Collaborations

National Aeronautics and Space Administration, Jet Propulsion Laboratory, USA

Uncertainty quantification for analysis and global prediction of carbon dioxide using data from the Orbiting Carbon Observatory-2 satellite
Spatial statistical modelling to estimate CO2 in atmospheric columns at any location on the globe, followed by estimation of near-surface fluxes and a statistical analysis of the uncertainties associated with these estimates. Click here for more details.


University of Missouri, Department of Statistics, USA

Spatio-temporal statistics for US federal agencies, particularly the Census Bureau, and for official statistics surveys
Developing methodological research and applied scientific knowledge for official statistics through the use of hierarchical spatio-temporal statistical modelling. Focus is on the American Community Survey conducted by the US Census Bureau. Click here for more details.


Commonwealth Scientific and Industrial Research Organisation (CSIRO), Australia

Soil-carbon dynamics
Predicting soil-carbon cycling in the presence of uncertainties in data, processes, and parameters. Click here for more details.


Academia Sinica, Institute of Statistical Science, Taiwan

Enhanced False Discovery Rate (EFDR) for detecting change
Transforming images to wavelet space and using EFDR to test the presence of signals in background noise. Click here for more details.


University of Bristol, School of Geographical Sciences, School of Chemistry, and School of Geographical Sciences, UK

Sea-level rise
Assessing the Antarctic contribution to sea-level rise using multivariate spatio-temporal models. Click here for more details.


University of Edinburgh, School of Informatics, UK, and Radboud University, Institute for Computing and Information Sciences, NL

Scalable inference with spatio-temporal models
Using sparse approximations within a message-passing framework in order to scale spatio-temporal inference with large, non-Gaussian spatio-temporal models. Click here for more details.


Texas A&M University, Department of Statistics, USA

Approximations of Gaussian process models for large spatio-temporal datasets
Using the Full-Scale Approximation to spatio-temporal covariance functions in order to scale spatio-temporal Gaussian processes. Click here for more details.

Last reviewed: 18 July, 2016