1. Objectives
The L-II Survey is applied to a major landing place, through which more than 50 percent of the total fish landings of a district are landed. In view of the importance of the major landing places a sample survey with objective measurement is applied to ensure the accuracy of data on catch landed through such landing places.
Depending on the availability of a fish market at the major landing place, the L-II Survey has two different objectives, as follows:
At a Major Landing Place Where a Fish Market is Available
In this case the record of fish marketing transacted through the fish market is available. However, it often happens that the total fish landings sold through the fish market is smaller than the actual total catch by fishing boats. This is because part of the catch might have been sold directly to fish dealers to reduce the amount of tax to be paid and other parts of the catch might have been consumed during a trip. Also, part of the catch might have been delivered to the owner and crew for home consumption. Therefore, at the major landing places where a fish market is available the L-II Survey aims at determining the quantity of catch which was not sold through the fish market.
At a Major Landing Place Where a Fish Market is Not Available
In this case there is no fish market record available. Therefore the L-II Survey aims at determining the total quantity of catch, which includes not only the catch sold to fish dealers but also the part of the catch consumed during the trip, the part of the catch delivered to the owner or crew of a boat for home consumption, etc.
2. Basic Design of L-II Survey
The L-II Survey aims at estimating catch for each type of fishing gear separately by applying two-stage sampling, using days as a primary sampling unit and trip (boat arrival) as a secondary sampling unit. According to the theory of two-stage sampling, it is well known that, as the number of sample days increases, the accuracy of catch estimates is improved. Whereas, the accuracy of catch estimates does not improve much by increasing the number of sample trips on a sample day. It is, normally recognized that, in two-stage sampling, two or five is a reasonable sample size of secondary sampling unit. This means that, in L-II Survey, two to five sample trips will be good enough for each type of gear.
In general, a sample survey for estimating catch is designed so as to ensure the annual estimate of catch with reasonable precision. In the L-II Survey it has been fixed that a sample day is selected once for every week, which means that 52 sample days will be selected throughout a year. These 52 sample days can be considered as more or less an appropriate sample size in L-II Survey for securing annual estimated catch with a reasonable accuracy. In L-II Survey it is intended to estimate the monthly catch, although the precision may be much lower than that of the annual catch. However, this may do no harm in terms of its usage, since the monthly catch is normally used to know seasonal changes in a fishery only.
3. Type of Fishing Gear to be Covered by Survey
In view of raising the survey efficiency, the coverage of the L-II Survey was limited to major types of gear, which play an important role in the volume of catches of the total catch landed through the major landing place. Thus, some minor fishing gears, whose total catch does not amount to 10 percent of the total fish landings of the major landing place, can be excluded from the sample survey. When the catch of these minor gears is reported to DGF, the catch recorded at the fish market is used as it is.
DFS is responsible for deciding, in consultation with PFS, the type of gear to be covered by L-II survey.
4. Sampling Design on Sample Day
4.1 Number of Sample Trips on Sample Day
In the L-II Survey it has been assumed that a two men team, consisting of one DFS official and one fishery extension officer (MP), is dispatched to the sample major landing place on every sample day. Of these two men, one is responsible for (i) recording the number of trips (boat arrivals) and (ii) selecting systematically a sample trip for each type of fishing gear separately. The other is responsible for observing catch and obtaining some other information of sample trips selected.
Assuming that 20 to 30 minutes are required for observing one sample trip and fish landing lasts for 7 to 8 hours per day, the maximum number of sample trips which can be covered by one enumerator may be 15 per day. Taking into account this maximum daily work load and theoretical sample size of secondary sampling unit as referred in 3 above, the standard number of sample trips on sample day is suggested as follows:
| Number of types of gear to be surveyed on sample day | Standard number of sample trips to be selected for each type of gear |
| 1 – 2 | 5 |
| 3 – 4 | 3 to 4 |
| 5 – 6 | 2 to 3 |
Note: After the new survey was implemented, it has been noted that, in the same major landing place, the number of the types of gear to be covered by the L-II Survey is normally less than two.
4.2 Determination of Sampling Interval and Selection of Sample Trips
The sampling interval is determined, for each type of gear separately, by dividing the expected number of boat arrivals by the number of sample trips. For example, when the number of sample trips is 3 and the expected number of boat arrivals is about 15 for a certain gear, the sampling interval will be 5.
The first sample trip is selected at random within such a sampling interval. For example, when the sampling interval is 5, then the first sample trip is selected from among the first 5 boat arrivals. If the 3rd boat arrival was selected as the first sample trip, then the 2nd and 3rd sample trips will be the 8th and 13th boat arrival.
5. Shifting Survey
When the situation of the major landing place involves one of the following two cases, the shifting survey is applied.
Case 1: Fish landing occurs twice, i.e. in the morning and in the afternoon independently
In the first week the survey is done only for the morning landing and in the second week it is done only for the afternoon landing. Then, this alternative sample survey is continued forward.
Case 2: Fish landing takes place for a very long period of hours
A good example is when a fish landing starts at 10.00 hours and ends at 24.00 hours. In such a case, the entire fish landing hours are divided into two periods e.g. (i) from 10.00 to 17.00 hours (ii) from 17.00 to 24.00 hours. Then, in the first week, the survey is done for the first 7 hours, and in the second week it is done for the second 7 hours. Thereafter, this alternative sample survey is continued.
