An Assessment of Ozone Effects in the Southern Appalachians Using a Multi-Stakeholder Process

 

Paper #463

 

N. S. Nicholas

 

Tennessee Valley Authority, 129 Pine Road. Norris, Tennessee  37828

 

P. F. Brewer

 

Southern Appalachian Mountains Initiative, 59 Woodfin Place, Asheville, NC  28801

 

D. A. Weinstein

 

Boyce Thompson Institute, Cornell University, Ithaca, New York 14853-1801

 

ABSTRACT

 

The Southern Appalachian Mountain Initiative (SAMI), a regional multi-stakeholder based initiative, was created to identify and recommend emissions strategies to remedy existing and prevent future adverse air quality effects in Southern Appalachia, with particular focus on Class I areas.  Air quality effects being assessed include acid deposition, visibility, and ozone.  The SAMI ozone assessment is being carried out in two phases and the results will be incorporated into an overall integrated assessment that will link emissions, atmospheric transport, exposures, environmental effects, and socioeconomic impacts.

 

Phase I of the ozone assessment was designed to determine indicator plant species and provide guidance on what levels of air quality changes are needed to detect changes in sensitive species response.  Phase I analyses recommended that a seasonal, cumulative, 24-hour, ozone exposure index be used as the most biologically relevant representation of exposures that impact forests.  The literature was reviewed for exposure-response relationships of regional tree species, based on controlled exposure studies with seedlings.  Where at least 10 percent growth loss was measured, Phase I of the assessment compared cumulative exposure levels to ambient exposures now found in the Southern Appalachian Mountains.  Phase I also outlined a number of different approaches for assessing forest response to changes in ozone exposures.

 

Elevated levels of ozone were found to reduce growth rate in some tree species.  Species with faster growth rates tended to have higher rates of ozone uptake and were more sensitive to ozone than slower growing species.  Because of definite species-specific responses to ozone levels, altered resource competition patterns could change succession processes and subsequent forest structure.  Prior to initiating Phase II of the ozone assessment, a demonstration study, using the individual tree behavior model TREGRO and the regional forest growth model ZELIG, was carried out to evaluate a methodology of extrapolating an estimate of forest response to various sustained ozone exposures over a 50 year period.  Phase II of the ozone assessment has been designed and initiated to predict forest response in the years 2010 and 2040 as a function of changes in ozone exposures.  The assessment addresses the entire SAMI geographic domain, but concentrate on the area’s ten Class I areas.

 

 

INTRODUCTION

 

The Southern Appalachian Mountain Initiative (SAMI), a regional multi-stakeholder based initiative, was created to identify and recommend emissions strategies to remedy existing and prevent future adverse air quality effects in Southern Appalachia, with particular focus on Class I areas.  Air quality effects being assessed include acid deposition, visibility, and ozone.  Design of the SAMI ozone assessment has been carried out by the SAMI Effects Subcommittee, a group open to all interested parties.  Active members have included representatives from regulatory agencies, federal land, state agencies, industry, environmental stakeholder organizations, and interested private citizens.  The SAMI ozone assessment was designed to be carried out in two major phases and the results will be incorporated into an overall integrated assessment that will link emissions, atmospheric transport, exposures, environmental effects, and socioeconomic impacts.

 

Phase I of the ozone assessment was designed to determine indicator species and provide guidance on what levels of air quality changes are needed to detect changes in sensitive species response.  The literature was reviewed for visible injury, growth, and physiological response and exposure-response relationships for regional tree species, using controlled exposure studies with seedlings; and where at least 10 percent growth loss was measured, compared cumulative exposure levels to ambient exposures now found in the Southern Appalachian Mountains.  Phase I also outlined a number of different approaches for assessing forest response to changes in ozone exposures.

 

Because of definite species-specific responses to ozone levels, altered resource competition patterns could change succession processes and subsequent forest structure.  Prior initiating Phase II of the ozone assessment, a demonstration study, using the individual tree behavior model TREGRO and the regional forest growth model ZELIG, was carried out to evaluate a methodology of extrapolating an estimate of forest response to various sustained ozone exposures over a 50 year period.  Phase II of the ozone assessment has been initiated and designed to predict forest response as a function of changes in ozone exposures, addressing the entire SAMI geographic domain, but concentrating on the area’s ten Class I areas.

