Date of Graduation

Fall 2018


Master of Natural and Applied Science in Agriculture


College of Agriculture

Committee Chair

Michael Goerndt


tree health, forestry, UAV, UAS, black walnut, plant stress, hyperspectral, photogrammetry, vegetation indices, remote sensing

Subject Categories

Agricultural Science | Agriculture | Environmental Monitoring | Forest Management | Forest Sciences | Natural Resources and Conservation | Plant Sciences


The development of compact sensors in recent years has inspired the use of UAS-based hyperspectral and aerial imaging techniques for small-scale remote sensing applications. With increasing concerns about climate change, spectrally-derived vegetation indices (VIs) have proven useful for quantifying stress-induced vegetation response. The goal of this study was to develop predictive models and assess methodology for modeling the biological response of a black walnut -dominant mixed hardwood stand to seasonal climate events using UAV-based hyperspectral remote-sensing. The derived VIs were evaluated against the means of four seasonal measures of climate calculated for a two-week period prior to the flight date. A best subsets regression was used to create best fitting linear regression models according to Bayesian Information Criterion (BIC). The highest-ranked model for total precipitation had an AdjR² of 0.0839 and RMSE of 0.0827 inches. The highest-ranked model for maximum air temperature had an AdjR² of 0.9922 and RMSE of 0.5485 °F. The highest-ranked model for average air temperature had an AdjR² of 0.9987 and RMSE of 0.2256 °F. The highest-ranked model for total solar radiation had an AdjR² of 0.9961 and RMSE of 0.06405 MJ/M². The results indicate that select VIs measured at the canopy level may be useful in estimating the response to at least some measures seasonal climate. The proposed regression models could help local researchers and landowners in making short-term management decisions, as well as further our understanding of climate-induced tree stress for maintaining sustainable forests in Missouri.


© Tyler G. Bradford

Open Access