20030179 Ver 6_Public Comments_20071025 (6)

OUTLINE: WNCA COMMENTS
1:
COVER LETTER
1. EXECUTIVE SUMMARY
II. INTRODUCTION
III. WHAT IS RIPARIAN BUFFER?
IV. WHY IS RIPARIAN BUFFER IMPORTANT?
V. WHAT DOES IT DO?
VI. HOW MUCH IS NEEDED?
VII. WHAT IS ITS DOLLAR VALUE?
VIII. WHAT IS ITS CONDITION AT LAKE GLENVILLE?
IN. THE HISTORY OF RIPARIAN BUFFER UNDER DUKE POWER
X. RECOMMENDATIONS
XI. ATTACHMENTS
a. Botanical analysis04
b. Holmes et al.
c. PICTS
THE WESTERN NORTH CAROLINA ALLIANCE [WNCA] 1/10/05
29 NORTH MARKET, SUITE 610
ASHEVILLE, NORTH CAROLINA 28801
Magalie R. Salas, Secretary
Federal Energy Regulatory Commission
888 FIRST ST. NE
Washington, DC 20426
Re: Scoping Comments for DUKE POWER NANTAHALA AREA [DPNA] EA/EIS
FERC Project Nos. P- 2601
P-2602
P-2603
P-2619
P-2686
P-2692
P-2698
Dear Secretary Salas:
These comments from the Western North Carolina
Alliance (WNCA) are primarily focused on the West Fork, P-2686, but we also
have serious concerns regarding all of the DPNA projects. We are concerned
about riparian habitat protection, public access, recreation, and
sedimentation.
The WNCA is a regional NGO with slightly more than 900 members of which
165 live within the DPNA generation area. The WNCA also sees itself as
representing the general public interest for these projects.
The WNCA has participated in these proceedings from the beginning and feel
that there were many positive gains made through this process. However,
we were not able to agree with all of the stipulations of the settlement
agreements and have been an active intervener. We feel the settlement
agreements failed to sufficiently protect the projects buffer strips and those
functions that it is to serve. We further feel that general public recreation in
and around the projects was insufficiently addressed and that recreational
opportunities on the East Fork, P-2698, were inadequately addressed as to
stream flows.
We have previously filed under separate cover a portion of our comments
and now we present the balance of these comments. Our comments in total
are two separate analyses of riparian buffer. First, is a presentation of
relevant Agencies positions on riparian buffer with an economic evaluation of
the impaired buffer strip at p-2686 and an estimate of the cost of
replacement of the vegetation using mitigation bank data. Second, is a
1
botanical analysis of the condition of the riparian buffer (strip) at Lake
Glenville, P-2686, which was previously submitted. The economic valuation
is based on the findings of the botanical analysis.
Thank you for this opportunity to provide these comments.
Sincerely,
William E. Lyons
1900 White Rock Rd.
Cullowhee, NC 28723-8556
mailto:welyons(cbearthlink.net
I certify that I have personally served all of the service lists for these
projects.
W. E. Lyons
mailto:[e-mail]
2
I. Exec. Summary:
These Comments are in four parts. The most important part of these being
"Analysis of vegetative communities and the effects of human disturbance
within buffer zones along the shoreline of Lake Glenville", attachment A, by J.
Dan Pittillo, Ph.D.; for it demonstrates significant losses of vegetative cover
and through that the consequences of the failure of Duke Power's shoreline
management practices and enforcement. The main body of these comments
is primarily a presentation of information published by various States
Agencies and Federal Agencies on the importance and requirements of
riparian buffers and some of the economic consequences of failing to
maintain sufficient riparian buffer. The photographs of the shoreline buffer
zone at Lake Glenville are a separate attachment. There is a Reference and
Bibliography section at the end of the main body for the readers'
convenience.
In the spring of 2004 The WNCA employed the services of J. Dan Pittillo,
Ph.D., then holder of the H.F. "Cotton" and Catherine Robinson Chair,
Professor in Biology at Western Carolina University, now retired, to perform
an assessment of the integrity of the riparian buffer at Lake Glenville, p-
2686. Dr. Pittillo has also been awarded the North Carolina Governor's Award
of Excellence (1992), the 1998 Elizabeth Ann Bartholomew Award (given by
the Southern Appalachian Botanical Society), and the 2004 Tom Dodd, Jr.
Award of Excellence, given by the Cullowhee Native Plant Conference.
Additionally, the WNCA and Dr. Pittillo acquired the statistical services of Mr.
Ben Prater, M.S., of The Southern Appalachian Biodiversity Project.
The purpose of the commissioned study was to determine if there is a
demonstrable loss of native plants and habitat within the buffer strip
bordering developed property adjacent to the Duke Power Nantahala Power
(DPNA) project boundary. The study's ultimate aim was to quantify any such
loss both numerically and monetarily (in dollars).
The "buffer strip" is defined for Lake Glenville and in general for Duke Power
Nantahala Area (DPNA) lakes, as the area surrounding the lake that is within
the project boundary and measured as ten vertical feet above the high water
mark for the lake. Note that the amount of project boundary land in square
footage will vary according to shoreline slope, with more land where the
slope is gradual and less land where the shore rises abruptly from the water.
It is estimated by the Jackson County Building Inspectors Office that (50%)
one half of the total shoreline of Lake Glenville has already been developed.
The value of the impaired fifty per cent of the buffer strip at Lake Glenville is
estimated as a $226,653,400.00 loss; this value being determined by use of
a Benefits Transfer Analysis ratio developed in a study of the value of
restored riparian buffer on the Little Tennessee River in Macon County, NC
and Rabun County, GA, attachment B, and an average mitigation bank dollar
3
amount [$200.00 per linear front foot] for the cost of restoration of riparian
buffer. Lake Glennville and the Tuckeseegee River are part of the Little
Tennessee drainage area.
STUDY FINDINGS:
The study reveals a dramatic loss of native vegetative cover, most
particularly in the shrub layer, which is primary in the prevention of erosion
and nutrient and toxin transfers. A 42% loss in abundance of trees is shown,
but even more alarming, there has been an 82% loss in species abundance
of shrubs in the sites adjacent to development.
Observations made during the study indicate:
1. a continual erosion of the shorelines due to wake action.
2. the loss of the vegetative cover.
3. The buffer strip, even if properly vegetated would not be of sufficient
depth from the water line to perform its functions optimally.
4. Many of the developed areas have shallow soils and bedrock has been
and is being exposed by the erosion, preventing significant natural re-
vegetation.
Native vegetation is the primary control of this erosion but has been
removed from many sites around the lake. Tree roots grow deepest
into the soil but are less effective for topsoil erosion control. The best
erosion and transport control occurs with the extensive root masses
associated with the native shrubs and from large woody debris.
5. Trees and shrubs which have been removed are not being replaced by
natural reproduction due to mowing. Grass and other lawn species are
being maintained on most developed sites and on the buffer strip
between those sites and the lake. These shallow rooted species are
not effective control against the undercutting action of the waves nor
are they effective sediment control on significant slopes. Lawn
grasses also are not significant for the uptake of toxins and nutrients.
6. Trees that die in the buffer zone and later topple into the lake as they
are undercut by wave action are often removed by the adjacent land
owners. While these trees and their branches could serve as
revetment against wave erosion, spawning grounds, and protective
habitat for fish, these functions are negated by the removal of the
trees and other large woody debris.
7. The use of rip-rap and rock walls by adjacent owners is not a suitable
replacement for natural vegetation.
The value of the buffer strip as riparian habitat goes far beyond
sedimentation and erosion control or even its value as habitat for birds and
animals.
This study did not examine in-pond or aquatic vegetation or animals or the
effects that the loss of vegetative cover has induced on them. However, in
discussions with two aquatic biologists they suggested that the lake in its
4
current condition and considering the current project management practices
of DPNA the lake could not likely support more than thirty percent of what its
normally expected biotic load would be.
The State of North Carolina and Jackson County, North Carolina both require
a buffer zone set back of thirty feet depth adjacent to any lake or stream
that is within a designated watershed. This requirement has been totally
ignored at Lake Glenville, which is within the Jackson County Watershed.
In a discussion of riparian buffer the slope of the terrain must be considered
when attempting to determine the needed depth from the waterline to permit
optimal functionality. The State did not consider severe slopes in setting its
minimum buffer width and Jackson County did not either. If the required
thirty feet of watershed buffer were in place at Lake Glenville the buffer
would still be inadequate due to the severe slope of the terrain.
We have not considered the cost of removing or destroying any bacteria,
fungi, antibiotics, toxins, or nutrients that maybe contained in the water.
Beginning in 1958 Duke Power began to remove the buffer strips from all of
its projects.
DPNA during the scoping hearing for the relicensing of p-2602 presented a
project map that did not show a buffer strip. Questioning revealed that it
was their intention to delete the buffer strip from their projects.
DPNA, a division of Duke Power, has failed in its responsibility to protect the
riparian buffer.
The value of riparian buffer as a Public Asset may exceed the value of
hydroelectric generation in some cases and must be given serious
consideration in determining the cumulative effects of projects.
It is obvious that greater protection must be given to the riparian buffer
around all of DPNA's lakes and that the impaired buffers must be repaired.
II. INTRODUCTION
The comments in this section are derived from a 1998 study by Marshall Flug
for the U.S. Geological Survey on "Ecosystem Resource Considerations in
Reservoir Management. "
"According to the Ecological Society of America (Lubchenco et al., 1991)
most current research focuses on commodity based managed ecosystems
with little attention to sustainability of natural ecosystems whose goods lack
a market value.
The true cost or value of ecosystem resources that include ecological
integrity and diversity remain unknown. Many water projects have been
developed over time to provide human and economic benefits, but without
5
accounting for the impacts and losses in water quality, biodiversity, wetlands,
riparian, and habitat area, to name a few resources that lack true market
value.
One of the recommendations cited in a paper on "America's Waters: A New
Era of Sustainability," by the Natural Resources Law Center in 1992 is:
"Federal hydropower pricing should reflect the full economic and
environmental cost of producing power, and revenues should be used to
assist in financing water conservation and ecosystem protection and
restoration."
The following comments and information are offered toward our concerns
over balancing power generation with the preservation of entire watersheds
and the ecosystems that sustain them.
III. WHAT IS THE RIPARIAN BUFFER:
[FROM:CHESALPEAKE BAY RIPARIAN HANDBOOK. A Guide for Establishing
and Maintaining Riparian Forest Buffers edited by Roxane S. Palone, and
Albert H. Todd, USDA FOREST SERVICE 1977]
"Riparian forest buffers are areas of trees, shrubs, and other vegetation
found next to stream channels and other waterways. They are modeled on
natural communities such as bottomland hardwood forest, coastal scrub, and
upland oak-hickory-pine forests. Conversion of these riparian forests to other
land uses has contributed to ecological problems in our waterways and the
Chesapeake Bay including sedimentation, nutrient and toxic chemical
pollution, and reduction of fish habitat.
Riparian wetlands are characterized by plant species adapted to periodic
flooding and/or saturated soils. They support a high diversity of plant and
animal species. More energy and materials, born by moving water, enter, are
deposited in, and pass through riparian ecosystems than any other wetland
ecosystem. Drier upland forests adjacent to waterways also provide many of
the same ecosystem values."
IV. The Function and Value of Riparian Vegetation:
From The Department of Natural Resources, The State of Washington.
"Stream or river banks are riparian areas, and the plants that grow there are
called riparian vegetation. Riparian vegetation is extremely important
because of the many functions it serves.
1. Bank stabilization and water quality protection
6
The roots of riparian trees and shrubs help hold stream banks in place,
preventing erosion. Riparian vegetation also traps sediment and
pollutants, helping keep the water clean.
2. Fish habitat
As dying or uprooted trees fall into the stream, their trunks, root wads,
and branches slow the flow of water. Large snags create fish habitat
by forming pools and riffles in the stream. Riffles are shallow gravelly
sections of the stream where water runs faster. Many of the aquatic
insects that salmon[fish] eat live in riffles. Salmon also require riffles
for spawning. They use pools for resting, rearing and refuge from
summer drought and winter cold.
3. Wildlife habitat
Over 80 percent of all wildlife species in western Washington use
riparian areas during some part of their life cycle. Riparian vegetation
provides food, nesting, and hiding places for these animals.
Unfortunately, forested riparian areas account for the smallest
percentage of forest land in Washington. [also true in NC]
4. Food chain support
Salmon and trout, during the freshwater stage of their life cycle, eat
mainly aquatic insects. Aquatic insects spend most of their life in
water. They feed on leaves and woody material such as logs, stumps
and branches that fall into the water from stream banks. Standing
riparian vegetation is habitat for other insects that sometimes drop
into the water, providing another food source for fish. [NC trout are
effected the same.]
5. Thermal cover
Riparian vegetation shields streams and rivers from summer and
winter temperature extremes that may be very stressful or even fatal,
to fish and other aquatic life. The cover of leaves and branches brings
welcome shade, ensuring that the stream temperature remains cool in
the summer and moderate in the winter. Cooler, shaded streams have
less algae and are able to hold more dissolved oxygen, which fish need
to breathe.
6. Flood control
During high stream flows, riparian vegetation slows and dissipates
floodwaters. This prevents erosion that damages fish spawning areas
and aquatic insect habitats.
7. Filtration:
The leaf litter acts as a filtration system by capturing sediment from
upland runoff. This action also helps to filter out phosphorous bonded
to sediment particles. The sediments and any nutrient or toxin, which
may be bonded to them, are trapped and become part of the forest
7
soil rather than clouding our waterways. Chemical and biological
processes of the forest remove nutrients, such as phosphorous and
nitrogen, and store them in the soil or as plant tissue. Pesticides are
also converted to nontoxic compounds by various chemical and
microbial activities within the forest. This helps to protect fish, which
are most threatened by pesticide pollution. Riparian forest soils act as
areas of water storage. Plants take up water into their tissues and
release it into the atmosphere."
As a primary filter for run-off the riparian buffer provides primary uptake
functions for the removal of nutrients and toxins contained in herbicides,
pesticides, fertilizers, and in human and animal waste. Therefore, we
conclude that any credible valuation of the cumulative impacts of project
operations must include the impacts on water quality and that impact on
public [drinking] water costs.
V. RECOMMENDED BUFFER WIDTHS:
8
lne
GamaIh...m 1rliI, aw,,1oo,g
Urals and ramannaffin man
raticall
203 Ora
buffer . 1.1 . 111. 1 orear
Urals
air
mFS Hainan Marat air gUmner
rancral
'150 If
air am Or mater III Par
Jra=� Rparan Polar ififfall Funalf and Or
ill
am Ri
1111 prItrali
11. 1 an Ilani 1.
of
r Nal
frater all flah .11111 al In
lor Inner 1.1
Or Or
.,.I
Ir Inarramn, faremarall., Ili lar
Applarfor Pirelli
renal Iranian
sami Or
Ill, I., Iranian,
filing Or Oal.
III all ramalm Oraral
an, in
Or Sai am Har PAI
artmar forn Odra Of Or mit Part
ill Or N.
I I
It am r later
I. IN. a 'Ogra, JW a
filial
X= =i2g Mr farina D
A Of GUMO I a..g airman Plan Rmr III, Iftrad Ima 11 IM
lne
GamaIh...m 1rliI, aw,,1oo,g

Putting Shorelands to Work
ow much of the shoreland needs to be protected?
H It depends on what you want the shoreland to be.
There is not a "one size fits-all" width that will
keep the water clean, stabilize the bank, protect fish and
wildlife, and satisfy human demands on the land. Many
localities have a minimum width requirement be sure to
check with your local planning or zoning office to find
out if one applies to your property. The minimum width
should provide acceptable levels of all needed benefits
at an acceptable cost. However, any amount of natural
shoreland is better than none. N
Width
i 7
i
r
r
r y?
t ?.
t
t
i pedestrian Pa
i
i
i
' ,
t
1
t
® Pier
Ordinary High
Water Mark
(OHWM)
How well a shoreland works depends on slope, soil,
natural vegetation, and other factors.
Width (feet)
Function 25 50 ]5 100 125 150 175 200 225 250 300 325 350 3X 400 425 q O 475 500
Water Quality
Flood Control
Shoreline Stability
Aquatic Habitat
Wildlife Habitat
Noise Reduction
M Unlikely to meet goal Meets goal with varying success depending M Exceeds goal
on site conditions

