Optimization Of Stocking Strategy In Fisheries

Most of the World’s key fisheries resources are fully or over-exploited, with many stocks declining or already collapsed, as a results of comercial fishing. The traditional input- and output-control management approach and restoration of habitat have their own shortages. Stocking fisheries is generally accepted as the most direct solution to improve the declining quality and quantity of resource. Modern stocking programs have been carried out for more than 100 years, and much progress has been researched on most of the fishery technologies related to breeding and releasing. However, the stocking fishery is more difficult and complicate than the exploitation of natural fishery resources. It covers a wide range of aspects from the selection of fish species to be released, feasibility study, and application of various releasing technologies, releasing test at different scales and optimization of releasing strategies to the implementation of stocking. These factors can have great impacts on the success of a stocking program. The process of stocking fishery has not been completely elucidated, and there is a lack of investigation on the changes of the stocking resources before fishing and after releasing, and evaluation of uncertainty resulting from the lack of understanding the above factors is often absent. The existing strategy for the stocking fishery is often based on limited prior research. The methods are not scientifically solid with little effort based on the thorough evaluation of releasing techniques and studies that incorporate the population dynamics and characteristics of released fishery resources, economic benefits and catch quality.In this study, stocking China shrimp(Fenneropenaeus chinensis) was used an example, where the uncertainties in length- weight relationships and von bertalanffy growth function were considered for evaluating their impacts on population dynamics. Two empirical formula based on the growth parameters were used to calculate the natural mortality, and compared to the reported natural mortality based on catch data. The changes in three natural mortalitys over time and their impacts on sex ratio structure were analyzed. I developed a yield per recruit(YPR) model to evaluate the stocking fishery, the factors affecting the yield of released fishery resources and impact of their uncertainty on the amount of resources and catch, as well as on biological reference points(Fmax and F0.1) were analyzed. Based on the developed YPR model, the maximum YPR and maximum economic benefits per recruit(i.e., the minimum input/ maximum output ratio) were used as an index to optimize releasing and fishing strategy, taking also into consideration of factors such as recapture rate and average catch weight. The YPR model was simulated using the age structures and the gender-specific length structures to reveal the response and strategic optimization difference in these factors to different combinations of releasing strategy under the three different structure models.Our results showed that uncertainty did not impact greatly the mean and median of the parameters in the relationship between body length and weight. With increasing uncertainty, the fluctuation of parameter estimation increased, and the stability of the fitting effect of the relationships decreased. With the increase of uncertainty in the length data, the mean and median of the L∞ from the growth equation trended to increase gradually. And the t0 gradually decreased when the length approaches zero, and the impact of uncertainty on the mean and median of K was not remarkable.Our study showed that the values of parameters in the relationship between length and weight were related to estimation methods. With the exponential increase of absolute residuals, parameter a increased, while parameter b was highly uncertain. For the stability of parameter estimates, it was better to use the least absolute deviation method than the least square method to estimate the parameters.Our calculations showed that estimation theory and data remarkably affected the estimation of natural mortality. Although the mating mortality was included in the natural mortality obtained based on the catch data(Ye et al., 1987), the M was smaller than those obtained with other two methods. The M was the greatest with the empirical formula proposed by Chen & Watanabe(1989). The difference was especially remarkable at earlier stage of fish, suggesting that the natural mortality of young fish might has been overestimated at early releasing stages. If the empirical formula based on the growth parameters was used to estimate the natural mortality, sex difference in fish might also impact the result. Using Gislason’s method for natural mortality, the female to male ratio, which was 1.21:1 when captured, reached 2.22:1 after they were released for one year. The ratios estimated with Chen & Watanabe’s formula were 1.15:1 and 4.44:1 at the same time points. These data suggest that when calculating the natural mortality gender-specifically, sex parameter should be included to ensure that the results are biologically meaningful.Four