useful for each stratum office to speed up calculations and reduce errors.

Equipment costs
6 Spring Dial Weighing Scales:	KShs. 28,800/-*
1 Calculator:	KShs. 1,250/-
Other Stationery:	KShs. 2,000/-
TOTAL:	KShs. 32,050/-

*Enquiries have to be made on the possibility of purchasing beam balances.

3.1.2. Malindi Stratum

From April 1984 to February 1985, 80% of the PSUs have been covered each month. Several mistakes were noted, particularly, in recording fishing hours and in catch data for specific cases such as a boat returning without any catch at all from a fishing trip.

As in the other strata, increased supervision of the recorders would improve the reliability of the estimates.

Personnel

1 Fisheries Officer in Malindi
1 Fish Scout in Malindi
1 Fisheries Assistant in Ngomani
1 Fish Scout in Watamu
1 Fisheries Assistant in Kilifi

Correct supervision would require assignment of a Fisheries Assistant to this specific task.

Transport

Malindi stratum has one landrover. A monthly visit to the PSUs could be organised if enough petrol was available. Distance to the PSUs are as follow:

Ngomeni:	30 kms
Kilifi:	60 kms
Watamu:	20 kms

Yearly Transport Costs

Gasoline and oil for landrovers: KShs. 5,200/-

Equipment

Equipment costs
6 Dial Weighing Scales:	KShs. 28,000/-
1 Calculator:	KShs. 1,250/-
Other Stationery:	KShs. 2,000/-
TOTAL	KShs. 32,050/-

3.1.3. Mombasa Stratum

From January 1984 to February 1985 95% of the PSUs were sampled each month. In future the survey of the Utange PSU, which was supposed to be a reserve PSU, will have to be sampled each month since it is the only PSU in stratum providing data on beach seine fishing. Several mistakes have been make in completing Forms CAS-KAF 2.1 and 2.2.

Personnel

1 Fisheries Officer (responsible for marine fisheries statistics) in Mombasa
4 Fisheries Assistants in Mombasa
2 Fish Scouts in Bamburi
2 Fish Scouts in Oldport
2 Fish Scouts in Port Reitz
2 Fish Scouts in Utange

Transport

Yearly transport costs for the supervision of the Fish Scouts and travelling expenses for the Officer in charge of the survey to Lamu, Malindi and Shimoni (Kwale stratum) were:

Yearly transport costs
Gasoline and oil for landrovers:	KShs 2,400/-
Per diem and travelling expenses:	KShs 3,500/-
TOTAL:	KShs 5,900/-

In Mombasa processing of the data collected in the 4 strata was performed using the Microcomputer of the Kenya Marine Fisheries Institute. The use of the computer was arranged with the Institute on the basis of one week per month.

Computer Expenses

At least three computer disks per month were needed for data storage using Visicalc software.

Yearly costs for the disks:	KShs 4,500/-
5 dial spring weighing scales:	KShs 24,800/-
1 calculator:	KShs 1,250/-
Stationery and others:	KShs 2,000/-
TOTAL:	KShs 32,550/-

3.1.4. Kwale Stratum

The main office of the Fisheries Department in this stratum is in Shimoni. From April 1984 to February 1985 76% of PSUs were sampled per month. Errors in data recording were made in this stratum also and more supervision is required in future.

Personnel

Shimoni stratum did not have a person working full time as supervisor.

Personnel involved in the survey were:

1 Fisheries Officer in Shimoni (part time)
1 Fisheries Assistant in Shimoni (part tine)
1 Fish Scout in Tiavi
1 Fish Scout in Kimodo
1 Fish Scout in Shimoni
1 Fish Scout in Chwaka-Kibujumi
1 Fish Scout in Vanga

Transport

Shimoni stratum has 1 landrover. Distance to the PSUs from Shimoni are as follows:

Tiwi:	60 kms
Kimodo:	40 kms
Chwaka:	15 kms
Vanga:	70 kms

Yearly transport costs:

Gasoline and oil for landrover: KShs 1,000/-

Equipment

The Fish Scouts in Kwale Stratum used weighing scales belonging to fishing cooperatives to weigh the fish. It is important that fish recorders have their own weighing scales since they must be able to record landings directly on the beach.

Equipment expenses

6 Weighing Scales:	KShs 28,000/-
1 Calculator:	KShs 1,2500/-
Stationery and others:	KShs 2,000/-
TOTAL:	KShs 31,250/-

4. COMPUTERISED DATA PROCESSING

Visicalc III software has been used to process the CAS data with an Apple III microcomputer. Visicalc is an electronic worksheet organised as a grid of columns and rows, each intersection of the columns and rows defines an entry position. In each entry position it is possible to enter an alphabetic title, a number or a formula relating to other entry positions. Two Visicalc forms have been designed to process the CAS data. The first form corresponds to the form CAS-KAF/2.3 of the manual processing system and takes input from the daily forms (CAS-KAF/2.1) and the monthly form (CAS-KAF 2.2). The second Visicalc form corresponds to the manual form CAS/KAF 2.4B and takes input from the previous Visicalc form and the monthly 2.2 form. The output from the second Visicalc form are monthly stratum estimates of catch and effort for each fishing gear.

5. DISCUSSION ON THE CATCH ASSESSMENT SURVEY

5.1. Stratum 01: Lamu
5.2. Stratum 02: Malindi
5.3. Stratum 03: Mombasa
5.4. Stratum 04: Kwale

5.1. Stratum 01: Lamu

The graphs of the monthly catch estimates are presented in appendix A figures 1-26.

From January to December 1984 an estimated total of 2,867 metric tonnes of fish were caught in Lamu stratum. This represents 29% of the total estimated annual landings of the marine artisanal fisheries. The most important gears, in relation to landings, are gill nets (37.1% of the catches), beach seines (24.2%) and cast nets (17.1%). Lobster, although representing only 4% of the total landings in the stratum, is very important because of its high value. 91% of the estimated total landings of lobster on the Kenya coast are from Lamu stratum. Figure 1 shows the trend of the estimated total catch per month in the stratum.

On average, catches were higher in the period September to March. The graph of monthly catch by hand line fishing boats show a very marked saw pattern.

5.2. Stratum 02: Malindi

Total catches in the stratum are estimated to be 1,856 metric tonnes. This represents 18.8% of total yearly landings from the marine artisanal coastal Fishery. The predominant gears in the stratum are gill nets with 53% and hand line 24% of the landings.

The trend of total landings follows mainly that of gill net catches (fig. 2 & 12). As for stratum 01 (Lamu), hand line catches show a saw toothed pattern.

5.3. Stratum 03: Mombasa

Sample data for the Monbasa stratum are available for most gears from January 1984 to February 1985. For the twelve months of 1984, total catches by all gears are estimated at 2,142 metric tonnes. This is 21.7% of the yearly estimated total catch for the artisanal fisheries in Kenya. The most important gears are the miscellaneous group "other gears" (32.2% of the stratum catches) and hand line (31.7%). The graph of total landings by month shows low catches in April and November and the highest catch in May (see figure 3). The "other gears" group is very anomalous with two very different gears mainly represented: fixed tidal traps and basket traps. The graph of hand line catches shows clearly defined period of good catches from April to June and September to October, low catches are estimated for the month of July.

5.4. Stratum 04: Kwale

Sample data are available for most gears from April 1984 to February 1985. To evaluate landings for all gears during a 12 month period, as in stratum 02, the first quarter catches have been estimated by interpolation, or by using regression methods by comparison with the estimates collected throughout the current national statistical system. The gear groups giving the bulk of the catches are the beach seine, handlines and the miscellaneous "other gears" respectively 29.5%, 24.9% and 24.2% of the total landings. The total catch for the 12 month period analysed was 2,998 metric tonnes, equivalent to 30.4% of the artisanal marine fisheries catch for the Kenya coast. The graph of total landings shows peaks in April and November - December and relatively low catches in August (see figure 4).

