PAGODA will focus on the global dimensions of changes in the water cycle in the atmosphere, land, and oceans. The overarching aim is to increase confidence in projections of the changing water cycle on global-to-regional scales through a process-based detection, attribution and prediction. The scientific scope prioritises themes 2,1,3,4 in the AO, adopting a focus on climate processes to extend our understanding of the causes of water source/sink uncertainty at the regional scale, which is where GCMs show huge variations concerning projected changes in precipitation, evaporation, and other water related variables. This model uncertainty is closely linked to shifts in large-scale circulation patterns and surface feedback processes, which differ between models. Furthermore, even where models agree with each other (for example, the suggested trend towards wetter winters and drier summers in Europe, connected to storm tracks and land surface processes), consistency with the real world cannot be taken for granted. The importance of quantitative comparisons between models and observations cannot be overstated: there is opportunity and urgent need for research to understand the processes that are driving changes in the water cycle, on spatial scales that range from global to microscopic, and to establish whether apparent discrepancies are attributable to observational uncertainties, to errors in the specification of forcings, or to model limitations. PAGODA will achieve its scientific objectives by confronting models with observations and reconciling observations, which possess inherent uncertainty and heterogeneity, with robust chains of physical mechanisms - employing model analysis and experiments in an integral way. Detection and attribution is applied throughout, in an iterative fashion, to merge the understanding from observations and models consistently, in order to isolate processes and identify causality. PAGODA is designed to focus specifically on the processes that govern global-to-regional scale changes in the water cycle, particularly on decadal timescales (the timescale of anthropogenic climate change). It addresses processes in the atmosphere, land and oceans, and brings together experts in climate observations, climate models, and detection and attribution. It seeks to exploit important new opportunities for research progress, including new observational data sets (e.g. ocean salinity reanalysis, TRMM and SSMIS satellite products, long precipitation records), new models (HadGEM3 & new capabilities for high resolution simulations), and the new CMIP5 model inter-comparison and to develop new methodologies for process-based detection, attribution and prediction.

Forecasting the weather from days to two weeks in advance has typically focused on the troposphere, the layer of the atmosphere closest to the ground. A typical weather forecast first attempts to estimate what the atmosphere is like now, and then extrapolates forward in time, using a complex model of the atmosphere based on the basic physical laws of motion. Over the last 15 years, evidence has been growing that different parts of the atmosphere and Earth system can also be exploited to improve weather forecasts. One of these regions is the stratosphere, the layer directly above the troposphere. Because, temperatures increase with height in the stratosphere, winds and weather systems are quite different, and a distinct community of scientific researchers who study the stratosphere exists around the world. Through the work of this community, many weather forecasting centres have been encouraged to look to the stratosphere to improve their weather forecasts and have been modifying their weather forecasting models accordingly. What has been missing, however, is a concerted effort to understand how best to make use of the stratosphere to improve weather forecasts and to determine how much weather forecasts might benefit. This proposal will fund a new international scientific network which will bring scientists from around the world together to study the stratosphere and how it might be used to improve weather forecasts. The network is made up of scientists from universities and weather forecasting centres around the world and is supported by two other international scientific research bodies. The network will allow scientists to come together to discuss current research in this area and to plan and carry out a new experiment which will compare the stratosphere and its impact on weather forecasts in their weather forecasting models. At the end of the research project, the network members will work together to produce a report which will provide guidance to all weather forecasting centres on the use of the stratosphere for weather forecasting.

The Arctic is undergoing rapid climatic change, with dramatic consequences for the 'Frozen World' (the 'cryosphere'), including reductions in the depth, extent and duration of sea ice, and seasonal snow cover on land, retreat of ice sheets/glaciers, and melting of permafrost ("ground that remains at or below 0 degrees C for at least two consecutive years"). This is important not only for local and regional ecosystems and human communities, but also for the functioning of the entire earth system. Evidence is growing that organic matter frozen in permafrost soils (often for many millennia) is now thawing, making it available for decomposition by soil organisms, with the release of carbon dioxide (CO2) and methane (CH4), both greenhouse gases (GHGs), as by-products. A major concern now is that, because permafrost soils contain 1672 petagrams (1 Pg = 1 billion tonnes) of organic carbon (C), which is about 50% of the total global below-ground pool of organic C, and permafrost underlies ~ 25% (23 million km2) of the N hemisphere land surface, a melting-induced release of GHGs to the atmosphere from permafrost soils could result in a major acceleration of global warming. This is called a 'positive biogeochemical feedback' on global change; in other words, an unintentional side-effect in the global C cycle and climate system. Unfortunately, the interacting biological, chemical and physical controls on CO2 and CH4 emissions from permafrost (and melting permafrost) environments to the atmosphere are the subject of much speculation because the scientific community does not know enough about the interactions between C and water cycling in permafrost systems. Warmer and drier soils may release more CO2, while warmer/wetter soils might release more CH4. Permafrost thawing also causes changes in the way water flows though the landscape (because frozen ground if often impermeable to water), and some areas may become drier, while others wetter. How the relative proportions of CO2 and CH4 emissions change, and their absolute amount, is critical for the overall 'global warming potential' (GWP) because these two gases have different potency as GHGs. Release of C from soils into freshwaters also needs to be taken into account because down-stream 'de-gassing' and decomposition of organic materials also influences releases of CO2 and CH4 from freshwater, or delivery of C to lakes/oceans. All-in-all, predicting the GWP of permafrost regions is scientifically challenging, and the interactions between the water (hydrological) and C cycles are poorly known. In this project we recognise the key role that hydrological processes play in landscape-scale C fluxes in arctic and boreal regions. In permafrost catchments in NW Canada (including areas where permafrost is known to be thawing) we will measure the capture of C from the atmosphere (through photosynthesis), its distribution in plants and soils, and the biological, physical and chemical controls of C transport and delivery from soils to freshwaters, and ultimately to the atmosphere as CO2 and CH4. In essence we wish to 'close the C cycle'. Field-based measurements of key processes in the water and C cycles, including geochemical tracer and state-of-the-art C, hydrogen and oxygen isotope approaches, will be linked by computer modelling. The project team, together with partners in Canada, the US and UK, is in a unique position to link the water and C cycles in permafrost environments, and we will deliver essential scientific knowledge on the potential consequences of climate warming, and permafrost thawing, for GHG emissions from northern high latitudes. Both for local peoples directly dependent on arctic tundra/boreal forest ecosystems for their livelihoods and cultural identity, and for the global community who must respond to, and anticipate, potential consequences of climate and environmental change, this project will represent a significant step forward in understanding/predictive capacity.

