EPA/600/A-97/049
General Circulation Model Scenarios for the Southern United States
Ellen J. Cooter1
Atmospheric Sciences Modeling Division
Air Resources Laboratory
National Oceanic and Atmospheric Administration
Research Triangle Park, NC 2771 1
1On assignment to the National Exposure Research Laboratory, U.S.
Environmental Protection Agency.
-------
-------
1. Introduction
This chapter provides background climatological information for the Southern
Global Change Program (SGCP) assessment. A brief description of the current
climatological setting of the Southern U.S. is provided; General Circulation Models
(GCM's), a fundamental tool for projecting possible future climate conditions, are
introduced, and their use in the development of the SGCP climate change scenario
data base is described. The discussion closes with the presentation of a suite of
possible future climate conditions (scenarios) for the Southern Global Change Program
study area.
2. The Current Climatological Setting
For the purposes of the SGCP, the "Southern U.S." extends roughly from 75°
to 100° West Longitude and from 30° to 37° North Latitude (Figure 1). Elevations
within the region range from near sea level along the Gulf and Atlantic coasts to more
than 1800 meters in the Appalachian Mountains. The climate of locations within the
Southeastern U.S. is determined primarily by latitude, proximity to the Gulf of Mexico
and the Atlantic Ocean and by altitude. Overall, the climate is temperate, becoming
largely subtropical near the coast. Summers are long, hot and humid with little day-to-
day temperature change. Late June to mid August receive local afternoon
Thundershowers. The coldest months are December, January and February, during
which the region is subject to frequent shifts between exposure to moist mild Gulf air
and cool, dry continental air. Severely cold weather seldom occurs. Except at higher
1
-------
elevations, temperatures of zero or lower are rare and occur only when there is snow
on the ground. The last spring freeze (T<-2.2°C) generally occurs between March 1 5
and April 1 5. The first fall freeze (T <-2.2°C) generally occurs between October 15
and November 15 (Koss, et al., 1988).
Precipitation is nearly all in the form of rain and varies greatly from year to year.
Snow can occur frequently at higher elevations, but is usually reported only once or
twice a year at most locations. Inter-annual rainfall patterns range from bi-modal in
the Southern Coastal regions of Mississippi, Alabama and Georgia, to nearly uniform
in North Carolina and Virginia. Nearly all precipitation is from local thundershowers,
which most frequently occur in the afternoon. During late August and in September,
summer conditions of atmospheric temperature and moisture persist, but thunder
showers become less frequent. However, late night and early morning
thundershowers, characteristic of late summer on the coast, continue until mid-
September. Rains during October are nearly always from showers or thundershowers
occurring ahead of temperature drops. Such changes become more frequent and more
pronounced as winter approaches. Dry, sunny weather prevails most of the time in
September and October, but from August through early October, heavy general rain
may occur when tropical disturbances or hurricanes move inland from the Gulf of
Mexico or Atlantic Ocean.
Droughts may occur any time during the growing season from late April through
October. Relatively long periods with little or no rain are more likely to occur in late
summer and autumn than at any other time, while a secondary maxima of such
2
-------
periods occurs in May and June. Severe local droughts occur nearly every year, but
severe region-wide droughts are relatively rare.
SGCP scientists needed climatological data which accurately reflected these
historical temporal and spatial characteristics, as well as potential changed climate
conditions under increased concentrations of atmospheric C02. A brief description of
the historical data base and methodologies employed to project changes from these
historical conditions will be presented. This is followed by a summary of region-wide
and within-region changes projected by four GCM's and applied to the Southern U.S.
historical data base.
3. A Historical Climate Data Base for Southern U.S. Applications
3.1 TEMPERATURE AND PRECIPITATION
A discussion of the current state of biological modeling of forests is presented
in Dale and Rauscher (1994). With few exceptions, the SGCP projects, which include
several of these models, focussed on the landscape or smaller spatial and temporal
scales (104 ha or smaller and centuries or shorter time). Using this information, it
was determined that the historical data base should provide information for the
Southern U.S., should represent an area as close to 100 km2 (approximately .1° lat)
as possible, and should contain long, continuous time series of daily maximum
temperature, minimum temperature, precipitation, humidity and solar radiation. The
3
-------
data base should adequately capture naturally occurring spatial and temporal inter-
correlation across time series and inter- and intra-annual variability within time series.
The Richman-Lamb historical data base meets many, but not all of these
requirements. The data (daily maximum temperature, minimum temperature and
precipitation) are spatially and temporally consistent, continuous through time, are of
the greatest spatial resolution practicable using observational data (a nearly regular 1°
latitude x longitude grid) and are easily accessed. Research using these data indicates
that they are of sufficient quality and are appropriate for regional impact assessment
applications (Richman and Lamb, 1985; Richman and Montroy, 1996), but should not
be used for the detection of subtle, small-scale changes or local long-term climate
trends. While failing to immediately fulfill all of the stated SGCP needs, the many
strengths of the data base outweigh its limitations. A detailed description of the
Richman-Lamb data base is provided in Cooter, et al. (submitted)
3.2 SOLAR RADIATION
One of the most important and least available climate variables required by
biological models is solar radiation. The most frequently reported radiation component
is global radiation. Global radiation is the sum of incoming direct and diffuse
shortwave radiation received by (incident upon) a unit horizontal surface. If this
surface is a leaf, then global radiation can be partitioned into that which is absorbed
(photosynthetically active, PAR), that which is reflected and that which is transmitted
(Bannister, 1976). Most biological calculations of leaf energy budgets begin on the
4
-------
basis of a plane, isolated and horizontal leaf in the air. These estimates can then be
adjusted for plant-specific physiology and canopy architecture.
Global radiation observations on the same spatial scale as the temperature and
precipitation data base were not available, and so several stochastic models of solar
radiation were considered. The model reported in Hodges et al, (1985) was selected.
The original model parameterizations were developed for use in regional crop
assessments with the CERES-MAIZE physiological model (Hodges et al., 1987). This
combination was later used for regional climate change impact assessment studies
(Smith and Tirpack, 1990; Cooter, 1990). It is well-suited for regional time series
studies because of its minimal input requirements and its response to precipitation
events and inter-annual variability.
The Hodges model was re-parameterized for the Southeastern U.S. using
observed and modeled global radiation data from the Solar and Meteorological Surface
Observation Network (SAMSON; NREL, 1992) which improved its representation of
spatial and temporal patterns of historical solar radiation time series. Details of the
model revision and performance in two maize process models are presented in Cooter
and Dhakhwa (1995).
3.3 VAPOR PRESSURE DEFICIT
SGCP biological models utilize a variety of different humidity estimates. Vapor
pressure deficit (VPD) and relative humidity are used most frequently. Humidity is
often modeled as directly impacting physiological plant processes through stomatal
5
-------
response to VPD. Indirect effects of atmospheric humidity are included through
simulated environmental stresses such as the availability and transport of moisture and
nutrients.
Vapor pressure deficit (VPD) was selected to represent atmospheric humidity
for the SGCP. Historical humidity observations are not available on the same spatial
scale as temperature and precipitation and so VPD is defined as:
VPD = es - e (1)
where VPD is day-average vapor pressure deficit in millibars. The vapor pressure, e,
is determined from the dewpoint temperature and is given by:
e _ g -J Q7g*e(17.269*SDEWj/(237.3 - SDEW)
The saturation vapor pressure, es is determined from the average temperature for the
day and is given by:
e _ g -| Q7g *g(17.269*STEMP)/!237.3 + STE.VIP) |2)
STEMP is the day average temperature and SDEW is the dewpoint in °C (Murray,
1967).
