10. LANDING SURVEYS

10.1 OBJECTIVES OF LANDING SURVEYS
10.2 SAMPLING REQUIREMENTS
10.3 A GENERAL-PURPOSE FORM
10.4 CASE STUDIES
10.5 TRAINING OF DATA COLLECTORS
10.6 BRIEFING OF DATA COLLECTORS
10.7 IMPLEMENTATION ASPECTS
10.8 FREQUENT PROBLEMS

In this section the methodological and operational aspects of sample-based Landing Surveys are presented, including:

• Objectives and basic data coverage
• Sampling requirements
• Example of a general-purpose form
• Case studies
• Training and briefing of data collectors
• Implementation aspects
• Data editing and checking
• Frequent problems

10.1 OBJECTIVES OF LANDING SURVEYS

Landings Surveys are conducted at landing sites with the purpose of collecting sample data on total catch and species composition, associated effort, and other secondary data such as prices and fish size (in weight units). In this handbook the following basic data are described:

• Catch of all species
• Associated fishing effort
• Overall CPUE
• Catch by species
• First-sale prices
• Number of fish in catch by species

The primary objective is to formulate, on a sample basis, overall CPUEs and species proportions within the estimating context of a minor stratum, a calendar month and a specific boat/gear category. Section 2.1 has provided guidelines in relation to this process.

10.2 SAMPLING REQUIREMENTS

In Landing Surveys sampling requirements and safe sample sizes to achieve minimum accuracy levels are different from those used in Boat Activity Surveys. This stems from the fact that the target populations of landings are much less demanding in sample size than those of boat activities.

Safe sample sizes in Landing Surveys are determined on the basis of:

• Desired accuracy level (with a minimum of 90%)
• Data population size (above 50,000 the population of landings is considered infinite)

The population size for landings is usually set at the theoretical maximum number of landings that can occur during a month. For instance, if 100 trawlers in a minor stratum operate in June 2001, then the maximum possible number of landings is 30 × 100 = 3,000 landings. Based on that limit and the desired level of accuracy it is possible to determine the number of samples that will be required at the end of a month.

Table 10.2 indicates recommended sample sizes for landings at a desired level of accuracy and as a function of data population size.

Table 10.2 Landings Surveys sampling requirements at varying accuracy levels and data population size

Accuracy (%)	90	91	92	93	94	95	96	97	98	99
Data Population size	Safe sample size for BACs
300	29	35	43	54	69	90	120	163	218	274
400	30	36	44	56	73	97	133	188	267	356
500	30	37	45	58	75	102	143	208	308	432
600	30	37	46	59	77	106	150	223	343	505
700	31	37	47	60	79	108	156	236	373	574
800	31	38	47	60	80	110	160	246	400	640
900	31	38	47	61	81	112	164	255	424	703
1000	31	38	48	61	82	114	167	262	445	762
2000	32	39	49	63	85	120	182	302	572	1231
* see notes 3000	32	39	49	64	86	123	188	318	632	1549
4000	32	39	49	64	87	124	191	327	667	1778
5000	32	39	50	64	87	125	192	332	690	1952
6000	32	39	50	65	88	125	194	336	706	2088
7000	32	39	50	65	88	126	195	339	718	2197
8000	32	39	50	65	88	126	195	341	728	2286
9000	32	39	50	65	88	126	196	342	735	2361
10000	32	39	50	65	88	126	196	343	741	2425
15000	32	39	50	65	88	127	197	347	760	2638
20000	32	39	50	65	89	127	198	349	770	2760
25000	32	39	50	65	89	127	198	351	776	2838
30000	32	39	50	65	89	128	199	352	780	2893
35000	32	39	50	65	89	128	199	352	782	2933
40000	32	39	50	65	89	128	199	353	785	2964
45000	32	39	50	65	89	128	199	353	786	2989
50000	32	39	50	65	89	128	199	353	788	3009
> 50000	32	40	50	65	89	128	200	356	800	3201

Notes on the table:

Sample sizes are interpreted as follows:

At a desired accuracy level of 95%, for a data population of 3000 (= 100 boats × 30 days) with an overall CPUE of 5 kg/day:

• 123 landings should have been sampled by the end of the month.
• The safe sample size of 123 corresponding to an accuracy level of 95% will assure that, at worst, the resulting CPUE estimates would be as high as 5.5 or as low as 4.5 kg/day.
• Assuming that sampling occurs during 10 days then about 13 landings should be sampled on each sampling day from the sampling sites in the minor stratum.

At a lower accuracy level of 90%, for the same data population of 3000 with an overall CPUE of 5 kg/day:

• the corresponding safe sample size is 32.
• At worst the overall CPUE would be then be estimated at between 4 and 6 kg/day.
• Assuming that sampling occurs during 10 days then about 4 landings should be sampled on each sampling day from the sampling sites in the minor stratum.

10.3 A GENERAL-PURPOSE FORM

There are five key areas that should form part of a Landings Survey form, although their detailed design depends on their data coverage and intended use. In the example below optional data are shaded.

