浑浊Ⅱ类水体叶绿素a浓度遥感反演(Ⅰ):模型的选择

Remote sensing retrieval for chlorophyll-a concentration in turbid case Ⅱ waters(Ⅰ):the optimal model

受高浓度悬浮物的影响,浑浊Ⅱ类水体叶绿素a浓度高精度定量反演一直是研究难点之一.利用2004年到2010年太湖4次实测光谱数据和水质参数,分别建立了两波段、三波段、改进三波段及四波段的叶绿素a估算模型;选择最优模型,利用巢湖2009年的实测数据进行独立验证.结果表明,四波段模型最适合高浑浊水体,线性相关性较好,决定系数R2在0.57～0.95之间,反演精度较高,RMSE在2.39～6.74μg/L之间.

There is a little trouble to retrieve chlorophyll-a concentration（Chl-a） through remotely-sensed imageries in turbid waters,which contains many suspended sediments largely affecting the signature of water-leaving radiance due to phytoplankton pigment.Based on the in situ measurements during the period of 2004~2010,the two-band,three-band,enhanced three-band and four-band models were,respectively,regionally parameterized for the application in Tai Lake.Then the four parameterized models were validated by the in situ data in Chao Lake,almost the same water environment as in Tai Lake.The strongest linear relationship between Chl-a and the four-band model（R2 varying in the range of 0.57 and 0.95,RMSE in the range of 2.39 and 6.74 μg/L） shows that the four-band model is the best for both the Tai Lake and the Chao Lake.