United States
Environmental Protection
Agency
Environmental Monitoring and Assessment
Program Center
Research Triangle Park, NC 27711
Research and Development
EPA/620/SR-94/006
June 1994
Project Summary
Forest Health Monitoring:
Southeast Loblolly/Shortleaf Pine
Demonstration Interim Report
Timothy E. Lewis and Barbara L Conkling
The U.S. Environmental Protection
Agency (EPA) has initiated a multiagency
program to determine the current status
and future trends in the nation's eco-
logical resources. The Environmental
Monitoring and Assessment Program
(EMAP) is focusing on broad resource
groups such as inland surface waters,
estuaries, arid lands, agroecosystems,
and forests. The forest component of
EMAP is a multiagency effort referred
to as Forest Health Monitoring (FHM).
The FHM program is jointly managed
and largely funded by the U.S. Depart-
ment of Agriculture (USDA) Forest Ser-
vice and EPA in cooperation with other
program participants.
The FHM program conducted a two-
year demonstration study to test a suite
of indicators considered important in
assessing forest health. The selected
study plots represented the loblolly/
shortleaf pine forest type. The study is
referred to in this report as the South-
east Loblolly/Shortleaf Pine Demonstra-
tion or SE DEMO. This interim report
serves to initiate the indicator evalua-
tion process and addresses the first
objective of the study. Six criteria have
been adopted by FHM and EMAP for
indicator evaluation. The FHM program
has developed an objective procedure
for determining the status of indicators
with respect to these six criteria. The
indicators were evaluated from a sta-
tistical arid biological perspective to
ascertain their suitability in Detection
Monitoring. After examination of data
collected during the first year of the SE
DEMO, several shortcomings were re-
vealed that necessitated the implemen-
tation of two additional companion
studies, the QA Reference Plot and
Georgia Remeasurement Studies.
These two studies will provide valu-
able information for addressing the in-
dex-period stability and high signal-to-
noise ratio criteria. The data from these
two studies, combined with the second
year's data from the SE DEMO, will be
used to evaluate the status of each
indicator. These findings will be re-
ported in the SE DEMO final report.
This Project Summary was developed
by EPA's Environmental Monitoring and
Assessment Program Center, Research
Triangle Park, NC, to announce key
findings of the research project that is
fully documented in a separate report
of the same title (see ordering informa-
tion at back).
Introduction
The U.S. Environmental Protection
Agency (EPA) has initiated a multiagency
program to determine the current status
and future trends in the nation's ecologi-
cal resources. The Environmental Moni-
toring and Assessment Program (EMAP)
is focusing on broad resource groups such
as inland surface waters, estuaries, arid
lands, agroecosystems, and forest. The
forest component of EMAP is a
multiagency effort referred to as Forest
Health Monitoring (FHM). The FHM pro-
gram is jointly managed and largely funded
by the U.S. Department of Agriculture
(USDA) Forest Service and EPA in coop-
eration with other program participants.
FHM partners provide additional financial
and personnel support and include partici-
pating state forestry agencies, the U.S.
Department of Interior's (USDI's) Bureau
of Land Management, the Tennessee Val-
ley Authority, and the USDA Soil Conser-
vation Service. Other cooperators include
Printed on Recycled Paper
-------
universities and three additional USD!
agencies: U.S. Fish and Wildlife Service,
U.S. Geological Survey, and the National
Park Service. The National Association of
State Foresters provides essential sup-
port, guidance, and assistance.
The FHM program conducted a two-
year demonstration study to test a suite of
indicators considered important in assess-
ing forest health. The selected study plots
represented the loblolly/shortleaf pine for-
est type. The study is referred to in this
report as the Southeast Lobloliy/Shortleaf
Pine Demonstration or SE DEMO. The
indicators were evaluated from a statisti-
cal and biological perspective to ascertain
their suitability in Detection Monitoring.
The objectives of the SE DEMO were
1.To evaluate selected indicators across
the Atlantic Coastal Plain and Piedmont
loblolly/shortleaf pine forest type, and
2.To evaluate their ability to provide an
assessment of condition in the loblolly/
shortleaf pine forest type.
The interim report initiates the indicator
development process for the SE DEMO.
This report addresses the first objective
as stated above. Operating concurrently
with the SE DEMO was the Southern Ap-
palachian Man and Biosphere (SAMAB)
Demonstration. Some indicator leaders
used data from the SAMAB region to aug-
ment their ability to address the indicator
development criteria, although the primary
focus was on the loblolly/shortleaf pine
forest ecosystem of the SE DEMO.
The interim report identifies problems in
data collection, quality assurance and qual-
ity control (QA/QC), and acquisition. The
report also identifies additional data
needed for indicator development. For pur-
poses of this report, most of the indicator
leaders treated the set of SE DEMO plots
as a simple random sample to facilitate
data analysis.
