Sensorweb Systems for Global High-Resolution Monitoring of Environmental Phenomena

Abstract

Many science phenomena are difficult to observe because they occur infrequently and without warning. Over the course of the last 20 years, the concept of a “sensorweb,” or a network of interdependent instruments that autonomously coordinate to observe phenomena of interest, has been developed and implemented. In one use case, sensorwebs use in situ and low- to moderate-resolution sensors for event detection and pointable sensors with high spatial resolution for subsequent observation of targets, thus accomplishing efficient high-resolution monitoring. These sensorwebs have previously been implemented to study flood extent and volcanic lava flow. In the flood monitoring case, mostly remote sensors have been used for low-resolution flood detection, while more in-situ sensors have been used for volcanic activity detection in the volcano monitoring application. To improve these sensorwebs, we integrate commercial satellite constellations for high-resolution observations, resulting in higher spatial and temporal resolutions than the individual assets used in previous work. We also spatially extend sensorwebs to the global scale, produce a high volume of alerts, and automatically generate useful high-resolution data products. These resulting data products include surface water extent products for our flood sensorweb and volcanic lava classifiers for our volcano sensorweb. Our sensorwebs represent the state of the art in automatic acquisition and processing of data for environmental phenomenon monitoring. We demonstrate the capabilities of these improved sensorwebs in our new flood sensorweb and volcano sensorweb systems and describe ongoing efforts to utilize additional in-situ and spaced-based sensors and generate additional data products.