Appendix D. Explanatory Notes For Figures

Figure 1: Consideration of both natural and human-induced influences on climate is leading to an understanding of the causes of changes in global average temperature since the mid-19th century. Carefully quality-controlled records of surface temperatures over land and ocean regions indicate that the global average temperature has increased about 0.5°C (about 1°F) since the mid-19th century and about 0.7°C (about 1.3°F) since the cool decade that opened the 20th century.

The sharp year-to-year variations in the observed record result both from the limited number of observation stations and from natural fluctuations. Natural fluctuations seem to be caused largely by air-sea interactions (e.g., El Niño events) and by cooling following major volcanic eruptions. Other random fluctuations may also be occurring.

Results from simplified climate models, which have been calibrated to comprehensive general circulation models used to study the climate, have been used to analyze which factors have been influencing global average temperatures. These results (shown in the smoother curves), which are calculated for climate sensitivities of 1.5, 2.5, and 4.5°C temperature increase at equilibrium for a doubling of the carbon dioxide concentration, suggest that several influences have been important. The increases in the concentrations of carbon dioxide and other greenhouse gases have created a strong warming influence. This warming influence has been countered by a regional-to-global cooling influence due to the reflection of a portion of solar radiation back into space by sulfate and other aerosols that have resulted mainly from coal combustion. In addition, changes in the intensity of solar radiation (as indicated by sunspot cycling) have created inter-decadal variations in temperature.

The somewhat chaotic nature of climate fluctuations prevents perfect agreement between model results and observations. Present understanding seems to be capturing most of the major changes. The remaining significant deviations between model results and observations suggest the possible need to consider the roles of additional natural influences, such as volcanic eruptions and snow and other atmospheric interactions with the land surface, which have mainly intra-decadal effects, as well as human influences, such as changes in land cover and tropospheric ozone, which may have multi-decadal effects. However, these influences, which are not yet included in climate models, do not appear to be the dominant factors in long-term climate change.

Figure 2: The central panel shows observations for winter (January-February-March mean) geopotential height anomaly for 1989. The geopotential height is a measure of the build-up of colder and warmer air masses in different regions of the atmosphere. The top and bottom panels show the model-simulated geopotential height anomalies (determined using a model that takes as input observed sea-surface temperatures) for the older and newer versions, respectively, of the atmospheric model of the Center for Ocean-Land- Atmosphere Studies (COLA). The contour interval is 150 m. The solid shading denotes a positive anomaly (indicative of warmer air) of 75 m; the cross- hatched shading denotes a negative anomaly (indicative of colder air) of 75 m. The solid contour denotes a positive anomaly of 225 m; the dashed contour denotes a negative anomaly of 225 m.

These results indicate that, if the sea-surface temperatures can be predicted accurately using ocean models, the present versions of the atmospheric models have the ability to accurately predict seasonal mean climate anomalies over North America and the adjoining areas, thereby allowing predictions of dry and wet and of cold and warm conditions, at least under some climatic conditions.

Figure 3: Reconstructions of the history of temperature and of the atmospheric CO2 concentration going back 150,000 years have been made from the air bubbles trapped in the ice of Antarctica. The baseline temperature (the "0" point) is for average conditions over the past few millenia for the ice cores and for the 19th century for the recent record. Note that the paleoclimatic temperature variations are for Antarctica, so will show larger swings than that for the global average temperatures shown for the period since 1850.

Considerable research has been conducted to understand the causes of the temperature changes. These records can be continued to the present with instrumental observations. For the period prior to human activities, there is general agreement that the natural cycles of solar radiation due to changes in the shape and orientation of the Earthís orbit are likely the major driving force for large-scale swings in temperature.

However, these variations are not enough to explain the glacial cycling unless the radiative effects of changes in the concentrations of CO2 and CH4 are amplifying the influence of variations in solar radiation on the climate. The high correlation in the CO2 and temperature curves suggests that this is indeed happening, so that warmth is associated with high CO2 concentrations and cool conditions with low CO2 concentrations, and that a number of feedbacks must be operating in determining the climatic response.

