Improving projections of rainfall trends through regional climate modeling and wide-ranging assessment
Grose, MR and Corney, SP and Katzfey, JJ and Bennett, JC and Bindoff, NL, Improving projections of rainfall trends through regional climate modeling and wide-ranging assessment, SUSTAINING OUR FUTURE: understanding and living with uncertainty, 12-16 December 2011, Perth, Western Australia, pp. 2726-2732. ISBN 978-0-9872143-1-7 (2011) [Refereed Conference Paper]
General Circulation Models (GCMs) are our best tool for assessing potential changes to our
climate on a global scale into the future. However, certain physical processes that influence rainfall at any
particular location operate at finer spatial scales than can be simulated by GCMs. Dynamical downscaling
using regional climate models (RCMs) addresses this problem by simulating the relevant processes at finer
scales, whilst retaining the important large-scale features of the original GCM. Assessment of the RCM
simulations is an essential task before using them as a guide for potential future rainfall trends. Assessing the
outputs of models both in terms of their simulation of relevant output variables (e.g. rainfall) and also the
dynamics gives a more comprehensive assessment than examining the variables alone. Similarly,
understanding the dynamics driving the projected trends from the model simulation can be used to gauge the
plausibility of that projected trend.
Here we present dynamically downscaled simulations of rainfall over the study site of Tasmania, Australia
(~10 km grid scale). These fine-scale projections were produced using a dynamical downscaling regional
climate model. Rainfall in the simulations is validated against a gridded climate dataset based on
observations, and the model simulation of broad climate fields such as mean pressure and wind fields were
assessed by comparison to reanalysis datasets. Additionally, the projected rainfall trends are interpreted in
terms of known climate drivers.
While GCM projections to 2100 show fairly uniform rainfall trends over all of Tasmania, the RCM
projections reveal distinct trends in the different districts of Tasmania. The spatial pattern of rainfall changes
also varies greatly in each season. The change in each district is driven by a unique combination of drivers
and processes. The main climate drivers of change include the alteration to mean circulation in response to
surface warming, changes to the strength and position of the subtropical ridge, shifts in atmospheric blocking,
the southern annular mode, as well as changes to the synoptic climatology of significant systems. At a scale
relevant to Tasmania the climate response to these processes is often poorly resolved in GCM simulations,
and is improved through regional modelling. In this paper we present a holistic view of rainfall changes over
Tasmania, linking large-scale drivers and finer scale processes to spatial and temporal changes in rainfall.
Such analyses are only possible with fine-scale dynamical downscaling. We conclude that including new
information from increased resolution using fine-scale dynamical downscaling provides more useful
projections of rainfall changes at a local scale. We propose that analysis of the relevant rainfall mechanisms
also helps to assess the confidence in which the climate simulations should be used as a tool for
understanding future changes in rainfall.