Dew point temperature observations are not widely available, but are often
estimated as being approximately equivalent to the 24-hour minimum temperature.
Friend (submitted) found that this assumption provided good estimates of global scale
humidities in Europe and North America, but not in Africa. Running et al. (1987) also
6
-------
found that the correlation between minimum temperature and dew point temperature
deteriorated in arid environments. Following up on this finding, Kimball et al. (1996)
examined the impact of this assumption on estimated humidity at first Order
Meteorological Station sites throughout the U.S., including the SGCP study area.
They found that, for data from Humid subtropical regions (which includes all of the
SGCP area, including sites as far west as Tulsa, Oklahoma and Austin, Texas), the use
of minimum daily temperature versus reported dew point temperature resulted in VPD
differences of less than .1 kPa. Therefore, while there are questions regarding the
validity of the dewpoint/minimum temperature assumption for locations outside the
SGCP study area, it appears to be reasonable for the present assessment.
4. GCM's - Tools for Predicting Change
In 1990 the IPCC (Intergovernmental Panel on Climate Change) announced
that, while recognizing the considerable uncertainty in the entire issue of climate
change, several statements concerning anticipated climate change in response to
increases in C02 and other trace gases could be made with some certainty {Table 1)
(IPCC, 1990S. These include statements concerning changes in temperature,
precipitation, soil moisture, snow and sea-ice.
In 1992, the IPCC issued a supplement to the original report in which current
findings concerning observed climate variability and change were reported. A partial
list of findings, which includes that of the observation of surface and mid-tropospheric
warming, is provided in Table 2. The executive summary concludes:
7
-------
"It is still not possible to attribute with high confidence all, or even a
large part of, the observed global warming to the enhanced greenhouse
effect. On the other hand, it is not possible to refute the claim that
greenhouse-gas-induced climate change has contributed substantially to
the observed warming." - (Folland, et al., 1992).
General Circulation models (GCM's) are widely considered one of the best tools
available for the exploration of future climate conditions under great uncertainty.
Although GCMs are far from being acurate tools of prediction, it has been
demonstrated that i) some are able to simulate the important large-scale features of
the climate system well, including seasonal, geographical and vertical variations of
forcing and dynamics in space and time; ii) many climate changes are consistently
projected by different models in response to greenhouse gases and aerosols and can
be explained in terms of physical processes which are known to be operating in the
real world and iii) GCM results exhibit "natural" variability on a wide range of time- and
space-scales which is broadly comparable to that observed (IPCC, 1996).
GCMs are three-dimensional (latitude, longitude, height) models of the climate
system. They simulate the physical processes that determine large-scale climate. All
attempt to solve a fundamental set of physical equations representing the conservation
of mass, momentum, and energy as well as equations of motion, state, and radiative
transfer; predictive equations for water vapor and heat energy balances at the earth's
surface are also included. The source terms in these equations include numerical
representations of turbulent transfer at the ground-atmosphere boundary, cloud
formation, condensation or rain, and transport of heat by ocean boundary currents.
All these process simulations are ultimately driven by the spatial and temporal
8
-------
distribution of solar radiation. The selection of which processes to model explicitly,
to parameterize, or to eliminate - as well as input and output variables and time
aggregation - vary with the model and model version.
The first widely released generation of GCM's (circa early to mid-1 980's)
estimated conditions over relatively few, very large grid cells and for equilibrium
double C02 conditions. The results of these models have been made widely available
to scientific and policy communities. As a result, their advantages, disadvantages,
caveats and assumptions are fairly well known. Some of the most notable areas of
uncertainty included coupling of the ocean and atmosphere, transient (time dependent)
greenhouse gas changes, explicit regional changes, and climate feedback and
sensitivity.
A second generation of GCM's emerged during the late 1980's and early
1990's which address many of these uncertainties. A summary of their findings are
reported in Gates et al (1992). These findings are updated yet again in IPCC (1996)
to reflect the now widespread use of transient, fully coupled atmosphere-ocean GCMs.
Many of the general findings reported in IPCC (1990) and Gates(1992) remain
unchanged, but the latest generation of GCMs now estimate a slightly lower mean
global temperature change; +1.5°Cto + 4.5°C (circa. 1990), +1°Cto +3.5°C by the
year 2100(circa. 1996). The 1996 estimate reflects model advances which address
the role of aerosols in the pre-1990 radiative forcing history, a revised understanding
of the carbon cycle and other important, but previously missing or overly simplified
atmospheric processes and attributes (Kattenberg, et al, 1996).
9
-------
In spite of notable model advances, confidence in regional change projections
remains low. The key factors that affect the regional performance of global coupled
models are their horizontal resolution and their physical parameterizations. Sources of
significant overall model uncertainty continue to be our inability to accurately
characterize cloud and radiation effects on the hydrologic balance over land and heat
flux processes at the ocean surface (Gates, et al., 1 996),
With the large number of GCM projections available to applications research
scientists (more than 20 in 1993), how does one identify "which model is the best?"
A common evaluation approach is to compare model results to historical climate
records, but this must be done carefully. Comparative performance in simulating
current and past climate varies from region to region and season to season. One
initiative, designed specifically to facilitate such evaluations, has been organized by
the World Meteorological Organization's (WMO) World Climate Research Programme
(WCRP) and is called the Atmospheric Model Intercomparison Project (AMIP). AMIP
evaluates the ability of atmospheric GCMs to simulate the global climate of the decade
1979-1988 (Gates, 1992). Twenty-five diagnostic subprojects have been examining
various aspects of the simulations. Reports have been prepared that compare, for
example, radiation, cloudiness, soil moisture, humidity, precipitation and vegetation
parameterization performance. Readers should contact the WMO-WCRP for a
complete list of AMIP reports and journal articles.
10
-------
5. The GCM Output
The SGCP assessment required the development of a set of double C02
application scenarios that parallel in space and time resolution, those for the historical
period. To assemble a set of climate change time series representing a range of
possibilities (rather than a single outcome), monthly equilibrium GCM summary output
arrays were obtained from the National Center for Atmospheric Research (NCAR)
archives for the NASA Goddard Institute for Space Studies (GISS; Hansen et a!.,
1983), the Geophysical Fluid Dynamics Laboratory (GFDL; Manabe and Wetherald,
1987), the Oregon State University (OSU; Schlesinger and Zhao, 1989) and the United
Kingdom Meteorological Office (UKMO; Wilson and Mitchell, 1987) GCMs. The
general characteristics of these second generation models are presented in Table 3.
Baseline (1xC02) and double C02 (2xC02) temperature, precipitation, humidity
and solar radiation fields produced by these models were retrieved. Unfortunately, the
temporal and spatial scales of these fields were inadequate for SGCP applications
(Cooter et al., 1993; Dale and Rauscher, 1994) and so, scenario generation techniques
were applied to provide more useful climate change information.
6. Change Scenarios for the South and Southeastern U.S.
The relative certainty expressed in Tables 1 and 2 all but disappears when we
attempt to explore the implications of such large scale findings for smaller regions
such as the Southern U.S. and scenario development techniques must be employed.
Robinson and Finkelstein (1991) define a scenario as one possible set of future climate
11
-------
conditions. A scenario should be internally consistent, developed using sound
scientific principles, but have no specific probability of occurrence attached.
