Managing Risk with Climate Prediction Products and Services

Demand for climate predictions on timescales of weeks to decades is accelerating as decision-makers in both private and public sectors increasingly recognize their relevance in building climate resilience and in climate change adaptation. Tailored climate services are sought by various types of users for longer-term decisions and planning, for early warning of potential hazards, and for climate variability and change adaptation and mitigation. Collaboration within the Global Framework for Climate Services (GFCS) is ensuring that providers of climate prediction products interact more effectively with users to meet this unparalleled demand for tailored climate services. 

Much progress is being made in developing climate prediction products. Model output data, climate predictions, complemented by inputs from stakeholders is yielding tailored climate services over many time scales. There are many examples of how users are benefiting from these products and services. 

Climate prediction 

National and international investments in climate observations, research and modelling have greatly improved climate predictions and projections over the past decades. They have also helped to advance overall scientific understanding of climate variability and change. This progress has provided a robust scientific foundation for climate services. 
Climate models base predictions on present conditions and assumptions concerning physical processes that will affect change. The predictions, outputs of the climate models, are probabilistic statements about future climate conditions on different time and spatial scales. Over global, regional or local spatial scales predictions cover time scales such as (see Manual on the global data-processing and forecasting system: Volume I - Global aspects):

Extended-range forecast – A forecast beyond 10 days and up to 30 days describing weather parameters, usually averaged and expressed as a departure from climate values for that period. Examples include 10-day and 2-week forecasts, extreme climate event probability predictions, 3 category (above normal, normal, below normal) probabilistic forecasts of rainfall and temperature, etc. 

Long-range forecast – A forecast from 30 days up to 1 year that generally includes monthly outlooks providing a description of averaged weather parameters expressed as a departure (deviation, variation, anomaly) from climate values for that month (not necessarily the coming month), 3-month or 90-day outlooks that provide a description of averaged weather parameters expressed as a departure from climate values for that period (which is not necessarily the coming 90-day period) and seasonal outlooks that provide a description of averaged weather parameters expressed as a departure from climate values for that season. Examples include corresponding climate forecasts with variables such as precipitation, temperature (2 metres above ground), sea level pressure, 500-hPa geopotential height; forecasts of Pacific and Atlantic tropical storms; and seasonal forecast of large-scale climate variability phenomena such as the El Niño-Southern Oscillation (ENSO) or the Madden-Julian Oscillation (MJO). 

Interannual prediction – A prediction from one to several years that describes the large-scale climatic state. This forecast is initialized with indicators of the current climate to capture the evolution of modes of internal climate variability, such as ENSO. Examples include prediction of the climatic trend of variables such as precipitation, temperature, sea level pressure and 500- hPa geopotential height.

Decadal prediction – A prediction of fluctuations in the climate system over the next 10 years, taking into account natural variability as well as human influences. This is achieved by initializing climate models with observations of the current climate state, in addition to specifying changes in radiative forcing due to greenhouse gases, aerosols and solar variability.

The distinctions between annual and decadal predictions are not clear and many climate centres use these terms interchangeably. 

From prediction outputs to services 

Climate model outputs – climate predictions – form the basis of tailored climate services; however, these can only be built through strong partnerships between providers, including national meteorological and hydrological services (NMHSs), and stakeholders. By being part of the process, stakeholders obtain climate information that they can understand, interpret and apply, to reduce the impacts of climate-related disasters, improve food security and health outcomes, enhance water resource management, and much more.

WMO Members provide a number of climate prediction services at global, regional and national levels to a variety of users, including individual decision-makers and policy-makers as well as organizations and humanitarian agencies. The WMO Global Producing Centres (GPCs) for Long-Range Forecasts take a lead in, and set the standards for, predicting climate and weather on global, regional and local scales. Their forecasts are downscaled by WMO Regional Climate Centres (RCCs) and by local forecasting centres, usually within the NMHS. The GPCs assist WMO Members to deliver better climate services and products, including regional long-range forecasts, and strengthen their capacity to meet national climate information needs. From global, regional and local climate predictions, GPCs, RCCs, NMHSs and other agencies generate tailored climate services for both the public and private sectors. Such services are offered by many NMHSs on the different time scales from seasonal predictions to the long-term projections for climate change impacts.
There are many examples of climate prediction related services from GPCs, we can only offer a few case studies.

