Climate & Climate Change

Greenhouse gases linked to past African rainfall

BOULDER — New research demonstrates for the first time that an increase in greenhouse gas concentrations thousands of years ago was a key factor in causing substantially more rainfall in two major regions of Africa. The finding provides new evidence that the current increase in greenhouse gases will have an important impact on Africa’s future climate. The study, led by the National Center for Atmospheric Research (NCAR), is being published this week in Science. “The future impact of greenhouse gases on rainfall in Africa is a critical socioeconomic issue,” said NCAR scientist Bette Otto-Bliesner, the lead author. “Africa’s climate seems destined to change, with far-reaching implications for water resources and agriculture.” The research drew on advanced computer simulations and analyses of sediments and other records of past climate. It was funded by the National Science Foundation, which is NCAR’s sponsor, and the Department of Energy Office of Science. A mysterious period of rain Otto-Bliesner and her co-authors in the United States and China set out to understand the reasons behind dramatic climate shifts that took place in Africa thousands of years ago. Lakes and other water features, such as the Ubari Oasis in southern Libya, were more prevalent across now-dry parts of Africa during past periods of more-plentiful precipitation. (Wikimedia Commons photo by Sfivat.) As the ice sheets that had covered large parts of North America and northern Europe started retreating from their maximum extent around 21,000 years ago, Africa’s climate responded in a way that has puzzled scientists. Following a long dry spell during the glacial maximum, the amount of rainfall in Africa abruptly increased, starting around 14,700 years ago and continuing until around 5,000 years ago. So intense was the cumulative rainfall, turning desert into grasslands and savannas, that scientists named the span the African Humid Period (AHP). The puzzling part was why the same precipitation phenomenon occurred simultaneously in two well-separated regions, one north of the equator and one to the south. Previous studies had suggested that, in northern Africa, the AHP was triggered by a ~20,000-year cyclic wobble in Earth’s orbit that resulted in increased summertime heating north of the equator. (In contrast, the northern hemisphere today is closest to the Sun in winter rather than summer.) That summertime heating would have warmed the land in such a way as to strengthen the monsoon winds from the ocean and enhance rainfall. But Otto-Bliesner said the orbital pattern alone would not explain the simultaneous onset of the AHP in southeastern equatorial Africa, south of the equator, since the wobble in Earth’s orbit led to less summertime heating there rather than more. Instead, the study revealed the role of two other factors: a change in Atlantic Ocean circulation that rapidly boosted rainfall in the region, and a rise in greenhouse gas concentrations that helped enhance rainfall across a wide swath of Africa. Tracing multiple causes of a wetter Africa As Earth emerged from the last Ice Age, greenhouse gases, especially carbon dioxide and methane, increased significantly—reaching almost to pre-industrial levels by 11,000 years ago—for reasons that are not yet fully understood. It was, the authors note, the most recent time during which natural global warming was associated with increases in greenhouse gas concentrations. (Because of feedbacks between the two, greenhouse gas concentrations and global temperature often rise and fall together across climate history.) The end of the last Ice Age also triggered an influx of fresh water into the ocean from melting ice sheets in North America and Scandinavia about 17,000 years ago. The fresh water interfered with a critical circulation pattern that transports heat and salinity northward through the Atlantic Ocean, much like a conveyer belt. The weakened circulation led to African precipitation shifting toward southernmost Africa, with rainfall suppressed in northern, equatorial, and east Africa. When the ice sheets stopped melting, the circulation became stronger again, bringing precipitation back into southeastern equatorial and northern Africa. This change, coupled with the orbital shift and the warming by the increasing greenhouse gases, is what triggered the AHP. To piece together the puzzle, the researchers drew on fossil pollen, evidence of former lake levels, and other proxy records indicating past moisture conditions. They focused their work on northern Africa (the present day Sahel region encompassing Niger, Chad, and also northern Nigeria) and southeastern equatorial Africa (the largely forested area of today’s eastern Democratic Republic of Congo, Rwanda, Burundi, and much of Tanzania and Kenya). In addition to the proxy records, they simulated past climate with the NCAR-based Community Climate System Model, a powerful global climate model developed by a broad community of researchers and funded by the National Science Foundation and Department of Energy, and using supercomputers at the Oak Ridge National Laboratory. By comparing the proxy records with the computer simulations, the study demonstrated that the climate model got the AHP right. This helps to validate its role in predicting how rising greenhouse gas concentrations might change rainfall patterns in a highly populated and vulnerable part of the world. “Normally climate simulations cover perhaps a century or take a snapshot of past conditions,” Otto-Bliesner said. “A study like this one, dissecting why the climate evolved as it did over this intriguing 10,000-year period, was more than I thought I would ever see in my career.” About the article Title: Coherent changes of southeastern equatorial and northern African rainfall during the last deglaciation Authors: Bette L. Otto-Bliesner, James M. Russell, Peter U. Clark, Zhengyu Liu, Jonathan T. Overpeck, Bronwen Konecky, Peter deMenocal, Sharon E. Nicholson, Feng He, Zhengyao Lu Publication: Science, doi: 10.1126/science.1259531