6. CRITICAL ANALYSIS OF THE RESULTS

As discussed in the Survey Revision plan, analysis of the results should be made in two phases:

i) Estimation of the table landings by quarter and by year (Estimate of Sampling errors etc.).

ii) Revision of the existing fisheries statistical.

7. COMPARISON BETWEEN THE CATCH ASSESSMENT SURVEY AND NATIONAL STATISTICS

7.1. Comparison by Stratun and by Month of Final Results

SURVEY RESULTS (Artisanal fishery)

The Fisheries Department has been collecting data for many years on landings from the main fishing villages along the coast from fishing cooperative. The figures obtained from the different districts are summed by regions. No attempt is made to estimate the catch of boats and villages not covered. For the marine fisheries the catch is given for four Departments: Lamu, Kilifi (Malindi), Mombasa and Kwale. These four regions correspond geographically with the four strata of the CAS. The total catch registered for the year 1984 for the four strata by the old statistical survey are around 4,995 metric tonnes, which is about 1/2 of the yearly estimates of the CAS. It must be remembered that the time periods covered for two of the CAS strata do not precisely correspond to the calendar year 1984 and the old survey data for 1984 are not yet official.

7.1. Comparison by Stratun and by Month of Final Results

Stratum 01: LAMU

Old survey estimates for this stratum give a total catch for 1984 of 1,194 metric tonnes, 2.4 times less than the estimate of 2,867 metric tonnes from the CAS. The graph in figure 1 shows the monthly total catch estimates for the two surveys. From January to June the trends are similar. Important differences arise for October and November where the old survey estimates do not show the peak found in the CAS estimates.

Stratum 02: MALINDI

The 1984 total catch is estimated by the old survey at 789 metric tonnes. This is almost 2.4 times less than the estimate of 1,856 metric tonnes given by the CAS for the period April 1984 to March 1984. The patter of the two graphs does not show much similarity (see figure 2)

Stratum 03: MOMBASA

Old survey and CAS estimates total catches for 1984 are, for this stratum, very similar: respectively 2,166 and 2,142. This is probably due to the fact that around Mombasa the old survey covers a high percentage of the landings sites. The graphs of monthly estimates for the two surveys show a quite different trend (see figure. 3).

Stratum 04: KWALE

The difference between the two estimates of total yearly landings is higher than in the other strata: The old survey estimate for the year 1984 is 820 metric tonnes while for the period April 1984 to March 1985 the CAS gives a figure 3.7 times higher, at 2,998 metric tonnes. Graph 4 shows the two curves of total catch by month. Although the curve of the old survey estimates is very flat a certain similarity can be seen in the two curves.

8. CONCLUSIONS AND RECOMMENDATIONS

8.1. Source of Error
8.2. Recommendations

8.1. Although the survey was successfully completed, it was clear that the results were affected by various sources of error. For future surveys effort should be made to avoid or minimise them, and we know by experience that a lot can be done especially in the field of non sampling errors.

8.1. Source of Error

a) Non sampling errors

From an analysis of the non sampling errors occured during this first experience, they can be divided in 4 categories:

i) The fish recorder does not collect information or fill the forms correctly.

ii) The fish recorder is not present at the beach when the fishing boats are landing and collects data by enquiry.

iii) The fish recorder cannot weigh the catches because he does not have proper weighing scales and estimates the landings.

iv) The fish recorder incorrectly identifies fish species.

For case (i) and (ii) the only solution is increased supervision. The presence on the bach of a supervisor from time to time will have double effect.

1. The supervisor will be able to check the work done by the recorder and correct possible mistakes directly from the source.

2. The recorder will have the feeling that the administration cares about the work he is doing and that fisheries data collection is not a meaningless routine.

Case (iii) has one possible solution, the purchase of appropriate weighing scales and buying whatever material is needed to keep them in good working condition.

Case (iv) become a very important source of error when catch estimates by species groups are produced. A possible solution is to organise local training courses for the fish recorders. These training courses could be useful also to present to and discuss with the enumerators how to solve problems arising in particular situations.

b) Sampling Errors

i) The Sampling Frame is used not only to select the PSU but also to calculate the raising factors used to raise sampling data from the PSU to total catch and effort for the strata. An error in the number of boats per stratum or in the gear distribution would therefore affect the estimates. A certain number of possible sources of errors in the sampling frame should be expected due to the difficulties encountered during the Frame Survey.

ii) The sampling Frame no longer reflects the reality of the fishing industry. In fact it is already two years old and will become more and more absolute, depending on the rate of change of the Artisanal Fisheries in the country or in individual strata.

iii) Boats changing the gear used depending on weather and seasonal changes of the fishing pattern and the fishery structure may affect the final estimates.

c) Other sources of error

Geographic boundaries and national borders do not mean much to the fishermen. It is well known that, along the East Coast of Africa, groups of fishermen migrate following the migration or abundances of fish and changes in general economic conditions in the different countries or areas along the coast. These movements would greatly affect the stratum estimates unless they were limited to the boundaries of one stratum.

In smaller countries of the region Catch Assessment surveys are designed in such a way as to continuously update the sampling frame. This system could not be applicable to countries like Kenya where the coast stretches for hundreds of kilometers and means of road transport are not always available.

Nevertheless, some possibility of updating the sampling frame must be envisaged:

i) The seasonal fluctuation in number of boats operating could be obtained by studying the fluctuation in the number of boats in operation in the PSUs.

ii) The supervisors could carry out regular visits to fishing villages which are not Primary Sampling Units to collect data on number and type of boats and gear used. This data could be correlated with those at point (i) to study seasonal fluctuations, and could also be used to update the Sampling Frame.

iii) The existing registration system of fishing boats could be upgraded and kept up to date.

Although many possible sources of error have been identified, the results of the first year of implementation of the CAS is thought to be a real improvement on the previous survey. The catch estimates provided by the CAS appear realistic, and seasonal patterns can be identified although a certain margin of error has to be assumed.

8.2. Recommendations

- It is highly recommended to continue the new catch assessment survey with some modifications in order to face more realistic situations and introduce improvements mainly in terms of efficiency of the survey system and also expand its coverage to the commercial fishery sector. In this respect some ad hoc assistance from FAO/FIDI/SWIOP can be foreseen.

- It is also strongly recommended to initiate a new Frame Survey with some improvement of the previous one.

- For this fishery statistical surveys a body of some 30 field staff and supervision received training and gained experience in statistical fishery data collection methods. This valuable investment should not be wasted but increased.

It is therefore advisable to arrange from time to time further on-job training for recorders and supervisors in data collection, species identification, basic statistical concepts and other practical topics.

To summerise, now that the new fishery statistical programme has been initiated the Kenyan Fishery Department should concentrate its efforts on:

- Updating the catch assessment survey system.

- Increasing field supervision/training.

- Providing better logistic assistance and equipment.

- Developing more routines for processing and presenting the results of the surveys.

- Expand the coverage of the survey to include the commercial fishery and other sectors of the fishing industry (i.e. fixed gear, traps etc.).

- Initiating a new Frame Survey.

- In a second instance initiating the preparatory work to implement the second phase and the adopted statistical survey programme (developing techniques for special surveys such as cost earning surveys, fish marketing surveys etc.).