Giorgi and Mearns (1991) describes a range of possible scenario generation
techniques that have been used in the past. A semi-empirical approach was adopted
for the SGCP scenario data base. Semi-empirical approaches attempt to translate
large-scale, GCM information into local statistics by using empirically derived
relationships between large-scale and local surface variables (Giorgi and Mearns,
1991). The basic assumption underlying this approach is that the inaccuracy from the
coarse GCM resolution is reduced when GCM-produced baseline and 2xC02 field
differences are applied to higher-resolution observed data. The classic semi-empirical
methodology applied here lacks sophistication, but has provided useful results (e.g.
Smith and Tirpack, 1990; Cooter, 1990). The application of this approach to each
historical time series, along with appropriate assumptions and caveats are described
below. A more detailed discussion is found in Cooter, et al. (submitted). One
example of more sophisticated statistical "down-scaling" scenario techniques is found
in von Storch et al. (1993) and Gyalistras et al. (1994).
6.1 TEMPERATURE
Two temperature scenario options considered were equal day and nighttime
warming and differential warming. For equal warming, SGCP 2xC02 temperature time
series were constructed by adding the difference between the 1xC02 and 2xC02 GCM
mean monthly temperatures to daily maximum and minimum historical temperatures.
12
-------
Equal warming scenarios were applied by Smith and Tirpack (1990) and Cooter
(1990), Differential warming scenarios do not assume that day and nighttime
temperatures warm by the same amount. To construct such scenarios requires that
the GCM include diurnal temperature projections. Table 3 indicates that only the GISS
and UKMO models include diurnal cycles and so a full suite of GCM-directed
differential warming scenarios could not be developed.
A second diurnal temperature scenario, based on the analysis of historical time
series, was proposed in Karl, et al. (1993). They determined that minimum
temperature (nighttime) warming has been approximately three times that observed
during the day (maximum temperature) and that the diurnal temperature range is
approximately equal to the mean monthly temperature change. Implementation of this
scenario was not possible for the SGCP assessment, but preliminary applications have
been explored and results for maize are reported in Dhakhwa et al. (In press).
The equal warming scenario was used for the SGCP. Temporally, the mean
intra-annual climate change reflected in each daily time series is dictated entirely by
changes between 1xC02 and 2xC02 GCM output fields. Physically reasonable spatial
detail is added by maintaining the original empirical relationships of the historical time
series. This approach assumes that spatial relationships characterized by the set of
historical time series does not change under double C02 conditions.
Several limitations are imposed by this scenario approach. Adding GCM
differences equally to temperatures on all days transposes the overall distribution of
temperature (maximum and minimum) upward (warmer). Within a month, for a given
13
-------
grid cell, diurnal temperature ranges and standard deviations remain unchanged.
Climate change differences estimated for some months are quite large and,
occasionally, result in daily temperatures above historical climate bounds. Some
biological models, developed exclusively from historical environmental conditions, may
produce unreasonable results using these scenarios and modifications or upper
response limits may need to be added before being used in climate change assessment
applications. Historical and GCM-directed region-wide scenarios are summarized in
Tables 4 through 6, Within-region SGCP scenarios are illustrated in Figures A-1.1.0
through A-2.2.4.
6.2 PRECIPITATION
For the precipitation scenarios, each daily rainfall total in the historical record
was multiplied by the ratio of mean monthly 1xC02 to 2xC02 GCM results. The mean
intra-annual climate change reflected in each resulting daily time series is dictated
entirely by changes between 1xC02 and 2xC02 GCM output fields. Spatial
relationships contained within the historical data base are maintained, thus increasing
spatial detail in a physically plausible fashion. This approach assumes that spatial
relationships characterized by the set of historical time series does not change under
double C02 conditions. The principle limitation of this scenario approach is that no
change in rainfall frequency or auto-correlation is assumed. Further, by multiplying
each event by a constant GCM ratio, the variance of the intensity process is altered
by the constant ratio, squared (Mearns, et al., 1996). Historical and GCM-directed
14
-------
region-wide precipitation scenarios are summarized in Tables 4 through 6. Within-
region SGCP scenarios are illustrated in Appendix Figures A-3.1.0 through A-3,2.4.
6.3 SOLAR RADIATION
GCMs are, ultimately, driven by spatial and temporal distributions of solar
radiation. Several studies have been conducted which compare radiation algorithms
across GCMs. For instance, Cess, et al.(1 990 and 1992) compared surface energy
fluxes and feedbacks and Fouquart et al., 1991 compared shortwave radiation fluxes
under an international initiative called the Intercomparison of Radiation Codes in
Climate Models (ICRCCM) project which predates AMIP. Each analysis indicates a
wide range of response (sensitivity) to carefully specified environmental changes.
Fouquart et al., 1991 conclude that "...many radiation algorithms could have
inherently unknown errors that may significantly affect the conclusions of the studies
in which they are used. This is true for climate modeling and weather forecasting
studies as well as for other applications, such as inferences from satellite
observations." Algorithmic improvements have been made and our understanding of
radiation processes improved since these analyses, but Cess et al. (1992) and
Fouquart (1991) reflect radiation algorithm performance for the generation of GCMs
used in the SGCP.
A few comparisons of GCM results with field observations have been made.
One of the earliest is Mearns, et all. (1989). It is reported that for the Great Lakes,
Great Plains and southeastern and Northwestern U.S., GCMs were able to simulate
15
-------
the annual solar cycle, but did not simulate the magnitude of solar (incident) radiation
well. A second study, Brazel, et al. (1993) reported similar results for the
southwestern U.S. Although GCM projections of incident solar radiation should not
be relied upon for applications research, these studies indicate that changes in average
radiation patterns might be meaningful.
Two radiation scenario approaches were considered; GCM-directed and SGCP-
modeled. GCM-directed scenarios were generated by modifying the baseline SGCP
radiation time series by the ratio of GCM baseline to 2xC02 projections. The SGCP
scenario approach assumes that historical statistical and physical relationships
represented by the model described earlier remain intact under changed climate.
Temperature dependent parameterizations for the model were re-calculated using the
GCM-directed temperature scenarios and new radiation time series were produced.
These two radiation scenarios were then compared and results reported in Cooter et
al. (submitted). We found that mean annual global radiation estimates generated by
the two approaches are not statistically different and so the time series of global
radiation obtained via the SGCP radiation model was accepted into the applications
data base. By doing so, internal consistency across climate variables assigned to an
SGCP grid cell is maintained, consistency between baseline and changed climate
scenarios is maintained, physical consistency between grid cells and among variables
is maintained and GCM internal and physical consistency is recognized and maintained
through the use of GCM-directed temperature scenarios to drive the radiation time
series. Other implications of this scenario choice is explored in greater detail in Cooter
16
-------
et al. (submitted). Historical and SGCP-generated region-wide change scenarios are
summarized in Tables 4 through 6. Within-region SGCP scenarios are illustrated in
Appendix Figures A-4.1.0 through A-4.2.4.
6.4 VAPOR PRESSURE DEFICIT
The suitability of direct GCM estimates of humidity for SGCP applications versus
simulation via equations 2 and 3 was considered. A recent AMIP evaluation of water
vapor (humidity) simulations across 28 GCMs suggests that estimates are within 5-
15% of observed historical values for the SGCP region at large spatial and temporal
analysis scales (4° grid spacing and decadal median annual, seasonal and monthly
estimates) (Gaffen et al., in press). Several sources of estimate bias have been
identified, but firm conclusions could not be made.