El Niño, La Niña & ENSO FAQ

Frequently asked questions What is El Niño? What is La Niña? What is ENSO and what does it stand for? How did these terms originate? How are El Niño and La Niña events defined? How do El Niño and La Niña affect weather in other parts of the world? Do El Niño and La Niña influence global temperature? Will climate change influence El Niño and La Niña? Advances in ENSO research Predicting El Niño and La Niña events What forces cause an El Niño to develop? Related stories Here comes El Niño—but what exactly is it?El Niño or La Nada? The great forecast challenge of 2014¡Hola, La Nada! What happens when El Niño and La Niña take a break? What is El Niño? Sea surface temperature and height anomalies in the eastern Pacific Ocean show the influence of expansion and warming during El Niño (November 1997, top) and cooling during La Niña (November 1998, bottom). (Images courtesy NASA.) El Niño is a warming of the central to eastern tropical Pacific that occurs every two to seven years, on average. The term also commonly refers to the atmospheric rearrangements that occur with the oceanic warming. During an El Niño event, sea surface temperatures across a watery expanse often as large as the United States can warm by 1–3°F or more for a period of from a few months to a year or two.    What is La Niña? La Niña, the counterpart to El Niño, is a cooling of the waters across the same region. As with El Niño, the term La Niña typically is used to refer to the associated atmospheric as well as oceanic patterns. It often lasts longer than El Niño, sometimes persisting or recurring for two or more years.  back to top What is ENSO and what does it stand for? El Niño and La Niña are normally accompanied by variations in the tropical Pacific Ocean’s Walker Circulation (see diagram), as well as a vast see-saw in atmospheric pressure—the Southern Oscillation—that modifies the Walker pattern. The term El Niño–Southern Oscillation, or ENSO, refers to the combination of atmospheric and oceanic effects associated with both El Niño and La Niña. El Niño and La Niña can be thought of as the ocean part of ENSO, while the Southern Oscillation can be thought of as the atmospheric component.  Through the atmosphere, teleconnections occur via waves to influence regions far from the tropical Pacific (see below). This diagram shows the Walker Circulation, a vast loop of air above the equatorial Pacific Ocean. See below for an alternate depiction. Click here or on the image to enlarge. (Illustration by Gabriel Vecchi, UCAR.) ENSO is a coupled phenomenon that would not occur without interactions between the atmosphere and ocean. Sometimes an atmospheric shift can occur in the tropical Pacific without the ocean fully responding, or vice versa. However, for an El Niño or La Niña to develop, the atmosphere and ocean must evolve in sync. For example, the usual east-to-west surface trade winds across the tropical Pacific (see diagram at left) tend to weaken during El Niño; this allows warm water to shift toward the eastern tropical Pacific. That warm water supports the eastward development of showers and thunderstorms, and the resulting atmospheric circulations tend to weaken the trade winds even further. Such positive feedbacks are crucial for the development of both El Niño and La Niña events. ENSO is one of the main sources of year-to-year variability in weather and climate around the world. Research to more fully incorporate this interannual variability into computer models is a major focus for improving long-range forecasting.  back to top How did these terms originate? The name El Niño originated in the region where one of the phenomenon’s local effects was first recognized. Fishing people and other coastal residents of Peru had long noticed a strange feature in the eastern Pacific Ocean waters that border their home. This region of tropical yet relatively cool water is host to one of the world’s most productive fisheries and a large bird population. In the first months of each year, a warm southward current usually modified the cool waters. But every few years, this warming started early (in December), was far stronger, and lasted as long as a year or two. Torrential rains fell on the arid land; as one early observer put it, “the desert becomes a garden.” Warm waters flowing south brought water snakes, bananas, and coconuts from equatorial rain forests. However, the same current shut off the deeper, cooler waters that are crucial to sustaining the region’s marine life. Because these strong events were often observed close to Christmas time, the phenomenon was dubbed El Niño (when capitalized, “the little boy” becomes “the Christ Child” in Spanish). Although “la niña” refers to a young girl, the capitalized term La Niña did not originate in the same way; it was adopted by researchers in the 1970s and 1980s to illustrate the relationship between warming and cooling events in the waters of the eastern tropical Pacific. Some researchers suggested “El Viejo” (the old man) or “anti–El Niño” as alternatives, but La Niña has won out as the standard term.  back to top How are El Niño and La Niña events defined? The most widely used definitions of El Niño and La Niña involve sea surface temperatures (SSTs) across the eastern tropical Pacific (see map below). To qualify as an El Niño event, according to the definition used by the National Oceanic and Atmospheric Administration, SSTs must remain at or above 0.5°C (about 1°F) for at least three months across the region labeled Niño3.4. NOAA provides historical records for temperatures in this key region. A “moderate” event would see readings holding at or above 1.0°C, and a “strong” event would be 1.5°C or higher. Sometimes the term “El Niño conditions” is used when SST readings have hit the El Niño threshold but haven’t been in place long enough to qualify as an El Niño event. Sea-surface temperature, averaged across the regions above, is used to track and diagnose El Niño and La Niña conditions. The Niño3.4 region is most commonly used. For reference, the vertical line at 180 degrees longitude represents the International Date Line. (Image courtesy Jan Null, Golden Gate Weather Services.) The Australian Bureau of Meteorology uses a Niño3.4 SST threshold of 0.8°C rather than 0.5°C to define El Niño, because a greater amount of SST warming is needed to produce impacts in Australia. The definitions for La Niña are the same as those for El Niño, except that the sea surface temperatures are below average rather than above. Each region highlighted in the map can be used to convey the nature of a particular event. Some El Niños show warming focused in the far eastern Pacific (Niño1+2). Others show the greatest oceanic warming across the Niño4 region, with substantially different atmospheric patterns than other El Niño events. This latter type is referred to as “El Niño Modoki” (from the Japanese for “similar but different”). Both El Niño and La Niña are most likely to intensify during northern fall, peak during the winter, and subside in the spring. For this reason, El Niño and La Niña events are often referred to by two adjoining calendar years (e.g., 2005–06). This graphic shows three-month running averages of sea-surface temperature across the Niño3.4 region. (Graph courtesy NOAA.) The strongest El Niño events on record, in 1982–83 and 1997–98, had three-month Niño3.4 averages peaking at 2.2°C and 2.4°C, respectively, according to NOAA’s Oceanic Niño Index table. The strongest La Niña event was in 1973–74, with a three-month Niño3.4 average reaching –2.0°C. An especially prolonged La Niña ran from mid-1998 to early 2001, with the index dipping to –1.7°C.  back to top How do El Niño and La Niña affect weather in other parts of the world? During an El Niño event, as warm surface water in the Pacific is shifted thousands of miles east of its usual location, the showers and thunderstorms nurtured by convection above this warm tropical water also change location. As the rising motion associated with the convection also shifts eastward, it causes other adjustments in atmospheric circulation, well away from the tropical Pacific. These persistent zones of rising and sinking air can shift into new locations for months, causing prolonged wet or dry conditions and related atmospheric heating anomalies. In turn, the anomalous heating sets up planetary-scale waves in the atmosphere that radiate away from the region, especially into the hemisphere experiencing winter. These are “teleconnections”—large-scale, long-lasting shifts in atmospheric circulation that can affect much of the globe. The effects extend throughout the Pacific Rim, across large parts of North America, and on to eastern Africa and other regions. La Niña brings a different set of teleconnections to these and other regions, with some but not all effects roughly opposite to those of El Niño. These maps show the most commonly experienced impacts related to El Niño (“warm episode,” top) and La Niña (“cold episode,” bottom) during the period December to February, when both phenomena tend to be at their strongest. (El Niño and La Niña images courtesy NWS/NCEP Climate Prediction Center.) The maps at right show the most common teleconnections associated with El Niño and La Niña during northern summer and winter. Not every warm- or cool-water event will produce all of these impacts, because other atmospheric features interact with each ENSO event to influence weather and climate around the globe. Weaker events, in particular, may look quite different from the prototypes shown here. El Niño Modoki events (where the warming is concentrated further to the west than usual) show significant differences in teleconnections from other El Niño events. In the United States, a strong El Niño event tends to produce milder- and drier-than-average conditions toward the north and cooler- and wetter-than-average conditions to the south. In California, a strong El Niño very often brings more moisture than usual. However, during the weakest El Niño events, San Francisco and Los Angeles are a bit more likely to be unusually dry than unusually wet. (Meteorologist Jan Null maintains a compilation of additional El Niño “myths and realities.”)  