Habitat discriminant analysis of short-neck clam (Ruditapes philippinarum) and Japanese mud shrimp (Upogebia major) on the tidal flats, South Korea

Abstract

A short-neck clam (Ruditapes philippinarum) is one of the most important commercial shellfish. The amount of the shellfish production has been severely reduced due to the unexpected invasion of the Upogebia major in some Korean tidal flats. Thus, it is highly required to develop a controlling technique for the invasion of the U. major. In this study, the diverse simulations of the habitat classification are carried out using the available habitat data of the U. major and R. philippinarum as part of the basic research. The data are the abundance of U. major and R. philippinarum, sediment grain size, and exposure time monitored in the central west coast of South Korea. In order to apply the discriminant machine learning models, the tidal-flat habitat was classified into four types based on the abundance Type 1 - U. major dominant, Type 2 - U. major and R. philippinarum mixed, Type 3 - No U. major and R. philippinarum, and Type 4 - R. philippinarum dominant. The models used in the discriminant analysis are the uniform range model (the simplest model), k-Nearest Neighbor (kNN) model, Support Vector Machine (SVM) model, and the Artificial Neural Network (ANN) model. Based on the simulation results, it is shown that the classification accuracies (in the calibration stage) of these models are 52%, 66%, 89%, and 95%, respectively in the order named above. (중략) Thus, it is highly required to develop a controlling technique for the invasion of the U. major. In this study, the diverse simulations of the habitat classification are carried out using the available habitat data of the U. major and R. philippinarum as part of the basic research. The data are the abundance of U. major and R. philippinarum, sediment grain size, and exposure time monitored in the central west coast of South Korea. In order to apply the discriminant machine learning models, the tidal-flat habitat was classified into four types based on the abundance Type 1 - U. major dominant, Type 2 - U. major and R. philippinarum mixed, Type 3 - No U. major and R. philippinarum, and Type 4 - R. philippinarum dominant. The models used in the discriminant analysis are the uniform range model (the simplest model), k-Nearest Neighbor (kNN) model, Support Vector Machine (SVM) model, and the Artificial Neural Network (ANN) model. Based on the simulation results, it is shown that the classification accuracies (in the calibration stage) of these models are 52%, 66%, 89%, and 95%, respectively in the order named above. (중략)