Resilience of coastal wetlands to extreme hydrologic events in Apalachicola Bay

Abstract

Extreme hydrologic events such as hurricanes and droughts continuously threaten wetlands which provide key ecosystem services in coastal areas. The recovery time for vegetation after impact from these extreme events can be highly variable depending on the hazard type and intensity. Apalachicola Bay in Florida is home to a rich variety of saltwater and freshwater wetlands and is subject to a wide range of hydrologic hazards. Using spatiotemporal changes in Landsat‐based empirical vegetation indices, we investigate the impact of hurricane and drought on both freshwater and saltwater wetlands from year 2000 to 2015 in Apalachicola Bay. Our results indicate that saltwater wetlands are more resilient than freshwater wetlands and suggest that in response to hurricanes, the coastal wetlands took almost a year to recover, while recovery following a drought period was observed after only a month.

Wetland Dynamics Inferred from Spectral Analyses of Hydro-Meteorological Signals and Landsat Derived Vegetation Indices

“…In this paper, we used the processed NDVI dataset developed by Tahsin et al for this study area [15]. The NDVI data time series are similar except for the four ecosystem classifications across the six locations used herein.…”

“…However, the complete NDVI time series data were limited by clouds and other effects. For instance, only 252 months of data were usable out of the 384 months of the study time period [15,29]. Since NDVI is released as a 16-day composite, when two images were available for a given month the one with less cloud coverage was selected.…”

“…For heavily vegetated areas, NIR reflectance was greater than red reflectance due to the tendency of chlorophyll to absorb red light. In those areas, NDVI values were expected to be greater than 0 [15,33]. The wetland classification was superimposed onto the wetland NDVI values to set NDVI boundaries for different CW classes.…”

“…This index is sensitive to the green vegetation biomass as affected by the type of wetland and season [14]. It has been well correlated with wetland greenness, for example, in Apalachicola Bay of Florida [1,15]. Landsat NDVI is also regarded as a reliable indicator for wetland greenness change detection [16].…”

Optical Cloud Pixel Recovery via Machine Learning

“…The tide gauge data were added to the predictor variable vector and used as a proxy for prevailing water levels. If a storm surge event was experienced, the value would be higher and according to our current hypothesis and previous work, would lower the NDVI of impacted areas, especially freshwater wetlands [54]. Examples of input data records used to train and test the algorithm are shown in Table 2.…”