back to top Do El Niño and La Niña influence global temperature? During El Niño, a deep pool of warm water usually restricted to the western tropical Pacific is replaced by a much larger, more shallow pool of warm water that covers most or all of the tropical Pacific. The expanded zone of warm sea surface temperatures allows more heat to be conveyed from the ocean into the atmosphere for months at a time. As a result, globally averaged temperatures often rise by a few tenths of a degree Fahrenheit during the latter stages of a strong El Niño event. Conversely, global temperatures can drop by a similar amount during a La Niña event. NCAR scientist Kevin Trenberth has likened El Niño to a “pressure valve” that releases built-up heat from the oceans into the atmosphere. The oceans cool during El Niño events, while the global atmosphere warms. The red line above shows surface temperatures (over land and ocean), in a running 12-month average, as calculated by NOAA. The vertical, shaded bars indicate El Niño (buff) and La Niña (sky blue) periods, based on Niño3.4 SST (sea surface temperature) anomalies. These anomalies appear in red (warm) and blue (cool) in the lower panel. (Figure 6, “An apparent hiatus in global warming?” Kevin E. Trenberth and John T. Fasullo, Earth’s Future, 2013.)   Scientists often account for ENSO by factoring out these bumps and dips in global temperature when analyzing the long-term trends related to climate change. For example, the first 15 years of this century saw more La Niña than El Niño influence, and global air temperatures showed little rise. Prior to that period, in the 1980s and 1990s, when El Niño events were more frequent, global temperatures rose more sharply. The map at right shows year-to-year fluctuations in global temperature (in red), the long-term trend of rising global temperature (in black), and the starts and stops of all ENSO events (shaded vertical bars). Note that other factors also influence global temperature, such as the eruption of Mt. Pinatubo in 1991. The volcano threw enough sun-blocking material into the atmosphere to cause a drop in global temperatures during 1992, despite the presence of El Niño. Shifts in global temperature, as well as in the likelihood of ENSO events, are closely associated with the state of the Pacific Decadal Oscillation (PDO), a pattern of ocean temperatures that reverses every 20-30 years. More La Niña events tend to be observed when the PDO is negative, and more El Niño events when it is positive. Scientists are not yet sure what prompts the PDO to shift modes.   back to top Will climate change influence El Niño and La Niña? Despite steady improvement, it remains difficult for many global climate models to simulate all aspects of ENSO events, so it has been a challenge to determine whether or how El Niño and La Niña will be affected by rising global temperatures. Some models have suggested that the central Pacific could warm enough to produce El Niño–like conditions on a semipermanent basis. The most recent (2013) assessment by the Intergovernmental Panel on Climate Change concluded: There is high confidence that the El Niño-Southern Oscillation (ENSO) will remain the dominant mode of interannual variability in the tropical Pacific, with global effects in the 21st century. Due to the increase in moisture availability, ENSO-related precipitation variability on regional scales will likely intensify. Natural variations of the amplitude and spatial pattern of ENSO are large and thus confidence in any specific projected change in ENSO and related regional phenomena for the 21st century remains low. [See IPCC AR5, Working Group I, PDF, page 21]  back to top Predicting El Niño and La Niña events Since recognizing more than 40 years ago that the oceanic and atmospheric parts of ENSO are strongly linked, scientists have investigated the combined phenomena from a variety of perspectives, gradually building a deeper understanding of this complex, influential atmospheric pattern. When a major El Niño event struck without warning in 1982–83, scientists recognized the event only after it had formed. This was because of limited observations, primitive computer models, the presence of simultaneous effects from a 1982 eruption of Mexico’s El Chichón, and an incomplete recognition of how El Niño development could vary from one event to another. Over the following decade, the United States, Japan, and other nations deployed instrumented buoys across much of the tropical Pacific. Since these buoys can detect the eastward spread of warm water months before an El Niño hits maturity, they serve as an invaluable early-warning tool. A regional model developed at Lamont-Doherty Earth Observatory generated some of the first model-based ENSO predictions starting in the late 1980s. Today, a number of computer models around the world use current ocean temperatures and atmospheric conditions to project the state of ENSO, looking a year or more into the future. Forecasters examine multi-model ensembles, scrutinizing where various models agree or disagree, in order to issue El Niño and La Niña forecasts. This graphic shows the likelihood of El Niño (red vertical bars), La Niña (blue bars), or neutral conditions (green bars) for overlapping three-month periods starting with November 2014–February 2015 (NDJ), with the likelihoods determined by a consensus of experts based on input from multiple computer models. The thin horizontal lines show the likelihood derived from climatology alone (the average occurrence, based on past experience). (Image courtesy International Research Institute for Climate and Society.) Among the leading sources of regular ENSO forecasts are NOAA’s Climate Prediction Center, the International Research Institute for Climate and Society at Columbia University (working with NOAA), and the Australian Bureau of Meteorology. (Note that the Australian BOM uses a higher threshold for El Niño development than the U.S. definition; see above.) Generally, the strongest ENSO events are predicted more accurately than weaker ones. A 2012 analysis in the Bulletin of the American Meteorological Society evaluated ENSO forecasts from 20 different prediction models for the period 2002 to 2011. The study found that the predictive skill of these models actually declined in the 2000s as compared to the 1980s and 1990s—not because of any loss in model quality (the models themselves had actually improved), but because the weaker, more variable ENSO events in this period made forecasting a far greater challenge. For more about the history of ENSO research, see Here comes El Niño—but what exactly is it?  back to top What forces cause an El Niño to develop? To figure out if a strong El Niño is on the way, forecasters must analyze a series of complex events and processes, ranging from the shifting temperature and structure of subsurface waters across the Pacific, to the evolution of trade winds as well as more localized wind bursts. The most intense Kelvin wave ever recorded across the equatorial Pacific Ocean brought a huge area of warmer-than-average water eastward from April to July 2014, but it failed to produce the long-lasting warm surface waters that define an El Niño event. These monthly cross sections show where water temperatures are departing from average within a slice of the equatorial Pacific, with Indonesia on the left and South America on the right. The legend at right is in multiples of 0.5°C. (Image courtesy Australian Bureau of Meteorology.) One ingredient that appears to be crucial is a multiyear buildup of warm water in the western Pacific. Two types of slow-moving disturbances can also push water upward and downward across the Pacific. Rossby waves travel westward, while Kelvin waves move eastward. In both cases, the sea surface rises, then falls, by up to several inches over a period of weeks to months as the wave passes. Often these effects are transient, but sometimes they can help nudge the Pacific into or out of an El Niño pattern. So can the Madden-Julian Oscillation, a pulse of atmospheric energy generated in the Indian Ocean every few weeks. MJO events can push clusters of showers and thunderstorms eastward across the tropical Pacific. In addition, to get an El Niño going, something must slow or reverse the trade winds. Once that happens, the displaced wind and water can then reinforce the El Niño state in a feedback process lasting for months. However, it’s not easy to turn around those persistent east-to-west winds in the first place. Westerly wind bursts can help do the trick. These clumps of west-to-east wind, pushing directly against the trade winds, can span hundreds of miles and can last a few days to several weeks. They can also kick off the Kelvin waves noted above. Although La Niña can develop independently of El Niño, it often materializes immediately after an El Niño event, as Kelvin waves and other phenomena cause the ocean to “bounce back,” overcorrecting the changes initially brought by El Niño. The La Niña pattern involves a strengthening rather than a partial reversal of both trade winds and the larger Walker Circulation. In part because La Niña more closely resembles the neutral state of the Pacific, it is somewhat easier for a La Niña event to last longer (up to 2–3 years) than an El Niño, which rarely persists for more than a year at a time. Scientific adviser: Kevin Trenberth Writer: Bob Henson Last updated : December 2014  back to top    