In light of this analysis and the need to provide climate change scenarios that
are spatially and temporally consistent with the historical time series, SGCP 2xC02
humidity scenarios were generated using the relationships described by equations 2
and 3. However, humidity changes simulated as being indirectly induced by increased
atmospheric C02 must be interpreted carefully. For instance, the SGCP scenario
approach assumes that daily maximum and minimum temperature are increased
equally. The vapor pressure and saturation vapor pressure, in turn, increase
proportionately (equations 2 and 3). VPD increases (i.e., the air is further from
saturation), but the relative humidity (approximated by the ratio of e to e3) remains
constant and the mixing ratio (the ratio of the mass of water vapor in the air to the
17
-------
mass of dry air with which the water vapor is associated) increases. Therefore,
depending on the measure chosen, humidity under 2xC02 conditions can increase,
decrease or remain unchanged from historical conditions. Direct GCM output indicate
increased mixing ratios in most cases which is in general agreement with the
empirically derived results presented here. Implications of these humidity scenarios
for loblolly pine stand evapotranspiration applications is presented in Cooter et al.
(submitted). Historical and SGCP-generated region-wide scenarios are summarized in
Tables 4 through 6. Within-region SGCP scenarios are illustrated in Appendix Figures
A-5.1.0 through A-5.2.4.
6.5 CRITICAL FINDINGS
region-wide
temperature
• Mean regional temperature increases range from
early growing season +3.7°C (OSU) to -f 6.1°C (UKMO)
late growing season +3.6°C (OSU) to +6.7°C (UKMO)
non-growing season +3.5°C (OSU) to +6.7°C (UKMO)
• Non-growing (November through March) and early (April
through June) growing seasons show decreased spatial
variability. Late growing season temperatures show
increased spatial variability.
precipitation
• Most2 models considered suggest increased growing season
precipitation and decreased non-growing season rainfall
2 In this discussion, "most" is used only when three of the four GCM's
considered in this analysis are in agreement.
18
-------
• Patterns of change within the growing season vary with model
early growing season late growing season
GISS + +
GFDL +
UKMO +
osu - +
global solar radiation
• All models considered indicate slightly increased early
growing season radiation, with most showing decreased
spatial variability.
• Most models suggest increased mean late season net
radiation. Spatial variability decreases in most cases.
vapor pressure deficit
• Vapor pressure deficit increases across all time periods and
change scenarios
• Greatest average vapor pressure deficit change occurs
during late growing season months
within-region
temperature
• Historical early growing season temperature patterns are
maintained in all scenarios. In most cases, magnitudes are
increased 2° to 4° C.
• Most models maintain late season temperature patterns and
values in North Carolina, South Carolina and Tennessee,
with 2° to 4° C increases elsewhere.
• Area of greatest temperature change from historical
conditions are projected for Louisiana, Mississippi and
Alabama.
19
-------
precipitation
• There is no model agreement concerning drying or
intensification of the early growing season moisture corridor
that, historically, extends from the Louisiana coast,
northward.
• Historical late growing season spatial patterns remain
essentially unchanged. GFDL and UKMO scenario rainfall
totals are also unchanged. GISS and OSU models indicate
more moist conditions throughout the region.
global solar radiation
• The models agree that although the pattern of mean
growing season solar radiation will remain unchanged,
values may increase slightly.
vapor pressure deficit
• Most models project early growing season VPD patterns
and values to remain unchanged in the Northern one-third
of the region and VPD increases in the interior of the
region. All models project no VPD changes along the Gulf
Coast.
• Late growing season spatial patterns are maintained.
Magnitudes along the Gulf Coast are maintained by all
models, and most others project uniform deficit increases
on the order of 2 mb.
20
-------
7. Summary
Historical and future climate conditions for the Southern U.S. have been
assembled and presented. While some statements concerning possible global changes
under double C02 conditions can be made with varying levels of certainty, the regional
expression of these changes remains highly uncertain.
The regional results derived for the SGCP suggest agreement regarding warming
in the South, with sensitivity tending towards the upper end of the IPCC range {Table
1). The area containing Louisiana, Mississippi and Georgia could experience the
greatest degree of change within the region. Although projected annual precipitation
increases could result in a more favorable biological environment for the South,
changes in the within-growing season distribution of precipitation, as well as
uncertainty concerning changes in spatial patterns of precipitation, make the ultimate
impact of these changes on biological productivity less certain. Uncertainty in the
GCM parameterization of cloud formation and cloud radiation feedback make
statements regarding changes in humidity and solar radiation highly speculative.
21
-------
REFERENCES
Bannister P. (1976) Introduction to Physiological Plant Ecology. New York: John
Wiley & Sons, pp36-61.
Brazel AJ, McCabe GJ JR, Verville HJ (1993) Incident solar radiation simulated by
general circulation models for the southwestern United States. Clim. Rsch.
2(3): 177-181.
Cess RD and others (1990) Intercomparison and interpretation of climate feedback
processes in 19 atmospheric general circulation models. J. Geophysical Rsch.
95 (D10): 16601-16615.
Cess RD and others (1992) Intercomparison and interpretation of surface energy
fluxes in atmospheric general circulation models. J. Geophysical Rsch. 97
(D4): 371 1-3724.
Cooter EJ (1990) The impact of climate change on continuous corn production in
the Southern U.S.A. Clim. Change 16: 53-82.
Cooter EJ, Eder BK, LeDuc SK, Truppi L (1 993) Climate change models and forest
research. J. of Forestry 91:38-43.
Cooter EJ, Dhakhwa GB (1 995) A solar radiation model for use in biological
applications in the South and Southeastern U.S. Ag. For. Meteorol. 78, 31-51.
Cooter EJ, Richman M, Lamb P (submitted) Development of a climate change data
base for biological assessments in the southern U.S. Climatic Change.
Dale VH, Rauscher HM (1994) Assessing impacts of climate change on forests: the
state of biological modeling. Clim. Change 28: 65-90.
Dhakhwa GB, Cooter EJ, LeDuc SK, Campbell CL (in press) Maize growth: Assessing
the effects of global warming and C02 fertilization with crop models. Ag. For.
Meteorol.
Folland CK, Karl TR, Nicholls N, Nyenzi BS, Parker DE, Vinnikov KYa (1992)
Observed climate variability and change. In: Houghton JT, Callander BA,
Varney SK (eds) Climate Change 1992: The supplementary report to the IPCC
scientific assessment. University Press, Cambridge, pp137-170.
Fouquart Y, Bonnel B, Ramaswamy V (1991) Intercomparing shortwave radiation
codes for climate studies. J. Geophysical Rsch. 96 (D5): 8955-8968.
22
-------
Friend AD (submitted) Parameterization of a global daily weather generator for
terrestrial ecosystem and biogeochemical modelling. Ecological Modelling.
Gaffen DJ, Rosen RD, Salstein DA, Boyle JS (in press) Evaluation of tropospheric
water vapor simulations from the atmospheric model intercomparison project.
• J. Clim.
Gates WL, Mitchell JFB, Boer GJ, Cubasch U, Meleshko VP (1992) Climate modeling,
climate prediction and model validation. In: Houghton JT, Callander BA,
Varney SK (eds) Climate change 1992, the supplementary report to the
IPCC scientific assessment. University Press, Cambridge, pp97-1 34.
Gates WL, Henderson-Sellers A, Boer GJ, Folland CK, Kitoh A, McAvaney GJ, Semazzi
F, Smith N, Weaver AJ, Zeng, Q-C (1996) Climate models - evaluation. In:
Climate change 1995, the science of climate change (IPCC WGI), Houghton JT,
Meira Filho LG, Callander BA, Harris N, Kattenberg A., Maskell K (eds.).