Part A - Document Identification

Landings Survey forms should always be identified to facilitate the organization and filing of hardcopy information, and cross-referencing between hardcopy forms and computer records.

Usually documents are identified by sequential numbers assigned as either:

• Pre-assigned numbers printed on the Landings Survey forms and distributed to data collectors. These numbers are also input during computer operations; or
• Numbers automatically assigned by the input procedure and then penciled on forms during input.

Part B - Sampling Activity

• Date (essential) - enables automatic grouping of landing data by month (or other period).
• Landing site (essential) - enables automatic grouping of landing data according to stratification criteria.
• Stratum (optional) - facilitates manual grouping of forms.
• Name of recorder (optional) - facilitates cross-checking, queries and evaluation of workload of data collectors. Its use is recommended.

Part C - Fishing operation

• Boat/gear type (essential) - directs grouping of data by boat/gear types as per survey standards.
• Number of Units landing (essential) - Usually it is 1. At times it can be greater than 1 to indicate the number of boats that operated together. It affects sample effort.
• Duration (essential) - specifies the number of days of a fishing trip. It affects sample effort.
• Total landing (optional). It is used when species composition is only a sub-sample of the total.

Part D - Species information

• Landing by species (essential) - Quantity of each species.
• Price (optional) - Highly recommended. When used on a sub-sample basis it provides prices and values by species and hence overall value of production.
• Number of fish in sample (optional). Highly recommended. When used on a sub-sample basis it provides useful data on average fish size, thus allowing various comparisons across gears, seasons and geographical areas.
• Sum of landings by species (essential) - manually computed as a control total to avoid input errors. It is also the basis for raising the sample to total landings when only a proportion was used for species composition.

Part E - Supplementary information and remarks

Information on fishers, fishing units, events occurring at sites, etc.

The data on this form show that:

Trip Duration was 3 days. All catch was sampled (45 kg). Total values of ‘other’ species was recorded as well as prices/kg by species. Associated sample effort is (1 unit) × (3 days), or 3 boat-days.

10.4 CASE STUDIES

The following paragraphs present a number of frequently encountered landing types and discuss the use of catch and effort parameters.

10.4.1 Sub-sampling for large landings

C. Fishing Operation
Boat/gear:	Handline
Number of Units:	1
Duration:	3
Total landing:	450 kg

If the total landing in the example form was 450 kg, this means that only 45 kg were sampled and recorded for species composition. The presence of the Total Landing field is now essential (not optional) because it indicates that catch by species will need to be raised by a factor of 450/45=10.

10.4.2 Boats landing together

C. Fishing Operation
Boat/gear:	Gillnets
Number of Units:	2
Duration:	3
Total landing:	120 kg

In this example two gillnetters operated and landed together. Sample effort is, therefore, 2 × 3 = 6 boat-days.

10.4.3 Fishing units with multiple operations per day

C. Fishing Operation
Boat/gear:	Beach seine
Number of Units:	1
Duration:	0.5
Total landing:	100 kg

In this example a beach seine made two hauls during a day but only one with 100 kg was recorded. Here the effort is set to 0.5 boat-days to indicate that a total of 200 kg would be expected during the whole day.

Such cases can be a problem. They generally apply to fishing units that can operate several times during a day, such as beach seines. The following points should be explained to the data recorder:

1) If the recording is done at the end of the day, the fisherman would specify how many times he operated during the day. This number will be used to calculate trip duration as a fraction of a fishing day.

2) If the recording is done earlier in the day, the fisherman would be asked to specify if more operations would follow and the total number then used to calculate trip duration as a fraction of a fishing day.

3) If there were doubt as to the reliability of the number of operations, it would be preferable to drop the sample from the dataset of sample landings.

If processed or packed fish landings occur occasionally then they should simply be ignored in data collection. However, if they are regular then conversion factors that enable calculation of whole fish weight from processed fish weight need to be developed and used in the estimation of total landings.

10.4.5 Catch sorted by commercial size category

If all landings have been sampled then they should all be combined into a single form as follows:

INPUTS
	Total landing:	100 kg. Effort: 1 boat-day.
	Total large species:	60 kg.
	Large species A:	40 kg.
	Large species B:	20 kg.
	Total small species:	40 kg.
	Small species A:	30 kg.
	Small species B:	10 kg.
RESULTS (on a single form)
	Total landing:	100 kg. Effort: 1 boat-day.
	Species A:	70 kg.
	Species B:	30 kg.

If, on the other hand, sub-samples are used the two alternative procedures are:

1. Using a single form
INPUTS
	Total landing:	1000 kg. Effort: 1 boat-day.
	Total large species:	600 kg.
	Sampled large species:	20 kg.
	Large species A:	15 kg.
	Large species B:	5 kg.
	Total small species:	400 kg.
	Sampled small species:	10 kg.
	Small species A:	6 kg.
	Small species B:	4 kg.
RESULTS (on a single form)
	Total landing:	1000 kg. Effort: 1 boat-day.
	Species A:	15/20 × 600 + 6/10 × 400 = 450 + 240 = 690 kg.
	Species B:	5/20 × 600 + 4/10 × 400 = 150 + 160 = 310 kg.