Here comes El Niño—but what exactly is it?

December 2, 2014 | It’s been a fascinating journey, watching how scientific and public views of El Niño have evolved over the last several decades. Sometimes you still see El Niño portrayed as a devastating creature, guaranteed to wreak havoc on everything in its supposed path. But the picture has gradually grown more sophisticated as scientists have learned more about the workings of this vast, complex phenomenon and how to communicate its impacts effectively. How did we get from there to here? A few landmarks stand out in the last 50 years of research and communication on El Niño, La Niña, and the El Niño–Southern Oscillation (ENSO).  (For quick definitions of each of these, check out our Weather-Maker Patterns Glossary. For frequentlly asked questions, see the El Niño, La Niña & ENSO FAQ) Linking atmosphere, ocean, and distant weather This diagram shows the Walker Circulation, a vast loop of air above the equatorial Pacific Ocean. (Illustration by Gabriel Vecchi, NOAA/GFDL.) Scientists in the mid-20th century were aware of the tropical Pacific Ocean's Walker Circulation (see diagram), as well as a vast see-saw in atmospheric pressure—the Southern Oscillation—that modified this prototypical pattern. And people living near the coast of Peru and Ecuador knew that the cold upwelling that nourished their fish-packed coastal waters sometimes disappeared for months at a time, especially around Christmas (thus the local name “El Niño,” or “the Christ Child”). These two phenomena were assumed to be unrelated until Jacob Bjerknes published a landmark paper in 1966 (see PDF). The concepts began to mesh further after a major El Niño event in 1972–73, when scientists began to draw firmer links between the oceanic warming of El Niño (which extended much further into the tropical Pacific than researchers had realized) and various atmospheric features that occurred across much of the globe (dubbed (“teleconnections”). The term El Niño–Southern Oscillation (ENSO) soon came into vogue. Pioneering social scientists such as Michael “Mickey” Glantz (then at NCAR, now at the University of Colorado Boulder) delved into the implications of El Niño for people on the ground and the potential usefulness of El Niño predictions for saving lives and livelihoods (see PDF, 1977). Building a detection system Extreme drought across Australia, fostered by a strong El Niño event, fed this wall of dust that swept across Melbourne on February 8, 1983. (Photo by Trevor Farrar, courtesy Australian Bureau of Meteorology and During the northern winter of 1982–83,  the most intense El Niño recorded up to that time brought devastating winter storms to much of California and severe drought and wildfire to Australia, among other widespread impacts. Scientists recognized this event only after it had formed, due to limited observations, primitive computer models, the presence of simultaneous effects from a 1982 eruption of Mexico’s El Chichón, and an incomplete recognition of how El Niño development could vary from one event to another. The next decade saw a concerted effort by the United States, Japan, and other nations to deploy instrumented buoys as part of the Tropical Ocean and Global Atmosphere program (TOGA). Since these buoys can detect the eastward spread of warm water months before an El Niño event hits maturity, they serve as an invaluable early-warning tool. (Unfortunately, it’s been a struggle at times to keep the buoys fully funded and maintained. In early 2014, Nature reported that “nearly half of the buoys in the Tropical Atmosphere Ocean array have failed because of delayed maintenance.”) Dozens of instrumented buoys were moored throughout the equatorial Pacific Ocean in the early 1990s, as part of what’s now called the TAO/TRITON array. The array gathers surface meteorological and oceanographic data and records ocean temperature to a depth of about 1,650 feet. (Photo courtesy NOAA.) Computer models also leapt forward in the 1980s, confirming the ocean-atmosphere linkage and pointing the way toward El Niño forecasts. A regional model developed by Mark Cane and Stephen Zebiak (Lamont-Doherty Earth Observatory) generated some of the first model-based ENSO predictions starting in the late 1980s. Other scientists began to simulate ENSO within global climate models, a task with its own challenges (some of which are still being tackled, as steady progress continues). A landmark field campaign hits the Pacific TOGA mounted one of the largest-ever atmospheric research projects, the Coupled Ocean-Atmosphere Response Experiment (TOGA COARE). Aircraft and ships canvassed the western Pacific tropics for months, and the resulting analyses showed how tightly the ocean and atmosphere components of ENSO were linked. The study is often cited in research papers to this day. El Niño of the century California's Russian River saw damaging floods  in March 1998 in the wake of heavy rain associated with that winter's intense El Niño event. (Photo by Dave Gatley, FEMA.) A massive El Niño, even stronger than the 1982–83 event, took hold in 1997–98. This time, thanks to the TOGA buoy network and more sophisticated satellites, forecasters saw the oceanic precursors well in advance and sounded the alarm ahead of the biggest impacts. These included record rains in California, severe drought in Indonesia, and a deadly tornado outbreak in Florida. The early warning not only helped decision makers charged with public safety get ready: it also piqued the interest of the public, many of whom hadn’t heard the term El Niño before. That label’s pithiness and catchy flair surely helped. Journalists had a field day with “El Niño,” whereas any headline writer would struggle with “the warm phase of the El Niño–Southern Oscillation.” This event’s high-water mark in pop culture came on October 25, 1997, as Chris Farley strode onto Saturday Night Live wearing a pro-wrestling outfit and bellowing, “I am El Niño!” La Niña takes center stage Sea surface temperature and height anomalies in the eastern Pacific Ocean show the influence of expansion and warming during El Niño (November 1997, top) and cooling during La Niña (November 1998, bottom). (Images courtesy NASA.) In surprisingly short order, the intense El Niño conditions of early 1998 segued into a strong La Niña event (1998–99). Over the preceding years, researchers hadn’t been in total agreement on a name for the cool-water counterpart to El Niño—some had proposed “El Viejo” (the old man”), “anti–El Niño,” or “cold event”—but by 1998, La Niña (“the young girl”) had begun to stick. That summer, NCAR’s Glantz brought a diverse group of scientists, journalists, and policy experts to Boulder for the first-ever La Niña Summit (see PDF for executive summary, United Nations University Press for collected papers). Thus far, the 21st century has been a quiet century for El Niño. The last 15 years have been more dominated by La Niña events, part of a Pacific-wide rearrangement of ocean circulation that tends to shift modes every 20–30 years. This phenomenon is referred to as the Pacific Decadal Oscillation or the Interdecadal Pacific Oscillation, depending on which part of the ocean is analyzed. It’s not yet clear what causes the PDO to switch modes from negative (cool sea surface in the central North Pacific) to positive (warm sea surface), or vice versa. But evidence is mounting that the world’s oceans tend to retain more heat deeper down, well below the surface, during the PDO’s negative phase—apparently a big reason why Earth’s atmosphere hasn’t warmed much since the late 1990s. Forecasts, busts, and probabilities Today, a number of computer models around the world use current ocean temperatures and atmospheric conditions to project the state of ENSO, looking a year or more into the future. Forecasters examine multi-model ensembles, scrutinizing where various models agree or disagree, in order to issue El Niño and La Niña forecasts. Among the leading sources of regular ENSO forecasts are NOAA’s Climate Prediction Center, the International Research Institute for Climate and Society at Columbia University, and the Australian Bureau of Meteorology. (Note that the Australian BOM uses a higher threshold for El Niño development than the U.S. definition.) El Niño hype returned in a big way early in 2014, when a huge surge of warm water spread eastward across the tropical Pacific. Some computer models suggested that a major El Niño event could develop by later in the year. But by the autumn of 2014, El Niño hadn’t yet materialized, a no-show that was widely interpreted as a forecast “bust.” Only in November did the signs of an approaching El Niño become more insistent, with sea surface temperatures in the Pacific’s Niño3.4 measurement region finally pushing well above the threshold of 0.5°C above average. Was the forecast really so bad? Northern spring is a notoriously difficult time to predict ENSO evolution. Moreover, the models and forecasters were never in lockstep agreement about what would happen later in 2014. The model consensus posted by the International Research Institute in April (in partnership with NOAA) wasn’t particularly dire, though some models did end up overshooting the mark. As conditions and model solutions evolved, NOAA's predicted likelihood that an El Niño would form by summer 2014 rose from 50% in the March outlook (PDF) to "exceeding 50%" in April (PDF), 65% in May (PDF), and 80% in June (PDF), with that June forecast applying to fall/winter 2014–15 rather than summer 2014. Update: December 2, 2014 | The above paragraph has been revised to incorporate a fuller range of NOAA products issued in early 2014. Lessons from 2014 As of mid-November 2014, the various prediction models examined by the International Research Institute for Climate and Society (IRI) tended to produce a weak to moderate El Niño for winter 2014-15, but with a fair bit of spread among the model solutions. (Image courtesy IRI.) The “missing” El Niño event of mid-2014 reminds tells us that we need to carefully manage our expectations about El Niño prediction. Once the oceanic and atmospheric fingerprints of El Niño begin to take shape in northern fall, as they're doing right now, then we can look more confidently to the impacts likely to occur the following winter, when El Niño events typically peak. However, this isn’t really predicting El Niño itself so much as “nowcasting” the emergence of El Niño’s oceanic warming and predicting the resulting impacts we’re likely to see. The oceanic shift itself remains much more difficult to forecast. The dance between atmosphere and ocean that nurtures and sustains El Niño is simply too complex for any model to capture with total accuracy. That’s why it’s so important to keep probabilities in mind. We can’t expect ironclad certainty on whether or when the eastern Pacific will warm into an El Niño event, but we can use multiple models to help gauge the odds. Similarly, and crucially, since each El Niño is different and since other weather and climate variables are always in the mix, the presence of El Niño doesn’t guarantee that Los Angeles will be doused with midwinter rain or Seattle will be dry and mild. However, it does shift the likelihoods in those directions. In that sense, we might want to think of El Niño less like an appealing buffoon from a comedy sketch—and more like a shrewd oddsmaker from a densely plotted thriller. Dive Deeper El Niño, La Niña & ENSO FAQ (NCAR/UCAR Backgrounder, December 2014) The "Almost" El Niño of 2014 (UCARConnect, December 2014) Writer/contact:Bob Henson, NCAR/UCAR Communications          