By modifying the amount of solar radiation absorbed at the land surface, bright snow and dark forests have strong influences on weather and climate; either a decrease in snow cover or an increase in forest cover, which shades underlying snow, increases the absorption of radiation and warms the overlying air. Computer models for weather forecasting and climate prediction thus have to take these effects into account by calculating the changing mass of snow on the ground and interactions of radiation with forest canopies. Such models generally have coarse resolutions ranging from kilometres to hundreds of kilometres. Forest cover cannot be expected to be continuous over such large distances; instead, northern landscapes are mosaics of evergreen and deciduous forests, clearings, bogs and lakes. Snow can be removed from open areas by wind, shaded by surrounding vegetation or sublimated from forest canopies without ever reaching the ground, and these processes which influence patterns of snow cover depend on the size of the openings, the structure of the vegetation and weather conditions. Snow itself influences patterns of vegetation cover by supplying water, insulating plants and soil from cold winter temperatures and storing nutrients. The aim of this project is to develop better methods for representing interactions between snow, vegetation and the atmosphere in models that, for practical applications, cannot resolve important scales in the patterns of these interactions. We will gather information on distributions of snow, vegetation and radiation during two field experiments at sites in the arctic: one in Sweden and the other in Finland. These sites have been chosen because they have long records of weather and snow conditions, easy access, good maps of vegetation cover from satellites and aircraft and landscapes ranging from sparse deciduous forests to dense coniferous forests that are typical of much larger areas. Using 28 radiometers, and moving them several times during the course of each experiment, will allow us to measure the highly variable patterns of radiation at the snow surface in forests. Information from the field experiments will be used in developing and testing a range of models. To reach the scales of interest, we will begin with a model that explicitly resolves individual trees and work up through models with progressively coarser resolutions, testing the models at each stage against each other and in comparison with observations. The ultimate objective is a model that will be better able to make use of landscape information in predicting the absorption of radiation at the surface and the accumulation and melt of snow. We will work in close consultation with project partners at climate modelling and forecasting centres to ensure that our activities are directed towards outcomes that will meet their requirements.

GOTHAM represents an ambitious research programme to gain robust, relevant and transferable knowledge of past and present day patterns and trends of regional climate extremes and variability of vulnerable areas identified by the IPCC, including the tropics and high-latitudes. It will achieve this by identifying the influence of remote drivers, or teleconnections, on regional climate variability, and assessing their relative impact. It will also assess the potential for improved season-decadal prediction using a combination of contemporary climate models, citizen-science computing and advanced statistical analysis tools. GOTHAM has the direct backing of many international weather and climate research centres, and will lead to the improved development of seasonal-decadal forecasts at the regional level. The improved knowledge and understanding of dynamical factors that influence regional weather and climate in the tropics/sub-tropics, and polar regions, will directly feed through to weather and climate forecast services to assist in their decisions on which priority areas of their model development to target in order to improve forecast skills. For example, GOTHAM will advise whether a model is missing or misrepresenting important global teleconnections that significantly influence regional climate in identified vulnerable regions. These impacts will be achieved through regular meetings with GOTHAM investigator groups and their extended collaborative networks, and extensive involvement in wider science and science-policy programmes with co-aligned strategies, such as the core projects within the WCRP. Improved seasonal to decadal scale forecasts will improve predictions of extreme events and natural hazard risks such as flooding that can have devastating impact on society. There is real potential for project results feeding through to impacts-related research, such as those involved in hydrological and flood forecast modeling, and these will be explored in liaison with identified partners in Asia and Europe.