southeast mega-region (chapter 5) from the foundation ... · southeastern united states physical...

137

CHAPTER 5

POTENTIAL CONSEQUENCES OF CLIMATEVARIABILITY AND CHANGE FOR THESOUTHEASTERN UNITED STATESVirginia Burkett1,2, Ronald Ritschard3, Steven McNulty4, J. J. O’Brien5, Robert Abt6,James Jones7, Upton Hatch8, Brian Murray9, Shrikant Jagtap7, and Jim Cruise3

Contents of this Chapter

Chapter Summary

Physical Setting

Socioeconomic Context

Ecological Context

Climate Variability and Change

Key Issues

Weather-related Stresses on Human Populations

Agricultural Crop Yields and Economic Impacts

Forest Productivity Shifts

Water Quality Stresses

Threats to Coastal Areas

Additional Issues

Climate Model Limitations

Water Resources

Impacts on Coastal Ecosystems and Services

Health Issues Related to Water Quality

Socioeconomic and Insurance Issues

Urban Issues

Crucial Unknowns and Research Needs

Literature Cited

Acknowledgments

1US Geological Survey; 2Coordinating author for the National Assessment Synthesis Team; 3University of Alabama inHuntsville; 4USDA Forest Service; 5Florida State University; 6North Carolina State University; 7University of Florida;8Auburn University; 9Research Triangle Institute

Climate of the Past Century

• Southeastern temperature trends varied betweendecades over the past 100 years,with a warmperiod during the 1920s-1940s followed by adownward trend through the 1960s. Since the1970s,temperatures have been increasing,with1990’s temperatures reaching peaks as high asthose of the 1920s-1940s.

• Annual rainfall trends show very strong increasesof 20-30% or more over the past 100 years acrossMississippi,Arkansas,South Carolina,Tennessee,Alabama, and parts of Louisiana,with mixedchanges across most of the remaining area. Thepercentage of the Southeast landscape experi-encing severe wetness increased approximately10% between 1910 and 1997.

Climate of the ComingCentury

• Climate model simulations provide plausible sce-narios for both temperature and precipitationover the 21 st century in the Southeast. Both ofthe principal climate models used in the NationalAssessment suggest warming in the Southeast bythe 2090s,but at different rates.

• The Canadian model scenario shows theSoutheast experiencing a high degree of warm-ing,which translates into lower soil moisture ashigher temperatures increase evaporation. TheHadley model scenario simulates less warmingand a significant increase in precipitation by2100 (about 20% on average,but with some dif-ferences within the region).

• Both climate model simulations indicate that theheat index (a measure of comfort based on tem-perature and humidity) will increase more in theSoutheast than in other regions. Heat index inthe Southeast is projected to rise by as much as8 to 15ºF (4 to 8ºC) in the Hadley model and byover 15ºF (8ºC) across the entire region in theCanadian model simulation by 2100.

CHAPTER SUMMARYRegional Context

The Southeast “sunbelt”is a rapidly growing regionwith a population increase of 32% between 1970and 1990. Much of this growth occurred in coastalcounties,which are projected to grow another 41%between 2000 and 2025. The number of farms inthe region decreased 80% between 1930 and 1997as the urban population expanded,but theSoutheast still produces roughly one quarter of USagricultural crops. The Southeast has becomeAmerica’s “woodbasket,”producing about half ofAmerica’s timber supplies. The region also producesa large portion of the nation’s fish,poultry, tobacco,oil,coal,and natural gas. Prior to European settle-ment,the landscape was primarily forests, grass-lands,and wetlands. Most of the native forests wereconverted to managed forests and agricultural landsby 1920. Although much of the landscape has beenaltered,a wide range of ecosystem types exists andoverall species diversity is high.

138

Potential Consequences of Climate Variability and Change

Key Findings

• The increase in the southeastern summer heatindex simulated by both the Hadley andCanadian climate models would likely af fecthuman activity and possibly demographics in theSoutheast during the 21st century.

• Agriculture could possibly benefit from increasedCO2 and modest warming (up to 3 to 4ºF, or 2ºC)as long as rainfall does not decline,but there aredifferences in individual crop responses.Management adaptations could possibly offsetpotential losses in individual crop productivitydue to increased evapotranspiration.

• Biological productivity of pine and hardwoodforests will likely move northward as tempera-tures increase across the eastern US. Hardwoodsare more likely to benefit from increases in CO2and modest increases in temperature than pines.Physiological forest productivity and ecosystemmodels suggest that,without management adap-tations,pine productivity is likely to increase by11% by 2040 and 8% by 2100 across theSoutheast compared to 1990 productivity. Thesemodels suggest that hardwood forest productivitywill likely increase across the region, by 25% by2090 compared to 1990 regional hardwood pro-ductivity.

• Under the Hadley model scenario,the region’sland use change in the next century is likely to

be dominated by non-climate factors such ascommodity prices and demographic forces.Urbanization will likely continue to convert for-est and agricultural land,while continued move-ment of land from agriculture to forest is alsoexpected. Under more extreme climate scenar-ios,land reallocation could possibly be more dra-matic as the productivity of land-based activitiessuch as forestry and agriculture is more pro-foundly influenced by climate.

• During the 21st century, the IPCC projects thatsea-level rise will likely accelerate 2- to 5 -foldcompared to the global average rate during the1900s (which was 4-8 inches). This would verylikely have dramatic effects on population cen-ters,infrastructure,and natural ecosystems in thelow-lying Gulf and South Atlantic coastal zone.

• Water and air quality are concerns given thechanges in temperature and precipitation thatare simulated by climate models.

• Changes in minimum temperature, rainfall,andCO2 will likely alter ecosystem structure,butinteractions are difficult to model or predict,par-ticularly relative to disturbance patterns.

• Changes in fresh water and tidal inflows intocoastal estuaries will likely alter the ecologicalstructure and function of these highly productiveand valuable ecosystems.

139

Chapter 5 / Southeastern United States

140


about 25% (NPA,1999). However, warming at high-er latitudes combined with increased heat stress inthe southeastern US may serve to decrease popula-tion migration towards the Southeast.

Based on the 1990 Census, about 61% of theSoutheast’s population is considered urban. Thenumber of farms in the region decreased 80%between 1930 and 1997 (USDA,1999). The 20thcentury was one of dramatic transition from anagrarian economy to one based on a combination ofnatural resources,manufacturing and trade,technol-ogy, and tourism. Roughly one half of 1990 employ-ment fell in the categories of manufacturing andwholesale/retail trade,compared with an average ofless than 5% in agriculture (US Bureau of theCensus,1994). Prior to 1950,corn and cotton werethe most important crops. Today agriculture is morevaried;soybean and hay outweigh cotton in acreageharvested and rice has become increasingly impor-tant. The Southeast still produces roughly one quar-ter of the nation’s agricultural crops,but timber har-vests are more valuable in terms of annual econom-ic impact in most states. Forest products industrieswere among the top four manufacturing employersin Mississippi,Alabama,North Carolina and Georgiain 1997. The Southeast has become America’s“woodbasket,” producing about half of America’stimber supplies. The region is also responsible for alarge portion of the nation’s fisheries,poultry, tobac-co,oil,natural gas,bauxite,coal,and sulfur produc-tion.

According to the 1990 census,18% of the popula-tion of the Southeast region lives below the povertylevel. The most poverty-prone areas include theLower Mississippi River Valley and parts ofAppalachia. While certain measurements,such asper capita income,have moved in the direction ofthe national averages,poverty rates in some areasare as much as two and a half times the nationalaverage. Levels of education of the population insome areas also lag behind national standards. Someof the smaller, more remote and geographically iso-lated areas of the region suffer from a lack of eco-nomic opportunities,have significant dependentpopulations,and lack the public institutions needed

POTENTIAL CONSEQUENCES OF CLIMATEVARIABILITY AND CHANGE FOR THESOUTHEASTERN UNITED STATES

PHYSICAL SETTING

The Southeast region (Fig.1) represents 15.4% ofthe land area of the US and 22.6% of its citizenry(US Bureau of the Census,1994). Although theSoutheast has considerable variation in landforms,itis possible to divide it into five fairly distinct regionsbased on physical geography. The lower third of thelow, flat Florida peninsula is a sub-tropical provincewith unique features such as the Everglades and theFlorida Keys. The Coastal Plain,which dominatesthe region,is a broad band of territory parallelingthe Gulf and South Atlantic seacoast from Virginia toTexas,with a deep extension up the lowerMississippi River valley. The Coastal Plain is relative-ly flat,with broad,slow-moving streams and sandyor alluvial soils. The Piedmont is a slightly elevatedplateau that begins at the “fall line,”where rivers cas-cade off the eastern edge of the plateau onto theCoastal Plain,and ends at the AppalachianMountains. This land is rolling to hilly, with manystreams. The Highlands comprise the inland moun-tain regions and include the southernmostAppalachian Mountains in the east and the Ozarkand Ouachita Mountains in the west. The InteriorPlains stretch into the north-central portion of theregion,including parts of Tennessee and Kentucky.The southern portions of the State of Virginia areincluded in aspects of this regional analysis. Thenorthern portions are included in the Northeastchapter with a discussion of the Chesapeake Baywatershed.

SOCIOECONOMIC CONTEXT The 544,000 square mile southeastern “sunbelt”isone of the fastest growing regions in the US. Thesoutheastern population increased by 21%,morethan double the national rate,between 1970 and1980 and another 11% between 1980 and 1990.Much of the historical population growth fromTexas to North Carolina occurred in the 151 coun-ties within the southeastern coastal zone,which areprojected to grow another 41% between 2000 and2025,compared to the projected national average of

141


southern boundaries while expanding west intoparts of eastern Oklahoma and Texas,and north intoparts of Missouri,West Virginia, Kentucky, andVirginia. Water stress and increased fire disturbancerestricts the Southeast forest under the Canadian cli-mate scenario,and large areas of the Southeast areconverted to savanna (grasslands with scatteredtrees and shrubs) and the Southeast forest movesinto the northcentral part of the US.

One of the biogeochemistry models (TEM) used inthe National Assessment projects large differences incarbon storage for the Southeast depending uponthe climate scenario used. Under the Canadian cli-mate model scenario, vegetation is projected to loseup to 20% of its carbon mass by 2030. However,under the Hadley climate scenario the same biogeo-chemistry model indicates that vegetation will addbetween 5 and 10% to its carbon mass over the next30 years. These differences in carbon storage reflectthe differences in climate scenario projections fortemperature and precipitation that are greatest inthe southeastern part of the US (Felzer and Heard,1999).

Ecological models used in the National Assessmentdo not simulate species-level response,nor do theysimulate land use changes,invasive species impacts,or other influences on ecosystems that cannot beeffectively modeled based on historical or empiricalevidence,unless the ecological models are linked

to support progressive development (Glasmeier,1998). These distressed counties present a profoundchallenge to policy makers concerned about climatechange mitigation strategies and issues,particularlyin the Appalachian coal-producing and Gulf Coastoil-producing regions.

ECOLOGICAL CONTEXTPrior to European settlement,the Southeast wasdominated by upland forests, grasslands,and wet-lands. Nearly one-third of the region may have beenwetland (Dahl,1990),but by 1990, wetlands hadbeen reduced to about 16% of the southeasternlandscape (Hefner, et.al.,1994). A wide range ofecosystem types is presently found in the region,ranging from coastal marshes to high-elevationspruce fir forests. Diversity of both plant and animalspecies is high compared to other regions consider-ing the extent of landscape alteration that hasoccurred. On an area basis,the Southeast has rela-tively high overall species richness indices (Ricketts,et.al.,1999). Vascular plant diversity is second onlyto Puerto Rico. Tree species richness is greatest inthe Southeast,with approximately 180 tree speciesfound in parts of South Carolina and more than 140tree species identified in most of the remainder ofthe region.

Forests still dominate parts of the Southeast;theshare of forestland in each state averages about 30%.About 20% of the present forests exist as pine plan-tations. Native longleaf pine was the predominantspecies in the Coastal Plain in the late 1800s butless than 3 million of the 60 million acres of south-eastern longleaf pine remain today (Boyer, 1979).More than 60 species of mammals occur in a rela-tively small area of the southern Appalachian moun-tains,while 40 or fewer mammal species are foundin the Coastal Plain (Currie,1991). The region hasvery high diversity of amphibians and reptiles.Roughly half of the remaining wetlands in the USare located in the Southeast,and more than three-quarters of the nation’s annual wetland losses overthe past 50 years occurred in the region.

