EUPORIAS periodic report
The European Commission has commissioned a four year project – EUropean Provision Of Regional Impacts Assessment on Seasonal and decadal timescales (EUPORIAS) - through its Framework Programme 7 (Grant Agreement number 308291). The main aim of the project is to demonstrate that the development of a suitable interface between users and providers of climate information can increase its societal usefulness, and ultimately strengthen the resilience of European society to climate variability and change. Whilst similar concepts would be applicable on a variety of time-scales, EUPORIAS focuses explicitly on climate predictions from one month to five years into the future.
The lack of a climate change signal emergence, and the large, unforced variability of the climate system in European latitudes, makes this one of the most challenging time-scales to work on. However the seasonal to decadal time-scale is also one of the most interesting from a decision-making point of view, as the effects, and hopefully benefits, of these decisions are realised relatively quickly by all involved.
This project began on 1 November 2012, coordinated by the United Kingdom’s Met Office, in conjunction with a consortium of 24 organisations representing world class European climate research and climate service centers; expertise in impact assessments and seasonal predictions; two United Nations agencies; specialists in new media (such as Twitter, Facebook, smartphones, and YouTube); and commercial companies in climate vulnerable sectors such as energy, water, and tourism. In addition, the consortium has strong links with end user organisations, many of which will be involved in the project’s activities through a stakeholder group and stakeholder activities.
In order to achieve the aim described above, the project was designed around the users of climate information and their needs. A Stakeholder Board was formed at the very beginning of the project. This comprises public and private organisations, who operate in Europe in a variety of sectors, including health, energy, water, forestry, agriculture and tourism. A series of in-depth interviews, dedicated workshops and online surveys were used to determine users’ needs and consequentially inform the development of the research agenda. One research theme in the project was specifically designed to reduce the known gap that exists between the ways in which climate information is provided and the formats that users require this information to be presented. Making the information useful in this way is considered very important. Three main strategies were formulated which would assist with making model outputs more usable by stakeholders. These were:
- .Address the resolution gap by developing suitable downscaling mechanisms (statistical and/or dynamical). Implicit in this is addressing the bias with respect to observations, which is an inevitable component of any model simulations;
- Use climate indices as simple impact models to increase the usability of seasonal predictions and
- Develop, and couple with climate prediction systems, complex process-based models able to address specific user-relevant impacts.
In the upcoming stages of the project, one of the main activities will be the development, the evaluation and the dissemination of the results associated with the climate service prototypes that will be developed within the project. Such prototypes are the antithesis of simply a climate information portal. Central to the idea of the project is the ethos that to be useful, the climate data needs to be adapted to the specific context of the users. Through a challenging but useful exercise the project has selected five prototypes of climate services that will be developed within EUPORIAS. These will address the needs of specific decision-makers in specific sectors. When developing these prototypes, the focus will be on the transformation of the data, its post-processing and its graphical representation. This will ensure that the data will be presented in the most useful way for each sector.
Work will also be done on the interaction of the climate prediction information, with the decision support systems, and assessing the marketability of the climate services developed. Also relevant to the work described above, the project identified uncertainty as a cross-cutting research theme. This was required in order to address the difficulties that arise when combining different sources of uncertainty and presenting them to decision makers. So far EUPORIAS has made progress in the understanding of the model skill; understanding of the impact model uncertainty; and in the assessment of the appetite that users have for different representations of the level of certainty associated with the impact predictions.
Expected fianl results and impacts
It may be useful here to distinguish the results and their potential impacts in terms of, acquisition of new knowledge, and technical improvements and achievements of direct relevance for the project.
1. Acquisition of new knowledge
The project provided one of the first comprehensive descriptions of the users’ landscape for climate predictions in Europe. Whilst other analyses have been conducted at both national and international level to our knowledge nothing equates to the depth or breadth of the analysis conducted within EUPORIAS. The two general workshops conducted with users and providers of climate information, the 75 in-depth interviews and the results of the online surveys, provided the scientific community in Europe with fresh data to better understand who the users of climate predictions are and what kind of information they would like to obtain. From a model perspective improvements have been made in our understanding of the current level of bias, skill and drift that seasonal predictions in Europe has. In particular an analysis of the skill of seasonal prediction for climate indices has been conducted. An analysis of the spatial and temporal structure of the model drift has been conducted for different variables. The results suggest that model
drift depends strongly on lead-time, start day, location and variable. Whilst bias-correction (drift removal) should still be considered as the natural way to analyse model output, the results suggest that in some very specific circumstances the analysis of direct model output can potentially be conducted.
2. Technical improvements and achievements
From a technical point of view there have been a number of major improvements achieved by EUPORIAS. On one hand the development of the ECOMS User-Data Gateway and the R-functions than sit on it, defined a new way of interaction between users and climate prediction data. Technical achievements of this nature can often be mis-evaluated, but they probably represent one of the most important long-term legacies of the project so far.
Another technical improvement has been the definition of a glossary of terms dynamically allocated on the web-page. Rather than develop yet another technical glossary; something that would duplicate the effort of other organisations such IPCC, NIPP, CORDEX, VALUE, etc., the emphasis here has been on the minimum set of terms that needs to be defined in order to minimise the misunderstanding between users and providers of climate information. The glossary, which is directly linked to other existing glossaries and which has been implemented as a Drupal Module, is updated regularly to reflect the feedback provided by the user community.
Significant improvements have also been obtained in the field of impact model and regional model initialisation for climate predictions. One of the challenges the project faced was related to the need to account for model drift. Defining a common protocol for impact model simulations, as well as a standard set of data to be used as a benchmark, will give EUPORIAS partners and others, a well-defined basis for inter-comparison of impact predictions. This can be used beyond the lifetime of the project. For dynamical downscaling; significant effort has been put into the definition of a suitable procedure for downscaling seasonal predictions.