Garcia, Carlos Javier (2017-08). Influence of Bark-Beetle-Induced Tree Mortality on Slope Erosion in the San Juan Mountains of Ouray, CO, USA. Master's Thesis.
Thesis
Degradation of the natural environment has resulted from the destruction by bark-beetles at various geographic locations around Earth. Presently, widespread tree mortality is occurring in the San Juan Mountains of southwestern Colorado. Methods of remote sensing for classifying mountain pine beetle-induced tree mortality have been developed, which will be applied to study the destruction of forest caused by the Douglas-fir beetle (Dendroctonus pseudotsugae), and fir-engraver beetles (Scolytus ventralis) that has been occurring in the Ouray, Colorado, area over the last several years. Infestations of bark-beetles result in wide-spread tree mortality, and the loss of vegetation can result in increased rates of surface runoff and slope erosion. This presents the problem identified: Does a causal relationship exist between the destruction of trees by bark-beetles and increased rates of surface runoff and erosion on the slopes in the Ouray, Colorado area? The question posed was answered by accomplishing the following three objectives: 1. Determine if a link exists between rate of tree mortality and rate of erosion. 2. Model the rate of tree mortality and rate of erosion in the study area. 3. Provide a first approximation of surface runoff and sediment production rate. Potential soil erosion, calculated by the model, ranged from ~0 Mg/ha/yr projected erosion in areas of low slope to a projected erosion value of ~2,300 Mg/ha/yr in areas with extremely steep slopes and areas of high drainage output conducive to flowing water. The movement of materials down slope poses a potential hazard for the town of Ouray, CO, which is situated at the bottom of the valley. Calculated changes in the normalized difference vegetation index (NDVI) from 2005 to 2016 showed a maximum negative change -0.53 and the maximum increase of 0.57 with a mean change of 0.08 and standard deviation of 0.09. A chi-squared test yielded a p-value of 0.00, allowing me to reject the null hypothesis and accept the alternative hypothesis, HA, that NDVI changed over the course of this time series. This study also provided a preliminary method for predicting surface runoff using an NDVI time series as a classification threshold input. Areas above 0.38 NDVI had a range of surface runoff values between -390,000,000 mm to -0.2 mm per tree within each pixel with a mean value between -146,916,924 mm and -5.2 mm and standard deviation of 12.6 mm - 83,674,617 mm. Runoff in areas below 0.38 NDVI had a range between 0.05 to 8.53 mm of runoff per tree with a mean of 2.15 mm and a standard deviation of 1.66 mm within each 900 m^2 pixel. Because of limitations, I was unable to conclude whether or not the null hypothesis can be rejected in that bark-beetle-induced tree mortality does not lead to an increase in surface runoff.
Degradation of the natural environment has resulted from the destruction by bark-beetles at various geographic locations around Earth. Presently, widespread tree mortality is occurring in the San Juan Mountains of southwestern Colorado. Methods of remote sensing for classifying mountain pine beetle-induced tree mortality have been developed, which will be applied to study the destruction of forest caused by the Douglas-fir beetle (Dendroctonus pseudotsugae), and fir-engraver beetles (Scolytus ventralis) that has been occurring in the Ouray, Colorado, area over the last several years. Infestations of bark-beetles result in wide-spread tree mortality, and the loss of vegetation can result in increased rates of surface runoff and slope erosion. This presents the problem identified: Does a causal relationship exist between the destruction of trees by bark-beetles and increased rates of surface runoff and erosion on the slopes in the Ouray, Colorado area?
The question posed was answered by accomplishing the following three objectives: 1. Determine if a link exists between rate of tree mortality and rate of erosion. 2. Model the rate of tree mortality and rate of erosion in the study area. 3. Provide a first approximation of surface runoff and sediment production rate.
Potential soil erosion, calculated by the model, ranged from ~0 Mg/ha/yr projected erosion in areas of low slope to a projected erosion value of ~2,300 Mg/ha/yr in areas with extremely steep slopes and areas of high drainage output conducive to flowing water. The movement of materials down slope poses a potential hazard for the town of Ouray, CO, which is situated at the bottom of the valley.
Calculated changes in the normalized difference vegetation index (NDVI) from 2005 to 2016 showed a maximum negative change -0.53 and the maximum increase of 0.57 with a mean change of 0.08 and standard deviation of 0.09. A chi-squared test yielded a p-value of 0.00, allowing me to reject the null hypothesis and accept the alternative hypothesis, HA, that NDVI changed over the course of this time series.
This study also provided a preliminary method for predicting surface runoff using an NDVI time series as a classification threshold input. Areas above 0.38 NDVI had a range of surface runoff values between -390,000,000 mm to -0.2 mm per tree within each pixel with a mean value between -146,916,924 mm and -5.2 mm and standard deviation of 12.6 mm - 83,674,617 mm. Runoff in areas below 0.38 NDVI had a range between 0.05 to 8.53 mm of runoff per tree with a mean of 2.15 mm and a standard deviation of 1.66 mm within each 900 m^2 pixel. Because of limitations, I was unable to conclude whether or not the null hypothesis can be rejected in that bark-beetle-induced tree mortality does not lead to an increase in surface runoff.