APPENDIX A - CAS MONTHLY ESTIMATES BY GEAR AND BY STRATUM

Monthly results by stratum and by gear
Graphs of monthly total catch estimates by strata
Graphs of monthly total catches by gear and by strata

Monthly results by stratum and by gear

TABLE 1 - MONTHLY CATCHES BY GEAR: STRATUM 1 (in kgs)

MONTH	CAST NET	GILL NET	BEACH SEINE	HAND LINE	OTHER GEAR	LOBSTER	TOTAL
JANUARY	(54070)	125287	73529	28205	37206	4607	322904
FEBRUARY	53984	107108	23709	43964	31555	8084	268404
MARCH	(40789)	66798	95018	12638	11747	16600	242590
APRIL	(22845)	56744	40000	(7078)	4764	5002	136434
MAY	(37292)	126364	32946	13477	(7463)	5161	222703
JUNE	(33198)	79704	(29340)	39675	11798	(4542)	198257
JULY	(30773)	65838	(42547)	13936	18156	(12524)	183774
AUGUST	(29523)	18133	55754	40472	11920	20506	176308
SEPTEMBER	(40348)	67240	84468	10767	31208	6924	240955
OCTOBER	(62861)	157604	94554	20374	22301	17707	375401
NOVEMBER	(37196)	66092	(54594)	41206	16206	6835	222129
DECEMBER	(46265)	125781	(67643)	19728	9176	7698	276291
TOTAL	489145	1062693	694102	291520	213500	116190	2867150

N.B: Figures in brackets are estimated.

TABLE 2 - MONTHLY CATCHES BY GEAR: STRATUM 2 (in kgs)

MONTH	CAST NET	GILL NET	BEACH SEINE	HAND LINE	OTHER GEAR	LOBSTER	TOTAL
JANUARY	7210	58541	10299	41313	28687	717	146767
FEBRUARY	3951	95790	4704	54126	15452	368	174391
MARCH	1166	183321	5480	51928	18154	377	260426
APRIL	7816	66963	4520	54990	20925	683	155897
MAY	11504	72277	8622	40900	26342	267	159912
JUNE	4669	86406	10425	18155	15314	608	135577
JULY	6028	82442	12496	31906	17337	716	150925
AUGUST	4518	76117	27583	18440	4468	622	131748
SEPTEMBER	5840	73270	22517	51427	10542	772	164368
OCTOBER	4875	41772	19985	20288	12967	773	100660
NOVEMBER	5498	67001	11256	34398	15012	764	133929
DECEMBER	6167	79564	13593	27952	13143	610	141020
TOTAL	69242	983464	151480	445823	198343	7268	1855620

TABLE 3 - MONTHLY CATCHES BY GEAR: STRATUM 3 (in kgs)

MONTH	CAST NET	GILL NET	BEACH SEINE	HAND LINE	OTHER GEAR	LOBSTER	TOTAL
JANUARY	30204	24222	17276	49595	57170	Ø	178467
FEBRUARY	20505	19528	5818	56930	58996	Ø	161777
MARCH	23995	24821	4026	56066	54059	Ø	162967
APRIL	35468	20734	5842	58248	38052	Ø	158344
MAY	49131	21384	7854	65224	105632	Ø	249225
JUNE	33464	27120	7000	58961	55200	Ø	181745
JULY	50620	30378	5445	64610	44571	Ø	195624
AUGUST	42447	32447	4151	53052	56196	Ø	188293
SEPTEMBER	42447	28594	5815	60907	54787	Ø	192550
OCTOBER	19338	25753	8105	63122	63320	Ø	179638
NOVEMBER	9395	23372	8902	42933	46419	Ø	131021
DECEMBER	22994	25582	6993	49603	56890	Ø	162062
TOTAL	380008	303935	87227	679251	691292	Ø	2141713

TABLE 4 - MONTHLY CATCHES BY GEAR: STRATUM 4 (in kgs)

MONTH	CAST NET	GILL NET	BEACH SEINE	HAND LINE	OTHER GEAR	LOBSTER	TOTAL
JANUARY	(28147)	28590	52636	64710	59336	(143)	233562
FEBRUARY	(34548)	15716	47773	48389	51724	(103)	198253
MARCH	(43669)	(18854)	(68639)	(54651)	(49916)	(173)	235902
APRIL	52790	21992	89505	60913	48107	(131)	273438
MAY	26438	27784	41250	43881	72770	130	212253
JUNE	(31218)	26318	53469	33685	35966	(269)	180925
JULY	(27180)	33079	26210	49526	68633	(376)	205004
AUGUST	(6614)	29803	30488	47741	36285	(321)	151252
SEPTEMBER	(4421)	24719	39767	64016	49677	488	183088
OCTOBER	7867	56627	140878	116213	66135.	(470)	388190
NOVEMBER	14297	(33000)	62257	92778	122363	(535)	325230
DECEMBER	(21047)	(23000)	232657	68637	65183	372	410896
TOTAL	298235	339482	885529	745140	726095	3511	2997993

N.B: Figures in brackets are estimated.

Graphs of monthly total catch estimates by strata

FIGURE 1

FIGURE 2

FIGURE 3

FIGURE 4

Graphs of monthly total catches by gear and by strata

FIG. 5

FIG. 6

FIG. 7

FIG. 8

FIG. 9

FIG. 10

FIG. 11

FIG. 12

FIG. 13

FIG. 14

FIG. 15

FIG. 16

FIG. 17

FIG. 18

FIG. 19

FIG. 20

FIG. 21

FIG. 22

FIG. 23

FIG. 24

FIG. 25

FIG. 26

FIG. 27

APPENDIX B

KENYA ARTISANAL FISHERY SAMPLING FRAME
BIG GEAR TABLE

KENYA ARTISANAL FISHERY SAMPLING FRAME

STRATUM: LAMU - CODE NO: 01

Administrative Division	Name of the Fishing village	Code No.	No. of			Total
			Dhows	Canoes	Cata.
KIUNGA	ISHAKANI	1205	16	14	2	32
	KIUNGA	1210	19	2	-	21
	MWAMBORE	1215	2	9	-	11
	RUBU	1?16	-	-	-	-
	MVUNGENI	1117	3	-	-	3
	MKOKONI	1220	5	1	1	7
SUB STRATUM 1
SUB TOTAL			45	26	3	74
FAZA	KIWAYUU IS.	1?21	15	1	-	16
	NDAN IS.	1122	-	-	-	-
	FAMBUZI	1223	-	3	6	9
	RASINI	1225	17	-	6	23
	KIZINGITINI	1330	67	21	-	88
	BWAJUMWALI	1235	-	-	5	5
	MYAMBOGI	1240	2	1	-	3
	CHUNDWA	1245	-	1	6	7
	SIYU	1250	-	-	5	5
	SHANGA	1251	3	-	10	13
	PATE	1252	-	1	6	7
	MTANGWANDA	1153	5	-	1	6
	DODORI	1154	2	7	15	24
	WANGE	1156	2	13	-	15
SUB STRATUM 2
SUB TOTAL			113	48	60	221
LAMU	MATONDONI	1225	-	25	-	25
	SHELLA	1260	33	-	-	33
	MANDA IS.	1263	1	3	8	12
	LAMU	1365	42	14	1	57
	KIPUNGANI	1170	2	6	-	8
	MUKOWE	1775	-	-	-	-
	MAGOGONI	1776	-	-	-	-
	NDUNUNI	1277	-	11	-	11
MPEKETONI	MKUNUMBI	1280	1	18	1	20
	NDAMBWE	1781	-	-	-	-
SUB STRATUM 3
SUB TOTAL			79	77	10	166
MPEKETONI	KIPINI	1385	7	13	2	22
	SHEKIKO	1?87	-	-	-	-
	ZIWAYUU Is.	1?90	-	18	3	21
	TENEWI Is.	1195	3	1	-	4
SUB STRATUM 4
SUB TOTAL			10	32	5	47
TOTAL			247	183	78	508

