Evaluation Of Coral Reef Ecological Resistance In Xisha Islands Based On Remote Sensing Technology

Coral reef ecosystems play an important role in the maintenance of marine biodiversity,the supply of marine biological resources and national defense construction.However,due to the intensified pressure on the global natural environment and regional human activities,the coral reef ecosystem in the South China Sea and even the world has been severely degraded,and measures to protect,restore and manage coral reefs are urgently needed.Coral reef ecological resistance evaluation can provide scientific reference for coral reef protection,restoration and management.At present,in the study of coral reef ecological resistance,the traditional field survey method has the problem of discontinuous ecological resistance data.Remote sensing is an effective means to obtain large-scale continuous ecological resistance data,but there are still problems that cannot directly obtain most ecological resistance data.Therefore,from the perspective of coral community structure and composition,this study selects four ecological resistance indicators:coral cover,stress-tolerant coral cover,coral species diversity and coral species evenness to construct a remote sensing evaluation system for ecological resistance in the Xisha Islands.Topographic,geomorphological and substratum zoning factors affecting ecological resistance indicators were obtained by remote sensing technology.The method of combining remote sensing numerical modeling and remote sensing spatial division was used to indirectly obtain the spatial distribution data of each index of ecological resistance in the study area,and further construct a normalized evaluation model of ecological resistance to evaluate the ecological resistance of the study area.The main conclusions are as follows:(1)Combining the"Ice,Cloud and Land Elevation 2"(ICESat-2)data with the Gaofen-1(GF-1)remote sensing images,this study proposes a multispectral remote sensing bathymetry inversion method for large-scale coral reef areas under geomorphological zoning.The results show that the single-band and band ratio binomial model is used for reef slope and deep lagoon geomorphology,and the multi-band ratio regression model is used for reef flat,shallow lagoon and point reef geomorphology,which can make the water depth remote sensing inversion in the geomorphology unit achieve the best effect.The correlation coefficient(R~2)between the model inversion bathymetry data and the field survey bathymetry data is 0.96;the comprehensive verification results of the field survey bathymetry data and the ICESat-2 deep water area data show that the model root mean square error(RMSE)is 1.33m,and the mean absolute error(MAE)was 1.06m and the mean relative error(MRE)was 16.76%.The geomorphology-based active and passive remote sensing bathymetry inversion method significantly improves the accuracy of the medium-resolution remote sensing bathymetry inversion model in coral reef areas,and can provide method support for high-precision bathymetry inversion in large-scale coral reef regions.(2)A DEM model was constructed based on the water depth data in the study area,and five topographic factors were obtained:water depth,bathymetry position index,roughness,profile slope and slope aspect.Based on the above factors,an indirect prediction model for the distribution of stress-tolerant coral cover,coral species diversity and coral species evenness index data was established in the reef slope area.The validation correlation coefficients(R~2)of the three ecological resistance index models were 0.64,0.46,and 0.42,and the RMSEs were 0.04,0.38,and 0.10,respectively.The spatial division of the study area was carried out by using remote sensing geomorphology and sediment,and combined with the average data of field surveys in each division,the spatially distributed data of coral cover in the study area,stress-tolerant coral cover in the non-reef slope area,coral species diversity and coral species evenness was obtained.This method of combining remote sensing spatial division with remote sensing numerical modeling can obtain effective ecological resistance index distribution data.(3)The weights of coral cover,stress-tolerant coral cover,coral species diversity and coral species evenness were obtained by the AHP combined with the coefficient of variation method,which were 0.25,0.35,0.25 and 0.15,respectively.It is suggested that when implementing specific work on coral reef protection and restoration in the Xisha Islands,priority should be given to the index of stress-tolerant coral cover.(4)The results of ecological resistance evaluation showed that reef slope area and reef flat coral prosperous area were high-level ecological resistance areas.Among them,the areas with the highest ecological resistance are distributed in the northern and northeastern reef slope areas of coral reefs,which are the best areas for protection and restoration work in this area.The point reef area and the sub-prosperous area of reef flat corals are medium-high ecological resistance areas.The reef flat coral sparse areas and lagoon coral areas are middle-low-level ecological resistance areas.The coral-free areas are low ecological resistance areas.The above findings can provide decision support for coral reef management priorities in the Xisha Islands.