Ecological Provinces In The South China Sea Based On Polynomial Mixture Model And Their Biogeochemical Characteristics Analysis

In the past few decades,despiting in the complex marine environment under physical and biological interactions,oceanographers have recognized the existence of distinct biogeographic provinces in the oceans,which can help us understand the ecological changes better in the ecological area characteristics and corresponding environmental evolution.Based on the 20-year satellite remote sensing chlorophyll data,this paper explores a reasonable and objective method to obtain ecological provinces in the SCS using the unsupervised-polynomial regression model,thenocean color and Bio-Argo data are used to verify the rationality of ecological provinces and analyze their biogeochemical characteristics.Firstly,this paper uses the unsupervised clustering method and polynomial regression model to test and analyze the error data in the SCS,using the remote sensing climatological data to determine the stable parameters.According to the selection of suitable parameter,the above-mentioned cluster analysis algorithm is used to divide the SCS into seven ecological provinces K1-K7:K1 is the coastal high value region,K2 is high value region controlled by mixing,K3 is the transitional region between high value and oligotrophic region,K4 is the winter bloom region in the northwestern part of Luzon,K5 is the southeast part of Indochina,K6 is the northern and the southeastern part of the SCS,K7 is the central sea basin region.Then,combined with satellite dataset including sea surface height,wind field,temperature,precipitation and mixed layer depth reanalysis data,using wavelet analysis to obtain different periods(seasonal,interannual)and trend items,and analysis of the change of physical factors to explain phenomenon,for all kinds of ecological provinces in the SCS.For the seasonal mode,since the SCS is dominated by the monsoon,there are still obvious half a year periodic changes in the the SCS,but the annual circulation mode is still the main body of the seasonal mode,the seasonal dominance factor of the ecological provinces are different.For the interannual mode,the first is the impact of ENSO climate force.when the El Nino event occurred,in the SCS,the overall sea surface temperature anomaly increased,making seawater stratification is obvious,accompanied by monsoon and wind stress curl weakened,the whole sea wind mixing strengthened and upwelling weakened,and nutrient which limits the growth of phytoplankton is lower than normal.For the trend mode,most of the CHL ecological provinces showed an increasing trend in the SCS.For the K1?K3 regions nearing the coastal area,the mixed layer depth is downtrend.The reason is that the trend of coastal upwelling becomes strong and enhances thegrowth of phytoplankton,at the same time the strong coastal upwelling oppresses the mixed layer depth,and makes it decline.For the K4?K7,the oligotrophic regines,which are mainly affected by the wind mixing strengthening,make the nutrient of subsurface entrainment into surface,so the concentration of chlorophyll increase.Finally,we use the Bio-Argo data to verify the rationality of ecological provinces clustered by the remote sensing data.In the K7 region the concentration of chlorophyll is the lowest and increases in the summer and winter accompany with a permanent SCM layer.In the K6 region,where algal blooms exist in winter,SCM layer is destruction so that the rich nutrient and chlorophyll can be taken to into surface.Closer to the K4 region,the algal blooms is even greater.There is a cold eddy and strong upwelling in the northwest sea of Luzon.Although the depth of the mixed layer is inhibited by upwelling effect,it will be easy to produce strong algae bloom,under the combined effect of mixing and upwelling.