STRATUM: MALINDI - CODE NO: 02

Administrative Division	Name of the Fishing village	Code No.	No. of			Total
			Dhows	Canoes	Cata.
KILIFI	MTO KILIFI	2?05	-	-	-	-
	NGOMENI	2310	33	4	-	37
	GONGONI	2?15	-	-	-	-
SUB STRATUM 5
SUB TOTAL		-	33	4	-	37
KILIFI	MAMBRUI	2120	-	1	-	1
	MALINDI	2225	7	14	1	22
WATAMU	MAYUNGO	2126	7	16	1	24
	WATAMU	2230	10	19	1	30
	?	2131	1	3	-	4
SUB STRATUM 6
SUB TOTAL			25	53	3	81
WATAMU	SITA	2132	2	23	-	25
	MIDA CREEK	2?34	2	33	-	35
	UYOMBO	2135	-	-	-	-
	?	2136	-	-	-	-
	KWASESE	2?38	-	-	-	-
SUB STRATUM 7
SUB TOTAL			30	108	4	142
WATAMU	SOKOKE	2240	-	24	-	24
WATAMU	?	2144	-	3	-	3
	KILIFI	2245	3	33	-	36
	VIPINGO	2?48	-	-	-	-
	TAKAUNGU	2150	-	-	-	-
	?	2151	-	4	-	4
	KAMBALA	2?55	-	-	-	-
	KANAMOI	2160	-	5	-	5
	MTWAPA	2165	-	7	-	7
SUB STRATUM 8
SUB TOTAL			3	76	-	79
TOTAL			91	141	7	339

STRATUM: MOMBASA - CODE: 03

Administrative Division	Name of the Fishing village	Code No.	No. of			Total
			Dhows	Canoes	Cata.
MOMBASA	MARINA	3105	9	31	1	41
	SERENA	3310	8	16	-	24
	UTANGE	3215	2	21	-	23
	BAMBURI	3220	1	34	-	35
	NYALI	3122	-	13	-	13
	OLD PORT	3225	14	12	1	27
	SHELLY BEACH	3135	-	20	-	20
SUB STRATUM 9
SUB TOTAL			34	147	2	183
MOMBASA	LIKONI	3230	5	12	-	17
	MTONGWE	3140	-	2	-	2
	PORT REITZ	3245	-	68	-	68
	MARITINI	3150	-	13	-	13
	TSUNZA	3155	-	39	-	39
	MWANDUDU	3160	-	14	-	14
	KINTINJE	3165	-	23	-	23
	JONVU	3270	-	14	-	14
	JONVU KUU	3175	-	24	-	24
	TUDOR	3180	-	4	-	4
	KITIJE	3185	-	-	-	-
SUB STRATUM 10
SUB TOTAL			5	213	2	218
TOTAL			39	364	2	401

STRATUM: KWALE: - CODE NO: 04

Administrative Division	Name of the Fishing village	Code No.	No. of			Total
			Dhows	Canoes	Cata.
TIWI	TIWI	4105	-	20	5	25
	?	4106	-	18	2	20
	?	4107	-	6	2	8
DIANI	DIANI	4110	5	10	6	21
MSAMBWENI	KINODO	4315	-	39	1	40
SUB STRATUM 11
SUB TOTAL			5	93	16	114
MSAMBWENI	GAZI	4120	-	12	-	12
	?	4124	-	34	1	35
	MSANBWENI	4125	3	25	1	29
	?	4126	-	17	1	18
	SHIRAZI	4130	-	6	-	6
	BODO	4135	-	23	2	25
	KIWAMBALE	4140	1	16	1	18
	FUNZI	4145	1	19	1	21
	RASANZWANI	4150	-	10	-	10
SHIMONI	SHIMONI	4255	-	33	2	35
	MKWIRO	4160	10	29	2	41
	CHWAKA-KIBUJUNI	4170	3	46	-	49
MAJORENI	MAJORENI	4275	-	36	3	39
	MKOKONI IS.	4?80	-	-	-	-
VANGA	VANGA	4285	19	41	16	76
	JIMBO	4?90	-	-	-	-
	NGOA	4191	-	10	-	10
	KIWEGU	4192	-	20	-	20
SUB STRATUM 12
SUBTOTAL			37	377	30	444
TOTAL			42	470	46	558

BIG GEAR TABLE

STRATUM	SUBSTR.	BEACH SEINE	USIO	OTHERS
01	1	3
01	2	2	3
01	3	9	5	3
01	4
TOTAL:		14	8	3
02	5
02	6	4	1	0
02	7	9	3	0
02	8	7	34	3
TOTAL:		20	38	3
03	9	10	18	5
03	10	3	77	3
TOTAL:		13	95	8
04	11	5	13	5
04	12	11	112	3
TOTAL:		16	125	8

APPENDIX C - LIST OF FORMS

CAS-KAF 2.1 - Monthly Inventory Characteristics
CAS-KAF 2.2 - Daily Catch Report
CAS-KAF 2.3 - Monthly Catch Record
CAS-KAF 2.4B - Estimation Process
CAS-KAF 2.5 - Total Monthly Catch Estimates by fishing methods

CAS-KAF 2.1 - Monthly Inventory Characteristics

FORM: CAS-KAF/2.1

CAS-KAF 2.2 - Daily Catch Report

FORM: CAS-KAF/2.2 - part = a

FORM: CAS-KAF/2.2 - part = b

CAS-KAF 2.3 - Monthly Catch Record

FORM: CAS-KAF/2.3

CAS-KAF 2.4B - Estimation Process

FORM: CAS-KAF/2.4B

CAS-KAF 2.5 - Total Monthly Catch Estimates by fishing methods

FORM: CAS-KAF/2.5

APPENDIX D - METHODOLOGICAL NOTES ON THE IMPLEMENTATION OF THE CAS

METHODOLOGICAL NOTES ON THE IMPLEMENTATION OF THE CAS

1. FIELD WORK (Recorders/Supervisors)

1.1 General Guidelines

- Each recorder will be responsible for the collection of information within the preselected village/landing place (PSU), following the instructions given to him.

- All the forms provided must be fully completed following the instructions below.

- At the end of each month, the recorder will report to his supervisor on the monthly activity and deliver the "Daily catch record" questionnaires (form CAS-KAF/2.2).

- Every month the recorder will be instructed by his supervisor on the gears to be sampled during the forthcoming month.

- The recorder must be present at the landing place every fishing day, before the arrival of boat landing, and weigh himself the each landed from each selected boat by species. He should also collect information on the fishing effort in terms of total fishing hours.

- If for any reason the recorder cannot be present at the landing place on a fishing day (or night) he must report this to his direct supervisor and indicate it accordingly on form CAS-KAF/2.1 (Point f).

1.2 Survey Operation

At the beginning of each calendar month, the recorder (eventually assisted by the supervisor) must go to the assigned fishing site/landing place to interview the head of the village and respective fishermen and complete parts a, b, d, e, g and h of form CAS-KAF/2.1. Part f will be completed together with the supervisor.