Burning questions about winter cold

November 10, 2014 | Old Man Winter seems to have gone maverick in the Northern Hemisphere over the last few years. Take 2014 as an example. It's on track to be the warmest globally in more than a century of record-keeping, with May, June, August, and September all setting world heat records for those particular months. Yet February only managed to tie for 21st warmest globally, mainly because of two regions of prolonged cold across North America and central Eurasia (see map below). Much of North America and central Eurasia were plastered with unusually cold air (blue regions) during February 2014, while most other land areas on Earth ran warmer than average. (Image courtesy National Climatic Data Center.) As people who've endured some of the worst cold and snow in decades try to reconcile that experience with the picture of a warming planet, they're also bracing for what 2014–15 might bring. A highly anticipated shift toward colder-than-average weather over the next few days across the central and eastern United States might be the first volley. With a weak to perhaps moderate El Niño now expected to take shape, the U.S. winter outlook from NOAA calls for temperatures across the Midwest and Northeast to be near or above average, in line with what's most common during El Niño events. However, AccuWeather is calling for recurrent bouts of cold and snow in roughly the same area. These outlooks aren't directly comparable—for example, AccuWeather doesn't specify whether the northeastern cold and snow would be worse than in an average winter—but the contrasting tones do suggest differences of opinion about what's most likely to steer our upcoming winter. Two takes on our upcoming winter: seasonal outlooks issued in October by NOAA (top) and AccuWeather (bottom).  Likewise, there's major disagreement among scientists on exactly what might be causing the more general tendency toward colder winters in places like the central and eastern United States, Europe, and Russia. One factor is the Arctic, where depleted sea ice may be playing a role. Another is the tropical Pacific, where a tendency toward cooler-than-average sea-surface temperatures may also be involved. The debate—one of the liveliest in weather and climate science today—is largely about which factor is most important. The role of the seas, from equator to pole The question of what's put the chill on northern midlatitude winters is tightly linked to debate over the "global warming hiatus," the widely publicized leveling of global temperatures that's run from the late 1990s into the early 2010s. As we've noted before, multiple studies indicate that the world's oceans have mopped up more heat than usual in this period and stored it at great depths, thus accounting for much of the hiatus. Scientists are still exploring which oceanic regions are most involved. This graphic shows globally averaged temperature anomalies (deviations from a 30-year average) in degrees Celsius since 1970 as compiled by NOAA. Most of the hiatus in atmospheric warming since the early 2000s has occurred during the months December through February (DJF, the orange trace above). (Graph adapted from Figure 3 in An apparent hiatus in global warming? Earth's Future, doi:10.1002/2013EF000165.) One fascinating aspect of the hiatus is that it's been concentrated in the months of December through February (DJF)—the period known as meteorological winter in the Northern Hemisphere. This interval has seen a slight drop in global temperature since the late 1990s, while the other nine months have held fairly steady (see map at right). "The strongest pause is in the northern winter," said NCAR's Kevin Trenberth, who recently analyzed seasonal aspects of the hiatus with three NCAR colleagues in the journal Nature Climate Change. While places like Chicago, New York, Berlin, and Moscow have had a few severe winters lately, they've also had some very mild ones. A much larger, more sustained region of cooling lies across the central and eastern tropical Pacific Ocean, where El Niño and La Niña play out (see map below). As it happens, La Niña—the periodic cooling of surface water across the eastern tropical Pacific—has predominated over El Niño since the late 1990s. Both El Niño and La Niña events tend to be at their strongest in DJF. Where it's really cool: The map above was produced by subtracting average November-through-March temperatures for 1976–1998 from the readings for 1998–2012. The most prominent cooling can be found over the northeast Pacific and the eastern equatorial Pacific. (Figure 3(f) from Trenberth et al., Seasonal aspects of the recent pause in surface warming, Nature Climate Change, doi:10.1038/NCLIMATE2341.) La Niña events typically last just a year or two, but a similar pattern can persist across the North Pacific for periods of 20 or 30 years—then flip to an opposite mode, more akin to El Niño, for another 20-30 years. This is the Pacific Decadal Oscillation, and it appears to have worked in concert with specific La Niña events to reinforce the cooler-than-normal waters at the equatorial surface since around 1998. During La Niña, stronger-than-usual trade winds keep warmer equatorial waters shunted toward the western Pacific. Such long-lasting rearrangements can goose the atmosphere into persistent responses that extend over a great distance. Modeling experiments carried out for the Trenberth et al. paper suggest that, especially in DJF, the unusually warm equatorial waters toward the western Pacific near the equator have led to a train of rising and sinking air pockets, called quasi-stationary Rossby waves. Analyzed at the jet-stream level (about 6 miles high), this wave train arcs northward into Alaska, eastward across the Canadian Arctic, then back south over the North Atlantic. Trenberth and colleagues argue that the wave train they identified could be the main culprit disrupting the "polar vortex" (the band of high-altitude winds that normally encircles polar regions), thus allowing for more southward intrusions of Arctic air and poleward surges of mild air. However, another prominent view is that the topsy-turvy pattern might be primarily caused not by the tropical ocean but by the Arctic itself. Some of the leading researchers in this camp include Jennifer Francis (Rutgers University), Judah Cohen (Atmospheric and Environmental Research, or AER), and James Overland (NOAA Pacific Marine Environmental Laboratory). The Arctic-as-driver viewpoint has gotten major play in media coverage over the last several years, and research on it continues to accumulate. How melting ice could lead to big snow Cohen began staking out theoretical territory starting more than a decade ago. Through a series of studies with colleagues at AER and the Massachusetts Institute of Technology, he found that greater-than-usual snowfall extent over Siberia during October correlates strongly with unusual cold and snow several months later, during the heart of winter, across the middle and higher latitudes of North America and Eurasia. The idea, as AER's website explains, is that heavy autumn snow encourages the build-up of deep, cold air masses over Russia, accompanied by a persistent trough, or dip, in the jet stream. As autumn moves toward winter, Rossby waves propagate upward from the trough into the stratosphere, disrupting the polar vortex. This would encourage increased "waviness" in the jet stream and help foster southward-plowing Arctic outbreaks. It would also help explain why the atmospheric hiatus has been focused in DJF, as noted by Cohen and colleagues in 2012 and more recently analyzed in the Trenberth paper. Arctic sea ice normally melts back to its minimum extent in September before regrowing in winter. The last 35 years show an unmistakable drop in September sea-ice extent across the Arctic. (Graph courtesy National Snow and Ice Data Center.) Why would this chain of events occur more often today than, say, 20 or 30 years ago? The biggest change in recent years at high northern latitudes has been the dramatic loss of summer and autumn sea ice, especially since 2007. Cohen hypothesizes that moisture from the Barents and Kara seas—which are now largely unfrozen well into autumn—is helping pump up snowfall totals over Siberia, located just to the southeast. Hunting for a middle ground While these two groups of scientists are often portrayed as being at loggerheads, it's possible that both of their hypothesized processes are at play. And perhaps it's too soon to tell. A review paper by Cohen, Francis, Overland, and eight other researchers, which appeared in Nature Geoscience in August, aims to identify what's generally accepted and what remains in question. "We certainly do not discount a tropical influence on the Arctic or on midlatitude wave patterns," said coauthor James Screen (University of Exeter) in an email to AtmosNews. "Indeed, this has been shown in a number of studies." Led by Cohen, the review paper focuses on Arctic amplification—the well-documented tendency of warming to be strongest at high latitudes, toward the North Pole—and its potential role in midlatitude weather extremes, especially during winter. This is a critical relationship to understand, especially since researchers agree that Arctic amplification is likely to continue. The current lack of consensus on the role of the Arctic in midlatitude weather, they suggest, "may help explain the media portrayal of a polarized view among scientists." When compared to the 1981–2010 mean value, snowfall was more extensive than usual for October (blue and purple areas, in percent above average) across much of the Northern Hemisphere last month. In Eurasia, the amount of snow cover was the second largest (behind only 1976) for any October since satellite monitoring began in 1967. Researchers have connected October snowfall in Eurasia to an increased risk of midlatitude cold outbreaks in the winter. (Map courtesy Rutgers Snow Lab.) The paper notes that both observational analyses and modeling studies lend some support to Cohen's hypothesis involving autumn snow cover in Eurasia. The review also suggests that, along with helping to produce increased snow cover, the delayed freeze-up of nearby sea ice could itself work in tandem with the Siberian cold and snow to displace the jet stream in autumn and produce a more variable polar vortex by early winter. One analysis published in September outlines the physics behind this chain of events, and another recent study suggests that the odds of severe winters in central Eurasia have roughly doubled because of the reduced Barents and Kara sea ice. "I think we can now say that this mechanism has transitioned from a hypothesis to a theory," said Jennifer Francis in an email to AtmosNews. On the other hand, Trenberth argues that the tropical oceans, though distant from the Arctic, hold a vastly larger store of energy, and thus any rearrangement of tropical water ought to have an outsized impact on mid- and high-latitude winter weather. Arctic amplification "is a consequence, not a cause" of the unusual jet-stream patterns, according to a recent essay by Trenberth. He joined John Wallace (University of Washington) and several other leading researchers in a letter published in Science last winter urging caution in attributing Arctic amplification and midlatitude extremes to sea-ice loss. Given enough time, this now-cooking debate could prove moot. A number of scientists expect that by later in the century, overall global warming—amplified in the Arctic—should predominate. A three-year project led by NCAR's Clara Deser and Lantao Sun is now working to quantify the longer-term picture. Their approach: Using two climate models (from NCAR and the UK Met Office) that extend up through the stratosphere, vary the amount of Arctic sea ice, and then look for the effects on mid- and high-latitude climate toward the latter part of this century. Initial results are now being finalized for a soon-to-be published paper. As for the coming winter, snow cover extent across Eurasia in October came in as the second most extensive on record—perhaps a chilly portent. Cohen will be posting a winter outlook in an annually updated report on the NSF website by the end of November. Meanwhile, according to a post by Mashable's Andrew Freedman, NOAA is considering a shift in its own winter outlook in a scheduled November 20 update, in light of the ample Siberian snowpack, El Niño's lackluster onset, and other factors. If global warming does eventually make the roughest winters increasingly rare, it's easy to envision today's youngsters telling their own grandkids about what conditions were like during the legendary cold and snow of the 2010s. Writer/contactBob Henson, NCAR/UCAR Communications    

