1Vermont’s Changing Forests
Key Findings on the
Health of Forested Ecosystems from the
Vermont Monitoring Cooperative
October 2009

VMC Staff
Donnie Ager, Web & Data Assistant
Lawrence Forcier, Principal Investigator
Joanna Grossman, Web & Data Manager
Mim Pendleton, Monitoring Technician & Site Operator
Judy Rosovsky, Monitoring Assistant
Carl Waite, Program Coordinator
2009 VMC Steering Committee Members
Anne Archie, USDA Forest Service
Douglas Lantagne, University of Vermont
Ed O’Leary, Vermont Agency of Natural Resources
Dennis May, USDA Forest Service
Steve Roy, USDA Forest Service
Robert Paquin, USDA Farm Service Agency
Charles Scott, USDA Forest Service
Steven Sinclair, Vermont Agency of Natural Resources
Mary Watzin, University of Vermont

2009 VMC Advisory Committee Members
Barbara Burns, Vermont Agency of Natural Resources
Nancy Burt, USDA Forest Service
Brian Keel, USDA Forest Service
Bill Keeton, University of Vermont
Jim Kellogg, Vermont Agency of Natural Resources
Rich Poirot, Vermont Agency of Natural Resources
Beverley Wemple, University of Vermont
Deane Wang, University of Vermont
Sandy Wilmot, Vermont Agency of Natural Resources
Vermont Monitoring Cooperative
Providing the information needed to understand, manage, and protect Vermont’s forested ecosystems
in a changing global environment.
The Vermont Monitoring Cooperative (VMC) was established in 1990. In 1996, a memorandum of understanding
was signed by the Vermont Agency of Natural Resources, the University of Vermont, and USDA Forest Service.
The partners agreed to work together to operate VMC to better coordinate and conduct long-term natural resource
monitoring and research within Mount Mansfield State Forest, the Lye Brook Wilderness Area of the Green
Mountain National Forest, and other relevant areas in Vermont.
The Vermont Monitoring Cooperative works in partnership with the USDA Forest Service State & Private Forestry as
part of the Cooperative Lands Forest Health Management Program. The majority of VMC operations are handled by
staff affiliated with the Rubenstein School of Environment and Natural Resources at the University of Vermont, the
Vermont Department of Forests, Parks & Recreation in the Vermont Agency of Natural Resources, and the USDA
Forest Service Green Mountain National Forest.
VMC Cooperators who participated in this project:
James S. Andrews, Vermont Reptile and
Amphibian Atlas Project
Lesley-Ann Dupigny-Giroux, University of Vermont
Steve Faccio, Vermont Center for Ecostudies
Jennifer Jenkins, University of Vermont
William Keeton, University of Vermont
Ron Kelley, Vermont Agency of Natural Resources
Jim Kellogg, Vermont Agency of Natural Resources
Trish Hanson, Vermont Agency of Natural Resources
Kent McFarland, Vermont Center for Ecostudies
Eric Miller, Ecosystems Research Group, Ltd.
Heather Pembrook, Vermont Agency of Natural Resources
Rich Poirot, Vermont Agency of Natural Resources
Chris Rimmer, Vermont Center for Ecostudies
Donald Ross, University of Vermont
Jamie Shanley, U.S. Geological Survey
Tom Simmons, Vermont Agency of Natural Resources
Christopher Still, University of California, Santa Barbara
Thomas Villars, Natural Resources Conservation Service
Beverley Wemple, University of Vermont
Sandy Wilmot, Vermont Agency of Natural Resources

Vermont Agency of Natural Resources
University of Vermont
United States Forest Service

Vermont’s Changing Forests
Key Findings on the
Health of Forested Ecosystems from the
Vermont Monitoring Cooperative
October 2009

2Introduction
Monitoring Vermont’s Forested Landscape
T he Vermont Monitoring Cooperative (VMC) was established in 1990 to track changes occurring in Vermont’s forests. Only limited information about
the health and baseline conditions of forested ecosystems
was available at that time. Vermont lacked the ability to
perceive subtle changes in ecosystem condition over
time and thus to be able to identify forces affecting forest
ecosystem health and productivity. In addition, there was
no dedicated, centralized, and stable
location for storing, maintaining, and
distributing important ecological data.
VMC was envisioned and created to
collect, assemble, and distribute high-
quality, documented data and information
to better understand environmental
changes and their impacts on forested
ecosystems. Understanding the interactive
nature of environmental changes required
ecosystem-scale, integrated, multi-
disciplinary monitoring and research
based on sound science. Those concepts
lie at the heart of the Vermont Monitoring
Cooperative’s existence.
This report offers a sampling of the
extraordinary amount of information VMC
has assembled in its first 18 years. While
VMC research focuses primarily on the
health of Vermont’s forests, forest ecosystems are complex
entities, affected by weather and climate, by natural and
anthropogenic disturbances, and by the long reach of time.
And everything in the forest relies on a web of connections,
many of which are just now beginning to be understood.
Consider the tiny, reclusive Bicknell’s thrush, a major topic
of VMC research efforts. Living at the top of Vermont in our
most isolated areas, the bird is nevertheless buffeted by
climate change, which alters its habitat both in New England
and its wintering grounds in the Caribbean; by atmospheric
mercury pollution, which has found its way into its blood
and feathers; and by the presence of happy skiers, who
build trails ever higher on the sides of mountains. A goal
of VMC is to learn how people can live side-by-side with the
thrush, the salamander, and the moose, using the manifold
resources provided by Vermont forests, always with an eye
toward a sustainable future.
This report represents the written contributions of 19
cooperators, but collective efforts of dozens of researchers
from varying backgrounds and disciplines working
collaboratively to compile and tell the stories contained
in this document. To the extent possible, the report is a
multidisciplinary synthesis. The majority
of datasets in the VMC data library can be
broadly characterized under the following
section titles of this report: The Health
of Our Forests, Weather and Climate in
Vermont, and Monitoring Atmospheric
Deposition. This report covers topics
including the effects of land-use change
on biodiversity, habitat and population
levels of many animal species, as well as
the effects of alpine development on the
environment. It includes current trends
in meteorological parameters such as air
temperature, relative humidity, and cloud
cover in the Champlain Valley and Vermont.
Also discussed are current conditions
and recent trends relating to atmospheric
deposition, including transport, acidification
of lakes and streams, ozone, and mercury in
the environment. Also explored are factors influencing forest
health, diversity, structure, productivity, and forests’ ability to
store carbon to help mitigate the effects of greenhouse gas
accumulation in the atmosphere.
The information in this report is intended as an
introduction to the body of research that has been amassed
over time, and which is growing daily. Numerous scientific
papers have been published using VMC data, some of which
are referenced here. We have tried to highlight successes
where VMC data have influenced state, regional, or national
policy or where research results have helped alter the
behavior of would-be polluters. As an organization, VMC
believes that it has an important and timely story to tell.
In This Report:
The Health of Our
Forests / 7
Weather and Climate
in Vermont / 20
Atmospheric
Deposition / 29
Selected VMC
Projects / 38
Resources / 39

3A Brief History of the Vermont Monitoring Cooperative
A major Vermont Forest Health Task Force assembled by then-Governor Madeleine
Kunin determined that Vermont needed to continuously monitor forest health and
important environmental variables. In 1990, there were very few models on which to build a
statewide monitoring and research program. That year, through the efforts of Vermont
Senator Patrick Leahy and his staff, the U.S. Congress appropriated $250,000 for the
establishment of VMC. Leaders of the three founding organizations—the Vermont Agency
of Natural Resources, the University of Vermont, and the U.S. Department of Agriculture
Forest Service—agreed to jointly form, nurture, and administer VMC as a long-term natural
resource monitoring program. VMC prided itself on a collaborative approach to
organization and administration, in both its daily operations and its longer-term monitoring
initiatives. VMC encouraged and embraced involvement and guidance from caring and
knowledgeable individuals working outside of the immediate partner organizations. In
1996, the three founding partners signed a formal memorandum of understanding in which
they agreed to work together to operate VMC with the goal to better coordinate and
conduct long-term ecological monitoring and research within the Mt. Mansfield State
Forest, the Lye Brook Wilderness Area of the Green Mountain National Forest, and other
relevant areas in Vermont. In the process of developing the memorandum, the broad and
efficient involvement of many Vermonters and other professionals was recognized as an
important outcome of this phase of the Vermont Monitoring Cooperative, and an element
that should be promoted as VMC continued to evolve.
VMC’s mission is: “To provide the information needed to
understand, manage, and protect Vermont’s forested ecosystems
within a changing global environment.” VMC seeks to serve
Vermont through improved understanding of annual conditions,
long-term trends, and interrelationships of the physical, chemical,
and biological components of forested ecosystems by collecting
and disseminating Vermont environmental data. VMC also
promotes the efficient communication and coordination of
multi-disciplinary environmental monitoring and research
activities among federal, state, university, and private entities
with common interests in the long-term health, management,
and protection of Vermont’s forests.
VMC has become an important database and information
management service for Vermont’s study of forest ecosystems
and environmental quality. The VMC data library contains 300
research and monitoring projects and datasets, collected over
nearly two decades by dozens of cooperators on a wide array of environmental topics (see
page 38 for a sampling of VMC projects and datasets; a more complete listing can be found
at the VMC web site: www.uvm.edu/vmc). VMC’s cooperators range from undergraduate
and graduate students doing research for class projects and theses to university, state,
federal, and private-sector research and monitoring scientists. Through VMC, all results are
assembled in a collaborative effort to advance knowledge and understanding of the
environment and Vermont’s forests. VMC data and information resources are used by
students, natural-resources managers, scientists, lay citizens, and policy makers. VMC
accomplishes its mission through a small professional staff, many other committed
professionals and volunteers, contemporary data management systems, education and
outreach programs, and continuing efforts to support and help coordinate Vermont’s
environmental data interests. VMC continually upgrades its data services to provide the
best data quality and accessibility possible.
VMC’s mission:
To provide the
information needed
to understand,
manage, and protect
Vermont’s forested
ecosystems within
a changing global
environment.
VMC cooperators have
conducted research
on Vermont forests for
18 years, amassing a
wealth of long-term
data.

4 A Steering Committee, whose members represent the three founding partner
organizations, guides the planning, administration, and policy development for VMC. An
Advisory Committee, with members from Vermont’s scientific research, monitoring, and
resource-management communities, helps inform decisions about operations, funding,
and support needs for research and monitoring projects, and advises on direction for the
organization in general.
Mount Mansfield was the first of two study sites selected by VMC for intensive forest
ecosystem monitoring and research (Figure 1). This site comprises over 5,500 acres of state
and university-owned land located between 1,300 and 4,300 feet in elevation. It includes
three watersheds, northern hardwoods and montane spruce-fir forest types, as well as an
alpine ecosystem. VMC utilizes
on-site laboratory facilities at the
University of Vermont’s Proctor
Maple Research Center (PMRC).
At PMRC, VMC operates field
collection sites for the National
Atmospheric Deposition Program
(NADP)/National Trends
Network, NADP/Atmospheric
Integrated Research Monitoring
Network, NADP/Mercury
Deposition Network, U.S.
Department of Agriculture
UV-B Monitoring and Research
Program, and the Vermont Acid
Precipitation Monitoring Program.
Other VMC facilities and
infrastructure include a 66-foot
walkup canopy tower; the
longest continually operating wet
atmospheric mercury monitoring
station that we are aware of in the
world; five meteorological
stations; three stream-gage stations; and a remotely accessed soil climate station. The
Vermont Air Pollution Control Division also operates a
co-located air quality monitoring station at PMRC.
The Lye Brook Wilderness Area, the second VMC intensive study site, provides a
southern Vermont complement to Mount Mansfield. It encompasses 15,000 acres of Green
Mountain National Forest land, ranges in elevation from 900 to 2,900 feet, and supports
northern-hardwood, spruce-fir, and paper-birch forests. The Lye Brook area’s surface waters
identify it as being sensitive to acid deposition. Lye Brook supports a remotely accessed
atmospheric monitoring station (ozone, wet and dry deposition, and meteorology), air
visibility monitoring, and a soil climate station.
VMC data collected and archived during the past 18 years have made significant
contributions toward our knowledge about climate change, land-use fragmentation, and
threats from insects, diseases, invasive plants, and air pollution. VMC data have been used
in Vermont to determine the best time to spray for forest tent caterpillars in order to protect
economically and ecologically important sugar maple trees, while minimizing the negative
effects on the endangered Indiana bat. Data collected by herpetologists tell us that while
numbers of the smooth greensnake may be declining in Vermont, the only viable population
of the North American racer may have already disappeared within the last two years. VMC
researchers found that Bicknell’s thrush tend to avoid crossing open areas such as open
ski trails and that male thrushes may be more vulnerable to predation when crossing these
open areas.
Mount Mansfield
Lye Brook
Figure 1: The major
VMC study sites at
Mt. Mansfield and
Lye Brook Wilderness
Area, indicated by
dots, are shown here
within the context of
the biophysical regions
of Vermont.
Source: Sandy Wilmot and
Beverley Wemple

5 VMC-supported projects have shown that high-elevation developments increase annual
water, sediment, and chloride yields over undeveloped watersheds. Also, we now know that,
on average, sugar maple budbreak is three days earlier and leaf-out five days earlier this decade
than in the 1990s; the 2008 budbreak was 12 days earlier and leaf-out was nine days earlier
than the baseline. A map of forest soil carbon assembled by VMC researchers which shows
areas of high or low carbon may suggest future guidelines for managing soil carbon. In a 2007
lawsuit filed by the U.S. Environmental Protection Agency and eight northeastern states
against a large midwestern utility company, data from PMRC in Underhill helped provide
compelling evidence that the company was keeping outdated power plants in service to avoid
the costs of producing cleaner power; in an out-of-court settlement, the company agreed to
spend $4.6 billion to retrofit several old power plants. Because of the pioneering work done
by VMC cooperators, Vermont is a strong candidate for a role as a pilot site in an anticipated
national mercury biomonitoring network. The Underhill monitoring site known as VT99 was
recently selected and funded by EPA as one of the initial sites in the new NADP/Mercury
Trends Network.
The work of VMC has helped Vermont’s leadership role on the national stage in certain
environmental issues. During the potentially uncertain times brought about by climate change,
continuing threats from air pollution, and changes in land use both locally and nationally, it is
more important than ever to support monitoring of environmental variables and conditions
and to efficiently and effectively share scientifically robust data among scientists, resource
managers, policy makers, and the public in an unselfish spirit of cooperation. Ecosystem
health is an essential element to achieve sustainability. The need for useful indicators of
forest ecosystem health recognized by Vermont and federal leaders 20 years ago is even
more important in 2009.
8
7
6
5
4
3
2
1
0
198
9
Hermit Thrush Decline
199
0
199
1
199
2
199
3
199
4
199
5
199
6
199
7
199
8
199
9
20
00

20
01

20
02

20
03

20
04

20
05

20
06

A
nn
ua
l I
nd
ex
Annual Trend = -6.3% (SE 0.024)
P = 0.011
The State of Vermont’s State Bird: Hermit Thrush Decline
R
Figure 2: Populations of hermit thrush have
declined by 6.3 percent annually at more than
two dozen VMC study sites.
Source: FBMP
esults from the VMC-supported Vermont Forest
Bird Monitoring Program (FBMP) indicate that
Vermont’s state bird, the hermit thrush, declined by
an average of 6.3 percent annually between 1989 and 2006
(Figure 2). FBMP monitors breeding birds at more than two
dozen forested study sites throughout the state, including
sites at Mt. Mansfield and Lye Brook. Several factors may
have contributed to this long-term decline, including habitat
alterations from deer overbrowse, soil calcium depletion
due to acid rain, and
habitat loss, especially on
the bird’s southeastern
U.S. wintering grounds.
n Hermit Thrush
During the
potentially
uncertain times
brought about
by climate change,
continuing threats
from air pollution,
and changes in
land use both
locally and
nationally, it is
more important
than ever to
support monitoring
of environmental
variables and
conditions.

6Figure 3: More than 80 percent of Vermont’s
forests are privately owned.
Source: Adapted from Wharton et al. 2003
The Landscape Over Time
T
Distribution of Timberland Area by Ownership
Individuals:
63%
Federal: 6%
Other Public: 8%
Forest Industry: 6%
Farmers: 6%
Corporations: 11%
he history of the Vermont landscape is marked by the influence of geologic forces that gave rise to the state’s
mountains over 400 million years ago and climatic forces that have produced dramatic changes in
environmental conditions. Roughly 11,000 years ago, glacial ice that had extended over the state for nearly two
million years began to retreat and trees began to appear on a landscape stripped bare of vegetation. These earliest
forests were dominated by black spruce and paper birch. Animal communities became established in new habitats
made available in the emerging forests. As the climate warmed, tundra vegetation retreated to only the state’s
highest peaks, while spruce and fir dominated the higher elevations and northern mountains. Lower elevations came
to be dominated by eastern white pine, maple, birch, hemlock, beech, oak, and hickory, creating today’s mosaic of
the mixed northern hardwood forest. Today forests cover roughly 4.6 million acres or roughly 78 percent of the state
(Wharton et al. 2003).
The current predominance of young to mature forests is an artifact of 19th-century clearing, subsequent
land abandonment, and secondary forest redevelopment on old-fields. With these changes have come shifts in the
types of ecosystem services provided by forested landscapes. For example, young to mature northern hardwood
forests provide less desirable habitats for late-successional wildlife species, lower levels of biomass and associated
carbon storage, and less dramatic effects on aquatic habitat structure in forest streams (Keeton et al. 2007). On the
other hand, a young, maturing forest has higher growth rates, providing opportunities for sustained yield timber
production. In addition, open lands and young forests provide abundant habitats for early-successional species,
some of which are now declining due to forest redevelopment.
Compared to the primary (or never cleared) forest systems, the secondary forests that have recovered across
Vermont’s landscape tend to have less complex canopies, lower densities of large trees (both live and dead), lower
volumes and densities of downed logs, smaller canopy gaps, and less horizontal variation in stand density (McGee
1999, Keeton et al. 2007). At larger scales, our land-use and forest-management history has converted landscapes
with complex patch mosaics, which are shaped by wind and other disturbances, to simpler configurations. Forest
patches are now less diverse in size and less complex in shape (Mladenoff and Pastor 1993). The relative abundance
of dominant tree species has also shifted as a result of land-use history (Cogbill et al. 2002). As we evaluate the
current conditions and trends in forest ecosystem health indicators, we need to keep in mind that land-use history
has profoundly shaped our current forested landscape.
Modern changes to the Vermont landscape accompany dramatic changes in the state’s population over the
last few decades. The population boom in the United States following the Second World War did not reach Vermont
until later, with a 14 percent population growth in the 1960s and a 15 percent growth in the 1970s, and rates slowing
to 10 percent or less in more recent decades. With this population growth, came a change in urbanized land,
which has grown by roughly 20 percent since 1960. Nearly one-third of urbanized land has come from conversion
of agricultural land and nearly two-thirds from conversion of forest land. Although much of this population and
development pressure has been focused in the Champlain Valley and more populous towns of the state,
a significant feature of Vermont’s recent growth has been a growth
in recreational development and second homes in Vermont’s
resort towns. Demographic changes and pressures associated
with property taxation have led to changes in land ownership and
increased parcelization of forest land across the state. Today, more
than 80 percent of Vermont’s forests are privately owned (Figure 3).
These changes in population, land ownership, and land use have
important implications for forest ecosystem health.
A 1999 EPA report estimates a loss of approximately 35
percent of our wetlands since European settlement, with more
recent annual losses of 200 to 400 acres per year (EPA 1999).
Undeveloped land provides valuable ecosystem services.
Degradation, fragmentation, or loss of these natural systems
adversely impacts forest ecosystem health and compromises the
ability of forests to provide habitat, clean water, pure air, and other
critical resources upon which we rely.

7 istorically forests were considered healthy simply
if they supported an abundance of healthy trees.
In a broader context, forests are ecosystems
supporting many organisms in addition to trees, and whose
functions are more inclusive than just human exploitation
of its resources. Forest ecosystems provide services to
humans without which we could not live. They purify
our air by capturing pollutants. They provide shade and
windbreaks that regulate and moderate temperatures. They
protect waters by purifying and keeping them cool. When
storms come, they mitigate the impacts from flooding.
They capture nutrients from the air, and renew soil fertility.
Tree roots adhere to soils and prevent major soil erosion
and sediments leaching into streams and lakes. Forests
provide habitats for game and nongame animals, harbor
natural control agents against pests, provide resilience
from disturbances. And they allow humans to benefit
economically from recreation, aesthetics, wood harvesting
and other forest products. A healthy forest ensures the
provision of these ecosystem services. A healthy forest
system is also dynamic in response to natural climate
variability, disturbances, and succession. But changes
in structure or function outside the historical range of
variability may signal stress.
Recent international agreements known as the Montreal
Process Criteria and Indicators of Forest Sustainability
have formed a foundation for describing and measuring
forest sustainability across the landscape. Five of these
indicators pertain to forest ecosystem health: biodiversity,
productivity, soil and water resources, carbon cycles, and
disturbances. These provide a framework for describing
current conditions and trends in forest ecosystem health
with particular focus on studies by the Vermont Monitoring
Cooperative (VMC).
THE HEALTH OF OUR FORESTS
H Biodiversity Biodiversity encompasses not only the composition of species, but also the complexity of structural features within
a forest stand and across landscapes. Vermont hillsides are
a patchwork of forests with trees of various sizes and ages.
This diversity of tree sizes and ages makes forests adaptable,
dynamic, better situated to recover from disturbance, and
also provides a broader range of habitats for other
organisms. One advisory group of Vermont foresters and
ecologists suggested that a healthy proportion of tree sizes
in a mature forest would be approximately 50 percent
saplings, 30 percent pole-sized trees, and 20 percent
sawtimber-sized trees (VTFPR 1999). Two time periods of
statewide forest inventory data show an increasing proportion
of trees in the smaller tree category (Figure 4).
Special characteristics of forests can make them suitable
as habitat for animal species. Vernal pools, caves, coarse
woody material, and legacy trees are examples of unique
structural elements that provide habitat for animals and
allow greater potential diversity. When animal survey data
are not available, habitat measures can be used as predictors
of potential species diversity. VMC data have been valuable
in building connections between animal-species diversity
and habitat features. Among other examples of this data
are VMC’s forest bird survey and moth-host plant database,
both of which are discussed in this report.

Composition of Tree Sizes in Vermont Forests
80
60
40
20
0
24.3
8.2
19.1
7.9
1983 1997
67.5 73
Figure 4: Studies in 1983 and 1997 show an increasing
percentage of Vermont trees are in smaller-size categories.
Source: VTFPR 1999

8Non-native Pests and Invasive Plants
ermont has been subject to many introductions of exotic insect
and disease organisms that have caused damage to forests
(Table 1). Some organisms have ecosystem-altering effects,
such as those responsible for Dutch elm disease and chestnut blight.
In other cases the impacts are slower, offering opportunities to develop
management strategies to mitigate impacts. VMC has been instrumental
in providing ongoing detailed information on non-native organisms, such
as pear thrips, gypsy moths, and beech bark disease, which are now well
established in Vermont, and has helped in understanding relationships
between these pests and environmental factors. For example, pear thrips
have been found to cause more damage when cold weather slows the
development of sugar maple leaves.
Other non-native insects are moving toward Vermont, and could cause
significant ecological effects. These include the emerald ash borer, with new
finds in Quebec; the Asian long horned beetle, a maple pest now in Worcester,
MA; and the hemlock woolly adelgid, now in forests in southeastern Vermont.
Long-term monitoring by VMC provides baseline data which will help assess
the impact of these insects on host species and associated organisms.
Invasive non-native plants, such as barberry, buckthorn, and honeysuckle,
continue to expand northward in Vermont forests, causing negative impacts on biodiversity and forest regeneration.
Forest monitoring plots, such as those measured at the VMC research sites, provide a mechanism to survey trends
in invasive plants and their long-term effects on forest health. The North American Maple Project survey results
highlight the growing incidence of invasive plant species in forests across Vermont (Figure 5). At the VMC sites at
Mt. Mansfield and Lye Brook, which are more removed from residential areas and their associated plantings of
non-native plants,no exotic invasive plants were found in a 2005 survey at 20 monitoring plots.
Table 1:
This list
indicates
the status of
non-native
insects and
diseases in
Vermont.
Source:VTFPR
60
50
40
30
20
10
0
Barberry Buckthorn Honeysuckle Multiflora Rose
Non-native Invasive Plant Species
Pr
es
en
ce
o
n
Pl
ot
s
(%
)
Figure 5: There is a growing
incidence of non-native invasive
plant species on monitoring plots
statewide.
Source: North American Maple Project
V
Non-native forest insects Status in Vermont
Japanese beetle Popillia japonica Established
Elm leaf beetle Pyrrhalta luteola Established
Pine shoot beetle Tomicus piniperda Recently detected
Eastern spruce gall adelgid Adelges abietis Established
Balsam woolly adelgid Adelges piceae Established
Hemlock woolly adelgid Adelges tsugae Recently detected
Beech scale Cryptococcus fagisuga Established
Introduced pine sawfly Diprion similis Established
Birch leafminer Fenusa pusilla Established
European spruce sawfly Gilpinia hercyniae Established
Mountain-ash sawfly Pristiphora geniculata Established
Larch casebearer Coleophora laricella Established
Gypsy moth Lymantria dispar Established
Pear thrips Taeniothrips inconsequens Established
Introduced basswood thrips Thrips calcaratus Established
Sirex wood wasp Sirex noctilio Recently detected
Emerald ash borer Agrilus planipennis Not detected
Asian long horned beetle Anoplophora glabripennis Not detected
Non-native forest diseases
Scleroderris canker Ascocalyx abietina Established
Chestnut blight Cryphonectria parasitica Established
White pine blister rust Cronartium ribicola Established
Dogwood anthracnose Discula destructive Established
Beech bark disease Nectria coccinea Established
Dutch elm disease Ophiostoma ulmi Established
Butternut canker Sirococcus clavigignenti-juglandacearum Established
Sudden oak death Phytophthora ramorum Not detected

9 Vermont is estimated to be home to 441 species of birds, mammals, amphibians,
and reptiles. A majority of these species are dependent on forests for all or part of their
life cycles. As we struggle to balance procurement of forest products with conservation
of wildlife, new models of forest management are emerging that protect specific
“focus” animal species. At VMC research sites, we have documented many forested
species and their associated habitat requirements. Biodiversity audits can help assess
the relative environmental impacts of various forestry and land-development practices
and of climate change. While biodiversity in communities is often seen as desirable,
it is important to remember that some ecosystems, such as in bogs or sandplains,
may have relatively low diversity but harbor highly valued species that are unusual
from a regional perspective.
Insect Diversity
While the enormous diversity and sheer numbers of insects excites entomologists, the
general public is familiar with only a relatively few species that have economic or health
impacts. Comparatively little or nothing is known about other insects. Insects comprise over
two-thirds of the approximately 2 million known species of living things on the earth, and
millions more species remain to be discovered. There are about as many species of butterflies
and moths on Mt. Mansfield as there are breeding bird species in all of Canada!
There are currently known to be approximately 2,000 species of butterflies and moths
(Lepidoptera) in Vermont, which reflects the great variety of habitats found here. The long,
varied, and sometimes arduous work that has produced this number involves methodical field
procedures combined with meticulous record-keeping. Additionally, existing curated insect
collections have been an invaluable source of baseline environmental data, with each specimen
vouching for the historical occurrence of a species at a particular time and place. Collectively,
this information allows us to retroactively track local arrivals and extinctions of various species,
and frames the study of the effects of human disturbance and climate change. For several
years in the 1990s, and again a decade later, Lepidoptera were surveyed at three elevations on
VMC sites on Mt. Mansfield. These efforts resulted in a count of close to 400 butterfly and
moth species. In the food-web, Lepidoptera larvae are a primary food source during the nesting
periods of the majority of birds that breed in Vermont.
What can insect communities tell us about the relative ecological health of an area?
Among other things, insects have roles in forests as pollinators, decomposers of leaf litter, and
as prey for breeding birds. Documenting the insect species living in a given habitat, along with
their immediate networks of organisms, brings us closer to evaluating the health of that natural
community. Monitoring indicator species may shed light on the vigor of a natural habitat. For
example, we know that the aquatic larvae of certain species of black fly are associated with
clean water, or that given species of springtails occur within a specific pH range.

