Hi, there.

I am a Postdoctoral Fellow at the Johns Hopkins Whiting School of Engineering working with Dr. Paul Ferraro. From 2019 to 2021, I was a Harvard Data Science Initiative Postdoctoral Fellow at the Harvard TH Chan School of Public Health. I earned a PhD in Biology and MS in Statistics from Auburn University.

I focus on the development and application of robust statistical and causal methods to estimate the consequences of environmental change, evaluate the role of interventions in modifying human and animal behaviors, and quantify the causes and effects of environmental and social changes on ecological, human, and planetary health.
Download CV

An overview of my research


Causality in ecology

Causal diagram of hypothesized causal relationships between market value, catch, and population size of commercially important marine fish.
Integrating and advancing causal inference methods for estimating nonlinear causal relationships in complex ecological systems

(Image: Causal diagram illustrating hypothesized causal relationships between market value, catch, and population size of a commercially important marine fish species)
Robust estimation for non- and semi-parametric models
Plots of coefficient functions of a single-index varying coefficients regression model
Developing and applying robust nonparametric statistical procedures for nonlinear modeling of climate change effects

(Image: Coefficient functions of a single-index varying coefficients regression model estimated using rank-based estimation with group LASSO selection)
Spatiotemporal effects of climate change on population dynamics
Pacific halibut swimming near the bottom of the ocean; Eurasian reindeer grazing in snow in Tromsø, Norway
Modeling spatial and temporal effects of ecological, environmental, and climatic change on population and species dynamics of marine and terrestrial species at regional to continental scales

(Image: Pacific halibut, left, and Eurasian reindeer, right)
Causal and machine learning methods for explaining persistence of agricultural practices
Decision tree showing how potential users transition into persisters or non-persisters over time
Predicting and evaluating the persistence of conservation practices, such as cover cropping, in agriculture using machine learning and causal inference approaches

(Image: Transition of conservation practice adoption to persistence over time, including churn of users dropping below a minimum threshold for persistent use of a conservation practice)
Examining the phenomenon of multiple paternity with an appropriate null model
Estimated frequency of multiple paternity across litter or broos sizes ranging from 2 to log(10) for a variety of species across four taxa
Examining evolutionary and environmental constraints to multiple paternity across taxa by contrasting empirical data on multiple paternity to estimates under a null model

(Image: Bayesian MCMC predictions of the frequency of multiple paternity across litter/brood sizes for species belonging to four taxa (colored circles). The solid red line represents the frequency of multiple paternity estimated using a zero-truncated binomial distribution)
Interpretable machine learning methods for identifying causes of HIV-related stigma
Map of East Africa with regions shaded to represent the level of association between formal education achievement and HIV-related stigma
Developing novel workflows combining interpretable machine learning methods and causal reasoning for identifying causes of and interventions for HIV-related stigma

(Image: Association between formal education and HIV-related stimga in Eastern Africa from Demographic and Health Survey data)