Progress on decadal climate prediction

September 8, 2014 | If today’s tools for multiyear climate forecasting had been available in the 1990s, they would have revealed that a slowdown in global warming was likely on the way, according to new research. The analysis, led by NCAR’s Gerald Meehl, appears in the journal Nature Climate Change. It highlights the progress being made in decadal climate prediction, in which global models use the observed state of the world’s oceans and their influence on the atmosphere to predict how global climate will evolve over the next few years. Such decadal forecasts, while still subject to large uncertainties, have emerged as a new area of climate science. This has been facilitated by the rapid growth in computing power available to climate scientists, along with the increased sophistication of global models and the availability of higher-quality observations of the climate system, particularly the ocean. After rising rapidly in the 1980s and 1990s, global surface air temperature has plateaued at high levels since around 2000. (Image courtesy NOAA National Climatic Data Center.) Although global temperatures remain close to record highs, they have shown little warming trend over the last 15 years, a phenomenon sometimes referred to as the “early-2000s hiatus”. Almost all of the heat trapped by additional greenhouse gases during this period has been shown to be going into the deeper layers of the world’s oceans. The hiatus was not predicted by the average conditions simulated by earlier climate models because they were not configured to predict decade-by-decade variations. However, to challenge the assumption that no climate model could have foreseen the hiatus, Meehl posed this question: “If we could be transported back to the 1990s with this new decadal prediction capability, a set of current models, and a modern-day supercomputer, could we simulate the hiatus?” Looking at yesterday's future with today's tools To answer this question, Meehl and colleagues applied contemporary models in a “hindcast” experiment using the new methods for decadal climate prediction. The models were started, or “initialized,” with particular past observed conditions in the climate system. The models then simulated the climate over previous time periods where the outcome is known. The researchers drew on 16 models from research centers around the world that were assessed in the most recent report by the Intergovernmental Panel on Climate Change (IPCC). For each year from 1960 through 2005, these models simulated the state of the climate system over the subsequent 3-to-7-year period, including whether the global temperature would be warmer or cooler than it was in the preceding 15-year period. Starting in the late 1990s, the 3-to-7-year forecasts (averaged across each year’s set of models) consistently simulated the leveling of global temperature that was observed after the year 2000. (See image at bottom.) The models also produced the observed pattern of stronger trade winds and cooler-than-normal sea surface temperatures over the tropical Pacific. A previous study by Meehl and colleagues related the observed hiatus of globally averaged surface air temperature to this pattern, which is associated with enhanced heat storage in the subsurface Pacific and other parts of the deeper global oceans. Letting natural variability play out A set of 262 model simulations for the last century that were assessed in the most recent IPCC report show the long-term warming trend produced by greenhouse gases, along with short-term trends produced by natural variability. A total of 10 simulations randomly produced variations for the period 2000–2013 that were similar to those actually observed during this period. Above is a map showing trends in sea surface temperature for those 10 model runs, with the characteristic cooling evident across the tropical Pacific. (Image courtesy Nature Climate Change.) Although scientists are continuing to analyze all the factors that might be driving the hiatus, the new study suggests that natural decade-to-decade climate variability is largely responsible. As part of the same study, Meehl and colleagues analyzed a total of 262 model simulations, each starting in the 1800s and continuing to 2100, that were also assessed in the recent IPCC report. Unlike the short-term predictions that were regularly initialized with observations, these long-term “free-running” simulations did not begin with any particular observed climate conditions. Such free-running simulations are typically averaged together to remove the influence of internal variability that occurs randomly in the models and in the observations. What remains is the climate system’s response to changing conditions such as increasing carbon dioxide. However, the naturally occurring variability in 10 of those simulations happened, by chance, to line up with the internal variability that actually occurred in the observations. These 10 simulations each showed a hiatus much like what was observed from 2000 to 2013, even down to the details of the unusual state of the Pacific Ocean. Meehl pointed out that there is no short-term predictive value in these simulations, since one could not have anticipated beforehand which of the simulations’ internal variability would match the observations. “If we don’t incorporate current conditions, the models can’t tell us how natural variability will evolve over the next few years. However, when we do take into account the observed state of the ocean and atmosphere at the start of a model run, we can get a better idea of what to expect. This is why the new decadal climate predictions show promise,” said Meehl. Decadal climate prediction could thus be applied to estimate when the hiatus in atmospheric warming may end. For example, the UK Met Office now issues a global forecast at the start of each year that extends out for a decade. “There are indications from some of the most recent model simulations that the hiatus could end in the next few years,” Meehl added, “though we need to better quantify the reliability of the forecasts produced with this new technique.” Meehl, Gerald A., Haiyan Teng, and Julie M. Arblaster, “Climate model simulations of the observed early-2000s hiatus of global warming,” Nature Climate Change (2014), doi:10.1038/nclimate2357 Writer/contactBob Henson, NCAR/UCAR Communications Collaborating institutionsCenter for Australian Weather and Climate Research National Center for Atmospheric Research FundersAustralian Bureau of MeteorologyNational Science FoundationU.S. Department of Energy Regional and Global Climate Modeling Program In Graphic Terms A series of "hindcasts" successfully captured the plateau in global air temperature observed in the last decade (yellow circle at right). Shown above are observed temperatures for each year (black asterisks) together with hindcasts (red circles) valid for a five-year period centered on each year. Each hindcast was issued based on conditions observed five years before the year indicated by the red circle. The red lines show the range of results for each set of hindcasts produced for that period. Blue circles show the forecasts that would have been produced by assuming the global temperature five years from now would be the same as today. Temperature anomalies for each year are shown relative to the average for the period 1986–2005. (Image courtesy Nature Climate Change.)

Why do La Niña events linger?

August 12, 2014 | For millions of people, the onset of El Niño or La Niña in northern autumn indicates whether they’re likely to face unusually warm, cold, wet, or dry conditions over the coming winter. A new modeling study pins down the process that apparently determines why La Niña events often last twice as long as typical El Niño events—a result with major implications for seasonal predictions extending more than a year out. “Our work suggests that we can make significant progress on predicting La Niña events and determining how long they might last,” said NCAR scientist Clara Deser. “Because of the disproportionate impact of La Niña on drought throughout the world, this could help lead to concrete social benefits.” Deser recently coauthored a paper with Pedro DiNezio (University of Hawaii) that is now in early release in the Journal of Climate. La Niña conditions were in place during the northern winters of 2010–11 and 2011–12, with major drought impacts observed in the United States. This 10-day average of ocean height across the tropical Pacific, collected by the U.S–French Ocean Surface Topography Mission/Jason-2 satellite, spans a period centered on September 3, 2010, when the La Niña event had already reached moderate levels. Lower ocean heights centered on the equator (blue and purple) indicate cooler-than-average conditions there. (Image courtesy NASA.) Although they’re often portrayed as a pair of opposites, El Niño and La Niña aren’t mirror images of each other. El Niño events—which involve warmer-than-usual water overspreading the eastern tropical Pacific—tend to last no longer than one year. However, in recent decades, almost half of La Niña events—which lead to strong trade winds and cooler-than-usual surface water—have returned for a second consecutive northern winter. This was the case during 2010–11 and 2011–12, two La Niña years that were both associated with devastating U.S. droughts. DiNezio and Deser examined the duration of La Niña in climate simulations performed with the Community Climate System Model (CCSM4). Computer models have long struggled to realistically simulate El Niño and La Niña, but CCSM4 is one of a small number of newer models in which El Niño and La Niña events are produced at a realistic pace and strength, making it suitable for the new research. Oceanic clues to La Niña's staying power Longstanding theories predict that the stronger trade winds associated with La Niña kick off very slow-moving ocean waves that move westward across the Pacific, bounce off the coast, and then, a year later, return to the central Pacific. These waves depress the thermocline—the boundary between frigid deep waters and warmer surface waters—by several dozen feet, which allows the ocean surface to warm and effectively ends La Niña. The team scrutinized this mechanism in 252 La Niña events produced by the CCSM4 in a simulation of 1,300 years of climate. In one out of every three simulated La Niña events, the waves did not succeed in ending the event, and La Niña persisted into a second year. As La Niña becomes stronger, the magnitude of this effect does not increase proportionally—a result observed in the model as well as in observations. This disproportionate, nonlinear behavior occurs because as the thermocline deepens, the resulting warming effect on surface waters reaches a limit. Thus, the depressed thermocline may need to persist for a second year in order to bring a strong La Niña event to an end. This result has implications for predicting two-year La Niña events. Notably, the team found that the depth of the thermocline six months before a La Niña begins was correlated with sea-surface temperature anomalies a year and a half later. “This could provide an 18-month lead time for predicting the return of La Niña conditions for a second year,” said Deser. Pedro N. DiNezio and Clara Deser, Nonlinear controls on the persistence of La Niña, Journal of Climate (2014), doi: Writer/contactBob Henson, NCAR/UCAR Communications Collaborating institutionsNCARInternational Pacific Research Center, University of Hawaii FundersJapan Agency for Marine-Earth Science and TechnologyNASANational Oceanic and Atmospheric AdministrationNational Science FoundationUniversity of Hawaii In Graphic Terms La Niña (left), neutral or "normal" conditions (middle), and El Niño (right) each have a distinctive signature of oceanic and atmospheric processes, as seen in this cross section of the tropical Pacific Ocean extending from Australia and Indonesia (left side of each panel) to the Americas (right side of each panel). Warm sea surface temperatures (reds and oranges) are shunted to the west during an El Niño event but extend eastward during a La Niña event. These warmer surface waters tend to be collocated with showers and thunderstorms and a deeper thermocline, the oceanic boundary separating warmer surface waters from colder, deeper waters. New research relates the depth of the thermocline in the central Pacific to the duration of a subsequent La Niña event. (Diagrams courtesy NOAA Pacific Marine Environment Laboratory.)        

Emperor penguins on the decline?