Productivity
Forest productivity, usually thought of as growth and abundance, is an endpoint of many
functional processes of forests that in one way or another contribute to growth. Measures
of forest processes include plant growth (biomass), reproductive success, timing of
developmental stages (phenology), and nutrient dynamics in air, biota, soil, and water.
Forest biomass is a measure of the effectiveness of the photosynthetic process. When
growth exceeds mortality, biomass accumulation increases. But if disturbances impact the
balance of growth to mortality, biomass accumulation decreases. At VMC’s Mt. Mansfield
site, there has been a decrease in above-ground live tree biomass on the east slope and
summit since 1997 (Figure 6), evidence of the impacts from a variety of stress events on
forest productivity. These events included the ice storm in January 1998 followed by excessive
moisture in the summer of 1998 and marked drought in 1999.
There are as many
species of butterflies
and moths on Mt.
Mansfield as there
are breeding bird
species in all of
Canada.
Atlantis fritillary

10
The timing of developmental stages provides a
baseline for comparing annual changes and allows
evaluation of stress impacts on functional processes
of plants and animals. When managing forest pests,
the timing of developmental stages of pest and host
is essential (see page 23 for a discussion of phenology
monitoring).
Soil and Water Resources
The soils supporting forests play a crucial role
in forest health, providing a myriad of services
ranging from nutrient recycling to physical support.
Characteristics important to forest health include
soil texture, drainage, depth, and fertility. With the
exception of soil fertility, these soil characteristics
don’t change significantly over time. Tunbridge soils
are typical of Vermont uplands, and not surprisingly
support the majority of northern hardwood forests (Table 2). Depth to bedrock or hardpan and
soil drainage affect root growth and availability of soil moisture, and determine areas that will
be most affected by droughts. Forest lands have been altered through time for a wide variety
of uses, and not all forests are growing in sites that represent their ideal growing conditions.
Soil-nutrient availability is a common measure of forest health, and where soil pH, base cations,
or aluminum concentrations are abnormal, ecological impacts can occur (see page 34 on
critical loads).
VMC supports a program of research and long-term monitoring
of streamflow and water quality dynamics of high-elevation forested
streams within the Lye Brook Wilderness Area and on the slopes of
Mt. Mansfield. These studies have shown that both the flow of water
and concentration of some solutes such as nitrate are highest during
snowmelt. As a result, most of the annual loss of these solutes from
forest to stream system is found during a relatively short time period
in the spring. This ongoing VMC monitoring will aid the detection of
environmental change due to changing climate and high-elevation
development such as ski resorts.
Carbon Cycles
One of the major greenhouse gases is atmospheric carbon in
the form of carbon dioxide. Through photosynthesis, CO
2
is removed
from the air by trees and forest plants, and stored as carbon in roots,
stems, and foliage. Forests play a huge role in carbon dioxide mitigation
in Vermont, so knowledge of current carbon storage and release is
essential to learning how to better manage our carbon budget. Carbon
can be stored for long time periods in live tree biomass, below ground in roots, in soils, and
in trees harvested for durable wood products. Vermont forests are being considered for future
policies aimed at sequestration of additional carbon emissions in attempts to reduce our
carbon footprint. In calculating carbon sequestration, researchers must consider a variety of
complex factors, including amount of carbon stored in forests, rate of carbon accumulation,
amount of biomass extracted for fuel, biomass used for durable wood products, impacts of
forest management on above- and below-ground forest carbon, and disturbances such as
insect defoliation.
160
140
120
100
80
60
40
20
0
B
io
m
as
s
(M
g/
ha
)
East West Summit
Above-ground Live Tree Biomass
Years
1993
1997
2001
2005
Natural Community Type Soil Name
Northern Hardwoods Tunbridge
Cabot
Berkshire
Lyman
Peru
Marlow
Rich Northern Hardwoods Buckland
Vershire
Dutchess

Oak-hickory Georgia
Stockbridge
Paxton
Pittsfield
Table 2: Common soil
series associated with
some of Vermont’s
forest communities, in
order of importance.
Source: Report by the
Governor’s Task Force on
Climate Change, 2007
Figure 6: At VMC’s Mt.
Mansfield site, there
has been a decrease in
above-ground live tree
biomass on the east slope
and summit since 1997.

11
Greenhouse Gas Mitigation
R apidly developing domestic and international carbon markets recognize three general possibilities
for forest carbon management. These are reforestation/
afforestation, avoided deforestation, and improved forest
management (Ruddell et al. 2007). The latter is concerned
with the carbon stored in managed forests, and focuses on
the concept of “additionality” or the potential for increasing
net carbon storage over a baseline level. But this has
proved challenging for scientists and forest managers alike,
particularly because of the complex carbon accounting
required to determine the net effects of a particular
management approach (Ray et al. 2009).
Actively managed forests provide carbon sequestration
benefits both within the forest ecosystem and in harvested
wood products. Biomass fuel produced as a by product of
forest management activities can help offset greenhouse
gas emissions from fossil fuels. Ongoing research is
exploring how to design forest management strategies
that optimize storage among these sinks. Rapidly growing,
younger or well spaced forest stands may have higher
rates of carbon uptake, but they have lower biomass per
unit area compared to older or less intensively managed
forests (Figure 7), and thus actually store less carbon than
high biomass forests with lower or stable rates of carbon
uptake (Harmon and Marks 2002). Research has shown
that as forests age they store more carbon, due to very high
levels of accumulated above and belowground biomass
(Keeton et al. 2007; Luyssaert et al. 2008). Forested
landscapes recovering toward an older, higher biomass
condition will store much higher quantities of carbon than
landscapes dominated by young to mature hardwood
forests (Rhemtullaa et al. 2009). Different management
approaches (frequency and intensity) result in different
amounts of average carbon storage over the long term.
Thus the choice of harvesting approach directly affects
not only emissions offsets but also long-term carbon
storage dynamics.
Research at UVM by VMC cooperators has used
simulation modeling to examine the impact of harvesting
frequency and proportion of post-harvest structural
retention on carbon storage and the significance of
including harvested wood products in carbon accounting
(Nunery and Keeton, in review). Carbon dynamics were
simulated under nine forest management scenarios,
spanning a range of increasing structural retention
and decreasing harvesting frequencies, including “no
harvest.” The simulations incorporated carbon flux
between aboveground forest biomass (dead and live
pools) and harvested wood products. The results
suggest that intensified regeneration cutting reduces
net carbon storage. Conversely, extended rotations or
entry cycles, high levels of post-harvest retention, and
practices favoring production of durable wood products
enhance average storage over multiple rotations or
entry cycles. The highest levels of storage are actually
achieved by passive management, even factoring in the
foregone storage in wood products. These conclusions
are valid only so long as the analysis does not include
the greenhouse gas emissions that would result from
substituting non-wood construction materials for wood
if production of the latter were reduced. Other work has
shown that factoring in “substitution effects” can make
some intensive management scenarios comparable to
the least intensive in terms of net carbon sequestration,
depending on the assumptions built into the analysis
(Malmsheimer et al. 2008).
To
ta
l C
ar
bo
n
St
or
ed
Stand Development Over Time
Figure 7: This figure shows carbon sequestration and
storage in relation to forest stand development. Note
that the highest rates of uptake occur in early to mid
stages of development, but the greatest levels of storage
are achieved late in development, resulting in a
substantial carbon reservoir.
Source: Jared Nunery and Bill Keeton, UVM

12
The estimated annual,
statewide accumulation of CO
2

in forests is 9.63 million metric
tons of carbon dioxide equivalent)
(Table 3). Keep in mind that in
1990, Vermont CO
2
emissions
were 8.1 million metric tons,
and by 2005 had increased to
9.1. Without emission-reduction
policies, emissions could grow to
10.67 million metric tons by 2030,
beyond forest-sequestration rates.







The two major carbon pools in forests are live trees and
soil. Forests that continue to increase live tree biomass are
considered sinks for atmospheric carbon. When the rate
of sequestration decreases to a point where net growth is
less than mortality and decomposition, such as in areas of
forest decline, forests become a source of carbon dioxide
emission. The VMC plot data shows that some locations on
Mount Mansfield may currently be a source of carbon dioxide
emissions, due to drought-induced mortality and a trend
toward decreasing biomass. This, however, could be reversed
as existing trees grow and new trees replace dead trees.
Soil carbon is less well studied. Vermont estimates may
be low because temperate forests are estimated to store nearly
double that of above-ground vegetation. Most of the soil
carbon in Vermont is organic carbon, allowing estimates of
soil carbon based on organic matter content. VMC researchers
assembled a map of forest soil carbon for Vermont (Figure 8);
areas of high or low carbon may suggest future guidelines
for managing soil carbon.
Disturbances
Forest health monitoring measures the ability of trees to
recover from damages inflicted by natural and anthropogenic
disturbances. Natural disturbances are an ecologically
important and intrinsic part of Vermont’s forested ecosystems. They shape the types, quantities,
and spatial distributions of habitats and strongly influence successional processes. In
contemporary ecology, natural disturbances are viewed as a “subsidy” to the system, creating
critical habitat structures and driving a host of ecological processes, including soil and nutrient
turnover, organic matter recruitment into streams, and carbon storage dynamics. But when
disturbance dynamics are perturbed by humans (for example, when people’s actions contribute
to severe flooding or large-scale insect outbreaks), then deleterious stress can be induced.
Disturbances also interact with human-induced stressors, such as acid deposition and climate
change, in determining trajectories of forest ecosystem change (North and Keeton 2008).
Vermont Forests Stored Carbon Annual
(MMt) Accumulation
of CO2 (MMt)

Soil 139 0.7
Forest floor 45.7 0.5
Down dead 12.2 0.4
Understory 3.2 0.03
Standing dead 11.4 0.3
Live trees 172.2 6.3
Wood products 1.4
Total 383.7 9.63

Table 3: Estimates of carbon stored in Vermont forests
and the annual removal of CO
2
from the atmosphere
through forest sequestration.
Figure 8: This map of
forest soil carbon in
Vermont, assembled by
VMC researchers, may
help develop guidelines
for managing forest
carbon in the future.
Source:Wilmot et al. 2008
Vermont Forest Soil Carbon

13
A variety of natural disturbance agents, including wind, ice,
insects, fungal pathogens, beavers, floods, and fire, have sculpted
our forested landscapes for centuries. Events that create gaps in
the canopy are the most common type of disturbance in Vermont’s
forests. Disturbance gaps usually involve death or damage to
individual or small groups of trees. Depending on size and
orientation, gaps can result in regeneration of intermediate to shade
tolerant species, release of advanced regeneration, or competitive
release and accelerated growth in proximate overstory trees.
In 1985, surveys of Vermont’s hardwood resources showed
nearly 14,000 acres of dead trees (Figure 9). Although subsequent
resurveys showed that acres of dead trees decreased and percent
of healthy trees improved to 90 percent or better, concern about
sugar maple tree health across the region—caused potentially by
impacts of acid deposition and the ill effects of tapping trees for
maple syrup production—led to the establishment of a Canada-United States sugar maple
health monitoring program (NAMP). In Vermont, one of the monitoring sites was at the base
of Mount Mansfield.
Sugar maple health monitoring has
provided solid information on trends in stress
events and recovery for this prominent tree
in Vermont forests. While initially this survey
showed areas of tree decline, with only 81
percent of sugar maples in Vermont plots
considered healthy, tree health improved in
the early 1990s (Figure 10). Tree health dipped
again in recent years following an outbreak of
forest tent caterpillar.
In 1990, a systematic forest health detection
program started in New England and grew to
become the national Forest Health Monitoring
program, aimed at detecting emerging
regionally significant forest health problems.
We have adopted key ecosystem methods
from this program for use on many of our
VMC monitoring plots. VMC provided a unique
opportunity to co-locate these forest health
measurements with those of atmospheric, weather, wildlife, water, and soils conditions, which
has improved our understanding of relationships among environmental changes, forest health,
and ecological dynamics (see Ozone, page 36).
A massive ice storm in
1998 devastated large
portions of Vermont’s
forests.
17,500
14,000
10,500
7,000
3,500
0
1985 1990 1995 2000
Acres of Hardwood Mortality in Vermont
Year
Figure 9: In 1985, a survey of Vermont’s
hardwood resources showed nearly
14,000 acres of dead trees. In later years
this acreage decreased dramatically.
Source: Kelley et al. 2002
25
20
15
10
5
0
MmcVi HVfea AaVeld
a  ,
1Y99 1Y9Y 1YY0 1YY1 1YY2 1YY3 1YY4 1YY5 1YY7 1YY8 1YY9 1YYY 2000 2001 2002 2003 2004 2005 2007 2008
e
r !
!
Figure 10: Sugar
maple health
monitoring has
provided long-
term information
on trends in stress
events and recovery.
While initially only
81 percent of sugar
maples were healthy,
tree health improved
in the early 1990s.
Source: Vermont Plots,
NAMP
Forest tent caterpillar

14
Precipitation is one of the key drivers of
forest health, both directly and indirectly. Trees
need adequate moisture for growth, but low
precipitation has also been correlated with rising
populations of insect defoliators. Periods of poor
sugar maple tree health have corresponded with
below-normal precipitation over the past two
decades. Following the 1999 and 2001 droughts,
forest tent caterpillar populations exploded,
resulting in extensive defoliation. Stress from a
variety of disturbances has affected sugar maple
health at Mt. Mansfield and around Vermont as
seen in the percent of trees with thinner than
normal foliage (Figure 11).
Monitoring has been instrumental in
determining thresholds for tree recovery. Periods of stress often lead to crown dieback, but trees
have the ability to recover if the stresses abate. Fertile sites high in calcium tend to recover more
quickly. This ability also varies with species. Sugar maple is good at recovering; yellow birch has
a moderate ability to recover, but paper birch has poor stress recovery. This information has
helped forest managers determine how and when to thin forests to maintain healthy, vigorous
trees, and was especially helpful in salvage cutting following the destruction from the 1998 ice
storm (Kelley et al. 2002). Tree recovery at VMC sites has followed this same trend; sugar maple
recovery from the ice storm and from forest tent caterpillar defoliation has been good, but white
birch recovery from ice damage and drought has not been as successful.
At the VMC study sites, weather factors play a significant role in shaping the growing
conditions of forests. Mt. Mansfield has elevations ranging from 1,300 to 4,300 feet and Lye
Brook Wilderness Area peaks at nearly 3,000 feet.
Winter storms are severe, with high winds and ice
covering trees. Soils are quite shallow in portions
of upper elevation forests. Also, air at these upper
elevations is laden with high amounts of acidic
compounds, ozone, and mercury. So monitoring
the health of these forests can show the most
extreme forest impacts from stress events, but
can also provide early indications of potential
problems that will affect a wider area of
Vermont’s forests.
Forests of the high-elevation spruce-fir
zone on Mt. Mansfield have been more prone
to mortality than other locations. The balsam
fir-dominated forests consistently have high
dieback and damage from broken branches.
Annual monitoring has also shown that heavy cone-producing years, while benefiting squirrel
populations, is detrimental to crown health and indirectly to bird reproduction, as squirrels are
predators of bird eggs. Birch trees are particularly vulnerable to drought. On the east slope of
Mt. Mansfield, birch condition has declined since the severe drought of 1998-99 (Figure 12).