The indicators evaluated in the SE
DEMO were:
Bioindicator Plants
Branch Evaluations
Crown Condition
Damage
Dendrochronology (not included in in-
terim report)
Dendrochemistry (not included in interim
report)
Foliar Chemistry
Lichen Communities
Lichen Chemistry (not included in in-
terim report)
Mensuration
Photosynthetically Active Radiation
(PAR)
Root Pathogens (not included in interim
report)
Soil Physiochemistry
Vegetation Structure
Wildlife Habitat
PAR indicator evaluation was done only
on the SAMAB plots. These plots were
not typically located in the loblolly/short-
leaf pine forest type but are nevertheless
included in the report as an appendix to
illustrate the ability of this indicator to meet
indicator development requirements. As
noted in the above list, some indicators
are not included in this interim report due
to problems with data availability and staff-
ing. All of the listed indicators will be in-
cluded in the SE DEMO final report. The
SE DEMO final report will also address
objective two above and will evaluate the
utility of the selected indicators for as-
sessing the condition of the loblolly/short-
leaf pine forest type.
The FHM indicators must progress from
candidate to pilot, to demonstration, and
finally to core status before they are con-
sidered "ready" for national implementa-
tion in Detection Monitoring. The progres-
sion involves many steps, which culmi-
nate in the ability of a selected indicator to
satisfactorily meet the indicator develop-
ment criteria. These steps include 1) iden-
tification of relevant societal values of con-
cern; 2) formulation of assessment ques-
tions to address; 3) selection of candidate
indicators; 4) review of scientific literature
and off-frame data; 5) testing of indicators
in pilot and demonstration studies; and 6)
evaluating the six indicator development
criteria.
FHM scientists selected the indicators
evaluated in the SE DEMO based on lit-
erature and peer reviews and expert opin-
ion. Each indicator addresses at least one
assessment question pertaining to certain
values that can be easily interpreted by
other scientists, policy makers, and the
general public. The selected indicators pro-
vide quantitative or qualitative links to the
key ecosystem processes and compo-
nents. An indicator thus serves as a met-
ric of society's concerns and its percep-
tions about forest health. Indicators are not
intended to demonstrate cause-and-effect
relationships but will do so in certain in-
stances. The preponderance of evidence
obtained from monitoring activities may
be convincing enough to implicate or clarify
certain causal hypotheses, but additional
off-frame data will usually be required to
verify or nullify these hypotheses.
The interim report describes the results
from the first year of the SE DEMO. Where
data were not sufficient from the first year
to answer some of the criteria, ancillary
data were used. The use of such data is
encouraged by EMAP. These data were
used to the extent possible to determine
the current status of each indicator with
respect to the six criteria. Recognizing the
lack of existing data, published or unpub-
lished, to answer some of the indicator
development criteria, the FHM team de-
signed two studies that were conducted in
the second year (1993) of the SE DEMO.
These two studies, the QA Reference Plot
Study and the Georgia Remeasurement
Study, will provide critical data for many
indicators to address the indicator devel-
opment criteria. Inasmuch as the final data
from these studies were not collected until
late August 1993, they are not available
for evaluation in this interim report. How-
ever, the final report will include these
data.
Procedures
The SE DEMO 1992 study area included
most of Virginia, Georgia, North Carolina,
and South Carolina. FHM Detection Moni-
toring plots had been established through-
out Virginia and Georgia during the 1991
field season. The 1991 Detection Monitor-
ing data were queried for all Virginia and
Georgia plots that had a forest type of
loblolly or shortleaf pine. Those plots that
were assigned to the 1992 rotation of the
one-quarter interpenetrating design (rotat-
ing panel sampling design based on
resampling each quarter every fourth year)
and did not fall in the SAMAB region were
resampled as part of the 1992 SE DEMO.
North and South Carolina had not been
part of the 1991 Detection Monitoring pro-
gram and therefore there were no existing
FHM plots. Instead, the USDA Forest
Service's Region 8 Forest Pest Manage-
ment Southern Atlas Project data set was
queried for all counties in the Carolinas
that had an average loblolly/shortleaf pine
basal area of >50 fWacre. State forestry
teams were sent to reconnoiter the plots
assigned to the 1991 and 1992 rotation of
the one-quarter interpenetrating design. A
plot was then included as part of the 1992
SE Demo if the plot was (1) forested (in
forest use, at least one acre in size, and
at least 120 ft wide), (2) not in the SAMAB
region, (3) had at least one subplot where
a majority of the overstory trees were
loblolly and/or shortleaf pine, (4) was as-
signed to the 1991 or 1992 rotation, and
(5) land owner permission could be ob-
tained.
Operating concurrently with the SE
DEMO was the SAMAB Demonstration.
As stated above, some indicator leaders
used data from the SAMAB region in the
evaluation of their indicator, although the
primary focus was on the loblolly/shortleaf
pine forest ecosystem of the SE DEMO.
-------
The SAMAB region includes 33 million
acres of southwest Virginia, eastern Ten-
nessee, western North Carolina, northern
Georgia, and parts of South Carolina and
Alabama.
Results and Discussion
Bioindicator Plants
The air pollution bioindicator plant indi-
cator has been a part of Detection Moni-
toring activities in the FHM program since
1990, when the first field plots were es-
tablished in six New England states. The
early testing of this indicator underscores
its importance with respect to the societal
values identified by FHM. In 1992, with
some improvement in field methodology,
the bioindicator plants indicator moved into
the two southeast demonstration projects,
the SE DEMO and the SAMAB Demon-
stration.