e_003 lrv the Bwld Hof Regent; of the UmVCL,i e: of tihxnnsin 5y,i.... .4II nghts level ed. rear i1'CVc10}eii 1!y' C11 i niani itii;9artI fdm Minc4. 111111"
Rnh.nK•,A. PJmn,: ronixrz.•alailrcm ticrtk.rir.mV11U111.11x1C1Inxni i&Ignr1 E.liring nlxi .irsi il•:[kr En..n nun.....d 8+zourrrs C.nrcc
?%5-Fite nsiou.
tetvEN PUB-CWQ038
DNR FLIP[-1VT-Y48 2003
Ensl or?
Liu r. rrd?
10
VI: RIPARIAN BUFFER VALUATION
The current total Public Asset loss in the riparian buffer of Lake Glenville
derived from this study is $227.5+ million. This value is determined quite
simply by multiplying the length of developed shoreline in feet by $200.00
per foot, the mitigation bank average cost to restore riparian habitat, then
multiplying by 15.5:1 which is the benefits transfer value for large scale
riparian habitat in the Little Tennessee River Watershed. This simple
arithmetic gives what would be the Public Asset Value of the Intact Riparian
Buffer. It should be noted that the $200.00 per foot is an average cost
factor for restoration of the riparian habitat derived from government
mitigation bank information and that the 15.5:1 factor is from a study by
Thomas P. Holmes et al that determined transfer benefit figures for value of
riparian buffers on the Little Tennessee River.
Note that our economic analysis does not contain an estimate of the volume
or dollar cost of the erosion and sedimentation caused by the removal of the
vegetative cover. Likewise, no estimate has been made of the quantity or
cost of abnormal nutrient or toxin transfers that have passed into the lake
and river system due to the removal of the native vegetative cover.
However, we have attached a study that developed the relative value of
riparian habitat in order to demonstrate the high value that such habitat
creates. [Contingent valuation, net marginal benefits, and the scale of
riparian ecosystem restoration; ECOLOGICAL ECONOMICS, October 2003, by
Holmes, Bergstrom, Huszar, Kask, and Orr, attachment B. Also in a separate
article Dr. Holmes has determined the cost of turbidity alone to be $113.50
per million gallons of treated water; (Thomas P. Holmes, "THE OFFSITE
IMPACT OF SOIL EROSION ON THE WATER TREATMENT INDUSTRY', LAND
ECONOMICS, NOV. 1988) and in 1985 Clark, Haverkamp, and Chapman,
determined the national cost to water utilities of turbidity due to agricultural
sediment turbidity to be 2.2 billion dollars.
MODEL PUBLIC DRINKING WATER UTILITY:
Assume a water treatment and delivery system with one million total users,
all residential.
In 1988, Thomas P. Holmes USDA/USFS, writing in "THE OFFSITE IMPACT OF
SOIL EROSION ON THE WATER TREATMENT INDUSTRY', LAND ECONOMICS,
NOV. 1988 determined that the cost of turbidity to a public water utility was
$113.50 per million gallons of treated water. The average public per capita
daily use of treated water is 100 gallons per day [Ross Bernstein, The Physics
Facts Book].
Obviously if there are one million people using 100 gallons of water per day
the daily usage will be 100 million gallons. And the daily cost due to turbidity
will be [$113.50 x 100 = 11,350.00 per day], and the annual cost to the
11
utility will be [$11,350 x 365 = $4,142,750.00 per year]. These cost break
down to $4.15 per person per year just to make drinking water clear; Not
safe to drink just appealing to look at. This also does not take into account
treated water used for industrial purposes which would raise the consumption
figure to 1316 gallons per day per person and would increase the per person
cost to $54.61 per year.
This model approximates the public water system for Metropolitan Charlotte,
North Carolina which is located on the Catawba River with the water intake at
Mt. Island Lake, p-2232.
Chattanooga, Tennessee which is roughly one half the size of Charlotte is
located downstream of Lake Glenville.
Charlotte as an industrial city would have an estimated cost due to turbidity
of 54.6 million dollars and Chattanooga at half the size but also an industrial
city would have an estimated cost due to turbidity of 27.3 million dollars.
COST OF REPLACEMENT
NCDOT and NC Clean Water Trust Fund Methodology cost estimate: for
replacing lost or damaged riparian buffers due to construction or other
causes.
Total shoreline = 27.7 miles
Developed shoreline = 13.85 miles
1 mile = 5280'1 5280' x 13.85 = 73,114 linear feet
73,114 linear feet times $200.00/foot = $14,728,000.00
The Holmes, Bergstrom, Kask, et al study "COMPARATIVE BENEFITS
TRANSFER OF RIPARIAN HABITAT" which was done in Macon County, N.C.
and Rabun County, Ga. indicates a benefits to cost ratio of 15.5:1 or higher
for restoration of contiguous habitat of eight miles or greater in length.
By inference from their study a ratio of 15.5:1 would indicate a value of
$3100.00 per foot of intact riparian habitat having a continuous length of
eight or more miles. By extension this would indicate a total Public Asset
value for an intact restoration of the impaired full shoreline of Lake Glenville
as being $453,306,800.00 as a public asset. The lost Public Asset value of
the impaired 50% of the shoreline would be $226,653,400.00
VII: EXISTING CONDITION OF LAKE GLENVILLE HABITAT:
In conclusion Dr. Pittillo's study shows that there is overwhelmingly
significant existing and ongoing physical damage to the natural riparian
vegetative cover of the buffer strip coincident with development adjacent to
the shoreline buffer zone of Lake Glenville and our analysis places the
imputed Social or Public Costs of this damage as at least $227.5 million.
12
Further, the nature of riparian habitat is that fragmentation destroys its
ability to be an effective filter progressively disproportionate to the percent of
fragmentation, and thus the fragmentation of the Lake Glenville riparian
border has rendered the remaining vegetative cover inadequate to perform
its natural functions of stabilization and filtration. This is a major
contributing cause of the continual erosion of the shoreline, which has led to
the loss of significant pieces of the designated buffer zone. This erosion
process has not been effectively managed by DPNA.
Without better and more intense management, and assuming further
development into the areas where natural vegetation still exists, any viability
of the buffer zone will likely be eliminated during the next relicensing term.
Mitigation for the lost vegetative cover and shoreline is indicated and better
management of the remaining intact vegetation is required.
VIII: HISTORY: Responsibility for Loss
In 1958 Duke Power parent company of DPNA applied for and was granted a
new 50 year combined license for all of their lakes on the Catawba River on
the strength of the construction of the new Lake Norman dam. Before the
license was issued Duke Power began to transfer all of the riparian buffer
property associated with these lakes and the river to another subsidiary or
division named Crescent Land and Timber, now Crescent Resources.
There are thirteen Duke Power lakes on the Catawba River having a
combined shore line length of over fifteen hundred miles. Of those fifteen
hundred miles there is no buffer strip within the project boundaries according
to the information that we have at present. Much of the Catawba River
shoreline is owned by Crescent Resources a former division of Duke Power
and now a sister division of Duke Energy and again there is little or no
protected riparian buffer. Only thirteen miles of shoreline on Lake Norman
and a small amount on Lake James have been preserved within State Parks
for public access. There is no riparian buffer strip on Duke's Catawba lakes
except for Mt. Island Lake, p-2232, and that buffer is on Crescent Resources
property. All of which are within the watershed for some public water
utility.
In 1997 DPNA filed a Shoreline Management Plan with the FERC for all of
their lakes. On July 1, 2003 DPNA put in place a new Shoreline Management
Plan. DPNA has failed to protect the riparian buffer under either of these
plans.
During the previous round of scoping hearings for the DPNA license renewals
DPNA presented project maps that did not show the buffer strip. The WNCA
caught this omission and pointed it out in the public hearing for the Dillsboro
Dam, p-2602. Hopefully this omission has been corrected, but we have no
proof of this since DPNA has declared this to be critical energy infrastructure
information under and has not provided definitive maps that would show the
buffer strip accurately.
13
We believe that this history establishes that neither Duke Energy, Duke
Power, or DPNA are sufficiently serious or protective of the Public Interest to
be entrusted with the care of a Public Asset as valuable and necessary as the
riparian buffer.
The loss of native vegetative cover in the riparian buffer has been a direct
result of the failure of DPNA to properly manage the buffer and to oversee
the actions of the land owners within project boundaries. The adjacent
owners were actively responsible for the removal of the vegetation, but
DPNA, as the licensee, must be held primarily responsible. In failing to
actively protect the buffer strip and enforce their Shoreline Management
Guidelines, DPNA facilitated and expedited the destruction of the protective
vegetative cover and habitat.
Today the effected riparian buffer at Lake Glenville serves none of the
purposes prescribed for it by either law or nature.
1. It is too small for habitat.
2. It is both to narrow and too sparse to be an effective filter.
3. It does not have the dense root mass or large woody debris
necessary to be an effective erosion control agent.
4. It has very limited access and is too narrow in many places to
provide either public access or public recreation.
In short the buffer strip at Lake Glenville serves none of the purposes that
are required of it and DPNA has made no effort to improve that situation or
to protect what little buffer there is.
The cumulative impacts of these projects can not fully be assessed without
accounting for the impact of the loss of the integrity of the riparian buffer
and the social and economic costs associated with that loss beyond the
bounds of these projects.
Both the State of North Carolina and Jackson County, North Carolina require
a thirty foot buffer strip be maintained within Watershed areas. All of DPNA's
East and West Fork lakes are within Jackson County designated Watershed.
DPNA has failed to protect that required thirty foot buffer.
IX: RECOMMENDATIONS
Many agencies have recommended a minimum buffer depth from water line
of fifty feet. TVA's new Shoreline Management Policy became effective in
November, 1999 and requires that a fifty foot shoreline management zone
[SMZ] will be established on TVA land that adjoins newly developed
residential areas. We recommend that due to the severity of the slopes this
same fifty feet of shoreline management zone should be required on all lands
around all DPNA projects as a minimum; not just newly developed or not yet
developed residential areas. We request that The Commission order the
14
insertion of such a requirement into the DPNA Shoreline Management
Guidelines or that The Commission make it a condition of the license. Section
10(a)(1) of the Federal Power Act allows inclusion of existing residential,
commercial, or other structures within the project boundaries for the purpose
of achieving beneficial public uses i.e. protection of public water supply. 18
CFR 4.41(h)(2) and 4.51(h)(2) state "The map must show a project
boundary enclosing all of the principal project works and other features... The
boundary must enclose only those lands necessary for operation and
maintenance of the project and for other project purposes, such as
recreation, shoreline control, or protection of environmental resources (see
paragraph (f) of this section (Exhibit E)). Existing residential, commercial, or
other structures may be included within the boundary only to the extent that
underlying lands are needed for project purposes (e.g., for flowage, public
recreation, shoreline control, or protection of environmental resources)..."
Due to Duke Power's history of failed enforcement and disregard for the
buffer strip and Public Assets, we request in addition that:
1. The Commission consider that the DPNA lakes buffer strips be placed
under a protective Conservation Easement to be held in trust by the
Southern Appalachian Highlands Land Trust or The Land Trust for The
Little Tennessee or other suitable third party and that the particulars
of this easement be the same as DPNA's Shoreline Management
Guidelines but to include a requirement for at least a 50 feet wide
riparian buffer and not more than a 20 feet wide view shed including a
10 feet wide access way for access to the lakes. The WNCA does not
wish to interfere with the adjoining land owners or the general publics
enjoyment of the lakes.
2. That if this buffer, once established, is not maintained by the
adjoining property owner then lake access and dock permits for said
owner or owners are to be withdrawn without review and
accompanying docks removed at the permitees expense.
3. We further request that The Commission require DPNA at its sole
expense to re-plant the vegetated buffer strip wherever it may be
impaired on all of its shorelines.
4. That the EA/EIS should include a study to determine the proper depth
from the waterline needed for the buffer zone considering the severity
of the slope at the DPNA projects.
5. The current project boundary should be expanded to include all lands
owned by Duke Power-Nantahala Area around all of its lakes and along
the rivers near their Dams. This expanded project boundary is
necessary to protect project purposes including recreation, shoreline
control, and protection of environmental resources.
15
6. In addition to Lake Glenville Bear Lake on the East Fork, p-2698, is
currently under intense development pressure and has a marginal
buffer strip with severe slopes. Bear Lake should also have a study to
determine the proper depth from water line for the buffer.
7. we applaud DPNA's efforts to accommodate the American Whitewater
Association and the Commercial rafting and kayaking businesses with
timed releases. We would however point out that many local citizens
and Western Carolina University and Southwestern Community College
students raft and kayak on the East Fork justifying the new put in and
take out points that DPNA plans to build. They would benefit
significantly from daytime flow increases that should not interfere with
the downstream scheduling currently planned by DPNA.
8. That the 35 cfs minimum flow for the East Fork should begin in mid
May or not later than the week before Memorial Day and should
continue through November 1.
9. That DPNA should pay annually to the Soil and Water Conservation
Districts of the five DPNA Counties for the life of the license Two
hundred thousand dollars. Said funds to be divided forty per cent to
Jackson County and the balance equally among the others and to be
used for stream bank restoration as mitigation for project operations.
We fully agree with comments submitted by North Carolina Wildlife
Resources Commission, NC Water Resources, US Fish &Wildlife Service and
THE US Department of Interior as presented for projects p-2619 and p-2603
accession #s: 20041213-5006, 20041213-5013,20041213-5044,20041213-
5060, and20041222-0158. We believe that the conditions existing in those
projects are also present in all of DPNA's Tuckeseegee projects as well and
that the recommendations these resource agencies propose for p-2619 and
p-2603 should also be applied to p-2686 and p-2698.
We believe the above recommendations are the least necessary for the
restoration of a functioning and viable riparian buffer zone at Lake Glenville
and other DPNA Lakes in order to better protect the publics' interests and
welfare and to mitigate damage due to project operations.
REFERENCE AND BIBLIOGRAPHY:
Comprehensive Plan - FPA, Section 10(a)(1) of the act specifies that a
hydroelectric project can be developed by an individual or corporation, or an
agency of a municipality or state provided it is "best adapted to a
comprehensive plan for improving or developing a waterway or waterways
for the use or benefit of interstate or foreign commerce, for the improvement
16
and utilization of water power, for the adequate protection, mitigation,
and enhancement of fish and wildlife... and for other beneficial public uses
including irrigation, flood control, watersupply, and recreationaland
other purposes...". FERC requires project applicants to identify all applicable
comprehensive plans developed by state and federal agencies and determine
if the project will comply with these plans (18 CFR 4.38). See section 3.1.d
for examples of these plans.
Project Boundary - 18 CFR 4.41(h)(2) and 4.51(h)(2)
. The map must show a project boundary enclosing all of the principal project
works and other features... The boundary must enclose only those lands
necessary for operation and maintenance of the project and for other project
purposes, such as recreation, shoreline control, or protection of
environmental resources (see paragraph (f) of this section (Exhibit E)).
Existing residential, commercial, or other structures may be included within
the boundary only to the extent that underlying lands are needed for project
purposes (e.g., for flowage, public recreation, shoreline control, or protection
of environmental resources)...
"A provision for a shoreline buffer zone that must be within the
project boundary, above the normal maximum surface elevation of
the project reservoir, and of sufficient width to allow public access to
project lands and waters and to protect the scenic, public
recreational, cultural, and other environmental values of the
reservoir shoreline" 18CFR4.41(f)(7)(iii)
"There is no Economic Value in clean water"
was one of Dr. Hugh McCauley's [professor of Economics, Clemson
University] favorite attention getters. There is no economic value in clean
water because in Economic terms there is no added value to clean water; it is
natural. What is the value of clean water today? Between $0.79 per gallon
and $8.00 per gallon because that is what we now pay for bottled water
because we are afraid of our public drinking water supply.
Water consumption:
When calculating the total amount of water consumed by the United States,
we must first define and separate water withdrawal from water consumption.
Withdrawal refers to water extracted from surface or ground water sources,
with consumption being that part of a withdrawal that is ultimately used and
removed from the immediate water environment. The majority of water
consumption comes from domestic use. It is surprising to hear that each
person in the United States uses approximately 380 liters of water each day
(100 gal). [Ross Bernstein, The Physics Fact Book. Hypertextbook.com]
There is a web site of water consumption data in the US from the year 1994
17
that states "In an industrial society such as the United States, personal
water consumption is between 200 and 300 liters per day (Fetter, 1994).
when the industrial and energy production usage is added in to the equation,
fresh water usage exceeds 5,000 liters per day on a per capita basis
(Fetter, 1994)." The website is:
http://www.emporia.edu/earthsci/student/eslick7/webpage.html
Benefits Transfer Bibliography
Alberini, A., and M. Cropper, Tsu-Tan Fu, A. Krupnick, Jin-Tan Liu, D. Shaw,
W. Harrington, 1997, "Valuing Health Effects of Air Pollution in Developing
Countries: The Case of Taiwan," Journal of Environmental Economics and
Management, 34, 107-126.
Asian Development Bank,1995, Economic Evaluation of Environmental
Impacts: A Workbook, prepared by Hagler Bailly Consulting Inc.
Atkinson, S.E., and T.D. Clocker, J.F. Shore, 1992, "Bayesian
Exchangeability, Benefit Transfer, and Research Efficiency", Water Resources
Research, Col. 28. No. 3.
Bergstrom, J.C. and De Civita, P., 1999, "Status of Benefits Transfer in the
United States and Canada: A Review," Canadian Journal of Agricultural
Economics 47, pg. 79-87.
Bingham, T., and Kealy, M.J., David, E., LeBlanc, M., Graham-Tomassi, T.,
Leeworthy, R. (eds.), 1992, "Benefits Transfer: Procedures, Problems, and
Research Needs," Proceedings from a 1992 Association of Environmental and
