Predicting El Niño and
Other Climate Variations
Successes & Remaining Challenges
Prepared by Richard M. Todaro and David M. Legler,
Office of the US Climate Variability and Predictability Program (CLIVAR)
20 June 2002

The major announcement on March 7, 2002 by scientists with the National Oceanic and Atmospheric Administration (NOAA) that another "El Niño" episode appears to be taking shape indicates the importance associated with forecasting this large-scale, often disruptive climatic event months ahead of time. Fifteen years ago, predictions of El Niño were left largely to adventurous researchers. Progress in predicting El Niño, months and occasionally a year in advance, has been facilitated by nearly twelve years worth of research sponsored by NOAA, NASA, the National Science Foundation, and other federal agencies, and coordinated by the United States Global Change Research Program (USGCRP).

Coordinated Research
Informs Decisions

Established initially under a directive issued by Pres. George H.W. Bush in 1989, the USGCRP was institutionalized by Congress in the Global Change Research Act of 1990 and tasked to coordinate research among various federal agencies into the broad, multi-faceted topic of global environmental change. This included the description and prediction of various forms of "natural variability" in the climate system, most notably the linked ocean-atmosphere El Niño-Southern Oscillation (ENSO) phenomenon.

The ENSO phenomenon manifests itself either in a warm ("El Niño") or cold ("La Niña ") phase, depending on whether the waters of the tropical Pacific from the coast of Peru westward to the International Dateline get warmer or colder than normal. Both warm and cold ENSO events, occurring irregularly every two to seven years, bring about disruptions to the winds that blow high in the atmosphere  as air pressure across the tropical Pacific Ocean shifts back and forth like a see-saw pattern called the  "Southern Oscillation". These changes in turn cause changes to the position of high and low pressure areas and the jet stream/storm track over the middle latitudes, including North America. The result can be dramatic climatic (weather) disruptions.

Prior to the winter of 1982-83, few people outside of the atmospheric and oceanographic research communities had heard of the ENSO phenomenon. In that season, the most powerful warm ENSO ("El Niño") episode ever recorded brought raging storms and devastating floods to the California and Peruvian coasts while bringing extreme drought and wild fires across South Asia, Indonesia, and Australia, as well as southern Africa. Globally, the extreme conditions caused by this warm ENSO caused about 2,000 deaths and $13 billion in damage as reported at the time.

In the wake of this destructive event, scientists all around the world began exploring ways to make accurate predictions about when ENSO events will occur, and it is here that USGCRP comes into play. The first comprehensive effort to understand and pursue prediction of ENSO was the international Tropical Oceans-Global Atmosphere (TOGA) research program. TOGA ran from 1985-1994 under the aegis of the UN World Climate Research Programme.  A global legacy of the TOGA program was the placement of the Tropical Atmosphere - Ocean (TAO) array of buoys that monitor ocean temperatures down about 1500 feet throughout the tropical Pacific Ocean. They also monitor meteorological parameters such as relative humidity, wind, and precipitation. 

The TAO support vessel and TAO buoy. Source: TAO Project Office, NOAA/Pacific Marine Environmental Laboratory.

With the creation of USGCRP in 1990, a new, more focused means for cooperation and coordination among the federal agencies carrying out and funding climate research was introduced and facilitated interagency cooperation. A more comprehensive ENSO observing system soon developed that included the 1992 TOPEX / Poseidon satellite and the 2001 JASON-1 satellite,  NASA's joint missions with the French CNES space agency to measure changes in sea-level height (which are intimately linked with changes in sea temperature as well as ocean circulation).

This system routinely monitors critical ocean changes and has enabled researchers to demonstrate the ability to predict the onset of ENSO-related warming of sea surface temperatures in the eastern Pacific up to a year in advance. Forecasts for changes in climate (e.g. -- for instance, "cooler and wetter across the Southeast") coupled to the predicted ocean temperature changes for ENSO are now being routinely produced. The skill and specificity of these forecasts are still quite crude, and while such forecasts may seem too general to be of much value, in fact they enable state and federal policymakers and ordinary citizens to make all manner of informed decisions. Such decisions often take the form of precautionary measures in an unexpectedly wide range of areas, including agricultural, home construction and repair, transportation infrastructure upkeep, water resource management, various forms of insurance, and coastal flood and hurricane preparedness.

