Environmental informatics uses large multidimensional, complex datasets to study environmental problems, which can be both discrete and continuous in space or time. These datasets and their requisite metadata can be managed by queryable databases. Geospatial Web application programming interfaces (APIs) provide remote access to dynamic subsets of environmental information. Persistent identifiers make data citable. The storage-computing trade-off is now heavily skewed in favor of moving calculations to the data. Provenance metadata help determine a data object's reliability and trustworthiness.Rising atmospheric CO2, the Antarctic ozone hole, and Gulf Stream warm-core rings were all discovered by analyzing long-term datasets. Similar work continues on mapping evapotranspiration and snow water equivalent. In these “fourth paradigm” problems, data (especially data collected operationally) drive hypothesis formation. Making data available requires new discovery mechanisms and policies favoring data sharing. Cloud computing and array-friendly databases will help bring processing to the data. Ubiquitous location sensing and geotagging will help turn citizen scientists into environmental information collectors.
The application of ultra-high spatial resolution imagery from small unpiloted aerial systems (sUAS) can provide valuable information about the status of built infrastructure following natural disasters. This study employs three methods for improving the value of sUAS imagery: (1) repeating the positioning of image stations over time using a bi-temporal imaging approach called repeat station imaging (RSI) (compared here against traditional (non-RSI) imaging), (2) co-registration of bi-temporal image pairs, and (3) damage detection using Mask R-CNN, a convolutional neural network (CNN) algorithm applied to co-registered image pairs. Infrastructure features included roads, buildings, and bridges, with simulated cracks representing damage. The accuracies of platform navigation and camera station positioning, image co-registration, and resultant Mask R-CNN damage detection were assessed for image pairs, derived with RSI and non-RSI acquisition. In all cases, the RSI approach yielded the highest accuracies, with repeated sUAS navigation accuracy within 0.16 m mean absolute error (MAE) horizontally and vertically, image co-registration accuracy of 2.2 pixels MAE, and damage detection accuracy of 83.7% mean intersection over union.
25 January 2000 in Journal of the American Society for Information Science