Resource Economists Workshop, Snowbird, Utah, June 3-5.
Bowker, J.M., and D.B.K. English, J.C. Bergstrom, 1997, "Benefit Transfer
and Count Data Travel Cost Models: An Application and Test of Varying
Parameter Approach with Guided Whitewater Rafting," Working Paper FS 97-
03, Department of Agricultural and Applied Economics, University of Georgia,
Athen, Georgia.
Boyle, K.J., and J.C. Bergstrom, 1992, "Benefit Transfer Studies: Myths,
Pragmatism, and Idealism", Water Resources Research, Cal. 28. No. 3.
Boyle, K.J., and G.L. Poe, J.C. Bergstrom, 1994, "What Do We Know About
Groundwater Values? Preliminary Implications from a Meta Analysis of
Contingent Valuation Studies", American Journal of Agricultural Economics,
76(5).
Brookshire, D.S., and H.R. Neill, 1992, "Benefit Transfers: Conceptual and
Empirical Issues", Water Resources Research, Vol. 28. No. 3.
18
Brouwer, R. and F. A. Spaninks, 1999, "The Validity of Environmental Benefit
Transfer: Further Empirical Testing," Environmental and Resource
Economics„ 14: 95-117.
Desvousges, W.H., and M.C. Naughton, G.R. Parsons, 1992, "Benefit
Transfer: Conceptual Problems in Estimating Water Quality Benefits Using
Existing Studies", Water Resources Research, Vol. 28. No. 3.
Desvousges, W.H., and F.R. Johnson, H.S. Banzhaf, 1998, "Environmental
Policy Analysis with Limited Information: Principles and Applications of the
Transfer Method," Northampton, MA, Edward Elgar.
Downing, M. and T. Ozuna, 1996, "Testing the Feasibility of Intertemporal
Benefits Transfers Within and Across Geographic Regions", Journal of
Environmental Economics and Management, 30, 316-322.
Feather, P. and D. Hellerstein, 1996, "Calibrating Benefit Function Transfer
to Assess the Conservation Reserve Program", American Journal of
Agricultural Economics, 79, February, 151-162.
French, D.D., 2001, "Three Essays on Benefit Transfer," PhD. Thesis, Ohio
State University, Columbus, Ohio, 43210.
Garrod, G. and K. Willis, 1999, "Benefit Transfer (Chapter 12)," in Economic
Valuation of the Environment: Methods and Case Studies, Edward Elgar
Publishing Limited, Cheltenham, UK.
Kask , S.B., and J.F. Shogren, 1994, "Benefit transfer protocol for long-term
health risk valuation: A case of surface water contamination", Water
Resources Research, Vol. 30, NO. 10.
Kirchhoff, S., and B.G. Colby and J.F. La France, 1997, "Evaluation the
Performance of Benefit Transfer: An Empirical Inquiry," Journal of
Environmental Economics and Management, 33, 75-93.
Leon, C.J. and F.J. Vazquez-Polo, N. Guerra, P. Riera, 2002, "A Bayesian
Model for Benefits Transfer: Application to National Parks in Spain," Applied
Economics, 34, 749-757.
Loomis, J.B., 1992, "The Evolution of a More Rigorous Approach to Benefit
Transfer: Benefit Function Transfer", Water Resources Research, Vol. 28. No.
3.
Loomis, J.B., and B. Roach., F. Ward, R. Ready, 1995, "Testing
Transferability of Recreation Demand Models Across Regions: A Study of
Corps of Engineer Reservoirs", Water Resources Research, Vol. 33. No. 3.
Lovett, A. A., and J.S. Brainard, I.J. Bateman, 1997, "Improving Benefit
Transfer Demand Functions: A GIS Approach," Journal of Environmental
Management, 51, 373-389.
19
McConnel, K.E., 1992, "Model building and judgement: Implications for
benefit transfers with travel cost models", Water Resources Research, Vol.
33. No. 3.
Moran, D. and D. Pearce, 2001, "Economic Valuation: Benefits Transfer",
Chapter 8 in "Handbook on the Applied Valuation of Biological Diversity,"
Organization for Economic Co-operation and Development, Paris, France.
Moran, D., 1999, "Benefits Transfer and Low Flow Alleviation: What Lessons
for Environmental Valuation in the UK,"Journal of Environmental Planning
and Management„ 42(3), 425-436.
Morrison, M. and J. Bennet, R. Blarney, J. Louviere, 1998, "Choice Modelling
and Tests of Benefit Transfer," paper presented at the World Congress of
Environmental and Resource Economists, Venice, Italy.
Naughton, M.C., and W.H. Desvousges, 1986, "Water Quality Benefits of
Additional Pollution Control in the Pulp and Paper Industry", Report prepared
for the U.S. Environmental Protection Agency.
Navrud, S., 2001, "Valuing Health Impacts from Air Pollution in Europe,"
Environmental and Resource Economics 20: 305-329.
Navrud, S., 2002, "Comparing Valuation Exercises in Europe and the United
States - Challenges for Benefit Transfer and Some Policy Implications,"
Chapter 4 in "Valuation of Biodiversity Benefits - Selected Studies," published
by the Organization for Economic Co-operation and Development, Paris,
France.
O'Doherty, R.K., 1995, "A Review of Benefit Transfer: Why and How," British
Review of Economic Issues, Volume 17, Number 43, October.
Pearce, D. and D. Whittington, S. Georgiou, D. James, 1994, "Ch. 10 in
Project and Policy Appraisal: Integrating Economics and the Environment",
Organization for Economic Co-operation and Development, Paris, France.
Piper, S., 1998, "Using Contingent Valuation and Benefit Transfer to
Evaluate Water Supply Improvement Benefits," Journal of the American
Water Resources Association, Vol. 34, No. 2, April.
Piper, S. and W.E. Martin, 2001, "Evaluating the Accuracy of the Benefit
Transfer Method: A Rural Water Supply Application in the U.S.A.," Journal of
Environmental Management,63, 223-235.
Ready, E., and Navrud, S.; Day, B., Duborg, R., Machado, F., Mourato, S.,
Spanninks, F., Rodrquez, M., 2002, "Benefit Transfer in Europe: Are Values
Consistent Across Countries?" working paper supported by the European
Union's Environment and Climate Research Programme.
20
Rosenberger, R. and J.B. Loomis, 2001, "Benefit Transfer of Outdoor
Recreation Use Values," U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station, Fort Collins, Coldorado.
Ruijgrokok, E.C.M., 2001, "Transferring Economic Values on the Basis of an
Ecological Classification of Nature," Ecological Economics39, 399-408.
Shrestha, R.K. and J.B. Loomis,2001, "Testing a Meta-Analysis Model for
Benefit Transfer in International Outdoor Recreation" Ecological Economics
39, 67-83.
Smith, V.K. and G.V. Houtven and S.K. Pattanayak, 2002, "Benefit Transfer
via Preference Calibration: "Prudential Algebra" for Policy," Land
EconomicsFebruary, 78 (1): 132-152.
Smith, V.K., 1992, "On Separating Defensible Benefits Transfers From
'Smoke and Mirrors"', Water Resources Research, Vol. 28. No. 3.
Vanden Berg, T.P., and G.L. Poe, J.R. Powell, 1995, "Assessing the Accuracy
of Benefits Transfers: Evidence from a Multi-site Contingent Valuation Study
of Groundwater Quality", Department of Agriculture, Resource and
Managerial Economics, Cornell University, WP95-01.
Walsh, R.G., and D.M. Johnson, J.R. McKean, 1992, "Benefit Transfer of
Outdoor Recreation Demand Studies, 1968-1988", Water Resources
Research, Vol. 28. No. 3.
21
Appendix 4
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
Andrews Park 0-10 m 10-2jOlan pitt;tjl?'30m 30-40m 40-50m
Control Acer rubrum 4 10.5 3.2 7.5 4 7 3.5 5 2.4 10
28-Jul-04 10.4 12 5 8 4.2 5.7 2 6.5
2.7 6.5
Prunus seroth 8.5 15
Pinus strobus 4 6
Liriodendron tulipifera 8.2 9.5
Tsuga canadensis 3.8 6
Carya glabra 3.2 4.5
Quercus alba 11.5 19
Betula lenta 7 8.5
Quercus rubra 3.5 7
Quercus coccinea 7 11.5
Quercus velutina 4.5 4.7
Nyssa sylvatica 2.5 5.7
Values in columns 1 are line intercepts (I) in m; col umns 2 are maxi mum diamete
Andrews Park 0-10m 10-201 20-30m 30-40m 40-50m
Experimental Quercus velut 5.2 5.5 14.5 18
28-Jul-04 Pinus strobus 11.4 8.5 10.5 13 3 3
Acer rubrum 6.9 9.5
Liriodendron tulipifera 7.7 10.5 6.9 9
Carya glabra 3 5
Robinia pseudoacacia 7.1 8.5
Oxydendrum arboreum 4 6
Quercus alba 6.5 6.5
Cornus florida 1 10.5
Amelanchier laevis 4.8 7.5
Pinus rigida 1 6
Values in columns 1 are line intercepts (I) in m; col umns 2 are maxi mum diamete
West 1 0-10 m 10-20m 20-30m 30-40m 40-50m
Control Carya glabra 2.3 5.7 5.3 5
2-Aug-04 Pinus strobus 0.8 4.3
Robinia pseuc 3.3 5.4 0.2 3.6
Acer rubrum 3.5 3.5 4.8 4.8 5.4 8
Betula lenta 6.6 10.8 2.5 6.7 4.6 5.4 1.5 4 6 6.8
4 5.2
Quercus rubra 5.8 7 7 7
Oxydendrum arboreum 4.6 6.6
5 5
Amelanchier laevis 4.2 5.7
Tsuga canadensis 6 7
Values in columns 1 are line intercepts (I) in m; col umns 2 are maxi mum diamete
West 1 0-10m 10-201 20-30m 30-40m 40-50m
Experimental Pinus strobus 7.4 7.42 11 112 11 11 4.2 11
2-Aug-04 Sassafras alb 4 4
Amelanchier 1 3 10
Nyssa sylvati( 5 8
Prunus serotina 5.8 9.3
East 1 0-10 m 10-20m 20-30m 30-40m 40-50m
Control Tsuga canade 1.2 5.4
7-A..,.-nn a ? a ?
Appendix 4
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
Betula lenta 3 19.2 OJ5Dan`4?atillo7.4 8.8 6 10 6 8.2
East 1 Control, cont.
3.7 8.4 0.5 6.5
4.5 8.7 7.2 79
12.6 12.6
Quercus rube 4.6 7.8 9.3 11.2 0.5 11.2
10.3 12
Aesculus flav? 0.5 8.5
Acer rubrum 3.5 4
4.4 10
Robinia pseudoacacia 5.8 8.9
Values in columns 1 are line intercepts (I) in m; columns 2 are maximum diamete
East 1 0-10m 10-201 20-30m 30-40m 40-50m
Experimental Tsuga canade 1.2 5.3
7-Aug-04 Liriodendron i 1.3 3.9 5 12
6 9
Quercus rube 3 15
Hamamelis vi 3 6.6
Quercus cocci 1.8 5.4
Pinus strobus 2.4 4.2 6.3 11.5
Acer rubrum 8.8 11 5.3 6.2 1 7.5
Magnolia fraseri 13 18.5
Values in columns 1 are line intercepts (I) in m; columns 2 are maximum diamete
West 2 0-10 m 10-20m 20-30m 30-40m 40-50m
Control Liriodendron i 6.8 13.5 8 11.3
9-Aug-04 Acerrubru m 9.1 11 2.4 7.2 9.2 10.8 10.1 13
4.2 5
Prunus seroth 9.4 14.3 9.2 10.7
Cornus floridz 0.2 11.3 0.2 3.4
3.2 4.5
ParthenoClssL 0.5 8.4
Vitis aestivali: 0.1 8.4 3.2 5.7
Robinia pseudoacacia 5.5 6.4 1.6 5.5
Pinus strobus 10 10
Values in columns 1 are line intercepts (I) in m; co lumns 2 are maxi mum diamete
West 2 0-10m 10-20m 20-30m 30-40m 40-50m
Experimental
9-Aug-04 Acerrubru m 2 7.8 8.8 11.5
Quercus rube 1 5.4 3.4 6.6
6.5 12.8
Liriodendron i 2.2 12.7 7 9.7 7 10
Prunus serotina 8.2 8.8
Pinus strobus 9 11.5 2.5 5.7
6.3 9
Robinia pseudoacacia 4.8 7 4 5
Cornus florida 5 7.4
Appendix 4
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
J. Dan Pittillo
East 2 0-10 m 10-20m 20-30m 30-40m 40-50m
Control Pinus rigida 0.4 4.3
10-Aug-04 Quercus velut 1.6 10.8 9.1 9.1 3.6 8.4 5.5 10.5
3.1 8.7
Pinus strobus 8.7 8.7 3.8 7.1 6.6 6.8 5.5 7.4
4.4 5.4 4.2 7 2.2 7.1
Prunus serotina 3.4 4.2
Carya alba 1 6.4 3.8 4 4.9 8.8
Quercus coccinea 4.8 6.7
Oxydendru m arboreu m 2.7 10.8
Values in columns 1 are line intercepts (I) in m; columns 2 are maxi mum dia mete
East 2 0-10m 10-20m 20-30m 30-40m 40-50m
Experimental Carya alba 4.7 5.4 1.2 6.4
9-Aug-04 Pinus strobus 2.2 8.5 13.2 13.2 9.9 9.9
Betula lenta 3 7.5
Quercus cocci 6.9 6.9 4.5 10.1 2.8 4.2
Quercus velut 2.5 6.9 6 11.2 2 5.4 5.3 8.5
1.9 3.1
1 5.9
Acer rubrum 6.5 6.5 3.9 5.4
5.9 6.9
Values in columns 1 are line intercepts (I) in m; columns 2 are maxi mum dia mete
West 3 0-10 m 10-20m 20-30m 30-40m 40-50m
Control Pinus strobus 8.2 8.2
16-Aug-04 Liriodendron i 0.5 8.9 9 11
Tsuga canade 2.6 3.8 9 11
1.6 3.4
Robinia pseuc 4.8 6
Acer rubrum 3.7 7.3
Betula lenta 3.7 5.4 4.2 7.4
8 9
Prunus serotina 4.7 5.1
0.2 6
Values in columns 1 are line intercepts (I) in m; columns 2 are maxi mum dia mete
West 3 0-10m 10-20m 20-30m 30-40m 40-50m
Experimental Carya alba 0.4 3.8
9-Aug-04 Liriodendron tulipifera 4 7.6 6.3 6.3
Quercus rubra 9 9.5
6.7 7.5
Robinia pseudoacacia 0.5 6.4
Prunus serotina 5.1 6.4
4.5 6.3
Values in columns 1 are line intercepts (I) in m; columns 2 are maxi mum dia mete
East 3 0-10 m 10-20m 20-30m 30-40m 40-50m
Control Pinus strobus 2.5 6.2 7.1 8.9 7 7 1.6 7.4
27-Aug-04 2.3 5 3.5 5.4 4.1 5.2 3 10.8
2 9.9
Oxydendrum 4.3 10.3
Acer rubrum 2.3 5.8 5.6 8.5
n........ ...I..l 7 a 7 9 1 n C 1 1 1' 1A C 1 o c a
Appendix 4
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
Experimental Quercus alba J. Dan Pittillo 10 10.8 0.2 6.8
26-Aug-04 Quercus coccinea 7 7.6 3.5 8 5.4 18.7
Values in columns 1 are line intercepts (I) in m; columns 2 are maximum diamete
Appendix 5
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
I Dan Pittillo
Andrews Park 0-Sm 5-10m 10-15m 15-20m 20-25m 25-30m 30-35m 35-4 0m 40-45m 45-50m
Control Hamamelis virgir 120 220 190 350 30 117
28-Jul-04 80 170 160 595 190 280
Vaccinium stamir 65 94 25 87 63 135
Clethra acu minata 102 156
Corpus florida 75 80
Acer rubrum 250 330
Calycanthus floridus 52 62 30 80
31 64
Fagus grandifolia 450 570
Linodendron tulipifera 170 170
Castanea pumila 92 190
Quercus rubra 124 140
Pinus strobus 15 125 55 130 55 90
Robinia pseudoacacia 290 290
Kalmia latifolia 280 350 55 220
Betula lenta 160 230
Oxydendrum arboreum 140 250
Nyssa sylvatica 10 145 160 220
32 105 100 320
180 255 90 330 115 175
30 145 15 35
Gaylussacia ursina 45 55 85 55
55 80 55 85
50 65 25 75
35 100 45 55
35 70 20 60
40 55 30 60
45 65 25 45
50 55
Sassafras albidu m 20 110
Rhododendron maximum 25 170
Smilax glauca 15 100
Smilax rotundifolia 15 120
For each line intrval, First column line cm distance covered (I); second column cm maximum plant diameter (M
Andrews Park 0-Sm 5-10m 10-15m 15-20m 20-25m 25-30m 30-35m 35-4 0m 40-45m 45-50m
Experimental Vaccinium stamir 80 80 62 117
28-Jul-04 24 45 11 33
Betula lenta 45 90
Amelanchier laev 28 31 10 19 9 19
Gaylussacia ursina 3 12
3 12
Sassafras albidum 10 74
Vaccinium pallid 73 97
30 60
25 50
Castanea pumila 67 240
49 145
Smilax glauca 3 3
For each line intrval, First column line cm distance covered (I); second column cm maximum plant diameter (M
West 1 0-Sm
Control Gaylussacia ursir 6
28-Jul-04 45
27
36
51
40
79
32
5-10m 10-15m 15-20m 20-25m 25-30m
33 83 100 67 67 43 56
97 37 60 86 116 10 64
63 19 64 52 52 34 72
88 23 70 44 80 28 70
77 45 80 31 80 13 55
42 50 100 45 70 50 50
88 1 100 46 47 44 55
67 45 45 43 80
190 320
30 30
47 47
42 60
8 55
57 57
27 55
30-35m 35-40m 40-45m 45-50m
28 62 38 50
16 44 42 45
14 40 22 70
50 70 43 57
30 67 34 51
11 40 1 58
40 49 24 64
4 56 23 30
31 35
Appendix 5
Lake Glenville Buffer Zo ne Study
Raw Data (Collected July 28-A ugust 26, 20 04)
J. Dan Pittillo
112 112
Gaylussacia ursir 20 37 25 55 20 47 9 40 10 40 27 47
West 1 Control, cont.
29 55 82 170 23 77 19 57
50 60 8 44
Rhododendron calendulacei 125 250 45 130 4 57
110 200 45 130
34 200
100 290
Quercus rubra 190 320
Hamamelis virginana 180 180 7 320
310 310
Clethra acuminata 170 250 31 80 168 192 45 54 31 120 24 52
16 160 1 116 48 80
53 215 21 150 55 110
50 125
40 90
32 115
Acer rubrum 140 395 24 30
200 280
230 400
Kalmia latifolia 70 275 65 100
115 250
Rhododendron maximum 200 300
Vaccinium corymbosum 128 170
For each line intrval, First col umn line cm distance covered (I); second column cm maximum plan t dia meter (M
West 1 0-5m 5-10m 10-15m 15-20m 20-25m 25-30m 30-35m 35-4 0m 40-4 5m 45-50m
Experimental Rhodendron cale 73 207
28-Jul-04 100 345
180 229
Hydrangea paniculata 49 78
Kalmia latifolia 112 150 51 113
59 114
Robinia pseudoacacia 17 113
Leucothoe fontanesiana 29 119
Amelanchier laevis 42 62
For each line intrval, First col umn line cm distance covered (I); second column cm maximum plan t dia meter (M
East 1 0-5m 5-10m 10-15m 15-20m 20-25m 25-30m 30-35m 35-4 0m 40-4 5m 45-50m
Control Kalmia latifolia 62 62 74 74
7-Aug-04 53 151
Betula lenta 310 480 50 280 210 210 380 410 210 305
Clethra acuminata 106 160 175 175 135 395
125 155 127 170
Halesia tetaptera 30 300 38 85
Ilex ambigua 76 124
Tsuga canadensis 118 410 180 350 105 307 84 480
165 270 40 210
Hamamelis virginiana 97 173 5 377 65 155 9 810 170 440
9 470 8 855
80 120
330 330
Calycanthus floridus 8 45 29 46
Gaylussacia ursina 2 54 92 110 72 100 8 63 30 72
24 84 51 59 32 65 16 61
26 92 49 82 56 101
Amelanchier laevis 45 45
52 67
Rhododendron maximum 39 280 180 450
32 240
25 115
12 63
440 610
41 52
49 61
Castanea dentata 17 119
Magnolia fraseri 190 220
East 1 O-Sm 5-SOm SO-SSnn 15-20m 20-25m 25-30m 30-35m 35-4 0m 40-4 5m 45-SOnn
Appendix 5
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
J. Dan Pittillo
Pyrularia pubera 1 65
Castanea dentata 290 570
Rubus canadensis 45 93
West 2 0-Sm 5-10m 10-15m 15-20m 20-25m 25-30m 30-35m 35-40m 40-45m 45-50m
Control Amelanchier laevis 10 81
9-Aug-04 Fraxinus americana 28 53
Pinus strobus 12 58
32 69
32 58
114 170
150 260
West 2 0-Sm 5-10m 10-15m 15-20m 20-25m 25-30m 30-35m 35-40m 40-45m 45-50m
Experimental Rhododendron or 34 110 115 157
9-Aug-04 44 182
Sassafras albidu m 90 15C
East 2 0-5r
Control Carya alba 0.3
10-Aug-04 Crataegus mono! 380
Quercus rubra 151
Corpus florida 100
Vaccinium stamir 52
23
a 5-10m 10-15m 15-20m 20-25m 25-30m 30-35m 35-40m 40-45m 45-50m
12 86 86 106 120
420 200 255
188
131
86 66 93
112
Quercus coccinea 86 155
Smilax rotundifolia 78 167 23 140
Amelanchier laevis 170 320 224 256 57 71 90 133
1 152 14 59
11 20
410 525
Vaccinium simulatum 198 198
107 283
Robinia pseudoacacia 64 118
Acer rubrum 65 82
10 57
3 79
Sassafras albidu m 11 59
Pinus strobus 39 42
9 46
Tsuga canadensis 86 200
Gaylussacia ursina 99 113
45 115
67 75
18 82
For each line intrval, First column line cm distance covered (I); second column cm maximum plant diameter (M
East 2 0-Sm 5-10m 10-15m 15-20m 20-25m 25-30m 30-35m 35-40m 40-45m 45-50m
Experimental Vaccinium stamir 50 53
10-Aug-04 Vaccinium corymbosum 69 157
Amelanchier laevis 330 570
300 470
Rhododendron hybrid 130 180 52 110
Forsythia virdissima 124 253
Quercus velutina 180 410
Kalmia latifolia 70 132
Acer rubrum 56 124
For each line intrval, First column line cm distance covered (I); second column cm maximum plant diameter (M
West 3 0-Sm 5-10m 10-15m 15-20m 20-25m 25-30m 30-35m 35-40m 40-45m 45-50m
Control Quercus rubra 90 139 44 56 90 90 1 33 38 66 1 117 173 430
9-Aug-00 56 145 103 183 70 93
73 146
Smilax rotundifol 47 116 26 140
76 234
Sassafras albidum 130 130 1 71 1 101 53 126
18 74 50 140 1 93 62 133
62 92 47 100 100 112 43 158
46 62 36 148 110 152 80 170
Appendix 5
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
J. Dan Pittillo
Carya alba 72 90 204 204
157 342
Amelanchier laevis 47 69
West 3 Control, cont.
Pyrularia pubera 44 44
Castanea pumila 115 158 53 238 154 168
Pinus strobus 38 83
Smilax glauca 6 7 6 60
Carya glabra 80 130 47 119
37 90 86 172
155 185
Acer rubrum 34 95 63 94
Robinia pseudoacacia 150 208
Prunus serotina 40 370 57 130
Ilex ambigua 2 135
1 69
Quercus alba 29 84
For each line intrval, First column line cm distance covered (I); second column cm maximum plant diameter (M
West 3 West 3
Experimental No shrubs present Experimental
65 65
28 128
5 41
65 65
210 210
258 350
120 120 120 207
127 127 40 154
81 81 1 55
1 40
For each line intrval, First column line cm distance covered (I); second column cm maximum plant diameter (M
East 3 0-5m 5-10m 10-15m 15-20m 20-25m 25-30m 30-35m 35-40m 40-45m 45-50m
Control Kalmia latifolia 20 400 35 35
27-Aug-04 90 190
Gaylussacia ursina 9 97 36 53 33 84 33 60
47 81 9 35 53 93
30 53 38 43 23 78
18 103 6 60 39 64
64 75 1 35 16 74
23 48 5 14 33 56
48 48 48 75
13 87 35 35
47 57 11 32
41 74
22 73
44 51
Rhododendron maximum 238 238
Acer rubrum 60 118 160 343 82 245 30 134 53 155 92 107 22 77
40 120 22 44 32 142 34 74 49 96 13 37
23 79 68 154 150 280
44 139 210 360
5 47
Castanea pumila
Tsuga canadensis
Prunus serotina
Robinia pseudoacacia
Castanea dentata
Fraxinus americana
62 167
17 90
25 76 18 125
67 100
43 97
48 222
85 85
51 57 26 140 21 95 31 60 12 17
11 42 1 119 1 37 97 126
37 158 9 30 9 60
Quercus rubra 24 110 23 305
Quercus alba 24 107
Corpus florida 50 116
Crataegus monosperma 30 30
Lindera benzoin 180 240
20 66
Amelanchier laevis 70 140
For each line intrval, First column line cm distance covered (I); second column cm maximum plant dia 4 109
East 3 O-Sm 5-SOm SO-SSnn 15-20m 20-25m 25-30m 30-35m 35-40m 40-45m 45-SOnn
Prunus serotina
Quemus rubra
Robinia pseudoacacia
Oxydend ru m arboreu m
Sassafras albidu m
Carya alba
Appendix 5
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
I Dan Pittillo
110 280
250 310
70 340 90 180
70 300
82 110
34 360
Appendix 6
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
Andrews Park 0-Sm5 -10m 10-15n 15:2AmMK21cn 25-30m 30-35m 3 5-40n40-45m45-50m
Control Prunus serotina 24 1 29 1
28-Jul-04 Fraxinus america 4
Vitis aestivalis 2
Amelanchier laev 4
Solidago 20
Potentilla simplex 7
Quercus alba 5
Agrostis 27
Vaccinium stamir 10 28
Acer rubrum 3
Smilax glauca 3
Carya glabra 13
Calycanthus floridus 3
Liriodendron tulipifera 4
Chimaphila maculata 5 1
Smilax rotundifolia 13
Hamamelis virginiana 5
Pinus strobus 20
Lysimachia quadrifolia 3
Tipularia discolor 1
Magnolia fraseri 9
Gaylussacia ursina 62 310 120
Quercus velutina 6
Rhododendron maximum 25
Nyssa sylvatica 14
Values are cm of a plant species covering each 5 m interval
