Skip to main content
Menu

Models used by the CPDN project

Scientists will be able to study events such as tropical storm Karl, which developed in the Atlantic in September 2016, using the OpenIFShome project. (Image: NASA Visible Earth, LANCE/EOSDIS Rapid Response team)

Models

Four climate models are currently available in climateprediction.net (CPDN for short), these are:

  • OpenIFS@Home – A global atmospheric model from the European Centre for Medium Range Weather Forecasts (ECMWF).
  • Weather@Home – A Met Office Hadley Centre global atmospheric model (HadAM3P) with a high resolution embedded regional model (HadRM3P) at either 50km or 25km.
  • HadAM4 – A high resolution Met Office Hadley Centre global atmospheric model. This runs at two resolutions on CPDN which corresponds to ~90km and ~60km.
  • HadCM3 – A Met Office Hadley Centre coupled model of the ocean and atmosphere.
  • HadSM4 - A Met Office Hadley Centre slab model.

 

OpenIFS model

OpenIFS@Home brings together two powerful tools: OpenIFS 43r3, an easy-to-use, supported version of European Centre for Medium-Range Weather Forecasts' (ECMWF) Integrated Forecasting System (IFS) widely used in research and education; and CPDN at the University of Oxford, a highly successful volunteer computing project that has been running since 2003.

See the results from the initial study where thousands of volunteer personal computers simulated slightly different realisations of the tropical storm Karl to demonstrate the performance of the large ensemble forecast.

Weather forecasting requires powerful computer systems and state-of-the-art computer models. The ECMWF Integrated Forecast System (IFS) is one of the world’s leading weather forecasting models. A version of their model, OpenIFS, is available to universities and research institutes for teaching and research.  As well as producing a 10-day weather forecast from the best estimate of the current weather, a large number of slightly different forecast scenarios, known as an ensemble, are created to allow a measure of certainty on the forecast to be provided.

Through OpenIFS@Home it is now possible to run a slightly different weather forecast on many hundreds or thousands of volunteer computers, making it possible to ask questions such as how predictable certain events are, particularly damaging extreme events such as intense rain or wind. The OpenIFS@Home facility offers researchers a new tool to study weather forecasts and related questions

 

Weather@Home models

Weather@Home is a group of regional climate modelling experiments within CPDN.

Thanks to your support of CPDN we are able to design experiments that answer questions we otherwise could not answer without large climate model ensembles. However, most extreme weather events take place on a much smaller scale that the global models can’t show. For this we need the weather@home project!

Weather@Home allows us to run regional climate models to answer the question: how does climate change affect our weather.

Weather@Home helps us, and scientists all over the world, to answer this question. It is a family of regional climate models for a growing number of regions around the world. With Weather@Home we can investigate how the odds of extreme weather events change due to man-made climate change and natural climate variability.

With Weather@Home you can run the model simulating the weather in your native part of the world. Weather@Home also makes CPDN a truly international project, as participants and the scientists who analyse the data come from all over the world. The fact that local scientists are collecting and analysing the data is important as it means that any results from the project are underpinned.

 

The Motivation for Weather@Home

You’ve heard of climate change, but what does that actually mean for the weather in the region where you live? Could it be that you are going to see an increase in the number of damaging weather events? Or could the weather actually be getting nicer? The Weather@Home experiment hopes to answer these questions, with your help.

For experiments which look at climate globally, Global Climate Models (GCMs) are used, as they have a coarse resolution and thus give an overall picture of climate. In this case, however, a Regional Climate Model (RCM) needs to be used so as to provide specific information about certain regions. RCMs increase the resolution of areas of interest, allowing scientists to make predictions regarding the local impact of climate change. Such models are also supplied with climate information such as temperature, winds and humidity, around the edges, so that the influence of the weather in other parts of the world is still taken under consideration. Essentially, this is achieved through the embedding of the regional model within a ‘driving’ global model.

Initially, three target regions are now available for download: the Western US, Southern Africa and Europe. These were chosen because the majority of CPDN participants (to date) live in Europe and the US, and because Southern Africa is a region thought to be particularly vulnerable to climate change.

 

The Science of Weather@Home

The overall experiment design is in five parts:

First, a large number of different versions of the global and regional models will be used to simulate the period from 1960 to 2010 using observed changes in sea surface temperatures, sea ice, atmospheric greenhouse gases and aerosols. The simulated climates and patterns of change in weather events from the models will then be compared with observations over the same period to select a range of realistic model versions and document their behaviour. If, for example, we find that a particular version of the model tends to over-do the number of storms, we can take account of this when using this model to forecast future changes in storminess.

The second experiment is to produce a forecast of changes in weather events by the 2020s and 2030s. Using output from many different models with evolving oceans to provide the forecast sea surface temperatures up to this time, the regional model will tell us about the potential changes to patterns of weather events through the next three decades in unprecedented detail. Features such as changes in the likelihood of drought, flood and extreme heat or cold are likely to be of particular interest.