The specific FHM societal values that
the bioindicator plants indicator addresses
include ecological integrity (e.g.,
biodiversity, productivity, and sustainability)
and aesthetics (e.g., injured foliage). The
link between the indicator and the FHM
conceptual model is made through the
following assessment questions, thus pro-
viding the essential rational for implement-
ing the bioindicator plant indicator in FHM.
1.1s there an association between the
number and distribution of plots with
foliar injury and air quality related val-
ues (e.g., aesthetics), by region or for-
est type?
2. How does the number and distribution
of plots with foliar injury change from
one year to the next, by region or
forest type?
3. Is there an association between ambi-
ent air quality data and plots with in-
jury and/or the plot-level measurement
index?
4.1s there an association between the
bioindicator response at ground level
and conditions in the upper canopy, by
region or forest type?
5. Is the change in number or distribution
of plots that show injury associated
with changes in biodiversity, by region
or forest type?
The relationship between the bioindicator
plants indicator and the six indicator devel-
opment criteria is summarized in Table 1.
Where there is uncertainty, the results of
the 1993 SE DEMO Project and the 1993
QA Reference Plot Study will most likely
provide the statistics needed to complete
the criteria evaluations.
Table 1. Status of Bioindicator Plant Indicator After First Year of the SE DEMO
Indicator Development Criteria Status
Unambiguously Interpretable
Conceptual Model
Indices
Nominal/Subnominal Threshold
Index-Period Stability
High Signal-to-Noise Ratio
Regional Responsiveness
Simple Quantification
Low Environmental Impact
High
Good
Good
Proposed - needs research
Needs Testing - revisit index
Needs Testing - assumed high
High
High - revisit index
High
Crown Condition
The crown portion of trees has always
been an indicator of the .health of forest
trees. Long before scientists attempted to
add the scientific process and rigor to the
estimation or measurement of crown vari-
ables, tree crown condition provided an
instantaneous visual estimation of the vigor
of individual trees, tree stands, and some-
times entire watersheds. Since tree crowns
form the basic structural architecture of a
forest ecosystem, they directly affect the
composition, processes, and vigor of the
forest understory floral and faunal compo-
nents.
The crown condition indicator addresses
FHM societal values of productivity and
aesthetics. Productivity has been shown
to be correlated with crown variables. Aes-
thetics of forest resources are dependent
upon the visual appearance of tree crowns.
The following assessment questions can
be addressed using the three indices for-
mulated during the 1992 SE DEMO.
1. What proportion of forests is growing
poorly due to subnominal crown con-
ditions across forest types or regions?
2. What proportion of forests is in aestheti-
cally poor quality due to subnominal
crown conditions across forest types or
regions?
The crown indices are currently meet-
ing many of the indicator development
criteria. Table 2 shows the relationship
between the crown condition indices and
the six indicator development criteria after
completion of the first year of the SE
DEMO. After completion of data analysis
of the 1993 SE DEMO, the QA Reference
Plot, and the Georgia Remeasurement Plot
Studies, more of the criteria will be more
adequately addressed, including index-period
stability. The necessary data will be avail-
able to evaluate the change in the crown
variables over the field season and the
subsequent effects of that change on the
crown indices. It will be possible to evalu-
ate the accuracy and precision for within-
and between-crew measurement error.
Approximately twice as many plots will be
available to calculate plot-to-plot variabil-
ity. More data from the QA Reference Plot
Study will also be available to quantify the
environmental impact to the plots. Evalua-
tion of the signal-to-noise ratio will begin
because data on the year-to-year variabil-
ity in the crown indices from three years
of the Georgia Remeasurement Plot Study
will be available.
Damage Indicator
Damage caused by diseases, insects,
air pollution, and natural and man-made
activities can affect forest growth and de-
Table 2. Status of Crown Condition Indicator After First Year of the SE DEMO
Indicator Development Criteria
Unambiguously Interpretable
Conceptual Model
Indices
Nominal/Subnominal Threshold
Index-Period Stability
High Signal-to-Noise Ratio
Regional Responsiveness
Simple Quantification
Low Environmental Impact
Status
Dieback Transparency
Good Good
Good Good
Good Good
Fair - needs research Good
Good Good
Good Good
Good Good
High High
Good Good ,
Structure
Good
Good
Fair -needs research
Fair -needs research
Good
Low- needs research
Good
High
Good
-------
velopment. Any of these agents, either
singly or in combination, can cause sig-
nificant declines in forest tree health. Dam-
age caused by the aforementioned agents
can reduce tree productivity, reduce stand
survivability and sustainability, and por-
tend reduced future growth and increased
mortality. The damage that is observed
on a tree may be an early-warning indica-
tion of reduced growth and may result in
the ultimate death of the tree. The dam-
age indicator may also be useful in as-
sessing the aesthetic value of forests. For
example, the recent public debate over
the decline of high spruce-fir forests in the
southern Appalachian Mountains that are
infested with the balsam wooly adelgid
(sulfur dioxide pollution has been factored
into the problem) has often been portrayed
from a tree damage perspective; photos
of large stands of brown-foliaged trees
were shown on television stations through-
out the country.