Andrews Park 0-Sm 5 -10m 10-15n 15-20n 20-25n 25-30m 30-35m 3 5-40n40-45m45-50m
Experimental Thelypteris noval 52
28-Jul-04 Liriodendron tulir 2
Danthonia spicat 1 33 58 92 64 88 108 46 23
Dicanthelium dichotomi 11 14 10 4
Amelanchier laevis 2
Aster lateriflorus 1 4 2 4 4 2
Ca rex 3
Houstonia purpurea 3
Potentilla simplex 6 5
Solidago cf. casiea 5
Rubus canadensis 4
Trifoliium repens 2
Viola primulifolia 5
Maianthemum racemosu m 4 20
Hieraceum paniculatum 2
Values are cm of a plant species covering each 5 m interval
West 1 0-Sm 5-10m 10-15n 15-20n 20-25n 25-30m 30-35m 35-40n40-45m45-50m
Control Gaylussacia ursir 46 62 138 110 82 202 139 31 90
28-Jul-04 Sassafras albidur 20
Thelypteris noval 24
Huperzia lucidula 9
Hamamelis virginiana 3
Polygonatum biflorum 1
Clethra acuminata 12
Pyrularia pubera 20 10
Appendix 6
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
J. Dan Pittillo
Values are cm of a plant species covering each 5 m interval
West 1 0-Sm 5-10m 10-15n 15-20n 20-25n 25-30m 30-35m 35-40n40-45m45-5Om
Experimental Dennstaedtia pur 28 18 42
28-Jul-04 Rhododendron ce 20
Viola blanda 20 5 4
Potentilla simple) 5 12 45
Sassafras albidur 7 6 14
Lysimachia quads 6 16
Carex sp. 5 9 12
Aster divaricata 12 3 8 70
Prenanthes altissima 18 3
Stachys lanata 2
Hemerocallis lilioasphoc 68 31 59
Quercus rubra 20
Polystichum acrostichoides 105 10
Ajuga repens 30
Hosta 61 47
Vitis aestivalis 5 20
Viola primulifolia 5 4
Rubus canadensis 17 137
Luzula echinata 7
Acer rubrum 3 17 18
Prunus serotina 6
Iris sp. 2
Rubus flagellaris 12 42
Dicanthelium lacuminatum v. faciculatum 4 17
Digitaria 12
Prunus pensylvanica 26
Betula lenta 7
Viola cucullata 16
Parthenocissus quinquefolia 16
Phytolacca americana 5
Aster cordifolia 7
Luzula echinata 12 2
Osmunda cinnamomea 12
Carex sp. 8 5
Goodyera pubescens 4
Calycanthus floridus 40 14
Quercus rubra 1 1
Hamamelis virginiana 7
Solidago curtisii 18 23
Arisima triphyllum 7 5
Eupatorium purpureum 6
Pyrularia pubera 18
Lysimachia quadrifolia 28
Galium lancifolium 16
Houstonia purpurea 3 3
Potentilla simplex 9 2 1
Polygonatum biflorum 8 11 4
Prunella vulgaris 10 8 61
Carex sp. 2 4 2
Values are cm of a plant species covering each 5 m interval
East 1 0-Sm 5-10m 10-15n 15-20n 20-25n 25-30m 30-35m 35-40n40-45m45-50m
Control Tiarella cordifolia 26 32 7 10
Appendix 6
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 20 04)
Anemone quinquefoliun 13 J. Dan Pittillo 2
Athyrium asplenioides 18 50 25 38
Aster divaricata 13 13 17 9 13
Luzula echinata 12 2
Osmunda cinnamomea 12
Carex sp. 8 5
Goodyera pubescens 4
Calycanthus floridus 40 14
Quercus rubra 1 1
Hamamelis virginiana 7
Solidago curtisii 18 23
Arisima triphyllum 7 5
Eupatorium purpureum 6
Pyrularia pubera 18
Lysimachia quadrifolia 28
Galium lancifolium 16
Houstonia purpurea 3 3
Potentilla simplex 9 2 1
Polygonatum biflorum 8 11 4
Prunella vulgaris 10 8 61
Carex sp. 2 4 2
East 1 0-Sm 5-10m 10-15n 15-20n 20-25n 2 5-30m 30-35m 3 5-40n40-45m45-50m
Experimental Dicanthelium 5 6
7-Aug-04 Luzula echinata 10
Festuca rubra 20 9 392 160 293 470 380 43 20 72
Danthonia spicat 7 12
Lobelia puberula 3
Festuca pratense 83 51 17
Digitaria 15 215 216 43
Pea pratense 46 5
Lespedeza repens 13
Oxalis stricta 3 33 30
Trifolium repens 5 29 10
Dactylis glomerata 12
Taraxacum officinalis 11
Geranium maculatum 4 20
Rudbeckia fulgida 4 3
Viola blanda 33
Angelica triquinata 10 20
Lactuca canadensis 3 2 3
Viola rotundifolia 6
Ageratina altissima 8 5 8
Osmunda cinnamomea 61 37
Viola sororia 5
Prosartes lanuginosum 10
Fraxinus americana 37 5
Carya glabra 22
Aster cordifolia 13
Viola primulifolia 4
Rubus canadensis 6
Hieraceum 28
Diervilla quadrifolia 10 12
Gaylussacia ursina 9 104 89
Appendix 6
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
J. Dan Pittillo
Values are cm of a plant species covering each 5 m interval
West 2 0-Sm 5-10m 10-15n 15-20n 20-25n 2 5-30m 30 -35m 35-40n 40-45m 45-50m
Control Parthenocissus q 61 88 88 90 26 6
9-Aug-04 Arisema triphyllu 12 5 4 9
Viola sororia 71 39 39 95 20 55 38 31 25
Anemine quinquefolia
Rubus canadensi 38
Sassafras albidur 13 31
Fraxinus americana 44 38 16 12 34
Sanicula trifoliate 3 6 13 9 17
Angelica triquina 3
Prunus serotina 5 11 11 9 13 7
Polypodium bifloitr tr
Vitis aestivalis 19
Smilax glauca 4 9
Lindera benzoin 18
Corylus americana 37
Botrychium dissectum 2
Pinus strobus 10
Malus coronaria 4 6
Carex sp. 10 10 39 23
Quercus velutina 17 19 20
Ageratina altissima 9
Crataegus monosperma 23 6
Liparis lilifolia 7 2
Galium circazana 2 7 3
Actaea pachypoda 1
Liriodendron tulipifera 29
Botrychium onedense 3
Goodyera pubescens 7 5 19
Potentilla simplex 9
Smilax rotundifolia 26
Danthonia spicata 2
Carex sp. 2 10
Lysimachia ciliata 16
Clematis virginiana 6
Vaccinium stamineum 11
Huperzia lucidula 12
Values are cm of a plant species covering each 5 m interval
West 2 0-Sm 5-10m 10-15n 15-20n 20-25n 2 5-30m 30 -35m 35-40n 40-45m 45-50m
Experimental Robinia pseudoacacia 18
9-Aug-04 Prunus serotina 6 5 1 5
Parthenocissus quinquefolia 13
Polygonatum biflorum 2 14
Pinus strobus 6 8
Danthonia spicata 13 170 210 37
Houstonia purpurea 29
Potentilla simplex 47 10
Hieraceum pilosella 19
Dicanthelium acuminatum var. fasciculatum 9 73 2
Agrostis cf. hyemalis 21
Solidago rugosa 8
Lysimachia terrestris 6
Appendix 6
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
Sassafras albidum J. Dan Pittillo 2
Arisaema triphyllum 16
Values are cm of a plant species covering each 5 m interval
East 2 0-Sm 5-10m 10-15n 15-20n 20-25n 25-30m 30-35m 35-40n40-45m45-5Om
Control Pinus strobus 2 2 28 15 28 4 22 10 48
10-Aug-04 Vaccinium stamir 38 30 34 20
Chimaphila macu 3 2 6 1 2
Prunus serotina 11 4 12
Danthonia spicata 28 16
Smilax rotundifolia 2
Viburnum cassinoides 9
Quercus coccinea 4 33
Gaultheria procumbens 187 120 213 146 94 42 14 5
Amelanchier laevis 7 28 9 5
Viola hastata 4
Quercus velutina 19 15
Rubus hispidus 9 33 11 23 7 1
Smilax glauca 3
Solidago caseia 19
Potentilla simplex 18 33 3
Platanthera cilliata 9
Acer rubrum 31 9 13 6
Gaylussacia ursina 42 120 58
Cimifuga racemosa 54
Maianthemum racemosum 29
Carya alba 26
Quercus alba 4
Tsuga canadensis 8 2
Quercus rubra 11
Hypopthis unilora 4
Oxydendrum arboreum 23
Kalmia latifolia 8
Rhododendron maximum 5
Lysimachia quadrifolia 13
Carex sp. 10
Values are cm of a plant species covering ea ch 5 m interval
East 2 0-Sm 5-10m 1 0-15n 1 5-20n 20-25n 2 5-30m 30-35m 35-40n40-45m45-5Om
Exprimental Lysimachia ciliate 17 3
10-Aug-04 Thelypteris novaboracensis
Aster umbellatus 7
Danthonia spicat 81 5
Potentilla simple) 15
Lysimachia quads 4
Pycnanthemum r 13
Dennstaedtia pur 270 375
Carya alba 8
Chimaphila macu 4 3 6
Acer rubrum 5
Smilax glauca 6 3 4
Vaccinium stamir 29 3
Vaccinium simulatum 50 5
Pinus strobus 41 4 1 2
Gaylussacia ursina 60 24 3
Hypopthys uniflora 3
Appendix 6
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
Microstegium vimineum J. Dan Pittillo 5
Lonicera japonica 3
Hemerocallis lilioasphodeloides 74 63
Tiarella cordifolia 6
Iris cristata 22 50
East 2, Experimental, cont.
Cordate Iv 40
Liriodendron tulipifera 1
Robinia pseudoacacia 16
Trifoliate Iv 3
Values are cm of a plant species covering each 5 m interval
West 3 0-Sm 5-10m 10-15n 15-20n 20-25n 25-30m 30 -35m 3 5-40n 40-45m 45-50m
Control Dennstadtia pun( 155 87 35 18
16-Aug-04 Crataegus mono! 8 1
Quercus rubra 4 7 24
Sassafras albidur 16 8
Pinus strobus 50 47 17 54
Prunus serotina 10 3
Acer pensylvanicum 6
Chimaphila maculata 5 2
Liriodendron tulipifera 11 8
Acer rubrum 37 18 61 6 1
Solidago caseia 13
Vaccinium stamineum 17
Gaylussacia ursina 9
Lysimachia quadrifolia 5 26
Amelanchier Iaevis 3
Trillium grandiflorum 3
Maianthemum racemosum 3
Medeola virginiana 4
Goodyera pubescens 5
Smilax glauca 3
Diphasiastrum digitarium 6 140
Carya glabra 14
Values are cm of a plant species covering each 5 m interval
West 3 0-Sm 5-10m 10-15n 15-20n 20-25n 25-30m 30 -35m 3 5-40n 40-45m 45-50m
Experimental Potentilla simple) 22 34 57 82 91 80 100 92 3
16-Aug-04 Hypericum gentie 1 2
Oenothera biflon 12
Danthonia spicat 5 19 21 100 175 212 80 44
Dicanthelium 15 42 22 4 9 33 28
Luzula acuminata 6
Houstonia purpurea 3 7 2 5
Hieraceum pilosella 8 27 3 10 10
Viola primulifolia 12 10
Veronica officinalis 3
Aster divaricaa 4
Andropogon virginicus 10 18 35
Lysimachia quadrifolia 2
Lysimachia ciliata 39
Dicanthelium 25 70
Hypericum stragulatum 24
Rumex acetosella 1
Leachea racemulosa 6
Betula Ienta 5 2 3
Acer rubrum 2 4 4
Appendix 6
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
Solidago sp. J. Dan Pittillo 5 3
Values are cm of a plant species covering each 5 m interval
East 3 0-Sm 5-10m 10-15n 15-20n 20-25n 25-30m 30-35m 35-40n 40-45m 45-50m
Control Thelypteris noval 15
27-Aug-04 Dennstadtia pun( 19
Gaulthera precut 26 54 11
Quercus alba 22
Prunus serotina 2 6 28
Smilax glauca 6
Gaylussacia ursina 37 54 110 49
Pinus strobus 34
Rubus hispidus 31 123 58 12 21 5
Fraxinus americana 12 6 14 60
Crataegus monosperma 5
Chimaphila maculata 2
Acer rubrum 3
Polystichum acrostichoides 72
Quercus velutina 5
Goodyera pubescens 18
Potentilla simplex 7 4 4 8
Solidago curtisii 13
Viola sororia? 18
Lindera benzoin 3
Maianthemum racemosum 14
Toxicodendron radicans 3
Corpus florida 20
Lonicera japonica 6
Amelanchier laevis 1
Values are cm of a plant species covering ea ch 5 m interval
East 3 0-Sm 5-10m 1 0-15n 1 5-20n 20-25n 25-30m 30-35m 3 5-40n 40-45m 45-50m
Experimental Digitaria 170 382 203 4
26-Aug-04 Festuca pratense 72 30 21 81
Oxalis stricta 11
Festuca rubra 95 43 150 5
Viola cuculata 2
Trifolium repens 6 7 11
Erigeron philadel 2
Lespedeza hirta 13
Rumex acetosella 10 2
Potentilla simplex 33 81 160 84 45 27
Setaria glauca 41
Prunella vulgaris 6 8 9 57
Smilax rotundifolia 2
Taraxicum officinalis 6
Achillea millefolium 4 47 5
Dicanthelium fuzzy 9 17 18 19 7 50 42 10
Plantago lanceolata 12 2
Acalypha gracilens 3
Leucanthemum 2
Paspalum 14
Lespedeza small 13 5 3
Lysimachia terrestris 15 42
Appendix 6
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
Viola primulifolia J. Dan Pittillo 2
Rubus canadensis 2
Chimaphila maculata 3 4
Lespedeza round 2
Danthonia spicata 110 20 118
Hypericum stragulatum 6 120 31
Rubus hispidus 55 96 60
East 3, experimental, cont.
Houstonia purpurea 3 1
Gaultheria procumbens 9
Calycanthus floridus 5
Solidago caseia 7 48 58
Andropogon virginicus 23 39
Polygala sanguinea 10
Lechea ramulosa 5 4
Lysimachia quadrifolia 6
Viola hastata 2
Epigaea repens 16
Dicanthelium #4 2
Heterotheca mariana 10
Vaccinium pallidum 7
Dicanthelium #5 2
Values are cm of a pla nt species covering each 5 m interval
Appendix 6
Lake Glenville Buffer Zone Study
Raw Data (Collected July 28-August 26, 2004)
J. Dan Pittillo
Control Plant List Experimental Plant List Combined Experimental Control
Acer pensylvan icu m
Acer rubru m
Actaea pachypoda
Ageratina altissima
Agrostis
Amelanchier laevis
Anemine quinquefolia
Anemone quinquefolium
Angelica triquinata
Arisema triphyllum
Aster d ivaricata
Athyrium asplenioides
Betula lenta
Botrychium dissectum
Botrychium onedense
Calycanthus floridus
Carex sp.
Carex sp.
Carex sp. 2
Carya alba
Carya glabra
Castanea dentata
Castanea pumila
Chimaphila maculata
Cimifuga racemosa
Clematis virginiana
Clethra acuminata
Cornus florida
Corylus americana
Crataegus monosperma
Danthonia spicata
Dennstadtia punctilobula
Diphasiastrum digitarium
Eupatorium purpureum
Fagus grandifolia
Fraxinus americana
Galium circazana
Galium lancifolium
Gaultheria procumbens
Gaylussacia ursina
Goodyera pubescens
Halesia tetaptera
Hamamelis virginiana
Houstonia purpurea
Huperzia lucidula
Hypopthis unilora
Ilex ambigua
Kalmia latifolia
Lindera benzoin
Liparis lilifolia
Liriodendron tulipifera
Luzula echinata
Lysimachia ciliata
Lysimachia quadrifolia
Magnolia fraseri
Maianthemum racemosum
Malus coronaria
Medeola virginiana
Nyssa sylvatica
Osmunda cinnamomea
Oxydendrum arboreum
Parthenocissus quinquefolia
Pinus strobus
Platanthera cilliata
Polygonatum biflorum
Polystichum acrostichoides
Potentilla simplex
Prunella vulgaris
Prunus serotina
Acalypha gracilens
Acer rubrum
Achillea millefolium
Ageratina altissima
Agrostis
Ajuga repens
Alnus serrulata
Amelanchier laevis
Andropogon virginicus
Angelica triquinata
Arisaema triphyllum
Aster
Aster cordifolia
Aster divaricata
Aster umbellatus
Athyrium asplenioides
Betula lenta
Calycanthus floridus
Carex
Carex sp.
Carya alba
Carya glabra
Castanea dentata
Castanea pumila
Cerastium fontinalis
Chimaphila maculata
Clethra acumuniata
Cordate Iv
Com us florida
Dactylis glomerata
Danthonia spicata
Dennstaedtia punctilobula
Dicanthelium
Dicanthelium
Dicanthelium dichotomum
Diervilla quadrifolia
Digitaria
Elaeagnus umbellata
Epigaea repens
Erigeron philadelphus
Festuca pratense
Festuca rubra
Forsythia virdissima
Fragaria virginiana
Fraxinus americana
Gaultheria procumbens
Gaylussacia ursina
Geranium maculatum
Hamamelis virginiana
Hemerocallis lilioasphodelus
Heterotheca mariana
Hieraceum pilosella
Hosta
Houstonia purpurea
Hydrangea paniculata
Hypericum gentianoides
Hypericum stragulatum
Hypopthys uniflora
Ilex montana
Iris cristata
Ins sp.
Kalmia latifolia
Lactuca canadensis
Leachea racemulosa
Lespedeza hirta
Lespedeza procumbens
Lespedeza repens
Lespedeza small
Leucanthemu m
Acalypha gracilens Acalypha gracilens
Acer pensylvanicum Acer pensylvanicum
Acer rubru m Acer rubru m
Achillea millefolium Acerrub rum
Actaea pachypoda Achillea millefolium
Ageratina altissima Actaea pachypoda
Ageratina altissima Ageratina altissima
Agrostis Ageratina altissima
Ajuga repens Agrostis
Alnus serrulata Agrostis
Amelanchier laevis Ajuga repens
Andropogon virginicus Alnus serrulata
Anemone quinquefolium Amelanchier laevis
Angelica triquinata Amelanchier laevis
Arisaema triphyllum Andropogon virginicus
Aster Anemone quinquefolia
Aster cordifolia Anemone quinquefolia
Asterdivaricata Angelica triquinata
Aster umbel latus Angelica triquinata
Athyrium asplenioides Arisaema triphyllum
Betula lenta Arisema triphyllum
Botrychium dissectum Aster
Botrychium onedense Aster cord ifolia
Calycanthusfloridus Aster diva ricata
Carex sp. Aster diva ricata
Carex sp. Aster umbel latus
Carex sp. 2 Athyrium asplenioides
Carya alba Athyrium asplenioides
Carya glabra Betula lenta
Castanets dentata Betula lenta
Castanea pumila Botrychium dissectum
Cerastium fontinalis Botrychium onedense
Chimaphila maculata Calycanthus floridus
Cimifuga racemosa Calycanthus floridus
Clematis virginiana Carex
Clethra acuminata Carex sp.
Clethra acumuniata Carex sp.
Cordate Iv Carex sp.
Cornus florida Carex sp. 2
Corylus americana Carya alba
Crataegus monosperma Carya alba
Dactylis glomerata Carya glabra
Danthonia spicata Carya glabra
Dennstaedtia punctilobula Castanea dentata
Dicanthelium Castanea dentata
Dicanthelium Castanea pumila
Dicanthelium dichotomum Castanea pumila
Diervilla quadrifolia Cerastium fontinalis
Digitaria Chimaphila maculate
Diphasiastrum digitarium Chimaphila maculata
Elaeagnus umbellata Cimifuga racemosa
Epigaea repens Clematis virginiana
Erigeron philadelphus Clethra acuminate
Eupatorium purpureum Clethra acumuniata
Fagus grandifolia Cordate Iv
Festuca pratense Cornus florida
Festuca rubra Cornus florida
Forsythia virdissima Corylus americana
Fragaria virginiana Crataegus monosperma
Fraxinus americana Dactylis glomerate
Galium circazana Danthonia spicata
Galium lancifolium Danthonia spicata
Gaultheria procumbens Dennstadtia punctilobula
Gaylussacia ursina Dennstaedtia punctilobula
Geranium maculatum Dicanthelium
Goodyera pubescens Dicanthelium
Halesia tetaptera Dicanthelium dichotomum
Hamamelis virginiana Diervilla quadrifolia
Hemerocallis lilioasphodelus Diaitara
Control Plant List Experimental Plant List Combined Experimental Control
Rhododendron maximum
Robinia pseudoacacia
Rubus canadensis
Rubus hispidus
Sanicula trifoliate
Sassafras albidum
Smilax glauca
Smilax rotundifolia
Solidago caseia
Solidago curtisii
Thelypteris novaboracensis
Tiarella cordifolia
Tipularia discolor
Trillium grandiflorum
Tsuga canadensis
Vaccinium corymbosum
Vaccinium stamineum
Veratrum parviflorum
Viburnum cassinoides
Viola hastata
Viola sororia
Vitis aestivalis
Lysimachia ciliata
Lysimachia quadrifolia
Lysimachia terrestris
Magnolia fraseri
Maianthemum racemosum
Medeola virginiana
Microstegium vimineum
Nyssa sylvatica
Oenothera biflorum
Osmunda cinnamomea
Oxalis stricta
Oxydendru m arboreu m
Parthenocissus quinquefolia
Paspalu m
Phytolacca americana
Pinus rigida
Pinus strobus
Plantago lanceolata
Poa pratense
Polygala sanguinea
Polygonatum biflorum
Polystichum acrostichoides
Potentilla simplex
Prenanthes altissima
Prosartes lanuginosum
Prunella vulgaris
Prunus pensylvanica
Prunus serotina
Pycnanthemum montanum
Pyrularia pubera
Quercus alba
Quercus coccinea
Quercus rubra
Quercus velutina
Rhododendron calendulaceum
Rhododendron hybrid
Rhododendron maximum
Robinla pseudoacacia
Rubus canadensis
Rubus flagellaris
Rubus hispidus
Rudbeckia fulgida
Rumex acetosella
Sassafras albidum
Smilax glauca
Smilax rotundifolia
Solidago curtisii
Solidago rugosa
Solidago sp.
Stachys lanata
Taraxacum officinalis
Tielypteris novaboracensis
Tiarella cordifolia
Trifoliate Iv
Trifolium repens
Tsuga canadensis
Vaccinium corymbosum
Vaccinium pallidum
Vaccinium simulatum
Vaccinium stamineum
Veronica officinalis
Viola
Viola blanda
Viola cucullata
Viola hastata
Viola primulifolia
Viola rotundifolia
Viola sororia
Vitis aestivalis
Hypericum gentianoides
Hypericum stragulatum
Hypopthys uniflora
Ilex ambigua
Ilex montana
Iris cristata
Iris sp.
Kalmia latifolia
Lactuca canadensis
Leachea racemulosa
Lespedeza hirta
Lespedeza procumbens
Lespedeza repens
Lespedeza small
Leucanthemu m
Leucothoe fontanesiana
Lindera benzoin
Liparis lilifolia
Liriodendron tulipifera
Lobelia puberula
Lon icera japonica
Luzula acuminata
Luzula echinata
Lysimachia ciliata
Lysimachia quadrifolia
Lysimachia terrestris
Magnolia fraseri
Maianthemum racemosum
Malus coronaria
Medeola virginiana
Medeola virginiana
Microstegium vimineum
Nyssa sylvatica
Oenothera biflorum
Osmunda cinnamomea
Oxalis stricta
Oxydendrum arboreum
Parthenocissus quinquefolia
Paspalum
Phytolacca americana
Pinus rigida
Pinus strobus
Plantago lanceolata
Platanthera cilliata
Poa pratense
Polygala sanguinea
Polygonatum biflorum
Polystichum acrostichoides
Potentilla simplex
Prenanthes altissima
Prosartes lanuginosum
Prunella vulgaris
Prunus pensylvanica
Prunus serotina
Pycnanthemum montanum
Pyrularia pubera
Quercus alba
Quercus coccinea
Quercus rubra
Quercus velutina
Festuca pratense
Festuca rubra
Forsythia virdissima
Fragaria virginiana
Fraxinus americana
Fraxinus americana
Galium circazana
Gahum lancifolium
Gaultheria procumbens
Gaultheria procumbens
Gaylussacia ursina
Gaylussacia ursina
Geranium maculatum
Goodyera pubescens
Halesia tetaptera
Hamamelis virginiana
Hamamelis virginiana
Hemerocallis lilioasphodelus
Heterotheca mariana
Hieraceum pilosella
Hosta
Houstonia purpurea
Houstonia purpurea
Huperzia lucidula
Hydrangea paniculata
Hypericum gentianoides
Hypericum stragulatum
Hypopthis unilora
Hypopthys uniflora
Ilex ambigua
Ilex montana
Iris cristata
Iris sp.
Kalmia latifolia
Kalmia latifolia
Lactuca canadensis
Leachea racemulosa
Lespedeza hirta
Lespedeza procumbens
Lespedeza repens
Lespedeza small
Leucanthemu m
Leucothoe fontanesiana
Lindera benzoin
Liparis lilifolia
Liriodendron tulipifera
Liriodendron tulipifera
Lobelia puberula
Lonicera japonica
Luzula acuminate
Luzula echinata
Luzula echinata
Lysimachia ciliata
Lysimachia ciliata
Lysimachia quadrifolia
Lysimachia quadrifolia
Lysimachia terrestris
Magnolia fraseri
Magnolia fraseri
Maianthemum racemosum
Rhododendron calendulaceui Maianthemum racemosum
Rhododendron hybrid Malus coronaria
Rhododendron maximum Medeola virginiana
Robinia pseudoacacia Medeola virginiana
Rubus canadensis Microstegium vimineum
Rubus flagellaris Nyssa sylvatica
Rubus hispidus Nyssa sylvatica
Rudbeckia fulgida Oenothera biflorum
Rumex acetosella Osmunda cinnamomea
Control Plant List Experimental Plant List Combined Experimental Control
Control total: 97
Solidago rugosa
Solidago sp.
Stachys lanata
Taraxacum officinalis
Thelypteris novaboracensis
Tiarella cordifolia
Tipularia discolor
Trifoliate Iv
Trfolium repens
Trillium grandiflorum
Tsuga canadensis
Vaccinium corymbosum
Vaccinium pallidum
Vaccinium simulatum
Vaccinium stamineum
Veratrum parviflorum
Veronica officinalis
Viburnum cassinoides
Viola
Viola blanda
Viola cucullata
Viola hastata
