GC13B-01 INVITED
Extreme Precipitation over the Continental United States
In recent years there has been much climatological literature devoted to the apparent increase in extreme precipitation events that may be a consequence of broader trends in climate due to increased greenhouse gases. However, most current papers still use rather simple statistical methods to study this phenomenon. Here, we apply state of the art extreme value methods based on exceedances over high thresholds, including covariates to represent trend and seasonality, to estimate 25-year return levels, and trends in those return levels over 1970-1999. This is done separately for nearly 5,000 stations in the US climatological network, then the results are combined across stations using spatial statistics. The results provide a detailed picture of how trends in rainfall extremes have varied across the United States, as well as summary statistics computed for 19 regions. They indeed confirm an overall increase in extreme rainfall levels, but it is by no mean homogeneous across the whole country. Separate analysis based on model runs from NCAR's Community Climate System Model provide a first insight into how such changes may project into the future.
GC13B-02
Precipitation Extremes as a Function of Resolution in the PRUDENCE Regional Climate Model Setup
The regional climate model HIRHAM has been used in resolutions of 50km, 25 km and 12km to perform 30-year long time slice climate change simulations in the setup of the European Union project PRUDENCE (http://prudence.dmi.dk). This project encompassed a systematic investigation of 10 different European regional models, which in one set of simulations were driven by one common set of boundary conditions. These came from the global atmospheric model HadAM3H and covered the time slices 1961-1990 and 2071-2100 according to the SRES scenario A2 for an area covering Europe and surrounding areas. This systematic regional model intercomparison has been continued in the ongoing ENSEMBLES project (http://www.ensembles-eu.org). A comparison of the HIRHAM results with observations shows that the higher-resolution simulation is more realistic. This applies to orographic rainfall and land-sea contrasts, but also to extremes, even when data are aggregated to the coarser-resolution grid. Climate change is shown to affect extreme precipitation to a higher degree than mean values for Europe in accordance with the conclusions reached in the PRUDENCE project; during summer, the average precipitation in Central and Southern Europe decreases by up to around 40%, but extreme precipitation is nevertheless generally increasing. It was a general feature of PRUDENCE simulations that summer extremes had a more positive climate change signal than the summer average, even though the absolute sign of the change in extremes varied. The conclusion of positive changes in extremes is very clear in the high-resolution simulation. Results from a pilot study of hourly precipitation shows that similar conclusions hold as in the case of daily precipitation.
GC13B-03
Assessing trends in observed and modelled climate extremes over Australia in relation to future projections
Multiple simulations from nine global coupled climate models were assessed for their ability to reproduce observed trends in a set of indices representing temperature and precipitation extremes over Australia. Observed trends over the 1957 to 1999 period were compared with individual and multi-modelled trends calculated over the same period. When averaged across Australia the magnitude of trends and interannual variability of temperature extremes were well simulated by most models particularly for the warm nights index. The majority of models also reproduced the correct sign of trend for precipitation extremes although there was much more variation between the individual model runs. A bootstrapping technique was used to calculate uncertainty estimates and also to verify that most model runs produce plausible trends when averaged over Australia. Although very few showed significant skill at reproducing the observed spatial pattern of trends, a pattern correlation measure showed that spatial noise could not be ruled out as dominating these patterns. Two of the models with output from different forcings showed that the observed trends over Australia for some of the temperature indices were consistent with an anthropogenic response but were inconsistent with natural-only forcings. Future projected changes in extremes using three emissions scenarios were also analysed. Australia shows a shift towards warming of temperature extremes particularly a significant increase in the number of warm nights and heat waves with much longer dry spells interspersed with periods of increased extreme precipitation irrespective of the scenario used.