1.3 Enumeration

Form CAS-KAF/2.1: 'Monthly inventory characteristics'

a. Indicate the stratum name and code number.

b. Indicate the name and code number of the fishing site/landing place, as per attached sampling frame.

c. Indicate the month (in letters) and the year.

d. Ask the head of the village or fishermen's representative (chairman) the "Total number of fishing boats haboured at the fishing site" and insert the number on the form. (Only active fishing boats should be considered).

e. Collect additional information from the fishermen/owners of the boats for the allocation of active fishing boats by fishing method (according to the gear used) and insert the respective numbers on the forms:

Gear Used:	01.	Cast net
	02.	Gill net
	03.	Beach seine
	04.	Hand line
	05.	Other
	09.	Lobster fishing

Indicate the number of boats using each gear type.

f. Insert the reference period for each fishing gear i.e. number of consecutive days within a month which have been allocated for the collection of information on a given gear; draw a Junction line under the surveying days.

The reference period must not be less than 7 consecutive days. The recorder must record the landings only of those fishing boats using a specific gear, as shown on form CAS-KAF 1 (point f). (E.g. If it has been agreed to record statistics of cast net fishing boat catch during the first ten days of the month, the recorder must try to interview for ten consecutive days all the cast net fishing boats landings). The enumerator will record under point the number of landings that occur during each sample fishing day (same as "L" in point f in the form CAS-KAF/2.2).

1.4 The reference period

The reference period has to be calculated as follows:

- For fishing villages with ten or less boats, irrespective of the fishing gear used: complete enumeration, specifying the predominant gear. Reference period: one month.

- For fishing villages with more than ten fishing units follow the examples given below:

i) Allocate two weeks for recording the catch and effort (hours) of boats using the predominant fishing method. (Enumeration of all active fishing boats using the specific gear and ignoring the remaining fishing units).

ii) Allocate one week for recording the catch and effort (hours) of boats using the second important fishing method. (Enumeration of all active fishing boats using the specific gear, ignoring the remaining fishing units).

iii) Allocate one week for recording the catch and effort (hours) of the remaining fishing units. (Enumeration of all the active fishing boats using any kind of gear, except those seen at point i) and ii) above.

1.5 Special Cases:

(i) In villages there is only one kind of fishing gear used: Complete enumeration with reference period: 1 month.

(ii) One very predominant fishing method and some others of minor importance. Allocate 2 weeks for the predominant (1st an 3rd Weeks). Allocate 2 weeks for the remaining (2nd and 4th Weeks).

(iii) Two or more predominant fishing methods of the same magnitude plus some minor ones:

Allocate 10 days for the first predominant gear
Allocate 10 days for the second predominant gear
Allocate 10 days for the remaining gear

In case of the recorder not being present at the landing place and therefore not being able to record the catch statistics, he must insert the code "00" in the box for that day under the number of landings. The code "0" will represent no landings on that day, (i.e. no fishing activity e.g. bad weather, festival day, etc..). At the end of the form indicate the date and get the supervisor's check and approval.

1.6 Form 2.2: Daily Catch Report:

Every fishing day the recorder must be present at the landing place until the landing process is terminated. Up to 10 fishing units can be recorded on one form CAS-KAF.

Part a:

One or more forms CAS-KAF 2.2 must be used to record items of information on ever fishing day. (If the fishing boats to be interviewed are between 11 and 20, then two forms are needed). Number the forms the same day as follows:

If one form is used:	Page 1/1
If two forms are used:	First form:	page 1/2
	Second form:	page 2/2
If three forms are used:	First form:	page 1/3
	Second form:	page 2/3
	Third form:	page 3/3 and so on

a) indicate the name and code number of the stratum;
b) indicate the name and code number of the fishing landing place (PSU);
c) indicate the year month (in letters) and the day of the month;
d) indicate the recorder's name and code number, the recorder's signature is required;
e) indicate the fishing method used and its code number.

Calculation of Landing Raising Factor

For the collection of catch statistics, we have used the census approach within each fishing method category, in other words, all the boats using a certain fishing gear, must be interviewed on the same day for a predetermined reference period. It is clear that, under normal circumstances, the number of boats landed must equal the number of boats interviewed. In some rare cases it will not be possible to interview all the landed boats on the same day (because too many landed at the same time or some refused to be interviewed, etc..). In these cases, item f should be completed.

f) At the end of the day, indicate:

- the number of fishing boats landed (Indicate as 'L' under the number of landings on form CAS-KAF 2.1, point f)

- the number of fishing boats interviewed (indicated as 'l'). The landing raising factor will be given by the ratio -

R1 = L/l

Part b:

Up to 10 fishing units can be recorded on one form CAS-KAF 2.1.

- Species are vertically listed and codified;

- The sequence of the fishing boats to be interviewed is horizontally organized. Insert the sequence number of the arrival.

- For each fishing unit interviewed, insert under its sequence number the type of fishing boat and the quantity and value of the catch in kg's and K.Shs against the appropriate species.

- The codification of the types of fishing boats is:

1. Canoes
2. Dhows
3. Others (specify)

Also, the total number of fishing hours for each fishing units must be inserted at the bottom of the column.

1.7 How to calculate Daily Landing Totals

Marginal Totals

At the end of the day make the necessary calculation by:

i) adding up vertically (by column) we get the daily total of fish landed by a single fishing unit irrespective of the species,

ii) adding up horizontally (by line) we get the daily sample total of all fishing units by species.

The horizontal totals must be written in the 'Sample totals' column. The "Daily totals by Species" column will be used if the values in f (L: number of boats landed) and (1: number of boats interviewed) are not equal. In this case, multiply the totals (including fishing hours) in the 'Sample Total' column, by the ratio L/l (R1) and record the result in the 'Daily Totals by Species column'.

Example: If 12 fishing units landed, and only 10 were interviewed (L = 12, l = 10), all the sample totals must be multiplied by 1.2.

R1 = L/l = 12/10 = 1.2

NOTE: The recorder must be aware of the importance of his task. The accuracy of the monthly estimates mostly depends on his skill and professional honesty. By following the plan as drawn on form CAS-KAF 2.1, once the first reference period is terminated the recorder must pass to the second predominant fishing method, and so on.

At the end of the month - see point 1.

2. OFFICE WORK - (Supervisors Office Staff)

List of forms to be used:

CAF-KAF 2.3 - MONTHLY CATCH RECORD - ?

CAF-KAF 2.4 - ESTIMATION PROCESS (STRATUM SAMPLE TOTAL)

CAF-KAF 2.5 - TOTAL MONTHLY CATCH ESTIMATES BY FISHING METHODS (BY STRATUM)

2.1 General Guidelines

- Read carefully all the instructions given below.

- The estimation process is the same for catch estimates, value estimates and fishing hours estimates (fishing effort). Amend forms CAS-KAF 2.1, 2.2, 2.3 for value estimates. Amend forms CAS-KAF 2.1, 2.2, 2.3 for fishing hours (effort).

- The supervisors must continuously supervise the field recorders.

- Check the content and arithmetic of the forms submitted monthly to the District Office by the field recorders.

- In your estimation process, note that the landing raising factors (R1 = L/l) can be equal or bigger than 1, never less than 1.

- After having completed the consistency and accuracy check of the source documents, transfer the data onto form CAS-KAF 2.3, following the correct instructions. Do not destroy the source documents, but keep them for record in the office, organized on a monthly file system.

- Errors can creep in at any stage of the processing phase, check carefully also when simply transferring the results from one form to another.

2.2 How to process the results Form CAS-KAF 2.1:

- Supervisors and office staff must check the content of this form and correct when necessary any discrepancies appearing in the completed forms. Pay particular attention to the number of landings per fishing day and the number of fishing days in the month (point f).

Form CAS-KAF 2.2:

- Check the codification system.

- Check the validity of the landing raising factory (R1) by comparing forms CAS-KAF 2.1 and 2.2.

