Harmel Tristan, Gilerson Alexander, Hlaing Soe, Tonizzo Alberto, Legbandt Tom, Weidemann Alan, Arnone Robert, Ahmed Samir
Optical Remote Sensing Laboratory, The City College of the City University of New York, New York City 10031, New York, USA.
Appl Opt. 2011 Oct 20;50(30):5842-60. doi: 10.1364/AO.50.005842.
The Long Island Sound Coastal Observational platform (LISCO) near Northport, New York, has been recently established to support validation of ocean color radiometry (OCR) satellite data. LISCO is equipped with collocated multispectral, SeaPRISM, and hyperspectral, HyperSAS, above-water systems for OCR measurements. This combination offers the potential for improving validation activities of current and future OCR satellite missions, as well as for satellite intercomparisons and spectral characterization of coastal waters. Results of measurements made by both the multi and hyperspectral instruments, in operation since October 2009, are presented, evaluated and their associated uncertainties quantified based on observations for a period of over a year. Multi- and hyperspectral data processing as well as the data quality analysis are described and their uncertainties evaluated. The quantified intrinsic uncertainties of HyperSAS data exhibit satisfactory values, less than 5% over a large spectral range, from 340 to 740 nm, and over a large range of diurnal daylight conditions, depending on the maximum sun elevation at the solar noon. Intercomparisons between HyperSAS and SeaPRISM data revealed that an overcorrection of the sun glint effect in the current SeaPRISM processing induces errors, which are amplified through the whole data processing, especially at the shorter wavelengths. The spectral-averaged uncertainties can be decomposed as follows: (i) sun glint removal generates 2% uncertainty, (ii) sky glint removal generates strong uncertainties of the order of 15% mainly induced by sun glint overcorrection, (iii) viewing angle dependence corrections improve the data intercomparison by reducing the dispersion by 2%, (iv) normalization of atmospheric effects generates approximately 4% uncertainty. Based on this study, improvements of the sun glint correction are expected to significantly reduce the uncertainty associated with the data processing down to the level of 1%. On the other hand, strong correlations between both datasets (R(2)>0.96) demonstrate the efficacy of the above-water retrieval concept and confirm that the collocated instrumentation constitutes an important aid to above-water data quality analysis, which makes LISCO a key element of the AERONET-OC network.
纽约北港附近的长岛海峡海岸观测平台（LISCO）最近建成，旨在支持海洋水色辐射测量（OCR）卫星数据的验证工作。LISCO配备了用于OCR测量的并置多光谱SeaPRISM和高光谱HyperSAS水上系统。这种组合为改进当前和未来OCR卫星任务的验证活动、进行卫星相互比较以及对沿海水域进行光谱特征分析提供了潜力。文中给出了自2009年10月起运行的多光谱和高光谱仪器的测量结果，并基于一年多的观测数据进行了评估，同时对其相关不确定性进行了量化。描述了多光谱和高光谱数据处理以及数据质量分析过程，并评估了它们的不确定性。HyperSAS数据的量化固有不确定性表现出令人满意的值，在340至740nm的大光谱范围内以及在大范围的昼夜日光条件下，不确定性小于5%，这取决于太阳正午时的最大太阳高度。HyperSAS和SeaPRISM数据之间的相互比较表明，当前SeaPRISM处理中太阳耀光效应的过度校正会导致误差，这些误差在整个数据处理过程中会被放大，尤其是在较短波长处。光谱平均不确定性可分解如下：（i）去除太阳耀光产生2%的不确定性；（ii）去除天空耀光产生约15%的强烈不确定性，主要由太阳耀光过度校正引起；（iii）视角依赖性校正通过将离散度降低2%来改善数据相互比较；（iv）大气效应归一化产生约4%的不确定性。基于这项研究，预计太阳耀光校正的改进将显著降低与数据处理相关的不确定性，使其降至1%的水平。另一方面，两个数据集之间的强相关性（R(2)>0.96）证明了水上反演概念的有效性，并证实并置仪器对水上数据质量分析具有重要辅助作用，这使得LISCO成为AERONET-OC网络的关键要素。