GC13B-04
Severe Extra-tropical Storms Under Climate Change And Related Impacts
Winter storms caused by extra-tropical cyclones over the Northeast Pacific and the Northeast Atlantic basins are important factors for property losses caused by natural hazards over Europe and North-America. The European storm series in early 1990 and late 1999 led to enormous economic damages (US-$14.2 bn and $18.5 bn, respectively) and insured claims (US-$9.8 bn and $10.75 bn, respectively). Although significant trends in North Atlantic / European storm activity have not been identified for the last decades, this study provide evidence that under anthropogenic climate change the number of extreme storms could increase, whereas the total number of northern hemispheric extra-tropical cyclones may be slightly reduced. This holds true for the Northeast Pacific as well as the Northeast Atlantic basin. The results from global climate models are well recognised in wind speed analyses from regional climate models. For parts of western Central Europe an increase in frequency and intensity of extreme wind speeds are identified. In this context, the analysis of climate models from the ENSEMBLES initiative offers the unique opportunity to investigate model to model variability for GCM and RCM simulation in more horizontal detail, leading thus to measures of uncertainty. Additionally, loss potentials derived from an ensemble of global and regional climate models using a simple storm damage regression model under climate change conditions are presented. For the two European regions (United Kingdom and Germany) ensemble-mean storm-related losses are investigated. Based on GCMs the ensemble mean is found to possibly increase by up to 37%. Furthermore, the interannual variability of extreme events will increase leading to a higher risk of extreme storm activity and related losses. In order to gain more regional information, RCMs have been forced with ECMWF-ERA40 for validation, and with several GCMs under IPCC SRES scenarios for future conditions.
GC13B-05 INVITED
Has Anthropogenic Forcing Caused a Discernible Change in Atlantic Hurricane Activity?
There is currently evidence both for and against the existence of a discernible anthropogenic impact on Atlantic hurricane activity. Emanuel's (pers. comm. 2007) Power Dissipation Index shows unprecedented high values in recent decades in the context of the past 60 yr, and correlates remarkably well with low-frequency tropical Atlantic SST variations. The limited record length, partial basin coverage by aircraft in the pre-satellite era, and lack of reconciliation with models limit the usefulness of this result for identifying possible anthropogenic influences. Landsea (EOS, 2007) uses landfalling storm statistics to infer no significant increase in basin-wide tropical storm counts since 1900. Landsea's critical assumption of a constant landfalling fraction over time limits confidence in this assessment. Nonetheless, an important finding is that U.S. landfalling hurricane activity (frequency and PDI) show no increasing trend over the past century or so. Holland and Webster (Phil. Trans. R. Soc. A 2007) conclude that basin-wide tropical cyclone and hurricane counts have increased dramatically during the past century, related to the rise in tropical Atlantic SSTs. Their key assumption is that the existing HURDAT data reliably portrays basin-wide statistics for tropical storms, hurricanes and major hurricanes, at least back to ~1900, which requires further substantiation. We use historical Atlantic ship track and storm track data to estimate the expected number of missing tropical storms each year in the pre-satellite era (1878-1965). After adjustment, the storm counts covary with tropical SSTs on multi-decadal time scales, but their long-term trend (1878-2006) is weaker than the trend in similarly normalized SSTs (though both are nominally positive). The linear trend in adjusted storm counts for 1900-2006 is strongly positive (+4.2 storms/century) and highly significant even after accounting for serial correlation. However, this trend begins near a local minimum in the time series and ends with the recent high activity, perhaps exaggerating the significance of the trend. The trend beginning from 1878 is weakly positive, and not statistically significant with p=0.3. The uncertainty in the late 1800s is larger than that during the 1900s--an important caveat on the results using the earlier start date. Tropical cyclone occurrence rates appear to have decreased in the western part of the basin (consistent with declining U.S. landfalling hurricane counts) but may have increased slightly in the central and eastern basin, suggesting a structural change such as shifts in storm tracks. Important assumptions of our methodology, such as that all landfalling storms since 1878 were detected and reported, require further investigation. In an attempt to reconcile the past observations with models, we have developed a regional modeling framework for downscaling Atlantic hurricane activity. Given observed large-scale atmospheric conditions and SSTs from reanalyses, the model reproduces several aspects of past Atlantic hurricane behavior (1980-2006). However, much further work is needed to produce simulations where hurricane activity changes can be confidently attributed, using such models, to various anthropogenic forcings or natural processes. Based on available evidence, we cannot yet conclude with high confidence that anthropogenic forcing has caused a discernible anthropogenic influence on hurricane activity to date.
GC13B-06
Simulations of the Synoptic Climatology of Extreme Precipitation for Current and Future Climates
The skill of regional climate models (RCMs) in simulating the synoptic conditions associated with extreme regional precipitation is assessed using observations from co-operative network observing sites and model results from 10-year RCM simulations of present and future-scenario climates. We focus on extreme daily precipitation events in the Upper Mississippi River Basin region during the cool season (September-March). Extreme regional events are defined as intensities in the top 0.05% that cover several observation sites or model grid points. For both the observed and simulated contemporary climate, nearly all extreme regional events occur when a slow moving, cutoff-low system develops over the Rockies and Great Plains and continually transports moisture into the Upper Mississippi region from the Gulf of Mexico. Results for the future-climate show similar circulation behavior for corresponding extreme events. The magnitude of daily precipitation in future-climate regional extreme events increases by 26%, which considerably exceeds the 16% increase in overall average daily precipitation and suggests that under climate change a greater fraction of cool-season precipitation will occur during extreme events. The results show the potential for RCMs to replicate the synoptic conditions associated with extreme regional precipitation events, supporting their use for projecting future changes in extreme events. The results also suggest robust circulation behavior for such extreme events even under climate change.
GC13B-07
Diagnosing El Niño Induced Drought in Seasonal Forecasts and Climate Change Projections
El Nino brings widespread drought to the tropics. Stronger or more frequent El Niño events in the future will exacerbate drought risk in already highly vulnerable areas. Even if the frequency and intensity of El Niño events do not increase in the 21st century, more generalized warming of the tropical Pacific may still produce a tropical teleconnection resembling that associated with present-day El Ni~o conditions. In this work, the patterns, spatial extent, and severity of El Ni~o induced tropical droughts are evaluated for a control period in the 20th century in seasonal forecasts, which have updated realistic initial conditions but fixed greenhouse gases (GHGs), and climate change projections, which have realistic GHG evolution but no observational updates. We examine the strengths and weaknesses of the model responses to large scale changes in sea surface temperature (SST) such as those due to ENSO or to climate change. The concern is that if the impact on precipitation fields from anomalous SST forcing on seasonal timescales is poorly simulated in climate change projections, then the impact on such precipitation fields from increasing greenhouse gas forcing cannot be trusted either. More optimistically, if the models contain robust information, and differences are due mainly to systematic biases, then the predictions/projections can be spatially recalibrated to provide more confident estimates of near-term and longer-term drought risk within the tropics. The results have implications for 21st century projected changes in the strength of the identified patterns of tropical drought.
GC13B-08 INVITED
Near-real-time attribution of extreme weather events
As the impacts of global climate change become increasingly evident, there is growing demand for a quantitative and objective answer the the question of what is "to blame" for observed extreme weather phenomena. In addition to considerable public interest, understanding how external drivers, particularly secular trends such as anthropogenic greenhouse gas forcing, is important for the correct quantification of current weather-related risks for the insurance industry. We propose a method of quantifying the contribution of external drivers to weather-related risks based on a twinned ensemble design. Under this approach, a large ensemble of simulations with a forecast-resolution atmospheric model is driven with observed sea surface temperatures and atmospheric composition over the period of interest. A second ensemble is then generated with the influence of a particular external agent, such as anthropogenic greenhouse gases, removed through modification of composition and surface temperatures. Conventional detection and attribution techniques are used to allow for uncertainty in the magnitude and pattern of the signal removed. The frequency of occurrence of the weather event in question can then be compared between the two ensembles. For the exploration of changing risks of the most extreme events, very large ensembles (thousands of members, unprecedented for a model of this resolution) are needed, requiring a novel distributed computing approach, relying on computing resources donated by the general public: see http://attribution.cpdn.org. We focus as an example on the events of Autumn 2000 which brought widespread flooding to many regions of the UK. Precipitation from the twin ensembles is used to force an empirical run-off model to provide an estimate of its contribution to flood risk. Results are summarized in the form of an estimated fraction attributable risk for the anthropogenic contribution to the flooding events of that year. http://attribution.cpdn.org