In 1992, damage and location were re-
corded for each tree. A severity rating
was not recorded in 1992 (except for dwarf
mistletoe) but was added in 1993 to pro-
vide a more reliable damage index. Three
different types of damage were recorded
for each tree. The location of injury is
important in ascertaining the degree of
threat to the survivability of the tree and
was recorded for all trees meeting the
aforementioned criteria. During the 1992
field season the crews were given the
liberty to record the cause of the observed
damage. Without having a well-trained for-
est pathologist or entomologist on each
crew, the diagnosis of cause was specie
lative, at best The diagnosis of cause
was deemed too variable to provide accu-
rate data for assessment of forest health
and thus was discontinued in 1993.
Using the ptot-Ievel damage index de-
veloped during the SE DEMO, the follow-
ing assessment questions can be formu-
lated.
I.What proportion of forest in the re-
gion is at risk of reduced growth due
to observed damages?
2.What proportion of forest in the re-
gion is at increased risk of mortality
due to observed damages?
3.1s the risk of reduced growth due to
observed damages increasing (or de-
creasing) in the region? .
4. Are the aesthetic qualities of forests
in the region improving or diminishing
due to observed damages?
The damage indicator had difficulty ad-
dressing many of the criteria primarily be-
cause of a lack of a reasonable plot-level
Index. The Index has not been the same
for more than one year. Since the index
has evolved over time, year-to-year vari-
ability is the most difficult component of
variability to address. Data need to be
gathered on the utility of the index across
multiple years through evaluation of indi-
cator performance in Detection Monitor-
ing. Table 3 summarizes the status of the
damage indicator with respect to the six
criteria.
Branch Evaluations
Observation of branches from the
ground provides general information on
tree crown condition and detects the more
severe injury symptoms. Close-up exami-
nation of foliage is required, however, to
obtain more specific tree vigor data and
detect early symptoms of branch damage
caused by air pollutants, disease, or in-
sects. The in-hand evaluation of branches
from the upper tree crown is FHM's branch
evaluation indicator.
The branch evaluation indicator ad-
dresses two important FHM societal val-
ues, productivity and aesthetics. The de-
velopment of relationships between the
branch evaluation metrics and important
assessment questions can be used as a
consistent way to set thresholds across
species, forest types, or other resource
groups. The following assessment ques-
tions can be addressed by the three branch
evaluation indices.
I.What proportion of forests has foliar or
wood damage resulting in a reduction in
growth across forest types or regions?
2. What proportion of forests has foliar or
wood damage resulting in a reduction in
the aesthetic qualities across forest types
or regions?
The branch evaluation indices currently
meet some, but not all, of the indicator
development criteria. Table 4 summarizes
the status of the branch evaluation indica-
tor with respect to the six criteria. After
completion of data analysis for the QA
Reference Plot Study, index-period stabil-
ity will be evaluated. After analysis of the
1993 SE DEMO and QA Reference Plot
data, the following will be evaluated: (1)
changes in the branch variables over the
field season and the subsequent effects
of such change on the branch indices; (2)
within- and between-crew measurement
error; and (3) plot-to-plot variability with
approximately twice as many plots. More
data will be available to quantify the envi-
ronmental impact on the specimen trees
from the QA Reference Plot study. In ad-
dition, signal-to-noise ratio will be evalu-
ated, because data on the year-to-year
variability in the branch indices will be
available for two years from all SE DEMO
plots and for three years from the Georgia
Remeasurement Plots.
Table 3. Status of Damage Indicator After First Year of the SE DEMO
Indicator Development Criteria Status
Unambiguously Interpretable
Conceptual Model
Indices
. NominaVSubnominal Threshold
Index-Period Stability
High Signal-to-Noise Ratio
Regional Responsiveness
Simple Quantification
Low Environmental Impact
Fair
Good
Low
Fair- needs research
' Fair- reevaluate index
Good - reevaluate index
Good
Good - reevaluate index
High
Table 4. Status of Branch Evaluation Indicator After First Year of the SE DEMO
Indicator Development Criteria Status
Unambiguously Interpretable
Conceptual Model
Indices
Nominal/Subnominal Threshold
Index-Period Stability
High Signal-to-Noise Ratio
Regional Responsiveness
Simple Quantification
Low Environmental Impact
Fair
Fair
Good
• Fair - needs research
Good-
Fair - needs research
Fair
High
Fair
-------
Foliar Chemistry
Foliar chemistry serves as both a con-
dition and a stressor indicator. The status
of macro- and micronutrients provides in-
formation on current condition. If these
nutrients are at or near their optimal or
critical levels, then a nominal nutrient sta-
tus condition is indicated. If the levels are
deficient or excessive, then a subnominal
condition is suggested. Selected trace el-
ements (e.g., Pb, Cd, Hg) are also mea-
sured in foliage and serve as indicators of
plant stress. As a stressor indicator, foliar
chemistry can be associated with other
condition indicators that may be related to
the status or change in forest condition
(e.g., visible injury, growth, soil productiv-
ity).