Two types of ecosystem models (biogeochemicaland biogeographical) show a wide range of poten -tial changes in vegetation in the Southeast duringthe 21st century, depending upon the climate sce-nario selected. One of the biogeographical models(MAPSS) projects significant shifts in major biomesunder the Canadian climate scenario,but not underthe Hadley climate scenario. Under the Hadley cli -mate scenario,the Southeast mixed forest retains its

Figure 1. The Southeast region includes all of nine states(Alabama, Florida, Georgia, Kentucky, Louisiana, North Carolina,Mississippi, South Carolina, and Tennessee), the southern portionof Virginia, and 50 counties in east Texas. Four subregional work-shops were conducted in the Southeast region. See Color PlateAppendix

Land Cover Map of the Southeast Region

in assuming what the net effects of CO2 enrichmentmight be across a region or biome.

CLIMATE VARIABILITY ANDCHANGEPast Century

The Southeast has some of the warmest conditionsin the US. However, it is the only region to showwidespread but discontinuous cooling periods of 2to 3.5ºF (1 to 2ºC) over almost the entire area dur-ing the past 100 years (Figure 2). Peninsular Florida,North Louisiana,and a few small areas in theAppalachian Mountains have shown a modest warm-ing of around 2°F (1ºC) since 1900. The reason theSoutheast temperature record shows a net coolingtrend over the past 100 years is that there was awarm period between the 1920s and 1940s,then asignificant downward trend through the late 1960s.The mid-1900s cooling trend may have been due tonatural variation. Human-caused sulfate aerosolemissions during this period may have also playedsome role. Sulfate aerosols reflect some sunlightback into space,thereby cooling the atmosphere(Kiehl,1999). Since 1970,the average annual tem-perature increased,with the most significantincreases occurring during the 1990s. Trends intemperature extremes over the past one hundredyears exhibit a decrease of about 5 days in the num-ber of days per year exceeding 90ºF (32.2ºC),and anincrease of 6 days in the number of days belowfreezing over the entire region. However, over thepast fifty years the average annual length of thesnow season decreased by 4 days.

The Southeast receives more rainfall than any otherregion. Annual precipitation trends show increasesof 20-30% or more over the past 100 years acrossMississippi,Arkansas,South Carolina,Tennessee,Alabama,and parts of Louisiana,with mixed changesacross most of the remaining area. The southernmountains of North Carolina,the southern tip ofTexas,and a couple of other small areas have slight-ly decreasing trends in annual precipitation. Muchof the increase in precipitation was associated withmore intense events (rainfall greater than 2 inchesor 5 cm per day). A small percentage of theincreased precipitation was associated with moder-ate rainfall events,which are generally beneficial toagriculture and water supply. Analysis of streamflow trends during 1944-1993 showed little changein annual maximum daily discharge,but significantincreases in annual median and minimum flows in

with other process models,both biological and eco-nomic. For example,Harcombe and others (1998)observed that Chinese tallow, a freeze-intolerantnon-native tree species,increased dramatically insoutheastern Texas over the past few decades.Chinese tallow increased by a factor of 30 between1981 and 1995,often out-competing native specieswhen canopy gaps form in mesic (medium mois-ture) and wet sites (Harcombe,et.al.,1998). Thesekinds of interactions and changes in forest dynamicsare difficult to simulate.

Mixed responses among species to fertilizationeffects of elevated atmospheric CO2 further con-found our ability to model ecosystem structure andproductivity. Several studies showed that elevatedCO2 increased photosynthesis rates and improvedwater use efficiency in many forest species and agri-cultural crops (Acock,et.al.,1985;Allen,et.al.,1989;Nijs,et.al.,1988). Two reviews of CO2 exposurestudies with deciduous and coniferous speciesfound that increases in growth rates varied widely,but that generally tree growth was stimulated byincreases in CO2 (Eamus and Jarvis,1989;NCASI,1995). However, limits on the availability of soilnutrients and water in many natural or semi-naturalecosystems can severely constrain the potentialimprovement in water use ef ficiency due to sup-pressed transpiration induced by enhanced CO2 lev-els,thereby offsetting potential gains in productivity(Lockwood,1999). Temperature,plant pests,air pol-lution,and light availability could also limit thepotential enhancement of growth by elevated CO2(NCASI 1995). Hence,one should be very cautious

142


Figure 2. Decadal average temperatures in the Southeast.Source D. Eastering, NOAA National Climatic Data Center. SeeColor Plate Appendix

Southeast US Annual Mean Temperature

the lower Mississippi Valley, and decreases in thesecategories in parts of Georgia and North Carolina(Linns and Slack,1999). Increased precipitationintensity in extreme events during the next centur yis suggested by climate models under doubled CO2

for the US (Mearns,et.al.,1995) and there is evi-dence that moisture in the atmosphere is increasingover the Caribbean region (Trenberth,1999).Heavy rains are less efficient (more water runs offinto the sea) and are more likely to cause flooding,which is a serious problem in the region.

Trends in wet and dry spells during the 20th centu-ry, as indicated by the Palmer Drought SeverityIndex (PDSI),are spatially consistent with theregion’s annual precipitation trends,showing astrong tendency to more wet spells in the GulfCoast states,and a moderate tendency in mostother areas. The percentage of the southeasternlandscape experiencing “severe wetness”(periodsin which the PDSI averages more than +3)increased approximately 10% between 1910 and1997.

Effects of El Niño on Climate in the SoutheastThe El Niño/Southern Oscillation (ENSO) phenome-non contributes to variations in temperature andprecipitation that complicate longer-term climatechange analysis in certain parts of the country, par-ticularly the Southeast. ENSO is an oscillationbetween warm and cold phases of sea-surface-tem-perature (SST) in the tropical Pacific Ocean with acycle period of 3 to 7 years. US climate anomalies(departures from the norm) associated with ENSOextremes vary both in magnitude and spatial distri-bution. El Niño events (the warm phase of theENSO phenomenon) are characterized by 2 to 4°F(about 1 to 2ºC) cooler average wintertime air tem-peratures in the Southeast. During the spring andearly summer months,the region returns to near-normal temperatures. Precipitation anomaly pat-terns following warm events indicate that GulfCoast states encounter wetter than normal winters(by about 1 to 2 inches per month). By the spring,the entire eastern seaboard shows increased precip-itation. In summer, climate impacts of warm eventsare more localized; for example,drier conditions arefound in eastern coastal regions,and from northTexas to northern Alabama. El Niño events also cre-ate upper atmospheric conditions that tend toinhibit Atlantic tropical storm development, result-ing in fewer hurricanes,while La Niña events havethe opposite effect, resulting in more hurricanes.Figure 3 depicts US Gulf of Mexico hurricane land-fall trends and the probability of hurricane landfallduring El Niño and La Niña years.

143


Figure 3(a). US Hurricane Landfall Trends in the Gulf of Mexico.This figure shows the number of US hurricanes making landfall inthe Gulf of Mexico by decade for the past 100 years. There werepeaks in activity during the 1910s and 20s, as well as a lower peakin the 1960s. The past 30 years have shown a decrease in the num-ber and intensity of Gulf hurricanes making landfall. See ColorPlate Appendix

Gulf Landfalling Hurricanes by Decade

Figure 3(b). Effect of ENSO Phase on Hurricane LandfallThis figure shows the probability of the number of hurricane land-falls on the US in a given hurricane season and ENSO phase (ElNiño, Neutral, La Niña). Based on the past 100-year record, theprobability of at least 1 hurricane landfall is similar for all threephases, with probabilities ranging from 78% for El Niño to 90% forLa Niña. For multiple landfalls, however, the differences caused byENSO phase become apparent. The probability of at least 2 land-falls during El Niño is 28%, but is 48% in neutral years, and 66%during La Niña. The probability of at least 3 landfalling hurricanesis near 0% for El Niño, 20% for neutral years, and 50% for La Niña.It is clear that El Niño years have few multiple hurricane strikes onthe US, while neutral years and La Niña years often see multiplehurricane strikes on the US coast. (Source: Florida StateUniversity, Center for Ocean-Atmosphere Prediction). See ColorPlate Appendix

US Hurricanes

144


During La Niña events (the cold phase of ENSO),theanomalies are sometimes reversed from those asso-ciated with warm events,but not everywhere.Above-average wintertime temperatures are presentEast of the Mississippi. By spring,the warmer anom-alies in the east are focused in the Ohio Valley andnorthern Florida,Georgia,and South Carolina.Wintertime precipitation patterns associated withcold events show increases (1-2 inches or 2.5-5 cmper month) in the band stretching from northernMississippi to southwestern Pennsylvania. In thespring,Gulf Coast areas have increased precipita-tion. In summer, the extreme southern US is colderthan normal and greater precipitation is evident inthe Southeast. Dry to very dry conditions are foundin parts of Texas and Louisiana. Thus,as suggestedby these results,the climate anomalies associatedwith opposite phases of ENSO are not in directopposition. For example, climate anomalies inFlorida are nearly opposite (for cold and warmevents) while those in many of the midwesternstates are of the same sign for both precipitationtotals and temperatures during both warm and coldevents. Further evidence demonstrates that climateanomalies associated with strong warm events arenot amplifications of normal warm events(Rosenberg,et al.1997).

Scenarios for the Future

The Hadley Centre climate model projects that by2030,maximum summer temperatures in the

Southeast will increase by about 2.3°F (1.3ºC) whilemaximum winter temperatures will increase by1.1°F (0.6ºC). The projected increase in mean annu-al temperature of 1.8ºF (1.0ºC) by 2030 and 4.1ºF(2.3ºC) by 2100 represents a smaller degree of pro-jected warming than for any other region. Thesmaller simulated warming rate is possibly due tothe buffering affects of the oceans,large amounts ofsurface water for evaporative cooling,and the sul-fate aerosol emissions that are prevalent throughoutthe eastern US. Sulfate aerosols may help explainthe mid-20th century cooling trend in parts of theUS;however, over the past two decades,sulfateemissions decreased,and the future cooling affect ofsulfate aerosols is not expected to be as importantdue to Clean Air Act restrictions. Although theincrease in temperature under the Hadley model issmall compared to other regions,the resultingincrease in the summer heat index by 8 to15ºF (4 to8ºC) (calculated from monthly maximum tempera-ture and relative humidity) would likely af fecthuman activity and,possibly,demographics in the21st century.

The Canadian Centre climate model projects highertemperature scenarios and higher southeastern heatindices by the end of the 21st century than does theHadley model. The Canadian model simulates anincrease in mean annual southeastern temperatureof about 3ºF (1.7ºC) by 2030 and 10ºF (5.5ºC) by2100. In the Canadian model,increases in maxi-mum summer temperature are the highest in theNation for both 2030 (5ºF or 2.8ºC) and 2100 (12°For 6.5ºC). The Canadian climate model simulates anincrease in average summer heat index above 15ºF(8ºC) across the entire region by 2100 (Figure 4).

Another important difference between the twomodels for the Southeast lies with the simulatedchanges in rainfall;the Canadian climate model simu-lates reduced average annual precipitation (10% lessthan present by 2090) while the Hadley model simu-lates more precipitation than present (20% more by2090). These differences have important implica-tions for hydrologic impacts on the Southeast,because the Canadian model simulates decreasedsoil moisture,while the Hadley model simulatesincreased soil moisture (Felzer and Heard,1999).Differences between the two models are illustratedin Figure 5,which depicts the simulated summer cli-mate in Georgia in 2030 and 2090. According to theHadley climate model scenario,the Southeast willremain the wettest region of the US for the next 100years. The precipitation changes projected by theHadley model by 2100 are consistent with otherparts of the eastern and midwestern US.

July Heat Index Change - 21st Century

Canadian Model Hadley Model+25ºF

+20ºF

+15ºF

+10º

+5ºF

0º

+25ºF

+20ºF

+15ºF

+10º

+5ºF

0º

Figure 4. The changes in the simulated heat index for theSoutheast are the most dramatic in the nation with the Hadleymodel suggesting increases of 8 to 15ºF for the southern-moststates, while the Canadian model projects increases above 20ºF formuch of the region. Heat indices simulated for the Southeast by2100. (source, NOAA National Climatic Data Center). See ColorPlate Appendix

145


The Max Planck Institute climate model (ECHAM4/OPYC3),one of a few models with suf ficient resolu-tion in the tropics to adequately simulate narrowequatorial upwelling and low frequency waves,sim-ulates more frequent El Niño-like conditionsand stronger La Niñas under a doubling of CO2,which is consistent with the Hadley model pro-jections with a doubling of CO2. The Max-Planck model also suggests that the mean cli-mate in the tropical Pacific region will shifttoward a state corresponding to present-day ElNiño conditions (Timmermann,et.al.,1999).McGowan,Cayan,and Dorman (1998) showedthat the frequency of warm sea surface eventsoff the western coast of North Americaincreased since 1977,but relationships betweenthis trend and reduced hurricane landfall in theGulf Coast region have not been established.

