Used spatial data, GIS, and R to investigate the causes and consequences of rarity in 14 species of the desert plant Encelia.Duties:- Applied data cleaning methods to filter erroneous geographic coordinates, ensuring accurate distribution maps.- Automated mapping in R with ggplot2 and geographic boundary layers to optimize parameters for precise species ranges.- Used resulting maps to determine which species were common versus rare.

Used machine learning (XGBoost, Random Forest Classification, Boosted Regression Trees), ArcGis Pro, and geospatial data to develop predictive models for arsenic competition in groundwater.Duties:- Designed models to assess risk of exposure for each county in the Colorado Plateau.- Evaluated which machine learning model most accurately predicted arsenic contamination.- Extracted/cleaned water quality data for GIS; compiled covariate layers: hydrology, geology, and soils into a raster surface in ArcGIS Pro. Extracted predictor values for each well location.- Prepared data for models and created arsenic-level training/testing datasets.- Tuned model hyperparameters using Supercomputing Cluster and 10-fold cross-validation for optimal accuracy.

- Reviewed the literature to develop a humane pain assessment plan for Black soldier fly larvae.- Performed and analyzed nociception tests on larvae. Assessed humane euthanization methods using particle plate sizes in a meat grinder.- Co-Authored in published work

Researched sexual dimorphism in wing patterns and forewing size in common buckeye butterfly (Junonia coenia) under different lighting.Duties:- Reared butterflies across 40 specialized enclosures, which required tracking lifecycles and upkeeping host plants- Used ImageJ to measure forewings, a spectrophotometer for reflectance analysis, and RStudio for data analysis and statistics (chi-squared, Generalized linear model)