| 基于Landsat 8影像估算新安江水库总悬浮物浓度 |
| 摘要点击 2875 全文点击 1355 投稿时间：2014-07-22 修订日期：2014-08-25 |
| 查看HTML全文
查看全文 查看/发表评论 下载PDF阅读器 |
| 中文关键词 Landsat 8 新安江水库 总悬浮物 经验方法 遥感估算 |
| 英文关键词 Landsat 8 Xin'anjiang Reservoir total suspended matters empirical method remote sensing estimation |
| 作者 | 单位 | E-mail | | 张毅博 | 南京师范大学虚拟地理环境教育部重点实验室, 江苏省地理信息资源开发与利用协同创新中心, 南京 210023
中国科学院南京地理与湖泊研究所湖泊与环境国家重点实验室, 南京 210008 | hbxgzyb@126.com | | 张运林 | 中国科学院南京地理与湖泊研究所湖泊与环境国家重点实验室, 南京 210008 | ylzhang@niglas.ac.cn | | 査勇 | 南京师范大学虚拟地理环境教育部重点实验室, 江苏省地理信息资源开发与利用协同创新中心, 南京 210023 | | | 施坤 | 中国科学院南京地理与湖泊研究所湖泊与环境国家重点实验室, 南京 210008 | | | 周永强 | 中国科学院南京地理与湖泊研究所湖泊与环境国家重点实验室, 南京 210008
中国科学院大学, 北京 100049 | | | 王明珠 | 中国科学院南京地理与湖泊研究所湖泊与环境国家重点实验室, 南京 210008
中国科学院大学, 北京 100049 | |
|
| 中文摘要 |
| 总悬浮物(total suspended matter,TSM)直接决定着水下光场分布,进而影响水体的初级生产力,其浓度也是水质和水环境评价的重要参数之一. 本研究构建了基于Landsat 8影像数据的较为清洁的新安江水库TSM的遥感估算模型,并给出了该水体TSM浓度的空间分布特征. 结果表明, 对该水体TSM浓度较为敏感波段为Landsat 8第二、三和八波段,线性相关的决定系数分别为0.37、0.51和0.42. 然而,以上任何一个波段都无法单独用于准确地提取该区TSM浓度,而利用以上3个波段构建的多元回归模型能够给出较为准确的估算结果,模型决定系数为0.92,平均相对误差为11%,均方根误差为0.16 mg·L-1. 新安江水库TSM浓度整体较低,变化范围为0.04~24.54 mg·L-1,平均浓度为2.19 mg·L-1. 高浓度部分位于湖的边缘区以及一些湖湾枝杈,如: 枫树岭水域、汾口水域、威坪水域、安阳水域、大墅水域、临岐水域等,主要是受入湖河流以及邻近水域采砂活动的影响. 因此研究认为利用Landsat 8数据的3个波段,采用多元回归模型能够较好地估算较清洁水体的TSM浓度. |
| 英文摘要 |
| Total suspended matter(TSM) plays an important role in determining the underwater light climate, which then affects the lake primary production. Therefore, TSM concentration is an important parameter for lake water quality and water environment assessment. This study developed an empirical estimation model and presented the spatial distribution of TSM concentration for the relatively clear Xin'anjiang Reservoir based on the in situ ground data and the matching Landsat 8 data. The results showed that Band 2, Band 3 and Band 8 of Landsat 8 data were the sensitive bands of TSM estimation in Xin'anjiang Reservoir with the linear determination coefficients of 0.37, 0.51 and 0.42, respectively. However, the linear models using Band 2, Band 3 and Band 8 could not give a reasonable and satisfying estimation accuracy. Therefore, a three-band combination estimation model of TSM concentration using Band 2, Band 3 and Band 8 was calibrated and validated to improve the TSM concentration estimation accuracy. The determination coefficient, mean relative error and root mean square error were 0.92, 11% and 0.16 mg·L-1, respectively for the three-band combination model. Overall, the TSM concentration was relatively low in Xin'anjiang Reservoir, ranging from 0.04 to 24.54 mg·L-1 with a mean value of 2.19 mg·L-1. Higher TSM concentrations were distributed in the nearshore zones and small bays such as Fengshuling bay, Fenkou bay, Weiping bay, Anyang bay, Dashu bay and Linqi bay, which were affected by input rivers rainfall and human dredging activity. Therefore, this study demonstrated that the combination of three bands using Landsat 8 data could be used to estimate the TSM concentration in the relatively clear Xin'anjiang Reservoir. |
|
|
|