Realizing the significant natural variabil-
ity in foliar chemistry that is inherent in the
measurement system, a carefully designed
sampling strategy was implemented for
the.SE DEMO. Only foliar chemistry for
loblolly pine is reported in the SE DEMO
interim report. Diagnostic Recommenda-
tion and Integrated System (DRIS) indi-
ces were calculated for the loblolly pine
foliar chemistry data obtained from the
1992 SE DEMO. Because of the way in
which the DRIS norms are derived, the
index is directly related to an assessment
question: "What proportion of the forests
has foliar nutritional levels capable of sup-
porting optimal growth in the Southeast-
ern United States?"
Using the current sampling protocols
and the DRIS indices for loblolly pine, the
foliar chemistry indicator was able to ad-
equately address the unambiguously in-
terpretable, simple quantification, and low
environmental impact indicator develop-
ment criteria (Table 5). With ancillary data
and off-frame research, the other indica-
tor development criteria will be addressed.
Lichen Communities
Hundreds of papers worldwide
(chronicled in the series "Literature on air
pollution and lichens" in the Lichenologist)
and dozens of review papers and books
published during the last century have
documented the close relationship between
lichen communities and air pollution, es-
pecially SO2, NO , and acidifying or fertil-
izing nitrogen and sulfur-based pollutants.
Although trees may respond to moderate,
chronic levels of air pollution, all of the
other influences on tree growth, such as
complex variation in soils, make the re-
sponse to pollutants difficult to measure.
Lichens provide, therefore, a clear indica-
tion of air pollution impacts, not only di-
rectly upon lichens, but also upon total
forest productivity. Because air pollution
affects long-term forest sustainability and
biodiversity, lichens also indicate trends in
assessing those ecosystem properties. In
addition, lichens themselves form a large
portion of the macrophytic species of many
forests.
For the purposes of the IFHM program,
the sampled flora is restricted to
macrolichens that occur on living or stand-
ing dead woody substrates. Only standing
woody substrates are included to stan-
dardize the measurements to a class of
substrates that can be found on all sites.
For example, although lichens commonly
are abundant and diverse on rocks, many
FHM plots will not contain exposed rock
as an available substrate.
Lichen communities provide information
relevant to several key assessment ques-
tions, including those concerning the con-
tamination of natural resources,
biodiversity, and sustainability of timber
production. Lichens not only indicate the
health of our forests, but there is a clearly
established linkage to environmental stres-
sors. Lichen communities are useful in
addressing the following assessment ques-
tions:
1. Is regional air quality (specifically ni-
trogen- and sulfur-based pollutants)
changing over time? If so, is it im-
proving or deteriorating? In what ar-
eas is it changing? In what fraction of
the area do lichens indicate the pos-
Table 5. Status of Foliar Chemistry Indicator After First Year of the S£ DEMO
Indicator Development Criteria Status
Unambiguously Interpretable
Conceptual Model
Indices
Nominal/Subnominal Threshold
Index-Period Stability
High Signal-to-Noise Ratio
Regional Responsiveness
Simple Quantification
Low Environmental Impact
Good
Fair
Good
High
Good
Fair- needs research
Low
Good
Fair
sibility of air pollution impacts on for-
est productivity and biodiversity?
2. Is the lichen component of biodiversity
changing over time? If so, is it in-
creasing or decreasing? In what ar-
eas is it changing?
It is believed that most criteria for indi-
cator development have been met (Table
6). Calibration of the lichen indices to lev-
els of ambient air quality in different re-
gions of the country is critical. The most
immediate tasks facing the implementa-
tion of the lichen community indicator are
stating threshold values for the lichen com-
munity indices and analysis of the QA
Reference Plot data for the evaluation of
high signal-to-noise ratio, index-period sta-
bility, and accuracy of the separation of
species by the field crews (simple quanti-
fication).
Mensuration Indicators
Forest mensuration is a crucial compo-
nent of any forest health determination.
The scope of forestry has widened due to
the advent of the sciences of ecology and
environmental biology. The area of men-
suration now covers topics essential to
forest ecosystem modeling, insect and dis-
ease incidence, wildlife habitat and wild-
life management, and recreation. This list
is, of course, not inclusive. Numerous for-
est succession and growth models have
been developed, all of which are predi-
cated on the forest ecosystem processes
of tree natality, growth, and mortality. Most
of these models depend on information
on the species present or expected in the
stand, thus highlighting the importance of
tree species diversity. This indicator pro-
vides information on habitat structure,
which strongly influences wildlife diversity
for a variety of taxa, including vascular
flora, birds, small mammals, and large
mammals. Upper canopy species diver-
sity is only one of the measurements
needed to comprehensively assess re-
gional forest ecosystem biodiversity. FHM
is preparing other measures of biodiversity.