 

 

PHASE I OF THE OZONE ASSESSMENT

 

Phase I of the ozone assessment was designed to determine indicator plant species and provide guidance on what levels of air quality changes are needed to detect changes in sensitive species response.  As a first step of Phase I, a survey of the peer reviewed literature was contracted, examining the effects of ozone on southern Appalachian trees.1, 2   The major findings from the literature covered visible injury, growth and physiological function, exposure-response, models, and assessment methodologies.

 

Visible Injury

 

The SAMI literature review found that tropospheric ozone has been shown to cause foliar injury on sensitive vegetation throughout much of the SAMI region. Foliar injury has been induced following ozone exposures delivered in laboratory, greenhouse, and field open-top chambers.  Foliar injury on hardwood species has been observed as mid-to late-season adaxial stipple, leaf reddening, and early leaf senescence.  Symptoms on conifers are less evident under ambient ozone exposures due to many mimicking symptoms.  However the literature review did not find any clearly defined association demonstrated between foliar injury and growth under natural growing conditions.1,2

 

Growth and Physiological Function

 

Both growth and physiological responses to ozone have been reported for tree species that occur in the SAMI region.  The majority of these investigations were short-term (<one year), under controlled conditions, and were with potted seedlings (< two years in age).  Growth effects due to ozone had been reported for 11 different coniferous species and 17 hardwood species that occur in the SAMI region.  Mature tree responses in the SAMI region have been reported for six species: black cherry (Prunus serotina Ehrh.), red maple (Acer rubrum L.), eastern white pine (Pinus strobus L.), loblolly pine (Pinus taeda L.), northern red oak (Quercus rubrum L.), and yellow-poplar (Liriodendron tulipifera L).  Ranking of species sensitivity to ozone (growth, physiology, and visible injury) is hindered by variation in environmental conditions, ozone exposures, study duration and objectives, tree age, and genetic variation within a species.  Ozone effects can be altered by other environmental and biotic factors such as water status, temperature, light, relative humidity, insects, and diseases.  Given the interaction between ozone and these factors, the authors of the SAMI report were unwilling to suggest that a reduction of 10-20% of ambient ozone levels could result in subsequent, measurable increases in growth and physiological function of forest trees in the SAMI region. 1, 2

 

Exposure-response

 

A review of 1983-1990 ozone data found that, within the southern Appalachian area boundary, ozone monitors usually measured fewer than 40 hours per year in which the hourly average ozone concentration was > 0.10 ppm.  The only year that deviated from this pattern was 1988.1, 2  A combination of the Palmer drought index and an analysis of southern Appalachian ozone data indicated that soil moisture may alter tree growth response to ozone exposures.3  Combination of exposure information with moisture availability and experimental exposure-response data identified areas that may have the greatest potential for possible vegetation effects.1, 3

 

Models

 

The literature review found that several models had been used to simulate growth and/or physiological responses to ozone for trees in the SAMI region.  These included a single tree model (TREGRO), several canopy models on ozone uptake and carbon fixation, a forest succession model (FORET), and a loblolly pine management model (AIRPTAEDA).  Each model had advantages and disadvantages.  For example TREGRO provided good information on ozone effects for a single open-grown tree but did not provide information regarding forest stand growth.  The canopy models all provided good information on ozone uptake and carbon fixation but do not provide information on whole tree responses.  The FORET model provided useful information on ozone effects to a forest stand but needs to be parameterized for ozone specifically and assumes that impacts to be equal within a species.  AIRPTAEDA was determined to have limited use for an assessment within the SAMI region since loblolly pine is a minor component of the region.1

 

Assessment Methodologies

 

Three assessment methodologies were considered for use in detecting risk due to ozone for different forest types in the SAMI region.1  The first methodology involved the use of kriging of ozone concentrations combined with data obtained from experimental results for nine tree species growing in the SAMI region to determine areas of possible concern.3  Since environmental factors such as soil moisture may influence tree response to ozone, the Palmer drought index was used to further subdivide the areas of concern.  The second methodology was an application of Geographical Information System (GIS) combined with estimated ozone exposures, other abiotic environmental variables, species distributions, simulation models and experimental data to predict areas with a likelihood of occurrence of adverse effects (i.e., biomass reductions).4,5  The third methodology focused on loblolly pine where the response of loblolly variations in rainfall and ozone exposures would be scaled up from a seedling to a forest stand and then long-term effects on loblolly pine productivity.6

 

A second Phase I report also evaluated different approaches for assessing forest response to changes in ozone exposures considered in the first report, adding information on unpublished work by M. Fulton, using the TREGRO model and regional environmental databases as an assessment tool.7  The response of northern red oak trees to ozone and drought stress was investigated using input data representing conditions over three years (1993-1995) at sites across the southeastern United States.  Overall growth rates in the absence of ozone showed wide variation between sites and years, and this variation was also seen in responses to ozone exposure.  In general, trees growing on sites and years with high evapotranspiration had high growth rates, high ozone uptake at any given exposure level, and large decreases in net photosynthesis in response to ozone exposure.  Trees grown in sites and years with low evapotranspiration responded less to ozone exposure in absolute terms but, because growth rates were already low, relative responses were comparable.  Variations in simulated responses were a function of ozone uptake, site differences, and climatic and edaphic conditions.

 

Phase I Recommendations

 

The primary reason a second Phase I report was commissioned was to evaluate which biologically relevant ozone exposure statistics could be used to describe a biological response.  Analyses recommended that a seasonal, cumulative, 24-hour, ozone exposure index be used as the most biologically relevant representation of exposures that impact forests.7  The literature was reviewed for exposure-response relationships for regional tree species, using controlled exposure studies with seedlings; and where at least 10 percent growth loss was measured, compared cumulative exposure levels to ambient exposures now found in the Southern Appalachian Mountains. Elevated levels of ozone were found to reduce growth rate in some tree species.  Species with faster growth rates tended to have higher rates of ozone uptake and were more sensitive to ozone than slower growing species.

 

The authors of the Phase I reports did not recommend any of the assessment technologies reviewed but did suggest that modifications might be necessary for the SAMI integrated assessment approach.1,7  They also cautioned that tremendous variability exists within natural systems and modeling ozone-induced exposure/responses across the geographic range of a species in the SAMI region posed a formidable task.  The influence of micro-site factors was considered to be most important in controlling response.  Available soil moisture was determined to be of  greatest single importance in controlling ozone uptake.1  Additional recommendations the consideration of the importance of taking into account the relative influences of insect pests, biotic pathogens, abiotic stressors (other than ozone), inter- and intra-specific competition for resources in contributing to the changes in forest health and productivity.  Otherwise an assessment might provide an unrealistic scenario of the relative importance of ambient ozone exposures to forests in the SAMI region.

 

 

DEMONSTRATION STUDY

 

One of the lessons from the Phase I ozone assessment was that due to species-specific responses to ozone levels, altered resource competition patterns could change succession processes and subsequent forest structure.  The SAMI Effects Subcommittee recognized that there was no previously completed technology example of a regional assessment that would address all of the subcommittee’s issues.  Therefore, prior to initiating Phase II of the ozone assessment, the SAMI Effects Subcommittee decided to work together with EPRI overseeing a demonstration study that used the individual tree behavior model TREGRO and the regional forest growth model ZELIG.  This previously planned EPRI-funded study by D. Weinstein was carried out to evaluate a methodology to assess the impacts of ozone on forest productivity.  The project objective was to understand the response of selected hardwood tree species growing in mixed forest stands under ambient environmental conditions and to estimate the forest response to various sustained ozone exposures over a 100 year period.  After discussions with the SAMI Effects Subcommittee, EPRI agreed to specify that the targeted study area to be within the SAMI region.  The study was then designed to have its selected tree species (black cherry, yellow-poplar, and red maple), to be ones prevalent in the SAMI forested mountainous region and represented a range of sensitivity to ozone.  The ozone, meteorological, and forest descriptive data used in the demonstration were all supplied from measurements from the Great Smoky Mountains National Park.8, 9 

 

TREGRO Model

 

The demonstration study first used TREGRO to start modeling the response of individual trees to varying ozone levels.  TREGRO is a single-tree model developed to examine plant physiological responses to environmental stresses.10  TREGRO tracks carbon allocation among plant tissues as measure of plant physiological responses.  The model can simulate shifts in carbon allocation, pools of total non-structural carbon, and tree growth rates under different environmental conditions and ozone exposure regimes.  TREGRO has been demonstrated to simulate the same ozone-induced shifts in photosynthate allocation away from roots to shoots as has been observed in field experiments with seedlings.10, 11, 12  TREGRO has also been demonstrated to simulate the responses of mature trees, which can never be thoroughly examined in experimental settings.  TREGRO uses an hourly time step for ozone and meteorological data.

 

ZELIG Model

 

The TREGRO modeled response of individual trees of the three targeted species were then added as an input to ZELIG model runs.  ZELIG is a stand growth model that represents the impact of competition for resources among individuals and species within forest stands, on stand productivity.  ZELIG accepts outputs from TREGRO that identify how expected growth rates under a given set of environmental conditions might be altered by ozone exposure.13  This ozone exposure impact for discreet years is propagated through ZELIG to predict ozone response over decades.

 

Demonstration Study Results

 

Separate TREGRO simulations were conducted on only the three selected tree species to determine their individual responses to different levels of ozone in the absence of competition.  The responses of all three species were simultaneously entered into ZELIG for a set of simulations at three different ozone levels (0.5 ambient, 1.0 ambient, 1.5 ambient), with the assumption that each tree of these three species in the forest would exhibit the corresponding ozone responses.  Because of the absence of available ozone response data for other species in the forest stand, it was assumed that no other forest tree species would show any direct alteration in growth, leaf area development, or drought sensitivity under altered ozone levels.  However, other species might have altered growth rates as an indirect consequence of the direct effect of ozone on black cherry, red maple, and/or yellow-poplar.

 

ZELIG-TREGRO modeling results showed that while the dynamics of the majority of the tree species were not altered, there was an impact on American beech (Fagus grandifolia L.), red maple, and yellow-poplar abundance at ambient ozone levels over 100 years.  At ambient as well as 1.5 x ambient levels, there  was little overall predicted change in the total basal area of the forest; abundance of American beech increased while red maple and yellow-poplar decreased.  The demonstration study was very useful in that it demonstrated an assessment methodology that estimated the magnitude of changes over time at the forest stand level that could be compared at varying ozone exposure levels.

 

 

PHASE II OF THE OZONE ASSESSMENT

 

Once the results from the Ozone Phase I reports and the results from the EPRI funded TREGRO-ZELIG demonstration study, the SAMI Effects Subcommittee began to design Phase II of the assessment.  The Effects Subcommittee reviewed a number of different assessment methods including risk characterization, various modeling approaches, and the critical loads approach.  The decision was made to not designate a specific approach in the Phase II Request For Proposals but to evaluate proposed assessment approaches based on the responsiveness to the Subcommittee objectives, focus, and product expectations. The objectives were to develop a protocol 1) to determine the relationships between ozone exposure, species responses, and modifying environmental influences to project changes in responses of key species to changes in emissions and 2) to apply spatial analysis and other statistical tools to summarize, display,  and compare exposures and forest responses for alternative future emissions strategies.  The primary focus was to be on Class I areas but to also consider the entire SAMI region if possible.  The Subcommittee also needed to be able to compare results across different emissions management strategies.  Additional product expectations were 1) information on forest health responses to ozone that can be used to estimate the socioeconomic benefits of alternative strategies to reduce ozone precursor emissions, 2) spatial display of any phase II results, and 3) consideration of how indicators of forest health might vary as a function of ozone exposure.

 

Following an open call for proposals, the SAMI Effects Subcommittee, by consensus, selected the proposal submitted by D. Weinstein and colleagues.  The selected Phase II approach has been designed to use ozone exposure projections from the atmospheric modeling component of SAMI to drive a multi-model projection of forest responses.  The approach specifically considers two critical processes that will determine how forests will respond to ozone: 1) the ability of trees to absorb ozone-induced injury when faced with co-occurring stresses, such as drought, infertile soils, and excess nitrogen deposition; and 2) the changes in the species composition of forests that will occur when one or more species is weakened by ozone.  First the experimentally-determined effect of ozone on leaf photosynthesis will be extrapolated to its subsequent effect on the growth of individual trees through the use of a single tree physiology model, TREGRO.  Then  the effects on individual trees will be extrapolated to the changes in forest development, by using the forest stand model, ZELIG, to predict changes in forest stand growth and species composition.  Finally effects on stand growth and species composition to the entire region will be extrapolated by linking predicted ozone exposures with predicted dose-response relationships, forest type location, and soil conditions within a Geographical Information System (GIS).  This procedure will be followed for each of the proposed SAMI emissions strategies.

 

 

Phase II Methods

 

Data assembly for the assessment is critical.  Hourly meteorological data (air temperature, humidity, vapor pressure, photosynthetically active radiation, and rainfall) and ozone are required for TREGRO parameterization.  ARC/INFOĆ software will be used to build a geographic information system for the SAMI region, which will be represented by 12 km grid cells.  Data sets for the region will include elevation, climate (precipitation and temperature), ozone, nitrogen availability (as high or low, based on deposition), soil texture (related to high or low water-holding capacity), and forest type distribution.

 

Response of forests to ozone is strongly dependent on the ozone-sensitivity of key species.8,14  Consequently, for this assessment forests of the SAMI region will be stratified into community types (characterized through dominance by key species).  Since the response of ozone response is strongly affected by the water availability, and soil fertility,9 each community type will be sub-stratified into two groups of water holding capacity and into two groups of fertility.  The USDA Forest Service Forest Inventory Analysis (FIA) forest classification15 will be used and only those forest types in this classification dominated by species known to be sensitive to ozone are specifically analyzed.  Table 1 shows the forest types in the FIA classification in the SAMI region, with ozone-sensitive species dominating a given forest type shown in bold.  Forest types that are dominated by containing ozone-sensitive species will be evaluated.

 

Table 1.  Detailed forest classification showing specific FIA forest types15 within each of the three main groups found in the SAMI region, Maple-beech -birch, Oak-hickory, and Conifer-dominated.  Those species shown to be responsive to ozone exposure are indicated in bold.  Community types in shaded cells contain ozone responsive species and will be evaluated.

 

Maple - Beech - Birch

Oak-hickory

Conifer dominated

Sugar maple-Beech- Yellow birch

Chestnut oak

Shortleaf pine

Black cherry

White oak-Red oak-Hickory

Virginia pine

Red maple-northern hardwood

White oak

Pitch pine

Red maple-Upland
Northern red oak

Table-mountain pine

Northern hardwood-Reverting field

Yellow-poplar-white oak-N. red oak
White pine

Mixed northern hardwoods

Sweetgum-tulip-poplar

White pine-Hemlock

 

Mixed central hardwoods

Hemlock

 

 

Balsam (Fraser) fir

 

 

Red spruce-Balsam (Fraser) fir

 

 

White pine-N. red oak-White ash

 

 

Shortleaf pine-Oak

 

 

Virginia pine-Southern red oak

 

The SAMI region will be divided into 12x12 km grid cells and the USDA Forest Service FIA Eastwide database15 used to identify the appropriate forest type to be analyzed in each grid cell of the region.  Each grid cell will be assigned the community type of the nearest FIA data point or, if multiple data points lie within the grid boundaries, each will be given a proportional abundance.  For example, for a given grid cell located in a Class I area the FIA data will be used to estimate the quantity of each potentially sensitive forest type within the cell.  The percentage of fine textured and coarse textured soils within this area will be estimated from the STATSGO (USDA) soils database.  Information on the soil type preferred by each forest type will be used to estimate the percentages of the forest type likely to be found on each of these two broad classes of soil texture.  The results of this classification will be reported in terms of the amount of forest across the SAMI domain and in representative Class I areas that are potentially sensitive to ozone. 

 

Phase II Assessment Design

 

Three steps are needed to extrapolate the effect of ozone on leaf photosynthesis to the effect of multi-year exposures of ozone on regional forests.  Each step must receive careful attention in the assessment because the processes at each biological scale greatly alters the ozone response.

 

Extrapolation from the effect of ozone on leaf photosynthesis to individual trees.  First, prediction of forest behavior is impossible without first explicitly considering the behavior of individual trees in exposure response to the mix of co-occurring stresses.  TREGRO predicts whether a given exposure of ozone is likely to force a tree beyond these limits, as well as the changes in growth pattern each tree will exhibit when this occurs.  This growth pattern will, in turn, determine whether a tree’s ability to compete within a forest canopy is altered.

 

The relationship between cumulative ozone exposure and hourly maximum potential photosynthesis rate will be entered into the TREGRO model of individual tree behavior.  Next, a separate simulation of the TREGRO model will be run for each species for three growing seasons.  For each of five species (red maple, sugar maple (Acer saccharum Marsh.), northern red oak, black cherry, and yellow-poplar), a prediction of the cumulative effect on overall tree growth, leaf canopy development, and the ratio of fine roots to leaves will be made.  This cumulative effect on individual trees will be assumed to represent the impact on an average mature tree of that species in the forest in the absence of competitive interactions.

 

Extrapolation from effects on individual trees to the ability of trees to successfully compete during forest succession.  The second critical process will be to determine if the changes in carbon accumulation patterns within individual trees will alter the competitive ability of the trees within a forest canopy.  Exposure to ozone could reduce the growth of a tree sufficiently to permit neighboring trees to grow taller and cut off access to full sunlight.  To predict the dynamics of plants in forest stands requires consideration of the regeneration dynamics, longevity, growth, and water and nutrient use patterns of co-occurring species. 

 

To examine the impact that the ozone-induced changes predicted by TREGRO would have on forest stand development, the results from TREGRO will be entered into ZELIG, a gap-succession model.  ZELIG uses estimates of the expected growth rates of different tree species to predict the effects of competition among individuals of different species.  ZELIG analysis will allow the assessment of the effect of ozone on competition among trees and evaluate whether trees in competition are more sensitive to ozone. 

 

ZELIG simulates the establishment of seedlings as a function of the average annual regime of environmental conditions and the ability of selected species to withstand these conditions.  The annual diameter increment of each tree in a section of forest is calculated in response to the regime of environmental conditions established each simulation year.  Since each tree species has a differential response to these conditions, competition is predicted based on which species can convert the available resources under the prevailing environmental conditions into the fastest growth rate.  Each tree is assumed to have a constant probability of dying each year, with the rate differing among species based on the percent of individuals of the species expected to survive through a time interval.  The probability of mortality is also increased for trees growing particularly slowly. 

 

Three predictions of the response of an individual tree to ozone made by TREGRO must be passed to ZELIG to evaluate whether changes in tree growth change the ability of trees to compete during forest development (Figure 1).  First, as ozone exposure reduces photosynthesis, carbon accumulates more slowly in a tree, slowing the height growth and crown expansion of the tree.  TREGRO assumes that the rate at which a tree accumulates carbon is directly related to the rate at which it grows in height and diameter.  Since, in the ZELIG model, the greater the height of a tree, the greater the likelihood that the tree will intercept more light, the ZELIG model must be given a prediction of the rate at which tree growth will be slowed under a given level of ozone exposure.

 

Second, if ozone alters the amount of leaves that a tree grows, it will alter the light environment for all surrounding trees.  An ozone-induced reduction of the canopy of a given tree will cause less shading to adjacent trees beneath its canopy.  TREGRO will predict the expected change in leaf area of each tree of a species given a cumulative annual exposure of ozone.  The relationship between ozone and leaf area change for each species will be entered into ZELIG, which in turn will calculate the consequences of a change in canopy leaf area.

 

Third, both field studies and previous simulations of tree response to ozone have demonstrated that the fine root biomass often declines under continuous ozone exposure.  This decline potentially has a major effect on the ability of a tree to withstand extended periods of drought, particularly if it is associated with a decline in the amount of water absorbing root tissue available to supply each unit of leaf tissue.  Therefore TREGRO will be used to calculate the amount of change in the ratio of fine root biomass to leaf biomass expected under a given annual exposure of ozone.  An assumption will be made that a decline in this ratio would be directly correlated with a decrease in growth at a given level of cumulative soil moisture availability during a given year.  ZELIG then calculates the consequences to stand development of this increase in drought sensitivity.

 

Extrapolation to region.  A regional assessment will be produced by scaling up the results of tree and stand data to the region on a 12 km grid cell basis.  In addition to providing predictions of the effect of ozone on overall forest growth, we will predict the effect of ozone on species composition of the stands.  The results from the TREGRO-ZELIG simulation set for the appropriate soil condition will be used to estimate the predicted sensitivity to ozone and the change in resource distributions over time in response to changes in ozone exposures.  Regression models of the response of each species to ozone will be developed for each of the combinations of soil texture and precipitation described above.  Based on the regional soil and precipitation data, the appropriate regression will be selected to predict the effect of ozone on each species in each grid cell.  The regional ozone data set will then be used as input to the regression equation to predict the forest-stand response to ozone.

 

Maps will be produced of both the distribution of potentially sensitive forests and the amount of predicted change in the abundance of the key species in these forests across the spatial range of the Southern Appalachian Mountains.  Forest distributions for 2010 and 2040 will be projected.  SAMI will supply assumptions for ozone exposure levels in a future baseline or reference case.  These tables and maps will be created for each of the emissions strategies provided by SAMI.  These results will also be reported in terms of the percentage of the forest in specified categories of sensitivity of forest growth to ozone across the SAMI domain and in representative Class I areas. 

 

 

CONCLUSIONS TO DATE

 

The SAMI ozone assessment is ongoing and is not expected to be completed until the summer of 2001.  However, over the past five years, a significant amount of work has been completed.  An ozone assessment approach has been selected to use ozone exposure projections from the atmospheric modeling component of SAMI to drive a multi-model projection of forest responses across the SAMI region.  Consensus findings from literature review and modeling analysis include:

·        Tropospheric ozone has been shown to cause foliar injury on sensitive vegetation throughout much of the SAMI region. However the literature review did not find any clearly defined association demonstrated between foliar injury and growth under natural growing conditions.

·        Growth and physiological responses to ozone have been reported for tree species that occur in the SAMI region.  The majority of these investigations were short-term, under controlled conditions, and were with potted seedlings.

·        Soil moisture may alter tree growth response to ozone exposures.  Combination of exposure information with moisture availability and experimental exposure-response data might be used to identify areas that may have the greatest potential for possible vegetation effects.

·        A seasonal, cumulative, 24-hour, ozone exposure index is the most biologically relevant representation of exposures that impact forests.

·        Elevated levels of ozone reduce growth rate in some tree species.  Species with faster growth rates tended to have higher rates of ozone uptake and were more sensitive to ozone than slower growing species.

·         The influence of micro-site factors was considered to be most important in controlling response to ozone and  available soil moisture was determined to be of  greatest single importance in controlling ozone uptake.

·        An ozone assessment approach has been selected to use ozone exposure projections from the atmospheric modeling component of SAMI to drive a multi-model projection of forest responses.  The approach specifically considers two critical processes that will determine how forests will respond to ozone: 1) the ability of trees to absorb ozone-induced injury when faced with co-occurring stresses, such as drought, infertile soils, and excess nitrogen deposition; and 2) The changes in the species composition of forests that will occur when one or more species is weakened by ozone.  Effects on stand growth and species composition to the entire region will be extrapolated by linking predicted ozone exposures with predicted dose-response relationships, forest type location, and soil conditions.  This procedure will be followed for each of the proposed SAMI emissions strategies.

·        There is the recognition that because of the tremendous variability exists within natural systems and modeling ozone induced exposure/responses across the geographic range of a species in the SAMI region posed a formidable task.  The importance of the relative influences of insect pests, biotic pathogens, abiotic stressors (other than ozone), inter- and intra-specific competition for resources in contributing to the changes in forest health and productivity must be considered in any assessment interpretation.  Otherwise an assessment might provide an unrealistic scenario of the relative importance of ambient ozone exposures to forests in the SAMI region.

 

 

LITERATURE CITED

 

1.  Chappelka, A. H., Samuelson, L. J., Skelly, J. M., Lefohn, A. S. Effects of ozone on forest trees in the southern Appalachians: an assessment of the current state of knowledge. 1996, Report prepared for the Southern Appalachian Mountains Initiative.

 

2. Chappelka, A. H., Samuelson, L. J. New Phytologist 1998, 139, 91-108.

 

3.  Lefohn, A. S., Jackson, W., Shadwick, D. S., and Knudsen, H. P. Atmospheric Environment 1997, 31, 1695-1708.

 

4.  Hogsett, W. E., Herstrom, A. A., Laurence, J. A., Lee, E. H., Weber, J. E., Tingey, D. T. Risk characterization of tropospheric ozone to forests. In: Proceedings of the 4th US/Dutch International Symposium: Comparative Risk Analysis and Priority Setting for Air Pollution Issue; Air and Waste Management Association, Pittsburgh, PA. 1993.

 

5. Hogsett, W. E., Weber, J. E., Tingey, D. T., Herstrom, A. A., Lee, E. H., Laurence, J. A., Environmental Management 1997, 21, 105-120.

 

6.  Luxmoore, R. J. In Transactions: the response of southern commercial forests to air pollution; Fagler, R. B., Ed.; Air and Waste Management Association, Pittsburgh, PA., pp 313-322, 1992.

 

7.  Lefohn, A. S. The identification of ozone exposures that result in vegetation visible injury and growth loss for specific species grown in the southern Appalachian Mountain region. 1998. Report prepared for the Southern Appalachian Mountains Initiative.

 

8.  Weinstein, D. A., Gollands, B., Retzlaff, W. A. The effects of ozone on regional productivity of lower slope hardwood forests of the Smoky Mountain region. Manuscript submitted to Forest Science, 2000.

9.  Weinstein, D. A., Gollands, B., Retzlaff, W. A. The influence of soil moisture on the effects of ozone in lower slope hardwood forests of the Smoky Mountain region. Manuscript, 2000.

10. Weinstein, D. A., Beloin, R. M., Yanai, R. D. Tree Physiology 1991, 9, 127-146.

11. Weinstein, D. A., R. D. Yanai,. J. Environmental Quality 1994, 23,: 418-428.

12. Weinstein, D. A., Samuelson, L. J., Arthur, M. A. Environmental Pollution 1998, 102, 307-320.

13. Urban, D. L., Bonan, G. B., Smith, T. M., Shugart, H. H. Forest Ecology and Management 1991, 42, 95-110.

14. Laurence, J. A., Ollinger, S., Woodbury, P. B. Regional impacts of multiple stresses on productivity and forest health. In Responses of northern U. S. forests to environmental change; Birdsey R, Hom, J., Mickler, R., Eds.; Springer-Verlag, NY, In press.

15. Hansen, M. H., Frieswyk, T., Glover, J. F., Kelly, J. F. USDA Forest Service Gen. Tech. Rep. NC-151, 1992, St. Paul, MN, USDA Forest Service, North Central Forest Experiment Station.

 

 

Paper presented at the AWMA 93rd Annual Conference & Exhibition, Salt Lake City, Utah, USA, June 18-22, 2000.