- The landing raising factor R1 is given by -

R1 = L/l

where L = number of landings per day per boat using the same fishing gear and l = number of sampled (interviewed) fishing units, among those landed (L) in that day.

PART B

- Check the consistency and accuracy of the tables and correct where possible major discrepancies,

- Cross check the calculations,

- Check correct application of the landing/raising factor,

- Check marginal totals and the "Daily Totals by Species" and correct mathematical errors.

Form CAS-KAF 2.3

1. Complete the heading of the form by indicating the name of the fish landing site (PSU) and its code number, the stratum number and the date (Month and Year).

2. Aggregate the "Daily Totals by Species" (in the case of a census approach use data from "Sample Data").

Once the summation and transfer are completed, work out the monthly total estimates.

The "Monthly Totals" for a given sample fish landing centre (PSU) are obtained by raising the "Sample Data" on a species basis to monthly totals (including fishing hours) by using the time raising factor (R2) which is given by:

R2 = M/D

where M = number of days in calendar month and D = number of days in the reference period (from form 2.1 Table f).

In the case where the non-recording code "00" appears in Table f on form 2.1., those days must be substracted from D (reference period days).

Calculate the time raising factors and indicate them in the appropriate box R2. Then multiply the values in column "Sample data" by "R2" and indicate the results under the "Monthly Total" columns. The form CAS-KAF 2.3 consists of two pages, the three most important gears to be considered are listed on the first page and the remaining three on the second page.

When the raising process has been completed, calculate the marginal totals on both pages. For the catch estimation process the totals obtained from the monthly totals column are used.

Form CAS-KAF 2.4B

To calculate the monthly estimated total catch and effort (hours) per stratum by gear used, transfer form CAS-KAF/2.3 to form CAS-KAF/2.4B the monthly sample totals for each PSU. Multiply the sample total catch and effort for each gear and for each village (PSU) by R3 to obtain different stratum estimates. R3 is the ratio between the total number of boats in the stratum (N) and the number of boats in the PSU.

R3 has to be calculated for each gear used and for each PSU.

For each PSU and gear used, different stratum estimates for catch and effort will be obtained. Compute the average of the different stratum estimates obtained for each PSU and each gear type.

The result will be one catch and one effort stratum estimate for each gear used. Form CAS-KAF 2.5

Transfer the total LANDED CATCH-STRATUM from the four forms CAS-KAF 2.4B (one for each stratum) to the appropriate "Stratum/gear used" box on form CAS-KAF 2.5. By adding up horizontally, we get the monthly total by stratum irrespective of stratum. The "Overall monthly total" is the estimated monthly catch for the sector. (Artisanal/Marine).

Repeat the estimation process every month.

IMPORTANT: UNDER NO CIRCUMSTANCES SHOULD THE SAMPLE (PSU) BE CHANGED

Suggestions:

Form CAS-KAF 2.5 must, definitely be compiled at headquarters in Mombasa.

Form CAS-KAF 2.1 and CAS-KAF 2.2 must be completed by the recorder/supervisor at the landing place. Completed forms CAS-KAF 2.2 must be kept in the District Office.

Form CAS-KAF 2.3 must be compiled in the stratum office (District Office) and then coupled with Form CAS-KAF 2.1.

Form CAS-KAF 2.4B must be analysed by the office (District or Headquarters) which collects forms CAS-KAF 2.1 and 2.3 together. (This task must be defined by the Department of Fisheries, according to the available qualified personnel and the human and financial resources).

APPENDIX E - AREA DISTRIBUTION

Figure 1 - COAST OF KENYA STRATUM: 1 (1-2)

Figure 2 - COAST OF KENYA STRATUM: 1 (3-4)

Figure 3 - COAST OF KENYA STRATUM: 2 (5-6)

Figure 4 - COAST OF KENYA STRATUM: 2 (7-8)

Figure 5 - COAST OF KENYA STRATUM: 3

Figure 6 - COAST OF KENYA STRATUM: 4

Fig. 7 - The coast of Kenya

APPENDIX F - STAFF SUPERVISORY

CATCH ASSESSMENT SURVEY 1984

STAFF SUPERVISORY

STRATUM 1
(1)	Mr. Andika Semo - F.O.
(2)	Mr. Ronald Deche - Kizingitini
(3)	Mr. Christopher Muhombe - kiunga
(4)	Mr. Mwangare - kipini
(5)	Recorders
STRATUM 2
(1)	Mr. Kithome - F.O.
(2)	Mr. Kasoso - Vanga
(3)	Recorders
STRATUM 3
(1)	Mr. L.N. Thairo - Overall Coordinator
(2)	Miss M. Mapinga
(3)	Mr. P. Musilu
(4)	Mr. B. Sanga
(5)	Recorders
STRATUM 4
(1)	Mr. Obadha - F.O.
(2)	Mr. A. Kombo - Vanga
(3)	Recorders

APPENDIX G - CATCH COMPOSITION IN CATCH ASSESSMENT SURVEY

Percentage species in the catch of January 1984 - February 1985
Graph of monthly distribution of important species and species groups
Monthly distribution of catches of Lobster
Percentages species in catches of stratum

Analysis of the data by:

M. VAN NIEROP
Biologist (RAF/79/065)

CATCH COMPOSITION IN CATCH ASSESSMENT SURVEY KENYA, JANUARY '84 - FEBRUARY '85

Data on catches by species, collected by the Catch Assessment Survey during the period January '84 - February '85, are processed on a Apple III personal computer with the software package Visicalc. Table 1 shows the final results: the percentages of 38 species catagories in the catch (weight) of cast net, gill net, beach seine, hand line, other geartypes and lobster fishing. A few remarks can be made on basis of these results:

- Rabbitfish, scavengers, sharks and rays, mixed demersal, sardines and prawns constitute 56 % of the total catch.

- 13 of the 38 species catagories represent less than one percent of the total catch. In terms of value, Beche-de-Mer and crabs (representing 0.16 and 0.14 percent of the total catch respectively) will probably have a more significant importance.

- 3 species catagories, gamefish, oyster and oyster shell grit, are not recorded at all.

- Large catches of both demersal and pelagic fish by gill net wil probably imply that two kinds of gill nets are in use.

- The catches of parrotfish, sardines, lobsters, prawns and Beche-de-Mer by hand line will probably originate from other gear types. Hand line fishermen will for example also haul some traps during an outing, resulting in a mixed catch.

- The high percentages of rabbitfish and prawns under "Other" will probably mean that this gear catagory consists mainly of basket (rabbitfish) and tidal (prawns) traps

Figures 1a-1f show the monthly distributions of species and species groups caught by the six gear catagories for all strata.

Some remarks can be made on these graphs:

Figure 1a, cast net

- Sardines are highly represented in catches during the period July - October.

- Prawns are highly represented in catches during the period February - June.

- In stratum 4 cast net catches mainly consists of demersal fish. Cast net data from this stratum were only available during the months April, August, November and December causing peaks in the graph "demersal fish".

- High catches of mullets during September in stratum 2, account for the peak in the graph "others".

Figure 1b, gill net

- Sharks and rays are mainly caught during the period December - January.

- The high percentages of both pelagic and demersal fish in the catches is probably the result of the fact that two different gill nets are in use: in stratum 2 mainly pelagics, sharks and rays are caught, while in stratum 3 mainly rabbitfish, scavengers, and parrotfish are represented in the catches (this catch composition is almost identical with the catch composition of the beach seine in this stratum).

- Kingfish are mainly caught during April - December.

- The peak in the graph "other pelagics" is mainly the result of high catches of Istiophoridae (marlin and sailfish) in stratum 2.

Figure 1c, beach seine

- The only seasonal trend which can be distinghuished in the graphs is the abundance of little mackerel (Rastrelliger kanagurta) during April - August.

- There is some difference in catch composition between the strata: in stratum 2 and in a lesser degree stratum 4, pelagics (mainly small) form an important part of the catch, while in stratum 1 and 3 mainly demersal fish are caught.

Figure 1d, hand line

No seasonal trend can be distinghuished in the graphs. Some remarks can be made on the differences in catch composition between the strata:

- In stratum 1 catfish forms an important part of the catch (more than 40 % of the total catch).

- In stratum 2 pelagics (particularly Carangidae, jacks) form the most important part of the catch, while in stratum 1, 3 and 4 mainly demersal fish are caught.

Figure 1e, other geartypes

- Prawns are caught mainly during the period February - June (like the catches of prawns by cast net). Prawns are mainly caught in stratum 1 and 3.

- Rabbitfish seems to be most abundant during the period September - November.

Figure 1f, lobster fishing

Because the catch consists almost uniformly of lobsters, the catch in weight is depicted per month. Unlike the other gear catagories, lobster fishing does not give constant monthly catches throughout the year. It seems that during March and during August - October lobster fishing is most profitable. Lobster fishing is mainly practised in stratum 1 and 2.

Tables 2 to 5 show the species composition by gear used for each stratum.

Percentage species in the catch of January 1984 - February 1985

Table 1. PERCENTAGES SPECIES IN THE CATCH OF JANUARY 1984 - FEBRUARY 1985

Species or species group	% CAST NET	% G. NET	% B. SEINE	% H. LINE	% OTHER	% LOBST.	% TOTAL
Rabbitfish	2.67	10.13	19.45	9.44	31.04	-	14.49
Scavenger	4.14	7.57	13.97	20.92	11.01	-	11.78
Snapper	1.00	1.76	3.63	5.74	4.18	-	3.31
Parrotfish	1.82	2.85	7.65	4.47	9.37	-	5.12
Surgeonfish	ni l**	ni l	1.26	ni l	ni l	-	ni l
Unicornfish	ni l	ni l	1.10	ni l	ni l	-	ni l
Grunter	ni l	ni l	ni l	ni l	1.11	-	ni l
Pouter	1.17	ni l	3.26	1.37	ni l	-	1.17
Black skin	ni l	ni l	1.85	ni l	1.30	-	ni l
Goatfish	ni l	ni l	ni l	ni l	1.41	-	ni l
Streaker	ni l	ni l	ni l	1.73	ni l	-	ni l
Rock Cod (Grouper)	ni l	2.35	1.54	5.91	2.24	-	2.69
Catfish	ni l	2.48	ni l	7.04	ni l	-	2.35
Mixed demersal	3.36	5.23	14.82	6.87	5.94	-	7.13
Jacks	ni l	1.56	3.80	7.03	ni l	-	2.87
Mullets	2.53	1.94	3.26	ni l	1.09	-	1.73
Little Mackerel	1.34	2.69	8.11	ni l	ni l	-	2.52
Barracuda	ni l	ni l	ni l	1.51	ni l	-	ni l
Milkfish	-***	1.23	ni l	ni l	ni l	-	ni l
Kingfish	ni l	13.24	ni l	2.47	ni l	-	4.28
Queenfish	ni l	2.69	1.47	ni l	ni l	-	1.14
Sailfish	ni l	3.51	ni l	1.84	-	-	1.38
Tuna like Fish	-	5.67	-	1.10	-	-	1.80
Dolphinfish	-	ni l	ni l	ni l	ni l	-	ni l
Mixed Pelagic	-	ni l	3.77	2.10	ni l	-	1.24
Sharks/Rays	2.18	27.86	ni l	8.12	ni l	-	9.84
Sardines	43.76	ni l	3.55	1.09	1.40	-	6.93
Mixed Fish/Others	2.56	1.38	3.31	5.01	9.01	-	4.09
Lobster	ni l	ni l	ni l	ni l	ni l	94.56	1.31
Prawns	27.79	ni l	-	ni l	13.15	-	6.02
Crabs	ni l	ni l	ni l	-	ni l	4.93	ni l
Gamefish	-	-	-	-	-	-	-
Oysters	-	-	-	-	-	-	-
Oyster Shell Grit	-	-	-	-	-	-	-
Beche-de-Mer	ni l	ni l	-	ni l	ni l	-	ni l
Octopus	-	1.56	ni l	2.23	2.95	0.51	1.49
Squid	ni l	ni l	ni l	ni l	-	ni l	ni l
Shells	-	ni l	-	-	ni l	-	ni l
TOTAL	100.00	100.00	100.00	100.00	100.00	100.00	100.00

**: less than 1% of the catch
***: no recordings

Graph of monthly distribution of important species and species groups

FIG. 1a - GRAPH OF MONTHLY DISTRIBUTION (as percentage of total catch) OF IMPORTANT SPECIES AND SPECIES GROUPS.

GEAR TYPE: CAST NET

Sardines

Demersal fish

Prawns

Others (mainly pelagics)

FIG. 1b - MONTHLY DISTRIBUTION (as percentage of total catch) OF IMPORTANT SPECIES AND SPECIES GROUPS.

GEAR TYPE: GILL NET

Sharks and rays

Other demersal fish

Rabbitfish

Kingfish

Other pelagic fish

FIG. 1c - MONTHLY DISTRIBUTION (as percentage of total catch) OF IMPORTANT SPECIES AND SPECIES GROUPS.

GEAR TYPE: BEACH SEINE

Rabbitfish

Scavengers

Other demersal fish

Little mackerel

Others (mainly pelagics)

FIG. 1d - MONTHLY DISTRIBUTION (as percentage of total catch) OF IMPORTANT SPECIES AND SPECIES GROUPS.

GEAR TYPE: HANDLINE

Pelagic fish

Sharks and rays

Scavengers

Rabbitfish

Other demersal fish

FIG. 1e - MONTHLY DISTRIBUTION (as percentage of total catch) OF IMPORTANT SPECIES AND SPECIES GROUPS.

GEAR TYPE: OTHERS (mainly traps)

Rabbitfish

Scavengers

Other demersal fish

Prawns

Others

Monthly distribution of catches of Lobster

FIG. 1f - MONTHLY DISTRIBUTION OF CATCHES OF LOBSTER (in tons).