Table 1: Billion-Dollar Southeast Weather Disasters, 1980-1999

Disaster Year Estimated Damages/ EstimatedCosts* Deaths**

AR/TN Tornadoes 1999 $ 1.3 billion 17TX Flooding 1998 $ 1.0 billion 31Hurricane Georges 1998 $ 5.9 billion 16Hurricane Bonnie 1998 $ 1.0 billion 3Southern Drought/Heat Wave 1998 $ 6.0-9.0 billion 200El Niño/Tornadoes and floods 1998 $ 1.0 billion 132MS/OH Valley Floods/Tornadoes 1997 $ 1.0 billion 67Hurricane Fran 1996 $ 5.0 billion 37Hurricane Opal 1995 $ 3.0 billion 27TX/OK/LA/MS Severe Weather 1995 $ 5.0-6.0 billion 32TX Flooding 1994 $ 1.0 billion 19Tropical Storm Alberto 1994 $ 1.0 billion 32Southeast Ice Storm 1994 $ 3.0 billion 9Summer Drought/Heat Wave 1993 $ 1.0 billion ***Hurricane Andrew 1992 $ 27.0 billion 58Hurricane Bob 1991 $ 1.5 billion 18TX/OK/LA/AR Flooding 1990 $ 1.0 billion 13Hurricane Hugo 1989 $ 9.0 billion 57Hurricane Juan 1985 $ 1.5 billion 63Hurricane Elena 1985 $ 1.3 billion 4Florida Freeze 1985 $ 1.2 billion 0Florida Freeze 1983 $ 2.0 billion 0Hurricane Alicia 1999 1983 $ 3.0 billion 21Total $ 83.7-87.7 billion 856

* not adjusted for inflation,** US only, *** undetermined (Source:National Climatic Data Center, 1999)

Figure 5. Illustration of how the summer and winter climates inGeorgia would shift under the Hadley climate scenario (HADCM2).For example, the summer climate in Georgia in the 2030s would bemore like the current climate of the Florida panhandle. Source:NOAA, National Climatic Data Center. See Color Plate Appendix

Summer Climate Changes from Hadley Centre Scenario

KEY ISSUES1. Weather-related Stresses on Human Populations 2. Agricultural Crop Yields and Economic Impacts3. Forest Productivity Shifts4. Water Quality Stresses5. Threats to Coastal Areas

Changes in average climate and weather extremeshave important economic implications in theSoutheast. There are several reasons why thisregion is of relatively high interest and concern.First,there is a strong ENSO signal,primarily inthe Gulf Coast states,that results in seasonal andyear-to-year variations in temperature and precipi-tation. Understanding potential future climatechange in the context of current natural variabili-ty can provide an important contribution to ongo-ing discussions of mitigation options. A secondconsideration is that the Southeast experiencesmany extreme climate events such as hurricanes,heat waves,tornadoes,ice storms, floods,andlightning storms that cause significant economiclosses to industry and local communities. The agri-culture and forestry sectors make substantial contri-butions to the regional and national economy andthese sectors are quite vulnerable to climate vari-ability. Water resources,air quality, coastalresources,and land use are other important regionalissues that may be strongly influenced by climatictrends and variability.

1. Weather-related Stresses onHuman Populations

The US experienced 42 weather-related disastersover the past 20 years that resulted indamages/costs in excess of $1 billion each;23 ofthese disasters occurred in Southeast states, result-ing in total damages/costs of about $85 billion.Most of the property damages were associated withfloods and hurricanes. Low-lying Gulf and SouthAtlantic coastal counties are particularly vulnerableto storm surge. Between 1978 and 1998,56% of theNational Flood Insurance Program (NFIP) policies inforce and 74% of total NFIP claim paymentsoccurred in southeastern coastal counties (HeinzCenter, 1999).

In addition to the projected shift towards more fre-quent El Niño-like conditions in the Southeast,someclimate models suggest that rainfall associated withEl Niño events will increase as atmospheric CO2increases. Increased flooding in low-lying coastalcounties from the Carolinas to Texas could possibly146


have adverse effects on human health. Floods arethe leading cause of death from natural disasters inthe Southeast and nationwide. Flooding,however, isnot the only problem stemming from unusual mete-orological events. The southern heat wave anddrought of 1998 resulted in damages in excess of $6 billion.

El Niño and the 1998 Florida WildfiresFlorida consistently ranks among the top five statesin terms of wildfire frequency and acreage affected,due largely to frequent thunderstorms and a warmmoist climate that promotes lush growth of volatileunderstory plants. To limit the accumulation offuels that promote disastrous wildfires,landownersin Florida routinely treat close to 2 million acres ayear with prescription fire. The unseasonably warmweather and copious rainfall brought about by ElNiño conditions during the winter of 1997-98 result-ed in even higher plant growth than usual and highsoil moisture that limited the acreage that could betreated effectively with dormant-season prescriptionfire. As El Niño conditions began to subside inMarch,1998, record breaking rainfall changed torecord-setting drought. Many prescription burnswere postponed further because of the increasingprobability of crown fires and root damage to trees.Lightning activity picked up as usual in May, butwith lower than average rainfall. By June 1,droughtindices reached record heights. During the ensuingsix weeks,more than 2,500 fires burned roughly500,000 acres in Florida,destroying valuable timberand damaging roughly 350 homes and businesses.Predicting El Niño conditions holds obvious benefitsfor fire preparedness and prevention. Changes in ElNiño patterns would have both ecological and eco-nomic implications.

Also of concern in the Southeast are the effects thatelevated surface temperatures have on humanhealth as a result of prolonged or persistent periodsof excessive summertime heat events coupled withdroughty conditions. For example,it is known thaturban surface temperatures in cities in the Southeastcan be elevated as much as 5 to 10ºF (approximate-ly 3 to 5ºC) over non-urbanized areas (Lo,et.al.,1997;Quattrochi and Luvall,1999). These elevatedurban surface temperatures are a heat stress tohumans and can significantly contribute to increas-ing both the duration and magnitude of photochem-ical smog,particularly ozone concentrations(Southern Oxidants Study, 1995;Quattrochi,et.al.,1998). Increases in maximum summer temperaturesare of particular concern among lower incomehouseholds that lack sufficient resources to improveinsulation and install and operate air conditioningsystems.

Adaptation OptionsUnderstanding the risks and vulnerability of commu-nities to weather-related hazards (considering hid-den and reported costs and the actual frequencywith which these disasters occur in the Southeast)is important to the quality of adaptation strategies.Across the region,intense precipitation hasincreased over the past 100 years and some modelssuggest that this trend will continue as the atmos-phere warms. Traditional approaches to mitigationsuch as flood proofing,elevated structures,andbuilding codes,are no longer adequate in them-selves,particularly in the coastal zone. Even ifstorms do not increase in frequency or intensity, sea-level rise alone will increase the propensity forstorm surge flooding in virtually all southeasterncoastal areas.

The National Oceanic and AtmosphericAdministration (NOAA) and the Federal EmergencyManagement Agency (FEMA) commissioned a studyon the true costs and mitigation of coastal hazardsin 1996. The report of this study calls for a strategicshift in hazard mitigation and focuses on modelstate programs developed in Florida and othersparts of the country to foster more disaster-resilientcommunities. Recommendations include improve-ments in disaster cost accounting and risk assess-ment,insurance/mitigation policy linkages,integrat-ed approaches to coastal management/develop-ment,and community-based mitigation planning(Heinz Center, 1999). Changes in climate and sea-level rise should be an integral consideration asSoutheast coastal communities develop strategiesfor hazard preparedness and mitigation. Severalstates have implemented permanent “set backs”or“rolling easements”to prevent further developmentin areas that will become more flood-prone as sealevel rises (see Coastal and Marine Resources chap-ter).

Health advisory systems,community-wide heat147


emergency plans,improved weather predictioncapabilities,and other adaptations that would likelyreduce urban heat stress and air pollution-relatedhealth effects are presented in the Northeast andHealth Chapters.

2. Agricultural Crop Yields andEconomic Impacts

Current ConditionsGreat agricultural changes have taken place in theSoutheast over the past 150 years. In 1849,theSouth produced more corn than the Midwest;Southeast acreage in corn was higher than in cot-ton. Cotton production expanded greatly after theCivil War, and by the late 1920s,cotton was moredominant in the South than it was a century before.There was complete mechanization of crop produc-tion in the Southeast after World War II and millionsof sharecroppers moved to the big cities in theNorth. This had an important impact,not only onlabor requirements,but on the whole economicstructure of agriculture. There has also been a shift-ing cropland base in this region. For example, overthe last 50 years soybeans changed from a minorforage crop to an agricultural staple second only tocorn in value of production. As soybeans and ricereplaced corn and cotton, farmers chose soils mostsuitable for the new crops. Drained wetland soils inArkansas were more productive in soybeans thanthe old Piedmont soils abandoned by cotton farm-ers.

In terms of agricultural potential,one of theSoutheast’s most important assets is its potential toexpand the acreage devoted to crops beyond thecurrent level. The land from which new croplandcan be drawn is currently about evenly dividedbetween pasture and forestland. Although theSoutheast could substantially increase acreage devot-ed to agriculture,it fares poorly in terms of native

Table 2: Principal Crops in the Southeast(103 acres)

(source,USDA,Census of Agriculture,1996).

1929 1949 1969 1987 1996

Corn 23,940 20,417 7,896 4,309 5,005

Cotton 23,228 13,031 4,711 3,345 5,931

Peanuts 2,207 2,348 1,046 971 927

Rice 598 1,011 1,194 1,654 2,156

Soybeans 1,321 2,599 13,894 25,645 12,303

148


soil fertility. In addition to having low fertility, mil-lions of acres of soils in the Southeast are moder-ately to severely eroded,the result of decades ofcontinuous corn and cotton production under poorsoil management. However, another of the region’sagricultural assets is its latitude and proximity tothe warm Gulf and Caribbean maritime influence.Overall,the Southeast has a consistent 30- to 90-daylonger growing season than the Corn Belt and theGreat Plains. The Southeast has enormous suppliesof fresh water in the form of rainfall,surface waterflowing through streams and creeks,and groundwa-ter. Water availability gives the Southeast some sub-stantial advantages,including irrigation possibilitiesthat have barely been exploited.

The Southeast’s mild climate and frequent rainfallpredispose the region to an array of agriculturalpest problems more serious than anywhere else in

the nation. Agricultural pests can reduce crop yieldsand raise production costs. Another consequence ofthe region’s pest problems has been relatively highuse of pesticides. Although the Southeast accountsfor only 14% of the nation’s cultivated cropland,itconsumes 43% of the insecticides and 22% of theherbicides used by farmers (USDA,Census ofAgriculture 1994).

Potential Impact of Climate Change on Crop Yieldand Water Use at the Field ScaleTwo families of mechanistic crop simulation modelsCROPGRO (for soybean and peanut) and CERES (forcorn,wheat, rice,and sorghum) in DSSAT V 3.5(Tsuji et al.,1998; Jones et al.,1999) were used tosimulate potential dryland and irrigated crop pro-duction using state- and crop-specific managementpractices throughout the Southeast. The HadleyCentre model was chosen for this analysis by theSoutheast working group because this model wasgridded properly for the region at the time theanalysis was begun,and because the model per-formed best,among those tested,at hindcastingsoutheastern ENSO-related climate anomalies.

Crop yield changes simulated under dryland (non-irrigated) conditions suggest that yields were sensi-tive to the Hadley climate change and CO2 fertiliza-tion and that the response varied by crops and loca-tions (Figure 6). Dryland crop yields generallydecreased along the Gulf Coast by 1 to 10% due towater stress. Furthermore,increased demand forwater by other sectors under higher temperatures islikely to amplify these impacts. Increases in atmos-pheric CO2 concentrations may reduce crop wateruse to some extent,due to increases in leaf stomataresistance,but cannot fully compensate for increas-es in crop water demand due to the higher tempera-tures in climate change scenarios.