On the right side of the figure, the CO2 concentration is extended to its present level, then to the level projected for the year 2100 using a central case scenario that assumes no special control measures are adopted (IPCC scenario IS92a). The inset shows the detailed record of the increase in the CO2 concentration over the past 40 years as recorded at the Mauna Loa Observatory in Hawaii by C.D. Keeling and by NOAA. This pioneering record first demonstrated that an increase in CO2 concentration was underway. Based on projected emissions scenarios, climate models predict a sharp rise in the atmospheric CO2 concentration out to 2100, resulting in a few-degree rise in the global average temperature.

Figure 4: Increased UV radiation can cause several types of damaging effects, depending on the energy level and intensity of the radiation and the susceptibility of the receptor of the radiation. The figure shows the percentage rate of increase (in percent per decade) in the exposure to UV radiation for the period 1979 to 1992 for three types of receptors.

This study used daily satellite estimates of cloud cover and information on local terrain height and reflectivity, and changes in ozone and aerosol concentrations to calculate the changes and seasonal variations in surface intensity of UV-radiation as a function of latitude. This constructed record of UV-radiation at the surface was then combined with the action spectra (i.e., the susceptibility to damage) for the skin, for genes, and for plants to estimate the weighted exposures and changes that resulted from 1979 to 1992. Trend lines were then fit so that rates of change could be calculated. Statistically significant increases in exposure are seen in middle and high latitudes, primarily in the spring and summer months when people and plants would be most exposed to sunlight.

Figure 5: The maps show the summer range and relative abundance of the bobolink relative to their maximum density in this domain, as estimated from an empirical-statistical model. The correspondence in predicted patterns with the abundances determined from the Breeding Bird Survey suggests that the empirical model for present conditions (top panel) captures many of the features of the observed abundance of this species over North America. The bottom panel shows the predicted range and abundance of the bobolink assuming the climate change response of a model with a doubled CO2 concentration (see J. Price, Potential impacts of global climate change on the summer distribution of some North American grasslands birds. Ph.D. dissertation, Wayne State University, Detroit, Mich., 1995, 540 pp.).

Figure 6: The wind field shown in the NSCAT figure for 21 September 1996 is typical of conditions near the solar equinox. The trade winds, which are quite strong and show up as intense blue, blow steadily from the cooler subtropical ocean areas toward the Intertropical Convergence Zone (ITCZ) that is located just north of the equator. Instead of blowing north to south, the trade winds are deflected westward by the Earthís rotation (which creates the Coriolis force). The air rises over the warm waters of the ITCZ then sinks in the subtropical regions (called the "horse latitudes"), forming the Hadley Cell circulation. Both the convergence area at the ITCZ and the divergence area in the horse latitudes have relatively low wind speeds (and so are indicated by the paler color).

Two typhoons are shown in the western Pacific. Typhoon Violet is just south of Japan; after these data were taken, Typhoon Violet struck the east coast of Japan causing damage and deaths. Typhoon Tom is located further east and is evolving into an extratropical (mid-latitude) storm.

The image is based on preliminary processing of the first set of NSCAT observations, using prelaunch model functions and calibration. Improvement is expected after the standard calibration and beam balancing procedures. The NSCAT data are objectively interpolated into 12-hourly and 1° longitude by 1° latitude grids (about 100 km) using the methodology described by W. Tang and W.T. Liu (Jet Propulsion Laboratory publication 96-19, 1996) using no other data for initialization.

This preliminary analysis demonstrates that the high spatial resolution of NSCAT and its observing capability under both clear and cloudy conditions will improve the monitoring of severe storms such as typhoons, whose location and intensity are usually not well-defined by conventional methods, and will therefore improve weather forecasts and public-warning capabilities. The NSCAT results also show that NSCATís global coverage will provide a much more detailed and reliable description of atmospheric circulation over the ocean, which will improve predictions of features such as El Niño warmings and other oceanic conditions that play an important role in causing seasonal average climates to change (e.g., because of shifts in the jet stream).

