Research On Coral Reef Bleaching Monitoring Based On Multi-source Remote Sensing Data

Coral reefs are important ecosystems in the tropical ocean,maintaining biodiversity and ecological balance.In recent decades,due to the environmental changes and the dual effects of human beings,coral bleaching events have occurred frequently and large-area coral reefs have been degraded.The global coral reef ecosystem is facing serious threats.Remote-sensing technology can rapidly,extensively and periodically obtain information on coral reefs around the world,especially in remote coral reef areas that humans cannot reach.The advantages of remote sensing technology make it widely used in coral reef monitoring.This paper uses the Xisha Islands area of the South China Sea as a research area,and tries to use multi-source remote sensing data,including optical remote sensing data with penetration characteristics,sea surface temperature data,and carbon dioxide concentration data in the atmosphere over the ocean,to monitor coral reefs for bleaching.The main research contents of this article are as follows:(1)Because there is currently evidence that an abnormal increase in temperature is an important cause of large-scale bleaching events,so first based on the 5-km grid size of SST data has been obtained in recent years and the most recent and most severe one during the El Nino 2014 to Coral reef bleaching in the hottest months of 2016.The specific method is to use the Hotspot and DHW index to obtain the extent and severity of the bleaching of coral reefs.Combining the calculation results of the two indices,we obtained the distribution image of coral reef bleaching warning and made a preliminary judgment on the bleaching of the coral reefs.We found that the coral reefs in the Xisha Islands are likely to be bleached.(2)According to the coral reef bleaching alert area map obtained in(1),use the high-resolution Landsat 8 data for further monitoring.Taking the Panshi Islet and Huaguang reefs as examples,we have obtained coral bleaching areas.Firstly,according to the spectral characteristics of different ground features of coral reefs,a preliminary decision tree algorithm was used to separate the underwater and underwater parts of coral reefs.Secondly,after the water depth of the underwater part of the coral reef distribution area is corrected,an object-oriented classification method is used to classify the reef.The coral reef reef plateau(ie,the distribution area of living coral growth bodies)was mainly extracted,and the extracted results were verified with Google Earth data.The result was more than 80%.Finally,the above technical process was used to extract the coral reef plate areas in the study area before and after El Nino in 2013 and 2016,and then the dynamic changes were obtained by the difference calculation to obtain the coral reef bleaching distribution image.(3)An increase in C02 concentration in the atmosphere will lead to the acidification of seawater and will exacerbate the degradation of coral reefs.It was found that the C02 concentration in the atmosphere near the surface of the study area has increased continuously from 2009 to 2015,which may also be one of the reasons for the deterioration of coral reef health.In China,coral reef health monitoring methods are currently based on in-situ observations based on field measurements.Remote sensing methods are less applied.In China,coral reef health monitoring methods are currently based on in-situ observations based on field measurements.Remote sensing methods are less applied.Due to the limitation of available data in spatial and spectral resolution,the accuracy of information extraction needs to be improved.However,the proposed method has a certain reference value for further theoretical research and has certain practical significance for coral reef management.