In the crop simulation model,dryland corn yieldincreased by 1 to 30% in the Coastal Plain anddecreased up to 10% in Louisiana and large parts ofMississippi,Arkansas,and Kentucky. The shortergrowing cycle reduced yield while increased CO2

and rainfall boosted yield. Due to lower water useefficiency, the model suggests that it could possiblybecome uneconomical to ir rigate corn,promptingfarmers to increase dryland corn production.Sorghum yields increased in the model results by 1to 30% in parts of Alabama,Georgia,South Carolina,and North Carolina where seasonal rainfallincreased by 5 to 15%. Simulated yields from irrigat-ed sorghum were 4 to 7% lower almost everywhere,even where higher yields were predicted under dry-land conditions,largely due to shorter growing sea-sons.

Figure 6. Dryland crop yield changes in 2030 (a) and 2090 (b) with-out adaptation for various climate sensitivity scenarios (source:Auburn University, Global Hydrology and Climate Center; Universityof Florida, Agricultural and Biological Engineering Department).See Color Plate Appendix

Dryland Crop Yield Changes

Simulated soybean and peanut yields increased by 1to 30% mostly within the Coastal Plain and mid-southand more than 30% in parts of North and SouthCarolina. Yields also increased,in parts of Arkansasand upper Mississippi,where dryland corn andsorghum yields decreased. Irrigated yields increased1 to 10% over almost all of the region includingwhere losses had been predicted under dryland con-ditions. The models simulated decreases of up to30% in dryland peanut yields in the lower Delta andalong the Gulf Coast,but when irrigation was addedin the same areas,yield increased by over 30%. If themodel-simulated changes were to occur slowly overnext 25 to 45 years, farmers would be likely to slow-ly increase irrigation as the marginal value of irriga-tion increased. The spatial variation in simulatedyield induced by climate change suggests that manyfarmers in the lower Delta and Gulf Coast may dropdryland production of peanuts while production ofthese crops may expand in other parts of the region.

Simulated winter-wheat yield increased in all regionsexcept in the lower Delta and parts of Arkansas.Simulated irrigated yields increased following a simi-lar trend as the dryland yield. Demand for irrigationincreased by 20 to 50% in the Delta where rainfalldecreased,and evaporation increased due to highertemperatures. Over the same period,irrigated yieldsdeclined. In parts of Arkansas and Louisiana,whereirrigated rice dominates,the model simulated 1 to10% yield losses by the 2030s and 3 to 39% increasesby the 2090s. One of the major threats to rice pro-duction is increasing competition for and increasingcosts of irrigation water.

Sensitivity Analysis of Crop Response to Climatic Change To avoid the narrow range of the climate conditionssimulated by the Hadley model, we conducted a sen-sitivity analysis in 10 agriculturally-dominant south-eastern areas using 25 combinations of anticipated

temperature and rainfall at 445 and 680 ppm CO2

levels superimposed on the current climate. Thesensitivity analysis identified the climate conditionsthat would be particularly damaging to the drylandproduction (see Table 3) and allowed us to considerto what extent yields can be maintained with cur-rent management practices. Results indicate thatyield would likely decrease compared to the currentvalues if temperature exceeds the current valuemore than the corresponding value for change inrainfall amount.

Figure 7 shows that under simulated 2030s CO2levels,+1ºC (1.8ºF) change in temperature increaseddryland production of soybean,peanuts,and wheatunder current rainfall,while yields of corn andsorghum declined. A 2ºC (3.6ºF) increase in temper-ature in 2030 resulted in further yield losses for allcrops simulated. In contrast,the effects of 2ºC(3.6ºF) temperature increase in 2090 with nochange in rainfall suggested a generally positiveeffect on crop yields. However, decreases in rainfallby 20% accompanied by temperature increases of 1to 2ºC (1.8-3.6ºF) almost doubled yield losses for alldryland crops studied. Under conditions of loweror the same growing season rainfall amounts,yieldsof all crops increased more due to increased CO2

levels than due to higher seasonal rainfall. For allcrops and combinations of temperature and CO2

changes,decreases in precipitation resulted in differ-ential decreases in crop yields. Changes in yieldswith 20% lower rainfall were of similar magnitude atall other temperatures and CO2 levels simulated.Furthermore, results showed that crop yields weremuch less sensitive to changes in temperature com-pared to changes in precipitation. An increase ingrowing season rainfall of about 20% almost com-pletely offset the negative ef fect of temperatureincrease. Irrigated yields were simulated assumingcurrent rainfall by varying only the temperatures

149


Table 3: Temperature Tolerance Limits for Various Cropsto the projected rainfall changes in 2030s and 2090s.

445 ppmv CO2 as in 2030s 680 ppmv CO2 as in 2090s Change in Current Rainfall Change in Current Rainfall

-20% 0% 20% -20% 0% 20%

Temperature Tolerance

Soybean +0 +1 +3 +1 +3 +4 Peanuts +0 +2 +3 +2 +3 +4 Corn +0 +0 +1 +2 +3 +3 Sorghum +0 +0 +1 +1 +1 +1 Wheat +0 +1 +2 +3 +3 +3

variety and planting date adaptations either by shift-ing from a decreased yield without managementchanges to an increased yield when managementwas changed. For most dryland crops,adaptationdid not completely eliminate yield loss under a 20%less rainfall condition. For corn,peanuts,sorghum,and wheat,there may be little need to change cur-rently adapted varieties under all combinations oftemperature and rainfall changes,in part due tostrong CO2 fertilization effects on crop yields.

Sub-regional Impacts on Productivity andProfitability The largest threat to crop production in theSoutheast appears to occur where decreases in pre-cipitation coincide with higher temperatures. Thiscombination of climate change would likelyincrease evapotranspiration demands in a climatewith lower water availability, which would increasewater stress and reduce crop growth and yields. Inmany areas of the Southeast,however, projectedtemperature rises of no more than 3 or 4ºF (2ºC orless) together with declining needs for irrigation,should enhance crop production. The Hadleymodel scenario for 2030 indicates improved condi-tions for water availability in the Tennessee Valley,coastal North and South Carolina,and the lowerMississippi Valley. Water supplies are projected tobe much worse in the Mississippi Delta and slightlyworse in Louisiana,Southern Alabama,the Floridapanhandle,and the Coastal Plain of Georgia. By2090 they also become much worse in NorthernMississippi,Southern Alabama,and SouthwesternSouth Carolina.

A farm management model was used to simulatechanges in crop mix, water use,and farm incomeassociated with the climate-induced yield changesdescribed above (Hatch et al.,1999). Of the majorcrop growing areas of the Southeast,the southernMississippi Delta and Gulf Coast areas are more neg-atively impacted,while the northern Atlantic CoastalPlain is more positively impacted. Analyses indicatethat farmers could possibly mitigate most of thenegative effects and possibly benefit from changesin CO2 and moisture that benefit crop growth. Thediscussion that follows is organized around the twoprincipal row crop growing areas of the Southeast,the Mississippi Delta and Atlantic Coastal Plain.

The southern portion of the Mississippi Delta isexpected to endure the severest negative impactswith the northern portion relatively less impacted.In both 2030 and 2090,simulated crop yield, wateruse,and income all are relatively worse off in thesouthern area of the Delta,particularly Louisiana,

from +1 to +5ºC (1.8 to 9ºF). If future rainfall ishigher, then irrigation requirements will likelydecrease,and if it is lower, irrigation needs will like-ly increase. Most irrigated crops were sensitive tosimulated climate changes and CO2 fertilization.

Implications for Field Scale AdaptationFuture climate change may strongly affect agricul-ture in the Southeast. In cropping systems,a widerange of low cost adaptations to climate change mayexist to maintain or even increase crop yields underfuture climates. Farmers may be able to respond tochanges in environmental conditions by choosingthe most favorable crops,cultivars,and croppingsystems. Our findings indicate that changes in plant-ing dates and maturity groups could possiblyincrease yield under the climate change conditionsstudied. In the simulations,all crops benefited from150


Figure 7. Dryland yields changes in 2030(a) and 2090(b) withoutadaption for various climate sensitivity scenarios. (source: AuburnUniversity, Global Hydrology and Climate Center; University ofFlorida, Agricultural and Biological Engineering Department). SeeColor Plate Appendix

Simulated Changes in Dryland Yields for Southeastern Crops based on the Hadley

(HADCM2) Scenario

Mississippi,and Arkansas. This picture contrastsrather sharply with the largely beneficial impacts inmuch of the Coastal Plain,especially the northerntier. Southern Alabama,the Panhandle of Florida,and southwest Georgia,the crop growing areas inproximity to the Gulf coast,are the areas of theCoastal Plain that are negatively impacted. The restof this important crop growing area,that stretchesfrom central Georgia to North Carolina,is expectedto see beneficial impacts from the climate-inducedyield changes and the resultant changes in farmmanagement.

Simulated changes in water use for ir rigation of rowcrops show a distinct north-south pattern. That is,the southern tier of the Southeast is expected toincrease its needs for irrigation water whereas thenorthern tier is expected to decrease its relativeneed for irrigation water. This pattern is somewhatevident in the 2030 simulation and very pro-nounced in the 2090 crop and management simula-tions.

Economic sensitivity to increased temperatures wasalso investigated. Two sensitivity scenarios wereanalyzed to provide an indication of sensitivity toincreased temperature without any changes in pre-cipitation. The sensitivity scenarios were “hot”and“very hot;”the former was an increase of 1ºC in2030 and 2ºC in 2090,and the very hot scenarioincreased temperature by 2ºC in 2030,and 5ºC in2090. These temperature changes were selectedbecause they roughly reflect the temperaturechanges associated with the Hadley and Canadianmodels, respectively, without the simulated changesin precipitation. The “hot”scenario had a slightlymore negative impact than the Hadley scenario inmany areas of the Southeast because the hot sce-nario did have the Hadley’s accompanying increasein moisture. The “very hot”scenario producedrather dramatic negative effects, again because thesewere not mitigated by additional moisture.

The heterogeneous growing conditions of theSoutheast and the great diversity of crops and man-agement systems used in the region make broadgeneralizations about regional climate ef fects onagriculture very difficult. The Southeast is oneregion of the nation that is very likely to experiencechanges in the mix of crops that can be profitablygrown. As a result,the Southeast is a region that willgain from improved information on climate effectsand on improved dissemination of this informationto farm managers. Improvements in understandingclimate and forecasting weather will improve theability of managers to deal effectively with these

and future changes, for example by providing themwith forecasts based on ENSO phase (Legler, et.al.,1999).

Additional Adaptation OptionsExpected changes in productivity and profitabilitywill very likely stimulate adjustments in manage-ment strategies. As yields change,commodity priceswill also change. Producers have several options bywhich they can adapt to changes in yield and priceexpectations. As previously pointed out,they canchange to alternative crops. They can also grow thesame crops,but adjust cultural practices, varyingplanting dates,seeding rates, row spacing,patternsof water usage,crop rotations,and the amounts,tim-ing,and application methods for crop nutrients andpesticides.

Technology can also be expected to respond withnew products and methods to optimize productionunder changing climatic conditions. Plant breederswill very likely respond by developing new varietiesto accommodate climatic changes. Combinations oftechnological advances and adaptive managementpractices could very likely minimize the potentialadverse effects and amplify the potential positiveeffects of climate change on agricultural productivi-ty in the Southeast.

3. Forest Productivity Shifts

Current ConditionsMost of the Southeast’s native forests were convert-ed to farmland by 1920,with a large percentage ofthis conversion occurring prior to the Civil War. By1860, about 43% of the total land area in theSoutheast was reported as farmland,but a substan-tial part of the farm holdings remained in forest,which was often used as a place for grazing live-stock (USFS,1988). With continued expansion ofsettlements,timberland continued to decline untilthe early 1920s. Significant changes in agriculturetook place after 1920 that caused abandonment oflarge areas of crop and pasturelands. These includedthe boll weevil infestation,which made cottongrowing unprofitable in many parts of theSoutheast. Some of the abandoned land was plantedwith trees,but the majority reverted naturally to for-est leading to increases in timberland acreage (USFS1988).

By the late 1950s and early 1960s,the decline intimberland began again in the Southeast,caused pri-marily by the clearing of forest for soybean andother crop production. Much of this timberlandreduction occurred in the bottomland hardwood 151


and their relationships with climate variability andchange remains an integral part of the economicwell being of the Southeast.

