School of Ocean Sciences, China University of Geosciences, Beijing, 100083, China.
Environ Monit Assess. 2013 Mar;185(3):2243-55. doi: 10.1007/s10661-012-2705-y. Epub 2012 Jun 19.
In this study, an improved Moderate-Resolution Imaging Spectroradiometer (MODIS) ocean chlorophyll-a (chla) 3 model (IOC3M) algorithm was developed as a substitute for the MODIS global chla concentration estimation algorithm, OC3M, to estimate chla concentrations in waters with high suspended sediment concentrations, such as the Yellow River Estuary, China. The IOC3M algorithm uses [Formula: see text] to substitute for switching the two-band ratio of max [R (rs) (443 nm), R (rs) (488 nm)]/R (rs) (551 nm) of the OC3M algorithm. In the IOC3M algorithm, the absorption coefficient of chla can be isolated as long as reasonable bands are selected. The performance of IOC3M and OC3M was calibrated and validated using a bio-optical data set composed of spectral upwelling radiance measurements and chla concentrations collected during three independent cruises in the Yellow River Estuary in September of 2009. It was found that the optimal bands of the IOC3M algorithm were λ(1) = 443 nm, λ(2) = 748 nm, λ(3) = 551 nm, and λ(4) = 870 nm. By comparison, the IOC3M algorithm produces superior performance to the OC3M algorithm. Using the IOC3M algorithm in estimating chla concentrations from the Yellow River Estuary decreases 1.03 mg/m(3) uncertainty from the OC3M algorithm. Additionally, the chla concentration estimated from MODIS data reveals that more than 90 % of the water in the Yellow River Estuary has a chla concentration lower than 5.0 mg/m(3). The averaged chla concentration is close to the in situ measurements. Although the case study presented herein is unique, the modeling procedures employed by the IOC3M algorithm can be useful in remote sensing to estimate the chla concentrations of similar aquatic environments.
在这项研究中，开发了一种改进的中分辨率成像光谱仪（MODIS）海洋叶绿素 a（chla）3 模型（IOC3M）算法，作为 MODIS 全球 chla 浓度估算算法 OC3M 的替代品，用于估算中国黄河口等高悬浮泥沙浓度水域的 chla 浓度。IOC3M 算法使用 [Formula: see text] 代替 OC3M 算法中最大双波段比 [R(rs)(443nm), R(rs)(488nm)]/R(rs)(551nm)的切换。在 IOC3M 算法中，只要选择合理的波段，就可以分离出 chla 的吸收系数。IOC3M 和 OC3M 的性能分别使用 2009 年 9 月在黄河口进行的三次独立航次采集的光谱上向辐射测量和 chla 浓度组成的生物光学数据集进行校准和验证。结果表明，IOC3M 算法的最佳波段为 λ(1)=443nm、λ(2)=748nm、λ(3)=551nm 和 λ(4)=870nm。相比之下，IOC3M 算法的性能优于 OC3M 算法。在黄河口使用 IOC3M 算法估算 chla 浓度，与 OC3M 算法相比，不确定性降低了 1.03mg/m(3)。此外，MODIS 数据估算的 chla 浓度表明，黄河口超过 90%的水域 chla 浓度低于 5.0mg/m(3)。平均 chla 浓度与现场测量值接近。尽管本文所提出的案例研究是独特的，但 IOC3M 算法所采用的建模程序在遥感中用于估算类似水生态环境的 chla 浓度可能是有用的。
