Understanding and predicting environmental conditions and hazards along our coastlines is important for management, coastal communities and economies. We provide scientific information and tools on environmental change and risk from hazards, including storms, rising seas, floods, changing air/water temperatures, and precipitation patterns. Satellite remote sensing technology and large-scale atmospheric patterns are used to test and develop coastal change indicators, metrics and machine-learning models to inform coastal managers on adverse situations affecting people, habitats, and valued resources.
Why We Care
Significant impact and damage to our coastal environments can occur due to increasingly more severe and frequent events. These weather–water events and subsequent damage are of great concern for people living in affected coastal communities and the valued assets we protect. We use large-scale data sets obtained nearly simultaneously over a relatively large area of the atmosphere and coastal water bodies from NOAA and other sources to create a framework to assess indicators of predictability of anomalous events damaging our coastal environments.
We determine how weather extremes interact with, or compound existing risks and impacts on coastal ecosystems. We determine how processes can be tracked, how impacts can be predicted and translated into environmental indicators that allow managers to take proactive measures.
What We Are Doing
We use satellite-derived data, in situ conditions and atmospheric patterns over large geographic areas as inputs to predict changes in coastal conditions. Results are analyzed and reassessed to eventually determine what weather-driven indicators can be used to measure or model a particular resilience characteristic. An example is using sea surface temperature (SST) data from large areas collected from satellites. The SST data were used to model weather-driven coastal change that can cause sea turtle hypothermia and mortality. Weather patterns have also been used to 1) evaluate satellite chlorophyll a and water clarity patterns in the Great Lakes and South Florida; 2) track kelp distribution, trends and changes in spatial patterns in Marine Protected Areas (MPAs) and 3) model and predict elevated sea levels and high-tide flood events in Mid-Atlantic and Southeast coastal areas.
- Evaluated linkages of weather patterns and water quality responses in South Florida using a synoptic climatological approach to evaluate algal bloom onset and severity, highlighting the contribution weather factors have on algal blooms. (2012)
- Modeled how regional winds in specific sequences contribute to daily, monthly, and seasonal changes in water clarity for the Southeastern U.S. (2014)
- Assessed cold snap mortality and hypothermia in loggerhead and green sea turtles in South Florida coastal ecosystems to develop a biological cold stress index using satellite sea surface temperature and weather pattern forcing. This work allows sea turtle rescue workers to be better prepared for when the next cold snap mortality or hypothermia event may occur. (2015)
- Subsequent South Florida work, defined water clarity patterns in South Florida coastal waters and their linkages to synoptic-scale wind forcing using satellite-derived light attenuation (Kd) index. (2016)
- Developed a non-linear autoregressive model with exogenous input to model water clarity relationships by reconstructing a historical water clarity index for the coastal waters of the Southeastern USA. (2016)
- In a multi-partner effort with CO-OPS and external partners, we used high-quality satellite altimetry and in situ data to model relationships between changing weather patterns, storm frequencies and nuisance flood events for NOAA high-priority areas, including sentinel sites and facilities. This work allows regional–local scale predictions of the probability of nuisance flood events. (2019)
- Continue collaborations with external research partners to create models to support outlooks of coastal inundation and flood risk for vulnerable areas within the North Carolina Sentinel Site Cooperative. This work will help guide restoration and adaptive planning in the Sentinel Site Cooperative using machine-learning models to better understand the combined effects of “upstream” and “downstream” sources of water level variability associated with changes in inundation that can affect marsh sustainability.
Weather patterns and extremes are undoubtedly linked to coastal effects such as floods, marsh inundation and movement, water quality, hypoxia, water clarity, and species mortality. Integrating downscaled weather data and information is critical to understand environmental impacts and predict and forecast future coastal changes, so managers and people living in coastal areas can consider measures to enable coastal resiliency.
Expected Outcomes
We will provide scientific information and tools using an array of integrated data sets, and weather–water machine-learning solutions. A full suite of interim milestones and end of project deliverables will include:
- Classification Maps (Self Organizing Maps). Historical daily to monthly weather–water types for all U.S. coastlines and surrounding areas
- Machine-learning Models. Neural Network–based indicator models linking atmospheric forcing predictor variables with aspects to coastal flooding and fisheries changes.
- Summary Report
- Improved understanding of environmental forcing on flooding and fishery recruitment and resiliency in all U.S. coastal systems
- Recommendations for informing fishery stock assessments
- Refereed Journal Article Manuscripts
- Manuscripts on weather linkages to blue crab fishery recruitment and resiliency and indicator models in Chesapeake Bay and surrounding ocean shelf environments.
Benefits of Our Work
This project will identify and link key atmospheric drivers to coastal hazards and impacts through scientific assessments and progressive machine-learning techniques. New tools and predictions of hazardous conditions developed in this project improve links to management through implementation of adaptive strategies for measuring and tracking rates and severity of change in systems.
Multiple management benefits include: improved response to increased hazard risk, thereby reducing further degradation of coastal ecosystems; more targeted monitoring of “at-risk” sites and impacted resources in protected areas (e.g., FKNMS; Great Lakes; Beaufort, NC); and improved guidance of cost-effective mitigation strategies for impacted shorelines, habitats, and resources. The science solutions and tools from this project will be integrated into existing NOS and academic partner decision-support frameworks including the CO-OPS Coastal Inundation Dashboard, and in federally protected areas.