基于MODIS数据的松嫩平原西部芦苇湿地地上生物量遥感估算

芦苇作为湿地生态系统中重要的群落类型,其地上生物量是衡量湿地生态系统质量的关键指标。应用面向对象的土地覆盖分类技术,基于多季相Landsat8 OLI遥感数据,提取松嫩平原西部芦苇湿地分布信息;依托野外实测芦苇地上生物量数据(AGB)和同期MODIS数据源的NDVI、EVI、RVI、MSAVI和WDVI 5种光谱植被指数,探讨不同光谱植被指数对芦苇AGB的敏感性,进而构建松嫩平原西部芦苇AGB遥感估算最优模型,并进行芦苇AGB遥感反演及空间格局分析。结果表明:2014年松嫩平原西部地区芦苇总面积为1653 km~2,其中扎龙湿地自然保护区内芦苇分布面积最大(1178km~2),占区域芦苇总面积的71.3%;所选取的5种植被指数均与芦苇AGB呈极显著正相关(P0.01),基于EVI构建的指数曲线模型为松嫩平原西部芦苇AGB反演的最优模型(R2=0.55)。研究区芦苇平均AGB为372.1g/m~2,AGB总量为6.14×105t,其中扎龙湿地自然保护区内芦苇AGB总量为4.38×105t;各保护区芦苇平均AGB由大到小依次为:向海保护区(469.7 g/m~2)大布苏保护区(454.1 g/m~2)莫莫格保护区(373.0 g/m~2)扎龙保护区(372.4 g/m~2)查干湖保护区(369.8 g/m~2);松嫩平原西部芦苇AGB总体呈现南高北低的分布格局,将为湿地生态系统管理与保护及芦苇资源的合理利用提供科学依据。

Remote sensing estimation of the aboveground biomass of reed wetland in the Western Songnen Plain, China, based on MODIS data

Reed is one of the important community types in wetland ecosystems, and its aboveground biomass (AGB) is a key index characterizing the quality of these ecosystems. In this study, the spatial distribution of reed wetland in the Western Songnen Plain was mapped using the object-oriented classification method combined with multi-seasonal Landsat8 OLI remote sensing data. Field observed reed aboveground biomass data and five spectral vegetation indices, i.e., normalized differential vegetation index (NDVI), ratio vegetation index (RVI), enhanced vegetation index (EVI), modified soil adjusted vegetation index (MSAVI), and weighted difference vegetation index (WDVI), which were derived from moderate-resolution imaging spectroradiometer (MODIS) products, were used to determine the sensitivity of different spectral vegetation indices to reed AGB and further to develop an optimal remote sensing model for reed AGB estimation and inversion. The results showed that, the total area of reeds in 2014 was 1653 km² in the Western Songnen Plain, and Zhalong Wetland Nature Reserve had the largest reed area (1178 km²) among the five wetland natural reserves. Additionally, strong correlations were found between each of the five vegetation indices and reed AGB (P < 0.01). The exponential curve model established using EVI was the optimal model (R² = 0.55) for reed AGB estimation and inversion. In the Western Songnen Plain, the mean reed AGB based on the exponential curve model was estimated to be 372.1 g/m² and the total reed AGB to be 6.14×10⁵ t. The maximum reed AGB was observed in Zhalong Nature Reserve with a value of 4.38×10⁵ t. The averaged reed AGB for each of the five wetland natural reserves can be listed in a decreasing order as Xianghai Natural Reserve (469.7 g/m²) > Dabusu Natural Reserve (454.1 g/m²) > Momoge Natural Reserve (373.0 g/m²) > Zhalong Natural Reserve (372.4 g/m²) > Chagan Lake Natural Reserve (369.8 g/m²). Higher AGB values were found in the southern area than in the northern. Results from this study could contribute to the protection and management of reed wetland, and the utilization of reed resource.