| 引用本文: | 杜雨春子,青松,曹萌萌,袁瑞强,顺布日,郝艳玲.乌梁素海沉水植物群落光谱特征及冠层水深影响分析.湖泊科学,2020,32(4):1100-1115. DOI:10.18307/2020.0418 |
| DU Yuchunzi,QING Song,CAO Mengmeng,YUAN Ruiqiang,SHUN Buri,HAO Yanling.Spectral features of submerged aquatic vegetation and water depths impact in Lake Ulansuhai. J. Lake Sci.2020,32(4):1100-1115. DOI:10.18307/2020.0418 |
|
| 摘要: |
| 沉水植物对于改善富营养化水体和重建水生生态系统起着至关重要的作用.应用遥感技术可以实时、大面积监测沉水植物的分布和生长情况，而冠层水深直接影响沉水植物在湖泊、河流中的准确遥感解译.本研究基于实测光谱数据，分析了乌梁素海沉水植物光谱特征，并研究了冠层水深对乌梁素海沉水植物反射光谱的影响，建立了乌梁素海沉水植物冠层水深反演模型.结果表明：1）挺水植物在短波红外1662 nm和2223 nm附近分别有一个反射峰，这是挺水植物区别于沉水植物和漂浮藻类的重要波段；0深度沉水植物（WDC=0）与漂浮藻类的光谱反射率非常接近，但是在绿波段（550~690 nm）有明显差异，因此，可以利用绿波段和短波红外波段的光谱特征来区分挺水植物、沉水植物和漂浮藻类.2）沉水植物群落的光谱反射率随冠层水深的增加而降低，在700~900 nm波段范围内变化最为明显，且在700~735 nm波段附近，沉水植物群落光谱反射率与冠层水深呈显著负相关.3）在建立的单波段/波段比沉水植物冠层水深反演模型中，波段比反演模型要优于单波段反演模型，波段比反演模型的决定系数R2>0.70，均方根误差<13.70 cm，平均相对误差<28%，反演精度较好，适用于10~60 cm沉水植物冠层水深的反演.4）利用波段响应函数，将实测光谱反射率积分到Landsat-8 OLI波段上，建立OLI了冠层水深反演模型，其中，波段比幂函数模型反演效果最好，R2为0.49，均方根误差为18.17 cm，平均相对误差40.05%.可用于精确大气校正后乌梁素海沉水植物冠层水深的反演. |
| 关键词: 沉水植物 光谱反射率 冠层水深 相关分析 回归模型 乌梁素海 |
| DOI：10.18307/2020.0418 |
| 分类号: |
| 基金项目:国家自然科学基金项目（41961057，61461034）、内蒙古自治区高等学校青年科技英才支持计划项目（NJYT-17-B04）和内蒙古自然科学基金项目（2019MS04013）联合资助. |
|
| Spectral features of submerged aquatic vegetation and water depths impact in Lake Ulansuhai |
|
DU Yuchunzi1, QING Song1, CAO Mengmeng1, YUAN Ruiqiang1, SHUN Buri1, HAO Yanling2
|
|
1.College of Geographical Science, Inner Mongolia Normal University, Hohhot 010022, P. R. China;2.School of Ecology and Environment, Inner Mongolia University, Hohhot 010021, P. R. China
|
| Abstract: |
| Submerged plants play an important role in improving eutrophic water and rebuilding aquatic ecosystem. Using remote sensing technology, the distribution and growth of submerged plants can be monitored in real time and in large area. The depth from the water surface to the plant canopy (WDC) directly affects the accurate remote sensing interpretation of submerged plants in lakes and rivers. Based on the measured spectral data, the spectral characteristics of submerged aquatic vegetation in Lake Ulansuhai were analyzed, and the effects of canopy water depth on the reflectance spectra of submerged plants in Lake Ulansuhai were studied. The retrieval model of canopy water depth of submerged plants in Lake Ulansuhai was established. The results showed that 1) There is a reflection peak near the 1662 nm and 2223 nm of the short wave infrared of the emergent plants, which is an important band that distinguishes the emergent plants from the submerged plants and the floating algae. The spectral reflectance of the submerged plants (WDC=0) and the floating algae is very close, but there are obvious differences in the green band (550-690 nm). Therefore, the spectral characteristics of the green band and the short wave infrared band can be used to distinguish emergent plants, submerged plants and floating algae. 2) The spectral reflectance of submerged plant community decreased with the increase of WDC, and it changed most obviously in the range of 700-900 nm, and there was a significant negative correlation between the spectral reflectance of submerged plant community and WDC near the range of 700-735 nm. 3) A single band / band ratio retrieval model of WDC is established, in which the band ratio retrieval model is better than the single band retrieval model. The R2 of the band retrieval model is more than 0.70, the RMSE<13.70 cm, the MRPE<28%, and the retrieval accuracy is high, which is suitable for retrieval of canopy water depth of submerged plants of 10-60 cm. 4) The situ spectral reflectance is integrated into the Landsat-8 OLI band by using the spectral response function, and the retrieval model of the water depth of OLI is established. Among them, the power function model of band ratio has the best retrieval effect, the R2 is 0.49, the RMSE is 18.17 cm, the MRPE is 40.05%, which can be used for the retrieval of the water depth of the submerged plant canopy in Lake Ulansuhai after accurate atmospheric correction. |
| Key words: Submerged aquatic vegetation spectral features water depth correlation analysis regression model Lake Ulansuhai |
