Dmitri Kavetski

Biography

One of the main contributions of Dmitri's research work has been the development of Bayesian Total Error Analysis (BATEA) - a comprehensive framework for parameter estimation and probabilistic prediction accounting for data and model uncertainties. BATEA is currently applied in hydrological modelling, with additional appplications developing in river system modelling, irrigation modelling and other areas of environmental engineering. Dmitri's broader interests include mathematical modelling of surface and subsurface hydrological systems, uncertainty estimation, the development of numerically robust rainfall-runoff and snow models, design of accurate and computationally efficient numerical algorithms for nonlinear differential equations and nonlinear optimization. During his postdoctoral research at Princeton University, Dmitri also made contributions to risk assessment of carbon geosequestration scenarios. Dmitri's international collaborations include the National Center for Atmospheric Research (NCAR, Boulder, USA), Swiss Federal Institute for Aquatic Science and Technology (EAWAG, Switzerland), CEMAGREF (Paris and Lyon, France), Centre Recherche Public Gabriel Lippmann (CRP-GL, Luxembourg), and other institutions worldwide. From 2012 onwards, Dmitri joined the School of Civil, Environmental and Mining Engineering at the University of Adelaide.

Research Interest

Bayesian inference and prediction in hydrology and environmental modelling. Development of novel Bayesian and Monte Carlo techniques for parameter estimation, uncertainty analysis and predictive applications. Areas of application have included hydrological models and river system models. Mathematical modelling in hydrology and environmental engineering. Selection of governing equations and process representations in hydrological and snow models, including scientifically defensible hypothesis testing. Model development, optimisation and testing. Applied numerical and statistical analysis in environmental engineering. Design and implementation of accurate, robust and computationally efficient numerical algorithms and software. Including solution of nonlinear differential equations, numerical integration, nonlinear optimisation problems, and others. Areas of application have included Richards equation for groundwater simulations, rainfall-runoff models, CO2 geosequestration models, and others.