Assessment of estuarine water-quality indicators using MODIS medium-resolution bands: Initial results from Tampa Bay, FL
Assessment of estuarine water-quality indicators using MODIS medium-resolution bands: Initial results from Tampa Bay, FL
复制标题
DOI：
10.1016/j.rse.2004.08.007
复制
发表时间：
2004-11-15
影响因子：
13.5
通讯作者：
Muller-Karger, FE
中科院分区：
文献类型：
--
作者：
Hu, CM;Chen, ZQ;Muller-Karger, FE
以佛罗里达州坦帕湾为例，我们探索了使用 MODIS 中分辨率波段（469、555 和 645 nm 处的 250 和 500 米数据）进行河口监测的潜力。 2003年10月21日至22日的实地调查表明，坦帕湾存在案例II水域，即盐度范围为24-32 psu时，(a)叶绿素浓度(11至23 mg m(-3))，(b)400 nm处有色溶解有机物(CDOM)吸收系数(0.9至2.5 m(-1))，以及(c)总悬浮沉积物浓度(TSS： 2 至 11 mg L(-1)) 通常不共同变化。 CDOM 是唯一与表面盐度呈线性反比关系的成分，尽管这种关系的斜率随海湾内的位置而变化。 MODIS 中分辨率波段虽然是为陆地使用而设计的，但其灵敏度比 Landsat-7/ ETM+ 数据高 4-5 倍，与 CZCS 数据相当或更高。使用多种方法从并发 MODIS 中分辨率数据中获取水成分的天气图。我们发现，各种大气校正算法的应用没有产生显着差异，这主要是由于传感器辐射校准和其他传感器伪影的不确定性。然而，在每个场景都可以被绕开的情况下，成分的原位观察和传感器辐射率之间的简单回归提供了合理的概要图。我们满足了改进传感器校准/表征、大气校正和生物光学算法的需求，以对这些中等分辨率波段进行操作和定量使用。 (C) 2004 Elsevier Inc. 保留所有权利。
Using Tampa Bay, FL as an example, we explored the potential for using MODIS medium-resolution bands (250- and 500-m data at 469-, 555-, and 645-nm) for estuarine monitoring. Field surveys during 21-22 October 2003 showed that Tampa Bay has Case-II waters, in that for the salinity range of 24-32 psu, (a) chlorophyll concentration (11 to 23 mg m(-3)), (b) colored dissolved organic matter (CDOM) absorption coefficient at 400 nm (0.9 to 2.5 m(-1)), and (c) total suspended sediment concentration (TSS: 2 to 11 mg L(-1)) often do not co-vary. CDOM is the only constituent that showed a linear, inverse relationship with surface salinity, although the slope of the relationship changed with location within the bay. The MODIS medium-resolution bands, although designed for land use, are 4-5 times more sensitive than Landsat-7/ ETM+ data and are comparable to or higher than those of CZCS. Several approaches were used to derive synoptic maps of water constituents from concurrent MODIS medium-resolution data. We found that application of various atmospheric-correction algorithms yielded no significant differences, due primarily to uncertainties in the sensor radiometric calibration and other sensor artifacts. However, where each scene could be aroundtruthed, simple regressions between in situ observations of Constituents and at-sensor radiances provided reasonable synoptic maps. We address the need for improvements of sensor calibration/characterization, atmospheric correction, and bio-optical algorithms to make operational and quantitative use of these medium-resolution bands. (C) 2004 Elsevier Inc. All rights reserved.