中文摘要：

水体叶绿素a含量的遥感反演是监测水体光学特性、评价水体污染的一个重要指标。本文以FY-3A/MERSI和AQUA/MODIS遥感影像为数据源,结合水体实测的叶绿素a含量,利用两类反射率模型,研究星载数据遥感反演叶绿素a的可行性。研究表明：基于FY-3A/MERSI和AQUA/MODIS可见光-近红外通道的光谱反演模型（R-λ11-R-1λ2）×Rλ3和R-λ11×Rλ3在太湖水体叶绿素a含量反演方面取得了较高的精度。基于MERISI通道的模型反演相关系数R2分别在0.60和0.72左右,基于MODIS通道的模型反演相关系数R2分别在0.55和0.65左右。通过比较分析,决定叶绿素a含量反演精度的主要因素包括两个方面,一是通道位置,即蓝波段和近红外波段是叶绿素a反演的敏感波段;其次,卫星空间分辨率,即较高的空间地面分辨率改善了混合像元。因此,MERSI比MODIS对应模型获得了较高的叶绿素a反演精度。这一结果将有助于FY-3A/MERSI遥感数据在环境监测和水体污染领域的进一步研究,为国产卫星的应用提供一定的参考。

英文摘要：

The Chlorophyll-a（Chl-a） concentrations in water are of great importance to the monitoring of water quality and ecosystem balance.Remote sensing offers a convenient and systematical tool for the observations of water at a long time scale.In this paper,we present a study of Chl-a estimation using the reflectance models（MOD3： R-1λ1-R-1λ2）×Rλ3and MOD2：R-1λ1×Rλ3） derived from the Medium Resolution Spectral Image（MERSI） onboard the newly launched FY-3A satellite and the Moderate Imaging Spectroradiometer（MODIS） onboard the AQUA platforms.Validation studies demonstrated that both models provided reliable estimates of Chl-a concentrations with determination coefficients R2 of 0.72~0.79（MOD2） and 0.52~0.76（MOD3） for MERSI standard band settings.This accuracy is slightly better than that of the MODIS results with R2 of 0.65～0.69 and 0.43~0.70 for MOD2 and MOD3,respectively.Comparison analysis between models and sensors indicated that the blue and near infrared wave ranges are of potential for Chl-a estimation.Besides,the higher spatial resolution of MERSI（250m） may explain the better performance for both models compared to that of MODIS.This research will be helpful for the development of future Chl-a estimation models using the satellite observations.