Viola primulifolia
Viola rotundifolia
Viola sororia
Vitis aestivalis
Experimental total: 143 All plot total: 170
Paspalum
Phytolacca americana
Pinus rigida
Pinus strobus
Pinus strobus
Plantago lanceolata
Platanthera cilliata
Poa pratense
Polygala sanguinea
Polygon atum biflorum
Polygonatum biflon rn
Polystichum acrostichoides
Polystichum acrostichoides
Potentilla simplex
Potentilla simplex
Prenanthes altissima
Prosartes lanuginosum
Prunella vulgaris
Prunella vulgaris
Prunus pensylvanica
Prunus serotina
Prunus serotina
Pycnanthemum montanum
Pyrularia pubera
Pyrularia pubera
Quercus alba
Quercus alba
Quercus coccinea
Quercus coccinea
Quercus rubra
Quercus rubra
Quercus velutina
Quercus velutina
Rhododendron calendulaceum
Rhododendron calendulaceum
Rhododendron hybrid
Rhododendron maximum
Rhododendron maximum
Robinia pseudoacacia
Robinia pseudoacacia
Rubus canadensis
Rubus canadensis
Rubus flagellaris
Rubus hispidus
Rubus hispidus
Rudbeckia fulgida
Ru mex acetosella
Sanicula trifoliate
Sassafras albidu m
Sassafras albidum
Smilax glauca
Smilax glauca
Smilax rotundifolia
Smilax rotundifolia
Solidago caseia
Solidago curtisii
Solidago curtisii
Solidago rugosa
Solidago sp.
Stachys lanata
Taraxacum officinalis
Thelypteris novaboracensis
Thelypteris novaboracensis
Tiarella cordifolia
Tiarella cordifolia
Tipularia discolor
Trifoliate Iv
Trfolium repens
Trillium arandiflorum
Control Plant List Experimental Plant List Combined Experimental Control
Species unique to control transects»»»»»»>
Species unique to experimental transects»»»»>
Species common to both control & experimental transe,
Additional releve species in control sites»»»»»
Additional releve species in experimental sites»»»
Total species observed during study»»»»»»:
Vaccinium stamineum
Vaccinium stamineum
Veratrum parviflorum
Veronica officinalis
Viburnum cassinoides
Viola
Viola blanda
Viola cucullata
Viola hastata
Viola hastata
Viola primulifolia
Viola rotundifolia
Viola sororia
Viola sororia
Vitis aestivalis
25 Vitis aestivalis
78 Unique to control sites>
72
107
149
431
25 78
'unique t
4
Andrews Park Control Andrews Park Experimental
Alnus serrulata Aster casiea
Aster divaricat, Aster lateriflorus
Aster undulatus Carya alba
Athyrium asplemoides Castanea dentat,
Betula alleghaniensis Dactylis glomerat,
Carex sp. Eupatorium purpureum
Castanea dentat, Juniperus virginiana
Corpus alternifolia Lysimachia quadrifolia
Corpus amomum Magnolia fraseri
Hieraceum paniculatum Parthenocissus virginiana
Ilex opaca Polygonatum biflorum
Ilex verticillata Prunella vulgaris
Impatiens capensis Solidago rugosa
Luzula echinat, Taraxacum officinalis
Lyonia ligustrina Toxicodendron radicans
Malus sylvestris Trillium grandiflorum
Medeola virginiana Jvularia sessilifolia
Monotropa uniflora
Osmunda cinnamomea West 1 Experimental
Oxypolis rigidior Abelia grandiflora
Parthenocissus quinquefolia Aesculus Pavia
Polystichum acrostichoides Agentina altissima
Prunella vulgaris Apios americana
Pyrularia pubera Arsema triphyllum
Rhododendron calendulaceum Bidens frondosa
Rubus flagellaris Carex laxiflorum
Solidago curtisii Carya glabra
Spiranthes tomentosa Cassia marilandica
Thelyptens novaboracensis Chimaphila maculat,
Uvularia sessilifolia Danthonia spicata
Viburnum cassinoides Dioscorea quaternat,
Viola cucullana Diphasiastrum digitatum
Viola primulifolia Erechtites hieracifolium
Viola pubescens Eupatorium compositifolium
Vitis labrusca Festuca rubra
Xanthorhiza simplissima Gnaphalium sp.
Hamamelis virginiana
West 1 Control Hedeoma pulgeoides
Calycanthus flondus Ilex glabra
Castanea dentat, Liatris spicat,
Hexastylis shuttleworthii Linodendron tulipifera
Lysimachia quadrifolia Lysimachia ciliat,
Magnolia fraseri
East 1 Control Maianthemum racemosum
Carya glabra Oxalis strict,
Chimaphila maculat, Polygonatum biflorum
Conopholis americana Pulmonaria'Roy Davidson'
Magnolia acuminat, Pyrulara pubera
Osmunda clayonii Rhododendron hybrids
Tilia americana heterophylla Sedum sp.
Trillium erectu m Smilax glauca
Uvularia sessilifolia Smilax rotundifolia
Viola rotundifolia Thelypteris novaboracensis
West 2 Control East 1 Experimental
Angelica purpurea Achalypa rhomboides
Arisema trphyllum quinatum Actaea pachypoda
Aster cordat, Apios americana
Aster d ivaricat, Ansaema triph yllu m q uinatu m
Botrychium virginianum Corpus amomum
Carya glabra Dicanthelium bossii
Chimaphila maculat, Dioscorea quaternat,
Corpus alternifolia Halesis tetaptera
Euphorbia collorat, Helianthus giganteus
Halesia tetaptera Impatiens capensis
Houstonia purpurea Leucothoe fontanesiana
Hypericum puntatum Linodendron tulipifera
Lactuca canadensis Malus anoustifolia
Rosa multiflora Veronica persica
Smilax herbacea
Solidago caseia West 2 Experimental
Solidago rugosa Amelanchier laevis
Tovaria virginiana Botrychium dissectum
Tsuga canadensis Carex venuosa
Carya glabra
East 2 Control Chimaphila maculat,
Acer pensylvanicum Crataegus monosperma
Athyrium asplenioides Dioscorea quaternat,
Betula lent, Diphasiastru m d igitariu m
Castanea pumila Goodyera pubescens
Cypripedium acaule Hemerocallis lilioasphodeloides
Galium triflorum Hypencum punctatum
Linodendron tulipifera Hyperzia lucidula
Luzula echinat, Kalmia latifolia
Polygonatum biflorum Lespedeza procumbens
Polystichum acrostichoides Lobelia inflate
Pteridium aquilinum latusculum Maianthemum racemosum
Pycnanthemum montanum Nyssa sylvatica
Quercus montane Oenothera biflorum
Solidago rugosa Parthenocissus quinquefolia
Toxicodendron radicans Perovskia x filigram
Trillium grandifloru m Polystichu m acrostichoides
Viola sororia Prenanthes altissima
Quecus velutina
West 3 Control Rhododendron hybrid
Ageratina altissima Rudbeckia fulgida
Aster divaricat, Smilax glauca
Corpus florida Solidago curtisii
Fraxinus americana Tpularia discolor
Hieraceum paniculatum Tsuga canadensis
Hydrangea arborea Vaccinium corymbosum
Magnolia frasen Vitis aestivalis
Nyssa sylvatica Vitis labrusca
Oxydendrum arboreum
Poe cf. Cuspidate East 2 Experimental
Polygonatum biflorum Anemone wood
Polystichum acrostichoides Athyrium painted fern
Prunella pensylvanica Clethra acuminate
Quercus velutina Corpus amomum
Rhododendron maximum Dicanthelium sp.
Tpularia discolor Goodyera pubescens
Viola rotundifolia Hosta spp.
Vitis aestivalis Hydrangea paniculat,
Impatiens sultana
East 3 Iris hybrid
Acer saccharu m Iris pseudocorus
Ageratina altissima Lamiaceae-thyme
Angelica triquinat, Mainthemum japonicum
Athyrium asplenioides Nyssa sylvatica
Betula lent, Osmunda cinnamomea
Botrychium dissectum Osmunda claytomi
Carya glabra Platanthera ciliate
Danthonia spicat, Rhododendron maximum
Impatiens capensis Rhododendron vaseyi
Microstegium vimineum Spiranthes japonicum
Monotropa uniflora Toxicodendron radicans
Oxalis strata
Parthenocissus quinquefolia
Platanthera cilliaris West 3 Experimental
Polygonatum biflorum Athyrium asplenioides
Polygonum cespitosum Carex sp. 1
Pycnanthemum montanum Carex sp. 2
Quercus rubra Carya glabra
Rosa multiflora Chimaphila maculate
Rubus canadensis Corpus florida
Sambucus canadensis Dioscorea quaternat,
Sassafras albidum Eunhorbia collorata
Quemus coccinea
Quemus velutina
Control Releve Species (all si Striped maple Rubus canadensis
Acer pensylvanicum White snakeroot Sassafras albidum
Ageratina altissima Smooth alder Smilax glauca
Alnus sermlata Angelica; Wild celery Thelypteris novaboracensis
Angelica purpurea Southern jack-in-the-pulpit Trillium grandiflorum
Arisema triphyllum quinatum Cordate-leaf aster Vaccinium pallidum
Aster cordat, Heart-leaf aster Veratrum parviflorum
Asterdivaricata Southern lady fern Verbascum thaspus
Athyrium asplemoides Yellow birch
Betula alleghaniensis Sweet birch
Betula lent, Rattlesnake fern East 3 Experimental
Botrychium virginianum Sweet shrub Acer rubrum
Calycanthus flondus Sedge Ambrosia artemisiifolia
Carex sp. Pignut hickory Aster undulatus
Carya glabra American chestnut Betula lent,
Castanea dentat, Chinquapin Botrychium dissectum
Castanea pumila Spotted wintergreen Cassia nictitans
Chimaphila maculat, Bear corn Castanea pumila
Conopholis americana Alternate-leaved dogwood Corpus amomum
Corpus alternifolia Branch dogwood Corpus florida
Corpus amomum Flowering dogwood Crataegus monosperma
Corpus florida Pink ladyslipper Dactylis glomerat,
Cypripedium acaule Poverty oat grass Caucus carob,
Danthonia spicat, Flowering spurge Dicanthelium boscii
Euphorbia collorat, White ash Eupatorium albidum
Fraxinus americana Sweet bedstraw Fraxinus americana
Galium triflorum Carolina silverbell Galium cimazans
Halesia tetaptera Large-flower heartleaf Galium tinctorium
Hexastylis shuttleworthii Rattlesnake weed Gaylullacia ursina
Hieraceum paniculatum Purple bluet Goodyera pubescens
Houstonia purpurea Wild hydrangea Helenium flexuosum
Hydrangea arborea St. John's wort Hypencum stragalum
Hypericum puntatum American holly Ilex opaca
Ilex opaca Winterberry Impatiens capensis
Ilex verticillata Jewelweed Juglans nigra
Impatiens capensis Wild lettuce Kalmia latifolia
Lactuca canadensis Tulip tree; yellow poplar Lespideza upright
Linodendron tulipifera Indian tobacco Lobelia inflat,
Lobelia inflat, Wood rush Oenothera biennis
Luzula echinat, Maleberry Oxydendmm arboreum
Lyonia ligustrina Whorled loosestrife Pinus strobus
Lysimachia quadrifolia Cucumber tree Platanthera ciliat,
Magnolia acuminat, Fraser magnolia Polygonum cespitosum
Magnolia frasen False solomon's seal; plumefl o' Prenanthes altissima
Maianthemum racemosum Common apple Quemus velutina
Malus sylvestris Indian cucumberroot Rhododendron maximum
Medeola virginiana Japanese grass Rudbeckia hirt,
Microstegium vimineum Indian pipe Salix babylonica
Monotropa uniflora Blackgum Salis sericea
Nyssa sylvatica Cinnamon fern Salvia lyrata
Osmunda cinnamomea Interrupted fern Smilax glauca
Osmunda clayonii Wood sorrell Solidago rugosa
Oxalis strata Sourwood Tsuga canadensis
Oxydendrum arboreum Cowbane Vaccinium stamineum
Oxypolis rigidior Golden ragwort Vernonia novaboracensis
Packera aurea Virginia creeper Veronica officinalis
Parthenocissus quinquefolia Yellow fringed orchid
Platanthera cilliaris Woodland bluegrass
Poa cf. Cuspidat, Mayapple
Podophyllum peltatum Solomon's seal
Polygonatum biflomm Smartweed
Polygonum cespitosum Christmas fern
Polystichum acrostichoides Selfheal
Pmnella vulgaris Pin cherry; fire cherry
Prunus pensylvanica Bracken
Pteridium aquilinum latusculu m Mountain mint
Pvcnanthemum montanum Oil nut: buffalo nut
Experimental Releve Species (all sites)
Abelia grandiflora
Acer rubrum
Achalypa rhomboides
Actaea pachypoda
Aesculus Pavia
Agentina altissima
Ambrosia artemisiifolia
Amelanchier laevis
Glossy abelia
Red maple
Three-sided mecury
Little doll's eyes
Red buckeye
White snakeroot
Common ragweed
Alleahanv serviceberrv
Rosa multiflora Blackberry Athyrium asplenioides Southern lady fern
Rubus canadensis Dewberry Athyrium painted fern Painted fern
Rubus flagellaris Elderberry Betula lent, Sweet birch
Sambucus canadensis Sassafras Bidens frondosa Beggar ticks
Sassafras albidum Carrion flower Botrychium dissectum Grapefem
Smilax herbacea Greenbrier Carex laxiflorum Sedge
Smilax rotundifolia Bluestem goldenrod Carex sp. 1 Sedge
Solidago caseia Curtis' goldenrod Carex sp. 2 Sedge
Solidago curtisii Rough-leaf goldenrod Carex venuosa Sedge
Solidago rugosa Hardtack Carya alba Mockernut hickory
Spiranthes tomentosa New York fern Carya glabra Pignut hickory
Thelypteris novaboracensis White basswood Cassia marilandica Wild senna
The americana heterophylla Cranefly orchid Cassia nictitans Sensitive plant
Tpularia discolor Jumpseed Castanea dentat, American chestnut
Tovaria virginiana Poison ivy Castanea pumila Chinquapin
Toxicodendron radicans Trillium Chimaphila maculat, Spotted wintergreen
Trillium erectum Large-flowered trillium Clethra acuminat, Sweet pepperbush
Trillium grandiflorum Canada or eastern hemlock Corpus amomum Branch dogwood
Tsuga canadensis Merrybells Corpus flonda Flowering dogwood
Uvularia sessilifolia Highbush blueberry Crataegus monosperma Hawthorn
Vaccinium simulatum Deerberry Dactylis glomerat, Orchard grass
Vaccinium stamineum Wild raised Danthonia spicata Mountain oat grass
Viburnum cassinoides Swamp violet Daucus carot, Queen Anne's lace
Viola cucullaria Primrose-leaved violet Dicanthelium bossii Panic grass
Viola primulifolia Yellow violet Dicanthelium sp. Panic grass
Viola pubescens Round-leaf violet Dioscorea quaternat, Wild yam
Viola rotundifolia Common blue violet Diphasiastrum digitatum Running cedar
Viola sororia Summer grape Erechtites hieracifolium Fireweed
Vitis aestivalis Summer grape Eupatorium albidum White bracketed thoroughwc
Vitis aestivalis Fox grape Eupatorium compositifolium Dog fennel
Vitis labrusca Yellowroot Eupatorium purpureum Purple joe-pye weed
Xanthorhiza simplissima Euphorbia collorat, Flowering spurge
Festuca rubra Creeping red fescue
Fraxinus americana White ash
Galium cimazans Forest bedstraw
Galium tinctorium Three-lobed bedstraw
Gaylussacia ursina Buckberry
Gaylussacia baccat, Huckleberry
Gnaphalium sp. Cudweed
Goodyera pubescens Rattlesnake orchid
Halesia tetaptera Carolina silverbell
Hamamelis virginiana Witchhazel
Hedeoma pulgeoides Pennyroyal
Helenium flexuosum Bitterweed
Helianthus giganteus Giant sunflower
Hemerocallis lilioasphodeloides Yellow daylily
Hieraceum paniculatum Hawkweed
Host, spp. Host,
Hydrangea paniculat, Peeges hydrangea
Hypericum punctatum St. John's wort
Hypericum stragalum St. Peter's wort
Hyperzia lucidula Shining clubmoss
Ilex opaca American holly
Ilex glabra Inkberry
Impatiens capensis Jewelweed
Impatiens sultana Sultana
Iris hybrid Iris
Iris pseudocorus Yellow flag
Juglans nigra Black walnut
Juniperus virginiana Red cedar
Kalmia latifolia Mountain laurel
Lespedeza procumbens Lespedeza
Lespideza upright Lespedeza
Leucothoe fontanesiana Dog hobble
Liatris spicat, Blazing star
Linodendron tulipifera Tulip tree; yellow poplar
Lobelia inflate Indian tobacco
Lobelia puberula Lobelia
Lvsimachia ciliate Frinaed loosestrife
Oenothera biennis
Osmunda cinnamomea
Osmunda claytonii
Oxalis strict,
Oxydend ru m arboreu m
Parthenocissus quinquefolia
Perovskia x filigram
Pinus strobus
Platanthera ciliat,
Platanthera ciliat,
Polygonatum biflorum
Polygonum cespitosum
Polygonum pensylvanicum
Polystichum acrostichoides
Prenanthes altissima
Prunella vulgaris
Pulmonaria'Roy Davidson'
Pyrularia pubera
Quemus coccinea
Quemus velutina
Rhododendron hybrids
Rhododendron maximum
Rhododendron obtusum
Rhododendron vaseyi
Robinia pseudoacacia
Rosa cultivated
Rubus canadensis
Rudbeckia fulgida
Rudbeckia hirta
Sells sericea
Salix babylonica
Salvia lyrata
Sassafras albidum
Sedum sp.
Smilax glauca
Smilax rotundifolia
Solidago caesia
Solidago curtisii
Solidago rugosa
Spiranthes japonicum
Taraxacum off icinalis
Thelypteris novaboracensis
Thymus serpyllum
Tpularia discolor
Toxicodendron radicans
Trillium grandiflorum
Tsuga canadensis
Uvularia sessilifolia
Vaccinium corymbosum
Vaccinium pallidum
Vaccinium stamineum
Veratrum parviflorum
Verbascum thaspus
Vernonia novaboracensis
Veronica off icinalis
Veronica persica
Vitis aestivalis
Vitis labrusca
Evening primrose
Cinnamon fern
Interrupted fern
Wood sorrel
Sou mood
Virginia creeper
White pine
Fringed yellow orchid
Yellow fringed orchid
Solomon's seal
Smartweed
Smartweed
Christmas fern
Rattlesnake weed
Self heal
Polmonaria
Oilnut; buffalo nut
Scarlet oak
Black oak
Rhododendron
Rosebay rhododendron
Azalea
Pink shell azalea
Black locust
Rose
Blackberry
Blackeyed susan
Blackeyed susan
Silky willow
Weeping willow
Lyre-leafed sage
Sassafras
Sedum
Catbrier
Greenbrier
Blue-stem goldenrod
Curtis' goldenrod
Rough leaved goldenrod
Japanese spirea
Common dandelion
New York fern
Thyme
Cranefly orchid
Poison ivy
Large flowered trillium
Canada or eastern hemlock
Merrybells
Highbush blueberry
Lowbush blueberry
Deerberry
False helebore
Common mullen
Ironweed
Speedwell
Blue eyes
Su mmer grape
Fox grape
Appendix 1: Species Lists & Analysis Codes
HERB SPECIES SHRUB SPECIES TREE SPECIES
Acalypha gracilens ACGR Acer rubrum ACRU Acer rubrum ACRU
Acer pensylvanicum ACPE Alnus serrulata ALSE Aesculus flava AEFL
Acer rubrum ACRU Amelanchier laevis AMLA Amelanchier laevis AMLA
Achillea millefolium ACMI Betula lenta BELE Betula lenta BELE
Actaea pachypoda ACPA Calycanthus floridus CAFL Carya alba CAAL
Ageratina altissima AGAL Carya alba CAAL Carya glabra CALL
Agrostis spp. AGSP Carya glabra CALL Corpus florida COFL
Ajuga repens AJRE Castanea dentata CADE Hamamelis virginiana HAVI
Amelanchier laevis ARLA Castanea pumila CAPU Liriodendron tulipifera LITU
Andropogon
virginicus ANVI Clethra acumuniata CLAC Magnolia fraseri MAFR
Anemine quinquefolia ANQU Cornus florida COFL Nyssa sylvatica NYSY
Crataegus
Angelica triquinata ANTR monosperma CRMO Oxydendrum arboreum OXAR
Parthenocissus
Arisema triphyllum ARTR Fagus grandifolia FAGR quinquefolia PAQU
Aster spp. ASSP Forsythia virdissima FOVI Pinus rigida PIRI
Aster cordifolia ASCO Fraxinus americana FRAM Pinus strobus PIST
Aster divaricata ASDI Gaylussacia ursina GAUR Prunus serotina PRSE
Aster umbellatus ASUM Halesia tetaptera HATE Quercus alba QUAL
Betula lenta BELE Hamamelis virginiana HAVI Quercus coccinea QUCO
Botrychium
dissectum BODI Hydrangea paniculata HYPA Quercus rubra QURU
Botrychium onedense BOON Ilex ambigua ILAM Quercus velutina QUVE
Calycanthus floridus CAFL Ilex montana ILMO Robinia pseudoacacia ROPS
Carex spp.1 CSPP Kalmia latifolia KALA Sassafras albidum SAAL
Leucothoe
Carex spp. 2 CASP fontanesiana LEFO Tsuga canadensis TSVI
Carex spp. 3 CARP Lindera benzoin LIBE Vitis aestivalis
Carya alba CAAL Liriodendron tulipifera LITU
Carya glabra CALL Magnolia fraseri MAFR
Cerastium fontinalis CEFO Nyssa sylvatica NYSY
Chimaphila maculata CHMA Oxydendrum arboreum OXIDE
Cimifuga racemosa CIRA Pinus strobus PIST
Clematis virginiana CLVI Prunus serotina PRSE
Clethra acuminata CLAC Pyrularia pubera PYPU
Cordate Iv COLV Quercus alba QUAL
Cornus florida COFL Quercus coccinea QUCO
Corylus americana COAM Quercus rubra QURU
Crataegus
monosperma CRMO Quercus velutina QUVE
Rhodendron
Dactylis glomerata DAGL calendulaceum RHCA
Danthonia spicata DASP Rhododendron hybrid RHHY
Dennstadtia Rhododendron
punctilobula DEPU maximum RHMA
Dicanthelium spp. DICS Robinia pseudoacacia ROPS
Dicanthelium DICA Rubus canadensis RUCA
1
dichotomum
Diervilla quadrifolia DIQU Sassafras albidum SAAL
Digitaria spp. DISP Smilax glauca SMGL
Diphasiastrum
digitarium DIDI Smilax rotundifolia SMRO
Elaeagnus umbellata ELUM Tsuga canadensis TSCA
Epigaea repens EPRE Vaccinium corymbosum VACO
Erigeron philadelphus ERPH Vaccinium pallidum VAPA
Eupatorium
purpureum EUPU Vaccinium simulatum VASI
Festuca pratense FEPR Vaccinium stamineum VAST
Festuca rubra FERU
Fragaria virginiana FRVI
Fraxinus americana FRAM
Galium circazana GACI
Galium lancifolium GALA
Gaultheria
procumbens GAPR
Gaylussacia ursina GAUR
Geranium maculatum GEMA
Goodyera pubescens GOPU
Hamamelis virginiana HAVI
Hemerocallis
lilioasphodelus HELI
Heterotheca mariana HEMA
Hieraceum spp. HISP
Hieraceum pilosella HIPI
Hosta spp. HOSP
Houstonia purpurea HOPU
Huperzia lucidula HULU
Hypericum
gentianoides HYGE
Hypericum
stragulatum HYST
Hypopthis unilora HYUN
Iris cristata IRCR
Iris sp. IRSP
Kalmia latifolia KALA
Lactuca canadensis LACA
Leachea racemulosa LERA
Lespedeza hirta LEHI
Lespedeza
procumbens LEPR
Lespedeza repens LERE
Lespedeza small LESM
Leucanthemum spp. LESP
Lindera benzoin LIBE
Liparis lilifolia LILI
Liriodendron tulipifera LITU
Lobelia puberula LOPU
Lonicera japonica LOJA
Luzula acuminata LUAC
2
Luzula echinata LUEC
Lysimachia ciliata LYCI
Lysimachia
quadrifolia LYQU
Lysimachia terrestris LYTE
Magnolia fraseri MAFR
Maianthemum
racemosum MARA
Malus coronaria MACO
Medeola virginiana MEVI
Microstegium
vimineum MIVI
Nyssa sylvatica NYSY
Oenothera biflorum OEBI
Osmunda
cinnamomea OSCI
Oxalis stricta OXST
Oxydendrum
arboreum OXAR
Parthenocissus
quinquefolia PAQU
Paspalum spp. PASP
Phytolacca
americana PHAM
Pinus strobus PIST
Plantago Ianceolata PLLA
Platanthera cilliata PLCI
Poa pratense POPR
Polygala sanguinea POSA
Polygonatum biflorum POBI
Polystichum
acrostichoides POAC
Potentilla simplex POSI
Prenanthes altissima PRAL
Prosartes
Ianuginosum PR LA
Prunella vulgaris PRVU
Prunus pensylvanica PRPE
Prunus serotina PRSE
Pycnanthemum
montanum PYMO
Pyrularia pubera PYPU
Quercus alba QUAL
Quercus coccinea QUCO
Quercus rubra QURU
Quercus velutina QUVE
Rhododendron
calendulaceum RHCA
