Coastal tidal wetland change in the northeastern United States
Coastal tidal wetlands are highly altered ecosystems at substantial risk due to widespread and frequent land-use change, coupled with sea-level rise, leading to disrupted hydrologic and ecologic functions and ultimately, significant reduction in climate resiliency. Knowing where and when the changes have occurred, and the nature of those changes and their associated risks, is paramount to coastal communities and natural resource management. Large-scale mapping of the costal tidal wetland changes is extremely difficult due to their inherent dynamic nature. To bridge this gap, we developed an automated algorithm for DEtection and Characterization of cOastal tiDal wEtlands change (DECODE) using dense Landsat time series. We used DECODE to track the status of coastal tidal wetlands in the northeastern United States from 1986 to 2020. The overall accuracy of land cover classification and change detection is approximate 95.8% and 99.8%, respectively. The vegetated wetlands and open water were mapped with 94.6% and 99.0% user’s accuracy, and 98.1% and 93.5% producer’s accuracy, respectively. The cover change and condition change were mapped with 68.0% and 80.0% user’s accuracy, and 80.5% and 97.1% producer’s accuracy, respectively. We discovered that approximately 3283 km2 (12%) of coastal tidal wetlands in the northeastern United States occurred at least one time of change but condition changes were responsible for majority (84.3%). Vegetated coastal tidal wetland decreased consistently (approximately 2.6 km2 per year) in the past 35 years, largely due to its conversion to open water.
Steps to reproduce
DECODE consists of four elements, including spectral break detection, land cover classification, change characterization, and tidal influence prediction. DECODE assembles all available Landsat observations and introduces a water level regressor for each pixel to flag the spectral breaks and estimate harmonic time-series models for the divided temporal segments. Each temporal segment is classified (e.g., vegetated wetlands, open water, and others – including unvegetated wetlands and uplands) based on the phenological characteristics and the synthetic surface reflectance values calculated from the harmonic model coefficients, as well as a generic rule-based classification system. This harmonic model-based approach has the advantage of not needing the acquisition of satellite images at optimal conditions (i.e., low tide status) to avoid underestimating coastal vegetation caused by the tidal fluctuation. At the same time, DECODE can also characterizing different kinds of changes including land cover change and condition change. Additional, DECODE can predict the degree of the tidal influence based on the coefficient of the water level regressor.