Identifying Hydrological Driving Forces of Coastal Wetland Degradation Using Random Forest Classifiers

Coastal wetlands provide critical ecosystem functions but are experiencing disruptions caused by inundation and saltwater intrusion under intensified climate change, sea-level rise, and anthropogenic activities. Recent studies have shown that these disturbances degrade coastal wetlands mainly through affecting their hydrological processes. However, research on what is the most critical hydrological variable for coastal wetland degradation and how it affects coastal wetlands is still in its infancy. In this study, we analyzed three types of wetland degradation that occurred in the past decades at the Alligator River coastal wetlands in North Carolina, US, including woody wetland loss, emergent herbaceous wetland loss, and woody wetlands degrading to emergent herbaceous wetlands. By using random forest classifiers, we determined the relative importance of different hydrological processes and the dominant variables in driving the three types of wetland degradation. Our results showed that random forest classifiers were accurate (> 97% accuracy) in classifying wetland degradation. Coastal hydrology was confirmed to play a crucial role in coastal wetland degradation. However, the dominant hydrological mechanisms of different types of wetland degradation varied. Woody wetlands were most susceptible to saltwater intrusion and were very likely to degrade if the saltwater table was shallower than 0.2 m below the land surface. In contrast, emergent herbaceous wetlands were most vulnerable to inundation and drought. The favorable groundwater table for emergent herbaceous wetlands was between 0.34 m above the land surface and 0.32 m below the land surface, beyond which the emergent herbaceous wetland tends to disappear. For degraded woody wetlands, their distances to canals played a crucial role in determining their fates after degradation. Our study shed light on the combined effects of natural disturbances and anthropogenic activities on coastal wetland degradation. The machine learning approach employed in this study provided critical knowledge about the thresholds of hydrological variables for different coastal wetlands. Such information can help guide effective and targeted coastal wetland conservation and restoration measures.