Evaluating Vegetation in the Sierra Nevada Using Continuous Landsat Time Sequences

Submitted by Anonymous (not verified) on Mon, 07/30/2018 - 14:12

Until recently, Landsat imagery has not been commonly used to investigate the link between long-term climate trends, extreme weather events, and vegetation response, in part because of the on-site computational limitations associated with compiling an imagery time series suitable for short- and long-term trend analyses. The USGS, in conjunction with the University of California (Davis) and the Desert Research Institute, has employed Google Earth Engine image processing to leverage the Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper (ETM+)  archive from 1985 to 2012 and corresponding climate records to better establish long-term relationships between vegetation dynamics and climate. Montane meadows, particularly wet meadows and fens, are believed to be particularly sensitive to climate-driven changes, and their relatively small size makes them excellent candidates for Landsat-resolution analysis.  Landsat-derived vegetation indices (i.e, normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), Tasseled Cap) and parameter-elevation regressions on independent slopes model (PRISM) temperature and precipitation data were summarized at the patch scale for roughly 5,000 meadows in the Sierra Nevada mountain range in California.  Index outputs were compiled every 16 days over a 27-year period, leading to nearly 600 observations for each meadow.

Data were post-processed to remove noise caused by snow, clouds, and shadows, and to exclude unrealistic index values resulting from anthropogenic influences or poor meadow boundary delineation. Scientists compiled long-term trends for each meadow from 1985–2012 by plotting noise-free data and summarizing index values by month and by year. Scientists also identified break points for each meadow, which are short-term departures from time series trends caused by events such as extreme rainfall, floods, and wildfire. For example, preliminary results suggest breaks in which a meadow exhibits a higher than normal NDVI for a given month may coincide with extreme rainfall events. A forthcoming assessment of vegetation trends with co-variates such as climate, meadow type, latitude, and altitude will address why different spatial and temporal patterns arise among meadows in the Sierra Nevada.

Author Name
Chris Soulard
Author Email