Document Type
Paper Presentation
Start Date
5-10-2023 10:30 AM
End Date
5-10-2023 11:45 AM
Event Website
https://iguide.illinois.edu/forum-2023/
Abstract
Technological advancement and the desire to better monitor shallow habitats in the Chesapeake Bay, Maryland, United States led to the initiation of several high-resolution monitoring programs such as ConMon (short for “Continuous Monitoring”) measuring oxygen, salinity, and chlorophyll-a at a 15-minute frequency. These monitoring efforts have yielded an enormous volume of data and insight into the condition of the tidal water of the Bay. But this information is underutilized in documenting the fine-scale variability of water quality, which is critical in identifying the link between water quality and ecological responses, partly due to the challenges in integrating monitoring data collected at different frequencies and locations. In a project to understand the environmental suitability of aquaculture sites and the future potential overlap between aquaculture and submerged aquatic vegetation, we developed a spatiotemporal synthesis of ConMon data with data from long-term, fixed-station seasonal monitoring. Here, we present our generalized additive model-based approach to predict salinity at high frequency (15 minutes) and fine spatial resolution (~100 meters) in the Maryland portion of the Bay, its major tributaries, and the shallow tidal creeks that exchange with the tributaries. Predictive performance was validated to be 1 PSU (practical salinity unit) in root mean square error using de novo monitoring. The resulting data provide insights into the environmental suitability of aquaculture, specifically the sensitivity of the Easter oyster (Crassostrea virginica) to low salinity stress. The spatiotemporal synthesis approach has potential applications for integrated monitoring and potential linkage with high-resolution water quality models for shallow habitats.
DOI
10.5703/1288284317664
Included in
Environmental Monitoring Commons, Natural Resources and Conservation Commons, Water Resource Management Commons
A Spatiotemporal Synthesis of High-Resolution Salinity Data with Aquaculture Applications
Technological advancement and the desire to better monitor shallow habitats in the Chesapeake Bay, Maryland, United States led to the initiation of several high-resolution monitoring programs such as ConMon (short for “Continuous Monitoring”) measuring oxygen, salinity, and chlorophyll-a at a 15-minute frequency. These monitoring efforts have yielded an enormous volume of data and insight into the condition of the tidal water of the Bay. But this information is underutilized in documenting the fine-scale variability of water quality, which is critical in identifying the link between water quality and ecological responses, partly due to the challenges in integrating monitoring data collected at different frequencies and locations. In a project to understand the environmental suitability of aquaculture sites and the future potential overlap between aquaculture and submerged aquatic vegetation, we developed a spatiotemporal synthesis of ConMon data with data from long-term, fixed-station seasonal monitoring. Here, we present our generalized additive model-based approach to predict salinity at high frequency (15 minutes) and fine spatial resolution (~100 meters) in the Maryland portion of the Bay, its major tributaries, and the shallow tidal creeks that exchange with the tributaries. Predictive performance was validated to be 1 PSU (practical salinity unit) in root mean square error using de novo monitoring. The resulting data provide insights into the environmental suitability of aquaculture, specifically the sensitivity of the Easter oyster (Crassostrea virginica) to low salinity stress. The spatiotemporal synthesis approach has potential applications for integrated monitoring and potential linkage with high-resolution water quality models for shallow habitats.
https://docs.lib.purdue.edu/iguide/2023/presentations/20