50
45
40
35
30
25
20
15
10
5
0
199
2
Percent of Trees with High Dieback
199
3
199
4
199
5
199
6
199
7
199
8
199
9
20
00

20
01

20
02

20
03

20
04

20
05

20
06

20
07

20
08

Balsam fir
Birch
Figure 12: This figure
indicates dieback of
birch and balsam fir
on Mt. Mansfield. A
decade of balsam fir
recovery follows a past
disturbance. Birch
trees are particularly
vulnerable to drought,
shown here in the
decline of birch since
the severe drought of
1998-99.
Source: Sandy Wilmot,
VTFPR
50
45
40
35
30
25
20
15
10
5
0
198
8
bhght wi ckf afrh ehh wt awtowotk brw on enot Pwrofkh
bhf eno Tgh cws
Hwkn Pwh eht Dfhorrf
198
9
199
0
199
1
199
2
199
3
199
4
199
5
199
6
199
7
199
8
199
9
20
00

20
01

20
02

20
03

20
04

20
05

20
06

20
07

Figure 11: Stresses from
a variety of events have
affected sugar maple
health. Following the
1999 and 2001 droughts,
for example, forest tent
caterpillar populations
exploded, resulting in
extensive defoliation.
Source: Vermonts Plots,
NAMP

15
orest managers in Vermont must contend with
myriad hot button issues. These include how to deal
with exotic diseases and insects, adapt to climate change,
safeguard riparian systems, minimize impacts associated
with forest roads, manage competing recreational interests,
and conserve forest biodiversity. Keeping working forests
economically viable, conserving open space, and
discouraging parcelization and sprawl are major concerns.
But 21st century foresters also face new opportunities, such
as growing interest in community-based forestry, forest
certification, rapidly developing carbon markets, and rising
demand for sustainably produced biomass fuels. These
opportunities may create economic and other incentives
for sustainable forest management, but also require
balancing tradeoffs between competing resource values.
An example is the inherent difficulty of how best to allocate
biomass among uses such as on-site carbon storage,
wood products, fuel, and habitat structure.
A central challenge in sustainable forest management
has been determining the best mix of management
approaches most capable of providing a broad array of
ecosystem functions while meeting landowner objectives.
VMC scientists are conducting research on a promising area
known as disturbance based silviculture, which emulates
naturally occurring forest processes (Keeton 2006).
A general finding of this research is that late-
successional characteristics in northern hardwood—
including vertically complex canopies and gaps—can be
promoted through a variety of modified silvicultural
approaches. But these management techniques are less
effective in other ways, tending to significantly inhibit certain
stand development processes, such as recruitment of large
trees, downed coarse woody debris, and the highest levels of
biomass and carbon storage. The results signal that
treatments can be modified to retain and even enhance
important elements of stand structural complexity, but of
course this involves tradeoffs in terms of reduced harvest
volumes (Figure 13).
The VMC-sponsored Forest Ecosystem Management
Demonstration Project (FEMDP) has also shown that low-
intensity, disturbance-based management can provide an
intermediate level of carbon storage, providing a margin of
enhanced on-site carbon sequestration compared to
conventional selection harvesting. Aboveground biomass is
predicted to increase over the next 50 years under all the
treatments, including controls. None of the experimental
stands is likely to attain the biomass they would have without
treatment, but some techniques are projected to attain 91.4
percent of the biomass levels of untreated stands.
FEMDP researchers have monitored a variety of wildlife
as indicators of biodiversity responses to sustainable forest
management. As predicted, responses to the silvicultural
treatments vary according to the habitat associations of
different groups of species. For instance, some early
successional bird species appear to have benefited from the
variably sized openings created by group selection logging,
while all the treatments have been effective in maintaining
the habitat needed by interior dwelling and late-successional
birds (Strong, unpublished data).
There are clear economic tradeoffs involved in modifying
silvicultural treatments to promote stand structural complexity
(Keeton and Troy 2006). FEMDP assessments examined
timber-related revenue only, whereas a more complete
economic analysis might include other potential revenue
sources, such as carbon credits. The profit margins incurred
by all the treatments were highly sensitive to site quality and
market conditions. Where these are poor, lower intensity
harvesting may generate only enough profit to cover expenses.
This might be acceptable in a limited number of settings,
for instance where disturbance-based approaches are employed
for restoration purposes in reserves. For commercial harvests,
however, the experimental approaches would be marketable
only where site quality is moderate to high and market
conditions are favorable. Under those conditions the
experimental treatments offer alternatives that provide
revenue from low-intensity harvest while also meeting
ecological management objectives.
Research on Sustainable Forest Management
F
P r e -H a r v e s t
P r e -H a r v e s t
P o s t - H a r v e s t
P o s t - H a r v e s t
Single-Tree Selection Unit
Structural Complexity Enhancement Unit
Figure 13: Visualization of stand structure changes
associated with two of the experimental treatments in
the VMC Forest Ecosystem Management Demonstration
Project.
Source: Keeton 2006
PRE-HARVEST
PRE-HARVEST
POST-HARVEST
POST-HARVEST

16
Biomonitoring
data collected on
mountain streams
below ski resorts
show that increased
sediment and
fluctuating pH can
affect the productivity,
abundance, and
species richness of
macroinvertebrates.
Ski Resorts and Land-Use Disturbance
Disturbances caused by conversion of upland forests to developed uses may have more
serious environmental implications than at low elevations. As Vermont ski resorts expand and
diversify activities over four seasons, their ecological impacts may be intensifying. Mountain
ecosystems are also increasingly subject to other disturbances and stresses such as climate
change, mercury deposition, acid precipitation, and development of wind and communication
towers. VMC-supported wildlife research has focused on three aspects of mountain
development: direct habitat loss, fragmentation, and modification. Hydrologic and aquatic
studies have examined the role of mountain development on streamflow, water quality, and
aquatic biota.
The response of wildlife species to forest fragmentation and loss is dictated by factors
such as home range size, sensitivity to habitat edge effects, and gap-crossing ability. Theoretical
modeling of bird species on two existing Vermont ski resorts, Stratton Mountain and Mount
Mansfield, showed dramatic decreases in population sizes as fragmentation and edge effects
increased (Strong et al. 2009). Modeling habitat requirements, using current ski resort
management scenarios, shows that species with one-hectare territories declined by 32-41
percent, while population declines in species with ten-hectare territories ranged from
64-73 percent. These results underscore the need to investigate actual impacts of ski resort
development on wildlife, in particular high-elevation forest birds like Bicknell’s thrush.
Research on songbirds has repeatedly shown that increased habitat fragmentation causes
“edge effects” which can lead to higher rates of nest predation and lower rates of nest survival.
Because ski trails and associated work roads fragment and increase the amount of edge habitat,
an important question is whether these modifications affect the nesting success of Bicknell’s
thrush. VMC researchers examined potential differences in nest predation and adult survivor-
ship on ski resorts compared to natural forest areas on Mansfield and Stratton (Rimmer et
al. 2004). Overall there was no strong evidence for a ski resort effect on nest predation or
survivorship. There was some evidence that male thrushes may be more vulnerable to predation
when crossing open ski trails. Radio telemetry data indicate that Bicknell’s thrushes tend to
avoid crossing large openings. Other species may also be reluctant to cross open ski trails,
as shown in research involving forest beetles (Strong 2009) and six species of songbirds
(Holmgren 2002).
Mountain watersheds are high energy settings, where water, solutes and sediment move
rapidly over steep slopes and thin soils. Decades of study in forests of the northeastern U.S.
and elsewhere indicate that logging and road construction affect annual water yields, peak
flows, sediment production, water quality, and aquatic habitat quality. The effects of ski resort
development, however, are not well known. In 2000, a team of VMC cooperators established a
Aerial infrared
photography showing
an increase in road
infrastructure and
land-use changes
in Charlotte over a
15-year period.
Source: Kelley et al. 2002
1985 2000

17
watershed study on Mt. Mansfield to examine the hydrologic effects of ski resort
development. The study uses a paired-watershed approach to compare streamflow,
water chemistry, and sediment yields for two adjacent watersheds on the eastern
slopes of Mt. Mansfield (Wemple et al. 2007; Shanley and Wemple 2009). Even
after accounting for small differences in watershed size and artificial snowmaking,
study results show that water runoff from the ski resort watershed consistently
exceeds that of the unmanaged watershed to a greater extent than forest harvesting
increased runoff over unmanaged forests in regional watershed studies. The
greater runoff appears to result from the combined effects of the more prodigious
snowfall in the ski-resort watershed and to differences in snowmelt delivery to
streams during spring runoff, and has important implications for storm water
management in mountain development projects. Annual sediment yield from the
ski-resort watershed exceeds that from the unmanaged watershed by roughly three
times (Figure 14), with much larger differences documented during periods of
construction and ground disturbance. Annual chloride yield is ten times higher in
the developed watershed, presumably due to road salting, relative to the control
(Figure 15).
VMC-sponsored research on ski resort development also focuses on aquatic
biota. Biomonitoring data collected on mountain streams below ski resorts show
that increased sediment and fluctuating pH can affect the productivity, abundance,
and species richness of macroinvertebrates. Increased flows often will cause a
scour effect temporarily decreasing macroinvertebrate abundance, results that are
seen in monitoring data across the state.
The three species of stream salamanders which inhabit the mountain forests of Vermont
serve as useful bioindicators of stream habitat quality. Salamander abundance and body sizes
were quantified within seven Vermont ski resorts and in adjacent undisturbed forest areas,
including Mt. Mansfield (Hagen 1998; Strong et al. 2009). Ski resort streams supported
significantly lower populations of spring salamanders and northern dusky salamanders, and
body lengths of northern dusky salamanders were shorter within ski resorts. These differences
may have resulted from clearing streamside vegetation and increased siltation rates in
streams within ski resorts.
VMC-supported research has shown that the effects of upland development, when
referenced against traditional forest practices of logging and road construction, have proved
to be larger than expected. Some of these effects can be mitigated through good management
practices, including storm water detention structures, waterbars on ski trails, and increased
riparian buffer widths. Other effects are an inevitable outcome of habitat fragmentation
associated with development of the mountain landscape.
Fragmentation Effects on Reptiles and Amphibians
Land conversion and fragmentation are key disturbances that adversely affect
the ability of animals to move safely through the landscape. As a result of
the statewide reptile and amphibian studies funded by VMC, we have refined
our knowledge of the distribution of all of Vermont’s reptiles and amphibians
and the effects of habitat fragmentation on these species. One of these
species, the eastern ratsnake, is among the largest snakes in the United
States, commonly reaching six feet in length. The eastern ratsnake has an
isolated northern population and a connected population in western Rutland
County. The northern population on the Monkton/New Haven/Bristol border
is entirely disconnected from any other eastern ratsnakes; hence, there
is no introduction of new genes into the population and no possibility of
recolonization after a local population extirpation. Consequently, it is at a
much greater risk of permanent loss than the Rutland County population.
250
200
150
100
50
0
Annual Suspended Sediment Yield
WY01 WY02 WY03
150
100
50
0
Annual Chloride Yield
WY01 WY02 WY03
Ranch Brook West Brook
Ranch Brook West Brook
Wood turtle
Figures 14 and 15:
In a paired-watershed
study at Mt. Mansfield
conducted in water
years 2001-2003,
both annual sediment
yield and annual
chloride yield were
higher on a ski-resort
watershed than on an
unmanaged watershed.
Source: Wemple et al. 2007;
Shanley and Wemple 2009

18
Among the best known migrating wildlife in Vermont are those species
of amphibian that breed in the early spring. With the first warm rains of late
March and early April, thousands of spotted salamanders and wood frogs that
have spent the winter in upland deciduous forests, head to the nearest beaver
ponds, vernal pools, and swamps to breed and lay their eggs. VMC-funded
work has discovered a few, large, remaining populations of the blue-spotted
salamander. One of these populations currently crosses a major shortcut to the
Williston big-box shopping malls. As a result, hundreds of adult blue-spotted
salamanders are killed on this road every spring. Eventually the amount of
mortality on the road will surpass the ability of the population to replace them,
and the population will disappear. Another amphibian, the eastern newt, relies
on a contiguous mosaic of often shifting ponds and forested uplands for its life
cycle. As these ponds and forests become isolated from each other by forest fragmentation,
the newts become increasingly at risk.
Wood turtles exemplify another conservation issue affecting long-lived species. Wood
turtles may live 30 years or more and do not become sexually mature until reaching about 14
years of age. Their survival as a species depends upon the turtles laying eggs for many years.
The long-lived strategy worked well for wood turtles until the introduction of cars, mechanized
haying equipment, and turtle collectors. Consequently, both their longevity and their terrestrial
behavior put them more at risk from increased development, habitat fragmentation, and human
population growth.
Another example of the impact of changing forested landscapes on reptiles is loss of
early successional habitat. Both smooth greensnakes and North American racers prefer open
grasslands or a mix of shrubs and grasses. Historically these might have been extensive beaver
meadows, flood plains, or areas cleared by rockslides, hurricanes, or other natural events. As
hilltop farms were abandoned, open areas first returned to forests, and now many are being
developed. Most of the lands that remained open were maintained by regular mowing, or used
as cropland. The machines that keep these lands open today have become faster and more
efficient over the years. Wildlife of all kinds is
mowed, raked, and baled along with the crops.
As a result, smooth greensnakes and North
American racers no longer inhabit cropland
or regularly mowed and baled areas. They
are, however, found along open power lines,
cleared margins of roadways, pastures (kept
open by animals), wet fields, and fields kept
open by occasional brush-hogging to maintain
views. Consequently, populations
of the smooth greensnake are
increasingly difficult to find. We know
of only one recent population
of North American racers and even
that one may have disappeared in
the last two years.

Blue-spotted salamander
Smooth greensnake
Eastern rat snake
VMC-funded work
has discovered
a few, large,
remaining
populations of
the blue-spotted
salamander.
Eventually the
amount of
mortality on the
road will surpass
the ability of the
population to
replace them.

19
Recommendations
Nearly two decades of VMC-supported research and monitoring provides a rich picture
of the health of our forests and a window into things to come. Strengths and weaknesses
in the health of our forest ecosystems have been highlighted in this report. Productivity
monitoring indicates stresses caused by air pollution, extreme weather events, and perhaps
changing climate conditions. Our forests produce an abundant supply of clean water, but
studies suggest that development will produce impacts on flows and water quality that will
have measureable effects on aquatic organisms. Biodiversity in our forests is high, but VMC
studies show clear effects of a landscape increasingly pressured by a growing population and
land-use practices that consume and fragment natural habitats.
VMC researchers studying forest-health issues recommend the following:
• Continue to study carbon dynamics in managed and unmanaged forests, and their
relationship to long-term forest recovery from historic land uses, anthropogenic stress,
natural disturbance dynamics, and forest management.
• Invest in equipment to better understand carbon flux and interactions between carbon
and other atmospheric pollutants as climate changes.
• Utilize VMC expertise and data to address the ecological effects of expanding biomass
harvesting for energy.
• Create demonstration areas for forest managers to measure forest carbon cycling and
understand carbon accounting and management options.
• Continue monitoring the responses of terrestrial and aquatic species to development
pressures leading to habitat fragmentation.
• Collect baseline data on new human and animal health risks emerging in Vermont,
including especially the deer tick, a vector of Lyme disease.
• Monitor natural disturbance impacts to better understand baseline ecosystem dynamics
and evaluate regime changes related to global climate disruption.
• Utilize VMC-supported watershed research to inform state permitting standards for new
mountain development projects.
• Provide data to inform guidelines for ski resorts and others conducting habitat-mitigation
efforts to maintain or restore ecosystem structure and function.
VMC Insect Data and Indiana Bats
ndiana bats are listed as an endangered species at state and
federal levels. Populations in Vermont depend on large old sugar
maple trees for nesting and rearing their young. A recent population
explosion of forest tent caterpillars defoliated thousands of acres of
sugar maple forest. Proposed control efforts to protect sugar maple
stands included aerial applications of a bacterial biological control agent,
Bacillus thuringiensis (Bt). Because Bt is known to kill Lepidoptera larvae in
general, concerns were raised about the effects of such treatments on the
bats’ food supply, which would have placed an additional stress on bats
during a critical life stage. Data on moth presence and abundance,
generated through VMC and other studies, were used to help time the
aerial sprayings so that effects on the Indiana bat populations would
be minimized.
I
Indiana bat
Nearly two
decades of
VMC-supported
research and
monitoring
provides a rich
picture of the
health of our
forests.

