Climate variability and coral reef damage from satellite imagery

目前对气候变化以及海冰和珊瑚遥感损害的可变性的研究侧重于两个主题：1极地海冰与气候系统之间的相互作用2气候压力对热带珊瑚的影响，从基于卫星的遥感可以明显看出。专题一的工作旨在分析由被动微波成像得出的北冰洋海冰集中的二十五年记录的变化和变异性。同事们对极地海冰中的时间信号的研究一直基于各种海洋的线性时间关系。在一些地区，随着时间的推移有一些变化，而在其他地区没有变化。认识到自然系统没有先验的理由假设线性关系，我们一直在研究一种程序来确定非线性时间信号，并将它们与大气变量联系起来，这些变量可能是解释过程强迫的候选因素。到目前为止，我们已经发表了几篇使用小波分析海冰的主分量场和850mb地球位势高度的会议论文，以探索这项技术并从我的同事那里获得反馈。关于珊瑚礁压力的工作(专题二)侧重于开发遥感程序，以便根据卫星数据评估珊瑚礁的健康状况。在上一次探索基金的申请中，我们报道了水下光谱学的开创性技术，以确定我们是否可以区分健康的珊瑚和受损的珊瑚。我们还使用一种基于变化检测的新方法在这个问题上取得了进展。通过在多日期图像上应用局域算子模型，例如Getis统计量，我们已经说明，我们可以通过应用变化检测程序来确定珊瑚礁健康状况随时间的变化。其目的是利用卫星或航空图像，以定量的方式确定环境压力对珊瑚礁的影响，以便对管理决策进行评估。

The ongoing research on climate change and variability of sea ice and coral damage from remote sensing focuses on two themes: 1/ Interactions between the polar sea ice and climate systems 2/ The impact of climate stress on tropical coral as evident from satellite based remote sensing. Work on topic one has been directed towards analysis of change and variability in the twenty-five year record of sea ice concentration for the Arctic seas derived from passive microwave imagery. Studies of the temporal signal in the polar sea ice by colleagues has been based upon linear temporal relationships for various seas. In some areas, there is some change through time, while there is no change in other areas. Recognizing that there is no a priori reason to assume a linear relationship for a natural system, we have been working on a procedure to determine non-linear temporal signals and relate them to atmospheric variables that may be candidates for explanation of the process forcing. To date, we have given several conference papers that have used wavelet analysis of the principal component fields of sea ice and 850 mb geopotential height to explore the technique and elicit feedback from my colleagues. Work on the coral reef stress (Topic two) has focused upon developing remote sensing procedures to assess the health of the coral reef from satellite data. In the last Discovery grant application we reported on ground-breaking techniques in underwater spectroscopy to determine whether we can distinguish healthy from damaged corals. We have also progressed on this problem using a novel approach based upon change detection. By applying Local Area Operator models, such as the Getis Statistic, on multi-date imagery we have illustrated that we can determine the change in the health of a coral reef through time by applying change detection procedures. The objective is to use satellite or airborne imagery to determine the effects of environmental stress on the reef in a quantitative manner that will allow assessment for management decisions.