Steven Crimp

Possible future changes in South East Australian frost frequency: an inter-comparison of statistical downscaling approaches

Anthropogenic climate change has already been shown to effect the frequency, intensity, spatial extent, duration and seasonality of extreme climate events. Understanding these changes is an important step in determining exposure, vulnerability and focus for adaptation. In an attempt to support adaptation decision-making we have examined statistical modelling techniques to improve the representation of global climate model (GCM) derived projections of minimum temperature extremes (frosts) in Australia. We examine the spatial changes in minimum temperature extreme metrics (e.g. monthly and seasonal frost frequency etc.), for a region exhibiting the strongest station trends in Australia, and compare these changes with minimum temperature extreme metrics derived from 10 GCMs, from the Coupled Model Inter-comparison Project Phase 5 (CMIP 5) datasets, and via statistical downscaling. We compare the observed trends with those derived from the “raw” GCM minimum temperature data as well as examine whether quantile matching (QM) or spatio-temporal (spTimerQM) modelling with Quantile Matching can be used to improve the correlation between observed and simulated extreme minimum temperatures. We demonstrate, that the spTimerQM modelling approach provides correlations with observed daily minimum temperatures for the period August to November of 0.22. This represents an almost fourfold improvement over either the “raw” GCM or QM results. The spTimerQM modelling approach also improves correlations with observed monthly frost frequency statistics to 0.84 as opposed to 0.37 and 0.81 for the “raw” GCM and QM results respectively. We apply the spatio-temporal model to examine future extreme minimum temperature projections for the period 2016 to 2048. The spTimerQM modelling results suggest the persistence of current levels of frost risk out to 2030, with the evidence of continuing decadal variation.

99.999 per cent certainty humans are driving global warming: new study

Credit: Skeptical Science under CC BY 3.0 licence Published in the journal Climate Risk Management last week, our research is the first to quantify the probability of historical changes in global temperatures and examine the links to greenhouse gas emissions using rigorous statistical techniques. Our new CSIRO work provides an objective assessment linking global temperature increases to human activity, which points to a close to certain probability exceeding 99.999 per cent. Our work extends existing approaches undertaken internationally to detect climate change and attribute it to human or natural causes. The 2013 Intergovernmental Panel on Climate Change Fifth Assessment Report provided an expert consensus that: