THE 13th INTERNATIONAL MEETING ON STATISTICAL CLIMATOLOGY
June 6-10, 2016 in Canmore, Alberta, Canada
The Schedule of Oral and Poster Presentations for the 13th International Meeting on Statistical Climatology is available, here.
Below is the list of sessions for the 13th International Meeting on Statistical Climatology with descriptions. See the Registration & Submissions section for instructions on submitting an abstract.
Friederike Otto
It is now widely accepted that the attribution of extreme weather and climate-related events is possible, albeit in a probabilistic sense and recognizing the role of multiple causal factors. Annual assessments of the role of climate change in individual weather and climate events are being compiled using a range of approaches (BAMS:Peterson et al, 2012 & 2013, Herring et al., 2014 &2015) and highlight that event definition and the exact framing of the attribution question are crucial factors determining the result. There is increasing interest in using event attribution in risk assessment, public communication, and, eventually, international negotiations. In order for the science to inform the latter, only attributing the role of climate change is likely insufficient and the relative roles of hazard and vulnerability need to be included when assessing impacts. This session aims to explore new developments in the science of attributing hazards as well as linking meteorological drivers and impacts and would seek to include contributions from the detection and attribution, climate impacts, and disaster risk communities.
Barbara Casati, Barbara Brown, Reto Knutti and Erich Fischer
Weather forecast verification and climate model evaluation are fundamental components of the development and application of environmental prediction systems. For example, process-based and physically meaningful verification diagnostics can provide guidance for model improvements, and robust and informative verification metrics are in demand to monitor progress and enable comparisons of different systems. Moreover, specific information on the strengths and weaknesses of the models can be valuable for forecast/projection end-users and decision-makers, to make optimal use of the forecast and/or climate projections, in responding, as an example, to a high-impact weather warning, or in planning a climate adaptation strategy. The past decade has witnessed the blooming of several novel techniques for weather forecast verification and metrics for climate model evaluation. Despite being focused on addressing the needs of specific (weather or climate) applications, many of these techniques can be envisioned and adapted to serve in a multidisciplinary context. This session welcomes contributions on statistical approaches for both weather and climate model evaluation (and associated applications), with the hope of enhancing scientific exchange and synergies between the weather and climate communities.
Andrea Toreti, Reik Donner and Alex Cannon
Nonlinear statistical and dynamical approaches applied to the characterization and prediction of climate extremes and rare events have been experiencing a growing interest. The introduction of corresponding concepts into the field of statistical climatology and their subsequent application to a variety of climatological problems have been fostered by close interdisciplinary collaborations between climatologists, physicists, applied mathematicians, computer scientists and statisticians, rapid increases in computing power, and developments from the data science community. This session aims at bringing together scientists working both in the development and application of nonlinear methods for the quantitative analysis and modeling of climate extremes. Contributions can be either theoretical (i.e. presenting new methodologies) or applied (i.e. showing well-established methods applied in an innovative way or to new data sets). Perspectives as well as review studies are also welcome. Also of interest is the "cross-pollination" of methods used to analyze extremes in other fields, for example econometrics and other geophysical sciences.
Xiaolan Wang, Lucie Vincent, Markus Donat and Lisa Alexander
The accuracy and homogeneity of climate data are indispensable for many aspects of climate research. In particular, a realistic and reliable assessment of historical climate trends and variability is hardly possible without a long-term, homogeneous time series of climate data. Accurate and homogeneous climate data are also indispensable for the calculation of related statistics that are needed and used to define the state of climate and climate extremes. Unfortunately, many kinds of changes (such as instrument and/or observer changes, and changes in station location and exposure, observing practices and procedure, etc.) that took place in the period of data record could cause non-climatic sudden changes (artificial shifts) in the data time series. Such artificial shifts could have huge impacts on the results of climate analysis, especially those of climate trend analysis. Therefore, artificial changes shall be eliminated, to the extent possible, from the time series prior to its application, especially its application in climate trends assessment.
This session calls for contributions that are related to bias correction and homogenization of climate data, including bias correction and validation of various climate data from satellite observations and from GCM and RCM simulations, as well as quality control/assurance of observations of various variables in the Earth system. It also calls for contributions that use high quality, homogeneous climate data to assess climate trends and variability and to analyze climate extremes, including the use of bias-corrected GCM or RCM simulations in statistical downscaling.
Seung-Ki Min, Sang-Wook Yeh and Wenju Cai
Understanding climate variability and its teleconnection mechanisms in the present climate to the future climate is critical to climate researches. While the climate variability and teleconnections provide a predictability source for intraseasonal-to-seasonal predictions for some regions, their modulation under global warming can increase the uncertainties of climate prediction over the same regions. One important but challenging question is how the modes of internal climate variability and their teleconnection patterns have been changing in the past climate and will change in the future climate. This session solicits recent studies addressing these and associated issues. All studies are welcome such as statistical analyses to better identify the modes of climate variability and teleconnection patterns, modeling studies to evaluate climate models for the simulations of climate variability and its teleconnections, and studies to better understand dynamics and governing mechanisms of the large-scale climate variability and its impacts.
Mohammad Reza Najafi, Upmanu Lall, Vijay Singh and Ashish Sharma
A rapidly growing number of observational datasets relevant to the study of the hydrological cycle have become available in recent years, ranging from ground based to satellite records, reanalyses and model simulations. These datasets provide the opportunity to better characterize the complex variability of the hydrologic cycle, although their spatial and temporal resolution often remains modest in relation to the regional scales at which most hydrologic studies are performed. Observational and process uncertainty, and the stochastic nature of the meteorological processes driving the hydrological cycle at short time scales implies that the study of its components requires the development and improvement of statistical approaches that properly characterize all sources of uncertainties, including observational uncertainty, and faithfully provide stochastic representations of hydrologic variability and processes. This session highlights presentations on the recent advancements in the application of statistical methods in the hydrologic analyses including methods that account for nonstationarity, uncertainty in hydrologic model inputs, structures and parameterizations, post-processing of hydrologic predictions, seasonal forecasts, multi-model approaches, risk and reliability analyses, effect of scaling and projected changes in the components of the hydrologic cycle. Many concerns with nonstationarity, related to climate and also to the modification of basin scale hydrologic processes have been raised. Papers addressing the long term variation of hydrology considering these factors are encouraged. Consistent with the IMSC focus on extremes, studies on all aspects of extremes in the hydrologic cycle, including floods and droughts, are also strongly encouraged.
Douglas Maraun, Trevor Murdock, Megan Kirchmeier-Young and Chris Jacks
Seasonal to decadal climate predictions as well as centennial scale climate projections are often based on rather coarse resolution general circulation models. Many users interested in climate predictions and projections, however, act on regional to local scales and often desire high resolution model output. This demand is especially evident for assessing the occurrence of climate and weather extremes. One way to bridge this scale-gap is by means of statistical downscaling, either by so-called perfect prognosis approaches or model output statistics, including statistical bias correction methods of global and regional climate models.
This session seeks to present and discuss recent methodological and conceptual developments in statistical downscaling. We especially welcome contributions addressing spatial-temporal and multi-variable variability (in particular of extreme events); the development of statistical models for sub-daily variability such as convective events; the integration of process understanding into bias correction methods; the selection of predictors to capture climate change; the performance and added value of downscaling methods on seasonal to centennial scales (including the ability to extrapolate beyond observed values); the development of process-based validation diagnostics for statistical downscaling; and the assessment of advantages and inherent limitations of different approaches. Mere applications will not be considered.
Nathan Gillett and Hideo Shiogama
Detecting and attributing human influence on extreme events and other climate variables remains a key policy-relevant topic in assessments of the Intergovernmental Panel on Climate Change (IPCC) and in the literature. The sixth phase of the Coupled Model Intercomparison Project (CMIP6) which is coordinating the climate model simulations underlying the next IPCC assessment, includes a sub-project focused on detection and attribution of climate change, DAMIP. Novel aspects of DAMIP compared to CMIP5 include future greenhouse-gas-only simulations to allow the future responses to greenhouse gases and other anthropogenic forcings to be separated and constrained using observations, simulations allowing the responses to greenhouse gases and other forcings to be separated in models with coupled chemistry, and perturbed forcing simulations to sample over forcing uncertainty. We invite abstracts on the experimental design of DAMIP and on methodological advances and new applications in climate change detection and attribution.
Gabi Hegerl, Sonia Seneviratne, Lisa Alexander, Xuebin Zhang and Francis Zwiers
Evidence is clear that the mean state and extremes of climate are changing. Increasingly, as extremes affect many aspects of our society, the climate community is being asked to provide skillful predictions and reliable projections of extremes at time scales from days to seasons and centuries and to explain the causes of recent extreme events. Significant advances in modelling, understanding and data availability are required to better identify and understand the multi-facet factors and mechanisms that determine the location, intensity, and frequency of climate extremes. Important improvements are underway, for example, due to high resolution modelling advances and improved physical understanding of feedbacks. The World Climate Research Program’s (WCRP) Grand Challenge on Understanding and Predicting Weather and Climate Extremes provides a framework to promote and to coordinate research on extremes and identify areas where quick progress is possible. This session aims to explore new developments related to the evaluation of processes including atmospheric dynamics and land surface feedback generating extremes in models, of teleconnections and their links to extremes in observations and models, of quantification and attribution of large scale changes in extremes, and evaluation of model ability to attribute extreme events. We invite particularly, but not exclusively, work that cuts across these areas. Research exclusively on event attribution and its framing is directed towards the session on event attribution, and work that exclusively focuses on model evaluation on the session of model evaluation for extreme events
Jana Sillmann, Philipe Naveau and Erich Fischer
This session focuses on advanced statistical methods to evaluate weather and climate extremes in climate model simulations. The combination of climate model simulations with observational and reanalysis products requires innovative analyses and metrics of performance that make use of information related to processes generating internal climate variability and amplifying weather and climate extremes. The sensitivities of model evaluation results to the chosen extremes index, reference dataset and considered region are likely related to model deficiencies in representing the spatio-temporal characteristics of physical processes and feedback mechanisms relevant to the occurrence of extremes, both in the climate models and in the weather forecasting models used in reanalyses. This session invites contributions that focus on the evaluation of the dependence of the location, magnitude, duration and frequency of extremes upon ambient climatic conditions. This includes, for instance, model evaluation methods that compare distributions rather than deterministic values and that are able to capture spatio-temporal characteristics; contributions that investigate classical performance metrics for univariate extremes and extend univariate performance metrics to multivariate; and contributions that discuss approaches to assess weights to individual models within large multi-model ensemble frameworks.
Seth Westra and Philipe Naveau
Extreme value theory provides a class of approaches for modelling the extremes of a random sample, enabling finite records to be used to estimate the probability of outcomes that may be more extreme than those that have been observed. Significant advances recently have been made in the statistical theory of extremes, particularly in the context of modelling multi-variable or ‘compound’ extremes (e.g. using multivariate extreme value models such copula-based approaches), spatial extremes (e.g. max-stable models, and latent variable models including Bayesian hierarchical approaches) and time-varying extreme value distributions. This session welcomes contributions describing recent theoretical developments, as well as practical implementations of extreme value approaches including intensity-duration-frequency (IDF) analysis, flood frequency analysis, and detection and attribution of change in hydrological and meteorological extremes. .