Publication Date


Date of Final Oral Examination (Defense)


Type of Culminating Activity


Degree Title

Master of Science in Geoscience



Major Advisor

Jen Pierce, Ph.D.


Nancy F. Glenn, Ph.D.


Mojtaba Sadegh, Ph.D.


I examined wildfire characteristics in the Frank Church Wilderness, central Idaho, between 1972-2012. Studying fire characteristics in the Frank Church Wilderness provides an opportunity to understand the history of wildfires in a federally designated wilderness area, largely devoid of management impacts with limited human access and activity. The ~958,000-hectare Frank Church Wilderness area encompasses the Middle Fork Salmon River. Vegetation cover ranges from high elevation (~2500-3200 meters) mixed conifer forests in the headwaters to low-elevation (~600-1000 meters) sagebrush-steppe and ponderosa pine (Pinus Ponderosa) forests. The Frank Church Wilderness is defined as unmanaged because effective fire suppression (e.g., vehicle and air-assisted fire suppression), logging, road access, and motorized vehicle use are extremely limited; therefore, this area provides an excellent location to examine historical changes in wildfire characteristics in the absence of substantial management influence. Studies of wildfires in the Western USA show an increase in area burned in the past several decades; however, the root cause of the trend is attributed to both historical fire suppression and a warming climate.

This research aims to understand fire characteristics and their correlation with a warming climate in the Frank Church Wilderness. Our research questions are:

  1. How do landscape fire metrics relate to warming trends in an unmanaged wilderness?
  2. How are landscape metrics of burned areas correlated with one another?

As a proxy for the influence of warming and drying on vegetation, I use vapor pressure deficit (VPD), which measures air aridity and is the difference between moisture pressure in the air and its value at saturation. The study uses fire atlas data from 1972-2012, remotely sensed data, and historical VPD records to test correlations among climate aridity, burn area, and other fire metrics.

This analysis shows that burned area in the Frank Church Wilderness increased between 1972-2012 and is significantly correlated with VPD, indicating that fires become larger as aridity increases. Severe fire years with large burn areas include 1988, 2000, and 2008. This work supports studies that attribute the growth in burned areas (1972-2012) to background warming and drying.

I used FRAGSTATS software and landscape metric calculations in a pilot study to better understand the changes to wildfire shape and total area burned in the Frank Church Wilderness. FRAGSTATS show a high positive correlation (Pearson correlation coefficient of 0.57) between total area burned and VPD (p-value of 0.001). The number of patches also positively correlated with VPD (p-value of 0.002). The landscape shape index had a positive correlation (Pearson correlation coefficient of 0.48) to VPD with a p-value of 0.01. Perimeter-area fractal dimension index metric had a negative correlation (Pearson correlation coefficient of -0.38) with VPD with a p-value of 0.05.

While additional work is needed, the scientific and land management communities can benefit from the nuanced understanding of the relationship between climate aridity and burned landscape patterns in an unmanaged region.