Rhododendron
maximum RHMA
Robinia
pseudoacacia ROPS
Rubus canadensis RUCA
3
Rubus flagellaris RUFL
Rubus hispidus RUHI
Rudbeckia fulgida RUFU
Rumex acetosella RUAC
Sanicula trifoliata SATR
Sassafras albidum SAAL
Setaria glauca SEGL
Smilax rotundifolia SMRO
Solidago spp SOSP
Solidago caseia SOCA
Solidago curtisii SOCU
Solidago rugosa SORU
Solidago sp. SOSP
Stachys Ianata STLA
Taraxacum officinalis TAOF
Thelypteris
novaboracensis THNO
Tiarella cordifolia TICO
Tipularia discolor TO
Toxicodendron
radicans TORA
Trifoliate Iv TRLV
Trifolium repens TRRE
Trillium grandiflorum TRGR
Tsuga canadensis TSCA
Vaccinium pallidum VAPA
Vaccinium simulatum VASI
Vaccinium
stamineum VAST
Veratrum parviflorum VEPA
Veronica officinalis VEOF
Viburnum
cassinoides VICA
Viola spp. VISP
Viola blanda VIBL
Viola cucullata VICU
Viola hastata VIHA
Viola primulifolia VIPR
Viola rotundifolia VIRO
Viola sororia VISO
Vitis aestivalis VIAE
4
PLOTS CODE LABEL
Andrews Park Control APC pl
Andrews Park
Experimental APE p2
East 1 Control E1C p3
East 1 Experimental E1E p4
East 2 Control E2C p5
East 2 Experimental E2E p6
East 3 Control E3C p7
East 3 Experimental E3E p8
West 1 Control W1C P9
West1 Experimental W1E p10
West 2 Control W2C pl 1
West 2 Experimental W2E p12
West 3 Control W3C p13
West 3 Experimental W3E p14
This is a list of all sample plots with the
corresponding code and labels for
reference in analysis.
5
Appendix 2: Indicator Speices Analysis Results
TREE ISA RESULTS:
Group 1 = Control
Group 2 = Experimental
RELATIVE ABUNDANCE in group, % of perfect indication (average abundance of a
given species in a given group of plots over the average abundance of that species in all
plots expressed as a %)
Group
Sequence: 1 2
Identifier: 3 4
# of terns: 2 4
Column Avg Max MaxGrp
1 ACRD 50 53 3 53 47
2 AEFL 50 100 3 100 0
3 AMLA 50 100 3 100 0
4 BELE 50 100 3 100 0
5 CAAL 50 100 4 0 100
6 CAGL 50 100 3 100 0
7 COFL 50 100 4 0 100
8 HAVI 50 100 4 0 100
9 LITU 50 83 4 17 83
10 MAFR 50 100 4 0 100
12 OXAR 50 100 3 100 0
13 PAQU 50 100 4 0 100
15 PIST 50 71 4 29 71
16 PRSE 50 100 4 0 100
18 QUCO 50 100 4 0 100
19 QURU 50 67 3 67 33
21 ROPS 50 60 3 60 40
23 TSCA 50 92 3 92 8
24 VIAE 50 100 4 0 100
Averages 40 72 34 45
1
Tree ISA cont.
RELATIVE FREQUENCY in group, % of perfect indication (% of plots in given group
where given species is present)
Group
Sequence: 1 2
Identifier: 3 4
# of Items: 2 4
Column Avg Max MaxGrp
1 ACRD 88 100 3 100 75
2 AEFL 25 50 3 50 0
3 AMLA 25 50 3 50 0
4 BELE 50 100 3 100 0
5 CAAL 13 25 4 0 25
6 CAGL 25 50 3 50 0
7 COFL 25 50 4 0 50
8 HAVI 13 25 4 0 25
9 LITU 75 100 4 50 100
10 MAFR 13 25 4 0 25
12 OXAR 25 50 3 50 0
13 PAQU 13 25 4 0 25
15 PIST 63 75 4 50 75
16 PRSE 38 75 4 0 75
18 QUCO 13 25 4 0 25
19 QURU 88 100 3 100 75
21 ROPS 88 100 3 100 75
23 TSCA 63 100 3 100 25
24 VIAE 13 25 4 0 25
Averages 31 48 33 29
2
Tree ISA cont.
INDICATOR VALUES (% of perfect indication, based on combining the above values
for relative abundance and relative frequency)
Group
Sequence: 1 2
Identifier: 3 4
# of Items: 2 4
Column Avg Max MaxGrp
1 ACRD 44 53 3 53 36
2 AEFL 25 50 3 50 0
3 AMLA 25 50 3 50 0
4 BELE 50 100 3 100 0
5 CAAL 13 25 4 0 25
6 CAGL 25 50 3 50 0
7 COFL 25 50 4 0 50
8 HAVI 13 25 4 0 25
9 LITU 46 83 4 8 83
10 MAFR 13 25 4 0 25
12 OXAR 25 50 3 50 0
13 PAQU 13 25 4 0 25
15 PIST 34 54 4 14 54
16 PRSE 38 75 4 0 75
18 QUCO 13 25 4 0 25
19 QURU 46 67 3 67 25
21 ROPS 45 60 3 60 30
23 TSCA 47 92 3 92 2
24 VIAE 13 25 4 0 25
Averages 23 41 25 21
3
Tree MONTE CARLO test of significance of observed maximum indicator value for
species. 1000 permutations. Random number seed: 42
IV From
Observed Randomized
Indicator groups
Column
-------- Value
----- (IV)
------ Mean
------- S.Dev p
1 ACRU 52.6 60.2 8.43 0.942
2 AEFL 50 33.4 11.85 0.336
3 AMLA 50 33 11.72 0.322
4 BELE 100 44.3 17.34 0.066
5 CAAL 25 33.5 11.91 1
6 CAGL 50 33 11.72 0.322
7 COFL 50 44.2 19.45 0.483
8 HAVI 25 33.5 11.92 1
9 LITU 83.3 59.8 10.55 0.066
10 MAFR 25 33.5 11.92 1
12 OXAR 50 33 11.72 0.322
13 PAQU 25 33.5 11.91 1
15 FIST 53.6 57.8 17.18 0.693
16 PRSE 75 49.3 23.66 0.407
18 QUCO 25 33.5 11.92 1
19 QURU 66.7 60 9.94 0.461
21 ROPS 60 60.1 10.3 0.671
23 TSCA 92.3 53.6 20.74 0.133
24 VIAE 25 33.5 11.91 1
* proportion of randomized trials with indicator value equal to or exceeding the observed
indicator value. p = (1 + number of runs >= observed)/(1 + number of randomized runs)
4
SHRUB ISA RESULTS:
RELATIVE ABUNDANCE in group, % of perfect indication (average abundance of a
given species in a given group of plots over the average abundance of that species in all
plots expressed as a %)
Group
Sequence: 1 2
Identifier: 1 2
# of Items: 3 4
Column Avg Max MaxGrp
1 ACRD 50 89 1 89 11
3 AMLA 50 90 1 90 10
4 BELE 50 60 2 40 60
5 CAAL 50 100 1 100 0
6 CADE 50 100 2 0 100
7 CAFL 50 100 1 100 0
8 CAPU 50 87 1 87 13
9 CLAC 50 69 2 31 69
10 COFL 50 100 1 100 0
11 CRMO 50 100 1 100 0
12 FAG R 50 100 1 100 0
15 GAUR 50 90 1 90 10
16 HATE 50 100 2 0 100
17 HAVI 50 80 1 80 20
19 ILAM 50 73 1 73 27
20 ILMO 50 100 2 0 100
21 KALA 50 57 1 57 43
24 LITU 50 100 1 100 0
25 NYSY 50 100 1 100 0
26 OXAR 50 100 1 100 0
27 PIST 50 95 1 95 5
28 PRSE 50 100 1 100 0
29 PYPU 50 100 2 0 100
30 QUAL 50 100 1 100 0
31 QUCO 50 100 1 100 0
32 QURU 50 95 1 95 5
34 RHCA 50 100 2 0 100
36 RHMA 50 69 2 31 69
37 ROPS 50 100 1 100 0
38 RUCA 50 100 2 0 100
39 SAAL 50 97 1 97 3
40 SMGL 50 80 1 80 20
41 SMRO 50 89 1 89 11
42 TSCA 50 94 1 94 6
44 VACO 50 100 2 0 100
45 VAPA 50 100 2 0 100
46 VASI 50 100 1 100 0
47 VAST 50 80 1 80 20
Averages 40 74 55 26
5
Shrub ISA cont.
RELATIVE FREQUENCY in group, % of perfect indication (% of plots in given group
where given species is present)
Column
1
3
4
5
6
7
8
9
10
11
12
15
16
17
19
20
21
24
25
26
27
28
29
30
31
32
34
36
37
38
39
40
41
42
44
45
46
47
Averages
Avg
ACRD
AMLA
BELE
CAAL
CADE
CAFL
CAPU
CLAC
COFL
CRMO
FAGR
GAUR
HATE
HAVI
ILAM
ILMO
KALA
LITU
NYSY
OXAR
PIST
PRSE
PYPU
QUAL
QUCO
QURU
RHCA
RHMA
ROPS
RUCA
SAAL
SMGL
SMRO
TSCA
VACO
VAPA
VASI
VAST
27
Group
Sequence:
Identifier:
# of Items:
Max
63
58
42
33
25
17
46
54
33
17
17
71
13
42
29
13
42
17
17
17
63
17
13
17
17
63
13
54
50
13
63
46
63
46
13
13
17
46
43
1
1
3
MaxGrp
100
67
50
67
50
33
67
75
67
33
33
75
25
50
33
25
50
33
33
33
100
33
25
33
33
100
25
75
100
25
100
67
100
67
25
25
33
67
35
2
2
4
25
50
50
0
50
0
25
75
0
0
0
75
25
50
25
25
50
0
0
0
25
0
25
0
0
25
25
75
0
25
25
25
25
25
25
25
0
25
1
1
2
1
2
1
1
2
1
1
1
2
2
2
1
2
2
1
1
1
1
1
2
1
1
1
2
2
1
2
1
1
1
1
2
2
1
1
19
100
67
33
67
0
33
67
33
67
33
33
67
0
33
33
0
33
33
33
33
100
33
0
33
33
100
0
33
100
0
100
67
100
67
0
0
33
67
6
Shrub INDICATOR VALUES (% of perfect indication, based on combining the above
values for relative abundance and relative frequency)
Column
1
3
4
5
6
7
8
9
10
11
12
15
16
17
19
20
21
24
25
26
27
28
29
30
31
32
34
36
37
38
39
40
41
42
44
45
46
47
Averages
Avg
ACRD
AMLA
BELE
CAAL
CADE
CAFL
CAPU
CLAC
COFL
CRMO
FAGR
GAUR
HATE
HAVI
ILAM
ILMO
KALA
LITU
NYSY
OXAR
PIST
PRSE
PYPU
QUAL
QUCO
QURU
RHCA
RHMA
ROPS
RUCA
SAAL
SMGL
SMRO
TSCA
VACO
VAPA
VASI
VAST
20
Group
Sequence:
Identifier:
# of Items:
Max
46
32
22
33
25
17
31
31
33
17
17
34
13
18
16
13
20
17
17
17
48
17
13
17
17
48
13
31
50
13
49
29
46
32
13
13
17
29
38
1
1
3
MaxGrp
89
60
30
67
50
33
58
52
67
33
33
60
25
27
24
25
21
33
33
33
95
33
25
33
33
95
25
52
100
25
97
53
89
62
25
25
33
53
32
2
2
4
1
1
2
1
2
1
1
2
1
1
1
1
2
1
1
2
2
1
1
1
1
1
2
1
1
1
2
2
1
2
1
1
1
1
2
2
1
1
9
89
60
13
67
0
33
58
10
67
33
33
60
0
27
24
0
19
33
33
33
95
33
0
33
33
95
0
10
100
0
97
53
89
62
0
0
33
53
3
5
30
0
50
0
3
52
0
0
0
8
25
10
7
25
21
0
0
0
1
0
25
0
0
1
25
52
0
25
1
5
3
2
25
25
0
5
7
Shrub MONTE CARLO test of significance of observed maximum indicator value for
species 1000 permutations. Random number seed: 42
IV from
Observed randomized
Indicator groups
Column Value (IV) Mean S.Dev p
1 ACRU 88.9 50.8 17.91 0.103
3 AMLA 59.8 52 19.83 0.375
4 BELE 30 44.6 18.98 1
5 CAAL 66.7 35.5 19.49 0.172
6 CADE 50 33.4 18.17 0.405
7 CAFL 33.3 28.4 4.2 0.415
8 CAPU 58 44.8 18.81 0.363
9 CLAC 51.9 49 15.75 0.476
10 COFL 66.7 35.1 19.15 0.16
11 CRMO 33.3 28.9 4.26 0.47
12 FAGR 33.3 28.4 4.2 0.415
15 GAUR 59.9 63.8 19.01 0.516
16 HATE 25 28.3 4.18 1
17 HAVI 26.7 46 18.95 0.89
19 ILAM 24.2 35.5 18.67 0.738
20 ILMO 25 28.5 4.21 1
21 KALA 21.4 42.8 18.3 1
24 LITU 33.3 28.4 4.2 0.415
25 NYSY 33.3 28.4 4.2 0.415
26 OXAR 33.3 28.4 4.2 0.415
27 PIST 94.5 51.6 19.32 0.063
28 PRSE 33.3 28.6 4.23 0.436
29 PYPU 25 28.5 4.21 1
30 QUAL 33.3 28.6 4.23 0.436
31 QUCO 33.3 28.9 4.26 0.47
32 QURU 94.5 57.4 19.58 0.124
34 RHCA 25 28.7 4.24 1
36 RHMA 51.9 49 15.75 0.476
37 ROPS 100 43.6 18.17 0.04
38 RUCA 25 28.5 4.21 1
39 SAAL 96.8 59.8 19.71 0.117
40 SMGL 53.3 42.9 18.56 0.363
41 SMRO 88.9 49.9 17.42 0.088
42 TSCA 62.4 49 19.56 0.388
44 VACO 25 28.7 4.24 1
45 VAPA 25 28.4 4.19 1
46 VASI 33.3 28.9 4.26 0.47
47 VAST 53.3 43.7 17.94 0.276
• proportion of randomized trials with indicator value equal to or exceeding the
observed indicator value. P = (1 + number of runs >= observed)/(1 + number of
randomized runs)
8
HERB ISA RESULTS:
RELATIVE ABUNDANCE in group, % of perfect indication (average abundance of a
given species in a given group of plots over the average abundance of that species in all
plots expressed as a %)
Note: Table continues for next 3 pages.
Group 1 = Control
Group 2 = Experimental
Group
Sequence: 1 2
Identifier: 1 2
# of Items: 4 4
Column Avg Max MaxGrp
1 ACGR 50 100 2 0 100
2 ACMI 50 100 2 0 100
5 ACRD 50 78 1 78 22
7 AGSP 50 56 1 56 44
9 AMLA 50 96 1 96 4
11 ANTR 50 100 2 0 100
12 ANVI 50 100 2 0 100
13 ARTR 50 100 2 0 100
15 ASDI 50 100 2 0 100
16 ASSP 50 100 2 0 100
18 BELE 50 100 2 0 100
21 CAAL 50 100 1 100 0
22 CAFL 50 100 1 100 0
23 CAGL 50 100 1 100 0
24 CASP 50 100 2 0 100
25 CEFO 50 100 2 0 100
26 CHMA 50 76 1 76 24
27 CIRA 50 100 1 100 0
28 CLAC 50 100 1 100 0
31 COFL 50 100 1 100 0
33 CRMO 50 100 1 100 0
34 CSPP 50 79 2 21 79
36 DASP 50 98 2 2 98
37 DEPU 50 100 1 100 0
38 DICA 50 100 2 0 100
39 DIDI 50 100 2 0 100
41 DISP 50 100 2 0 100
42 ELUM 50 100 2 0 100
43 EPRE 50 100 2 0 100
44 ERPH 50 100 2 0 100
45 FEPR 50 100 2 0 100
46 FERU 50 100 2 0 100
47 FRAM 50 100 1 100 0
48 FRVI 50 100 2 0 100
51 GAPR 50 99 1 99 1
52 GAUR 50 100 1 100 0
9
54 GOPU 50 100 1 100 0
55 HAVI 50 100 1 100 0
57 HEMA 50 100 2 0 100
58 HIPI 50 100 2 0 100
59 HISP 50 100 2 0 100
60 HOPU 50 100 2 0 100
62 HYGE 50 100 2 0 100
63 HYST 50 100 2 0 100
64 HYUN 50 100 1 100 0
66 KALA 50 100 1 100 0
67 LEHI 50 100 2 0 100
68 LEPR 50 100 2 0 100
69 LERA 50 100 2 0 100
71 LESP 50 100 2 0 100
72 LIBE 50 100 1 100 0
74 LITU 50 67 1 67 33
75 LOJA 50 100 1 100 0
77 LUAC 50 100 2 0 100
79 LYCI 50 100 2 0 100
80 LYQU 50 67 1 67 33
81 LYTE 50 100 2 0 100
83 MAFR 50 100 1 100 0
84 MARA 50 64 1 64 36
87 NYSY 50 100 1 100 0
88 OEBI 50 100 2 0 100
90 OXAR 50 100 1 100 0
91 OXST 50 100 2 0 100
92 PAQU 50 100 2 0 100
93 PASP 50 100 2 0 100
95 PIST 50 94 1 94 6
96 PLCI 50 100 1 100 0
97 POAC 50 100 1 100 0
98 POBI 50 94 2 6 94
99 POPR 50 100 2 0 100
100 POSA 50 100 2 0 100
101 POSI 50 93 2 7 93
104 PRSE 50 87 1 87 13
105 PRVU 50 100 2 0 100
107 PYPU 50 100 1 100 0
108 QUAL 50 100 1 100 0
109 QUCO 50 100 1 100 0
110 QURU 50 100 1 100 0
111 QUVE 50 100 1 100 0
113 RHMA 50 79 1 79 21
114 ROPS 50 100 2 0 100
115 RUAC 50 100 2 0 100
116 RUCA 50 100 2 0 100
118 RUHI 50 61 1 61 39
119 SAAL 50 91 1 91 9
121 SEGL 50 100 2 0 100
10
122 SMGL 50 100 1 100 0
123 SMRO 50 88 1 88 12
124 SOCA 50 86 2 14 86
125 SOCU 50 100 1 100 0
126 SOSP 50 61 1 61 39
128 TAOF 50 100 2 0 100
129 THNO 50 57 2 43 57
131 TIDI 50 100 1 100 0
132 TORA 50 100 1 100 0
135 TRRE 50 100 2 0 100
136 TSCA 50 77 1 77 23
137 VAPA 50 100 2 0 100
139 VAST 50 100 1 100 0
140 VEOF 50 100 2 0 100
142 VIAE 50 100 1 100 0
144 VICA 50 100 1 100 0
145 VICU 50 100 2 0 100
146 VIHA 50 67 1 67 33
147 VIPR 50 100 2 0 100
149 VISO 50 95 1 95 5
Averages 35 67 32 39
RELATIVE FREQUENCY in group, % of perfect indication (% of plots in given group
where given species is present)
Group
Sequence: 1 2
Identifier: 1 2
# of Items: 4 4
Column Avg Max MaxGrp
1 ACGR 13 25 2 0 25
2 ACMI 13 25 2 0 25
5 ACRD 63 75 1 75 50
7 AGSP 25 25 1 25 25
9 AMLA 50 75 1 75 25
11 ANTR 13 25 2 0 25
12 ANVI 25 50 2 0 50
13 ARTR 13 25 2 0 25
15 ASDI 25 50 2 0 50
16 ASSP 13 25 2 0 25
18 BELE 13 25 2 0 25
21 CAAL 13 25 1 25 0
22 CAFL 13 25 1 25 0
23 CAGL 13 25 1 25 0
24 CASP 13 25 2 0 25
25 CEFO 13 25 2 0 25
26 CHMA 50 75 1 75 25
27 CIRA 13 25 1 25 0
11
28 CLAC 13 25 1 25 0
31 COFL 13 25 1 25 0
33 CRMO 13 25 1 25 0
34 CSPP 25 25 1 25 25
36 DASP 63 100 2 25 100
37 DEPU 13 25 1 25 0
38 DICA 25 50 2 0 50
39 DIDI 13 25 2 0 25
41 DISP 25 50 2 0 50
42 ELUM 13 25 2 0 25
43 EPRE 13 25 2 0 25
44 ERPH 13 25 2 0 25
45 FEPR 13 25 2 0 25
46 FERU 13 25 2 0 25
47 FRAM 25 50 1 50 0
48 FRVI 13 25 2 0 25
51 GAPR 38 50 1 50 25
52 GAUR 50 100 1 100 0
54 GOPU 13 25 1 25 0
55 HAVI 25 50 1 50 0
57 HEMA 13 25 2 0 25
58 HIPI 25 50 2 0 50
59 HISP 13 25 2 0 25
60 HOPU 50 100 2 0 100
62 HYGE 13 25 2 0 25
63 HYST 25 50 2 0 50
64 HYUN 13 25 1 25 0
66 KALA 13 25 1 25 0
67 LEHI 13 25 2 0 25
68 LEPR 13 25 2 0 25
69 LERA 25 50 2 0 50
71 LESP 13 25 2 0 25
72 LIBE 13 25 1 25 0
74 LITU 25 25 1 25 25
75 LOJA 13 25 1 25 0
77 LUAC 13 25 2 0 25
79 LYCI 13 25 2 0 25
80 LYQU 50 50 1 50 50
81 LYTE 25 50 2 0 50
83 MAFR 13 25 1 25 0
84 MARA 38 50 1 50 25
87 NYSY 13 25 1 25 0
88 OEBI 13 25 2 0 25
90 OXAR 13 25 1 25 0
91 OXST 13 25 2 0 25
92 PAQU 13 25 2 0 25
93 PASP 13 25 2 0 25
95 PIST 50 75 1 75 25
96 PLCI 13 25 1 25 0
97 POAC 13 25 1 25 0
12
98 POBI 25 25 1 25 25
99 POPR 13 25 2 0 25
100 POSA 13 25 2 0 25
101 POSI 88 100 2 75 100
104 PRSE 50 75 1 75 25
105 PRVU 13 25 2 0 25
107 PYPU 13 25 1 25 0
108 QUAL 38 75 1 75 0
109 QUCO 13 25 1 25 0
110 QURU 13 25 1 25 0
111 QUVE 38 75 1 75 0
113 RHMA 38 50 1 50 25
114 ROPS 13 25 2 0 25
115 RUAC 25 50 2 0 50
116 RUCA 25 50 2 0 50
118 RUHI 38 50 1 50 25
119 SAAL 25 25 1 25 25
121 SEGL 13 25 2 0 25
122 SMGL 38 75 1 75 0
123 SMRO 38 50 1 50 25
124 SOCA 25 25 1 25 25
125 SOCU 13 25 1 25 0
126 SOSP 38 50 2 25 50
128 TAOF 13 25 2 0 25
129 THNO 38 50 1 50 25
131 TIDI 13 25 1 25 0
132 TORA 13 25 1 25 0
135 TRRE 25 50 2 0 50
136 TSCA 25 25 1 25 25
137 VAPA 13 25 2 0 25
139 VAST 25 50 1 50 0
140 VEOF 13 25 2 0 25
142 VIAE 13 25 1 25 0
144 VICA 13 25 1 25 0
145 VICU 13 25 2 0 25
146 VIHA 25 25 1 25 25
147 VIPR 38 75 2 0 75
149 VISO 25 25 1 25 25
Averages 16 26 15 16
13
Herb ISA results cont. for neat 3 pages
INDICATOR VALUES (% of perfect indication, based on combining the above values
for relative abundance and relative frequency)
Group
Sequence: 1 2
Identifier: 1 2
# of Items: 4 4
Column Avg Max MaxGrp
1 ACGR 13 25 2 0 25
2 ACMI 13 25 2 0 25
5 ACRD 35 59 1 59 11
7 AGSP 13 14 1 14 11
9 AMLA 37 72 1 72 1
11 ANTR 13 25 2 0 25
12 ANVI 25 50 2 0 50
13 ARTR 13 25 2 0 25
15 ASDI 25 50 2 0 50
16 ASSP 13 25 2 0 25
18 BELE 13 25 2 0 25
21 CAAL 13 25 1 25 0
22 CAFL 13 25 1 25 0
23 CAGL 13 25 1 25 0
24 CASP 13 25 2 0 25
25 CEFO 13 25 2 0 25
26 CHMA 31 57 1 57 6
27 CIRA 13 25 1 25 0
28 CLAC 13 25 1 25 0
31 COFL 13 25 1 25 0
33 CRMO 13 25 1 25 0
34 CSPP 13 20 2 5 20
36 DASP 49 98 2 1 98
37 DEPU 13 25 1 25 0
38 DICA 25 50 2 0 50
39 DIDI 13 25 2 0 25
41 DISP 25 50 2 0 50
42 ELUM 13 25 2 0 25
43 EPRE 13 25 2 0 25
44 ERPH 13 25 2 0 25
45 FEPR 13 25 2 0 25
46 FERU 13 25 2 0 25
47 FRAM 25 50 1 50 0
48 FRVI 13 25 2 0 25
51 GAPR 25 50 1 50 0
52 GAUR 50 100 1 100 0
54 GOPU 13 25 1 25 0
55 HAVI 25 50 1 50 0
57 HEMA 13 25 2 0 25
58 HIPI 25 50 2 0 50
59 HISP 13 25 2 0 25
14
60 HOPU 50 100 2 0 100
62 HYGE 13 25 2 0 25
63 HYST 25 50 2 0 50
64 HYUN 13 25 1 25 0
66 KALA 13 25 1 25 0
67 LEHI 13 25 2 0 25
68 LEPR 13 25 2 0 25
69 LERA 25 50 2 0 50
71 LESP 13 25 2 0 25
72 LIBE 13 25 1 25 0
74 LITU 13 17 1 17 8
75 LOJA 13 25 1 25 0
77 LUAC 13 25 2 0 25
79 LYCI 13 25 2 0 25
80 LYQU 25 33 1 33 17
81 LYTE 25 50 2 0 50
83 MAFR 13 25 1 25 0
84 MARA 21 32 1 32 9
87 NYSY 13 25 1 25 0
88 OEBI 13 25 2 0 25
90 OXAR 13 25 1 25 0
91 OXST 13 25 2 0 25
92 PAQU 13 25 2 0 25
93 PASP 13 25 2 0 25
95 PIST 36 70 1 70 2
96 PLCI 13 25 1 25 0
97 POAC 13 25 1 25 0
98 POBI 13 24 2 1 24
99 POPR 13 25 2 0 25
100 POSA 13 25 2 0 25
101 POSI 49 93 2 6 93
104 PRSE 34 66 1 66 3
105 PRVU 13 25 2 0 25
107 PYPU 13 25 1 25 0
108 QUAL 38 75 1 75 0
109 QUCO 13 25 1 25 0
110 QURU 13 25 1 25 0
111 QUVE 38 75 1 75 0
113 RHMA 22 39 1 39 5
114 ROPS 13 25 2 0 25
115 RUAC 25 50 2 0 50
116 RUCA 25 50 2 0 50
118 RUHI 20 31 1 31 10
119 SAAL 13 23 1 23 2
121 SEGL 13 25 2 0 25
122 SMGL 38 75 1 75 0
123 SMRO 24 44 1 44 3
124 SOCA 13 21 2 4 21
125 SOCU 13 25 1 25 0
126 SOSP 17 20 2 15 20
15
128 TAOF 13 25 2 0
129 THNO 18 21 1 21
131 TIDI 13 25 1 25
132 TORA 13 25 1 25
135 TRRE 25 50 2 0
136 TSCA 13 19 1 19
137 VAPA 13 25 2 0
139 VAST 25 50 1 50
140 VEOF 13 25 2 0
142 VIAE 13 25 1 25
144 VICA 13 25 1 25
145 VICU 13 25 2 0
146 VIHA 13 17 1 17
147 VIPR 38 75 2 0
149 VISO 13 24 1 24
Averages 13 25 12 13
25
14
0
0
50
6
25
0
25
0
0
25
8
75
1
16
MONTE CARLO test of significance of observed maximum indicator value for species
1000 permutations. Random number seed: 42
Note: Herb data continues for next 3 pages
IV from
Observed Randomized
Indicator groups
Column
-------- Value (IV)
----- -----
- Mean S.
------- Dev p
1 ACGR 25 25 0.79 1
2 ACMI 25 25 0.79 1
5 ACRU 58.7 57.6 17.45 0.63
7 AGSP 14.1 29.3 17.79 1
9 AMLA 72.3 53.7 18 0.255
11 ANTR 25 25 0.79 1
12 ANVI 50 28.4 18.5 0.423
13 ARTR 25 25 0.79 1
15 ASDI 50 29.9 17.34 0.428
16 ASSP 25 25 0.79 1
18 BELE 25 25 0.79 1
21 CAAL 25 25 0.79 1
22 CAFL 25 25 0.79 1
23 CAGL 25 25 0.79 1
24 CASP 25 25 0.79 1
25 CEFO 25 25 0.79 1
26 CHMA 56.9 45.6 17.55 0.374
27 CIRA 25 25 0.79 1
28 CLAC 25 25 0.79 1
31 COFL 25 25 0.79 1
33 CRMO 25 25 0.79 1
34 CSPP 19.7 32.3 14.99 1
36 DASP 97.7 49.9 14.96 0.023
37 DEPU 25 25 0.79 1
38 DICA 50 29.2 17.76 0.423
39 DIDI 25 25 0.79 1
41 DISP 50 34.5 13.68 0.439
42 ELUM 25 25 0.79 1
43 EPRE 25 25 0.79 1
44 ERPH 25 25 0.79 1
45 FEPR 25 25 0.79 1
46 FERU 25 25 0.79 1
47 FRAM 50 34.5 12.84 0.408
48 FRVI 25 25 0.79 1
51 GAPR 49.5 43.9 17.11 0.439
52 GAUR 100 45.2 16.17 0.023
54 GOPU 25 25 0.79 1
55 HAVI 50 29.6 16.91 0.407
57 HEMA 25 25 0.79 1
58 H I PI 50 32.1 15.45 0.428
59 HISP 25 25 0.79 1
17
60 HOPU 100 46.6 17.56 0.023
62 HYGE 25 25 0.79 1
63 HYST 50 33.6 14.03 0.423
64 HYUN 25 25 0.79 1
66 KALA 25 25 0.79 1
67 LEHI 25 25 0.79 1
68 LEPR 25 25 0.79 1
69 LERA 50 29.8 17.31 0.423
71 LESP 25 25 0.79 1
72 LIBE 25 25 0.79 1
74 LITU 16.7 31.1 16.55 1
75 LOJA 25 25 0.79 1
77 LUAC 25 25 0.79 1
79 LYCI 25 25 0.79 1
80 LYQU 33.3 46.2 17.24 0.819
81 LYTE 50 34.6 13.63 0.439
83 MAFR 25 25 0.79 1
84 MARA 32.1 39.3 15.41 0.712
87 NYSY 25 25 0.79 1
88 OEBI 25 25 0.79 1
90 OXAR 25 25 0.79 1
91 OXST 25 25 0.79 1
92 PAQU 25 25 0.79 1
93 PASP 25 25 0.79 1
95 PIST 70.4 49.3 17.91 0.142
96 PLCI 25 25 0.79 1
97 POAC 25 25 0.79 1
98 POBI 23.5 34.6 13.1 1
99 POPR 25 25 0.79 1
100 POSA 25 25 0.79 1
101 POSI 92.7 64.7 18.48 0.104
104 PRSE 65.6 43.8 17.2 0.142
105 PRVU 25 25 0.79 1
107 PYPU 25 25 0.79 1
108 QUAL 75 39.8 18.03 0.142
109 QUCO 25 25 0.79 1
110 QURU 25 25 0.79 1
111 QUVE 75 41.6 17.31 0.142
113 RHMA 39.5 39.5 18.11 0.723
114 ROPS 25 25 0.79 1
115 RUAC 50 34.4 13.34 0.423
116 RUCA 50 30.4 16.44 0.414
118 RUHI 30.6 39.6 15.91 0.736
119 SAAL 22.7 34.1 13.5 1
121 SEGL 25 25 0.79 1
122 SMGL 75 39.1 15.61 0.142
123 SMRO 44.1 42.3 17.5 0.742
124 SOCA 21.4 33.1 14.11 1
125 SOCU 25 25 0.79 1
126 SOSP 19.7 39.5 17.48 1
18
128 TAOF 25 25 0.79 1
129 THNO 21.4 39.5 16.72 1
131 TIDI 25 25 0.79 1
132 TORA 25 25 0.79 1
135 TRRE 50 31.6 15.42 0.414
136 TSCA 19.2 32.2 15.23 1
137 VAPA 25 25 0.79 1
139 VAST 50 32.5 15.38 0.437
140 VEOF 25 25 0.79 1
142 VIAE 25 25 0.79 1
144 VICA 25 25 0.79 1
145 VICU 25 25 0.79 1
146 VIHA 16.7 30.4 16.43 1
147 VIPR 75 41.2 17.38 0.136
149 VISO 23.7 35.3 13.12 1
* proportion of randomized trials with indicator value equal to or exceeding the observed
indicator value. p = (1 + number of runs >= observed)/(1 + number of randomized runs)
19
Appendix 8: Relative Dominance Histograms for Trees and Shrubs
Andrews Park Control (Trees)
c
ro
c
0
0
Y
ro
45
40
35
30
25
20
15
10
5
0 F-I F-1 F-1
t`J?SJ? ?o.?? ? e? ???
G
0t CO C'
Cg' J5 &
P ?J4? Q?t` 6t0 ?? C3? CJ Q? ?J0 CGJ? ?tGJ ?cs?
0
Adrews Park Experimental (Trees)
a
u
ro
0
0
m
ro
?fia ?J5 ?J? A??`' qua
z ?J ?o
0?J ot
SC3? J5 `fit J`'
5 5? 0S 'tJ ?a ?o
C,
Je?? 4`?J P?' ??so Ga Q eJ ?? J? `fi ??•o
30 -
25
20
15
10
5
F-I
0
West 1 Control (Trees)
35
30
a?
U
c 25

