Climate Variability
and Change
USGCRP Fiscal Years 2004-2005 Accomplishments

The following are selected highlights of recent research supported by CCSP participating agencies (as reported in the fiscal year 2006 edition of the annual report, Our Changing Planet).

Ancient Climate of the Western United States [1, 19]

Evidence indicates that the climate of northern California over the past 3,500 years has been dominated by El Niño cycles. Paleoclimatologists infer this from the width of annual growth rings of redwood trees, the contents of packrat middens (debris piles), and the chemistry of tiny planktonic foraminifera (shelled, single-cell organisms preserved as fossils in the geologic record). In addition, these scientists concluded that abundance variations in the planktonic foraminifer Globigerinoides sacculifer in marine cores from the western and northern Gulf of Mexico can serve as an effective proxy for the southwestern U.S. monsoon on millennial and sub-millennial time scales. The marine record confirms the presence of a severe multi-century drought centered at approximately 1,600 years before present, as well as several multi-decadal droughts that have been identified in a long tree-ring record spanning the past 2,000 years from west-central New Mexico. The marine record further suggests that the southwest monsoon circulation, thus summer rainfall, was enhanced in the middle Holocene (approximately 6,980-4,710 years before present). The marine proxy record provides the potential for constructing a highly resolved, well-dated, and continuous history of the southwest monsoon for the entire Holocene (approximately the last 10,000 years). Records of U.S. paleoclimate such as those described above provide measures of natural variability and a context for attributing the causes of recent climate change.

Borehole Climate Reconstructions Show 500-Year Warming Trend [18]

A new analysis indicates that the average temperature of the Northern Hemisphere has increased by 1°C since A.D. 1500. This estimate is somewhat greater than Northern Hemisphere temperature reconstructions based on most inferences from tree and ice cores. Vertical temperature profiles obtained from 695 boreholes were used to make this estimate. Boreholes can be used to infer past temperatures because temperature changes at the surface affect the subsurface temperature. These subsurface changes slowly propagate downward, so that temperatures farther underground correspond to temperatures farther back in time.

North American Drought Atlas Indicates Increased Dryness during the Medieval Warm Period [6]

Analyzing paleoclimate records in the western United States over the past 1,200 years, scientists found evidence that elevated aridity in the U.S. West may be a natural response to climate warming. The study revealed that a 400-year-long period of elevated aridity and epic drought occurred in what is now the western United States during the period A.D. 900-1300. This corresponds broadly to the so-called "Medieval Warm Period," a time in which various paleoclimate records indicate unusual warmth over much of the Northern Hemisphere. The study's authors used tree ring records to reconstruct evidence of drought, and also examined a number of independent drought indicators ranging from charcoal in lake sediments to sand dune activity. The team then used published climate model studies to explore mechanisms that link warming with aridity in the western United States. The authors of the new study postulate that certain mechanisms associated with warming may lead to increased prevalence of drought in the western interior region of North America. A CD-ROM summarizing the data, called the North American Drought Atlas. It is the first of its kind, providing a history of drought on this continent. The atlas contains annual maps of reconstructed droughts over North America, including an animation of those maps showing aridity over time.

Observed Thinning of West Antarctic Glaciers [22, 23]

In the wake of the Larsen B Ice Shelf disintegration in 2002 (see Figure 5), glaciers in the Antarctic Peninsula have thinned and their movement to the Weddell Sea has accelerated. According to another study conducted by U.S. and Chilean researchers, glaciers in West Antarctica are shrinking at a rate substantially greater than was observed in the 1990s. They are losing 60% more ice into the Amundsen Sea than they accumulate from inland snowfall. The ice loss from the measured glaciers corresponds to an annual sea-level rise of 0.2 mm (0.008 in), or more than 10% of the total global increase of about 1.8 mm (0.07 in) per year. Ice shelves in the Amundsen Sea appear to be thinning, offering less resistance to their tributary glaciers. The findings suggest that ice shelf breakup may lead to an increased rate of sea-level rise.

Projected Heat Wave Intensity, Frequency, and Duration [16]

Results under a "business-as-usual" scenario using a global coupled climate model (the Parallel Climate Model, PCM) project a geographic pattern to future changes in heat waves. This model projects that areas in Europe and North America could experience more intense, more frequent, and longer lasting heat waves in the second half of the 21 st century. Observations show that present-day heat waves over Europe and North America coincide with specific atmospheric circulation patterns that are projected by the PCM to be intensified by increases in greenhouse gases (see Figure 6).