Stand structure and stand diversity are also
important in understanding the effects of
the incidence of pests and diseases. The
forest ecosystem factors of regeneration,
growth, mortality, and overstory diversity
comprise the suite of mensuration indica-
tors.
The mensuration indicators address
three major FHM societal values, biotic
integrity, aesthetics, and extent. However,
the SE DEMO report only addresses bi-
otic integrity, which incorporates such is-
sues as sustainability, productivity (in the
ecological sense), and biodiversity.
Sustainability of a forest ecosystem is con-
-------
Tablo 6. Status ofUchen Communities
Indicator After First Year of the SE
DEMO
Indicator Development Criteria
Status
Unambiguously Interpretable High
Conceptual Model Good
Indices High
Nominal/Subnominal Threshold Good
Index-Period Stability Good
High Signal-to-Noise Ratio Good
Regional Responsiveness High
Simple Quantification High
Law Environmental Impact Good
ceptually dependent on the processes of
natality, growth, and mortality, as well as
sufficient diversity to withstand the typical
catastrophic events that impact forests.
Forest productivity, in the silvicultural and
ecological senses, is highly related to these
same three processes. Using the four men-
suration indicators and their associated
indices, the following assessment ques-
tions can be addressed:
t.What proportion of the population is
subnominal with respect to
a. productivity, as represented by
plot-level basal area increment?
b. forest natality and regeneration,
as represented by seedling re-
cruitment?
c. sustainability, as represented by
high overstory mortality rate?
d. resiliency, as represented by low
overstory species diversity?
2.1s there a regional change in "a"
through "d*?
3.What is the spatial distribution of "a"
through "d"?
Table 7 shows the relationships between
each mensuration indicator and the six
indicator development criteria. The data
from the QA Reference plots will permit
evaluation of regeneration for index-period
stability. Overstory diversity has already
been shown to meet this criterion. Over-
story diversity now passes all the criteria.
Growth and mortality require more than
one year of data to calculate, so
Index-period stability cannot be evaluated
based on one year of data from these
plots. A revisit to the QA Reference plots
or further literature search will be needed
to answer this question for these two indi-
cators. However, the available literature
suggests that the growth indicator has the
potential to meet the statistical criteria of
index-period stability and high
signal-to-noise ratio. Mortality already
passes all the critical criteria, except
index-period stability. It is believed, based
on the analysis of the New England De-
tection Monitoring data, that mortality will
pass this criterion as well. All of the men-
suration indicators meet the criteria of re-
gional responsiveness and low environ-
mental impact.
Soil Classification and
Physiochemistry
There are several conceptual models upon
which to base the FHM soil indicator develop-
ment activities, depending on the specific in-
dex of interest. For instance, the nitrogen
cycling model is commonly used to show the
interactions of N among soil, biota, and the
atmosphere. Regarding forest productivity,
models focusing on soil-climate interactions
are considered crucial to our understanding of
soil's influence on tree growth. From a hydro-
logic standpoint, several models of water flow
theory are used to assess hydro-physiographic
dynamics on forested slopes. At this time, the
number of possible models are too innumer-
able to present here. Instead, a conceptual
model coordinated across all soil indicators
will be presented in the SE DEMO final report
in 1994.
Since the development of plot-level in-
dices for this indicator is in progress, many
of the indicator development criteria could
not be addressed. The status of the indi-
cator as shown in Table 8 reflects this
shortcoming. Once suitable indices have
been formulated the indicator should be
capable of satisfactorily meeting most of
the criteria. Given the critical role soil plays
in numerous forest ecosystem processes,
it is difficult to envision a forest monitoring
program without some type of edaphic
indicator.
Vegetation Structure
The vegetation structure indicator aug-
ments other FHM on-plot vegetation mea-
surements (i.e., trees and lichens) with
measurements of non-tree, understory veg-
etation that cpmprises most of the vascu-
lar plant species diversity in forests. Fur-
ther, this indicator provides information
about habitat structure, which strongly in-
fluences wildlife diversity. Monitoring veg-
etation structure helps provide information
to evaluate the effectiveness of environ-
mental and/or forest resource policies. The
vegetation structure indicator supports this
effort by supplying information relevant to
policies concerning plant species diversity
(e.g., the Endangered Species Act of 1973)
and habitat diversity (e.g., Federal Policy
and Management Act of 1976). The ability
to assess habitat within FHM is important,
since habitat loss and alteration are the
most immediate and widespread threats
to biodiversity and may exacerbate the
predicted impacts of global climatic
change.