Potential Impacts of Climate ChangeAs part of this Assessment,PnET-II,a physiologically-based forest process model,combines climate,soil,and vegetation data to simulate annual soil waterstress,drainage and biological productivity in south-ern US forests (Aber, et.al.,1993;McNulty, et.al.,1994). Using Forest Inventory and Analysis datafrom the USDA Forest Service,predictions of forestproductivity per unit area were projected intoregional growth using current volume for southernpine forests (USDA,1988). Model projections offuture forest productivity included the influence ofdoubled CO2. The Hadley climate scenario (the wet-ter of the principal climate models used in theNational Assessment) was used for reasons cited ear-lier. Total changes in standing volumes were calcu-lated by multiplying growth per unit area by thetotal area of pine forest across the region.

The PnET-II model indicated a 12% increase in over-all southeastern forest productivity by 2100,butthere were important differences between hard-woods and pines. In model simulations using theHadley scenario,southern pine plantations experi-enced an 11% increase in productivity by 2040 andan 8% increase by 2100,while hardwood and mixedpine hardwood forest (which represent 64% of thetotal forest area) experienced a 22% increase in pro-ductivity by 2040 and a 25% increase by 2100 com-pared to 1990 productivity estimates. This differ-ence would likely be accentuated under the warmertemperatures simulated by the Canadian model.This is significant because pines (used for pulp andpaper),presently account for almost two-thirds ofthe region’s forest industry land and about half ofthe nation’s softwood inventory. Climate modelsused as input in both the forest and VEMAP modelssuggest a northward shift in forest productivity overthe next century, but they do not consider changesin management that could potentially ameliorateadverse effects. At least two ecosystem models runwith the Canadian climate scenario suggest thatthere will be a 25 to 50% increase in fires and thatpart of the southeastern pine forest will be replacedby pine savannas and grasslands due to increasedmoisture stress (see Vegetation and BiogeochemicalScenarios Chapter).

Figure 8 shows PnET-II model outputs for southernforest net primary productivity in the Southeast atpresent (baseline),and at 2040 and 2090 decadalaverages under the Hadley model climate scenario.

forest areas of the Mississippi Delta. Forest reduc-tions were further fueled by growth in urban areas,highways,power lines,and related development.Throughout the 1970s,timberland was cleared foragricultural use and for an expanding export mar-ket. In the decade 1982-92,the National ResourcesInventory reports roughly a half million-acre loss(less than 1%) in forestland in the Southeast.

Land use changes in the region are sensitive to anyprojected changes in the value of agricultural andforest lands. Expansion of urban and built-up areasin the Southeast also represents a significantdemand for land,but one that will continue to besmall relative to the total land base. For example,although developed land increased about 27% in thedecade 1982-92,the total land use in this categoryrepresents only 8% of the total land use in theSoutheast. Future land use changes are likely tohave major impacts on things which do not havemarket prices:wildlife and habitat,topsoil,aesthet-ics,pollution of groundwater by agricultural chemi-cals,soil erosion,sedimentation,and loss of wet-lands. The management of these natural resources

152


Hadley Model Southern Har dwoods 2000 Hadley Model Southern Har dwoods 2100

(10 year average)

300 - 800800 - 12001200 - 15001500 - 18001800 - 21002100 - 2500

grams/meter square/year300 - 800800 - 12001200 - 15001500 - 18001800 - 21002100 - 2500

grams/meter square/year

Figure 8. Potential net primary productivity (NPP) of loblolly pineand southern hardwoods simulated by the PnET model with theHadley climate scenario (HADCM2). (Source: USDA Forest Service,Southern Global Change Program). See Color Plate Appendix

Hadley Model Southern Pines 2000 Hadley Model Southern Pines 2100

(10 year average)

Potential Southern Pines and Hardwoods NetPrimary Productivity (NPP)

300 - 800800 - 12001200 - 15001500 - 18001800 - 21002100 - 2500

grams/meter square/year300 - 800800 - 12001200 - 15001500 - 18001800 - 21002100 - 2500

grams/meter square/year

The principal factor influencing the lower increasein pine relative to hardwood productivity by 2090 isthe fact that pines have greater water demands thando hardwoods on a year-round basis. Even with theincreased water use efficiencies associated withincreased atmospheric CO2, the southern pines arelimited by water as evapotranspiration rates increasewith air temperature.

The impact of climate change on the distribution andimpact of forest pests and diseases remains uncer-tain. Southern pine beetles caused over $900 millionin damage to southern US pine forests between 1960and 1990. Higher winter air temperatures willincrease over-wintering beetle larva survival rate,andhigher annual air temperatures will allow the beetlesto produce more generations per year. Both of thesefactors could increase beetle populations. On theone hand, field research has demonstrated that mod-erate drought stress can increase pine resin produc-tion and therefore reduce the colonization successrate of the beetle. However, severe drought stressreduces resin production and greatly increases thesusceptibility of trees to beetle infestation.Insufficient evidence currently exists to predictwhich of these factors will control future beetle pop-ulations and impacts (McNulty, et.al.,1998).

Potential Effects on Timber MarketsThe Sub-Regional Timber Supply (SRTS) model (Abt,et.al.,2000) was developed to link forest inventorymodels with timber market models. The model usesestimated relationships between prices,harvest,andinventory to model market impacts of shifts in sup-ply or demand. The SRTS model uses the spatiallyexplicit and species specific growth changes fromPnET-II to modify inventory accumulation. Thecumulative nature of inventory tends to dampen themarket impacts of the variability found in annualgrowth rates.

This analysis of the future of the forest sector comesat an important turning point in historical trends.Since the turn of the century, southern inventorieshave been increasing due to recovery from exploita-tion in the 1920s and the emergence of industrialforestry in the 1950s. During the last decade,removals of both hardwoods and softwoods haveincreased rapidly and are approximately equal togrowth. This implies that even subtle climate changeimpacts may influence the direction of future inven-tory changes. Overall,the SRTS model (using theHadley climate scenario) indicated that climatechange would more likely favor the Mid-Atlantic overthe East Gulf, and hardwoods over softwoods,andthat growth over the 2000-2020 period would be sig-

nificantly lower than over the 2020-2050 time peri-od. This,along with currently favorable growth/removal ratios in the Mid-Atlantic region,led toshifts northward in pine and hardwood harvest inall model runs.

Beyond the spatial and market adjustments to cli-mate change within the forest sector, land-use feed-backs from the agricultural sector, discussed below,also tend to move inventory and harvest to areaswith comparative advantages. Sensitivity analysis tohigher temperatures (Hadley +2ºC,or 3.6ºF) indicat-ed that the northern shift in inventories and marketsbecame more pronounced and regional pricesincreased as the mid-Gulf region experienced signifi-cant growth declines.

Potential Effects on Forestland AreaAlthough forests and agriculture dominate theSoutheastern landscape,the effect of a changing cli-mate on the relative productivity of these activitiesis just one of many factors that will determine howthe region’s land will be used in the 21st century.Urban and other developed uses,while currently arelatively small part of the regional land base,haveexpanded substantially in the last two decades andare likely to continue to do so in the future. Inrecent years, much of the forest area lost to develop-ment in the region was about equally offset by gainsfrom forest establishment on previously agriculturalland due to the decline in agricultural returns. Thishas tended to stabilize net forest area trends whileexacerbating losses in agricultural land.

Without accounting for climate change, forest area isprojected to remain fairly stable in aggregate to2040. But within the region,there are expected tobe areas with substantial land use change (Figure 9).Urbanized areas in the North Carolina and Georgiapiedmont and southern Florida are projected tocontinue the conversion of forestland and agricul-tural land to developed use,but on a regional basis,these losses are expected to be offset by movementof land from agriculture to forest in other areas,such as the Mississippi delta.

Relative changes in forest and agricultural returnsbrought on by climate change could possiblychange the pattern of stable forest areas in thefuture if, as some scenarios suggest, agriculture canadapt to climate change in some parts of the regionbetter than forests can. Under the Hadley base cli-mate scenario,our model simulations suggest rela-tively little change in the way that land is allocatedbetween forests and other uses between now and2040,though some northern migration of forest area

153


Adaptation OptionsIn general,the biological productivity of southeast-ern forests will likely be enhanced by atmosphericcarbon enrichment,as long as precipitation doesnot decline or air temperature increase soil moisturestress to a level that would offset potential CO2 ben-efits on productivity. The modeling systememployed to analyze regional impacts of climatechange captures some adaptation through its model-ing of economic behavioral responses to changes inthe biophysical conditions in forestry and agricul-ture. For instance,the northward shift in forest pro-ductivity is projected to lead to relative increases inthe proportion of regional timber harvests thatcome from the northern reaches of the region. Thiswill tend to compensate for the biophysical effectsof climate by reducing harvest pressures in themore negatively affected southern parts and increas-ing pressure in the more positively af fected north-ern parts. In addition,landowners are projected toshift land between forests and agriculture in placesand at times where the change in relative productiv-ity warrants it.

Other potential adaptation strategies not modeled inthis Assessment include genetic and silvicultural sys-tem improvements that increase water use efficien-cy or water availability. The knowledge of the roleof fire,hurricanes,droughts,and other natural distur-bances will be important in developing forest man-agement regimes and increasing stand productivityin ways that are sustainable over the long term.Under a hotter, drier climate,an aggressive fire man-agement strategy could prove to be very importantin this region.

Timber productivity associated with increased tem-perature, growing season length,and carbon enrich-ment may be further enhanced by improved geneticselection,bio-engineering,use of marginal agricul-tural land for tree production,and more intensiveforest management. Reduction of air pollutants(e.g.,ozone,nitrogen oxides) could also be animportant strategy for increasing forest productivity.

4. Water Quality Stresses

Current ConditionsThe Southeast has abundant surface waterresources,most of which are intensively managed.Almost all major river systems in the region havebeen dammed and there are few minor streams thathave not been af fected by landscape alteration,channelization,surface or ground water with-drawals,or other human activities. Based on 50-yearor longer streamflow records at 395 stations,Lins154


Timberland Acreage Shift

Figure 9 (a). Changes in land use based on timberland acreageshift for 1993-2040: baseline without climate change. Forestlandlosses are projected in the more urbanized areas of the Southeast,from northern Virginia through the Georgia piedmont and southernFlorida. The movement of land from agriculture to forest is project-ed in many parts of the mid-South. See Color Plate Appendix

Figure 9(b). Timberland acreage shifts by 2040 due to Hadley cli-mate change. In 2040, forestland is projected to be slightly higherwith Hadley base climate change than without climate change insome of the northern reaches of the Southeast, but slightly lowerunder climate change in parts of the deep South. Year 2040 landallocation effects in most of the region are fairly neutral. (Source:North Carolina State University, Department of Forestry; ResearchTriangle Institute, Center for Economics Research). See Color PlateAppendix

is expected (Figure 9). However, sensitivity analysisreveals that substantial variation from the Hadleybase scenario (e.g.,+2 or 4ºC,3.6ºF or 7.2ºF) wouldlikely have more dramatic effects on land allocationand lead to larger net losses in forest area. Theseprojections should be evaluated with caution,how-ever, as the land use model has employed more lim-ited information on the sensitivity of agriculturaleconomic returns to climate change than on on eco-nomic returns to forestry.

5% - 25% Decline<5% Change5% - 25% Increase

>25% Decline5% - 25% Decline<5% Change5% - 25% Increase

and Slack (1999) found that the conterminous US isgetting wetter. Exceptions were noted in parts ofthe Southeast and the Pacific Northwest,wheresome stream gauges showed a decrease in minimumdaily discharge. Parts of Florida and Georgia appearto be experiencing a trend towards decreased mini-mum flows,while the Lower Mississippi River Valleystations showed an increase in both annual mediandaily and annual minimum daily discharge (Lins andSlack,1999). In Louisiana,when Keim and others(1995) modeled historical streamflow based on pre-cipitaion,streamflow per unit drainage areaincreased significantly since 1900.

Potential Impacts of Climate Change Changes in climate that result in decreased runoffduring early summer generally reduce water qualityin the Southeast (Mulholland,et.al.,1997). Summerlow flows occur when water quality (particularlydissolved oxygen) of many southeastern streamsand rivers is at its lowest (Meyer, 1992). Reduceddissolved oxygen during summer months can resultin massive fish kills and harmful algal blooms inboth coastal and inland waters.