July 30, 2014 | Emperor penguins, the large, charismatic birds known from their frequent film and TV appearances, are in danger. A collaborative research project is drawing attention to the impending plight of the emperors. By 2100, according to a new study, their numbers will have fallen by around 19% and will continue to decline, qualifying the species for endangered status. Emperor penguin communities entirely ring the continent of Antarctica. Of the 45 known colonies, only one has been extensively studied for decades, and most of the others have never been visited by humans, nor are they likely to be. Emperors live on sea ice off the coast of the continent, and the amount of ice plays a major role in determining the health of a colony. Too much ice and the penguins have a long, debilitating walk to the sea and food; too little ice and the colony is more exposed and vulnerable to predation. A colony of emperor penguins at Australia's Snow Hill Island, October 2009. (Photo by Jenny Varley, Wikimedia Commons.) Stéphanie Jenouvrier (Woods Hole Oceanographic Institution) is an expert on penguin life, and she wanted to project the size of emperor populations into the future as Earth’s climate warms. The problem, she says, was her “limited background on climate science.” Meanwhile, at NCAR, senior scientist Marika Holland is a climate scientist with a longstanding specialty in modeling sea ice changes, although she has never been to Antarctica and has never seen an emperor penguin. Aware of Holland’s previous work, Jenouvrier contacted her and Julienne Stroeve at the University of Colorado’s Cooperative Institute for Research in Environmental Sciences. The three of them collaborated on preliminary studies published in 2009. Jenouvrier received a fellowship from CIRES and worked in Boulder for almost a year, collaborating closely with Holland, Stroeve, Mark Serreze at the CIRES National Snow and Ice Data Center, and other scientists on a follow-up study, published in 2012, and on their most extensive update, recently published in Nature Climate Change. The biologists used long-term data from the one well-studied emperor colony, off the coast of Terre Adélie, to estimate the relationship between sea ice and rates of breeding success and survival of chicks. They used the record of penguin population and sea ice concentrations at Terre Adélie to estimate vital rates (births/deaths) and population dynamics at each colony. Learning each other’s languages: biology and climate The next challenge was to project sea ice changes over the rest of the 21st century and relate that to the health of each penguin community. Sea ice off Antarctica does not behave uniformly: although the total area of sea ice around the continent has increased somewhat in recent years, the trends vary by region during the course of a year and over longer periods. Sea ice must therefore be studied in the relatively small segments that host individual colonies, in order to assess the viability of penguin populations. At first, says Holland, the biologists and climate modelers spoke two different languages, which was “a bit of a barrier.” The frequent interchanges during Jenouvrier’s year in Boulder helped bridge that gap, she adds. The sea ice scientists began with a group of 20 or so climate models and settled on a widely used midrange scenario of emissions produced for the Intergovernmental Panel on Climate Change called SRES A1B. Once the penguin population models and sea ice change models were set, climate projections were fed into the penguin population models. Due to inherent uncertainties in the models, Jenouvrier ran tens of thousands of computer simulations to achieve the results that the team published. They found that sea ice will generally decline and its variability will increase by the end of this century. As a result, the simulations indicate that emperor populations will increase by around 10% through midcentury, but then decline to 19% below current levels by 2100. One group of 7 colonies facing the Ross Sea will still be non-threatened by that time, although with a reduced population. On the other side of the continent, facing the Indian Ocean and Weddell Sea, 10 colonies will face quasi-extinction. Most of the rest will qualify as endangered. The collaborative nature of a study like this, Holland says, allows the expertise of NCAR scientists to inform such other fields as biology and economics to better understand the global system. The researchers conclude that the emperor penguin is “fully deserving of Endangered status due to climate change, and can act as an iconic example of a new global conservation paradigm for species threatened by future climate change.” WriterHarvey LeifertContactDavid Hosansky, NCAR & UCAR Communications Collaborating institutionsNational Center for Atmospheric ResearchUniversity of AmsterdamUniversity College LondonUniversity of Colorado/Cooperative Institute for Research in Environmental SciencesUniversity of La RochelleWoods Hole Oceanographic Institution FundersAlexander von Humboldt FoundationEuropean Research CouncilGrayce B. Kerr FundNational Oceanic and Atmospheric AdministrationNational Science FoundationPenzance Endowed Fund in Support of Assistant ScientistsWoods Hole Oceanographic Institution   Dive deeper Stéphanie Jenouvrier, Marika Holland, Julienne Stroeve, Mark Serreze, Christophe Barbraud, Henri Weimerskirch and Hal Caswell, Projected continent-wide declines of the emperor penguin under climate change, Nature Climate Change (2014), doi:10.1038/nclimate2280 In Graphic Terms Annual mean change of sea ice concentrations (SIC) between the twentieth and twenty-first centuries and conservation status of emperor penguin colonies by 2100. SIC projections were obtained from a subset of atmosphere-ocean general circulation models. Dot numbers refer to each colony evaluated, with dot color showing conservation status (red = quasi-extinct, orange = endangered, yellow = vulnerable, green = not threatened).  (Figure 1 from Jenouvrier et al., Projected continent-wide declines of the emperor penguin under climate change, doi:10.1038/nclimate2280; image courtesy Nature Climate Change.)    

Climate experts estimate risk of rapid crop slowdown

BOULDER – The world faces a small but substantially increased risk over the next two decades of a major slowdown in the growth of global crop yields because of climate change, new research finds. The authors, from Stanford University and the National Center for Atmospheric Research, say the odds of a major production slowdown of wheat and corn, even with a warming climate, are not very high. But the risk is about 20 times more significant than it would be without global warming, and it may require planning by organizations that are affected by international food availability and price.  “Climate change has substantially increased the prospect that crop production will fail to keep up with rising demand in the next 20 years,” said NCAR scientist Claudia Tebaldi, a co-author of the study. Stanford professor David Lobell said he wanted to study the potential impact of climate change on agriculture in the next two decades because of questions he has received from stakeholders and decision makers in governments and the private sector. “I’m often asked whether climate change will threaten food supply, as if it’s a simple yes or no answer,” Lobell said. “The truth is that over a 10- or 20-year period, it depends largely on how fast the Earth warms, and we can’t predict the pace of warming very precisely. So the best we can do is try to determine the odds.” A storm looms behind wheat fields in eastern Colorado, where recurrent drought has had major impacts on agriculture over the last 15 years. (©UCAR, photo by Carlye Calvin. This image is freely available for media & nonprofit use.) Lobell and Tebaldi used computer models of global climate, as well as data about weather and crops, to calculate the chances that climatic trends would have a negative effect of 10 percent on yields in the next 20 years. This would have a major impact on food supply. Yields would continue to increase but the slowdown would effectively cut the projected rate of increase by about half at the same time that demand is projected to grow sharply. They found that the likelihood of natural climate shifts causing such a slowdown over the next 20 years is only 1 in 200. But when the authors accounted for human-induced global warming, they found that the odds jumped to 1 in 10 for corn and 1 in 20 for wheat. The study appears in this month’s issue of Environmental Research Letters. It was funded by the National Science Foundation (NSF), which is NCAR’s sponsor, and by the U.S. Department of Energy (DOE). More crops needed worldwide Global yields of crops such as corn and wheat have typically increased by about 1-2 percent per year in recent decades, and the U.N. Food and Agriculture Organization projects that global production of major crops will increase by 13 percent per decade through 2030—likely the fastest rate of increase during the coming century. However, global demand for crops is also expected to rise rapidly during the next two decades because of population growth, greater per-capita food consumption, and increasing use of biofuels. Lobell and Tebaldi set out to estimate the odds that climate change could interfere with the ability of crop producers to keep up with demand. Whereas other climate research had looked at the crop impacts that were most likely, Lobell and Tebaldi decided to focus on the less likely but potentially more dangerous scenario that climate change would reduce yield growth by 10 percent or more. The researchers used simulations available from an NCAR-based climate model (developed by teams of scientists with support from NSF and DOE), as well as several other models, to provide trends in temperature and precipitation over the next two decades for crop-intensive regions under a scenario of increasing carbon dioxide. They also used the same model simulations without human-caused increases in carbon dioxide to assess the same trends in a natural climate. In addition, they ran statistical analyses to estimate the impacts of changes in temperature and precipitation on wheat and corn yields in various regions of the globe and during specific times of the year that coincide with the most important times of the growing seasons for those two crops. The authors quantified the extent to which warming temperatures would correlate with reduced yields. For example, an increase of 1 degree Celsius (1.8 degrees Fahrenheit) would slow corn yields by 7 percent and wheat yields by 6 percent. Depending on the crop-growing region, the odds of such a temperature increase in the next 20 years were about 30 to 40 percent in simulations that included increases in carbon dioxide. In contrast, such temperature increases had a much lower chance of occurring in stimulations that included only natural variability, not human-induced climate change. Although society could offset the climate impacts by planting wheat and corn in cooler regions, such planting shifts to date have not occurred quickly enough to offset warmer temperatures, the study warned. The authors also found little evidence that other adaptation strategies, such as changes in crop varieties or growing practices would totally offset the impact of warming temperatures. “Although further study may prove otherwise, we do not anticipate adaptation being fast enough to significantly alter the near-term risks estimated in this paper,” they wrote. “We can’t predict whether a major slowdown in crop growth will actually happen, and the odds are still fairly low,” said Tebaldi. “But climate change has increased the odds to the point that organizations concerned with food security or global stability need to be aware of this risk.” About the article Title: Getting caught with our plants down: the risks of a global crop yield slowdown from climate trends in the next two decades Authors: David B. Lobell and Claudia Tebaldi Publication: Environmental Research Letters – doi:10.1088/1748-9326/9/7/074003

