Our research employs mechanistic and empirical modeling approaches for assessing the severity and causes of environmental impairments, particularly those related to surface water quality.  A common theme of our research is to provide rigorous uncertainty quantification, so that policy makers and the public can be presented with the ranges of likely outcomes associated with different future scenarios, allowing for more informed decision-making.  Model uncertainty quantification is also critical to the scientific community, as it provides for rigorous hypothesis testing and suggests where additional research and data collection would be most beneficial.

Our research also aims to reduce the predictive uncertainty of models by more effectively leveraging available information, such as field monitoring data, satellite imagery, and results from previous experiments and related biophysical modeling studies.  This auxiliary information is incorporated through various methods, such as the geostatistical fusion of multiple spatial data layers, and the specification of prior probabilities and multiple calibration endpoints using Bayesian statistics.