The third experiment returns to changes seen over the last 50 years, and attempts to quantify to what degree these changes can be attributed to the effects of human interference in the climate system. The driving conditions fed into the models are modified to reflect what they would have been like if we had not produced the greenhouse gas and aerosol emissions that we have over the past century. The difference between these simulations and the initial `baseline’ runs will provide the basis for assessing the human contribution to recent weather trends.

The fourth experiment returns to forecast mode but runs beyond the timeline of the second experiment, providing detailed information about changes in weather features in a world 2, 3 and 4 degrees warmer than global temperatures today, representing a range of climates that might be encountered towards the end of this century or beyond. This experiment will provide some of the most detailed information to date on regional weather in such possible future worlds, which is essential to assess the range of potential impacts of climate change.

Finally, the fifth experiment looks back into the past – looking at snapshots of the weather at intervals over the past 10,000 years, a period of Earth’s history called the ‘Holocene.’ This is the first time large numbers of regional models have been applied to such ‘paleoclimate’ (past climates) simulation: an unprecedented opportunity to explore the evolution of the weather over recent Earth history.

Please note: The project team aims to run as many of these experiments as possible, but not all experiments will necessarily be performed for all regions.

 

The W@H Regions

region_plot (1)
This map shows all currently running regions under Weather@Home.

At the start of the project in 2010 initially, three target regions where chosen and described here in more detail: the Western US, Southern Africa and Europe. These were chosen because the majority of CPDN participants (to date) live in Europe and the US, and because Southern Africa is a region thought to be particularly vulnerable to climate change. All regional experiments will ultimately be managed from our partners within the region, but the development of the models (to date) takes place in Oxford.

The ultimate aim is to have model simulations for all CORDEX regions in the world, the Coordinated Regional Climate Downscaling Experiment, is a World Climate Research Programme (WCRP) framework to evaluate regional climate model performance through a set of experiments aiming at producing regional climate projections.

Our regional models for Weather@Home were initially deployed as separate applications e.g: hadam3p_eu, hadam3p_anz, hadam3p_afr, hadam3p_pnw. More recently we have developed a single application (Weather@Home2 / WAH2) that can run different regions on demand. The region is specified in the workunit and will appear as part of the workunit name e.g. wah2_eu25, wah2_sas50.

Results from our regional modelling experiments are freely available to anyone investigating the impacts of weather and climate changes across the available regions.

Europe (hadam3p_eu: EU25 and EU50)

The Weather@Home application hadam3p_eu is a 50km region matching the EURO-CORDEX domain. This is our own home region and thus also our experimental region with currently the largest data sets available.

The WAH2 EU25 region was developed for the MARIUS project and differs slightly from the EURO-CORDEX domain.

The WAH2 EU50 region differs slightly again and is rotated slightly to avoid including Greenland at the edge of the domain.

Western US (PNW25 and WUS25)

We have two Western US regions:

The first is labelled PNW for Pacific North West, is an unusual region as it is smaller than our other regions and has a resolution of 25 km. The model region encompasses the Eastern Pacific Ocean and the Western US, extending inland as far as the Rocky Mountains. North to south it stretches from Canada down to the Gulf of California, and includes the Coast Range of mountains.

The second is labelled WUS and is a slightly larger region than PNW, stretching slightly further north and east.

Central America (CAM25, CAM50)

We have two central America regions, a 50km region and a slightly smaller and rotated 25km region.

This Central America experiments are conducted in collaboration with our partner Ruth Cerezo-Mota at Universidad Nacional Autónoma de México. It will be used for TITAN and other future projects.

South America (SAM50)

We have a 50km South America region which covers the CORDEX South America domain.

This South America region will be used for future projects.

Australia & New Zealand (hadam3p_anz)

This region includes the whole of Australia and New Zealand at 50 km resolution.

Results of this experiment go to our partner David Karoly at the University of Melbourne and analysis will also be undertaken by Sue Rosier at NIWA (The National Institute of Water and Atmospheric Research), Wellington, New Zealand.

South Asia (SAS50)

This is the CORDEX region South Asia which stretches from the Horn of Africa and the Arabian Peninsula in the West to Vietnam and Cambodia in the East. In North-South direction it includes the Himalayas as well as Lake Victoria, India and the surrounding Ocean is in the centre of the region.

East Asia (EAS50)

We have developed an East Asia region to be used for future projects. This differs from the CORDEX East Asia region as it goes further south and includes all of Indonesia and Papua New Guinea.

Africa (hadam3p_afr: NAWA25)

The hadam3p_afr region is a 50km region over most of Africa apart from Southern Africa (excluding countries from Namibia, Botswana and Zimbabwe southward). This region was developed for the ACE-Africa project.

We additionally have a smaller 25km region over North Africa and West Asia for use in future projects.