Projected Changes in Hurricane Intensity and Rainfall [13]

A recent study makes use of nine independent computer simulations of global climate change produced by different research institutions from around the world. In all, 1,300 simulated "hurricanes" were generated using a higher resolution version of a current operational hurricane model forced by 80-year future "business-as-usual" climate model projections. Results of this study indicate a modeled link between surface oceanic warming and a change in the intensity of simulated tropical storms. By 2080, the model-projected changes resulting from approximately doubled carbon dioxide concentrations could cause an average increase of approximately one-half a category in intensity (see Figure 7) and an 18% increase in rainfall within 60 miles of the storm's center. Better understanding is needed of how natural climate variations influence the frequency, severity, and favored paths of hurricanes, how the climate system responds to increased greenhouse gas concentrations, and how key physical processes (e.g., convection) are best represented in climate and hurricane models. It is essential that these advances be accompanied by improved understanding of the factors affecting societal vulnerability to hurricanes, including levels of coastal development, preparedness, and response systems.

Soot from Fossil Fuels Changes Snow Reflectivity [10]

Soot produced as a by-product of fossil-fuel burning can be carried for hundreds of miles before being deposited on the ground. Soot that falls on snow increases the snow's absorption of solar energy, thereby increasing its melting rate. Exposing bare ground may lead to further warming since it generally absorbs more solar energy than snow. A new study indicates that this effect may have contributed to some of the global warming of the past century, including a portion of the trend toward early springs in the Northern Hemisphere, thinning Arctic sea ice, and melting land ice.

Subpolar North Atlantic Circulation Changes during the 1990s [8]

The giant vortex of an ocean current, or gyre, in the northwestern North Atlantic appears to have slowed. Observations of sea-surface height reveal that significant changes have occurred over the past decade in the mid- to high-latitude North Atlantic Ocean. TOPEX/Poseidon altimeter data show that the average sea-surface height increased during the 1990s in this region. The same measurements were used to infer that ocean surface current velocities likely declined during the 1990s in the subpolar gyre. Combining the data from earlier satellites, researchers found that this circulation pattern may have been weakening since at least the late 1970s. Direct observations in the boundary current of the Labrador Sea support this interpretation. The direct observations also indicate that the associated deep underwater current is weakening as well. These changes are potentially significant because the Atlantic Ocean circulation patterns are responsible for redistributing a substantial portion of Earth's heat from low latitudes to high latitudes, making Europe much warmer than it would be without these currents.

New Insights into Predictability of El Niño Events [3]

Forecasts of El Niño climate events are routinely provided and distributed, but the limits of how accurately and how far in advance El Niños can be predicted are still a subject of debate. Previous studies suggest that forecast abilities are largely limited by the effects of high-frequency atmospheric variations, or by the growth of small initial errors in model simulations. In a recent study using an advanced coupled ocean-atmosphere model, researchers made retrospective forecasts of the interannual climate fluctuations in the tropical Pacific Ocean for the period 1857 to 2003. This is several times longer than any previous experiment of this kind. The model demonstrated significant skill in predicting El Niño events back to the 19 th century. Furthermore, strong El Niño events had some predictability up to 2 years in advance. The research suggested that El Niño events may be more predictable than previously thought. Additionally, the study suggests that one of the keys to better El Niño predictions is accurate initial conditions (i.e., the state of the ocean-atmosphere system as determined by observations of ocean temperatures, sea surface temperature, surface winds, etc., at the beginning of the model simulation).

Tropical Sea Surface Temperatures Affect Northern Hemisphere Winter [11]

Recent research suggests that long-term variability in tropical sea surface temperature (SST) has had an important effect on regional changes in the winter climate of the Northern Hemisphere during the last half of the 20th century. The warming of tropical Indian and western Pacific Ocean surface temperatures since 1950 has been related to unusual changes in the winter North Atlantic and European climate. The changes are characterized by a trend in indices of the North Atlantic Oscillation, the most recurrent regional pattern of atmospheric variability in the Northern Hemisphere mid-latitudes.   The changes also include decadal-scale climate variability over the North Pacific Ocean and adjacent continents that affects agricultural harvests, water management, energy supply and demand, and fisheries yields. The link between northern climate variability and the tropical oceans suggests a potential basis for improving climate predictions. This work also indicates the potential value of determining the future course of tropical SST patterns for improving projections of regional climate changes.

