Feasibility Project awarded funding by SECURE

 Statistical software to identify spatiotemporal patterns and coherence over river networks

 Dr Claire Miller1  (PI), Prof. Marian Scott1, Kelly Gallacher1, Mr John Douglass2, Dr Linda Pope2, Dr Mark Hallard3, Dr Graeme Cameron3 and Dr Robert Willows1

 1School of Mathematics and Statistics, University of Glasgow, 2 Environment Agency (EA), 3Scottish Environment Protection Agency (SEPA)

 Environment agencies invest a wealth of resource in monitoring river nutrient levels in order to protect water quality and for compliance reporting to Europe. Advanced statistical models incorporating the river network structure can aid  in the  identification of common spatiotemporal nutrient patterns and potentially provide efficiency savings in the monitoring budget. In this project, statisticians and environmental scientists from The University of Glasgow, the Scottish Environment Protection Agency (SEPA) and the Environment Agency (EA) will collaborate to develop and implement novel statistical software tools, for use within the environmental science (ES) community, to identify spatiotemporal and coherent river water quality patterns. Implementation of the software and benefits of the methods will be demonstrated to the ES community through a workshop in Spring 2016.

 The project will run from 1st November 2015 until 30th April 2016 and the objectives are:

  • to identify the features of the developing statistical methodologies for identifying spatiotemporal patterns within river networks which are of greatest relevance and utility to the ES partners;
  • to develop statistical software tools in the R programming language (R Core Team,2014, www.R-project.org),  and  associated  documentation, to apply the identified statistical approaches to new data sets and demonstrate their utility and benefits to ES partners;
  • to demonstrate the integration of these tools within TIBCO Spotfire (http://spotfire.tibco.com) commonly used software within SEPA), through the use of the TIBCO Enterprise Runtime for R engine.

Final report and talk given at SECURE Annual Conference 2016 can be viewed below: