Multi-Scale Sensor Fusion With an Online Application: Integrating GOES, MODIS, and Webcam Imagery for Environmental Monitoring

摘要

Webcam and satellite imagery are integral sensor web components, but there are few websites that directly synthesize these products. Additionally, websites as a means for reviewing multiple images in an organized way and for analyzing time-series imagery have not been fully explored. This paper describes how web applications can facilitate review of image archives for multi-scale sensor data fusion, with a test case of an automatic panning camera at the highest peak on Santa Cruz Island (in Channel Islands National Park off the Santa Barbara, CA coast) and MODIS and GOES-11 satellite imagery. We developed the application PanOpt (http://zulu.geog.ucsb.edu/panopt/index.html) for filtering this image database and outputting selected images as a static matrix or as an animation. On-demand product generation includes time-series charts and scatter plots of image metrics for regions of interest, with data markers linked to views of corresponding satellite and camera imagery via event-based web scripting. This is useful not only for exploratory analysis but is well suited for pedagogical applications and imparting a greater synoptic understanding of environmental phenomena. Demonstration of this website's utility includes multi-modal characterization of clouds as well as ground- and satellite-based plant phenology time-series. With this tool, the quantitative and qualitative aspects of image exploration are synergistic, making it easier to discern underlying trends and anomalies and to identify directions for further analysis.