Gund Graduate Fellow, Rubenstein School of Environment and Natural Resources

Kristen’s research involves the development of tools to model and evaluate river catchments as complex and dynamic systems.  She applies machine learning algorithms and Bayesian statistics to better understand spatial and temporal patterns in pollutant loading, to inform sustainable design of infrastructure for geomorphic and ecological compatibility, and to guide conservation and restoration activities for reduced flood losses. 

Kristen earned her MS in Geosciences from Pennsylvania State University, and worked as a consulting Professional Geologist prior to founding a small Vermont-based business in 2000 to conduct watershed assessments and river restoration.  In her cross-disciplinary work with local, state and federal stakeholders, she identified a need for advanced data-driven tools to support restoration and conservation decision-making.  This inspired Kristen’s return to graduate school in the UVM Civil & Environmental Engineering program, with a focus on data science.  She is interested in emergent properties and patterns extracted from environmental data that can be linked to drivers of underlying hydrologic and biogeochemical processes and used to guide management and conservation strategies in support of sustainable development.

Dissertation: Computational Tools to Evaluate Sensitivity of River Networks to Natural and Human Disturbance Regimes

Co-Advisors: Donna M. Rizzo and Mandar M. Dewoolkar

Contact

Areas of Expertise and/or Research

River and floodplain dynamics, spatial analysis, geostatistics, Bayesian inference, machine learning, ecosystem services

 

Education

  • MS, Geosciences (Hydrogeology), Pennsylvania State University
  • MS, Geosciences (Hydrogeology), Pennsylvania State University