Variable Bounds Analysis of a Climate Model Using Software Verification Techniques

Authors

Peter Revesz, University of Nebraska-LincolnFollow
Robert Woodward, University of Nebraska–LincolnFollow

Date of this Version

5-1-2014

Citation

P. Z. Revesz, R. Woodward, Variable bounds analysis of a climate model using software verification techniques, Applications of Information Systems in Engineering and Bioscience (includes Proc. 13th International Conference on Software Engineering, Parallel and Distributed Systems), J. Balicki, ed., WSEAS Press, pp. 31-36, 2014.

Comments

OPEN ACCESS

Abstract

Software verification techniques often use some approximation method that identifies the limits of the possible range of values that variables in a computer program can take during execution. Current climate models are complex computer programs that are typically iterated time-step by time-step to predict the next value of the climate-related variables. Because these iterative methods are necessarily computed only for a fixed number of iterations, they are unable to answer many long-range questions that may be posed regarding climate change, for example, whether there are natural fluctuations or whether a tipping point is reached after which there is no return to normal. In a departure from the usual step-by-step climate models, we propose to use software verification techniques to predict absolute bounds, i.e., predict maximum and minimum values that could ever arise according to a climate model. Hence we can answer some of the long-range climate prediction questions that were not possible to answer before.