Predictive uncertainty in water resource assessments

Abstract

Water resource management The statement that ‘there are three kinds of lies: lies, damned lies and statistics’ has been attributed to the 19th century British Prime Minister, Benjamin Disraeli (1804-1881) to convey suspicion about the use of statistical models to support arguments in the face of little data. There is little doubt that examples of the unscrupulous use of predictive models (whether statistical or other types of models) can be found throughout the history of science, however, we have also come to rely on models to organise and manage or lives. Nowhere is this truer than with the sciences that involve natural environmental systems (including hydrology) which are complex and difficult to measure. While most hydrologists and water resource engineers would get (justifiably?) upset if their outputs were considered little better than clever statistics (and therefore no better than lies according to Disraeli), they would all be willing to admit that the outputs of their models contain uncertainties.

Authors

Categories