An assessment of southeastern water quality associ-ated with changes in climate was conducted usingEPA’s GIS-based BASINS model to evaluate currentand future water quality conditions under bothmean and extreme hydrologic conditions in theSoutheast. Analyses were conducted for each of theUS Geological Survey’s eight-digit Hydrologic UnitCodes (HUCs). Water quality indices included dis-solved oxygen,nitrogen and nitrates,and pH. Theassessment included three steps:identification ofbasins with current and potential water qualityproblems,prediction of general change in streamflow conditions under scenarios of future climate,and re-evaluation of affected basins using the Hadleyclimate model scenarios for 2030 and 2100.

While water quality problems across the Southeastare not critical under current conditions,qualityattainment status is not met in several cases duringthe majority of the year, and can become criticalunder extreme low flow conditions during someportions of the year. Stresses on the water quality ofthe Southeast appear to be associated with intensiveagricultural practices,urban development,coastalprocesses,and possibly mining activities. As mightbe expected,the impacts of these stresses appear tobe more frequent during extreme conditions,proba-bly associated with dry weather. Analysis of the cur-rent status (based on 1990-97 observations) of thewatersheds in the Southeast for dissolved oxygenrevealed few problems under average conditions.

However, it must be recognized that because theBASINS database is indexed to major USGS hydro-logic units,it necessarily consists of observations ofconditions on the larger streams in the regions.Thus,smaller tributaries may exhibit water qualitydegradation that was not apparent at the larger scaleof this analysis.

The analysis suggested that only scattered HUCwatersheds in a few states currently exhibit dis-solved oxygen (DO) conditions below, or nearlybelow the recommended 5 mg/l during averageconditions. However, dissolved oxygen problemsunder extreme low flows do arise at a few locationsin most,if not all,states. Nitrate levels in streams inthe Southeast are used as an indication of nutrientcontent in these HUCs. While some nutrients areessential for ecosystem health, excessive levels canresult in harmful conditions such as algal blooms,which can negatively impact DO levels. The currentstatus (1990-97) based on observations for totalaverage nitrate nitrogen content for streams of theSoutheast reveals that many exhibit levels above 0.5mg/l and in some cases above 4 to 5 mg/l,which is3- to 4-times higher than levels common in mostsoutheastern streams.

One interesting observation is that streams that cur-rently exhibit low dissolved oxygen levels do notcorrespond to those basins where nutrient levelsappear to be high. However, in both cases,theproblems are most prevalent in watersheds withintensive forestry or agricultural operations.

Climate scenarios for the southeastern US providecontrasting results in terms of temperature and pre-cipitation estimates over the region,so that in somecases conditions may improve while in others theymay degrade. The Canadian model results show lit-tle overall change until 2030, followed by drierweather in most of the region over the next seventyyears. On the other hand,the Hadley Centre modelpredicts a slight decrease in overall precipitationover the region during the next 30 years,afterwhich precipitation increases significantly. TheHadley model results also show significantlydecreased precipitation during the first six monthsof the year with rainfall returning to normal,or nearnormal, for the last six months,particularly by theend of the centur y. These results are particularlystriking for the immediate Gulf Coast region andindicate that this area may be exceptionally vulnera-ble to degraded conditions. Intensive agriculturalactivity including disking and planting in the earlyspring, fertilizer application in the late spring/earlysummer, and harvesting in the fall may significantly

155


5. Threats to Coastal Areas

Current ConditionsFew regions have the combination of special charac -teristics and vulnerabilities found in southeasterncoastal areas. The interaction of sea-level rise,storms,beach erosion,subsidence,salt water intru-sion,urban development,and human populationgrowth,and shifts in the transition zone where landmeets ocean,creates conditions for potential adverseeffects on the largest segments of the southeasternpopulation. Large cities located in the coastal zone(such as Houston,Charleston,and New Orleans)already suffer frequent and severe flood damages.

Potential Impacts of Climate ChangeSea-level rise is regarded as one of the more certainconsequences of increased global temperature,andsea level has been rising gradually over the past15,000 years. Globally, average sea level rose 4 to 8inches (10 to 20 cm) during the past 100 years andthis average rate is projected to accelerate 2 to 5-foldover the next 100 years (IPCC,1998). Parts of theCity of New Orleans that are presently 7 feet belowmean sea level may be 10 or more feet below sealevel by 2100,due to a combination of rising sealevel and subsidence of the land surface.

Low-lying marshes and barrier islands of the south-eastern coastal margin are considered particularlyvulnerable to sea-level rise,but all are not equallyvulnerable. Cahoon and others (1998) found thatsome Gulf coastal marshes and one mangrove site insouth Florida are being gradually submerged becausethey do not accumulate sediment quickly enough tokeep up with present rates of sea-level rise. Incoastal Louisiana,landforms created by MississippiRiver sediment deposition over the past 8,000 yearsare naturally de-watering and compacting. As sealevel rises,inundation or displacement of coastalwetlands and barrier islands is occurring. Theimpacts of subsidence (the lowering of the land rela-tive to sea level) are aggravated by human activitiessuch as levee construction along the MississippiRiver, ground water withdrawals,and canal dredgingthough marshes,passes,and barrier islands. Changesin tidal amplitude have caused salt-water intrusioninto many formerly fresh and brackish water habi-tats. Roughly one million acres of south Louisianawetlands have been converted to open water since1940,and Louisiana’s barrier islands have eroded totwo-thirds of the size they were in 1900.

exacerbate water quality conditions during this period.

Preliminary hydrologic analyses based on the Hadleyscenario suggest that streamflow in the Southeast(particularly along the Gulf Coast) may decline by asmuch as 10% during the early summer months overthe next 30 years (Cruise,et.al.,1999;Ritschard,et.al.,1999). These results lead to the conclusions thatwater quality conditions may become critical duringmore frequent periods of extreme low flow.Correlation of the hydrologic analyses with the landuse in basins where water quality problems alreadyexist suggests that the problems may be most acutein areas of intensive agricultural activity, in coastalareas,or near coastal streams (Cruise et al.,1999).

Many of the basins with high nitrate levels form theboundary between two states. The Chattahoocheeboundary between Georgia and Alabama and theTombigbee boundary between portions of Alabamaand Mississippi are two outstanding examples. TheHadley scenario suggests decreased water availabilitythroughout much of this region over the next 50years. As streamflow and soil moisture decrease,intensity of fertilizer application may increase andirrigation needs may become critical. These issueswould likely lead to intense competition for scarcewater resources and conflicts between these statesover runoff treatment and water quality.

Water quality is also a concern in nearshore marineenvironments. Both the Canadian and Hadley cli-mate models suggest an increase in rainfall in theUpper Mississippi Valley (see Great Plains andMidwest chapters). A large (8,000 to 18,000 km2

during 1985-1997) zone of oxygen-depleted (hypox-ic) coastal waters is found in the north-central Gulfof Mexico and is influenced in its timing,duration,and extent by Mississippi River discharge and nutri-ent flux (Rabalais,et.al.,1999; Justic et al.,1997).Nitrate delivered from the Mississippi Basin to theGulf of Mexico,principally from non-point agricul-tural sources,is now about three times larger than itwas 30 years ago as a result of increases in nutrientloading per unit discharge (Goolsby, et.al.,1999).Hypoxia,which is most prevalent in the lower watercolumn,can adversely affect marine life and is agrowing concern to those who harvest and manageGulf fisheries. An increase in Upper MississippiBasin streamflow, where the majority of the nitrogenand phosphorus loading occurs,portends anincrease in the hypoxic zone offshore.

156


If sea-level rise accelerates during the 21st centuryas predicted by the Intergovernmental Panel onClimate Change (IPCC),many other southeasterncoastal areas will experience shoreline retreat andcoastal land loss. Under the IPCC’s “best estimate”of average global sea-level rise over the next 100years,the Big Bend area of the Florida Gulf coastwill likely undergo extensive losses of salt marshand coastal forest (Doyle,1998, Figure 11). Since1980,losses of coastal forests in parts of Florida,South Carolina,and Louisiana have been attributedto salt water intrusion and/or subsidence. Since1991,landowners and public land managers inFlorida have observed extensive die-offs of Sabalpalm along a 40-mile stretch of coast between CedarKey and Homosassa Springs. Williams and others(1999) attribute the forest decline to salt waterintrusion associated with sea-level rise. Since 1852,when the first topographic charts were prepared ofthis region,high tidal flood elevations haveincreased approximately 12 inches.

Rising sea levels due to climate warming may alsoaffect estuaries and aquatic plant communities. Sea-level rise reduces the amount of light reaching sea-grass beds (light penetration decreases exponential -ly with water depth),thereby reducing growthrates. Some marine grass beds may be eliminatedbecause their shoreward migration is impeded byshoreline construction and armoring (Short andNeckles,1999). Increased tidal range associatedwith sea-level rise may have deleterious effects onestuarine and fresh water submerged aquatic plantsby altering both salinity and water depth. (SeeCompounding Stresses on Major Estuaries and Baysin the Northeast chapter for a discussion of sea-levelrise effects on salinity in the Chesapeake Bay.)

Coastal Wetland Vulnerability Thresholds at which sea-level rise results in coastalwetland loss vary among sites due to differences inrates of vertical accretion (sediment build up) andlocal subsidence or uplift processes. Cahoon andothers (1998) estimated the potential for submer-gence of 10 southeastern wetlands by simultaneous-ly measuring surficial sediment accretion and soilsurface elevation changes,and then comparingthese rates to observed sea-level change from tidegauges. Three of the 10 sites are experiencing a netelevation deficit relative to sea-level rise. The othersites are presently accumulating enough sediment tokeep pace with sea-level rise. If sea-level rise wereto accelerate 4-fold,the Oyster Bayou site would besubmerged by about the year 2045. The OysterBayou site would not be submerged if sea-level riseincreases 3-fold or less,unless the site is impactedby a hurricane or other disturbance. Both long-termprocesses (e.g.,accretion,compaction,and decom-position) and episodic events (e.g.,hurricanes)affect the threshold at which coastal wetlands aresubmerged by sea-level rise.(See Figure 10)

157


Figure 10. Relationship of vertical accretion and marsh surface elevation change withlocal relative sea level-rise for sites located in a) low salt marsh, b) high salt marsh,c) mangrove forest and d) brackish marsh. The diagonal line indicates parity betweenaccretion or elevation change and sea-level rise. (source: Cahoon et al., 1998) SeeColor Plate Appendix

Sea Level Rise and Marsh Changes

number of deaths attributed to hurricanes declineddramatically since the 1950s (Pielke and Pielke,1997).

Adaptation OptionsThere are few practical options for protectingcoastal communities and ecosystems as a wholefrom increased temperature, changes in precipita-tion,or rapidly rising sea level. Still,a variety ofmanagement measures could be applied on a site-by-site basis to increase the resiliency of specificcommunities and ecosystems or to reduce or partial-ly compensate for impacts. Many of these measurescould be justified based solely upon non-climatethreats to coastal regions. For example,increasedprotection for existing coastal wetlands and removalof other stresses (such as dredge-and-fill activitiesand water pollution) may not only reduce the sensi-tivity of coastal communities, wetlands,and barrierislands to small changes in average sea level but alsoachieve broader conservation goals (Burkett andKusler, 2000).

Other no-risk measures for achieving broader objec-tives and reducing climate change impacts include:limiting construction in areas where coastal wet-lands may be displaced as sea level rises;installingsediment diversions for dams;linking presently frag-mented wetlands and waters to provide the corri-dors needed for plant and animal migration;usingwater control structures for some wetlands toenhance particular functions and address decreasedprecipitation and/or increased evaporation;increas-ing management programs for invasive species con-trol;and implementing various coastal restorationmeasures (Burkett and Kusler, 2000).

ADDITIONAL ISSUESSix additional climate-related issues for theSoutheast region are:

• Climate Model Limitations: Existing general cir-culation models cannot adequately resolve somecomponents of climate or certain geographic ortopographic features that are important becauseof their interaction with regional climate fea-tures. For coastal regions, much uncertaintyexists about the effects of global climate changeand variability on tropical storms,the mostimportant natural hazard affecting regional vul-nerability. Effects of climate change on areas ofhurricane origination and threshold for hurri-cane formation,intensity, frequency, and tracksare poorly understood.

