97th ESA Annual Meeting (August 5 -- 10, 2012)

PS 74-99 - Using remote sensing vegetation indices to assess moisture stress and long-term responses of subalpine grasslands to climate change

Thursday, August 9, 2012
Exhibit Hall, Oregon Convention Center
Jordan Holsinger1, Jack Brookshire1 and T. Weaver2, (1)Land Resources and Environmental Sciences, Montana State University, Bozeman, MT, (2)Department of Ecology, Montana State University, Bozeman, MT
Background/Question/Methods

Given their restricted setting and relatively rapid plant-soil interactions, high elevation grasslands are thought to be good indicators of the impact of climate change on ecosystem function. Here we focus on a high elevation (~2600 MASL) fescue (Festuca idahoensis) grassland in the Bangtail range in SW Montana, home to one of the longest running snow manipulation experiments in the world (BTL). Recent studies report a synchronous snowpack decline across the Northern Rocky Mountain cordillera affecting snowmelt runoff, peak snowpack, and number of snow free days. Our unprecedented 40-year record at BTL shows long-term directional changes in the grassland system regardless of snow manipulation treatment, effectively controlling for the declining snowpack of the region. We hypothesize that these long-term changes are not directly related to a declining snowpack but instead are due to changes in growing season characteristics resulting from climate change which lead to greater water stress. To evaluate this hypothesis, vegetation indices (NDVI and EVI) from MODIS and Landsat were used to assess changes in BTL growing season characteristics and dryness in the BTL grasslands over time and their relation to the observed long-term changes. 

Results/Conclusions

Initial results using MODIS vegetation indices from 2001-2011 indicate shifts in growing season characteristics including delayed green-up and lower maximum NDVI and EVI values with time as well as a reduced growing season range of NDVI and EVI (e.g. NDVImax-NDVImin). When MODIS EVI data is paired with snowpack data from a nearby SNOTEL site we found a weak trend of decreasing EVImin with a greater number of snow free days, implying that a greater number of snow free days may cause greater absolute dryness (lower EVImin). Clear trends may be obstructed by decadal variations and therefore we conducted and will present analyses using Landsat data over a longer observational window (1972-present).