DrMatthew Clark
Honorary Research Associate
Centre for Environmental Policy - Faculty of Natural Sciences
Orcid identifier0000-0002-3217-1192 (opens in a new tab)
- Honorary Research AssociateCentre for Environmental Policy - Faculty of Natural Sciences
- 503, Weeks Building, South Kensington Campus, United Kingdom
RESEARCH
The focus of my collaborations at Imperial are to understand how some conservation interventions "scale out," or spread from person to person, local to regional scales, or across institutions. This project is led by Dr. Morena Mills (Centre for Environmental Policy at Imperial College) and Dr. Arundhati Jagadish (Nature Conservation Foundation), and largely relies on empirical data gathered from ongoing projects run by Conservation International. The end goal of this project is to deliver a predictive model to forecast the long-term adoption rates of various community-based conservation initiatives in a variety of locations across the globe. This project is funded by the Leverhulme Trust-funded grant: The race to environmental sustainability
The main methods I use to collect data for my research are satellite-based remote sensing, household surveys, participatory mapping, and accessing data from a variety of publicly available sources. I analyze these data using hierarchical Bayesian regression approaches and by assessing them as consistent with synthetic data produced via agent-based simulations. I am currently interested in developing methods to perform model selection for alternative agent-based simulations using approximate Bayesian computation.
The main methods I use to collect data for my research are satellite-based remote sensing, household surveys, participatory mapping, and accessing data from a variety of publicly available sources. I analyze these data using hierarchical Bayesian regression approaches and by assessing them as consistent with synthetic data produced via agent-based simulations. I am currently interested in developing methods to perform model selection for alternative agent-based simulations using approximate Bayesian computation.