论文部分内容阅读
本研究以大亚湾为实验区，以陆地卫星ＴＭ数据为信息源，结合表层海水叶绿素浓度实测资料建立模型。在对叶绿素光谱特征及遥感估算叶绿素浓度机理研究基础上，选取了ＴＭＩ—ＴＭ４波段的７５种波段组合为子因素，以叶绿素浓度为母因素，利用灰色系统理论，分析各波段组合与叶绿素浓度之间的关联度。将关联度最大的５种波段组合分别建模，得到５个估算表层海水叶绿素浓度的反演模型。误差分析表明，各模型的最大相对误差在１９％以下，平均绝对相对误差在１１．２％以下，相对标准误差在６．７％以下，模型精度较高。研究表明：（ＴＭ３×ＴＭ４）是估算沿岸海水表层叶绿素浓度的最佳波段组合，采用（ＴＭ３×ＴＭ４）与ＴＭ１、ＴＭ２或ｌｎ（ＴＭ十ＴＭ２）、Ｉｎ（ＴＭ１×ＴＭ２）之比值并不能改善估算精度。
In this study, the Daya Bay experimental area, using land satellite TM data as a source of information, combined with surface chlorophyll concentration measured data to establish a model. Based on the spectral characteristics of chlorophyll and the mechanism of chlorophyll concentration estimation by remote sensing, 75 combinations of TMI-TM4 bands were selected as sub-factors. Based on the chlorophyll concentration as the parent factor and the gray system theory, The degree of correlation between. Five kinds of band combinations with the highest degree of correlation were respectively modeled, and five inversion models for estimating chlorophyll concentration in surface seawater were obtained. Error analysis shows that the maximum relative error of each model is below 19%, the average absolute relative error is below 11.2%, the relative standard error is below 6.7%, and the model accuracy is high. The results show that: (TM3 × TM4) is the best band to estimate chlorophyll concentration in coastal seawater. The ratio of (TM3 × TM4) to TM1, TM2 or In (TM10 TM2), In (TM1 × TM2) Estimated accuracy.