Material for NERC e-Science Glossy publication

GCEP needs to produce 300-400 words of material for a "glossy" brochure promoting the achievements of the NERC e-Science programme to the public and the government. This will form a single page in the brochure. We need an introduction, some good "stories" from GCEP and some suggested figures (please upload).

The headings are semi-fixed but the text is up to us. Please remember that this is for a general audience and so should be written as plainly as possible.

Title: Can anyone suggest something snappy and/or amusing?

Suggestion 1: Studying the feasibility of 10-year forecasts

Suggestion 2: Advancing 10-year climate forecasts

Suggestion 3: It's too hard to predict tomorrow's rainfall, so let's try next year's mean temperature instead!

Introduction

(I dreamt this up - please edit if inaccurate! Hopefully this will give an idea of the general tone and technical level of writing.)

Policy-makers and many private enterprises can gain immense benefits from knowing how the earth's climate will change over the next 5-10 years. It can affect many facets of society from policy decisions on investment in flood defences to the price of home insurance. Climate models can help to make the necessary predictions but they need to start from an accurate knowledge of the Earth’s present state. This is not a trivial problem: despite the relentless growth of the quality and quantity of observations from satellites and ground-based instruments, not all necessary observations are available. There is also a great need to improve the way this large amount of data is incorporated into the model.

It is the slowly-varying properties of the Earth system that are most likely to be important for predicting the decadal climate. These include the oceans, sea ice, snow and soil moisture. The Grid for Coupled Ensemble Prediction (GCEP) project is employing Grid technology to harness the computing power that is required to run large numbers (ensembles) of climate models. The aim is to investigate the effects of the slowly-varying properties on decadal climate prediction and also to develop new ways to incorporate observations into the model.

Stories

These topics are suggested from the GCEP poster. Better titles are needed! Please fill in the details and feel free to add any other stories. Editing can come later. Only 50-100 words on each.

Sea ice sensitivity

Sea ice is an important feedback and affects the coupling between the atmosphere and the ocean. But how sensitive is the climate to the initial state of the sea ice? Model runs show that sea ice recovers rapidly, within a few years, to an extreme perturbation - the removal of the entire sea ice cover. This is because the underlying climate is still "conditioned" to expecting sea ice. More extreme effects - warming the oceanic surface waters - lead to longer-lasting anomalies. This suggests a more passive role for sea ice than has previously been thought.

Perturbing Nordic Sea conditions

The temperature in Western Europe is highly dependent on the heat that the ocean brings from the south in the North Atlantic Current. The strength of this current is in turn dependent the conditions in the Nordic Sea where the water sinks and returns to the south. By perturbing the amount of salt in the Nordic Seas in climate models, GCEP scientists have been able to see how signals propagate down the coast and ultimately control the strength of the North Atlantic Current. If more knowledge is gained about these processes in the Atlantic it could lead to more accurate decadal forecasts of Western European temperatures.

Taking account of climate change in predictions

It is not only the initial state of the climate which can affect a 10-year forecast, but also the various external forcings it experiences during those ten years. These forcings include radiation from greenhouse gases and solar variations, but also volcanoes and intermittent climate phenomena such as El Nino. While only the forcing from greenhouse gases can be reliably predicted, it is an important driver of the climate and it is important that a 10-year forecast includes this. To enable scientists to study the effect of climate change on the predictablility of various climate properties, GCEP has set up a system which provides scientists with many different climate scenarios which are all likely in our present climate. These scenarios can then be used in climate models to determine the predictability of the climate.

ENSO/snow cover.

Anomalies in snow cover extent and snow depth are known to affect the climate, both locally and over long distances. But is this effect overwhelmed by climate drivers such as the El Nino-Southern Oscillation (ENSO)? By running the model many times in different configurations, the effect of ENSO on snow cover and vice versa can be explored. Figure X shows that deeper snow in the Himalayas occurs in the years when ENSO is positive, and similarly the negative phase brings less snow. But can snow conditions affect the weather in this region over and above the effect of ENSO, and will we get forecasts wrong if we don't take this into account?

Key points

A series of bullet-points is required to highlight key results or innovations.

• Running and managing many large climate simulations over multiple computer clusters
• Developing new ways to incorporate observations into climate models
• Improving knowledge on the impact of snow cover and sea ice on climate
• Studying the impact of changes in the Nordic Seas on the Atlantic circulation
• Providing a framework for studying decadal prediction in the current changing climate

-- JonBlower - 19 Sep 2007