Cambridge University Press, Cambridge, UK. pp 233-284.
Giorgi F, Mearns LO (1991) Approaches to the simulation of regional climate
change: A review. Rev. Geophys 29: 191-216.
Gyalistras D., von Storch H, Fischlin A, Beniston M. (1994) Linking GCM-simulated
climatic changes to ecosystem models: case studies of statistical down-scaling
in the Alps. Clim. Rsch. 4(3): 167-189.
Hansen J, Russell G, Rind D, Stone P, Lacis A, Lebedeff S, Ruedy R, Travis L (1983)
Efficient three-dimensional global models for climate studies: Models I
and II. Mon. Weather Rev. 3: 609-662.
Hodges T, French V, LeDuc S (1985) Estimating solar radiation for plant simulation
models. AgRISTARS Technical Report JSC-20239; YM-15-00403, 21pp.
Hodges T, Botner D, Sakamoto C, Hays-Haug J (1987) Using the CERES-Maize model
to estimate production for the U.S. combelt. Ag. For. Meteorol. 40:
293-303.
IPCC (1990) Climate change: The IPCC scientific assessment, Houghton JT,
Jenkins GJ, Ephraums JJ (eds.). Cambridge University Press, Cambridge, U.K.
IPCC (1996) Summary for Policymakers and Technical summary of the Working Group
I Report, Cambridge University Press, Cambridge, UK.
23
-------
Karl TR, Jones PD, Knight RW, Kukla G, Plummer N, Razuvayev V, Gallo KP, Lindseay
J, Charlston RJ and Peterson TC (1993) Asymmetric trends of daily maximum
and minimum temperatures. Bull. Am. Meteor. Soc. 74:1007-1023.
Kattenberg A, Giorgi F, Grassl H, Meehl GA, Mitchell JFB, Stouffer RJ, Tokioka T,
• Weaver AJ Wigley TML (1996) Climate models - projections of future climate.
In: Climate change 1995, the science of climate change (IPCC WGI), Houghton
JT, Meira Filho LG, Callander BA, Harris N, Kattenberg A., Maskell K (eds.).
Cambridge University Press, Cambridge, UK. pp 289-257.
Kimball J, Running SW, Nemani R (1996) An improved method for estimating surface
humidity from daily minimum temperature. Ag. For. Meteorol. (In press).
Koss WJ, Owenby JR, Steurer PM, Ezell DS (1988) Freeze/frost data. United States
Department of Commerce National Oceanic and Atmospheric Administration
National Climatic Center Climatography of the U.S. No. 20, Supplement No. 1.
Manabe S, Wetherald RT (1987) Large-scale changes in soil wetness induced by an
increase in carbon dioxide. J. Atmos. Sci. 44: 1211-1235.
Mearns LO, Schneider SH, Thompson SL, Daniel LR (1989) Analysis of climate
variability in general circulation models: comparisons with observations and
changes in variability in 2xC02 experiments. In: Smith JB, Tirpak DA (eds) The
potential effects of global climate change on the United States: Appendix I -
Variability. U.S. Environmental Protection agency, Office of Policy, Planning,
and Evaluation, Washington, DC, p 1-59.
Mearns LO, Rosenzweig C, Goldberg R (1996) The effect of changes in daily and
interannual climatic variability on CERES-wheat: A sensitivity study. Clim.
Change 32: 257-292.
Mitchell JFB, Manabe S, Meleshko V, Tokioka T (1990) Equilibrium climate change
and its implications for the future. In: Houghton JT, Jenkins GJ and
Ephraums JJ (eds) Climate change, the IPCC scientific assessment.
University Press, Cambridge, pp134-172.
Murray FW (1967) On the computation of saturation vapor pressure. J. of Appl.
Meteorol. 6: 203-204.
NREL (1992) User's Manual, National Solar Radiation Data Base, (1961-1990).
National Renewable Energy Laboratory, Golden, Colorado. 93pp.
24
-------
Richman MB and Montroy DL (1996) Nonlinearities in the signal between el nino/la
nina events and North American precipitation and temperature. In Preprints of
the 13th conference on Probability and Statistics in the Atmospheric Sciences,
February 21-23, 1996, San Francisco, CA, 90-97.
Richman MB, Lamb PJ (1985) Climatic pattern analysis of three- and seven-day
summer rainfall in the central United States: some methodological
considerations and a regionalization. J. Clim. Appl. Meteorol. 24: 1325-
1343.
Robinson PJ, Finkelstein PL (1991) The development of impact-oriented climate
scenarios. Bull. Am. Meteorol. Soc. 72: 481-489.
Running SW, Nemani RR, Hungerford RD (1987) Extrapolation of synoptic
meteorological data in mountainous terrain, and its use for simulation forest
evapotranspiration. Can. J. For. Rsch. 17: 472-483.
Schlesinger ME, Zhao Z-c (1989) Seasonal climate changes induced by doubled C02
as simulated by the OSU Atmospheric GCM/Mixed-layer Ocean Model.
J. Clim. 2: 459-495.
Smith JB, Tirpack DA, eds (1990) The Potential Effects of Global Climate Change on
the United States. Hemisphere Pub. Corp., New York.
von Storch H, Zorita E, Cubasch U (1993) Downscaling of global climate change
estimates to regional scales: an application to Iberian rainfall in wintertime. J.
Clim. 6: 1161-1171.
Wilson CA, Mitchell JFB (1987) A doubled C02 climate sensitivity experiment with
a global climate model including a simple ocean. J. Geophys. Res.
92(D1 1): 13,1315-13343.
25
-------
Table 1. Main equilibrium changes in climate due to doubling C02 deduced from
a suite of GCMs. Items listed were assigned three or more *'s on a five * scale. Five
*'s indicate virtual certainties, one * indicates low confidence {Mitchell, et al., 1990).
Rating
Description
Temperature
*****
the lower atmosphere and Earth's surface warm;
*****
the stratosphere cools;
* * *
near the Earth's surface, the global average warming lies
between +1.5°C and +4.5°C, with a "best guess" of 2.5°C;
* * *
the surface warming at high latitudes is greater than the global
average in winter but smaller than in summer, (In time
dependent simulations with a deep ocean, there is little warming
over the high latitude southern ocean);
* * *
the surface warming and its seasonal variation are least in the
tropics.
Precipitation
* * * *
the global average increases (as does that of evaporation), the
larger the warming, the larger the increase;
* * *
increases at high latitudes throughout the year;
* * *
increases globally by 3 to 15% (as does evaporation);
Soil Moisture
* * *
increases in high latitudes in winter;
Snow and
sea-ice
* * * *
the area of sea-ice and seasonal snow-cover diminish
-------
Table 2. Selected findings of the IPPC concerning observed climate variability and
change (from Folland, et al., 1992).
Temperature
• Continuing research into the nineteenth century ocean temperature
record has not significantly altered our calculation of surface
temperature warming of 0.45 ± 0.1 5°C since the late nineteenth
century.
• A new analysis of radiosonde data confirms that mid-tropospheric
warming has occurred over the past several decades.
• Microwave Sounding Unite (MSU) data provide a more complete
satellite-based global data set for tropospheric and stratospheric
mean temperatures, but the record is still too short for a meaningful
assessment of trends.
Precipitation
• Precipitation variations of practical significance have been
documented in a number of regions on many time and space scales.