In this process manual calculations are required for raising catch by species.

2. Using multiple forms
Form 1
	Total landing:	600 kg. Effort: 0.5 boat-days.
	Sampled:	20 kg.
	Large species A:	15 kg.
	Large species B:	5 kg.
Form 2
	Total landing:	400 kg. Effort: 0.5 boat-days.
	Sampled:	10 kg.
	Small species A:	6 kg.
	Small species B:	4 kg.

In the above process raising of catch by species will be done automatically. The effort used in the multiple forms should be split in order to add to the actual 1 boat-day.

10.4.6 Non-fishing boats landing catch

Generally, such landings need not be sampled because they do not provide information on the sample effort associated with the catch.

10.4.7 Migration of fishing units

In theory, the migration of fishing units only affects effort-related surveys. There should be no reason for not sampling landings from boats that operate from sites different from the one being visited. In fact, this is the correct approach at locations that are only landing sites and not homeports. Usually it would be preferable to give priority to local boats and include non-local boats only when the total number of samples is below safety limits.

10.5 TRAINING OF DATA COLLECTORS

Compared to effort-related data collection schemes, Landings Surveys are less demanding in sample size but require more skills on the part of data recorders. Lack of adequate training has direct implications on the reliability of data relating to total landings, catch by species, prices, values, sample effort and fish size. The following major points should be considered in this respect:

Fishing operations

• Identification of boat/gear type
• Cases when samples should or should not be taken
• How to obtain representative samples from boats that are landing
• Effective ways of measuring or eye-estimating total catch
• How to correctly record sample effort data

Species composition

• Species identification
• Effective ways of measuring, or estimating by eye, catch and number of fish by species
• When and how to obtain information on first-sale prices or values

10.6 BRIEFING OF DATA COLLECTORS

Providing data collectors with precise and unambiguous instructions is fundamental for the reliability of the data obtained through Landing Surveys, including:

• Explaining in full detail the recording forms, their use in data collection and their purpose and utility in the overall sampling programme
• Clarifying the issues concerning boat/gear types, such as sequential and concurrent use
• Planning visits to landing sites
• Methods for approaching fishers and village authorities in order to obtain complete and reliable information

10.7 IMPLEMENTATION ASPECTS

Implementation of successful Landings Surveys needs careful planning including:

Pilot phase: small scale testing (e.g. within one or two landing sites) to identify possible design and operational drawbacks and to assess likely timeframes for full-scale operations.

Testing of the entire sampling programme: combine a pilot implementation of the Landings Survey with the entire sampling programme for catch/effort assessment (normally 6-12 months) to ensure both meet their requirements

Evaluation and revision phase: forms and classifications are revised and new instructions issued to data collectors.

Supervision and assistance: Supervision of data collectors is essential for ensuring that data collection is conducted according to planned procedures and schedules. Typical supervisory functions involve:

• Ensuring that recorders visit landing sites according to work schedules and perform their job as instructed
• Checking the way data are sampled and recorded
• Ensuring that recorders are equipped and make effective use of items essential for their job
• Back-reporting of problems relating to movement, timing of visits and duration of stay at sites

Data editing and checking: Data collection forms are usually reviewed prior to processing, including:

• Organizing field documents to facilitate subsequent processing by grouping forms by Month, Minor stratum, Landing site, Boat/gear type, Date, or in any other sequence that will be convenient to the data operators
• Checking that sites and boat/gear types are recorded according to survey standards
• If species are not printed as a standard list, checking the species names recorded by the data collector
• Calculating species catch totals for cross-checking purposes
• Spotting suspiciously high or low values in catch, prices and sample effort data
• Controlling the number of samples for each estimation context (minor stratum, month, boat/gear type)

10.8 FREQUENT PROBLEMS

Timing of field activities

• If few or no landings occur during the allocated time at a landing site, and recorders remain “idle” for long periods, then the work schedule for that site should be reviewed and better use of their time should be made.

Selection of landing sites

• Sites are not representative of all boat/gear types and for certain fisheries no samples can be collected. Sampling sites should be reviewed.
• Sites are not representative of the population of landings. For instance, catch and effort data are atypically high or low. Sampling sites should be reviewed.
• Very important sites do not show individually in the estimates since estimates are produced at minor stratum level. To remedy such reporting problems important landing centers ought themselves to be considered as minor strata.

Concurrent use of fishing gear

• In recording sample effort of a landing it may happen that a boat has operated different gears in one fishing trip. Section 8.4 provides some guidance on this aspect.

SUMMARY This section presented Landings Surveys, thus completing the discussion on methodological and operational aspects of surveys that are involved in the estimation of total catch, including Objectives and basic data coverage Sampling requirements and safe sample size limits achieving minimum accuracy levels Example of a general-purpose form for the recording of landings Case studies. Commonly used sampling techniques in artisanal fisheries Training and briefing of data collectors Implementation aspects Data organization, editing and checking Frequent problems in field operations