基于植被指数的海南岛霸王岭热带森林地上生物量空间分布模拟

热带森林在全球碳循环方面扮演着重要的角色, 预测其生物量分布可以加深对碳循环过程的理解。然而目前基于植被指数模拟技术进行热带森林生物量分布的研究报道较少。该文以海南岛霸王岭林区热带森林为研究对象, 在基于遥感影像和135个公里网格样地调查的基础上, 分别选取归一化差异植被指数(NDVI)、短红外湿度植被指数(MVI5)、中红外湿度植被指数(MVI7)和比值植被指数(RVI)与总物种生物量、顶极种生物量和先锋种生物量做相关分析, 并利用逐步线性回归分析分别构建了基于植被指数的生物量回归模型; 利用残差图对模型的有效性进行检验。结果表明, MVI7和MVI5与总物种和顶极种生物量关系显著, 而NDVI和RVI对先锋种生物量具有较好的指示作用; 总物种、顶极种和先锋种生物量预测精度较高的区域分别占总面积的69.24%、73.98%和88.08%, 表明3个生物量模型均具有较好的拟合精度; 模拟结果表明总物种和顶极种生物量主要集中于研究区中部、北部和西南部区域, 而先锋种生物量无明显的分布规律, 是不均衡地散布于整个研究区域, 反映了群落组成结构、干扰历史、地形及气候因素等的影响。

MODELLING THE SPATIAL DISTRIBUTION OF ABOVEGROUND BIOMASS BASED ON VEGETATION INDEX IN A TROPICAL FOREST IN BAWANGLING, HAINAN ISLAND, SOUTH CHINA

Aims The biomass of tropical forests plays an important role in the global carbon cycle; however, the distribution of tropical forest biomass based on vegetation index is seldom explored. Our objectives were to evaluate relationships between biomass and vegetation indices and to determine the spatial distribution of the aboveground biomass of tropical forest in Bawangling, Hainan Island, South China. Methods Using measurements of forest biomass from 135 sample plots distributed over the study area, we correlated four vegetation indices (normalized difference vegetation index (NDVI), moisture vegetation index using Landsat’s 5 (MVI5), moisture vegetation index using Landsat’s band 7 (MVI7) and ratio vegetation index (RVI)) with aboveground biomass (total biomass, climax species biomass and pioneer species biomass) using the Pearson correlation method. We also developed models describing the relationships between forest aboveground biomass and vegetation indices using stepwise linear re-gression analysis. Three maps of biomass components were produced using the developed models, and residual maps were used to test the validity of the models. Important findings MVI7 and MVI5 are most effective for total biomass and climax species biomass, whereas NDVI and RVI seem to be good indices of pioneer species biomass. The strongly predictive percent areas for total species, climax species and pioneer species biomass models were 69.24, 73.98 and 88.08, respectively. Simulated biomasses of total species and climax species were distributed in the center, north and southwest parts of the study area; however, simulated biomass of pioneer species was scattered.