Owing to data coverage and inhomogeneity problems, however, we
cannot yet say anything new about global-scale changes.
• Evidence continues to support an increase in water vapor in the
tropical lower troposphere since the mid-1 970's, though the
magnitude is uncertain. However, we cannot say whether the
changes are larger than natural variability.
Snow and sea-ice
• Northern Hemisphere snow cover continues its tendency to be less
extensive than that observed during the 1970s when reliable
satellite observations began.
• No systematic change can be identified in global or hemispheric
sea-ice cover since 1973 when satellite measurements began.
-------
Table 3. Attributes of four General Circulation Models used for SGCP
scenario development.
Model
GFDL
GISS
OSU
UKMO
Date of output
generation
1988
(Q-flux)
1982
(Q-flux)
1984-1985
1986
numerical solution
technique
Spectral
(R-15)
Finite
difference
Finite
difference
Finite
difference
horizontal resolution
(Lat x Long)
4.5° x 7.5°
7.8° x10.0°
4.0° x 5.0°
5.0° x 7.5°
vertical resolution
9
9
2
11
initial C02
concentration
300
315
326
320
solar constant
(W/m2)
1467
1367
1354
1395
diurnal cycle
no
yes
no
yes
surface
characterization
uniform
fractional
uniform
uniform
land cover
6
vegetation
types
8
vegetation
types
snow, sea
ice, 6 veg.
types
sea, sea-ice,
snow, snow-
free land
convective
parameterization
moist
adiabatic
penetrating
convection
penetrating
convection
penetrating
convection
cloud cover
on/off,
RH > 99%
on/off,
RH > 100%
on/off,
RH > 100%
fractional
Ocean
parameterization
60-meter
slab,
prescribed
65-meter
slab,
prescribed
60-meter
slab,
prescribed
50-meter
slab,
prescribed
-------
Table 4. Early growing season (April, May, June) historical and climate change conditions across the
Southern Global Change Program's study region.
maximum
temperature (C°)
minimum
temperature (C°)
precipitation (mm)
vapor pressure
deficit (mb)
solar radiation
(W/m2)
mean
s.d.
mean
s.d.
mean
s.d.
mean
s.d.
mean
s.d
History
27.4
1.8
14.4
2.5
329.7
41.9
8.33
.9
239
10.1
GISS
31.5
1.7
18.4
2.4
372.3
54.8
10.4
1.1
244
9.1
GFDL
31.2
1.9
18.2
2.6
386.1
84.2
10.3
1.2
244
11.9
OSU
31.1
1.7
18.0
2.5
302.8
38.7
10.2
1.1
240
9.3
UKMO
33.5
1.7
20.4
2.2
350.7
68.5
11.6
1.4
245
9.7
-------
Table 5. Late growing season (July, August, September) historical and climate change conditions
across the Southern Global Change Program's study region.
maximum
temperature (C°)
minimum
temperature (C°)
precipitation (mm)
vapor pressure
deficit (mb)
solar radiation
(W/m2)
mean
s.d.
mean
s.d.
mean
s.d.
mean
s.d.
mean
s.d
History
31.6
1.7
19.2
2.1
327.7
91.3
10.2
1.3
230
15.0
GISS
35.8
2.0
23.3
2.1
379.3
84.1
12.7
1.9
234
12.7
GFDL
35.9
1.7
23.4
1.9
308.5
114.8
12.8
1.8
234
15.4
OSU
35.2
1.9
22.7
2.3
398.6
111.8
12.4
1.7
231
13.4
UKMO
38.3
2.2
25.8
2.2
318.5
109.4
14.5
2.2
229
12.6
-------
Table 6. Non-growing season (January, February, March, October, November, December) historical
and climate change conditions across the Southern Global Change Program's study region.
maximum
temperature (C°)
minimum
temperature (C°)
precipitation (mm)
vapor pressure
deficit (mb)
solar radiation
(W/m2)
mean
s.d.
mean
s.d.
mean
s.d.
mean
s.d.
mean
s.d
History
17.0
2.9
4.1
3.2
581.8
143.6
4.8
.8
138
10.5
GISS
21.5
2.7
8.7
2.8
518.8
130.8
6.3
.9
137
10.1
GFDL
21.1
2.7
8.0
3.0
642.9
164.7
6.1
.9
136
10.7
OSU
20.5
2.7
7.5
3.0
540.3
128.7
5.8
.9
138
9.7
UKMO
23.7
2.6
10.7
2.8
550.8
141.8
7.0
1.1
138
10.1
-------
LIST OF FIGURES
Figure 1. Map of the Southern Global Change Program study region.
-------
75 W
WESJ
VIRGIN
, VIRGINIA
MISSOURI
KENTUCKY
I
I
1 i
temcssbe
NORTH
C^ROLNA
( ""
4--
SOUTH \
CAROLINA
OKLAHOMA
ARKANSAS
GEORGIA
MISSISSIPPI
ALABAMA
RORD
LOUISIANA
25 N
lotrvr
"W
-------
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
APPENDIX
This appendix contains 1°latitude x longitude maps of historical and double C02
scenarios for the Southern Global Change Program study area. Plotting categories
have been selected to facilitate comparison across models for the specified time
period. They are not necessarily appropriate for comparison across time periods. In
most cases, the number of plotting categories and interior plotting category size for
a climate variable has been maintained, but category limits may be shifted between
time periods. Plotting categories rather than isolines are presented to facilitate
analysis of changes in the magnitude and spatial distribution of climate conditions
throughout the region. The uncertainty inherent in making regional statements from
global-scale model output discourages reliance on absolute values of specific climate
variables.
A-1
-------
LIST OF APPENDIX FIGURES
A-1.1.0 Historical early growing season maximum temperature.
A-1.1.1 GISS early growing season maximum temperature.
A-1.1.2 GFDL early growing season maximum temperature.
A-1.1.3 OSU early growing season maximum temperature.
A-1.1.4 UKMO early growing season maximum temperature.
A-1.2.0 Historical late growing season maximum temperature.
A-1.2.1 GISS late growing season maximum temperature.
A-1.2.2 GFDL late growing season maximum temperature.
A-1.2.3 OSU late growing season maximum temperature.
A-1.2.4 UKMO late growing season maximum temperature.
A-2.1.0 Historical early growing season minimum temperature.
A-2.1.1 GISS early growing season minimum temperature.
A-2.1.2 GFDL early growing season minimum temperature.
A-2.1.3 OSU early growing season minimum temperature.
A-2.1.4 UKMO early growing season minimum temperature.
A-2.2.0 Historical late growing season minimum temperature.
A-2.2.1 GISS late growing season minimum temperature.
A-2.2.2 GFDL late growing season minimum temperature.
A-2.2.3 OSU late growing season minimum temperature.
A-2
-------
A-2.2.4
UKMO late growing season minimum temperature.
A-3.1.0 Historical early growing season precipitation.
A-3.1.1 GISS early growing season precipitation.
A-3.1.2 GFDL early growing season precipitation.
A-3.1.3 OSU early growing season precipitation.
A-3.1.4 UKMO early growing season precipitation.
A-3.2.0 Historical late growing season precipitation.
A-3.2.1 GISS late growing season precipitation.
A-3.2.2 GFDL late growing season precipitation.
A-3.2.3 OSU late growing season precipitation.
A-3.2.4 UKMO late growing season precipitation.
A-4.1.0 Historical early growing season solar radiation.
A-4.1.1 GISS early growing season solar radiation.
A-4.1.2 GFDL early growing season solar radiation.