Real-world success of these forecasts occurred in the winter of 1997-98 when an El Niño as powerful to but longer-lived than the 1982-83 occurred. According to a 1998 report by the NOAA Office of Global Programs [PDF] and a 1999 World Meteorological Organization report [PDF], this event caused 24,000 deaths and $33 billion in damage globally in storms, floods, and droughts. This included $2 billion in agricultural losses and $2.6 billion in property losses in the United States (New York Times, March 12, 2002).

There are many  case studies of regional ENSO-related impacts that have motivated the development of seasonal climate forecasts for specific regions in advance of ENSO events. In Florida, El Niño years bring flooding rains in prime agricultural areas. Significant impacts are found on the sugarcane, tomato, bell pepper, and snap bean crop yields, not to mention on the state's famed citrus fruit industry. Dr. Jim O'Brien at Florida State University cites the example of how ENSO seasonal forecasts enabled a few potato farmers in Florida to mitigate, through relatively inexpensive field preparations, the otherwise costly impacts of flooded fields that El Niño-related heavy rains usually bring. In California, the state established the  El Niño Information web site to provide citizens information, including forecasts and predictions, impacts and effects, and contacts to the various state agencies tasked with assorted preparations in numerous sectors (e.g. emergency response, food and agriculture, public utilities) for the anticipated impact of this El Niño. Local officials up and down the coast convened public hearings to ensure El Niño community preparedness.

It was only through careful observations, development of prediction models, and production of climate forecast products made possible by the wide range of USGCRP-facilitated research, that both ENSO climate forecasts and assessments of the societal impacts are now even possible.

But the work is not done.

Stiff Challenges for Researchers

Scientific evidence has indicated that variability of the other tropical oceans influence  the climate of surrounding continents. The  U.S. Climate Variability and Predictability program (CLIVAR) -- one of the components of the USGCRP that focuses  on these types of natural variability-- faces stiff challenges. These include:

• Improving the lead-time and skill of ENSO ocean temperature predictions -- Currently the best models can predict six months in advance (out to a year in special cases) if an ENSO event is likely; however, predictions of how ocean temperatures will change during the course of an ENSO event are inadequate. This fundamentally limits the  capabilities to build better climate-forecast models. In addition, there is the unanswered question of how ENSO is affected by longer-term trends in the global climate system.

• Developing better climate-forecast systems -- Even if ocean temperature predictions were perfect, models used to forecast regional climate effects associated with these changes need vast improvement in order to be able to  deliver useful products to elected officials, policymakers, and ordinary citizens. Improved climate forecast models that consider how the land influences the climate, as well as other critical components of the climate system, must be developed. Such a system needs to be able to faithfully predict the likelihood of extreme (e.g. droughts, floods, heat waves) events in particular regions.

• Describing and predicting other tropical variability -- Variability of the tropical Atlantic is thought to be related in part to the strength of the trade winds and to tropical convection (thunderstorm activity). Tropical Atlantic Variability (TAV) has been shown to influence the climate of North and South America and other regions. Likewise, the variability of the Indian Ocean amy have impacts on South Asia and portions of eastern and southern Africa. Yet we do not understand the fundamental physical mechanisms governing these modes (patterns) of variability, nor the extent to which these patterns impact our ability to make predictions. We do recognize that interactions between the ocean, atmosphere, and land surfaces are very important.

• Modeling the interconnectedness between tropical and mid-latitude variability -- Evidence shows that there is a connection between ENSO, an ocean-air phenomenon of the tropical Pacific, and TAV. Likewise, there appears to be relationships between them and other types of variability outside of the tropics. An example is the connection of ENSO and TAV to another type of natural variability in the atmosphere called the North Atlantic Oscillation (NAO). The NAO is a see-saw pattern in air pressure across the Atlantic Ocean between the Azores and Iceland that has a profound effect on the position of the storm track across eastern North America and Europe -- with resulting big effects on the winter weather in these places. All of these need to be explored to improve climate forecast model capabilities and skills.

Further research in all these areas is vital to determine the over-arching, critically important of question of how the global climate system will evolve over the next century and addressing the question of how human civilization will have to adapt. And USGCRP has a key role to play in coordinating the various components of all this important research.