S20

WEATHER AND CLIMATE IN VERMONT
ituated just south of the 45th parallel, Vermont
experiences a humid, continental climate that
is characterized by a high degree of variability. The
state’s latitude and location relative to air masses and
frontal systems that originate to its northwest, west, and
south are key factors in both the weather fluctuations that
are observed seasonally and annually, as well as the nature
of the storms that move across it. Precipitation is equally
distributed throughout the year and the cloud shield that
affects most of the region makes Vermont one of the
cloudiest places in the United States.
Weather and climate refer to two very different
phenomena. Weather describes the condition of the
atmosphere (for example, temperature, precipitation,
cloud cover, and humidity) over a short time frame of
minutes to about a week. Climate, on the other hand, refers
to the longer-term averages—months to millennia and
longer—of the variations in the atmosphere, biosphere,
and hydrosphere over time. Climate variability incorporates
the naturally occurring fluctuations in the atmosphere over
these long time periods, while climate change refers to
a long-term change in the statistics of climate variables
(such as winds and temperature) due to either natural
climatic processes, changes in earth-sun characteristics,
or anthropogenic mechanisms (American Meteorological
Society 2000).
Natural climatic hazards that plague Vermont include
temperature extremes, flooding, drought, tornadoes,
damaging winds, severe thunderstorms, winter storms and
forest fires. Each hazard has seasonal characteristics and
temporal cycles, with some occurring more frequently in
certain decades than others. Hazards can be exacerbated by
the state’s terrain and mountain barriers (its orography), in
particular the north-south trending Green Mountains and
Taconics. Orographic precipitation refers to the enhanced
precipitation totals that occur when a storm system is
forced to rise as it encounters a mountain barrier. This is
often observed across Vermont and can lead to flooding
episodes in the summer or areas of increased snowfall
accumulation in the winter. On a local and regional scale,
Vermont’s mountains and valleys also affect the funneling
of wind flow, the creation of cold/frost hollows in valleys,
and the line between freezing precipitation and rain or snow
during complex winter storms. These terrain influences are
translated into the distribution and diversity of species types
as well as the location and magnitude of storm damage on
forests (Dupigny-Giroux 2002).
Meteorological Trends
In Vermont there is a saying that one extreme follows
another. This is particularly true for droughts and floods.
Very severe droughts are rare and tend to be statewide,
spanning a number of years. These include the severe
droughts of the mid-1960s, 1998-1999, and 2001-2002—
all were noteworthy in terms of their effects on the state’s
forests. The 1998-1999 event affected 85,000 acres (34,425
hectares) statewide with symptoms such as leaf scorch, leaf
yellowing, and early leaf color. Less severe droughts occur
more frequently and tend to be localized in extent. Species
that are susceptible to drought include red and sugar maple
(Dupigny-Giroux 2002). Temperature extremes can occur
in every season, including extreme cold and frost during
the summer which can be detrimental during the growing
season. In the spring, marked fluctuations following
budding can influence flowering in species such as crab
apples. In 2001, spring temperatures varied from at least
90 degrees F to about 20 degrees F. This accelerated the
production of apple blossoms, which were then destroyed
by the low nocturnal temperatures. Moisture extremes
can be conducive to insect outbreaks as well as diseases
such as tar spot that repeatedly struck Norway maples in
2007 and 2008.
Storms and other severe weather can act as
disturbances in forested ecosystems, the most noteworthy
of which in recent decades was the ice storm of January
1998, which damaged more than 951,000 acres in Vermont
alone. Other stressors such as drought and hurricane-
induced rainfall have also caused physical and chemical
changes in plant response. When these stressors and

21
disturbances occur consecutively or coincidentally, they raise questions about
long-term forest health and viability.
Meteorological observations are taken at six Vermont Monitoring Cooperative
(VMC) sites. The lowest in elevation are Colchester Reef (125 feet or 38 meters)
and Diamond Island (148 feet or 45 meters), located on Lake Champlain. Both
stations report 15-minute averages with Colchester Reef beginning in July 1996
and Diamond Island in May 2004. In September 1996, the station on the western
slopes of Mt. Mansfield began operations at an elevation of 2,800 feet (853
meters). A second station, also at 2,800 feet on the eastern side of Mt. Mansfield,
became operational in July 1999. Fifteen-minute data are available for these stations
as well as for the Proctor Maple Research Center Air Quality site (1,309 feet or
399 meters). Finally the National Weather Service supervises daily records at the
summit of Mt. Mansfield (4,300 feet or 1,310 meters), which has been in operation
since 1954.
Daily readings of temperature and relative humidity taken at the Colchester
Reef station versus Mt. Mansfield West highlight the stations’ physical and
geographical differences (Figures 16 and 17). The moderating influence of Lake
Champlain is clearly observed at Colchester Reef in contrast to the more continental and
higher elevation on Mt. Mansfield. Variations in relative humidity between these two sites are
determined by both the temperature and water vapor pressure of air and largely reflect the
different air characteristics at the sites. The range of relative humidity values is larger at
40
30
20
10
0
-10
-20
-30
199
7
Air Temperature – Colchester Reef
199
8
199
9
20
00

20
01

20
02

20
03

20
04

20
05

20
06

20
07

20
08

20
09

Year
A
ir
T
em
pe
ra
tu
re
(
C
)
30
20
10
0
-10
-20
-30
-40
199
7
Air Temperature – Mt. Mansfield
199
8
199
9
20
00

20
01

20
02

20
03

20
04

20
05

20
06

20
07

20
08

20
09

Year
A
ir
T
em
pe
ra
tu
re
(
C
)
Figure 16: Daily average air temperatures at
Colchester Reef (top) and Mt. Mansfield West.
Source: Christopher Still
100
90
80
70
60
50
40
30
20
10
0
199
7
Relative Humidity – Colchester Reef
199
8
199
9
20
00

20
01

20
02

20
03

20
04

20
05

20
06

20
07

20
08

20
09

TeYr
C
em
Yt
i
e
A
up
ia
it
)
(-
Relative Humidity – Mt. Mansfield
100
90
80
70
60
50
40
30
20
10
0
199
7
199
8
20
00

20
01

20
02

20
03

20
04

20
05

20
06

20
07

20
08

20
09

TeYr
C
em
Yt
i
e
A
up
ia
it
)
(-
Figures 17: Daily average relative humidity at Colchester
Reef (top) and Mt. Mansfield West.
Source: Christopher Still
The Colchester Reef
monitoring station.

23
Forest Processes in Spring
Forests are dynamic. Thus, understanding and measuring
what is normal can be challenging, requiring a long-time
commitment. One measure of forest processes is the timing
of spring events such as tree leaf and flower development,
ephemeral plant flowering, and pear thrips (a forest insect
pest) emergence and feeding. Over the past 18 years, sugar
maple leaf and flower development have been monitored
annually, and other spring phenological events have been
monitored periodically, at Mt. Mansfield, Lye Brook, and other
locations, as indicators of forest health.
Forests in spring are teeming with activity as trees, plants,
and animals come alive from their long winter’s dormancy.
Everything happens quickly and the timing of events
determines an organism’s survival and propagation. Herbaceous ephemeral plants (plants that
are only seen in the spring) are a clear example of a quick, well timed forest process. Sunshine
penetrates forest canopies that haven’t yet produced their leaves, warming the forest floor.
This provides an opportunity for plants that normally live in the shade of forest trees to take
advantage of the warmth and open canopy to quickly emerge, gather sunlight for energy,
flower, and set seeds. By the time tree leaves have come out, these plants are finished their
crucial life stages and retreat into dormancy once more. The timing of these events has been
monitored over the last 18 years at the base of Mt. Mansfield in a sugar maple-dominated
forest, providing a baseline for monitoring change over time (Table 4).
Pear thrips was discovered as a new sugar maple pest in 1987, and in 1988 thrips defoliated
nearly 250,000 acres of forestland in
Vermont. This tiny insect emerges
from the soil in the early spring and
searches for open sugar maple buds
to enter and feed upon. If spring bud
development is rapid, thrips have
little time to enter buds, feed, and
damage leaf tissue. But if weather
conditions stall bud development,
thrips can destroy young leaves,
causing significant forest injury. In the
early 1990s, research on pear thrips
at the UVM Entomology Research
Laboratory produced monitoring
methods for sugar maple leaf and
flower development. Subsequent tree
and insect monitoring at the VMC
Mt. Mansfield site have contributed
to understanding population and
management strategies of this insect
in relation to sugar maple processes. In 1993 and 1995, pear thrips populations were significant,
emergence and feeding developed sooner than sugar maple leaf development, and this resulted
in large areas of forest injury. Normally, sugar maple leaf emergence outpaces thrips feeding
and little leaf injury occurs.
Additional monitoring of other hardwood species has increased our understanding of the
progression of spring development among species, at different elevations and within forest
canopies. Managing forest health in Vermont requires answering questions about the timing
of tree species development. For example, when VMC researchers received a request for
information on the timing of sugar maple pollination, the answer was readily available: between
April 28th and May 15th. Another request for information came from the maple syrup industry,
Figure 20: Standardized
Precipitation Index (SPI)
for the western climate
division of Vermont
1895-2008. Periods of
excessive precipitation
and droughts have been
noted.
Source: Courtesy the National
Drought Mitigation Center
4
3
2
1
0
-1
-2
-3
-4
Standardized Precipitation Index 1895-2008
SP
I v
al
ue
s
18
95
18
99
19
03
19
07
19
11
19
15
19
19
19
23
19
27
19
31
19
35
19
39
19
43
19
47
19
51
19
55
19
63
19
67
19
71
19
75
19
79
19
83
19
87
19
91
19
95
19
99
20
03
20
07
Months
1927 1938
1955
1998,1999
2005
1995
19871980
1978
19631915
Table 4: Flowering
dates of understory
spring plants at
Proctor Maple
Research Center in
Underhill. Data results
are from 18 years of
monitoring (1991-
2008).
Source: Sandy Wilmot and
Tom Simmons, VTFPR

24
which was interested in expanding tapping to include red maple trees and needed information
on the timing of red maple bud development in comparison to sugar maple. Red maple has
been traditionally thought of as flowering early in spring and so could potentially add a “buddy
flavor” if mixed with sugar maple sap. Although red maple flower buds begin to swell on
average two days earlier than sugar maples, budbreak (when leaves start to emerge) is actually
later. Sugar maple bud development is on average 37 days from bud swell to full leaf out,
whereas red maple is 51 days. Monitoring results indicated that red maple, on average, leafed
out 14 days later that sugar maple, so timing of leaf out was not a valid reason for excluding
red maple sap.
Spring development, however, is highly variable
from year to year and location to location, depending
on temperature, precipitation, snow pack, and other
environmental factors. On Mount Mansfield with its
colder temperatures, increased snow pack, and other
environmental differences, leaf development of sugar
maple forests located at 2,200 feet was 7-10 days later
than at 1,400 feet.
Because of interest in climate change, researchers
are often asked if the timing of leaf development is
earlier or later than normal. Sugar maple budbreak
(green tip of leaf emerging from buds) and leaf out
(full leaf development) have been earlier in the decade
of the 2000s compared to the 1990s. While there is
much variability from year to year, eight of the nine
years of earlier than normal leaf development occurred since 2000. On average, budbreak is
three days earlier and leaf out is five days earlier this decade than in the 1990s (Figure 21). In
2008, monitoring showed that budbreak was 12 days earlier and full leaf out was nine days
earlier than the baseline. While this may seem like an indication of climate change effects,
there have not been significant changes in spring temperatures over this same time period.
Although temperature and bud development are closely linked, there may be other reasons for
bud development differences such as late snow cover in the spring, soil moisture deficit, or
other weather conditions. As with many weather-dependent processes, more time is needed
to discern what lies within the realm of normal variation, and what will prove to be a long-term
trend.
29-May
24-May
19-May
14-May
9-May
4-May
29-April
24-April
19-April
199
2
Comparison of Sugar Maple Leaf Out Dates
199
3
199
4
199
5
199
6
199
7
199
8
199
9
20
00 20
01

20
02

20
03

20
04

20
05

20
06

20
07

20
08

Year
D
at
e
Annual Leaf out
1991-1999 Average
2000-2008 Average
Figure 21: On average,
budbreak of sugar
maples at Proctor Maple
Research Center is three
days earlier and leaf
out is five days earlier
this decade than in
the 1990s.
Source: Sandy Wilmot and
Tom Simmons, VTFPR
Development stages
of red maple flowers
(top row) and leaf
buds (bottom row).
The maple
syrup industry
was interested
in expanding
tapping to
include red maple
and needed
information on
timing of bud
development.

25
Weather and Climate Influences on Reptiles and Amphibians
Reptile and amphibian abundance and health are dependent in part on the amount and
distribution of heat and moisture in their surroundings. Unlike mammals or birds, reptiles
and amphibians do not generate their own heat. They control their internal
temperatures by moving into or out of the sun, warm water, or into or out of
contact with warmer objects. They are essentially solar heated. Their growth,
digestion, healing, and all other physiological processes work best within a fairly
narrow range of temperatures and water conditions.
In our work on Mt. Mansfield and near the Lye Brook Wilderness, we have
often found wood frog and spotted salamander eggs or larvae drying up at the
bottom of what were vernal pools earlier in the year. Vernal pools fill with water
during the spring but often are dry by early fall. If the pool dries before the
amphibians hatch and metamorphose (approximately the third week in July for
spotted salamanders at our sites), they will all die. This is a fine balance. Even
if the amount of annual rainfall is adequate, if it is dry for three weeks in May,
the pools will dry and the amphibians will die.
Changes in the timing or amount of snowfall in winter also control populations. If snow
cover is non-existent or minimal during an extended cold period in the winter, even freeze-
tolerant species such as spring peepers and wood frogs that winter in the leaf litter will freeze
and die.
Soil Climate Analysis Network
he USDA Natural Resources Conservation Service (NRCS) Soil Climate Analysis Network (SCAN) is
a cooperative nationwide data collection system designed to support natural resource assessments and
conservation activities. As part of the long-term soil monitoring program, VMC and NRCS partnered to install
SCAN stations near Lye Brook Wilderness and on the west flank of Mt. Mansfield in September 2000.
In addition to supporting the national objectives of the SCAN program, the objectives of the VMC
SCAN sites are to collect long-term data on local weather and soil moisture and temperature to complement
measurements of physical, chemical, and biological parameters at long-term soil monitoring sites located nearby.
Remote sites like Vermont’s are designed to provide near real-time data from a variety of sensors. The
above-ground sensors collect information required for climate analysis and evapotranspiration calculations,
including precipitation, air temperature, relative humidity, wind speed and direction, solar radiation, barometric
pressure, snow depth, and snow water content. Below-ground sensors collect soil temperature and soil moisture
data at five depths (2, 4, 8, 20, and 40 inches). At Lye Brook, two sets of below-ground sensors are installed, one
set in a sunny forest opening another in shade under the forest canopy.
SCAN data show that in the summer, the upper layers of soil are the warmest, but in the winter, the deeper
layers are warmest. In July and August, there are daily soil temperature fluctuations of up to 3 degrees C at
the surface, while at 40 inches, daily temperature changes are on the order of about 0.1degrees C or less. At
some point in the month of April, the soil has virtually the same temperature throughout the soil profile as the
upper layers begin to warm up. In September, the same temperature equalization happens as the upper layers
begin to cool down. Very few soil temperature readings of below 0 degrees C have been recorded, which raises
the question of whether these soils actually freeze in winter, as is commonly believed. The soils typically have
the highest moisture content reading in April. This seems to be more attributable to snowmelt than increased
precipitation. All soils dry out in the summer months, regardless of precipitation levels. Although not as distinct
as in summer, there is a noticeable drop in soil moisture in winter.
T
Wood frog

26
Reptile and amphibian distribution is also controlled in part by climate. Many of our rarest
reptile species such as North American racers, timber rattlesnakes, and common five-lined
skinks are at the northern extremes of their ranges. As the climate warms, these species may
benefit. Their numbers may increase and their ranges may expand if they can move safely
through the landscape to colonize new areas. But it may be difficult or impossible for them
to move safely to the nearest piece of appropriate habitat. Habitat fragmentation may have
resulted in large distances between appropriate habitat locations. Mortality from roads and
other sources may be too frequent to allow adequate numbers to move. As a result, it may be
that only more mobile southern species, with relatively unspecialized habitat requirements,
expand their ranges in the state.
Several amphibian species are at the southern extreme of their range in Vermont. The
boreal chorus frog may have already been extirpated from Vermont. It was last heard in Alburg
in 1999. This may be the result of drainage of temporary low-elevation wetlands and an inability
to compete with wood frogs. As the climate warms, our remaining northern species, the mink
frog, may disappear from its current limited range in northeastern Vermont. It is unclear why
it is limited to that portion of our state. It may be unable to compete with more aggressive
southern species such as bullfrogs, or perhaps colder water temperatures keep other diseases,
parasites, or predators at bay. Whatever the case, the mink frog’s climate is warming and its
environment and surrounding species mix will change. It may soon disappear from Vermont.
The Ice Storm of January 1998
In January 1998, a prolonged, widespread icing episode affected much of the northeastern
United States and adjacent Canada. Damage resulting from this severe ice storm was aerially
mapped on one-fifth of the forest land in Vermont. Damage was most severe to hardwoods,
recently thinned stands, and pole-size trees. Damage consisted mainly of branch breakage,
broken main stems, and bent trunks, with fewer uprooted trees. The storm impacted about 20
percent of the forest plots used for forest health monitoring in the state, killing more than 20
percent of the hardwood trees on these plots. Damage affected parts of both Mt. Mansfield
and Lye Brook, and prompted ecological studies of damage patterns and forest response.
Most trees have an amazing ability to recover from a natural disturbance such as this.
For hardwood species, this is especially true if the damage occurs during the winter months
when most of the trees’ energy reserves remain in the roots. Recovery for trees that weren’t
immediately killed by the storm was documented for damaged sites at Mt. Mansfield and
elsewhere in the state between 1999 and 2003 by doing vertical photography of crown canopy
cover and oblique photography of individual tree crowns.
I
Following the destruction caused by the ice storm in 1998, sugar maples recovered
rapidly, as shown in these series of photos of crown improvement of 8-inch diameter
trees in Strafford, Vermont.
In addition to
tracking long- and
short-term climate
and weather
changes and
events, knowing
the potential
significance of
these changes to
native species,
communities,
and working
ecosystems will
hopefully generate
more informed
action.

28
Modeling Bicknell’s Thrush and Climate Change

icknell’s thrush, a rarely seen small bird with a flute-like song, is a mountain species of high
conservation concern. It nests only in high-elevation “islands” of spruce-fir forests in the
northeastern United States and southeastern Canada. There are probably fewer than 50,000 individuals.
On its Caribbean wintering grounds, where an estimated 90 percent of the global population is
concentrated on Hispaniola, loss of forested habitats has been severe and is ongoing. The species is
at risk from a variety of threats to its breeding habitat, including development, atmospheric pollution,
and potentially the most damaging, climate change.
Because the extent of Bicknell’s thrush breeding habitat is controlled primarily by climate, projected
warming has the potential to alter the distribution and abundance of this species. Changes in vegetation
communities along elevation gradients in the Northeast are strongly influenced by temperature (Spear
1989; Botkin et al. 1972). Warmer growing seasons could gradually elevate forest ecotones and confine
high-elevation plant and animal communities to progressively higher, smaller, and more isolated patches.
To project effects of climate change on Bicknell’s thrush habitat, VMC researchers first modeled
mean July temperature in mountainous areas of New York
and northern New England (Ollinger et al. 1995). Next, they
identified the temperature range that corresponds with a
model of the birds’ current breeding distribution (Lambert et
al. 2005). Then researchers simulated warming, in one-degree
increments, and measured changes in the availability of suitable
habitat. Finally, predicted future changes in temperature were
used to assess potential impacts on the amount of suitable
habitat available under different climate-change scenarios
(Hayhoe 2007).
Regional warming of even 1 degree C will reduce potential
Bicknell’s thrush habitat by more than half and an increase of
2 degrees C may be sufficient to eliminate all breeding sites
from most of Vermont (Figure 23). A 3 degree C increase
in growing season temperatures has the potential to nearly
eliminate Bicknell’s thrush habitat in the entire Northeast.
Summer temperatures are projected to rise on average by 2.8-5.9 degrees C under a range of carbon-
emission scenarios (UCS 2006). It’s not clear how long it will take for ecotones to shift once temperature
changes occur; scientific opinion ranges
from decades to as long as centuries.
VMC modeling using the temperature
envelope approach should be interpreted
to present the range of possible changes
in spruce-fir habitat.
B
12000
10000
8000
6000
4000
2000
0
Projected Reduction in Bicknell’s Thrush Habitat
Temperature (C) Change Relative to 1950-1990
B
ic
kn
el
l’s
T
hr
us
h
B
re
ed
in
g
H
ab
ita
t (
H
a)
0 1 2 3 4 5 6
Predicted Lower
Emissions
Scenario (2.8C)
Predicted Higher
Emissions
Scenario (5.9C)
Figure 23: This chart shows reduction
in Bicknell’s thrush habitat under
simulated warming conditions.
Simulations raised mean July
temperature in 1 degree C increments
and assumed even warming.
Source: Rodenhouse et al. 2008
Bicknell’s thrush

29
A irborne pollutants can travel thousands of miles from their origins, but are eventually removed from the air and deposited to the earth’s surface
through various atmospheric deposition processes. Wet
deposition removes pollutants such as sulfuric and nitric
acids which are dissolved in rain, snow, or cloud water. Dry
deposition occurs when gaseous pollutants like ozone or
sulfur dioxide, or particle pollutants like lead or black carbon,
are deposited directly to plant leaves, surface waters, or other
environmental surfaces. Wet and dry deposition processes
provide a valuable service by cleansing the air of pollutants
which would otherwise build to intolerably high atmospheric
concentrations. However, atmospheric deposition also brings
adverse environmental consequences in the form of acids,
oxidants, hazardous organic compounds, mercury, and other
toxic metals which accumulate in aquatic and terrestrial
ecosystems.
Atmospheric deposition monitoring provides an
important foundation for environmental research, as
atmospheric inputs can accumulate on the landscape
over time and have cascading impacts on ecosystems.
Long-term atmospheric deposition monitoring has not
only been a cornerstone that supports other VMC research,
but has also been an important research focus in its own
right. Analyses of VMC air quality and deposition data have
shown that Vermont’s pollution originates from a variety
of local and distant sources. For example, the sulfate that
impairs our visibility and acidifies our rain comes partly from
Canadian smelters (which also contribute arsenic), partly
from East Coast oil burning (which also contribute nickel and
vanadium), and partly from midwestern coal burning (which
also contributes selenium). All of the above pollutants have
declined as emissions from these various source categories
have been reduced. VMC has also conducted pioneering
research on the trends, origins, and effects of ambient
concentrations and deposition of atmospheric ozone,
aerosols, and mercury.


MONITORING ATMOSPHERIC DEPOSITION
The core VMC site for atmospheric deposition
monitoring is the Proctor Maple Research Center
(PMRC), in Underhill. PMRC hosts several precipitation
and air quality monitoring programs in response to
regional and national scale issues such as acid rain,
airborne toxic substances (such as benzene), cancer-
causing ultraviolet radiation, haze, mercury, and ozone.
PMRC is intensively instrumented, and features a 66-
foot tower to support measurements and research on
pollutant gradients within and above the forest canopy.
Air and deposition monitoring is also conducted near
VMC’s site at the Lye Brook Wilderness Area.
The segments to follow feature key VMC
atmosphere-related research on acid rain, mercury, and
ozone, and on their effects on vegetation, soils, and lakes.
Also included is a segment on critical loads assessment,
which is research at the interface of atmosphere and
ecosystems designed to assess how much pollutant
deposition ecosystems can withstand.
Acid Deposition
Atmospheric deposition monitoring at the PMRC
in Underhill includes stations in three different statewide
or national “acid rain” monitoring networks: the Vermont
Acid Precipitation Monitoring Program (VAPMP, since
1980), the National Atmospheric Deposition Network
(NADP, since 1984), and the NOAA Atmospheric
Integration Research Monitoring Network (AIRMoN,
since 1993). VMC has a second atmospheric site at Lye
Brook in southern Vermont. Lye Brook was formerly a
CASTNET site for dry deposition, and wet deposition
data from a nearby NADP site in Bennington are used to
assess the composition and trends of acid deposition in
Lye Brook.

31
demonstrated that much of the mercury deposited in Vermont originated
in areas with high densities of coal-fired electric power plants (Keeler et
al. 2005; Miller et al. 2009). Emissions from coal-fired power plants have
remained stable over the period (Miller et al. 2009).
VMC and EPA-funded observations of ambient air concentrations of
reactive gaseous mercury (RGM), particulate mercury (HGP), and gaseous
elemental mercury (GEM) at VT99 since 2004 provide an even clearer
identification of out-of-state mercury sources. Because the air samples
represent 2-hour averages, air-mass back-trajectories associated with them
are more accurate in comparison to the complex multiple trajectories
that contribute mercury to any given rainfall event. These observations
establish that all three forms of mercury experience net transport to a
greater degree than suggested by current generation EPA emissions-
transport models used to establish policy. These measurements make
plain that emissions-dense regions are associated with the highest air
concentration events in northern Vermont (Figure 26, Miller et al. 2009).
These unique and detailed records have provided additional insights
such as the importance of climate variations influencing wet deposition, a strong seasonality
to both wet and dry deposition, and less enrichment of mercury in cloud water relative to
rain as compared to other pollutants. VMC also coordinated and executed several studies
comparing the performance of different wet-deposition collector types on behalf of the
national network and research partners. The core mercury monitoring activities at VT99 have
facilitated numerous other studies, including determination of dry-deposition rates of mercury
to forests and retention of mercury in watershed soils.
MICB MDN
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
14
12
10
8
6
4
2
0
A
nn
ua
l H
g
W
et
-D
ep
os
iti
on
(
ug
/m
2 )
Year
Figure 25: Wet
deposition of mercury
at Underhill. There was
no significant trend
in deposition over the
period.
Source: Miller et al. 2009

Reactive Gaseous Mercury (pg/m
3 )
Anthropogenic Hg2+
Emissions (g / y)
0
20
40
60
80
100
120
140
109 110 111 112 113 114 115 116
R
ea
ct
iv
e
G
as
eo
us
M
er
cu
ry
(
pg
/m
3 )
Day of Year
Figure 26: High reactive
gaseous mercury (RGM)
levels in Vermont are due
to the transport of air
from emissions-dense
regions to the south and
west. The inset panel
shows RGM measurements
at VMC’s Underhill site,
VT99, for seven days in
2007. The main panel
shows the motion of air
parcels for the 72 hours
prior to their arrival at
VT99 during the same
period overlain on a
map of RGM emissions.
The gray squares show
the air transport paths
resulting in high RGM
measurements. The
blue and green circles
(shown to the northwest
and northeast of VT99)
show the air transport
paths resulting in low
RGM concentrations in
Vermont.
Source: RGM emissions
courtesy of Mark Cohen, NOAA;
map by Eric Miller

32
Mercury in wet and dry deposition is assimilated into plants and soils, allowing mercury
to gain entry to both aquatic and terrestrial food webs. Surveys of Vermont inland waters
document many lakes and ponds with elevated mercury levels in water, plankton, and fish.
Consequently, the Vermont Department of Health has issued advisories against consumption
of specific game fish in Lake Champlain and various lakes and ponds. Elevated mercury in
fish such as yellow perch is translated into high mercury burdens in fish-eating wildlife such
as loons and river otters. A coordinated effort between VMC and NOAA-funded research
identified the rapid impact of major atmospheric mercury deposition events on Lake
Champlain water and zooplankton mercury levels. Atmospheric mercury loading dominates
the overall mercury budget of Lake Champlain.
The nexus of two long-term VMC studies (atmospheric mercury and Bicknell’s thrush
demographics) resulted in the first ever observations and documentation of mercury
bioaccumulation in a purely terrestrial food web (Rimmer et al. 2005; Rimmer et al. in review).
Bicknell’s thrush breeding in Vermont’s high-elevation forests exhibit elevated mercury
concentrations in the blood and feathers compared to other songbirds. Components of the
Bicknell’s montane food web show relative mercury levels similar to patterns associated
with differing food-web positions documented for aquatic ecosystems (Figure 27). Because
Bicknell’s thrush is a strictly terrestrial insectivore, these discoveries have raised concerns
about the health of other songbird populations and whether mercury toxicity may cause
immune suppression and interact with other ecological stressors such as soil calcium
depletion from acid rain. Additional findings that blood mercury levels are up to three
times higher in overwintering birds than in breeding individuals suggest worrisome rates
of deposition and dietary uptake on the species’ winter range. The seasonal decline of
blood mercury in breeding birds may reflect both the gradual loss of mercury carried from
the wintering grounds and an early- to late-season diet shift, corresponding to local prey
availability. Early-returning migrants likely consume disproportionate numbers of carnivorous
invertebrates (spiders and harvestmen, for example), with elevated mercury tissue levels, then
switch to plant-eating moth and caterpillar larvae low in mercury as these emerge in mid-
summer. Much work remains to be done on the dynamics of mercury bioaccumulation by
Bicknell’s thrush and other terrestrial animals during different periods of their annual cycle.
Atmospheric mercury deposition seems likely to continue to pose significant risks for
Vermont’s aquatic and terrestrial ecosystems until upwind power-plant mercury emission
sources are reduced. The existing and continuing mercury research coordinated by the VMC
provides an opportunity to detect atmospheric deposition changes and subsequent ecological
recovery. The detailed atmospheric mercury studies at Underhill provide a crucial check
on EPA modeling efforts used in designing emissions reduction programs. The extensive
Additional
findings that
blood mercury
levels are up
to three times
higher in
overwintering
birds than
in breeding
individuals
suggest
worrisome rates
of deposition
and dietary
uptake on the
species’ winter
range.
Early Season Food Web
Deciduous Foliage
Hg 0.008 ug/g
Deciduous Herbivores
Hg 0.019 ug/g (17.9%)
Detrital Herbivores
Hg 0.077 ug/g (11.3%)
Detrital Omnivores
Hg 0.108 ug/g (2.9%)
Leaf Detrital Layer
Hg 0.254 ug/g
Bicknell’s Thrush
May
Blood Hg 0.124 ug/g
Raptors
Blood Hg 0.490 ug/g
All Levels Carnivores
Hg 0.123 ug/g (40.6%)
Varied Level Varied Diet
Hg 0.032 ug/g (27.3%)
Mammals
?
Blood
Sucking
Figure 27: Relative
mercury burdens
and inferred percent
contribution (in
parentheses) to
Bicknell’s thrush
diet during the early
breeding season.
Mercury burdens
increase at higher
levels of the food web.
Source: Rimmer et al. 2005;
Rimmer et al. in review