Table 7. Status of Mensuration Indicators After First Year of the SE DEMO
Status
Indicator Development Criteria
Unambiguously Interpretable
Conceptual Model
Indices
Nominal/Subnominal Threshold
Index-Period Stability
High Signal-to-Noise Ratio
Regional Responsiveness
Simple Quantification
Low Environmental Impact •
Regeneration
High
High
High
Good
Good
Good
High
High
Good
Mortality
High
High
High
Good
Good
High
High
High
Good
Growth
High
High
High
Good
Good
Good
High
High
Good
Diversity
High
High
High
Good
High
High
High
High
Good
Table 8. Status of Soil Classification and Physiochemistry Indicator After First Year of the
SEDEMO
Indicator Development Criteria
Status
Unambiguously Interpretable
Conceptual Model
Indices
Nominal/Subnominal Threshold
Index-Period Stability
High Signal-to-Noise Ratio
Regional Responsiveness
Simple Quantification
Low Environmental Impact
Fair
Fair
Low
Low
Fair
Fair
Fair
Good
Fair
-------
The vegetation structure indicator pro-
vides information on vascular plant com-
munity structure that will be used to ad-
dress the following assessment questions:
1. What percentage of the forest resource
in Region X contains vegetation in a
subnominal condition as evaluated with
the vegetation structure indicator?
2. Is the percentage of the forest resource
in Region X with subnominal vegeta-
tive condition increasing, decreasing,
or unchanging as evaluated with the
vegetation structure indicator?
3. What are the relationships between
vegetation attributes (e.g., amount, dis-
tribution, or composition) measured by
the vegetation structure indicator and
natural stresses or human-induced pol-
lution?
Table 9 shows the relationship between
the indicator and the six criteria.
At present, this indicator meets the low
environmental impact, simple quantifica-
tion, and regional responsiveness (in the
NE and SE) indicator development crite-
ria. All of the conditions for unambigu-
ously interpretable have been satisfied
except for determining thresholds of nomi-
nal/subnominal for lower strata. In addi-
tion, complexity must be incorporated into
the indicator. Index-period stability will be
addressed in more detail after data from
the QA Reference Plot Study have been
evaluated. Signal-to-noise ratio will be ad-
dressed using data from the Georgia
Remeasurement Study.
Wildlife Habitat
The wildlife habitat indicator is based on
empirical relationships between bird com-
munity composition and local vegetation
features. Natural disturbance regimes in
the eastern U.S. typically lead to landscapes
dominated by mature forest, with bird com-
munities dominated by forest-interior spe-
cies. The abundance and diversity of
disturbance-sensitive bird species are af-
fected by anthropogenic forest disturbance
and indicate the ecological health of the
local bird community. Bird communities of
undisturbed landscapes have high biologi-
cal diversity and are dominated by spe-
cies that are vulnerable to population de-
clines. Because the distribution patterns
of most bird species are strongly affected
by vegetation structure, habitat measure-
ments can be used to predict characteris-
tics of the local bird community such as
dominance by disturbance-sensitive spe-
cies.
The wildlife habitat indicator directly re-
lates to ecological integrity. The rationale
Table 9. Status of Vegetation Structure
Indicator After First Year of the SE
DEMO
Indicator Development Criteria
Status
Unambiguously Interpretable High
Conceptual Model High
Indices High
Nominal/Subnominal Threshold Fair
Index-Period Stability Good
High Signal-to-Noise Ratio Fair
Regional Responsiveness High
Simple Quantification High
Low Environmental Impact Good
for this approach is based on the fact that
the distribution of most bird species is
strongly affected by the structure and spe-
cies composition of vegetation. Distur-
bance can therefore significantly alter a
bird community. The absence of an im-
portant vegetation feature can often ren-
der habitat unsuitable for a species, and
vegetation variables often account for more
than 75% of the variance in spejcies' abun-
dances across sites. Using data from the
1992 SE DEMO project, an index of forest
bird community health has been developed
that can be calculated by comparing local
population levels of disturbance-sensitive and
disturbance-tolerant bird species. A multiple
regression model was developed to pre-
dict the value of this index from informa-
tion about local vegetation structure. The
predicted values derived from the regres-
sion model are used as the wildlife habitat
indicator. Using the wildlife habitat index
the following assessment questions can
be addressed:
LWhat proportion of the region has
subnominal breeding bird assem-
blages that are related to changes in
vegetation?
2. What is the spatial distribution of these
subnominal areas?
Table 10 summarizes the status of the
wildlife habitat indicator with respect to
the six indicator development criteria.
Thresholds for the index need to be ex-
amined. Index-period variability in the bird
community index and wildlife habitat indi-
cator does not appear to be significant.
However, more data are needed to thor-
oughly address index-period stability and
high signal-to-noise ratio criteria. The cu-
mulative distribution function for the wild-
life habitat indicator suggests that a larger
sample size is needed to address the
regional responsiveness criterion. The in-
dicator meets most of the remaining crite-
ria.
Photosynthetically Active
Radiation
Measurements of solar radiation inter-
cepted by the forest canopy have proven
fundamental to interpreting plant commu-
nity productivity and function. Photosyn-
thetically active radiation (PAR), the quan-
tity of light between the 400-700 nm wave-
lengths of the spectrum, is the part of the
spectrum plants use for photosynthesis.
The percentage of PAR transmitted by a
plant canopy can be estimated by the
ratio of PAR under the canopy to ambient
incoming PAR. This ratio is related to
canopy condition and canopy leaf area
and can be combined with growth mea-
surements to estimate growth efficiency,
an important indicator of forest health. PAR
can also be combined with other indica-
tors such as mensuration, vegetation struc-
ture, foliar chemistry, crown assessment,
damage, and/or remote-sensing data to
assess canopy condition with multivariate
indicators.
