中西太平洋鲣鱼围网渔业资源的空间热点和空间异质性分析

Abstract

The western and central Pacific Ocean is one of the most productive purse seine skipjack tuna( Katsuwonus pelamis) fisheries in the world. To understand the structural characteristics of skipjack tuna resources in the western and central Pacific Ocean,based on a Chinese fleet purse seine skipjack tuna catch and effort data from 2005—2009,the datawere summarized by year and by yearly average for 0. 5° × 0. 5° areas,and the spatial autocorrelation and spatial heterogeneity of skipjack tuna resources were calculated by using a spatial statistical method. The hot spot areas of skipjack tuna resources were analyzed with marine environment factors( sea surface temperature and chlorophyll concentration) to determine the reasons for hot spot formation. The methods and conclusions are as follows.( 1) Through conventional statistics calculations we obtained the skewness value,Kurtosis number,coefficient of variation,and global spatial autocorrelation Geary c index of skipjack tuna resources,and found that the density of skipjack tuna resources was low in most of the western and central Pacific,but high density areas could be found in some"hot spot". Fish resources exhibited spatial distribution differences and showed strong concentration distribution spatial patterns. As a whole,there was medium spatial autocorrelation.( 2) Through hotspot analysis,we found that local spatial autocorrelation of skipjack tuna resources was strong. There was more than one"hot spot"and"cold spot"that were statistically significant.( 3) When studying the spatial heterogeneity and directional variation of skipjack tuna resources by using geostatistical methods,we obtained the semivariograms parameters and best-fitting semivariogram models. The best-fitting semivariogram model was the sphere model. The average spatial correlation distance( the geostatistical range) was about 1000 km. Approximately 50% variation was explained by spatial autocorrelation,and the rate of spatial autocorrelation was moderate. In directional variation,the structure of the south-north direction was stronger than that of the east-west direction. The rate of structure variation was67% in the south-north direction and 49% in the east-west direction. This conclusion is consistent with marine environmental structure in which spatial distribution has better regularity and structure in the south-north direction compared with the east-west direction. The fractal dimension value of different directions was from 1. 876 to 1. 9; therefore,spatial autocorrelation was weak.( 4) This paper regards skipjack tuna resource hotspot areas as regional fishing grounds.Respectively mapped fishing grounds overlapped with sea surface temperature and chlorophyll concentration. The fishing grounds of skipjack tuna resources in the western and central Pacific Ocean were divided into two different local fishing grounds: cold-tongue and warm-pool ground. The cold-tongue fishing ground,which is caused by the Middle East Pacific equatorial upwelling,provided more abundant primary productivity in the frontal zone,and hence more food for fish. The warm-pool fishing ground is nearer to the mainland and islands where coastal upwelling system can provide primary productivity.( 5) Finally,we compared the method of hot spot analysis with that of fishing ground gravity and habitat suitability index,and identified advantages and disadvantages of those methods. Future studies will involve using the space residual model to solve spatial autocorrelation problems.