Pollution, wildfires, and a warming California

June 25, 2014 | California will likely experience more large fires in forested areas this century because of rising temperatures and changes in precipitation along with development patterns, new research finds. The resulting increase in wildfire activity could increase some types of fire-generated air pollution by more than half. The computer modeling study, led by Matthew Hurteau of Pennsylvania State University with co-authors including NCAR scientist Christine Wiedinmyer, used several development and climate scenarios to project where burned areas are likely to occur based on changes in hydrology, human population, and land-surface characteristics. The authors then estimated the emissions, which can have a significant impact on air pollution. The work was published earlier this year in Environmental Science & Technology. A smoky future To estimate the potential impacts on air pollution, the researchers turned to a computer model known as Fire Inventory from NCAR, or FINN. The model enabled the team to simulate emissions of greenhouse gases (carbon dioxide and methane) and fine particulate matter from the amount of burned plant materials. They also used FINN to estimate emissions of non-methane organic compounds that, when combined with nitrogen oxides (NOx), can react to form ground-level ozone, a pollutant that harms plants and causes airway and lung irritation. The results for a climate scenario projecting a medium-high increase in temperature suggest that, by 2050, total California wildfire emissions would increase by a median value of 24 percent over 1961–1990 values. The median value reaches 56 percent by the year 2085, with a range from 19 to about 100 percent, depending on how development and population evolve. The largest increases come from projected fires in the national forests of northern California. The team also examined the amount of carbon dioxide stored in plants that would be emitted to the atmosphere by burning wildfires each year. They found that, by the year 2085, half of the modeled scenarios projected emissions within a range of 16.5 to 19.6 teragrams of CO2 per year, compared to 10.7–12.2 Tg per year in 1970. Emissions of particulates, NOx, and non-methane organic compounds were found to increase in all emission scenarios modeled, and they would be likely to significantly affect California’s air quality. “Based on what we know now, the probability of having increased fires in California is pretty high. That’s going to lead to a lot more air pollution from fire, which could have big impacts on air quality,” Wiedinmyer said. “Moving forward, these wildfire effects need to be considered in air quality management plans.” The group is now working on modeling future wildfire and the resulting air pollution for the entire western United States. Matthew D. Hurteau, Anthony L. Westerling, Christine Wiedinmyer , and Benjamin P. Bryant, Projected Effects of Climate and Development on California Wildfire Emissions through 2100, Environmental Science & Technology 2014, 48, 2298−2304 DOI: Writer/contactDavid Hosansky, NCAR & UCAR Communications Collaborating organizationsUniversity of California, MercedNational Center for Atmospheric ResearchPennsylvania State UniversityRAND Corporation FundersBureau of Land ManagementCalifornia Energy CommissionNational Oceanic and Atmospheric AdministrationNational Science FoundationU.S. Department of Agriculture

A deeper dive into 21st-century climate

June 4, 2014 | Twice as much flour and water should give you twice as much bread. Earth’s atmosphere doesn’t respond to greenhouse gases quite that simply. But researchers are finding new ways to work with some aspects of climate change that are surprisingly linear. This work could point the way to saving time and money in future climate research while producing a richer set of studies—and a richer range of information for policy makers—on various climate change scenarios and how society might mitigate and/or adapt to the changes. NCAR senior research associate Tom Wigley, now at the University of Adelaide, delivered a keynote address at NCAR on April 23 as part of a workshop on approaches to pattern scaling. Wigley teamed with Benjamin Santer (Lawrence Livermore National Laboratory) in 1990 on the first paper describing the technique of pattern scaling. Wigley’s MAGICC/SCENGEN software package, which uses the technique to allow researchers to scale model output for various emissions levels, has been downloaded by thousands of users. (©UCAR. Photo by Bob Henson. This image is freely available for media & nonprofit use.) Several dozen experts gathered at NCAR in April to discuss “pattern scaling,” one of several statistical approaches that allow scientists to interpolate between two highly detailed simulations and get simpler but still-useful results that fill in the gap between them. Pattern scaling is designed to map out how climate might evolve if emission pathways differ from those that have been studied intensively through full simulations with global climate models. More than four possible futures Part of the challenge in preparing for climate change is that it can take thousands of hours of costly supercomputer time to produce a single projection of how climate might unfold over the next century. Moreover, nobody knows the extent to which nations and industries might act to stem the rise of greenhouse gases. Because of this, it takes multiple simulations—with emissions allowed to plummet in some runs and to grow unchecked in others—in order to provide a broad range of possible trajectories for global climate. For example, the modeling carried out for the recent Fifth Assessment Report from the Intergovernmental Panel on Climate Change involved four different “representative concentration pathways.” Each of these RCPs specify the amount of carbon dioxide in the atmosphere expected by a certain point (such as the year 2100) (see graphic). Pattern scaling aims to fill in the gaps between these simulations. The technique could assist policy makers who want to know the impacts of a different emission level, such as something in between two of the four existing RCPs. It’s also widely used in other types of research on climate impacts. Where does it work best? Pattern scaling appears most effective in projecting changes at large regional scales, where scientists expect a climate model to be most skillful. For example, it’s well established that emitting more greenhouse gases will tend to make subtropical areas drier and subpolar regions wetter. The models also show polar regions warming more quickly than lower latitudes, a trend already being observed. These changes should be roughly proportional to how much global average temperature rises, which in turn is a function of how much total greenhouse gas is emitted. The latter relationship is easily simulated with very simple and inexpensive models. Pattern scaling allows those results to be translated into projections of future climate change across large regions in a computationally efficient way. Not all phenomena lend themselves to the pattern-scaling approach. It doesn’t capture the effects of internal (natural) climate variability, which can temporarily mask the signal of human-caused changes. As an example, much of the southeastern United States has seen little temperature rise over the last century. Climate models didn’t predict this regional “warming hole,” since they’re not designed to track the ups and downs of internal variability. It’s also difficult to use pattern scaling to determine how quickly the warming hole will diminish, as it’s eventually expected to do. Pattern scaling also has its limits when warming pushes climate past a major threshold, such as the melting of Arctic summertime sea ice. Participants came out of the workshop agreeing that it’s time for a more thorough evaluation of pattern scaling. They want to determine where and how pattern scaling might be used more extensively. For example, it could play a role in research leading up to the next major assessment from the Intergovernmental Panel on Climate Change. “Pattern scaling can’t completely replace the need for comprehensive Earth system model runs, but perhaps we can rely on it more than we think we can now,” says NCAR scientist Brian O’Neill, who helped coordinate the workshop. Writer/contactBob Henson, NCAR & UCAR Communications Workshop coordinatorsBrian O’Neill, NCARClaudia Tebaldi, NCARJames Murphy, UK Met OfficeTim Carter, Finnish Environment Institute FundersNational Science FoundationWorld Meteorological Organization Dive deeper Workshop website: "Pattern Scaling, Climate Model Emulators, and their Application to the New Scenario Process," April 23-25, 2014 In Graphic Terms Pattern scaling helps researchers interpolate between results from widely differing model scenarios, including the four representative concentration pathways (RCPs) used in the 2013-14 Fifth Assessment Report from the Intergovernmental Panel on Climate Change (IPCC). This image is drawn from FAQ 12.1 in Chapter 12 of the Working Group I report. Left: Global mean temperature change averaged across all Coupled Model Intercomparison Project Phase 5 (CMIP5) models (relative to 1986–2005) for the four RCP scenarios: RCP2.6 (dark blue), RCP4.5 (light blue), RCP6.0 (orange) and RCP8.5 (red); 32, 42, 25 and 39 models were used respectively for these four scenarios. Likely ranges for global temperature change by the end of the 21st century are indicated by vertical bars. (Note that these ranges apply to the difference between two 20-year means, 2081–2100 relative to 1986–2005, which accounts for the bars being centered at a smaller value than the end point of the annual trajectories.) Right: For the highest (RCP8.5) and lowest (RCP2.6) scenario, these maps show examples of surface temperature change at the end of the 21st century (2081–2100 relative to 1986–2005) from two CMIP5 models. These models are chosen to show a rather broad range of response, but this particular set is not representative of any measure of model response uncertainty. (Image courtesy IPCC.)


Subscribe to Climate & Climate Change