Climate Champion Program – The Managing Climate Variability (MCV) Climate Champion program of the Australian Bureau of Meteorology (BOM) aims to help farmers manage climate risk by providing them with the best climate tools, products, practices and seasonal outlooks, and by helping them to gain an understanding of how they might use that information in their farm business. It also aims to give climate researchers a chance to interact with farmers and get feedback about what they need from research. Twenty farmers, representing most of the major agricultural commodities, are taking part in the program through which they can:

  • talk with researchers about the tools and informa- tion they need to help them manage climate risks;
  • test early research products and practices, and possibly influence the research,
  • influence how research findings are communicated to farmers, and
  • help farmers in their region and commodity area to learn how to deal with climate variability and change.

The story of Gillian Taylor, from Bibbaringa Farm in New South Wales, offers a good example. Gillian’s 990-hectar farm is devoted to beef cattle. Gillain explains how Bib-baringa Farm uses the information from BOM. “Because we’ve set up our farm to be easy to manage, we don’t hand feed our cattle. We’re firm believers in destocking before it’s too late. I believe ‘too late’ is when you’re six months into a drought, you’ve run all your cattle through your paddocks and you have to either buy feed or sell your stock in a market where everyone else is in the same situation, so the price is low. Taking too many cattle out of our mob slows down operations, but we at least know we have enough feed in our paddocks for the cattle we have left and we are getting a good price for the cattle we sell.” 

“It’s already May and we’ve only had 111 millimetres [of precipitation] this year [2013], which is well below average [250 mm]. We’ve been looking at the forecasts on Elders and the Bureau of Meteorology, and been speaking to a number of friends in the area who access forecasting from private forecasters and everyone is saying that there’s going to be no significant rainfall coming in the next couple of months. We like to check everything against the seasonal forecasts on the Bureau of Meteorology website. With all this in mind, we decided to sell 100 cows and 50 weaners [weaned calves] last week to leave us with 350 cows and calves. If we don’t receive any significant rain in the next 6 weeks we’ll take more cattle out.”11

Climate service for the Three Gorges area – Over the last 20 years, the Yangtze River Basin has experienced increasingly disastrous flood and drought incidences that have challenged the safe operation and scientific control of reservoir waters. The benefits – flood control, drought resistance, power generation, water storage and shipping, among others – of the Three Gorges Reservoir Project, completed in 2009, are under threat. Tailored hydrometeorological service are mitigating the threat. 

Precipitation anomaly predictions by percentage for June- August 2013 (top) and the observed results (bottom) Precipitation anomaly predictions by percentage for June- August 2013 (top) and the observed results (bottom) - Precipitation anomaly predictions by percentage for June- August 2013 (top) and the observed results (bottom) In early April 2013, a climate trends report was released for the flood season that covered June-August. The predictions accurately distinguished the abnormally low rainfall as a main trend in most part of the Yangtze River Basin, while warning the possible appearance of two precipitation centres in the Min-Tuo River Basin and the Jialing River Basin. Observations showed that the heavy rainfall triggered severe flooding in July. These prediction services for key period are critical to users. (Source Hubei Provincial Meteorological Bureau)

For example, the Yangtze River Basin Meteorological Centre initiated a range of consultation meetings to predict the water storage volume for the Three Gorges Reservoir, when it reported a reduced total inflow in August 2013 compared with the same period the previous year. The meetings led to the production of detailed precipitation predictions for September upstream of the Yangtze River and precipitation trends for the September-November storage period across the Yangtze River Basin. Climatologists suggested that the Three Gorges Cascade Dispatching Centre start retaining water ahead of the usual schedule, so as to reduce the post-storage period pressure on the reservoir, secure a smooth transition of water levels, and use the water resources at the end of the flood season. The water retaining time was advanced to the end of August, two weeks earlier than in the regular schedule. During the storage period, the Yangtze River Basin Meteorological Centre kept a close watch on the precipitation trends across the upper streams of the Yangtze River, and provided timely precipitation forecasts for the extension period. Observed precipitation was largely consistent with predictions that 8-11, 13-16 and 22-24 September would experience precipitation. Thus, the predictive products provided a fairly accurate guideline for flood control scheduling and storage.

