Date of this Version
Allen, R. H. (2022).Coupling Dendrochronology and Remote Sensing Techniques to Assess the Biophysical Traits of Juniperus virginiana and Pinus ponderosa within Grassland Communities in the Semi-arid Grasslands of the Nebraska Sandhills (Master's thesis, University of Nebraska-Lincoln, Lincoln, Nebraska, USA). Retrieved from https://digitalcommons.unl.edu/
Woody species encroachment is occurring within the sandhills region in Nebraska, primarily driven by Juniperus virginiana and Pinus ponderosa, altering ecosystems and the services they provide. Effective, low cost, and cross-scale monitoring of woody species growth and performance is necessary for integrated grassland and forest management in the face of climate variability and change. In this study, we sought to establish a relationship between remote sensing-derived vegetation indices (VIs) and dendrochronological (raw and standardized tree ring width) measurements to assess the performance of encroaching woody J. virginiana and P. ponderosa located within the Nebraska National Forest in the sandhills. We hypothesized that the environmental stresses that impact foliage growth and photosynthetic capacity of J. virginiana and P. ponderosa will also affect tree ring growth, which can be detected using remote sensing techniques, and that the impact of abiotic stresses on tree performance would be decoupled from that of adjacent grasslands due to differences in physiology, response to stress, and rooting depth. We evaluated relationships between the abiotic environment (precipitation, temperature, Palmer Drought Severity Index (PDSI), and soil water content 0 – 3 m depth), tree ring growth, and VIs. Our results indicate that precipitation, temperature, and PDSI were significant (p < 0.05) predictors of J. virginiana and P. ponderosa growth based on dendrochronological and VI measurements (1984 – 2013), while soil water content from 40 – 300 cm was a significant predictor of J. virginiana performance (2005 – 2013). Out of six VIs, four (NDVI, GCI, GRVI, and LSWI) were significant predictors of tree ring growth for both species. R2 values between grassland VIs and growing season climate were greater than those of J. virginiana or P. ponderosa, while grassland performance was decoupled from soil water content. We determined via Least Absolute Shrinkage and Selection Operator (LASSO) regressions that previous year climate was an important determinant of current year growth of both tree species but did not affect current year grassland performance. This study provides evidence for the efficacy of VIs in monitoring interannual variations in the growth of woody species, while determining abiotic factors significantly impacting the growth of grassland vegetation, J. virginiana, and P. ponderosa in the sandhills region.
Advisor: Tala Awada