Ocean Salinity Mapping Based on Adaptive Ant Colony Algorithm to Optimize Kriging Interpolation

In order to solve the problem of scarcity of marine salinity data, this paper uses the adaptive ant colony optimization Kriging interpolation method to fill the data gap. Kriging interpolation is a widely used optimal interpolation technique. Its parameter selection of the mutation fitting function directly affects the accuracy of interpolation. The ant colony algorithm adopts a global search strategy, and it can effectively explore the large search space and find the global optimal solution of the parameters. In order to improve the optimization accuracy, we introduce the adaptive ant colony algorithm in this paper. It can dynamically adjust the pheromone importance factor, pheromone volatilization rate and ant number according to the current search state during the search process, so as to obtain more accurate interpolation results. In data interpolation simulation experiments, the ocean salinity in the South China Sea and some areas of the Yellow Sea in 2021 was selected as sample data, and the mean absolute error (MAE) and root mean square error (RMSE) were used as performance evaluation criteria. We used the spline interpolation method, the inverse distance interpolation method, the ordinary Kriging interpolation method, the particle swarm optimization Kriging interpolation method, the genetic algorithm optimization Kriging interpolation method, and the adaptive ant colony optimization Kriging interpolation algorithm in the experiments. Experimental results show that the adaptive ant colony optimization Kriging interpolation method exhibits the lowest RMSE and MAE, and presents a certain superiority in terms of prediction accuracy. The results of this study will help to fill the gap in ocean salinity observations, and will be crucial for unraveling scientific questions such as ocean dynamics, the water cycle, climate change, and global energy balance.