A-4.1.3 OSU early growing season solar radiation.
A-4.1.4 UKMO early growing season solar radiation.
A-4.2.0 Historical late growing season solar radiation.
A-4.2.1 GISS late growing season solar radiation.
A-4.2.2 GFDL late growing season solar radiation.
A-3
-------
A-4.2.3
A-4.2.4
OSU late growing season solar radiation.
UKMO late growing season solar radiation.
A-5.1.0 Historical early growing season vapor pressure deficit.
A-5.1.1 GISS early growing season vapor pressure deficit.
A-5.1.2 GFDL early growing season vapor pressure deficit.
A-5.1.3 OSU early growing season vapor pressure deficit.
A-5.1.4 UKMO early growing season vapor pressure deficit.
A-5.2.0 Historical late growing season vapor pressure deficit.
A-5.2.1 GISS late growing season vapor pressure deficit.
A-5.2.2 GFDL late growing season vapor pressure deficit.
A-5.2.3 OSU late growing season vapor pressure deficit.
A-5.2.4 UKMO late growing season vapor pressure deficit.
A-4
-------
MAXIMUM TEMHiRATURK (O
ESI GREATER THAN 34.9
B 350 TO 34.9
[~ 31.0 TO 319
19 29.0 TO 30.9
H LESS THAN 29.0
fl-l.l-0
-------
*4
i • ¦ i---',
L.. ; •.
MAXIMUM TEMPERATURE (C)
¦ GREATER THAN 34.9
ES 33.0 TO 349
~ 31 0 TO 32.9
¦i 29.0 TO 30.9
LESS THAN 29.0
.v->
-------
MAXJMUM THMPI:RATI JRI; (C}
grkaiik than u.g
® 33.0 TO 34.<;
D 31.0 TO 3Z9
® 29.0 TO 30.9
® I.ESS THAN 29.0
-------
MAXIMUM TtMPtRA I URJi
® (•Ri-ATER TIIAN 34.9
Si 33.0 TO 34.9
C3 3i.O TO 329
" 29-0 TO 30.9
® UiSS THAN 29.0
4-1.1.3
-------
MAXIMUM TF.MPKRATIJRF. (C)
m
GRI1ATER TlIAN 34.9
ii
33 0 TO 34.9
~
31.0 TO 329
¦
29.0 TO 30.9
BP
LESS THAN 29.0
A -/./•
-------
MAXIMUM TEMVtiRATlIRK (CI
99 GREATER THAN 34.9
EH 33.0 TO 349
~ 31.0 TO 32.9
B 29.0 TO 30.9
IH LHSS THAN 29.0
-------
MAXIMUM TEMPERATURE (O
¦B GREATER THAN 38.9
ES 37.0 TO 38.9
d 35.0 TO 36.9
H 33.0 TO 34.9
80 LESS THAN 33.0
A I O I
-------
Maximum temperature (O
US GREATER than 38.9
37.0 TO 38.9
O 35.0 TO 36.9
® 33.0 TO 34.9
® 1.1 iss THAN 33 0
-------
MAXIMUM TF.MPERATIJRK (O
N GRliAH'R THAN 38.9
ED 37.0 TO 38.9
~ 35.0 TO 36.9
¦ 33.0 TO 34.9
HI l.KSS THAN 33.0
/4-I.2-3
-------
S®lSPIIfs
*"{¦
S:
I
maximum temperature (r>
¦ GREATER THAN 38 9
m 37.0 TO 38.9
O 35.0 TO 36.9
® 33.0 TO U.9
M LESS THAN 33.0
1 I
-------
1 fi 1 i
'XsaM.
;.-u-es
rasl!
e#Sltl§ili
!®1C'3&^!4m®
!SP=?ft\
Ifelilifi
M
are
•' Sal# K^SKij
i.>
-------
MINIMUM TKMPBRATURi; (CI
13 GREATER THAN 21.9
E3 20.0 10 21.9
I I 18 0 TO 19.9
SI 16.0 TO 17.9
EH l.iaS THAN 16.0
v
fi-Z-l-l
-------
MINIMUM TEMPERATURE (CI
53 GRfiATER THAN 21.9
lS! 20.0 TO 21.9
~ IR.O TO IV)
®B1 16.0 TO 17.D
¦ I.ESS THAN lft.0
fl- I.h £
-------
MINIMUM TEMPKRAT11RK (O
SKf (iRRATF.R THAN 21.9
P'l 20.0 TO 21.9
I I IR0 TO 19.9
S39I 160 TO 17.9
jlj l.liSS THAN 16.0
sag
gfecissi^'i' ¦ 11 11 _"tT>
ft-2.1.5
-------
MINIMUM TEMPJiRATURK (C)
Hfl GRKATER THAN 219
fTrvi 20.0 TO 21.9
[ I 18.0 TO 19.9
SB 16.0 TO 17.9
fH LESS THAN 16.0
-------
sfigyjj
W?»li
minimum temperature (C)
[EH GREATER THAN 26.9
RH 75.0 TO 20.9
1 ] 7.3.0 TO 2-1.9
H! 21.0 TO 9
ffl LESS THAN 21.0
ft-z. 2.0
-------
MINIMUM TEMPliRA'JURI: (O
G3 GREATER THAN 26.9
E2I 25X1 TO 26.9
~ 23.0 TO 24.9
889 21.0 TO 22.9
H I .ESS THAN 21.0
Vr ,
igKSsSSSgS
ipfs
fl-2.2.1
-------
MINIMUM TCMPERATURt (C)
I® CiKRATKR THAN 26.9
M 25.0 TO 26.9
C J 23.0 TO 24.9
R3 2J.0 TO 22.9
® I-liSS THAN 21.0
- ? . 2 ¦ ?„
-------
MINIMUM TF.MPKRATIIRE (O
® GRIiATKK THAN 20.9
IP 3.0 TO 76.M
~ 23.0 TO 24.»
633 21.0 TO 22.9
IS I.F.SS THAN 21.0
-------
MINIMUM TEMPKRATURK (O
® ORtiAlVM THAN 36,9
® 25.0 'I'D I6,i)
C-1 23.0 TO 24 g
BS 21.0 TO 22.0
ft .. z d
-------
PRF-CirrrATION (mm)
ES GREATER THAN 449.9
91 400.0 TO 449,9
[~ 350.0 TO 399.9
E3 300.0 TO MM
S3 LESS THAN 300.0
ft
- 2,1-0
-------
PRECIPITATION (mm)
Sffl GREATHR THAN 4«.9
¦ 100.0 TO 44M
I 1 350.0 T0 3W.9
HI 300.0 T0 3«.9
Hi LESS THAN 300.0
{] -3.1. I
-------
PRECIPITATION (mm)
@i GRI'AiliR THAN 449.9
¦ 400.0 TO 449.9
f~l 350.0 TO 3W.9
FT] 300.0 TO 349.0
B LESS THAN 300.0
A-3, !¦ a
-------
PRECIPITATION (mm)
m GREATER THAN 449.9
IS 100,0 TO 449.9
LM 350,0 TO 399.9
El 300.0 TO 349.9
ES LESS THAN 300.0
-------
PRECIPITATION (mm)
Si (WKATGR TIUN 449,9
H «X>.<> TO 449.9
O 350,0 TO 399.9
ffl 300,0 TO 349.9
Hi I.ESS THAN 300.0
A- 3.