factors influencing mortality, i.e., environment change, predation, intake sea water and illegal fishing, were shown to have similar influence on resource biomass. They had greater impacts on resource reduction when the mortality was high. Under the same model, the difference increased between the mortality factors as the time increased for the released resource and reached the maximum before fishing. Of the models were the amounts of resources estimated based on the age-structure YPR and male length-structure were the greatest and smallest, respectively. The difference was more evident after fishing. At the lower level, the impact of the four factors on cumulative catch was greater than at higher level, with increasing difference over time. At the same level, the cumulative catch yield per unit releasing was the smallest when calculated based on the age-structure YPR. In the beginning of fishing, the cumulative catch yield was slightly higher based on male than on female length-structure YPR models, till the shrimp was grown for 130 to 140 days, then the situation reversed.Natural and mating mortality occurred across the fishing period. At the beginning and end of fishing, the amounts of resources and catches based on Ye et al method were over 2 and near 3 times the amounts estimated using the Gislason’s empirical formula, and about 60 and 87 to100 times of the values estimated with Chen and Watanabe’s method. At the same mating death rate, the impact on catch quantity was greater when the mating occurred earlier. Therefore, we suggest to increase fishing efforts or to start fishing earlier as long as the catch is higher enough to meet the commercial requirements. These would reduce the mating death impact for higher catch quantity.In the absence of uncertainty, the level of mortality factors occurring before fishing did not impact the biological reference points; after the introduction of uncertainty, there was a big difference between F0.1 and Fmax based on the age-structure YPR model. This is true for results obtained with the female length-structure YPR model. In addition to the natural mortality factors, changes in the level of other factors resulted in underestimated F0.1 and Fmax when the age-structure YPR model was used, and overestimated Fmax when the female length-structure YPR model was applied. In addition, the F0.1 means based on the age- and male length-structure YPR models was 10 and 20 times the median based on the female length-structure YPR model, respectively. The F0.1 medians obtained with the three modes were distributed around 5% percentile, suggesting that they shifted left, which was more evident for the F0.1 median obtained with the female length-structure YPR model. Our data also showed that the F0.1 estimates were more conservative based on the age- than length-structure YPR model, while the Fmax estimates were relatively consistent based on the age- and female length-structure YPR models.When the three YPR models were applied to estimate the YPR、EPR、NPR and average catch weight under different releasing and fishing strategies, the indices from different models showed basically a similar trend of change with the values differing. The results of YPR and NPR did not differ between the models nor showed any regularity. EPR was the greatest when calculated based on the male length-structure YPR model, followed by the female length-structure YPR model. For average catch weight, the biggest came from the age-structure YPR model, and the smallest from the male length-structure YPR model.Among the 30 strategies simulated and evaluated, six best combinations with the maximal YPR were releasing on June 10 with the length of 5 cm; releasing with the length of 5 cm and fishing on August 16; releasing with the length of 1 cm(age structure YPR) and 5cm(female length structure YPR), and with the fishing mortality of 0.21; releasing on June 10 and fishing on August 16; releasing on June 10 with the fishing mortality of 0.21; and fishing on August 16 with the fishing mortality of 0.09. As shown in the contour map, the optimal strategies for the combinations of the releasing length for the maximal YPR and fishing effort, and fishing time and fishing effort are the combinations of the longest releasing length and the greatest fishing effort as well as the earliest fishing time and maximal fishing mortality. The difference from the expected results likely resulted from the uncertainty of some parameters during the simulation. Based on the contour map, within the range of the parameters simulated, it is clear that the optimal releasing and fishing strategy for the maximal YPR is to release on June 10 with shrimp of 5 cm long and catch them on August 16 with the fishing mortality of 0.21. For the maximal EPR, the optimal strategy combinations are to release shrimp of 1 cm long between May 20-25; release shrimp of 1 cm long and catch them around August 28; release shrimp of 1 cm long with fishing mortality coefficient of 0.1; release and fish around May 5 and August 13, respectively; release around May 15 with fishing effort of 0.27; and fish on September 13 with fishing effort of 0.27.