The Estimation Of Mangrove Biomass And Carbon Storage Using Remote Sensing Data In Beibu Gulf Coast

Mangrove is located in tides swamp, distributed in intertidal zone around seashore,so sample survey is too difficult for measure trees for mangrove that the study on mangrove biomass is limited severely. In order to improve the efficiency of the study on mangrove biomass, by Landsat TM imagines,with large number of sample survey about mangrove biomass in Beibu Gulf coast, this paper studied the estimation models of aboveground biomass and carbon storage of five kinds of mangrove community (Rhizophora stylosa,Bruguiera gymnoihiza, Vicennia mariana, Aegiceras corniculatum, Kandelia candel) and mixed species community.In this paper, multiple regression analysis and KNN(K-Nearest Neighbor) methods were used, and the two methods were compared and evaluated.The results were as follows:(1) There is a strong relationship between mangrove biomass and texture features,however, a weak relationship between mangrove biomass and variables usually used in biomass estimation by remote sensing, such as NDVI,TM3,TM4.(2) This paper build the aboveground biomass regression models of Rhizophora stylosa, Bruguiera gymnoihiza, Vicennia mariana, Aegiceras corniculatum, mixed species,but not Kandelia candel forest. Though models have certain errors, provide a method for mangrove biomass estimation using remote sensing data. The study is of great practical significance.(3) In pixel scale, the multiple regression analysis is better than KNN in biomass estimation using remote sensing data.(4)Estimatiing mangrove biomass by KNN methods, the accuracy increases by the scale, and K=10 is better than K=5 in the accuracy.(5) The order of the accuracy of biomass estimation by both the multiple regression analysis and KNN(K-Nearest Neighbor) is that:Rhizophora stylosa and Bruguiera gymnoihiza>Vicennia mariana>Aegiceras corniculatum>mixed species, while multiple regression analysis Rhizophora stylosa is slightly better than Bruguiera gymnoihiza, while KNN, Bruguiera gymnoihiza is slightly better than Rhizophora stylosa. (6) The conversion coefficients between biomass and carbon storage of mangroves are:Bruguiera gymnoihiza:47.47%,Rhizophora stylosa:43.18%, Vicennia mariana: 41.47%, Aegiceras corniculatum:41.84%,Kandelia candel:42.32%,mixed species 43.26%。The texture features extracted remote sensing and KNN methods rarely used in forest biomass estimation, are used in the study on estimation mangrove biomass and carbon storage firstly, and getting certain findings. The estimation models of mangrove biomass and carbon storage provide a convenient and efficient method for mangrove ecosystem field.