-------
PRECIPITATION (mm)
11 GRliATKK THAN 349.9
B 300.0 TO34U.0
~ 250.0 TOW
El 200.0 TO 249 9
IS l.KSS THAN 200.0
3. 2.0
-------
PRKCIPITATION (mm)
E*) GREATER T1IAN 519.9
H 3ttl0TO3#.9
~ 250.0 TO 299.9
HI 200,0 TO 249.9
B31 1,HSS THAN 200.0
A-2>. 1.1
-------
PRKCII'ITATON (nun)
[SS] GREATER than 34').9
IS 300.0 to ;w.9
ni 250.0 10 299.9
EI3 200 0 1 0 249.9
GSi) LIXS THAN 200.0
Q' 3. 2. Z
-------
MU-CIPITATION (mml
H CiRJKATUR THAN 349.9
¦ 300.0 TO 349.9
~ 250.0 TO299.9
E3 300.0 TO 249.9
9 I.ISS THAN 200.0
(4 •• 'J. 2. 3
-------
lUKCDTTATION (nun)
fM GREATER IIIAN 349.0
BSI ;oo.o to:>4o o
.'>0.0 TO 299.9
l.„j 200.0 TO 249.9
13 LESS THAN 200,0
U •
, - ¦i""' x
ft • 3, 2. '/
-------
SOLAR RADIATION (w/[e2>
OREA'll'lR THAN 251.9
M 245.0 TO 2S4.0
1 I 235.0 TO 744 9
Kg 225.0 TO 234.9
Mi LESS THAN 225.0
4 I O
-------
SOLAR RADIATION (w/m2>
OREATKR THAN 254.9
© 245.0 TO 254.9
L""l 235.0 TO 244. •>
81 225.0 TO 234.9
53 LESS THAN 225.0
A-M.l.l
-------
SOi-AR RADIATION l»-5.0 TO WM
H >25,0 TO 23 1.9
@3 LESS THAN 225.0
jf) - M, l.2~
-------
SOIAR RADIATION 5.0 TO 2H.9
¦I 225.0 TO 234.9
SI1 I.F.SS THAN 225.0
i I '
-------
SOLAR RADIATION (w/m2>
HI ORKATF.R THAN 254.9
B! 245.0 TO 254.9
I I 235.0 TO 244.9
BB 225.0 TO 114.9
BS LESS THAN 225.0
'3 -h.\4
-------
SOUR KADIATION (w/m2)
E2 (iRRATI-R THAN 2-H
fel JJ.S.0 TO 244,0
1 I 225.0 TO IU.9
HB .'I i (i TO 224.9
@8 1-I SS THAN 214.0
ft -H.2.0
-------
U.•
V'.'..'.
SOIAR RADIATION (w/m2)
(13 GRILATHR THAN 244 0
IS 235.0 TO 244.9
IU 225.0 TO 234.9
Gffl 214.0 TO 224.9
(Ml LESS THAN 214.0
-------
SOLAR RADIA TION (w/m2)
US GREATER THAN 2M.9
[33 235.0 TO 244.9
LI) 225.0 TO 2JM
@8 214.0 TO 2249
El LESS THAN 214.0
*4-2.Z
-------
SOLAR RADIATION (w/m2>
H GREATER THAN 2449
H 235.0 TO 244.0
O 225.0 TO 234.9
3£ 214.0 TO 224.9
1! l.ESS THAN 214.0
-------
SOl-AR RADIATION (w/m2)
Kffl GRI1A I1..R THAN 244.9
El 235.0 TO .'44.9
mo TO 234.9
BS3 214.0 TO 224.9
O I.ESS THAN 214.0
-------
VAPOR PRESSURE DEFICIT (mW
¦ GREATER THAN 12.4
II 11.5 TO 1X4
1 I 10.5 TO 11.4
¦ 9.5 TO 10.4
¦ LESS THAN 9.5
-------
-.\N\ *>*'"*
6^1
sn?t teiyift«it m
«•
VAPOR PRESSURE DEFICIT (mh)
¦I GREATER THAN 12.4
¦ 1I.S TO 124
( J 103 TO 11.4
¦I 9.5 TO 10.4
¦ LESS THAN 9.5
-------
£xv.vn
VAPOR PRESSURE DF-FJOT fob)
¦ GREATER THAN 12.4
H 11.5 TO 114
~ 10.5 TO II.*
Bi M TO 10.4
H LESS THAN 9.5
-------
VAPOR PRESSURE DEMCIT (rob)
¦i
GREATER THAN 1X4
m
11.5 TO 12.4
~
10.5 TO 11.4
¦
9.5 TO 10.4
¦
LESS THAN 9.5
-------
vapor pressure mncn (mb)
¦ GREATER THAN 114
63 11.510 12.4
O 10.5 TO 11.4
¦ 9.5 TO 10.4
Wk LESS THAN 9.5
m
-------
VAPOR PRESSURE DEFICIT (mb)
¦ GREATER THAN 15.9
H3 14.0 TO 15.9
LJ 12.0 TO 13.9
¦ 10.0 TO 11.9
¦ LESS THAN 10.0
-------
H
VAPOR PRESSURE DEFICIT (mb)
¦ GREATER THAN 15.9
S! 14.0 TO 15.9
~ 12.0 TO 13.9
¦ 10.0 TO 11.9
¦ LESS THAN 10.0
-------
C§&SSK
¦ GREATER THAN 15.9
81 14,0 TO 15.9
~ 12.0 TO 13.9
¦ 10.0 TO 11.9
¦ LESS THAN 10.0
-------
VAPOR PRESSURE DEI ICrr (mb)
" (iKEATER THAN 15.9
SS 140 TO 15.9
~ 12.0 TO 13.9
Hi 10.0 TO 11.9
¦ LESS THAN 10.0
-------
VAPOR PRESSURE 0EFJCTT tab)
Hi GREATER THAN 15.9
B 14.0 TO 15.9
Q 12.0 TO 13.9
B 10.0 TO H.9
® less THAN 10.0
-------
TECHNICAL REPORT DATA
1. REPORT NO.
EPA/600/A-97/049
2.
4. TITLE AND SUBTITLE
General Circulation Model Scenarios for the Southern United States
3. R
5.REPORT DATE
6.PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
COPTER, Ellen J.
8.PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Same as Block 12
10.PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
National Exposure Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13.TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/9
15. SUPPLEMENTARY NOTES
16. ABSTRACT
/
^This chapter provides a climatological summary arid,perspective against which specific research resuits presented in later
chapters can be viewed. A regional data base containing historical daily climate time series and climate change scenarios for the
Southeastern United States was developed for the U.S. Forest Service Southern Global Change Program (SGCP). Daily historical
values of maximum temperature, minimum temperature and precipitation and empirically derived estimates of vapor pressure
deficit and solar radiation across a uniform 1^ latitude x ^longitude grid were obtained. Climate change scenarios of
temperature, precipitation, vapor pressure deficit and solar radiation were generated using semi-empirical techniques which
combined historical time series and simulation field summaries from GISS, GFDL, OSU and UKMO General Circulation Model
(GCM) experiments. An internally consistent Platitude x 1° longitude climate change scenario data base was produced in which
vapor pressure deficit and solar radiation conditions were driven by the GCM temperature projections, but were not constrained to
agree with GCM calculated radiation and humidity fields. Map summaries of these historical and climate change conditions are
presented. ^
17.
v.
/
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/ OPEN ENDED TERMS
c.COSATI
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
EPA-2220
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21.NO. OF PAGES
20. SECURITY CLASS (This Page)
UNCLASSIFIED
22. PRICE
------- |