<div>Background:&nbsp;Portunus trituberculatus is an important economic crab in the East China Sea. With the increase of tonnage and power of offshore fishing vessels, fishing intensity also increases, which has caused great pressure on P. trituberculatus resources. Protecting P. trituberculatus and achieving sustainable utilisation of resources are urgent problems that need to be solved. Therefore, protection and rational development of P. trituberculatus resources are important to accurately understand its spatial and temporal distribution. 
 
Methods: In this study, the temporal and spatial distribution predictive models of P. trituberculatus in the northern East China Sea were built on the basis of three analysis methods (generalised additive model [GAM], random forest [RF] and artificial neural network [ANN]) using bottom trawl survey data and environmental data from 2006 to 2007. The fitting and prediction performances of these three models were compared. 
 
Result: Season and sea bottom temperature were the most important factors on the distribution of P. trituberculatus. The fitting performance of ANNs was better than those of GAMs and RFs, but its predictive performance was worse than those of GAMs and RFs. Therefore, RFs was the appropriate model in predicting the distribution of P. trituberculatus in the northern East China Sea. The abundance of P. trituberculatus was significantly higher in summer than in other seasons (P&lt;0.01) and generally higher in the northern part of the study area than in the southern part in all seasons.</div>

​Spatial and Temporal Distribution of Portunus trituberculatus in the Northern East China Sea based on Different Modelling Approaches

Spatial and Temporal Distribution of Portunus trituberculatus in the Northern East China Sea based on Different Modelling Approaches