33
observations, research network, and experience of VMC mercury
researchers positions Vermont as a strong candidate for a role
as a pilot site in an anticipated national mercury biomonitoring
network. VT99 has recently been selected and funded by EPA as
one of the initial sites in the new NADP/Mercury Trends Network.
Lake Trends over Time
During the late 1970s, the chemistry of lakes was surveyed
throughout Vermont. Concern was mounting that remote,
high-elevation lakes in geologically sensitive areas were either
already acidified or risked acidification due to the long distance
transport of atmospheric pollution. U.S. and European studies
had confirmed that the discharge of pollutants including sulfur
dioxide and oxides of nitrogen from coal fired power plants and
vehicles caused lake and stream acidification. A waterbody’s sensitivity to acidification is
determined not only by the load of pollutants entering from the atmosphere, but also by
landscape characteristics such as watershed size, soil type, and slope of the watershed. As a
result, not all Vermont lakes are affected by acid rain. Initial monitoring within the southern
Green Mountains indicated that this region of Vermont was sensitive to acidification and
that a high proportion of the undeveloped lakes were notably acidic. By far, the greatest
concentration of acidic lakes and streams was found in Bennington and Windham counties.
In 1982, Vermont entered into a cooperative agreement with the U.S. Environmental
Protection Agency’s Long-Term Monitoring (LTM) Project to assess 36 lakes. In 1993,
VMC established a southern Vermont monitoring site in the Lye Brook Wilderness. Long-
term chemistry trends have since been established for Branch and Bourn ponds. Federal
mandates under the Clean Air Act require Class I Wilderness Areas (like Lye Brook) to
protect air-quality related values.
There have been
several striking trends
on Vermont’s acid lakes,
including a decline in
sulfate concentrations and
an increase in pH (implying
a decrease in acidity). As
the primary measurement
of acidity, pH has steadily
climbed in most Vermont
acid lakes since the
passage of the 1990 Clean
Air Act Amendments.
These amendments mandated a reduction in the amount of acidifying pollutants and it
was expected that acidity levels on lakes would drop, as they did, for example, at Branch
and Bourn Ponds (Figure 28). An unexpected result, however, was the reduction in calcium
and magnesium in lakes. With the reduction of sulfate deposition on the landscape, less
sulfate is moving through the soil, and therefore less calcium and magnesium are leached
out of soils and bedrock. Calcium and magnesium are essential for healthy ecosystems, and
concentrations need to increase before Vermont’s acid lakes can be restored to their native
biological diversity.
Acid-sensitive lakes have improved significantly since the 1990 Clean Air Act
Amendments. But because of the loss of calcium and magnesium throughout the
watershed and in the lake itself, sensitive fish, snail, and insect species will not recover until
further reductions in acidifying agents occur. Also, time is needed to allow bedrock and soil
to resupply calcium and magnesium to the lakes through the weathering process.

140
120
100
80
60
40
1980 1990 2000 2010
6.5
6.0
5.5
5.0
1980 1990 2000 2010
Bourn Pond
Figure 28: Bourn Pond
trends in sulfate and
pH.
Source: Heather Pembrook,
VTDEC

34

Forest and Surface Water Critical Loads
Vermont’s mountain ridges and deep glaciated valleys form a diverse
landscape with widely varying geologic, climatic, and ecological zones.
These biophysical contrasts coupled with differing land-use practices
across the state lead to a complex mosaic of ecosystem tolerance for
the effects of atmospheric deposition of sulfur and nitrogen. The critical
load of sulfur plus nitrogen acidity is the level of deposition below which
no harmful ecological effects occur for an ecosystem. The exceedance
of the critical load is the difference between the critical load and current
atmospheric deposition loading. The magnitude of the exceedance
indicates the severity of ecosystem risk posed by sulfur and nitrogen
deposition.
The critical sulfur and nitrogen load for Vermont’s terrestrial
ecosystems was established relative to the sustainability of the essential
plant nutrients calcium, magnesium, and potassium, with data and
collaboration provided by VMC (Miller et al. 2005). When these nutrients are
lost in streamflow faster than they are supplied from precipitation and mineral
weathering, inadequate levels of nutrients may develop in both soils and plants.
Poor calcium nutrition underlies a wide range of forest health problems including
reduced growth rates, inadequate plant response to climate stress, pest and
pathogen stress, and increased mortality (Schaberg et al. 2008) as well as
decreased breeding success of songbirds. The critical load for aquatic ecosystems
was established with respect to the acid neutralizing capacity (ANC) of water, which in
turn is related to pH (acidity) and levels of toxic aluminum (Pembrook 2003). Low pH
and elevated aluminum impair the health and cause die-offs of many fish and plankton
species, leading to a loss of diversity in aquatic ecosystems.
The lowest critical loads in Vermont were found along the spine of the Green
Mountains and in the Northeast Kingdom, where soils are developed in thin and patchy tills
derived from rocks with low buffering capacity. Not surprisingly, this is where acid precipitation
exceeded the critical loads by the greatest amount. Areas of calcium-rich rocks and soils
scattered throughout the region support the highest critical loads where forests and surface
waters are well buffered against acidic deposition.
Total sulfur plus nitrogen deposition ranged widely across the state. High-elevation areas
received the highest sulfur and nitrogen deposition due to orographically enhanced precipitation
and cloud water inputs. Deposition was also high in the southern and western parts of the
region due to their proximity to emission sources. Despite the substantial decrease in sulfur
deposition documented in Vermont over the past 25 years, combined sulfur and nitrogen
deposition during 1999-2003 continued to exceed the critical load for sulfur and nitrogen acidity
for 30 percent of Vermont forests and 29 of 30 acid-sensitive ponds (Figure 29). Vermont’s 30
acid-sensitive ponds are found in areas of nutrient poor soils with high atmospheric deposition
loads (Figure 30). While deposition had fallen below the critical load in one pond and within
6 percent of the critical load in two other monitored ponds by 2003, the anticipated biological
recovery has been delayed in these waters due to coincident
reductions in calcium concentrations, depriving aquatic animals
of an essential nutrient.
Bourn Branch Little Mud (Winhall)
Excess Acid Deposition
Critical Load
m
eq
/m
2/
yr
120
100
80
60
40
20
0
Figure 29: Yellow
through red indicates
where atmospheric
deposition exceeds
the critical load, thus
potentially damaging
forest ecosystems. In the
years 1999-2003, the
critical load for sulfur
and nitrogen acidity was
exceeded for 30 percent
of Vermont forests and
29 of 30 acid-sensitive
ponds (black circles).
Source: after Miller et al.
2005
Figure 30: Three of the thirty Vermont acid-sensitive ponds are
within or adjacent to the Lye Brook Wilderness Area: Bourn,
Branch, and Little Mud Ponds. The top of each stacked bar
represents the total annual acid input. The tan part of each bar
represents the Critical Load each pond can withstand without
adverse biological effects. The brown part of each bar represents
the excess acid loading.

36
The success of the long-term soil
monitoring project will depend in part on
how well it responds to future challenges.
The relevant environmental concerns can
change over time; for example, atmospheric
deposition of sulfur may become less of an
issue, while soil warming and soil carbon
storage may become bigger concerns in
the next century.
Ozone
Ground-level ozone pollution causes
serious human health effects, which can be
especially severe for individuals with existing
cardiovascular or respiratory diseases. Ozone
pollution also damages forest plants and
agricultural crops by reducing plant growth
and vigor, reducing seed production, and
increasing susceptibility to insects and
disease. As with humans, certain plants are especially sensitive to pollution. Black cherry, white
ash, and yellow poplar are ozone-sensitive and show visible symptoms of injury following
exposure to prolonged periods of ozone pollution. Unlike the human health effects which result
from short-term, eight-hour peak concentrations, the effects of ozone on plant species occur
from the cumulative effects of repeated exposures over the entire growing season.
Ground-level ozone pollution has shown a gradual improvement over the past decade at
sites in northern and southern Vermont. Biologically relevant ozone pollution is expressed in
terms of a cumulative seasonal index. EPA scientific advisors recently suggested 7 ppm-hours
as a threshold below which adverse effects on sensitive vegetation are substantially reduced.
Ozone at the Underhill VMC site has dropped below the EPA threshold since 2000, while in
southern Vermont, similar improvements were seen after 2002. These decreases in ground-level
ozone have been accompanied by parallel and dramatic reductions in visible damage in ozone-
sensitive plants (Figure 33). Plant injury in southern Vermont has been consistently greater, in
keeping with the higher atmospheric ozone levels. These results show that our state and federal
clean air strategies are working to reduce airborne concentrations of ozone pollution. Despite
this progress, additional improvements are needed, since ozone-induced plant injury continues
at southern Vermont sites, and since ozone damage can occur at concentrations below those
that cause visible foliar injury.
In 2008, the EPA Clean Air Scientific Advisory Committee
strongly encouraged the agency to set a cumulative ozone
standard in the range of 7-15 ppm-hours to protect sensitive
vegetation. (The former EPA administrator decided not to take
this advice.) By combining two VMC long-term monitoring
datasets—ozone concentrations in ambient air and foliar ozone
injury—VMC cooperators have shown that foliar ozone injury
in ozone-sensitive plants tracks cumulative seasonal ozone
exposures, not isolated high ozone concentration events. VMC
results further suggest that foliar ozone damage is substantially
reduced when the ozone exposure falls below about 7 ppm-
hours. These findings strongly support the recommendations
of EPA’s scientific advisors, and should provide valuable
information as a new EPA administrator reconsiders the
adequacy of the current ozone standards.

14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
20
18
16
14
12
10
8
6
4
2
0
3-
Ye
ar
A
ve
ra
ge
W
12
6
O
zo
ne
(
pp
m
-h
ou
rs
)
3-
Ye
ar
A
ve
ra
ge
P
er
ce
nt
P
la
nt
In
ju
ry

Estimated Ozone
Injury Threshold
Southern Vermont W126 Ozone
Southern Vermont % Plant Injury
Northern Vermont W126 Ozone
Northern Vermont % Plant Injury
Figure 33: Ground level
ozone trends during
the growing season for
southern (Bennington)
and northern
(Underhill) Vermont.
Trends in the percent
of ozone sensitive
plants with injury
(curved lines, right
axis) at monitoring
sites in northern and
southern Vermont,
including VMC sites.
Dry summers in 1995,
1999 and 2001 are
partly responsible for
lower ozone injury in
those years.
Source: From VT Air
Pollution Control Division
and USFS/FIA
Ozone injury to white ash.

37
Recommendations
Vermont has helped lead the nation in atmospheric deposition monitoring, and with
that in mind the authors of this section recommend the following:
• VMC environmental and biological monitoring efforts should continue despite recent
improvements in air quality.
• Continue VMC monitoring of total and methyl mercury in wet deposition, ambient
air mercury speciation, and dry deposition. These long-term datasets are essential for
documenting changes in mercury deposition rates in response to emissions policies,
and for continuing studies on mercury cycling in Vermont’s ecosystems.
• Vermont should continue seasonal chemical monitoring of Branch and Bourn Ponds.
These water bodies are ideal indicators to document trends resulting from emission
reductions required by the Clean Air Act. Conversely, new sources of downwind
emissions can degrade these water bodies and overwhelm any expected improvement.
• Biological monitoring of sensitive aquatic groups like mayflies, clams, and snails
should be conducted every five years unless a change in water chemistry warrants more
frequent analysis. Biological assessments can be examined on an ongoing basis as part
of the present Long-Term Monitoring Project.
• Use VMC data to inform national ozone standards. A cumulative, seasonal ozone
standard in the range of 7-15 ppm-hours will protect most sensitive vegetation.
Vermont has
helped lead
the nation in
atmospheric
deposition
monitoring.
An Acid Rain Settlement
n the early 1970s, scientists observed high mortality of red spruce trees in Vermont and other northeastern
states. In response to an early hypothesis that acid rain was damaging trees and the soils that supported them,
programs were developed to monitor the chemistry of precipitation. Proctor Maple Research Center in
Underhill was one of the first stations to begin long-term
measurements in 1981. Precipitation chemistry combined with
meteorological measurements did indeed reveal the “smoking
gun”: sulfur emissions from electric generating facilities in
the Midwest were the predominant cause of acidic deposition
in Vermont. Without long-term observation of atmospheric
chemistry, there would be no way to tell whether regulatory
programs designed to clean up air pollution were working
or not.
In 2007, the EPA and eight northeastern states filed
a lawsuit against a large utility company using data from
Underhill and other northeastern sites. This company
operates several power plants that predate the Clean Air Act
Amendments of 1977, and thus were “grandfathered in” to
operate without pollution controls. The lawsuit alleged that
the utility was keeping these plants alive to avoid the costs
of producing cleaner power. The data from Underhill helped
provide compelling evidence, and the company settled out-of-
court, agreeing to spend $4.6 billion to retrofit several old power plants. This process is a prime example of how
VMC science has helped shape policy and safeguard human and ecosystem health.
I
Atmospheric samplers at Proctor Maple Research
Center.