c
20
E
0
0
15

0
>
10

0

a? Sp4
? ?J
J?`
`
`e 07
46
5
0
0
C
?
P
`?
Q?GJ
O? `?` Oa?J
Q ,?Q J
C
4 ??tJ {?
?r? J?0
¢o???,
O
West 1 Experimental (Trees)
70
60
m 50
U
C
Rf
E
40
0
30
cu
Z 20
10
0
Pinus strobus Sassafras Amelanchier Nyssa sylvatica Prunus serotina
albidum laevis
West 2 Control (Trees)
40
35
u
= 30
e?
= 25
0 20
15
> 10
e?
5
0 F-I
S
CP 4,
Ile
Iq
West 2 Experimental (Trees)
25
v 20

tC
= 15

E

10

m 5
m

0

a
o?`?a o°J? ?G? ?os`aa
11
<<J `GJ? fi 1Q ?5-Z
?`S10 0?1?
4
West 3 Control (Trees)
25
20
m
E
15
0
0 10
d
5
d
0
o10JS
51 ?J?Q ?a0 oaC`? SSJ4 `? ??t0
v 4
West 3 Experimental (Trees)
4
4
3
m
3
c
E
= 2
0
2
1
Carya alba Liriodendron Quercus rubra Robinia Prunus
tulipifera pseudoacacia serotina
East 1 Control (Trees)
35-
30-
25 c m
20
0
0 15
10
5
0 F-1 Eli
??a? ? ??J?a `GJ5 GJ?JS Ott J?oa
Jo'?G o?agoP Qy 4Je P?? ?. aQ?o
0'
4
East 1 Experimental (Trees)
45
40
35
30
c
E 25
D 20
> 15
10
5
0
Ilei
2i
a
e?a?oC,
e
c
??? ?4i
East 2 Control (Trees)
50
45
40
35
30
0 25-
20-
15-
lo-
5-
0-
K" ?c 0
4`fiJ try SJ? ?J?? c A
O+
East 2 Experimental (Trees)
35
30

m
25
m
E 20

0
15

10

a
5

0

?av? ? oX)J
J
o-J0 ?
East 3 Control (Trees)
45
40
35
30
25
20
15
10
5
0
Q6p
0
Andrews Park Control (Shrubs)
18-
16-
14--
12--
10-
8--
6- y 4
2-
0- ?tcj, S,
? c?•? a??? 4 G ago oast. g{c`
Andrews Park Experimental (Shrubs)
40

35

L
30
N
25

20

15

0
10

a?
> 5
Y

0

West 1 Control (Shrubs)
30
U 25
c
20
0 15
10
5
a?
0
Oj
?a J? G? ,1a , jet GJ.
0
G ao G ??, G
?O
L ?
P? CIO
G?
West 1 Experimental (Shrubs)
60
50
c
40
0 30
20
10
0
?4
J
a
4
West 2 Control (Shrubs)
8-
7

6

U
? 5
c
a
c
E
4
0
0
v
3

R
2

1

0

Amelanchier Fraxinus Pinus
laevis americana strobus
West 2 Experimental (Shrubs)
80
70

60

w
U
C 50

C
0 40

m
30

20

10

0

Rhododendron maximum Sassafras albidum
West 3 Control (Shrubs)
25
20
U
C
Rf
E
E 15
0
10
m
T,
m
ca •.? ?a a? ?a aca , a
?a?? fiaa?fia aQJ?a??? +?aJ S?J?t
0t ?a fi? a ,a ?e a?
d? ??J?a Gala ?.fi Ica ??fi
East 1 Control (Shrubs)
25
20

U
C

15

E
0
10

5

0
J
.a?a??o?aaea?a
Ja
+ac?
J?
G1a
a?a
ea
as
01
? Series1
East 1 Experimental (Shrubs)
U
C
c
E
0
0
m
m

+ "
e
12 P
.
a ?a+
?e? 5?
G
eta a?a I \5
QJ? ?efi? ?e0
,a a as
??J?rpt a?aae ?J?Ga
G 4J
60
50 -
40
30
20
10
0- F-1

