Monitoring coastal marshes for persistent saltwater intrusion

Abstract

Saltwater flooding of coastal marshes by storm surge, rising sea level, and subsidence is a primary cause of wetland deterioration and habitat loss. The objective of this study is to provide resource managers with remote sensing products that support ecosystem-forecasting models requiring inundation data. This investigation employed time-series indices derived from 250-meter NASA Moderate Resolution Imaging Spectroradiometer (MODIS) and 30-m Landsat imagery to map flooding and saltwater stress in the Sabine Basin in southwest Louisiana, before and after Hurricane Rita in 2005. After nearly 20 feet of storm surge inundated the area during Hurricane Rita, Hurricane Ike produced a storm surge of almost 14 feet in the same area and flooded areas as far as 30 miles inland. The study design of this investigation centered upon the use of vegetation and wetness (water) indicators to map flooded areas. The study team assigned a vegetation index to marsh areas of concomitant vegetation and water. We derived daily MODIS time series of Normalized Difference Vegetation Index, Normalized Difference Water Index, and Normalized Difference Soil Index from the NASA Stennis Space Center Time Series Product Tool, which provides the capability to compute phenological parameters and performs temporal modeling at ecosystem scales. We estimated the extent of flooding as the percentage of time the MODIS index was water; i.e., below a certain threshold. The percentages indicate areas of persistent flooding over certain time intervals, thereby informing planners of areas with a high probability of conversion to open water. The study team used Landsat 5 and 7 data for the years 2004 through 2006 to produce an 8-day time series of vegetation and wetness indices. We evaluated these Landsat-based flood maps with lidar data and in situ elevation data collected by the U.S. Geological Survey (ÜSGS) and Louisiana Department of Natural Resources Coastwide Reference Monitoring System for the Sabine Basin. Finally, we combined salinity data collected in situ from the USGS and from the National Oceanic and Atmospheric Administration with our flooding estimates to map areas of persistent saltwater intrusion. The combination of these data are useful for habitat switching modules that predict the migration of marsh species from one salinity regime to another from estimates of the annual percent inundation and the mean annual salinity.