A method of applying daily GCM outputs in assessing climate change impact on multiple day extreme precipitation for Brisbane River Catchment

Abstract

Urban river floods are normally caused by consecutive days of extreme precipitation. It is expected that climate change will have significant impact on extreme precipitation because the warming atmosphere will alter the precipitation pattern by changing the global/regional hydrological cycle. The Intergovernmental Panel on Climate Change (IPCC) finds that there is a tendency for an increase in daily heavy precipitation events in many regions in the world, including some in which the mean precipitation is projected to decrease. Currently, the General Circulation Model (GCM) is still the most reliable tool for generating the future climate change scenarios, but GCM is still facing with the problems in simulating daily precipitation at regional or local scales, particularly in extreme precipitation events simulation, due to its coarse spatial resolution and the current incomplete understanding of the climate system. The poor performance in regional/local precipitation simulation makes it difficult of direct using GCM outputs in climate change impact on extreme precipitation change studies, because extreme precipitation event is most likely a localised phenomenon. Recently, research efforts have been put into the downscaling the GCM to support localised impact assessment. However, such downscaling models are either complex so computational demanding (dynamic downscaling) or require extensive of observed data (statistical downscaling). That has leads to a limited availability of model results which in most cases are not sufficient to fulfil a localised impact assessment needs. Since the publication of IPCC AR4, many GCM daily simulation outputs have become publically available, which provide an opportunity to study the change impact on total precipitation amount at daily bases, either for one day or multiple days. Given that the direct application of GCM simulation data in assessing climate change impact on extreme precipitation is yet to be examined, the important question is then how to make use of the daily GCM results in order to obtain local daily precipitation statistics and their changes in the future. This is of major importance for the extreme precipitation properties since the upper tail of the precipitation distribution suffers most from the coarse resolution representation in the GCMs. Thus the focus of this study was given to examine the climate change impact on the extreme precipitation by linking the GCM daily simulation results with the local extreme precipitation observations based on a statistical approach. Due to the current incomplete understanding of the climate system, the GCM's precipitation generation mechanism is associated with high uncertainties. Pattern scaling has been proved to be an economic and efficient method to cope with such uncertainties in generating the range of future climate change scenarios from different GCMs. Based on this method, together with a GCM-ensemble probabilistic prediction, we developed a future extreme events generation method that is capable of addressing the range of uncertainties that caused by different GCM precipitation generation mechanisms. The method was applied to the study the climate change impact on the Brisbane River rainfall and flooding. Since the historical flooding in Brisbane River were caused by multipleday (typically 4 - 7 consecutive day) extreme rainfall, the method was applied to analysis the 5-day maximum total rainfall for the 17 long term historical observation stations in the catchment. The results reveal a reinforced trend toward more intense extreme precipitation events into the future under climate change. The shortening of return periods for extreme precipitation events and greater intensity of such events has implications for planning and decision making of durable infrastructure along with emergency services planning, landuse regulation and building codes. This relates not only to possible flood mitigation strategies such as the potential need for additional flood mitigating infrastructure but also for the current built environment.