Date of Completion


Embargo Period



Global Navigation Satellite System (GNSS), Real-time kinematic (RTK), Accuracy, Geographic Information System (GIS), Habitat suitability model, Model comparison, Analysis of covariance (ANCOVA), Friedman's rank sum test, Cochran's Q test, Binary classification analysis

Major Advisor

Dr. Thomas H. Meyer

Associate Advisor

Dr. John C. Volin

Associate Advisor

Dr. Jason Vokoun

Associate Advisor

Dr. Chuanrong Zhang

Field of Study

Natural Resources: Land, Water, and Air


Doctor of Philosophy

Open Access

Open Access


Theory and previous research suggest that sky obstructions and humidity can degrade global navigation satellite system (GNSS) positioning accuracy for static observation sessions. It is reasonable to suppose that these effects might be even worse for real-time kinematic (RTK) positioning, because the observer will likely collect fewer observations for RTK positioning than for a static occupation, so the statistics used to estimate the positions have fewer data to work with. These effects have not been thoroughly studied, which provided the motivation to conduct several experiments to quantity the effects. Temperature and humidity are variables of interest, so the first experiment establishes whether a digital weather station is an acceptable replacement for a sling psychrometer. The second experiment quantifies RTK positioning accuracy affected by broad-leaf canopy conditions with the effect of ground-level absolute humidity and the effect of sky obstruction as determined using analysis of covariance; this is to study the RTK position-accuracy degradation caused by the water content in the atmosphere and by the possible signal blockage from physical structures around the occupation site. These results were then applied to extend previously published work about a work on habitat-suitability and environmental favorability maps for bentgrass species in Connecticut with logistic regression analysis from GNSS data. The information gained from the experiments was used to study four new biological habitat suitability and environmental favorability models by comparing their strengths and weaknesses using GIS mapping and multiple comparison statistics of the created maps.