Using the Landsat archive and gridded climate data to assess restoration of mountain meadows

Mountain meadows provide a variety of ecosystem services, and the impacts of these ecosystem services propagate beyond the boundaries of the meadows themselves . Conservation and restoration efforts frequently focus on these ecosystems. Meadow monitoring data is both expensive and difficult to collect, and projects often operate on limited budgets. Cloud computing has made it possible to access and analyze remotely sensed data at high spatial and temporal resolution. Here we use the Landsat archive to develop robust and repeatable assessments of meadow restoration projects in the Lake Tahoe Basin. The duration (35+ years), 8-16 day overpass frequency, and 30 meter pixel size of Landsat imagery make it an ideal tool to establish baseline pre-restoration conditions in the absence of extensive field data and to assess changes following a restoration project. We focused on the normalized difference vegetation index (NDVI), which integrates a wide range of phenomena (climate and land use change, geomorphological processes, fire and other disturbances, vegetative succession, etc.) into a single metric . Because it is vegetation-based, NDVI is extremely sensitive to changes in hydrologic regime, and such changes are frequently goals of meadow restoration projects (e.g., arrest incision, restore floodplain connectivity). We retrieved spatially and temporally integrated annual values of NDVI and gridded climate data for more than 350 meadows throughout the basin, and also examined the distribution of individual pixels within those meadows. Annual NDVI was controlled for inter-annual climate variability. Using non-parametric statistical analyses that considered both whole-meadow integrated NDVI and the intra-meadow variability of NDVI, we examined several restoration sites within the basin. Results are consistent with field-based reports on both successful and unsuccessful restoration projects. The data used during this study are freely available and continually collected. While remote sensing analysis cannot substitute for boots on the ground field visits, this work offers a low-cost opportunity to collect some data that may not be available through other means and to fill data gaps between infrequent field visits.