A Sampling of Previous Research Projects from GEO/WRI 201

As Climate Warms, Spruce Fail to Recover after Severe Beetle Outbreak on Utah’s Markagunt Plateau
William Atkinson, '18

Photograph of field work on the Markagunt Plateau

Field work on the Markagunt Plateau

Extended insect infestations in susceptible forests, such as the 1990s spruce beetle outbreak in Utah’s Markagunt Plateau, can kill over 95% of the area’s dominant tree species, transforming forest biodiversity and impacting climate change feedbacks. Recent analysis of satellite imagery indicates that continued warm and dry conditions are inhibiting spruce recovery on the plateau, but uncertainties of low-resolution imagery must be resolved by ground-truthing. While other researchers have analyzed long-term recovery after older outbreaks, spruce recovery after modern climate change-enhanced outbreaks is less well understood. By analyzing data from ground surveys, tree dating, and high-resolution UAV imagery, I test my hypothesis that spruce are being succeeded by younger fir and aspen, particularly in areas unaffected by humans. Results indicate abnormally few young spruce at the site compared to similar studies, but plentiful young and fast-growing fir and aspen, confirming the hypothesis that spruce are being replaced during this modern era of climate change.


Invasive Tamarisk Narrow the Colorado River in Meander Canyon, Utah
Artemis Eyster, '19

Photograph of field work at Meander Canyon

Field work at Meander Canyon

Invasive tamarisk trees have spread along waterways in the American Southwest, decreasing biodiversity and groundwater abundance, and increasing salinity, over-bank flooding, and fire frequency. Increases in temperature could impact tamarisk recruitment by changing peak snow-melt induced floods as well as tree flowering time. Between Canyonlands National Park and Moab, Utah, a floodplain of the Colorado River has been restricted by 49% since 1986. I analyzed hydraulic and climate variables, Landsat satellite images, UAV derived elevation model, and tree cores. Tamarisk colonized and stabilized a sandbar between 1988 and 1991, leading to an expansion of the floodplain behind the sandbar which has now become populated by native willow.


Using Landsat Satellite Images to Map Our Changing World
Alec Getraer, '19

Photograph of Alec Getraer, '19

Alec Getraer, '19

Global vegetation cover is changing dramatically as a result of deforestation due to increased human development and rising forest mortality caused by anthropogenic climate change. To quantify these changes, an increasing number of researchers are using the Normalized Differential Vegetation Index (NDVI) derived from remote-sensed Landsat imagery. However, long term studies require the use of multiple Landsat databases, currently Landsat 5, 7, and 8, which produce statistically distinct NDVI datasets. This study shows that calibration of NDVI data sets between Landsat 5 and Landsat 7 is critical to making accurate long term assessments of vegetation cover.


Topography-Based Snowpack Extrapolation over Utah Lake Watersheds Estimates Yearly Water Supply Trends
Benjamin Getraer, '19

Photograph of Benjamin Getraer, '19

Benjamin Getraer, '19

Many places on Earth depend heavily on the gradual melt of seasonal snowpack for fresh water. Predictions of serious snowmelt changes and watershed changes based on Snow Water Equivalent (SWE) only observe data from individual sites; my method extrapolates site data into a watershed estimation using the landscape features of the watershed. I also found that almost all of the recent studies looking at changes in streamflow and SWE either fail to take key sources of error into account, or make predictions based on datasets that terminated almost 30 years ago.


The Role of Topography and Vegetation Type in Post-Fire Forest Recovery, Fishlake National Forest, Utah
Emily Geyman, '19

Photography of Emily Geyman, '19

Emily Geyman, '19

In recent decades, fuel accumulation, warmer temperatures, and longer fire seasons have driven up the frequency and severity of wildfires worldwide. An improved understanding of forest recovery dynamics—and their dependence on temperature, precipitation, topography, and vegetation type—is essential for developing targeted wildfire management techniques that effectively mitigate fire damage in each landscape. Using Landsat satellite data, drone-derived aerial imagery, and a variety of image analysis techniques including image classification and difference mapping, I investigate how four variables—elevation, aspect, slope, and vegetation type—have influenced the rate of forest regeneration following a 15,000-acre fire in Fishlake National Forest, Utah. I predict that steeply-sloped areas will have a slower NDVI recovery because those regions endure the most soil erosion and nutrient loss. Shallow north-facing slopes and areas populated by deciduous trees experienced the fastest NDVI rebound, while elevation had no significant effect on the rate of vegetation recovery.