基于优化模型的莱州湾三疣梭子蟹(Portunus trituberculatus)栖息地适宜性研究

OPTIMIZATION OF HABITAT SUITABILITY MODEL FOR PORTUNUS TRITUBERCULATUS IN LAIZHOU BAY

三疣梭子蟹(Portunus trituberculatus)是莱州湾最重要的经济蟹类,其资源丰度受栖息环境影响显著,为了解不同的环境因子对其栖息地分布的影响,根据2010~2020年夏季底拖网调查数据,研究莱州湾三疣梭子蟹栖息地适宜性及其影响因子。利用广义可加模型(generalized additive models,GAM)选取变量因子,通过提升回归树模型(boosting regression tree,BRT)对因子进行权重分析,构建了4种栖息地适宜性指数(habitat suitability index,HSI)模型,并通过实测值和预测值的Pearson检验,对模型比较和验证。结果表明,GAM和BRT优化的HSI模型好于其他3种模型(未优化模型,GAM优化HSI模型,BRT优化HSI模型),以生物量表征资源丰度的模型好于尾数表征资源丰度的模型,算术平均法构建的HSI模型的整体预测准确率和相关系数高于几何平均法,对莱州湾三疣梭子蟹栖息地有较强的预测能力。底层水温、底层盐度和水深对三疣梭子蟹栖息地影响较大,夏季栖息地适宜性较高的海域(HSI>0.7)主要分布在莱州湾南部、东南部和东北部海域,中部和西南部海域分布较少。研究结果为莱州湾三疣梭子蟹增殖放流工作提供了科学依据。

Portunus trituberculatus is an important economic crab in Laizhou Bay, Shandong. The abundance of P.trituberculatus is significantly affected by the environmental factors. To understand the effects of different environmental factors on the habitat distribution, the habitat suitability of P. trituberculatus in Laizhou Bay and its influencing factors were studied based on samples collected by bottom trawling from 2010 to 2020. Generalized additive models(GAMs) were used to determine which environmental variables should be included in the HSI(habitat suitability index) models. The relative contribution(%) was weighted in the HSI model to the total deviation explained by the boosted regression tree(BRT). Four HSI models(non-optimized model, BRT informed HSI model, GAM informed HSI model, and BRT+GAM informed HSI model) were constructed, and the Pearson test on measured and predicted values were used to compare and verify the models. Results show that among the four models, the BRT informed and the GAM informed HIS models showed the best performance. The model in which biomass was selected to indicate P. trituberculatus abundance showed better performance. The model of biomass-based resource abundance is better than that of the mantissa-based. The overall prediction accuracy and correlation coefficient of HSI model constructed by arithmetic average method are higher than those of geometric average method, showing a strong ability to predict the habitat of P. trituberculatus in Laizhou Bay.Bottom water temperature, bottom salinity, and water depth have great influence on the habitat of P. trituberculatus. The area of high habitat suitability(HSI>0.7) of P. trituberculatus distributed mainly in the southern, southeastern, and northeastern waters of the bay, while less in the central and southwestern waters. This study provided a scientific basis for the stock enhancement of P. trituberculatus.