PS 53-195 - Snow cover estimates in the semi-arid southwest using Landsat satellite images

Friday, August 12, 2016
ESA Exhibit Hall, Ft Lauderdale Convention Center
Mariah T. Ashley1, Temuulen T. Sankey2 and Jason McVay2, (1)College of Engineering, Forestry, and Natural Sciences, Northern Arizona University, Flagstaff, AZ, (2)Northern Arizona University, Flagstaff, AZ
Background/Question/Methods

Snowmelt in the semi-arid southwest is crucial in supplying water for the ecosystem and human population. Spring snowmelt contributes to increased groundwater recharge, providing up to 85% of annual water supply to Arizona (Serreze et al., 1999). However, because pine trees have large tree canopies, most of the snow is intercepted and sublimated before it can replenish the soil, but efforts like the Four Forest Restoration Initiative (4FRI), conducted by the U.S. Forest Service, will restore approximately 2.5 million acres of ponderosa pine forests. Through prescribed fires and mechanical thinning techniques, 4FRI’s effort will attempt to decrease the threat of large-scale catastrophic wildfires. The purpose of our study is to use remote sensing satellite data to estimate snow accumulation, snow water equivalence (SWE), and snow retention as a result of forest restoration treatments. Spanning 26 years, from 1988 to 2014, 66 Landsat 5TM/ETM+ images were used to estimate snow accumulation at five local sites near Flagstaff, Arizona. Each of the five sites, approximately 10 hectares in size, contains three treatment types: control, thin, and thin-and-burn. Landsat images were analyzed using the Normalized Difference Snow Index (NDSI) algorithm which estimated snow accumulation at each site. To ground-truth our remote sensing data; time-Domain Reflectometer (TDR) probes were established at 500 feet transects to measure soil moisture, two NRCS snow courses were also established at each treatment type, and SNOTEL (SNOw TELemetry) station data measurements of snow accumulation and precipitation were used.

Results/Conclusions

Landsat NDSI analysis showed that the thin and thin-and-burn sites had greater snow accumulation than the control sites and Landsat-based NDSI also correlated very strongly with NRCS and SNOTEL data. Forest management efforts do have an effect on snow accumulation of which was a previously underdeveloped study. These forest management efforts have significantly shown an increase in snow accumulation and retention which can then feed into the spring season, possibly contributing to groundwater recharge while also reducing catastrophic wildfires (Sankey et al., 2015). Understanding snow cover is vital to the region because it is projected that temperatures increase and snow season length decrease due to global climate change. In the desert regions of Arizona, spring snowmelt is important for growing metropolitan areas and ecosystems. 4FRI will continue to manage and restore ponderosa forests for the next couple of decades, allowing for more data and more research.