Storm surge is intensified as sea level rises and natu-ral coastal defenses deteriorate. It is important tonote that even if there is no significant climate-change-driven increase in the frequency or intensityof Atlantic hurricanes,these storms will be moredamaging when making landfall on coastal regionsas sea level rises and coastal landforms erode. SouthAtlantic and Gulf coastal populations increased 15%between 1980 and 1993. During this period,thevalue of insured coastal property in the USincreased by 179%(Insurance Institute for PropertyLoss Reduction and Insurance Research Council,1995). Florida topped the list of coastal states withpotential hurricane damages,with $872 billion ininsured properties. Insurance Research Councildemographic projections suggest that the number ofpersons living in the most hurricane prone countieswill have increased to 73 million by 2010,a dou-bling since 1995 (Insurance Institute for PropertyLoss Reduction and Insurance Research Council,1995).

During the past 30 years,the Gulf of Mexico hasseen a decrease in the number of hurricanes makinglandfall. Hurricanes Andrew, Hugo,and Brett are theonly Category 4 storms to make landfall since 1969,but since 1900,35% of all hurricanes hit Florida andalong the middle Gulf Coast,and 50% or more of allhurricanes making landfall are Category 3 or higher(Heinz Center, 1999). Property losses due to hurri-canes increased from less than $5 billion per decadebetween 1900 and 1940 to about $15 billion perdecade during the 1960s to 1980s;however, the158


Figure 11. Changes in Florida’s Big Bend region forest, marshes, and open water underIPCC (1998) sea-level rise scenarios. (Source: Doyle, 1998). See Color Plate Appendix

Changes in Florida’s Big Bend

38 in (95 cm) - High Estimate

20 in (50 cm) - Mid Estimate

6 in (15 cm) - Low Estimate

159


issues,traffic patterns,and evacuation/shelter

infrastructure are all important areas for potential

mitigation and future research. The potential

impacts of climate change on oil,natural gas,and

navigation infrastructure is also of concern in the

region.

CRUCIAL UNKNOWNS ANDRESEARCH NEEDS

Precipitation Uncertainties

The Southeast is the only region for which currentclimate models simulate large and opposing changesin precipitation patterns over the next 100 years.The range of differences is so great that it is difficultto state with any degree of confidence that precipi-tation will increase or decrease in the Southeastover the next 30-100 years as atmospheric CO2increases. Until climate models are improved (oruntil there is a way to validate and compare theaccuracy of the existing models),people in thisregion must consider a wide range of potentialfuture changes in soil moisture and runoff, as wehave in this Assessment.

Human Health

There are serious human health concerns related tothe plausible increases in maximum temperature forthis region,particularly among lower income house-holds. Similarly, the flood prone nature of coastalcounties from the Carolinas to Texas could have sig-nificant human health implications in addition to theeconomic losses discussed earlier. Approaches formodeling human health effects should include theseaspects of the population in addition to the climaticscience. Water quality degradation could possiblybecome a more serious human health problem in theregion;improvements in our capability to modelstreamflow and water quality in both inland andcoastal waters are needed.

Agriculture and Forestry

Our current understanding of the potential conse-quences of climate change on agriculture in theSoutheast is focused on a few key row crops. In thefuture,it will be important to include the affects ofclimate variations on other high value crops (e.g.,cit-rus and vegetables) and on animal management prac-tices. Furthermore,the role of climate on pests andpest management systems needs to be included infuture assessments. It will also be necessary to

• Water Resources: Fresh water plays an importantrole in many sectors including coastal resources,health, agriculture,estuarine fisheries,andforestry. Competing demands from urban devel-opment, agriculture,and recreation for alreadystressed ground water systems would likely beexacerbated by changes in precipitation and saltwater intrusion due to sea-level rise.

• Impacts on Coastal Ecosystems and Services:Sea-level rise, changes in fresh water delivery tocoastal estuaries,and increased atmospheric tem-perature and CO2 all portend changes in thestructure and function of coastal and estuarinesystems. Losses of coastal marshes and sub-merged aquatic vegetation will have impacts athigher trophic levels. Gulf coastal states current-ly produce most of the Nation’s shrimp, oysters,and crabs,and each of these estuarine fisheries isdependent upon the primary productivity ofcoastal ecosystems.

• Health Issues related to Water Quality: Theeffects on surface waters of changes in precipita-tion have important health implications in theregion. Increased precipitation promotes thetransportation of bacteria as well as otherpathogens and contaminants by surface watersthroughout the region. Health consequencesmay range from shellfish infections transmittedto humans to ground water contamination asso-ciated with saltwater intrusion.

• Socioeconomic and Insurance Issues: In the ulti-mate analysis,the issue of climate change and theneed for an assessment of potential conse-quences will be relevant to the degree that it isplaced on a human scale. To this end,the poten-tial societal impacts of climate change must beidentified and understood. Insurance exposureand/or the insurability of coastal and island facili-ties are issues that should be examined in thecontext of climate change and variability.

• Urban Issues: A distinctive characteristic of

southeastern coastal regions is their current high

level of urbanization and rate of population

growth,which could possibly be affected by

changes in climate that are presented in this

Assessment. The urban environment will have its

own responses to the impacts of climate change

and variability. While responses will be driven by

stakeholder and policymaker decisions,they

need to be evaluated and understood within the

framework of different potential regional scenar-

ios of climate change consequences. Design and

construction factors,building code issues,infra-

structure and lifelines,energy use,structural vul-

nerability to natural hazards,land use and zoning

Aquatic and Coastal Resources

If precipitation patterns continue to change on ascale similar to that observed over the past 100years,many southeastern aquatic ecosystems,includ-ing estuaries,will be affected by changes in stream-flow. There are several additional unknown variablescorresponding to future conditions that might affectthe quality of southeastern water resources. They fallinto two categories:future pollutant loadings (naturaland anthropogenic) and biophysical reactions. Thepollutant loadings,both point and non-point,will bedirectly related to changes in land use activity includ-ing the presence of confined animal systems andgrowing population centers. Also,atmospheric depo-sition of nitrogen will be tied to continued emissionsof nitrogen oxides (NOx). The biophysical reactionson the land surface that might serve to uptake nitro-gen and other constituents will be associated withland cover conversion and vegetation. Futureresearch programs that include these three criticalunknowns seem crucial to gaining a clearer under-standing of the relationship between variations in cli-mate and water quality in the Southeast.

Quantitative data describing the response of nativesoutheastern plant communities to atmospheric car-bon enrichment, water quality (e.g.,salinity),andchanges in temperature and soil moisture are limitedto a few key species. Moreover, very few studieshave addressed the potential interactions among cli-mate variables and between plant species,and evenfewer studies deal with climate effects at secondaryand higher levels in the food web. Several recentstudies suggest that a number of invasive species willbe favored by climatic change in the Southeast,suchas the freeze-intolerant Chinese tallow, which is nowa serious invader in the Gulf coastal plain. Modelsthat integrate environmental change,species respons-es,and interactions among species are needed todescribe pathways that are likely to alter plant andanimal community structure. Research is also need-ed to determine secondary and higher-order effectson ecosystem goods and services. Carbon sequestra-tion is one ecological function that has been poorlydescribed in this region of abundant wetlands andforests that play potentially significant roles as car-bon sinks.

Research and demonstration projects are needed toidentify and prioritize methods that may be imple-mented to minimize the adverse ecological effects ofclimatic change on native southeastern flora andfauna. Monitoring is needed to evaluate long-termtrends in the abundance and distribution of native

develop, validate,and evaluate new technologycapabilities such as new genetic varieties that willhelp farmers cope with any changes in future cli-mate. The effects of biotechnology (e.g.,transgeniccrops) on future agricultural productivity in thisregion,including the benefits of using less fertilizer,pesticides,or water, need to be evaluated in light ofplausible climate scenarios. Because the incorpora-tion of new climate-related technological capabili-ties into agriculture is relatively new and yetunproven,future pilot studies should explore thecommunication of such information to the agricul-ture community.

The potential effects of climate change on agricul-tural prices are an equally complex interaction ofphysical effects and managerial responses world-wide. Spatial equilibrium economic models thatwould address these market issues require that theinformation from all regions be reasonably similar.In the case of climate change,detailed informationfrom one region set against very general informationfrom many important competing crop growingareas would not provide a consistent framework forunderstanding worldwide response in the agricul-ture sector. Thus,it would be very useful to investi-gate in greater detail climate-induced productioneffects in major international crop areas to integratesuch farm management results from importantgrowing areas worldwide to address potential cli-mate-induced price effects.

Although extensive laboratory and field research hasbeen completed on the individual impacts of chang-ing air temperature,precipitation,ozone,carbondioxide,and nitrogen availability on forest produc-tivity, water use,and carbon sequestration,there isstill little understanding of the synergistic impacts ofenvironmental change on southern forests. Fieldexperiments with multiple treatment factors (e.g.,variables) are quite costly, and there are scalingproblems associated with laboratory experiments.Therefore,improved development,testing,and vali-dation of integrated stress impacts through comput-er modeling are crucial future research needs.Models can provide a mechanism for examiningchanging atmospheric and socioeconomic impactson forest structure and function. However, beforeany confidence can be given to such model projec-tions,priority needs to be given to testing,modelverification,and analysis.

160


and non-native species, focusing on species andgroups that are considered highly sensitive to therange of predicted climatic changes in the region(e.g.,amphibians in pine flatwood ponds,benthicinvertebrates in coastal estuaries,and salt-tolerantinvasive aquatic plants that could out-competenative plant species that are important wildlife foodsources).

Extreme Events/DisturbancePatterns

Changes in disturbance patterns (e.g.,hurricanes,floods,droughts) are possibly more significant interms of potential economic losses than longer-termchanges in precipitation or temperature.Ecosystems are also impacted by climate-related dis-turbance. Disturbance is a natural process that,inmany cases,not only structures ecosystems but sus-tains them as well. Our limited understanding ofthe role of disturbance in natural ecosystems andour inability to predict climate extremes is problem-atic for those interested in mitigating the potentialadverse impacts of climate change. Research shouldbe undertaken to examine potential changes in dis-turbance regimes that may be expected as the cli-mate warms and precipitation patterns change.Disturbance topics should not be limited to weatherevents,however. Fire,harmful algal blooms,andinsect outbreaks are ecological disturbances thatmay be heavily influenced by climatic conditions.The ecological effects of these types of disturbancesare difficult to model or predict because they areoften poorly understood. Basic information is need-ed to identify ecological changes that are likely tooccur as the type,frequency, and spatial patterns ofdisturbance are altered as the climate changes.

LITERATURE CITEDAber, J. D.,C.Driscoll,C.A. Federer, R.Lathrop, G. Lovett,J. M.Melillo, P. Steudler, , and J.Vogelmann, A strategyfor the regional analysis of the effects of physical andchemical climate change on biogeochemical cycles innortheastern (U.S.) forests, Ecological Modeling, 67, 37-47,1993.

Abt,R.C., F.W. Cubbage,and G. Pacheco, Forestresource assessment using the subregional timber sup-ply model (SRTS), Forest Products Journal, 50(4),25-33,2000.

Acock,B.,V. R.Reddy, H. F. Hodges, D. N. Baker, and J. M.McKinion,Photosynthetic response of soybeancanopies to full-season carbon enrichment, AgronomyJournal, 77, 942-47,1985.

Allen,L.H.,K. J. Boote, J.W. Jones,A. J. Rowland-Bamford,G. Bowes, D.A.Graetz,K.R.Reddy,Temperature andCO2 effects on rice:1988. US Department ofAgriculture,Carbon Dioxide Research Division,Washington,DC,1989.

Boyer,W. D.,Regeneration of natural longleaf pine for-est, Journal of Forestry, 77, 572-575,1979.

Burkett,V.,and J. Kusler, Climate change: Potentialimpacts and interactions in wetlands of the UnitedStates, Journal of American Water ResourcesAssociation, 36(2),313-320,2000.

Cahoon, D. R., J.W. Day, D. J. Reed,and R.S.Young,Global climate change and sea-level rise: estimating thepotential for submergence of coastal wetlands,inVulnerability of Coastal Wetlands in the SoutheasternUnited States: Climate Change Research Results , USGeological Survey, Biological science reportUSGS/BRD/BSR 1998-0002, pp.21-34,1998.

Cruise, J. F.,A.S.Limaye,and N.Al-Abed,Assessment ofthe impacts of climate change on water quality in thesoutheastern United States, Journal of American WaterResources Association, 35(6),1539-1550,1999.

Currie, D. J.,Energy and large-scale patterns of animal -and plant-species richness, American Naturalist,137(1),27-49,1991.

Dahl,T. E.,Wetland losses in the United States:1780’s to1980’s,US Department of Interior, Fish and WildlifeService,Washington DC, 21 pp.,1990.

161


Insurance Institute for Property Loss Reduction andInsurance Research Council, Community exposure andcommunity protection: Hurricane Andrew’s legacy(IIPLR and IRC).Wheaton,IL (IILPR) and Boston,MA(IRC),1995.

Intergovernmental Panel on Climate Change, TheRegional Impacts of Climate Change: An Assessmentof Vulnerability, Cambridge University Press,UnitedKingdom,514 pp.,1998.

Jones, J.W.,S.S. Jagtap,and K. J. Boote,Climate change:implications of soybean yield and management in theUSA,paper presented at the World Soybean ResearchConference VI,Chicago,Illinois,August 7,1999.

Justic´, D., N. N. Rabalais,and R.E.Turner, Impacts of cli -mate change on net productivity of coastal waters:Implications for carbon budget and hypoxia, ClimateResearch, 8, 225-237,1997.

Keim,B. D., G. E. Faiers, R.A.Muller, J. M.Grymes,andR.V. Rohli,Long-term trends of precipitation and runoffin Louisiana,USA, International Journal ofClimatology, 15, 531-541,1995.

Kiehl, J.T.,Climate change enhanced: Solving theaerosol puzzle,Science, 283(5406),1273-1275,1999.

Legler, D. M.,K. J. Bryant,and J. J. O’Brien,Impact ofENSO-related climate anomalies on crop yields in theUS, Climatic Change, 42(2),351-375,1999.

Lins,H. F.,and J. R.Slack,Streamflow trends in theUnited States, Geophysical Research Letters, 26(2),227-230,1999.

Lo,C. P., D.A.Quattrochi,and J. C.Luvall,Application ofhigh-resolution thermal infrared remote sensing andGIS to assess the urban heat island effect,International Journal of Remote Sensing , 18, 287-304,1997.

Lockwood, J. G.,Is potential evapotranspiration and itsrelationship with actual evapotranspiration sensitive toelevated CO2 levels?, Climatic Change, 41(2),193-212,1999.

McGowan, J.A., D. R.Cayan,and L.M.Dorman,Climate-ocean variability and ecosystem response in the north-east Pacific, Science, 281, 210-217,1998.

Doyle,T.W.,Modeling global change effects on coastalforests,in Vulnerability of Coastal Wetlands in theSoutheastern United States: Climate ChangeResearch Results, US Geological Survey, BiologicalScience Report USGS/BRD/BSR 1998-0002,Lafayette,Louisiana,pp.67-80,1998.

Eamus, D.,and P.G. Jarvis,The direct effects of increasesin the global atmospheric CO2 concentration on natu-ral and commercial temperate trees and forests,Advances in Ecological Research , 19, 1-55,1989.

Felzer, B.,and P. Heard,Hydrological implications ofGCM results for the US national assessment, Journal ofAmerican Water Resources Association, 35(6),1327-1339,1999.

Glasmeier,A.,The history of Central and SouthernAppalachia: A socioeconomic analysis,paper presentedat the Central and Southern Workshop on ClimateVariability and Change,Morgantown,West Virginia,May27-28,1998.

Goolsby, D.A.,W.A.Battaglin, G. B.Lawrence,R.S.Artz,B.T.Aulenbach,R. P. Hooper, D. R. Keeney, and G. J.Stensland, Flux and source of nutrients in theMississippi-Atchafalaya River Basin, In Review,WhiteHouse Office of Science and Technology Policy,Committee on Environment and Natural Resources,Hypoxia Working Group,May 1999, pp.13-37,2000.

Harcombe, P.A.,R.B.W. Hall, J. S.Glitzenstein,E.S.Cook,P. Krusic,and M.Fulton, Sensitivity of Gulf Coastforests to climate change,in Vulnerability of CoastalWetlands in the Southeastern United States: ClimateChange Research Results, US Geological Survey,Biological Science Report,USGS/BRD/BSR 1998-0002,Lafayette,Louisiana,pp.47-67,1998.

Hatch, U.,S. Jagtap, J. Jones,and M.Lamb, Potentialeffects of climate change on agricultural water use,Journal of American Water Resources Association,35(6),1551-1562,1999.

Hefner, J. M.,Southeast wetlands:Status and trends,mid-1970’s to mid-1980’s,US Department of Interior, Fishand Wildlife Service,Atlanta,Georgia,32 pp.,1994.

Heinz Center, The Hidden Costs Of Coastal Hazards:Implications For Risk Assessment And Mitigation,Island Press,Washington,DC, 220 pp.,1999.

162


McNulty, S. G., P. L.Lorio,Jr.,M. P.Ayres, and J. D. Reeve,Predictions of Southern Pine Beetle populations usinga forest ecosystem model,in The Productivity andSustainability of Southern Forest Ecosystems in aChanging Environment, edited by R.Mickler and S.Fox, Springer Publishing,pp.617-634,1998.

McNulty, S. G., J. M.Vose,W.T. Swank, J. D.Aber, and C.A.Federer, Landscape scale forest modeling: Data basedevelopment,model predictions,and validation using aGIS, Climate Research , 4, 223-231,1994.

Mearns,L. O., F. Giorgio,L.McDaniel,and C.Shields,Analysis of daily variability of precipitation in a nestedregional climate model: Comparison with observationsand doubled CO2 results, Global Planetary Change,10, 55-78,1995.

Meyer, J. L.,Seasonal patterns of water quality in black-water rivers of the coastal plain,southeastern UnitedStates,in Water Quality in North American RiverSystems, edited by C. D. Becker and D.A.Neitzel,Battelle Press,Columbus,Ohio,pp.249-276,1992.

Mulholland, P. J., G. R.Best,C.C.Coutant, G. M.Hornberger, J. L.Meyer, P. J. Robinson, J. R.Stenberg,R.E.Turner, F.Vera-Herra,and R. G.Wetzel,Effects of climatechange on freshwater ecosystems of the southeasternUnited States and the Gulf Coast of Mexico,Hydrological Processes, 11, 949-970,1997.

National Climatic Data Center, Billion dollar US weatherdisasters,1980-1999, PDF version,www.ncdc.noaa.gov/ol/reports/billions.html.July 20,1999.

National Council of the Paper Industry for Air andStream Improvement (NCASI),Global change and for-est responses: theoretical basis,projections,and uncer-tainties,NCASI,New York,Technical Bulletin No.690,pp.11-18,1995.

Nijs,I.,I.Impens,and T. Behacghe,Leaf and canopyresponses of Lolium perenne to long-term elevatedatmospheric carbon-dioxide concentration, Planta177, 312-320,1988.

Pielke,Jr.,R.A.,and R.A.Pielke,Sr.,Vulnerability to hur-ricanes along the US Atlantic and Gulf coasts:Considerations of the use of long-term forecasts,inHurricanes: Climate and Socioeconomic Impacts , edit-ed by H. F. Diaz and R.S.Pulwarty, Springer Publishing,New York,147-184,1997.

Quattrochi, D.A., J. C.Luvall,M. G. Estes,Jr.,C. P. Lo,S. Q.Kidder, J. Hafner, H.Taha,R. D. Bornstein,R.R.Gillies,and K. P. Gallo,Project ATLANTA (ATlanta Land useANalysis:Temperature and Air quality): A study of howthe urban landscape af fects meteorology and air qualitythrough time,Preprints, American MeteorologicalSociety, 104-107,1998.

Quattrochi,D.A.,and J. C.Luvall, Urban sprawl andurban pall: Assessing the impacts of Atlanta’s growthon meteorology and air quality using remote sensingand GIS, Geographical Information Systems, 9, 26-33,1999.

Rabalais, N. N.,R.E.Turner, D. Justic´, Q. Dortch,and W.J.Wiseman,Characterization of hypoxia, Topic #1, Gulfof Mexico Hypoxia Assessment, NOAA CoastalProgram Decision Analysis Series, National Oceanicand Atmospheric Administration,Coastal Ocean Office,Silver Spring,Maryland,144 pp.plus 3 appendices,1999.

Ricketts,T. H.,E.Dinerstein, D. M.Olson,and C.Loucks,Who’s where in North America? Patterns of speciesrichness and the utility of indicator taxa for conserva-tion, BioScience, 49(5),369-381,1999.

Ritschard,R.L., J. F. Cruise,and L. U. Hatch,Spatial andtemporal analysis of agricultural water requirements inthe Gulf Coast of the United States, Journal ofAmerican Water Resources Association, 35(6),1585-1596,1999.

Rosenberg, N. J.,R.C.Izaurralde,M.Tiscareño-Lopéz, D.Legler, R.Srinivasan,R.A.Brown,and R. D. Sands,Sensitivity of North American agriculture to ENSO-based climate scenarios and their socio-economic con-sequences: Modeling in an integrated assessmentframework, Pacific Northwest National Laboratory,Seattle,Washington,146 pp.,1997.

Short, F.T.,and H.A.Neckles,The effects of global cli-mate change on seagrasses, Aquatic Botany, 63, 169-196,1999.

Southern Oxidants Study Team. 1995. The state of theSouthern Oxidants Study: Policy-relevantFindings inozone pollution research,1988-1994. SouthernOxidants Study, Raleigh,North Carolina,94 pp.

Timmermann,A., J. Oberhuber,A.Bacher, M.Esch,M.Latif, and E.Roeckner, Increased El Niño frequency in aclimate model forced by future greenhouse warming,Nature, 398, 694-697,1999.

163


ACKNOWLEDGMENTSDedicationThis chapter is dedicated to Dr. Ronald Ritschard,a good friend, colleague, and scientist, withoutwhose efforts this Assessment would not have beenpossible. Ron will be deeply missed.

Many of the materials for this chapterare based on contributions from participants on and

those working with the

Central and Southern Appalachians Workshop TeamWilliam T. Peterjohn (PI),West Virginia UniversityRichard Birdsey, USDA Forest ServiceAmy Glasmeier, Pennsylvania State UniversitySteven McNulty, USDA Forest ServiceTrina Karolchik Wafle,West Virginia University

Gulf Coast Workshop and Assessment TeamZhu Hua Ning*,Southern University and A & M

CollegeKamran Abdollahi*,Southern University and A & M

CollegeVirginia Burkett,USGS National Wetlands Research

CenterJames Chambers,Louisiana State UniversityDavid Sailor,Tulane UniversityJay Grymes,Southern Regional Climate CenterPaul Epstein,Harvard UniversityMichael Slimak,US Environmental Protection Agency

Southeast Workshop and Assessment TeamsRon Ritschard*,University of Alabama – HuntsvilleJames O’Brien*,Florida State UniversityJames Cruise*,University of Alabama – HuntsvilleRobert Abt,North Carolina State UniversityUpton Hatch,Auburn UniversityShrikant Jagtap,University of FloridaJames Jones,University of FloridaSteven McNulty, USDA Forest ServiceBrian Murray, Research Triangle Institute

* signifies Assessment team chairs/co-chairs

Trenberth,K.E.,Conceptual framework for changes ofextreme of the hydrological cycle with climate change,Climatic Change, 42, 327-339,1999.

Tsuji, G., G. Hoogenboom, P.Thornton,(Eds.),Understanding Options for Agricultural Production,Kluwer Academic Publishers,399 pp.,1998.

US Bureau of the Census,County and City Data Book,US Government Printing Office,Washington,DC,GPO#003-024-08753-7,1994.

US Department of Agriculture,Census of Agriculture,National Agricultural Statistics Service,Washington,DC,1994.

US Department of Agriculture,Census of Agriculture,National Agricultural Statistics Service,Washington,DC,1996.

US Department of Agriculture, Census of Agriculture,National Agricultural Statistics Service,Montgomery,Alabama,1999.

US Forest Service,The South’s fourth forest:alternativesfor the future, US Department of Agriculture, ForestService,Washington DC, Forest Resource Report No.24, 512 pp.,1998.

Williams,K.,K.C.Ewel,R. P. Stumpf, F. E.Putz,and T.W.Workman,Sea-level rise and coastal forest retreat onthe west coast of Florida, Ecology, 80(6), 2045-2063,1999.

164


southeast mega-region (chapter 5) from the foundation ... · southeastern united states physical...

Documents