MODELING WETLAND RESPONSE TO FUTURE SEA-LEVEL RISE IN THE PAMLICO AND CROATAN SOUNDS, NORTH CAROLINA

Abstract

Coastal habitats are among the world's most vulnerable environments to climate change and are highly sensitive to the impacts of future SLR. During the course of this century sea-level rise (SLR) enhanced by global climate change will become a major issue affecting coastal wetlands. Predicted SLR in the future could have major impacts on estuarine systems and will likely force changes in wetland spatial extent, geographic location, and type. Coastal wetlands located along the Pamlico and Croatan Sounds in eastern North Carolina will undoubtedly be greatly affected by future SLR due to their large spatial extent and high vulnerability, and will need to be closely monitored and mapped to determine their future locations and rates of change, including erosion, accretion, and loss. Research assessing the impacts of future SLR on coastal wetlands is vital for determining ways to conserve and protect these natural resources. The use of GIS-based, ecological SLR modeling is essential in order to analyze and explore the potential habitat changes of coastal wetlands during long-term SLR. The purpose of this study is to determine the relative accuracy of the Sea-Level Affecting Marshes Model (SLAMM) in predicting wetland response to future SLR in the Pamlico and Croatan Sounds, North Carolina. SLAMM accuracy was determined by performing a model hindcast and outputs were compared to current wetland maps utilizing point and cell-based accuracy assessments, as well as various descriptive statistics. Accuracy results from model hindcasting were deemed acceptable to run model forecasts through 2100 using varying SLR scenarios. Future wetland change in both spatial extent and type were assessed using both quantitative and visual analysis. Model forecast results predict major changes within the study area, even devastating ones ecologically to wetlands and all interlinked habitats and ecological systems. Additional studies should be conducted using SLAMM utilizing hindcasting for calibration of model parameters and implementing higher-quality input data to yield better model outputs and accuracy.