The CDI Monthly Meeting in March focused on three 2019 CDI-funded projects: a national-scale map of sinkhole subsidence susceptibility; a project that will allow collection of near real-time eDNA surveillance of invasive species or pathogens; and an overview of SEINed - a tool for Screening and Evaluating Invasive and Non-native Data.
Jeanne Jones at the Western Geographic Science Center shared progress on the creation of a Subsidence susceptibility Map for the conterminous U.S. that will identify hotspots for sinkholes and areas susceptible to developing sinkholes. Sinkholes can pose major issues by focusing contaminated and/or polluted surface water into groundwater and creating instability in the foundations of buildings and roads. As such, a consistent map for the identification of sinkhole hotspots is vital in order to anticipate and manage risks.
The goals of this project include creating the first nationwide digital dataset of sinkhole hotspots, incorporating this dataset into the SHIRA (CDI) Risk map for use by DOI emergency agencies, and providing access to the dataset for external use by emergency managers, land use planners, and public works agencies. To meet these goals, the project team used The National Map, the National Hydrography Dataset (NHD), the Yeti supercomputer, and other data.
Jeanne shared challenges and solutions involved with several steps of the process. For instance, data had to be screened visually and manually in order to identify gaps in spatial coverage, and to screen out wetlands, open water, urban areas, and other non-karst landscape features.
Jeanne also posed a question to the CDI: How does flow accumulation processing with DEMs compare across Arcpy, TauDem, RichDem in terms of speed, consistency of results, max size of raster for high performance computing? You can respond to her at email@example.com.
The project presented by Elliott Barnhart tackles the problem of rapid detection and prediction of biological hazards. USGS collects a massive amount of near real time data with stream gauges, but the analysis of this data can take much longer. To solve this problem, the project team created a cloud-hosted digital database that combines all the collected data, and can easily incorporate eDNA and other data streams into models that indicate the presence or absence of organisms.
Challenges faced during the course of this project included creating effective quality control filters for funneling in data from multiple sources, and linking the benefits and capabilities of several different systems (like the MBARI Environmental Sample Processor, Department of Energy Systems Biology Knowledgebase, and more).
The Nonindigenous Aquatic Species (NAS) is the central repository for spatially referenced accounts of introduced aquatic species. NAS tracks over 1,290 aquatic species and stores over 600,000 observations from across the U.S. and spanning from the 1800's to the present. The SEINeD tool was developed to solve the problem: How does the NAS database get non-native occurrence data from groups not focused on invasive species? The SEINeD tool allows stakeholders to upload a biological dataset (fish, inverts, plants, etc.) collected anywhere in the conterminous US, Alaska, Hawaii, or US Territory that can then be screened for invasive or non-native aquatic species occurrences.
The SEINed tool helps to filter out inaccuracies due to incorrect taxa and spatial identifications before checking the indigenous status of the species against the sighting location. The tool flags non-native species that are exotic (from other countries/continents), AND non-native species from within the U.S. (for example, rainbow trout native to the west coast on the east coast) The data is then enhanced with the addition of spatial information like hydrological unit codes (HUCs) and returned to the user. The user can then submit the enhanced/corrected CSV to the NAS program.
The SEINed tool launches May 4th! Watch the NAS website for updates.
See the recording and slides at the March Monthly Meeting page.