Selected VMC Projects and Datasets
Project Title Contact Location Duration
AIR
Ambient Air Monitoring for Ozone Benjamin Whitney Mt. Mansfield 1986-present
Bennington 1986-present
Assessment of Dry Deposition Kathleen Weathers Lye Brook 2002-2007
Atmospheric Mercury Deposition Monitoring Eric Miller Mt. Mansfield 1992-present
Basic Meteorological Monitoring:
Colchester Reef Meteorology Carl Waite Lake Champlain 1996-present
Proctor Maple Research Center Carl Waite Mt. Mansfield 1988-present
Mt. Mansfield – mid slope (884m) Carl Waite Mt. Mansfield 1996-present
Mercury Flux at PMRC Eric Miller Mt. Mansfield 2004-present
NADP/National Trends Network Carl Waite Mt. Mansfield 1984-present
Vermont Acid Precipitation Monitoring Program Heather Pembrook Mt. Mansfield 1980-present
FOREST
Alpine plant communities Tammy Gilpatrick Mt. Mansfield 2000-2001
Eric Hazelton
Rick Paradis
Forest Ecosystem Management Demonstration Project:
Silvicultural experimentation Bill Keeton Mt Mansfield 2001-present
Soil and biodiversity responses Bill Keeton Mt Mansfield 2001-present
Forest Health Monitoring Sandy Wilmot Lye Brook 1991-present
Mt. Mansfield 1991-present
Forest Damage Aerial Survey Barbara Burns Mt. Mansfield 1991-present
Barbara Burns Lye Brook 1991-present
Forest Pest Monitoring Trish Hanson Mt. Mansfield 1991-present
Long-term vegetation monitoring William Howland Mt. Mansfield 1991-1995
Tree Phenology Monitoring Sandy Wilmot Mt. Mansfield 1991-present
SOIL
Soil Climate Analysis Network Sites:
Soil Thomas Villars Lye Brook 2000-present
Thomas Villars Mt. Mansfield 2000-present
Meteorology Thomas Villars Lye Brook 2000-present
Thomas Villars Mt. Mansfield 2000-present
Long-term Soil Monitoring: Sampling Thomas Villars Lye Brook & 1999-present
Scott Bailey Mt. Mansfield 1999-present
Don Ross
Long-term Soil Monitoring: Soil Mercury Thomas Villars Lye Brook & 1999-present
Scott Bailey Mt. Mansfield 1999-present
Don Ross 1999-present
WATER
Biological & Chemical Survey of Selected Surface Waters in Lye Brook
Macroinvertebrate communities Jim Kellogg Lye Brook 1993-1995
Long-term Water Quality & Biological Monitoring Project for Lake Champlain Eric Smeltzer Lake Champlain 1992-present
Paired Watershed Study on the East Slope of Mt. Mansfield:
Hydrologic Monitoring Jamie Shanley Mt. Mansfield 2000-present
Jon Denner
Stream Chemistry of Ranch Brook and West Branch Beverley Wemple Mt. Mansfield 2000-present
Don Ross
WILDLIFE
Amphibian Survey & Monitoring Jim Andrews Lye Brook 1991-present
Mt. Mansfield 1991-present
Abby Pond 1991-present
Bicknell’s thrush Population Demographics and Ecology:
Assessing Levels of Methyl Mercury in Montane Forest Bird Community Chris Rimmer Mt. Mansfield 2000-present
Kent McFarland
Ski Resort Development Effects on Montane Forest Bird Community Kent McFarland Stratton Mt. 1995-1997
Mt. Mansfield 1995-1997
Forest Bird Surveys Steven Faccio Lye Brook 1991-present
Mt. Mansfield 1991-present
Insect Diversity on Mt. Mansfield Scott Griggs Mt. Mansfield 2000
John Grehan Mt. Mansfield 1990-1995
James Boone Mt. Mansfield 1991-1994

38

39
Resources
The Vermont Monitoring Cooperative maintains a data library containing more than 300 research and monitoring
projects and datasets. For more information, visit the VMC web site: www.uvm.edu/vmc.
Literature Cited
American Meteorological Society Glossary of Meteorology. 2000. http://amsglossary.allenpress.com/glossary.
Botkin, D.B., et al. 1972. Some ecological consequences of a computer model of forest growth. Journal of Ecology. 60:849-872.
Cogbill, C.V., P.S. White. 1991. The latitude-elevation relationship for spruce-fir forest and treeline along the Appalachian Mountain
chain. Vegetation. 94:153–175
Dupigny-Giroux, L-A. 2002. Climate variability and socioeconomic consequences of Vermont’s natural hazards: a historical
perspective. Vermont History, Volume 70, Winter/Spring, pp. 19-39.
EPA. 1999. Vermont State Program Unit of the Environmental Protection Agency. Keeping in touch, September report, 1999.
Freedman, J.M., D.R. Fitzjarrald, K.E. Moore, R.K. Sakai. 2001. Boundary layer clouds and vegetation-atmosphere feedbacks. Journal
of Climate. 14(2), 180-197.
Hagen, Kimberly. 1999. The effects of ski area development on populations of stream salamanders in central Vermont. M.S. thesis.
Antioch University New England.
Harmon, M. E. and B. Marks. 2002. Effects of silvicultural practices on carbon stores in Douglas-fir-western hemlock forests in the
Pacific Northwest, USA: results from a simulation model. Canadian Journal of Forest Research. 32:863-877.
Hayhoe K., C. Wake, J. Bradbury, et al. 2007. Past and future changes in climate and hydrological indicators in the U.S. Northeast.
Climate Dynamics. 28:381-407.
Holmgren, A. 2002. The territorial response of six songbirds to ski trails at Smugglers’ Notch Resort and Stowe Mountain Resort,
Vermont. Senior Thesis. University of Vermont.
Keeler, G.J., L.E.Gratz, K. Al-Wali. 2005. Long-term atmospheric mercury wet deposition at Underhill,Vermont. Ecotoxicology. 14, 71-83.
Keeton, W.S. 2006. Managing for late-successional/old-growth characteristics in northern hardwood-conifer forests. Forest Ecology
and Management. 235: 129-142.
Keeton, W.S. and A. R. Troy. 2006. Balancing ecological and economic objectives while managing for late-successional forest
structure. L. Zahvoyska, editor. Ecologisation of economy as a key prerequisite for sustainable development. Proceedings, Sept. 22
-23, 2005. Ukrainian National Forestry University, L’viv, Ukraine.
Keeton, W.S., C.E. Kraft, D.R. Warren. 2007b. Mature and old-growth riparian forests: Structure, dynamics, and effects on Adirondack
stream habitats. Ecological Applications. 17:852-868.
Kelley, R.S., J. Lackey, E.L. Smith, B. Frament, F. Peterson. 2002. The Health of Vermont’s Hardwood Resource: 1985-2001. Vermont
Department of Forests, Parks and Recreation.
Lambert, J.D., K.P. McFarland, C.C. Rimmer, S.D. Faccio, J. L. Atwood. 2005. A practical model of Bicknell’s Thrush distribution in the
northeastern United States. Wilson Bulletin. 117:1-11.
Luyssaert, S., E.D. Schulze, A. Borner, A. Knohl, D. Hessenmoller, B.E. Law, P. Ciais, J. Grace. 2008. Old-growth forests as global
carbon sinks. Nature. 455:213-215.
Malmsheimer, R.W., P. Heffernan, S. Brink, D. Crandall, F. Deneke, C. Galik, E. Gee, J. A. Helms, N. McClure, N. Mortimer, S. Ruddell,
M. Smith, J. Stewart. 2008. Preventing Greenhouse Gas Emissions through Wood Substitution. Journal of Forestry.106:132-135.
McGee, G.G., D.J. Leopold, R. D. Nyland. 1999. Structural characteristics of old-growth, maturing, and partially cut northern
hardwood forests. Ecological Applications. 9:1316-1329.
Miller, E.K. 2005. Assessment of forest sensitivity to nitrogen and sulfur deposition in New Hampshire and Vermont. Final Technical
Report. NH Dept. of Environmental Services. 19pp.
Miller, E.K., M.B. Burkins, R. Poirot, J. Shanley, S. Lawson, C. Waite, M. Pendelton, A. VanArsdale, M. Cohen. 2009. Atmospheric
mercury in Vermont and New England: ambient air mercury speciation studies. Final Report to USEPA-ORD-HEASD.
Min, Q. and S. Wang. 2008. Clouds modulate terrestrial carbon uptake in a midlatitude hardwood forest. Geophysical Research
Letters. 35, L02406.

40
Mladenoff, D.J. and J. Pastor. 1993. Sustainable forest ecosystems in the northern hardwood and conifer forest region: concepts and
management. In: G.H. Aplet et al. (eds.). Defining Sustainable Forestry. Island Press.
North, M.P. and W.S. Keeton. 2008. Emulating natural disturbance regimes: an emerging approach for sustainable forest management.
In R. Lafortezza et al., editors. Patterns and Processes in Forest Landscapes-Multiple Use and Sustainable Management. Springer.
Nunery, J.S. and W.S. Keeton. In review. Forest carbon storage in the northeastern United States: effects of harvesting frequency and
intensity including wood products.
Ollinger S.V., J.D. Aber, C.A. Federer, et al. 1995. Modeling the physical and chemical climatic variables across the northeastern U.S.
for a Geographic Information System. USDA Forest Service General Technical Report. NE-1913.
Pembrook, H. 2003. Calculating critical loads of acidity and exceedances for acid-impaired lakes in Vermont using the Steady State Water
Chemistry (SSWC) model. Appendix A of Vermont Dept. of Environmental Conservation TMDL for 30 acid impaired lakes.

Ray, D.G., R.S. Seymour, N.S. Scott, W.S. Keeton. 2009. Mitigating climate change with managed forests: balancing expectations,
opportunity, and risk. Journal of Forestry. January/February Issue: 50-51.

Rimmer, C.C., K.P. McFarland, J.D. Lambert, R.B. Renfrew. 2004. Evaluating the use of Vermont ski areas by Bicknell’s Thrush:
applications for Whiteface Mountain, New York. Vermont Institute of Natural Science. Woodstock, VT.
Rimmer, C.C., K.P. McFarland, D.C. Evers, E.K. Miller, Y. Aubry, D. Busby, and R.J. Taylor. 2005. Mercury concentrations in
Bicknell’s Thrush and other insectivorous passerines in montane forests of northeastern North America. Ecotoxicology. 14:223-240.
Rimmer, C.C., E.K. Miller, K.P. McFarland, R.J. Taylor, S.D. Faccio. In review. Mercury bioaccumulation and trophic transfer in the
terrestrial food web of a montane forest.
Rhemtulla, J.M., D. J. Mladenoff, M.K. Clayton. 2009. Historical forest baselines reveal potential for continued carbon sequestration.
Proceedings of the National Academy of Sciences. 106: 6082-6087.

Rodenhouse, N.L., S.N. Matthews, K.P. McFarland, J.D. Lambert, L.R. Iverson, A. Prasad, T.S. Sillett, R.T. Holmes. 2008. Potential effects
of climate change on birds of the Northeast. Mitigation and Adaptation Strategies for Global Change. 13: 517-540.
Ross, D.S., 2007. A carbon-based method for estimating the wetness of forest surface soil horizons. Canadian Journal of Forest Resources.
37:846:852.
Ruddell, S., R. Sampson, M. Smith, R. Giffen, J. Cathcart, J. Hagan, D. Sosland, J. Godbee, J. Heissenbuttel, S. Lovett, J. Helms,
W. Price, R. Simpson. 2007. The role for sustainably managed forests in climate change mitigation. Journal of Forestry. 105:314-319.
Schaberg, P.G., Miller, E.K., Eagar, C. 2008. Assessing the threat that anthropogenic calcium depletion poses to forest health and
productivity. USDA Forest Service General Technical Report. Southern and Pacific Northwest Research Stations.
Also in: www.threats.forestencyclopedia.net.
Shanley, J. B. and B. Wemple. 2009. Water quality and quantity in the mountain environment. In J. E. Milne et al., editors.
Mountain Resorts: Ecology and the Law. Ashgate Publishing.
Spear, R.W. 1989. Late-quaternary history of high-elevation vegetation in the White Mountains of New Hampshire. Ecological
Monographs. 59:125-151.
Still, C.J. et al. 2009. The influence of clouds and diffuse radiation on ecosystem-atmosphere CO
2
and CO18O exchanges.
Journal of Geophysical Research-Biogeosciences. 114, G01018, doi:10.1029/2007JG000675.
Strong, A.M., C.C. Rimmer, K.P. McFarland, K. Hagan. 2009. Effects of mountain resorts on wildlife. In J. E. Milne et al., editors.
Mountain Resorts: Ecology and the Law. Ashgate Publishing.
Strong, Allan. Unpublished data. University of Vermont.
UCS (Union of Concerned Scientists). 2006. Climate change in the U.S. Northeast: a report of the Northeast climate impacts
assessment. UCS Publications.
Vermont Department of Forests, Parks & Recreation. 1999. In: A forest that works for all: the Vermont forest resource plan, 1999-2008.

Wemple, B., J.B. Shanley, J. Denner, D. Ross, K. Mills. 2007. Hydrology and water quality in two mountain basins of the northeastern
U.S.: assessing baseline conditions and effects of ski area development. Hydrological Processes. 21: 1639-1650.

Wharton, E.H., R.H. Widmann, C.H. Barnett, T.S. Frieswyk, A.J. Lister, B. DeGeus. 2003. The forests of the Green Mountain State.
Resource Bulletin NE-158. USDA Forest Service, Northeastern Research Station.

Wilmot, S., M. Stuart, T. Villars, D. Ross, J. Juillerat, E. Engstrom. 2008. Vermont forest soil carbon map (unpublished).

Acknowledgments
The Vermont Monitoring Cooperative would like to thank the three people who supervised all
aspects of this project: Larry Forcier and Carl Waite of the University of Vermont, and Barbara Burns
of the Vermont Department of Forests, Parks and Recreation. Six VMC Cooperators took leadership
roles in producing the report’s three sections: Beverley Wemple and Sandy Wilmot (The Health of
Our Forests); Lesley-Ann Dupigny-Giroux and Christopher Still (Weather and Climate in Vermont);
and Eric Miller and Jamie Shanley (Monitoring Atmospheric Deposition).
Additionally, VMC thanks its many Cooperators, past and present, who have entrusted VMC with
their data and donated, in many cases, their time and expertise for the betterment of Vermont and
beyond, and its Stakeholders, some of whom helped guide the production of this report and are the
real reason for VMC’s existence.
Finally, VMC would like to give special thanks and recognition to Jennifer Jenkins and Sean Lawson,
both of whom were instrumental in initiating and moving forward the idea and process that has
culminated in this report, and David Sleeper for his excellent leadership, experience, and guidance
that has made production of this report possible.
Funding for this project was provided by the USDA Forest Service State & Private Forestry,
Forest Health Protection.
Project Manager and Editor: David Sleeper, Sleeper Associates LLC
Report design: RavenMark, Inc.
Printing: Queen City Printers, Inc.
Report Photos:
(Cover Photos) Large photo: Emily Sloan; inset photos: Sandy Wilmot, Jim Andrews, Ron Kelley, and Ejla
Page 2: Sandy Wilmot
Page 3: Sandy Wilmot
Page 4: Top: Sandy Wilmot; bottom: Heather Pembrook
Page 5: Bryan Pfeiffer/www.wingsphotography.com/
Page 7: Ron Kelley
Page 9: Kent McFarland
Page 13: Ron Kelley
Page 18: top, Amy Alfieri; bottom: both C. Slesar
Page 19: John Chenger
Page 20: Ejla
Page 21: Carl Waite
Page 24: Sandy Wilmot
Page 25: Jim Andrews
Page 26: Ron Kelley
Page 28: Bryan Pfeiffer/www.wingsphotography.com/
Page 29: VMC
Page 30: VMC
Page 33: Kellie Merrell
Page 36: Gretchen Smith
Page 37: VMC
Inside back cover: Emily Sloan
The stock used for this report is FSC certified...

705 Spear Street
South Burlington, VT 05403
(802) 656-0683
www.uvm.edu/vmc