| 引用本文: | 岳程鹏,李 兴,包龙山,等.基于Landsat8 OLI遥感数据反演乌梁素海浮游植物生物量[J].灌溉排水学报,2020,39(8):122-128. |
| ,et al.基于Landsat8 OLI遥感数据反演乌梁素海浮游植物生物量[J].灌溉排水学报,2020,39(8):122-128. |
|
| 摘要: |
| 【目的】评价乌梁素海多个季度浮游植物生物量反演模型的适用性，为乌梁素海水质治理与改善提供一定的理论依据。【方法】利用乌梁素海Landsat8 OLI遥感数据，结合实测水体的叶绿素a质量浓度数据，采用回归分析，构建乌梁素海浮游植物生物量反演模型，对反演模型精度和普适性进行验证。【结果】春季以b5（近红外）/b4（红光）为自变量的二次多项式回归方程拟合度较差，决定系数为0.463，实测数据与预测数据的均方根误差为6.88 mg/m3；夏季以b5（近红外）/b4 （红光）为自变量的二次多项式回归方程拟合度最优，决定系数为0.816，实测数据与预测数据的均方根误差为3.67 mg/m3；秋季以（b5-b4）/b3为自变量的二次多项式回归方程拟合度适中，决定系数为0.602，实测数据与预测数据的均方根误差为4.63 mg/m3。【结论】同步采集水样与高光谱数据，利用细胞体积转化法计算浮游植物生物量，是提高浮游植物生物量反演模型精度的重要前提条件。 |
| 关键词: 浮游植物；生物量；遥感反演；叶绿素a；乌梁素海 |
| DOI：10.13522/j.cnki.ggps.2019001 |
| 分类号: |
| 基金项目: |
|
| Using Remote Sensing to Estimate Seasonal Variation in Phytoplankton Biomasses in the Lake Wuliangsuhai |
|
YUE Chengpeng, LI Xing, BAO Longshan, WEI Jingting
|
|
1.College of Geography Science, Inner Mongolia Normal University, Hohhot 010022, China; 2.Water-saving and Agriculture Center, Inner Mongolia Normal University, Hohhot 010022, China;
3.Forestry and Grassland Bureau, Ordos 016100, China; 4.Inner Mongolia Environmental Science and Technology Co. Ltd, Hohhot 010011, China
|
| Abstract: |
| 【Background】The Lake Wuliangsuhai is located at Wulateqianqi and is the eighth largest freshwater lake in China. It is also a reservoir receiving discharge of domestic sewages, drainage water from proximal irrigated farmland, and treated industrial wastewater in Hetao Irrigation District. The rapid development in industry and agriculture over the past few years in this region has led water quality of the lake to deteriorate, and the emergence of eutrophication has made the lake worst polluted in Inner Mongolia. Phytoplankton is a primary productivity of the lake and its community structure such as species, quantity and biomass could be used as indicators to evaluate lake water quality. Traditional water quality monitoring systems are costly and laborious, and modern technologies such as remote sensing are able to timely monitor eutrophication emergence and prevent its development at a fraction of the cost of the traditional systems.【Objective】The objective of this paper is to evaluate the feasibility of using satellite imageries to inversely estimate phytoplankton biomass in spring, summer and autumn, respectively, in the Lake Wuliangsuhai.【Method】Water samples were taken 0.5 m below the water surface from points located by GPS. MgCO3 solution was added to the water samples on site prior to being shipped to laboratory for chemical analysis. The mass concentration of chlorophyll a in water samples was measured using a spectrophotometer. The Landsat8 OLI imageries were pre-processed using the ENVI 5.5 software and the derived data, along with the measured chlorophyll a content, were used to establish and test an inverse model to calculate the phytoplankton biomass in the Lake Wuliangsuhai.【Result】The relationship between the content of the chlorophyll a and the satellite imagery index can be described by a quadratic polynomial regression equation, but the accuracy of the model varied with season. The decisive factor and root-mean-square error of the model were 0.463 and 2.71 mg/m3 for spring, 0.86 and 3.67 mg/m3 for summer, and 0.602 and 4.67 mg/m3 for autumn, respectively.【Conclusion】The model established based on satellite imageries to estimate phytoplankton biomass was reasonably accurate and the key to improve its accuracy is the synchronous data and choosing a proper inversion method. |
| Key words: phytoplankton biomass; remote sensing inversion; Chlorophyll a; Lake Wuliangsuhai |
