Simulation analyses of abundance estimation of Portunus trituberculatus 

based on set gillnet for different sampling designs

Abstract: The set gillnet has the characteristics of low cost, high selectivity, being easy to use, clean and environment friendly. It is one of the important fishing gears for fishing Portunus trituberculatus in the East China Sea. It is also the main operation method in the pilot waters of Zhejiang Province’s quota fishing management system. In addition, set gillnet is also used for fishery resource monitoring and estimation of fishery resources. Based on the catch of set gillnet, scientific and reasonable assessment of P. trituberculatus resources can not only reduce the cost of assessment, but also be a useful supplement to the investigation of conventional bottom trawl resources.

In order to investigate the effects of different sampling designs of set gillnet on resource estimation of Portunus trituberculatus，an algorithm model is created to estimate the capture probability by different current speeds, moving speed of crab and net length. Based on this algorithm, four sampling survey methods including fixed station sampling, simple random sampling, stratified fixed station sampling and stratified random sampling are simulated. Considering the number of sampling stations of 16 and 25, four seasons, 50, 80 and 150 columns of set gillnet, a total of 72 sets of simulation experiments are performed. The python programming language is used to repeat simulation calculations for each set of experiments 1000 times. The results are as following:

1) In terms of sampling method, fixed station sampling with the increase of the number of set gillnet, the REE and RB values gradually decrease. In the random sampling scheme, the REE and RB values change in a smaller range. When the number of gillnets and the number of sampling stations are the same, the effect of stratified sampling is better than that of nonstratified sampling, and stratified random sampling has the best effect in this resource evaluation simulation experiment; lower resource density and higher ocean current speed make the REE value relatively small, and result in better results. In the case of a certain survey site, stratified sampling is superior to the fixed station and simple random sampling, and the result of resource estimation is more accurate.

2) In terms of the number of sampling stations, for the fixed station sampling and stratified fixed station sampling methods, the results of 25 sampling sites are better than those of the number of 16 sampling sites, showing a trend that the more the number of sites, the more accurate the fishery resource survey results. For the random sampling and stratified random sampling methods, the error values of the scenario of 25 sampling stations are smaller than those of the number of 16 sampling stations, but the error range is not greatly different. It shows that the survey results of random sampling methods tend to be stable, and the influence of the number of sampling stations is not obvious.

3) In terms of the number of set gillnet, besides the scenarios of the spring fixed station sampling, whose REE and RB values decrease with the increase of the number of set gillnet, other sampling scenarios all appear with the trends that the RB value changes from positive to negative with the number increasing, and there is a certain deviation in the estimation results. Therefore, when assessing fishery resources based on set gillnet, larger number of nets may not have better results. It must be comprehensively considered based on the structure and size of the actual sampled species, seasonal temporal and spatial distribution characteristics, as well as the environmental characteristics of sampling sea area.

4) In terms of seasonal characteristics, P. trituberculatus has the characteristics of seasonal temporal and spatial distribution, and seasonal factors have a certain impact on the accuracy of the survey of P. trituberculatus resources. In seasons with high resource density, a stratified sampling design is suitable. The resource density of P. trituberculatus in autumn is high and unevenly distributed, and the ocean current is the slowest among four seasons. Therefore, the error in the estimation results of resources in autumn is also the largest. The resource density of P. trituberculatus in winter is low, and the error in the estimation results of winter resources is smaller than those in other seasons. The estimation results are the closest to the “true value” among all sampling plans.

The use of set gillnet for fishery resource surveys can reduce the damage of fishing gears to the ocean bottom, and plays a positive role in maintaining the marine ecosystem and promoting the sustainable development of fishery resources. However, as a passive fishing gear, set gillnet has strong selectivity, and has certain requirements for the selection of sampling methods and the number of sampling stations. It may overestimate or underestimate the abundance of resources. Therefore, it is necessary to comprehensively consider the impact of other possible factors in application. 

Keywords: crab; northern East China Sea; station quantity; stock assessment