GEAR TYPE: LOBSTER FISHING

Percentages species in catches of stratum

Table 2. PERCENTAGES SPECIES IN CATCHES OF STRATUM I

Species or species group	% CAST NET	% G. NET	% B. SEINE	% H. LINE	% OTHER	% LOBST.	% TOTAL
Rabbitfish	-	7.20	35.07	ni l	30.23	-	14.39
Scavenger	-	8.58	20.68	17.54	14.07	-	12.43
Snapper	9.73	4.38	1.43	7.20	ni l	-	3.66
Parrotfish	-	ni l	1.76	ni l	1.61	-	ni l
Surgeonfish	-	ni l	1.06	-	ni l	-	ni l
Unicornfish	-	ni l	ni l	-	ni l	-	ni l
Grunter	-	ni l	1.37	1.70	1.72	-	ni l
Pouter	-	ni l	1.45	ni l	ni l	-	ni l
Blackskin	-	ni l	4.07	ni l	1.16	-	1.11
Goatfish	-	-	1.15	-	ni l	-	ni l
Streaker	-	ni l	ni l	ni l	-	-	ni l
Rock Cod (Grouper)	-	4.03	2.48	18.00	3.80	-	5.20
Catfish	-	5.30	ni l	33.22	1.77	-	7.08
Mixed demersal	-	10.50	19.04	4.01	12.78	-	11.12
Jacks	-	2.61	2.96	7.73	2.10	-	3.13
Mullets	-	2.77	2.46	1.55	6.73	-	2.73
Little Mackerel	-	1.82	ni I	ni l	-	-	ni l
Barracuda	-	ni l	ni l	ni l	-	-	ni l
Milkfish	-	ni l	-	ni l	-	-	ni l
Kingfish	-	5.00	ni l	ni l	-	-	2.42
Queenfish	-	1.99	ni l	ni l	-	-	ni l
Sailfish	-	2.06	-	-	-	-	ni l
Tunalike Fish	-	7.73	-	ni l	-	-	3.79
Dolphinfish	-	ni l	ni l	ni l	ni l	-	ni l
Mixed Pelagic	-	ni l	ni l	ni l	ni l	-	ni l
Sharks/Rays	-	29.51	ni l	3.30	ni l	-	14.71
Sardines	-	-	-	ni l	-	-	-
Mixed Fish/Others	-	2.94	2.78	1.15	ni l	-	2.29
Lobster	-	-	-	-	-	100.00	4.20
Prawns	86.86	-	-	-	20.23	-	4.03
Crabs	3.41	-	-	-	-	-	ni l
Gamefish	-	-	-	-	-	-	-
Oysters	-	-	-	-	-	-	-
Oyster Shell Grit	-	-	-	-	-	-	-
Beche-de-Mer	-	-	-	-	-	-	-
Octopus	-	-	-	-	-	-	-
Squid	-	-	-	-	-	-	-
Shells	-	-	-	-	ni l	-	ni l
TOTAL	100.00	100.00	100.00	100.00	100.00	100.00	100.00

Table 3. PERCENTAGES SPECIES IN CATCHES OF STRATUM II

Species or species group	% CAST NET	% G. NET	% B. SEINE	% H. LINE	% OTHER	% LOBST.	% TOTAL
Rabbitfish	-	5.81	3.13	ni l	87.78	-	12.87
Scavenger	-	ni l	ni l	7.68	1.23	-	2.13
Snapper	-	ni l	ni l	6.22	ni l	-	1.67
Parrotfish	-	ni l	ni l	-	5.81	-	ni l
Surgeonfish	-	ni l	-	ni l	ni l	-	ni l
Unicornfish	-	ni l	-	ni l	ni l	-	ni l
Grunter	-	ni l	-	-	ni l	-	ni l
Pouter	-	ni l	ni l	ni l	ni l	-	ni l
Black skin	-	-	-	ni l	ni l	-	ni l
Goatfish	-	ni l	-	-	ni l	-	ni l
Streaker	-	ni l	-	1.96	ni l	-	ni l
Rock Cod (Grouper)	-	ni l	ni l	4.09	ni l	-	1.25
Catfish	-	ni l	ni l	5.59	ni l	-	1.76
Mixed demersal	ni l	4.03	20.25	16.34	1.09	-	7.78
Jacks	3.65	ni l	ni l	16.32	ni l	-	4.58
Mullets	11.00	ni l	28.42	ni l	1.08	-	3.44
Little Mackerel	-	1.86	20.31	ni l	ni l	-	2.69
Barracuda	-	ni l	-	ni l	-	-	ni l
Milkfish	-	ni l	-	ni l	-	-	ni l
Kingfish	ni l	32.37	-	6.55	ni l	-	18.50
Queenfish	-	4.98	ni l	ni l	ni l	-	2.85
Sailfish	-	1.27	-	2.41	-	-	1.24
Tuna like Fish	-	6.53	-	1.48	-	-	3.76
Dolphinfish	-	-	-	ni l	-	-	ni l
Mixed Pelagic	-	ni l	-	7.04	-	-	1.72
Sharks/Rays	8.45	35.53	-	20.42	-	-	23.82
Sardines	11.61	1.61	ni l	-	-	-	1.37
Mixed Fish/Others	-	ni l	23.88	ni l	ni l	-	2.40
Lobster	2.75	ni l	-	-	-	85.92	ni l
Prawns	61.51	ni l	-	-	-	-	2.74
Crabs	-	ni l	ni l	-	-	13.17	ni l
Gamefish	-	-	-	-	-	-	-
Oysters	-	-	-	-	-	-	-
Oyster Shell Grit	-	-	-	-	-	-	-
Beche-de-Mer	-	-	-	-	-	-	-
Octopus	-	ni l	-	-	-	ni l	ni l
Squid	-	ni l	ni l	ni l	-	ni l	ni l
Shells	-	-	-	-	-	-	-
TOTAL	100.00	100.00	100.00	100.00	100.00	100.00	100.00

Table 4. PERCENTAGES SPECIES IN CATCHES OF STRATUM III

Species or species group.	% CAST NET	% G. NET	% B. SEINE	% H. LINE	% OTHER	% LOBST.	% TOTAL
Rabbitfish	ni l	28.12	21.26	20.24	12.40	-	15.25
Scavenger	2.54	21.31	16.78	16.84	9.45	-	12.45
Snapper	-	ni l	ni l	1.83	ni l	-	ni l
Parrotfish	ni l	26.60	29.59	10.84	12.38	-	12.08
Surgeonfish	ni l	-	ni l	ni l	-	-	ni l
Unicornfish	ni l	-	ni l	ni l	ni l	-	ni l
Grunter	ni l	-	-	1.13	1.62	-	ni l
Pouter	1.73	-	ni l	1.83	ni l	-	1.06
Black skin	-	-	-	ni l	-	-	ni l
Goatfish	ni l	-	-	-	ni l	-	ni l
Streaker	ni l	-	-	1.09	-	-	ni l
Rock Cod (Grouper)	ni l	-	-	2.38	ni l	-	ni l
Catfish	ni l	ni l	-	-	ni l	-	ni l
Mixed demersal	2.58	4.34	9.23	4.95	3.78	-	4.28
Jacks	-	-	2.73	5.24	ni l	-	2.15
Mullets	ni l	-	ni l	ni l	ni l	-	ni l
Little Mackerel	2.35	-	7.65	ni l	ni l	-	1.17
Barracuda	ni l	-	-	?.46	ni l	-	ni l
Milkfish	-	-	-	ni l	ni l	-	ni l
Kingfish	ni l	-	-	1.62	-	-	ni l
Queenfish	ni l	-	-	ni l	ni l	-	ni l
Sailfish	ni l	-	-	-	-	-	ni l
Tuna like Fish	-	-	-	ni l	-	-	ni l
Dolphinfish	-	-	-	-	-	-	-
Mixed Pelagic	-	-	-	ni l	ni l	-	ni l
Sharks/Rays	1.60	-	-	7.21	ni l	-	2.84
Sardines	58.68	-	ni l	1.68	3.58	-	11.86
Mixed Fish/Others	4.58	-	-	11.55	19.63	-	11.24
Lobster	ni l	-	-	ni l	ni l	-	ni l
Prawns	21.91	-	-	ni l	24.31	-	11.77
Crabs	ni l	-	-	-	ni l	-	ni l
Game fish	-	-	-	-	-	-	-
Oysters	-	-	9.23	-	-	-	-
Oyster Shell Grit	-	-	2.73	-	-	-	-
Beche-de-Mer	ni l	2.59	ni l	ni l	ni l	-	ni l
Octopus	-	16.47	7.65	6.02	7.50	-	6.35
Squid	ni l	-	-	-	-	-	ni l
Shells	-	ni l	-	-	-	-	ni l
TOTAL	100.00	100.00	100.00	100.00	100.00	-	100.00