Feasibility of Constructing Three-Dimensional Climate Analyses Prior to 1948 [25]

A recent study has shown, for the first time, that it is possible to produce a three-dimensional depiction of the atmosphere back through at least the beginning of the 20th century. A number of such three-dimensional "reanalysis" data sets have been produced by integrating past observations together within state-of-the art climate models – but never before 1948, which is when upper-air observations made by weather balloons became broadly available. Extending the three-dimensional reanalyses to earlier time periods would significantly increase their potential to support a variety of applications that to date have not been possible for the first half of the 20th century, including improving descriptions and understanding of long-term variability in mid-latitude storm systems; improving understanding of the atmospheric circulation associated with extreme events such the 1930s drought in central North America or the prolonged wet period that led to overestimates of precipitation, hence over-allocation of water resources, in the Colorado River; and creating a longer period over which to evaluate climate models driven by changes in greenhouse gases, aerosols, and solar activity. The results suggest that reanalysis is feasible extending back as far as the late 19th century and may yield useful analyses from the surface through much of the lower troposphere at daily time resolution.

Developments in Climate Modeling

A new generation of climate models has significantly improved representations of physical processes, as well as increased resolution, putting them at the forefront of international research. New simulations of climate change during the 20th century have been completed using these models. Various high-end modeling centers sponsored by DOE, NASA, NOAA, and NSF developed and tested the new models. All show significant improvements compared to their predecessors a decade ago.

Figure 8 shows an example of output from recent simulations using four of the leading U.S. climate models, compared with satellite-based observations. Although the detailed evolution of temperature differs in models and observations, both show a common picture of gradual stratospheric cooling, which is caused by the combined effects of ozone depletion and increases in well-mixed greenhouse gases. Superimposed on this overall cooling are the short-term (1- to 2-year) warming signatures of the El Chichón and Pinatubo volcanic eruptions.

U.S. climate modeling capability has advanced significantly in the last 4 years. Output from the four U.S. models shown in Figure 8 is available for the CCSP synthesis and assessment products and the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report . An extensive database, including the output from these models, is archived and made accessible to interested climate researchers through an enabling technology (Earth System Grid) and the Program for Climate Model Diagnosis and Intercomparison (PCMDI). It is anticipated that upcoming analyses of these climate model projections will yield fresh insights into climate variability and change. A special interagency grants program was implemented to accelerate analyses of the 20th century historical simulations. A subset of the data useful for research on the potential effects of climate change on climate-sensitive resources and systems is being made available through the IPCC Data Distribution Center (DDC). A major effort in the coming years will be to assess the capabilities of these and related models for simulating regional climate change. Preliminary indications are that these models possess some skill at regional scales.

Despite the recent model improvements, there are still significant uncertainties associated with aspects of climate models. One of these is the representation of clouds, which remains one of the weakest links in modeling the physical climate system. New integrated approaches have been developed to address this challenge, taking advantage of new high-resolution satellite data, field observations, and small-scale cloud models. Important work in this area is being carried out by the Climate Variability and Predictability (CLIVAR) program's Climate Process Teams supported by CCSP. The Climate Change Prediction Program–Atmospheric Radiation Program Parameterization Testbed (CAPT) project is also addressing the cloud modeling problem through a novel approach that includes analyzing the ability of a climate model to accurately simulate weather events, diagnose the errors, and subsequently improve the model. One example of success in this work is an improved model representation of the processes that trigger precipitation. An improved model representation of vertical cloud overlap has been incorporated in the Geophysical Fluid Dynamics Laboratory (GFDL) and Canadian general circulation models as well as the European Centre for Medium-Range Weather Forecasts numerical weather forecast model.

As noted elsewhere in this report, improvements are being made in understanding and modeling other components of the Earth system, including atmospheric chemistry, ecosystems, and carbon cycling. Efforts are underway to integrate these efforts in increasingly comprehensive Earth system models.