Hazards outlooks for areas exposed to reduced crop production or livestock vulnerability8 – Drought is one of the greatest challenges in the developing world. In order to allay its impacts, the international community has develop Famine Early Warning Systems Network (FEWSNET) to bring safe food and water to populations in need. Over the past decades, it has focused attention on advance risk planning in agriculture and water, which requires frequent updates of weather and climate outlooks. In this context, the Climate Prediction Centre (CPC) of the U.S. National Oceanic and Atmospheric Administration (NOAA) provides monthly climate out- looks highlighting areas where crop production might be reduced or livestock may be vulnerable due to flooding or drought for Africa, Central America, the Caribbean and Central Asia, valid for one week.

Preliminary hazard outlook bulletins are prepared every week and distributed to partners within FEWSNET, including field representatives with expert knowledge of on the ground conditions with whom a teleconference is organized. These discussions permit participants to finalize the hazards outlooks that are then dissemi- nated on the Internet and to an e-mail distribution list. United States Agency for International Development (USAID) uses this information for decision-making for humanitarian response planning for threats to food security. Tailored food security climate outlooks permit the development of contingency plans for safe and timely delivery of food and to populations afflicted by drought or flood related disasters. 

The Hydrological Outlook in the UK – The Met Office of the United Kingdom of Great Britain and Northern Ireland (UK) is currently developing added value operational climate products about likely future hydrological conditions on a monthly timescale. Based on the current status of river flows and groundwater levels, a number of techniques are used to predict conditions going forward over coming months. The outputs include a map with highlighted areas and a detailed text interpretation of the forecasts as well as raw precipitation, temperature, groundwater and river flow forecasts.

Stakeholders – government and private water company representatives – participate with meteorological and hydrological experts in the development of the initial outlooks. As a result, the Met Office understands their decision-making processes and information requirements. A wide range of users are now accessing these outlooks, which are readily available via email (register on the Hydrological Outlook UK website).

Forecasting dengue fever risk for the FIFA World Cup9 – The dengue fever forecast service was developed as part of the EURO-BRazilian Initiative for improving South American seasonal forecasts (EUROBRISA) during the 2014 World Cup. The EUROBRISA multi-model system seasonal precipitation forecasts for the period from March to May and an empirical model seasonal temperature forecasts were input into a dengue risk model to produce probabilistic forecasts for the disease in Brazil for June, month of the World Cup. 

Closer collaboration to widen service 

Climate prediction information services are important and relevant to a wide-range of users. The overview of currently available services and related benefits in this article emphasizes the importance for close cooperation between providers and stakeholders in their development. By collaborating through the GFCS, both the private and public sector are able to set up successful new projects to support the implementation of climate services, especially in least developed countries or regions.



Climate data: Historical and real-time climate observations along with direct model outputs covering historical and future periods. Information about how these observations and model outputs were generated (“metadata”) should accompany all climate data.

Climate product: A derived synthesis of climate data. A product combines climate data with climate knowledge to add value. Climate information: Climate data, climate products and/or climate knowledge.

Climate service: Providing climate information in a way that assists decision making by individuals and organizations. A service requires appropriate engagement along with an effective access mechanism and must respond to user needs15


WMO Global Producing Centres of Long-Range Forecasts (GPCLRFs)


1 Met Office, United Kingdom
2 Climate Prediction Centre / National Oceanic and Atmospheric Administration
3 Beijing Climate Centre, China Meteorological Administration
4 Centre for Weather Forecasts and Climate Studies / National Institute for Space Research, Brazil
5 The Australian Bureau of Meteorology
6 Climate Prediction Division, Earth Environment and Marina Department, Japan Meteorological Agency
7 Filipe Lucio, C. Tamara Avellán and Zhiqiang Gong, Global Framework for Climate Services Office and Sylvie Castonguay, Communications and Public Affairs Office
8 NOAA/Climate Prediction Centre
9 Lowe R, C. Barcellos, C.A.S. Coelho, T.C. Bailey, G.E. Coelho, R. Graham, T. Jupp, W.M. Ramalho, M.S. Carvalho, D.B. Stephenson, X. Rodó, 2014: Dengue outlook for the FIFA World Cup in Brazil: an early warning model framework driven by real-time seasonal climate fore- casts. Lancet. Infectious Diseases (Print), v. 14, p. 619 - 626.

Share this page