基于环境一号卫星的太湖叶绿素a浓度提取

综合环境一号小卫星的CCD数据和同步地面水质监测数据,发现可见光红波段与近红外波段的波段组合与叶绿素a实测浓度存在较高相关性,并以此为基础建立了3个提取水体表层叶绿素a浓度的遥感信息模型.经验证分析,基于近红外波段与红波段比值的模型用于叶绿素a浓度反演提取的精度良好,RMSE达到了6.04mg/m3.将该模型应用于环境一号卫星CCD数据,生成了2009年5~12月共8幅太湖水体叶绿素a浓度分布图,并对其进行了时空分析,结果符合实际,并与以往的研究结果相一致.但模型不适用于水生植被覆盖较多区域叶绿素a浓度估算.

Quantitative estimation of Taihu chlorophyll-a concentration using HJ-1A and 1B CCD imagery

In this study, models estimating water chlorophyll-a concentration have been proposed and calibrated, based on the regression analysis between HJ-1 satellite CCD images and synchronous in-situ water quality monitoring data, ranging from September to December in 2008 on Taihu, Jiangsu Province. The red and near infrared bands of HJ-1A and 1B CCD data, with central wavelength at 660 nm and 830 nm respectively, have shown to be sensitive to the chlorophyll-a concentration. Three models has been built and validated. The results illustrate the high potentials of the models to estimate chlorophyll-a concentration in inland waters for operational applications from satellite HJ-1 satellite CCD data, which will be of immense value for environment monitoring. Among the three models been proposed, the one based on index R830/R660 has the highest estimation precision, 6.04mg/m3. This model has been applied to eight images of HJ-1 satellite CCD, generating chlorophyll-a concentration distributions from May to December 2009.