5.1.1.1 Powered boats
For fishery administration, a very accurate number of powered boats is particularly required. To meet such a requirement, a powered-boat card, which can be used for ten years, is prepared for every powered boat. In the first year of the new survey these powered.. boat cards were prepared based on a list of powered boats prepared for the 1974 DGF fishery census (see survey Form SL-1).
Thereafter, at the end of each year, the card is brought up-to-date and the type of fishing gear employed during the past one year is recorded, based on an interview with the respective fishermen.
5.1.1.2 Nonpowered boats
At the end of each year a complete list of fishing households, together with the number of boats in use, is prepared with the use of Survey Form SL-2 for every sample village which was selected for the L-III Survey. During the survey year the number of fishing units by type of fishing gear is surveyed on a quarterly basis, mainly for the estimation of catch and the number of trips for the L-III Survey.
At the end of each year, the list is brought up-to-date and the number of fishing establishments, the number of nonpowered boat by sizes and the number of fishing units by type of gear are counted from SL-2.
5.1.1.3 Estimation at district level
The date for three survey items which were counted from SL-1 and SL-2 are transferred to Estimation Form EL-1. Then, estimation of these items at district level is done on this form.
This survey is divided into three surveys, according to the difference in the nature of survey objects.
5.1.2.1 L-I survey
This survey is applied only to some 30 fishing companies, whose catches are mainly for export. As these companies keep a good record of fishing operations, only Survey Form SL-3 is delivered to the companies requesting them to fill them out. The list of such companies was prepared by the DGF.
5.1.2.2 L-II Survey
This survey is applied to major fish landing places, where a large quantity of the catch is landed by trawler, purse seiner, drift gillnetter, etc. The selection of such major landing places is illustrated in Appendix 7. At present, throughout the country some 100 major landing places are selected for the L-II Survey.
Although there are some 5 000 fish landing places (marine fishing villages) throughout the country, it is assumed that about a half of the total marine catches are landed through such major landing places. With the progress of fishing boat mechanization, the role of such major landing places will become more important in terms of the fish landed. For this reason, a sample survey with objective measurement is applied to ensure the accuracy of catch data collected.
Two-stage sampling, using a day as a primary sampling unit, and a trip (boat arrival) as a secondary sampling unit is applied. In principle, one day is selected from each week as a sample day. On a sample day all boat arrivals are counted and at the same time a sample of trips are selected for the measurement of catch (See Survey Forms 4 and 5).
The numbers of trips and catch are estimated on a monthly basis using Estimation Form EL-2. When a record of catch sold to a fish market is available, ratio estimation is applied. When there is no such record available, simple estimation is applied.
The sample design of L-II survey is illustrated in more details in Appendix 8. How the L-II Survey is effective in calculating catch estimates is shown in Appendix 10.
5.1.2.3 L-III survey
This survey is applied to all marine fishing villages, excluding major landing places for L-II Survey. The nature of the marine fishery covered by this survey is the small-scale fishery, which is undertaken by a huge number of small fishing households with nonpowered boats or without boats. The catches are landed all along the sea coast and such landing places are not always easily accessible.
Owing to such a difficulty as referred to above, a sample survey with subjective measurements is applied, although such a survey method is not always very appropriate in terms of accuracy of the calculated estimates. The 1973 CBS agriculture census provided, through its complete enumeration method, a complete list of the marine fishing villages together with the number of fishing households for each village. However, for areas which were not covered by the CBS agriculture census, a similar sampling frame was obtained through the 1974 DGF marine village survey.
L-III Survey is based on “cluster sampling”, using the marine village as a sampling unit. “Probability proportional to the size (PPS) of marine fishing village” was used for choosing the sample villages. Specifically, for the collection of the sample villages, the names of all marine villages and the number of fishing households were at first transferred to Form KL, which is seen in Appendix 6. Thereafter, the number of fishing households was accumulated in the order of marine villages listed in the Form. Then, the sample villages were chosen by means of PPS. Further details on this procedure are given in Appendix 7.
L-III Survey is a quarterly survey and actual field enumeration is confined to the sample fishing villages. For every sample village, a complete list of fishing households is prepared using Survey Form SL-2, through which the number of fishing units by type of gear is sought on a quarterly basis. On the other hand, with the use of Survey Form SL-6, average catch per trip and average number of trips per quarter is obtained by means of an interview with some sample households. Thus, total catch per quarter for each type of gear is estimated as a product of total number of fishing units, average number of trips per quarter and average catch per trip. Further details on this procedure are given in Appendix 9.
To estimate the number of trips and catch throughout a district, Estimation Form EL-3 is used. As seen in EL-3, the method of estimating is a ratio estimation, and such a ratio is obtained by dividing the total number of fishing households throughout a district by the number of fishing households through all sample villages. All these figures are obtainable from Form KL.
From the viewpoint of the users of such data, the statistics do not always require very high accuracy. What is more important is to keep a close link between total catch in live weight, catch disposed of through various channels and, subsequently, the quantity of fisheries commodities produced. When a survey on the disposition of catch and quantity of fisheries commodities produced is done independently from a fishery production survey, a great discrepancy among these three figures often occurs. Therefore, the idea of combining this survey with the fishery production survey is to eliminate any such unreasonable discrepancy.
Out of the total catch landed, what percentage is disposed of for fish processing is roughly known by the district fishery officers. Therefore, catch disposed through various channels is estimated, based on such already known information. Then the quantity of fisheries commodities produced is estimated by applying a relevant conversion factor to the catch disposed through a certain channel (See Estimation Form EL-6 and Reporting Form LL-5).
A sampling frame for the inland water fishery was established with the use of Form K, which is seen in Appendix 6. This ad hoc frame survey was undertaken in the middle of 1976 with the help of the heads of villages.
Survey methods applied to the inland open water fishery are exactly the same as those developed for the marine fishery. However, major fish landing places, such as are seen in the marine fishery, are rarely seen in this fishery, so that in the majority of districts only the L-III Survey is applied.
Since species caught by the inland open water fishery are quite different from those of the marine fishery, different reporting forms are used. (See Reporting Form LP-1, LP-2 and LP-3 in Appendix 6).
A sampling frame for brackishwater culture was established through the complete enumeration of the 1973 CBS agriculture census, whereas, a sampling frame for fresh water culture was established through the ad hoc frame survey, as mentioned above.
In principle, the survey method for culture is the same as the L-III Survey for the marine fishery. Therefore, the survey for culture is a sample survey with subjective measurement, and sample villages are selected by means of PPS (See Form KB in Appendix 6). For the survey of inventory items like the number of culture households, area of fish pond, etc., a complete list of culture households is prepared for every sample village. The survey for such inventory items is done only once a year at the end of each year (See Survey Form SB-1).
The survey for yield is done on quarterly basis with the use of Survey Form SB-2. The survey is confined to sample villages, within which ten sample culture households are randomly chosen, based on Survey Form SB-1. Then an interview is conducted with these sample households.
Estimation of inventory statistics and yield are done with the use of Estimation Form EB-1 and EB-2, respectively. Reporting Forms LB-1 and LB-2 are used for reporting statistics to the DGF. Unlike the marine fisheries and the inland open water fishery, the yield from culture is seldom used for fish processing, therefore, no estimation is done for disposition of the catch.
