Understanding Quaternary Fluvial Landscape Evolution: The challenge of combining observations and models Seminar

Quaternary fluvial archives contain a vast array of data which could be used to test increasingly sophisticated numerical landscape evolution models (LEMs). Unfortunately the existing generation of LEMs can only produce generalised outputs relating to sediment budgets, which give only an impression of the response of a catchment to different frequency-magnitude discharge regimes. These outputs are not accurately testable against the sedimentary record, which has led many to abandon hope of achieving model validation and to settle for the less satisfactory model corroboration goal. Such generality ultimately however, raises questions concerning the ‘meaning’ of model outputs.

What is needed is a new dialogue between the archive and modelling communities, with the aim of arriving at a set of common ‘attributes’ which can be measured in the field and produced directly from the models. This talk will discuss the current problems surrounding data-model comparison and signpost how the communities might address these problems.