| 520 |
|
|
|a In order to determine the favourable oceanographic conditions which influence fish aggregation areas, the integration of remote sensing and GIS technique was applied. This paper aims to classify the spatial distribution and abundance of R. kanagurta in the South China Seas (SCS) using principal component analysis (PCA) and cluster analysis (CA). Remotely-sensed satellite oceanographic data of chlorophyll-a concentration (chl-a), sea surface temperature (SST) and sea surface height (SSH) together with high catch fish data were used to characterize seasonal abundance of the R. kanagurta. PCA identified two principal components that had eigenvalues >1 (PC1 and PC2) which accounted for 59.3% of the cumulative variance. Factor loading in the PCA proved that all environmental variables used in this study; chl-a (PC1), SSH and SST (PC2) had influenced the CPUE of R. kanagurta. Using CA, two clusters of CPUE abundance were identified. In cluster 1, an average CPUE of 350.7 kg/m³ with highest catch were recorded in January, April, May, July and October. Meanwhile, in cluster 2, an average CPUE of 1033.9 kg/m³ with highest catch were recorded in April, May, September and October. Preferred range for fish aggregations showed SST, SSH and chl-a were observed in between 29-31°C, 1.12-1.28 m and 0.24-0.42 mg/m3, respectively. Binary habitat suitability index was used to model the potential aggregation areas. The highest potential fish aggregations areas of R. kanagurta were found located along the coast of Peninsular Malaysia in early and late Southwest monsoon (at accuracy of 83.68% with kappa of 0.7).
|
