Application of Advanced Software Engineering Tools and Methods in the Environmental Sciences


Grodon Blair, Lancaster University, UK
Benoit Combemale, University of Rennes 1, IRISA, France


Computational methods have advanced significantly over the last decade or so in support of earth and environmental sciences. For example, significant advances have been made in the performance and availability of High Performance Computing facilities or cloud computing resources. There has also been significant advances in running potentially complex environmental models over such computational resources, and in representing uncertainty in such computations. Nevertheless, a significant problem-implementation gap still exists whereby developers have to implement software using abstractions that are at a lower level than those used to express the problem. Put simply, scientists often spent far too much time hacking Linux scripts for instance, when they would much rather be focusing on their science and their modelling experiments. In contrast, there has recently been significant advances in software engineering that have great potential to reduce this problem-implementation gap. Examples of areas of innovation include model-drtiven engineering, software frameworks, software and infrastructure as a service, and advances in middleware.

This unique workshop aims to bring together researchers from the fields of environmental sciences and software engineering/computer science to look at new techniques to support computational methods, particularly related to water resources. The workshop will i) seek to understand the requirements in terms of contemporary computational methods, ii) explore cutting edge techniques in software engineering, and iii) investigate the potential of emerging techniques to address the core challenges in computational methods in the context of water resources. The format of the workshop will reflect these three goals, with plenty of scope for discussion. The expected outcomes will be a research roadmap to reduce the problem-implementation gap, the stimulation of exciting new collaborations around this roadmap, and a greater awareness of contemporary software engineering practices that are supportive of the sciences.

Comments are closed