The PAR indicator addresses the envi-
ronmental value of ecological integrity for
the FHM program. Measurements of solar
radiation intercepted by the canopy are
fundamental to interpreting the productiv-
ity and function of plant communities and
can be used to assess the effects of stres-
sors. Using PAR measurements to esti-
mate leaf area, the following assessment
questions can be addressed:
1. What is the status of forest canopy (in
terms of leaf area) over the region?
2. How is leaf area changing over time
for the region?
3. What is the growth efficiency (i.e., bio-
mass productivity per unit of solar ra-
diation) of forested ecosystems over
the region?
4. How is growth efficiency changing over
time for the region?
This indicator needs additional research on
analyzing the index formula and developing
thresholds (unambiguously interpretable), as
well as evaluating post-stratification procedures
Table 10. Status of Wildlife Habitat Indicator
After First Year of the SE DEMO
Indicator Development Criteria
Status
Unambiguously Interpretable Good
Conceptual Model High
Indices High
Nominal/Subnominal Threshold Fair
Index-Period Stability Good
High Signal-to-Noise Ratio Low
Regional Responsiveness High
Simple Quantification High
Low Environmental Impact High
-------
(regional responsiveness). Off-frame data
and data from the QA Reference Plot
Study will be used to evaluate index-period
variability. The off-frame data should pro-
vide useful variance components for evalu-
ating the criterion of high signal-to-noise
ratio. Based on preliminary analyses, data
from the off-frame research should indi-
cate that the PAR-derived leaf area index
can meet the statistical criteria. The indi-
cator satisfactorily meets the low environ-
mental impact and simple quantification
criteria. Table 11 shows the status of the
PAR indicator with respect to the six crite-
ria.
Conclusions
The FHM program has successfully de-
ployed a full suite of indicators in the south-
eastern U.S. loblolly/shortleaf pine eco-
system during a two-year demonstration
study. This interim report serves to initiate
the indicator evaluation process and ad-
dresses the first objective of the study.
Six criteria have been adopted by FHM
and EMAP for indicator evaluation. The
FHM program has developed an objective
procedure for determining the status of
indicators with respect to these six crite-
ria. After examination of data collected
during the first year of the SE DEMO,
several shortcomings were revealed that
necessitated the implementation of two
additional companion studies, the QA Ref-
erence Plot and Georgia Remeasurement
Table 11. Status of PAR Indicator After First
Year of the SE DEMO
Indicator Development Criteria
Status
Unambiguously Interpretable Good
Conceptual Model High
Indices Fair
Nominal/Subnominal Threshold Low
Index-Period Stability Good
High Signal-to-Noise Ratio Low
Regional Responsiveness Good
Simple Quantification High
Low Environmental Impact High
Studies. These two studies will provide valu-
able information for addressing the
index-period stability and high signal-to-noise
ratio criteria. The data from these two stud-
ies, combined with the second year's data
from the SE DEMO, will be, used to evalu-
ate the status of each indicator. These
findings will be reported in the SE DEMO
Final Report. In the final report recom-
mendations will be made as to which
indicator(s), if any, are ready for national
implementation, i.e., core status. A sec-
ond report, envisioned to be similar to
FHM's Annual Statistical Summaries, will
be prepared. This report will use as many
of the SE DEMO indicators as practical,
to address the second objective of the SE
DEMO study.
The information in the full report was
funded in part by the U.S. Environmental
Protection Agency under Interagency
Agreement No. DW12935103-01 with the
USDA Forest Service, Interagency Agree-
ment No. DW14935509-01 with the USDI
Bureau of Land Management, and Con-
tract No. 68-CO-0049 with Lockheed En-
gineering Sciences and Technologies.
The full report represents data from one
year of field operations of the EMAP. Be-
cause the probability-based scientific de-
sign used by EMAP necessitates multiple
years of sampling, there is uncertainty as-
sociated with these data. This uncertainty
will decrease as the full power of the ap-
proach is realized. Similarly, temporal
changes and trends cannot be reported,
as these require multiple years of obser-
vation. Please note that the full report
contains data from demonstration studies
in one geographic region. Appropriate pre-
cautions should be exercised when using
this information for policy, regulatory, or
legislative purposes.
•frV.S. GOVERNMENT PRINTING OFFICE: IW4 - 5MHWV8017I
-------
-------
-------
-------
TimothyE. Lewis (USDI Bureau of Land Management, Research Triangle Park, NC)
and Barbara L. Conkling (North Carolina State University, Raleigh, NC) are
presently located at the Southeastern Forest Experiment Station, Research
Triangle Park, NC 27709.
Samuel A. Alexander is the EPA Project Officer (see below).
The complete report, entitled "Forest Health Monitoring: Southeast Loblolly/Short-
leaf Pine Demonstration Interim Report," (Order No. PB94-152386; Cost: $44.50,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Forest Health Monitoring National Office
Forestry Sciences Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27709
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268
Official Business
Penalty for Private Use $300
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
EPA/620/SR-94/006
------- |