Home > Overview > Key Findings > Key Outcomes > GCE III - Remote Sensing

GCE III - Key Finding in 2016

    Filtering out tidal flooding improves remote sensing of marsh vegetation

    Remote sensing in tidal marshes can provide synoptic assessments of marsh extent, phenology, primary production, and carbon sequestration. However, periodic tidal flooding reduces spectral reflectance, especially in the near and short-wave infrared wavelengths, which can make existing satellite products inaccurate and noisy. O'Connell et al. (2017) developed the TMII (Tidal Marsh Inundation Index) that could be used to filter MODIS imagery. When the TMII was included in a MODIS workflow it produced vegetation composites for S. alterniflora pixels that were consistent with expected patterns, whereas existing MODIS products (e.g. MOD13) were noisy and lacked seasonality (Fig. 1). We have tested the TMII on MODIS marsh pixels in both the GCE and Plum Island LTER domains as well as the Gulf coast, and expect it to be broadly useful for producing vegetation time series in tidal marshes. In addition to allowing one to sample on a much larger spatial scale than could be done in the field, another advantage of remote sensing is the ability to go back in time. O'Donnell and Schalles (2016) used geospatial techniques to scale up in situ measurements of aboveground S. alterniflora biomass to landscape level estimates using 294 Landsat 5 TM scenes acquired between 1984 and 2011. When Landsat-derived vegetation estimates were compared with abiotic drivers, they found that river discharge, precipitation, temperature and sea level all had positive relationships with biomass. There was also evidence for a long-term decline in biomass (about 34%) that appears to be related to increased frequency of drought and associated soil stressor responses in recent years. We are using both of these studies to inform our process-based plant and soil models.


    Fig. 1 NDVI time series for a Sapelo Island MODIS pixel, shown as (A) the standard MODIS 16-d composite series (MOD13),(B) the raw daily MODIS data, classified as flooded vs dry pixels based on the Tidal Marsh Inundation Index (TMII), and (C) an "optimal" (e.g. TMII filtered) 16-d composite series. From O'Connell et al. (2017).


This material is based upon work supported by the National Science Foundation under grants OCE-9982133, OCE-0620959, OCE-1237140 and OCE-1832178. Any opinions, findings, conclusions, or recommendations expressed in the material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.