Figure 7: Worldwide sea-level rise is caused as water is added to the oceans when glaciers and ice sheets melt and because the ocean waters increase in volume as they warm (as do most substances due to their positive thermal expansion coefficient). The flooding of islands colonized several centuries ago is evidence that the local sea level has been rising in the Chesapeake Bay.

Over the past 100 years, sea level has risen ~0.3 m (~12 in) in the U.S. mid-Atlantic region. About half of this change has been due to worldwide sea- level rise and about half has been due to a slow sinking of the land due to local factors, including a very long-term response to the removal of glacial ice and a shorter term land-subsidence response due to significant pumping of groundwater. Climate change is projected to cause the rate of global sea-level rise to increase, implying an overall rise from both human influences and the continuing preindustrial rate of rise of ~0.65 m (~26 in) by 2100 in the mid- Atlantic region. Populated place and associated elevation data for this figure were taken from the Geographic Names Information System (GNIS). The GNIS is the official Federal repository of domestic geographic feature names information. USGSís 1:24,000-scale topographic maps are one of several sources of data for the GNIS, which is currently in Phase II of its compilation. The elevations in the GNIS are typically read or interpolated for a single point at or near the center of the populated place. To construct this map, the GNIS was queried for populated places with elevation entries of 1 m or less.

Figure 8: The Crary Ice Rise (83°S, 174°W) is situated in the southeastern corner of the Ross Ice Shelf, Antarctica. Surrounded by about 400-m-thick ice floating on the Ross Sea, the Crary Ice Rise is an ice-capped island that causes the ice to rise nearly 50 m above the adjacent ice shelf. The rise rests in the downstream ice flow from two West Antarctic ice streams. As such, some investigators conjecture that the ice rise presents an obstruction to ice flow, effectively damming the upstream ice and contributing to the stability of the inland ice sheet.

Recent studies suggest that the Crary Ice Rise is changing, which alters the effect that the ice rise has on local ice flow. Based on ice thickness and ice surface velocity data, the shelf immediately downstream of the ice rise is estimated to be thinning by about 1 m/yr. Southwest of the ice stream (downward on the figures), the ice shelf is believed to be thickening by a similar amount. These calculations have led authors to speculate that the orientation of the ice rise may be shifting. Photographs taken by DODís Corona satellite in 1963 (and recently declassified as part of the MEDEA project) and then, 32 years later, by the commercially operated SPOT satellite support this idea. Seemingly fresh, sinusoidal cracks are present in 1963, along the northeastern "hook" of the ice rise. By 1995, the cracks have evolved into segmented blocks that are being rotated by the shearing flow past the ice rise.

Large portions (100-km scale) of this sector of Antarctica are susceptible to dramatic changes in ice flow. Ice streams, large rivers of ice that drain the West Antarctic Ice Sheet into the Ross Ice Shelf, are known to turn "on and off" on time periods of many hundreds of years. This in turn leads to local changes in the grounding line, which is the location where grounded ice starts to float free of the descending ocean floor. It is believed that rapid changes in ice sheet behavior are in fact due to changes in the internal dynamics of the ice sheet. Changes in glacier physical properties, such as subglacial water flow patterns, are presumed to be the important feedback mechanisms controlling rapid changes in the ice sheet, rather than external climate forces.

With an extended delay, the structure of the Crary Ice Rise responds to changes in the inland ice sheet by growing or retreating as the upstream ice flux shifts. The present pattern of thinning along the northeastern margin of the ice shelf and thickening along the southwestern margin may have been repeated in the past. Evidence for this comes from ice thickness patterns far downstream of the ice rise. Because of this, the ice shelf acts as a "tape-recorder," preserving remnants of ice thickness patterns of an earlier time. Paired ice thickness hollows and domes similar to the pattern seen today are located about 200 km downstream of the ice rise. Using present day ice velocities, these observations suggest a similar process occurred around the ice rise some 400 years ago. The Corona and SPOT data suggest researchers may now have an opportunity to study a repeat of that same, and perhaps episodic, process.


Go To: Appendix E: Contact Information
Return To: Table of Contents