East 2 Control (Shrubs)
25
20

u
c
15

E
0
10

M
5

0

E

(0 U N E T E U)
N
Z5
(6 (6
0) .6
620c

co N c U
Z
U E n c s

(n V O
V M2

N
N
N
H N
4 (
ts5 2
E >
0 N (
6
E > U ) S6

a Q Q C7
East 2 Experimental (Shrubs)
50
45
40

35

30
0 25
20

15

10

5

0

ci?
o? Sal c3?? `ati ??t
r•?s
c?
`?? ago ?a J?? ??? P
??`g,, ,??J, oao 4?t5 C?
East 3 Control (Shrubs)
40
35
= 30
25
0 20
0
a 15
10
5
0
??o?•a `Sa ?a? ??a ??ca ?`•a Gamma a`,?a ?`?a
?a . aQ oa ??` of Q
C?a?? ao?et? Ga Q o?1?.a4s?-? ?a??a
?o ¢ Goa
East 3 Experimental (Shrubs)
30
25

u
20

0 15
0
> 10

5
F1
0
F-1

a??'a Spa J?a`a ole°?a GaG,a `eJCa .?J aava
CIO' ?? 5e? oa5 boa ?° a act
ci?
OJ Ga5 aJ a?Ja J?? a?? a ?`a5 G
O??t ? 4s t3 SaQ ??a? aaaa
Analysis of vegetative communities and the effects of human
disturbance within conservation zones along shoreline of Lake
Glenville
J. Dan Pittillo, Ph.D. and Ben Prater, M. E. M.
10 January 2005
Project supported by the
Western North Carolina Alliance
29 Market Street, Suite 610
Asheville, NC 28801
www.wnca.org
Analysis of vegetative communities and the effects of human
disturbance within conservation zones along shoreline of Lake Glenville
Abstract:
A study of the buffer strip vegetation surrounding Lake Glenville, a hydropower
impoundment of the Duke Power Nantahala Area, was commissioned by the Western
North Carolina Alliance. The "buffer strip" is defined as the area surrounding the lake
that is within the project boundary, measured as 10 vertical feet above the high water
mark for the lake. Seven paired study sites were sampled using the Line Intercept
Method, a technique selected because the buffer strip was usually limited in width. Each
study site was selected based on the ability to locate 50-meter line lengths along a
developed zone in front of homesites (experimental) with an adjacent undeveloped 50-
meter strip (control). Three sites were chosen south of the dam on the west side, another
three sites along the east side of the lake, and one site in Andrews Park. All vascular plant
species were tallied for coverage of the intercept line and trees and shrubs were measured
for the longest crown diameter. An additional listing of plants, or releves, were made for
species occurring within 5 meters of the intercept lines. Data on slope, aspect, compass
direction, and global position system were taken and photographs of the sites were made.
A total of 431 vascular plant species were observed in the study with 175 (40% of total)
of these sampled by the line intercepts. The data were analysed by the PCORD-
multivariate method. The greatest significance was the reduction in the woody plants in
the developed or experimental sites, especially for buckberry (Gaylussacia ursina), green
brier (Smilax rotundifolia), white pine (Pious strobus), and black locust (Robinia
pseudoacacia). There were moderately significant more yellow poplar present in the
developed sites accompanied by significantly more shallow rooted oat grass (Danthonia
spicata) and purple bluet (Houstonia purpurea). We conclude that erosional differences
observed in front of the developed sites relates to the loss of the root masses of shrubs
and trees compared to adjacent, undeveloped control sites.
Introduction:
Southern Appalachia is a physiographic region categorized by the southern
portion of the Appalachian mountain range extending across parts of Virginia, Tennessee,
and North Carolina. The species associated with this region comprise some of the
richest species diversity in the northern hemisphere. Some of the most unique residents
of this region are the numerous species of plants. These plants are ecologically
significant in terms of their diversity and the unique communities they form.
Species protection is a principle concern for conservationists. Conservationists
strive to resolve potential conflicts between ecological protection of resources and human
impacts. To effectively mitigate the impacts of human activities, such as development,
resource management regulations have been established. These regulations must be
complied with in order to achieve conservation goals. One of the most frequent
regulatory frameworks applied in the Southern Appalachians is for the protection of
water resources. An example of this are the conservation zones maintained along the
shores of impoundments in which native plants serve a key function.
2
Lakeshore erosion is a problem for impoundments, especially in steep mountain
terrain. This is compounded by windy areas and especially with motorboat wakes. The
greater the amount of slope and the greater amount of motorboat activity means that
erosion can be serious enough to scour the soil from lake margins. This silt then affects
the aquatic life quality of the shallow water zone and sedimentation of the reservoir as
well. For this reason, maintenance of shoreline vegetation with adequate root mats to
resist erosion is an important resource that needs protecting.
Objective:
The objective of this study was to assess the effects of development on shoreline
vegetation. The choice of a reservoir that has residences adjacent to segments of
undeveloped shores was desired. Lake Glenville represents such an area with extensive
shoreline development and is readily accessible by boat and in some areas, by road as
well. The hydroelectric reservoir was built in 1940 and began hydroelectric operation in
1941. It is managed as the Nantahala Power division of Duke Power which purchased
Nantahala Power Company in 1988.
[http://www.wcu.edu/mhc/npl/CompanyTimeline.htm]
To achieve this objective a vegetation survey was conducted and the results
analyzed in an effort to answer the following research questions:
• Is there quantifiable distinction between plant communities found in developed
and undeveloped areas?
• How are species associated with the presence/absence of development along
shorelines?
• Has development altered the species composition of plant communities and
compromised their ecological integrity and function?
• How can adverse modification be quantified in economic terms using replacement
cost estimates?
• What mitigation measures must be taken?
Methods:
Data Collection
A general survey of the east and west shores of Lake Glenville indicated
numerous cleared areas. Many of these were contiguous. Interspersed were areas of
vegetation that has apparently not been cleared since the establishment of the dam. A
line-intercept method (Cox 1985) was chosen as the method of studying the vegetation
due to the restriction of the zone owned by Duke Power, defined as ten elevational feet
above the high water level. While many of the gentle slopes extended back as far as 100
feet, the desired 10 in (33+ feet) x 10 in standard plot size could not be fitted in many
steeper sites. The line intercept requires only the width of a measuring tape (ca. 1 cm)
and thus could be fitted along he borders. However, the line needs to be longer and a
50m (165 ft) length was chosen. Three levels of vegetation were taken: 1) tree level,
defined as any plant exceeding 10 cm (6 inches) in diameter at breast height (dbh); 2)
shrub level, defined as any woody plant above 0.5 in (33 in.) but less than 10 cm. dbh;
and 3) herb level, defined as any plant less than 0.5 in. Two values each of trees and
shrubs were recorded, the distance crossed by the line and the maximum crown diameter
(M value), projected from visual perpendicular projection to the canopy edge. Trees
were recorded as 10 in segments while the smaller herbs were recorded in 5 in segments.
3
Herbs were tallied by different species for each 5 m. segment and their maximum
diameters were not recorded due to time constraints
Values of tree and shrub levels were calculated for importance values. This
technique uses the intercepted line coverage values for determining dominance (D = I/t)
where D is dominance, I is total distance of the canopy for a given species crossing the
line, and t is the transect length (50 m). The density (d, or the number of plants per unit
area) is calculated from the sum of the reciprocal M values times the unit area (here
defined as 100 sq. m.) divided by the total transect length. The frequency (f, encounters
of a species in a sample unit) is calculated as sum of the reciprocal M values divided by
the number (n) of M values. The weighted frequency (F) is the product of the frequency
values (f) times the number of transect intervals in which a species occurs. These three
values, dominance, density, and frequency are converted to relative values (100) and the
sum of them is the Importance Value (300). In addition to the species intercepted by the
lines, a 5-m strip along each side of the line and line ends were surveyed for additional
species, called releves. Thus the lists give a fair representation to the number of species
occurring in each line sample.
This technique allowed us to sample each pair at one location. The sample taken
on the developed (cleared) section was designated the experimental while the sample
taken in the unmanipulated woodland the control. A total of three sites were chosen on
each side (East /West) of the lake and an additional sample was taken at Andrews Park.
A total of fourteen sites were surveyed and 178 different species recorded and measured.
Data Analysis
In addition to producing the ecological metrics that were derived from the field
measurements, various multivariate analyses were conducted to examine relationships
between plant species and the sample plots. To perform these analyses the data was
distilled and input into a matrix that included the abundance of all recorded species
within the surveyed plot. For trees and shrubs abundance or quantity was recorded of the
total number of stems within each plot (N). For herbs no N was calculated so coverage
relativized by the species maximum was used as a surrogate. Table 1 provides an
abbreviated version of the data matrix for a visual of the input format. Appendix 6
provides a key to the abbreviations and various codes used in the analyses. All analyses
were carried out using a software program developed exclusively for multivariate
analysis entitled PC-ORD (McCune et al. 2002) developed by MjM software design.
QUAL QUCO QURU QUVE
P1 1 1 1 1
p2 1 0 0 2
p3 0 0 3 0
p4 0 1 1 0
p5 0 1 0 5
p6 0 3 0 4
p7 0 0 0 4
p8 2 3 0 0
P9 0 0 2 0
P10 0 0 0 0
p11 0 0 0 0
p12 0 0 2 0
p13 0 0 0 0
p14 0 0 2 0
Table 1: Abbreviated example of
data matrix showing oak species in
the tree layer
The analysis preformed to investigate the associations among species and the
study plots involved three multivariate techniques. These techniques included a non-
metric multi-dimensional scaling (NMS) ordination, a hierarchal cluster analysis and an
indicator species analysis. Each of the analyses were preformed in succession and
provided a robust analysis of the relationships among species and their respective habitats
using the available data.
NMS is considered to be the most effective method of describing ecological
communities and should be the preferred method unless other analyses are required
(McCune & Grace 2002). The reason this method should be the ordination of choice is
that it makes no assumptions about the data. NMS ordinates all samples relative to each
other all at once in a reduced dimensional space defined by the user. Therefore patterns
may be expressed as clusters of similarity based on ecological distance. Optimizing the
relationships between species and the environment into these discrete patterns is the goal
of NMS. These discrete patterns can then be further analyzed using the next technique.
Cluster analysis is a hierarchal classification technique used in ecology to relate
entities that are similar in composition. In this case plots described by species are
grouped or "linked" based on compositional dissimilarity and are quantified in a distance
matrix. The term hierarchal clustering relates to the fact that plots are grouped starting
with the most similar followed by the next similar. As clusters are expanded (total
number of clusters reduced) the associations among groups expand. An optimal number
of clusters or levels are sought in an effort to most adequately represent community
associations. To understand how species identify with these discrete groups the next
technique was used.
Indicator species analysis is used to attribute species to particular environmental
conditions based on the abundance and occurrence of that species within a group. In this
study the groups were defined through the cluster analysis and plant species were
analyzed for their associations and fidelity to specific groups. A species that is a "perfect
indicator" is faithful to a particular group without fail. Indicator values range from 0 to
100 with 100 being a perfect indicator score. Because indicator species analysis is a
statistical inference, a test of significance is applied to determine if species are significant
indicators of the groups to which they are associated. Indicator species analysis is a
technique used to differentiate between groups.
In addition to these multivariate techniques, species data was used to produce
metrics of ecological importance. In calculating the importance value, the relative
dominance is included. Relative dominance is the dominance calculated as a percent of
the total dominance values among all species. Relative dominance for every species
associated with each plot across all three vegetation layers was calculated. This metric is
well suited for illustrative purposes when comparing plots. A comparison of dominance
can be displayed graphically and serve to exhibit the contribution of species, in terms of
abundance, to the plant community within the various plots. It will also provide a better
understanding of what species are present or absent from manipulated plots.
Results:
A total of 431 species were observed along the transects and adjacent areas, called
releves, within 5 meters of the intercept lines (Appendices 1-5). The control transects
had 25 species unique to the controls and an additional 78 unique species were
intercepted by the lines in the experimental plots. In addition, there were 72 species
found in common with both control and experimental sites. The releve controls had an
additional 107 species while the releve experimental sites had an additional 149 species.
Non-metric Multidimensional Scaling
To determine the relationship between the plant species and the various sample
plots, an NMS was used. Figures 1, 2 and 3 illustrate the results of the NMS ordination
for each surveyed layer of vegetation using a six-dimensional solution. By examining the
proximity of each variable on the graph as it relates along the ordinated axes, it is evident
that the species are more closely associated with respective plots. The plots (triangles)
that are closer together are more similar in terms of species composition. The species (+)
they share in common are the ones closest in proximity to the plots.
Cluster Analysis
To determine if distinct groups existed in the data, a cluster analysis was
performed. Cluster analysis was conducted at a level of 13. This means that the data was
initially clustered into 13 discreet groups. With only 14 plots a level of 13 allows for the
possibility of total dissimilarity between virtually all of them. The level of clustering
used is arbitrary and the user can adjust the level to show the best relationships among
plots and species. Adjusting the level of clustering revealed that at a level of 6 the best
representation of the data is shown. The outputs of the cluster analysis are purely
graphical and include a dendrogram and a color coded NMS. The dendrograms (Figures
4, 5, 6) illustrate the mechanics of how the hierarchical clustering was performed by the
algorithm. Each connection indicated by the lines show the magnitude of difference
between groups and connects similar groups together. These similar groups are color
coded to represent the most distinct associations.
The color coded NMS's (Figures 7, 8, and 9) gives us a look at how these groups
associate in ordination space and provides an idea of which species belong to which
group. Each of these groups is arranged along axes based on ecological distance as a
measure of dissimilarity. This concept is well translated in the NMS as groups closely
associated with each other are the same color.
The cluster analysis resulted in the final selection of 6 clusters as the best
representation of species and plot associations for all three vegetation layers. This
grouping level was used in the indicator species analysis as a starting point.
6
Indicator Species Analysis (ISA)
The final multivariate technique takes the analysis one step farther. We have
illustrated how species and the sample plots are associated and we know which groups of
plots are similar compositionally, now we can determine which species best represent
these groups as indicators. The indicator species analysis produces output which provide
indicator values based on relative frequency, abundance, and the combination of both.
The values themselves indicate a species fidelity to specific groups. Species with high
indicator values are said to be good indicators of that group.
Appendix 7 provides the indicator values for all the tree, shrub, and herb species
as they relate to the two groups delineated by the ISA. Four of the six groups delineated
through the cluster analysis were dropped from the analysis because they included only
one plot associated with no species. This group reduction occurred during all three ISA's
for each vegetation layer. Similarly, several species were also removed from each of the
ISA's because of their low frequency of occurrence. These species are dropped because
species that occur in only one plot will always be perfect indicators of that plot making
the results less informative for the user. By not including these "rare" species the final
result of the analysis presents an unbiased indicator of species fidelity.
The indicator values were based on relative abundance, relative frequency, and a
combination of both (Appendix 7). A Monte Carlo test of significance on each species
indicator value is also provided. Based on the Monte Carlo test of significance (Dufrene
& Legendre 1997), eight species were moderately to highly significant indicators of
groups they were associated with.
The non-significant indicators can still be useful to understand what Groups 1 and
2 might characterize in terms of disturbance. Those species with perfect indicator scores
can be attributed to either the control or experimental plots within the original data
matrix. Based on this cross checking it is evident that for all vegetation layers Group 1
represents the control plots and Group 2 represents the experimental plots.
Below is a listing for the species that are statistically significant indicators of the plots:
Tree species of moderate significance include:
Betula lenta (BELE or sweet birch) for control sites
Liriodendron tulipifera (LITU or yellow poplar) for experimental sites
Significant shrub level indicators include:
Robinia pseudoacacia (ROPS or black locust) for control sites
Shrub level species of moderate significance include:
Pinus strobus (PIST or white pine) for control sites
Smilax rotundifolia (SMRO or green brier) for control sites
Significant herb level indicators include:
Danthonia spicata (DASP or mountain oat grass) for experimental sites
Gaylussacia ursina (GAUR or buckberry) for control sites
Houstonia pupurea (HOPU or purple bluet) for experimental sites
The significant results are summarized below:
Significant Indicator Values (p<0.05):
Using Monte Carlo Test of Significance
Significant Indicator Values (p<0.05):
Using Monte Carlo Test of Significance
Tree Species
IV from
--' --'
Shrub Species
IV from
Species that are considered significant indicators provide insight as to how
species relate ecologically to development disturbances. A species that occurs in all or
nearly all transects is expressing the result of influx of seeds (or propagules) and the
ability of the habitat to sustain this species against other potential species. In the more
mature forests, ecologists usually refer to those that either dominate all others (such as
oaks and hickories) or are regularly found in the stand (such as buckberries or mountain
laurels). Based on the species data there is evidence that supports the ecological
distinction between plots that have been disturbed and those that have not.
Additional Analysis
As stated previously the data collected in the field can be used to derive a wide
variety of ecological metrics that aid in the analysis of impacts to communities. On
average, plots that had been manipulated lost 42% of tree species abundance. Plots that
were developed experienced a dramatic 82% loss in species abundance. The herbaceous
layer was analyzed across plots using coverage. Based on coverage the herbaceous layer
experienced a 33% increase in herbs on experimental plots. This is likely contributable
to an increased level of sunlight reaching the forest floor after the loss of shrubs and
trees. It is important to note here that herbs also included seedlings of trees and shrubs
which are further stimulated by canopy openings.
To gain a sense of what these trends illustrate graphically the relative dominance
for each species is presented in a series of histograms (Appendix 8) comparing each of
the 14 plots. These histograms corroborate the trends described above illustrating a
decrease in species dominance and species diversity amongst trees and shrubs.
Figure 1
Tree NMS
2
A
O.EFL
A,
p14

p13
B ELE 0
p9 + TSCA

+URI} OpS
+
p12
A
Romi
LIITU COFL
?
PR
+ SE +
OXAR
+
p21 AC RU
AM LA +
+
CAGL

+
7 PIST
p + CAAL WPR
A + AS
P1
A
NY SY
QUVE +
+
?A6? L

P5

p6
A
QUCO QUAL
+ +
P8

Axis 1
9
Figure 2
Shrub NMS

PH RNIRM
A
A
2
A
CLAC
+
3?TE
SIN RO U SI UE
+ HAVA
P ISY+
ILAM + GA UP

9 13
BELE ANI LA
Q ©
+ SMGL §6AL
VAS+ T
++ CAAL
+ +
CAGE
+ QURU
RHMA CAPU +
+ +
COFL

aPA
gh RHCA
+
VACO
+
ROPS
4PRSE
CRMO +

p12
A
W IXE'
A
OXAR
+
QUCO
MIRA
it
QUAL
ACRU +
KA LA
}
FRAM
+
pI® E
2N
Axis 1
10
Figure 3
Herb NMS
R
A
GAUR

QUAL
+
Ski RO
+ OLIVE
AST
GA RMUO
PWM? SO LIBE CRMO + ?M
+ + + FRAM PRSE +
+
T3PE
N +
O
& + FIST CA(AL RHMA
+AQU + SAAL SM GL +LOJA + YPU ++ +

HYUN
+
ACRU M EVI
+ CHMA.
QURU F*vl +A GAL
+ + +

DBFN MARA SOSP
IA E
+
A GSP +
+
+ + plum
VIHA

? RUHI OS I HISP
SPP ASCO + +
+ + @y
LYOU VIBL + RE
UCA+ASP ELI +
+
A RTR +
+
POAC
+
A« SOCU ANTR TA OF
+ + +
V1CU LUEC

TRGR
+
ASDI
+
LYCI
FEPR
+
P 081 F DISP
+ ROPS
EDF
+
CAFL THNO + -+ SOCA
+ICO + +
-f

PRVU
+
BELE POPR
+ VIP R +
+ POSI
+ LYTJ
p12 DASP + V
d HOPU + DICA
+ .}}- LER.A
-SP-4ANVI
HIPI
+ ?y
Axis 1
11
Dendrograms
Figure 4 Tree Cluster Analysis
Distance (Objective Function)
ME-02 4.6E-01 8.5E-01 1.2E+00 1.6E+
Information Remaining (%)
100 75 50 25 0
p1
f!-
saRg6
P'
Ir'
1 5 3

e
p

Figure 5 Shrub Cluster Analysis
Distance (Objective Function)
1.4E-01 0E-01 1.7E+00 2.4E+00 3.2E+
Information Remaining (%)
100 75 50 25 0
clstei6
p13 1 11
P1

P
1

Figure 6
Herb Cluster Analysis
Distance (Objective Function)
8E-02 8.3E-01 11+00 2.3E+00 31 E+
Information Remaining (%)
100 75 50 25 0
P1
PB

V ?leifi
0
p
1,a
p1
0
P.

12
Figure 7
Tree Cluster Analysis
2
PAJAL p14

COFL

PR9F
F
?R1A LITU
F
*SY
p7 + ROPS p13
+
PIS+

QURU
+
\RR
A.L
C +
O
R ACRU
+ + TSCA
QUVE +
+
CA GL B ELE
+
+ P9 +?
P5
a
P1
a
PG

QUCQ
+
QUAL
+
p8
A
AXIS 1
13
Figure 8a (plots labeled)
Shrub Cluster Analysis
2
P4
P11
A

P3
5 1
+

P9 p13
+
+

-F
p2

p12
+
P10
+
+
p6 PB

+

P7

Axis 1
14
Figure 8b (species labeled)
Shrub Cluster Analysis
d
2
y?rc
CLAC
HATE SM RO,
+ HAV fA SI FU992
PIST+
ILAM + GAUR
+ +
BELE © AMLA Q
+ SM GL ° L
VA S+ ?- CAAL
+ +
CADE
+ GURU
RHMA CAPU +
COFL
VA PA
IN RH CA
+
?'FilEf
+
OXAR
CUCO
+
AL
A
VA CO
+
ROPS
CRMO 4P RSE
+ +
H3f :%
DUAL
ACRU +
+
KA LA
FRAM
+
LIBE
+
Axis 1
15
Figure 9a (plots labeled) Herb Cluster Analysis
P1
A

P11
+
+
+
+.

++
p13 + +

+
+ + +
+ +
+
-4+
+
+
+

+
pt0
+
+ +
+

+

+ f
P3
'f
+
p7 +
P5
+ + ++ +
+
+
+
++
P9
+ ++ q
+ p
6 +
P

+
+

+

+ +
+

+ 8
p12 +
'F
+ p2 +
+ +
+
PA
+
Axis 1
16
Figure 9b (species labeled)
Herb Cluster Analysis
CL
As
l?f?fi1R
GAUR

QUAL
+
SM RO TORE
+ OUVE VAST
AST
N{?B1?
CRM O + -'
I
+ + + FRAM PRSE +
P
+
TII
C+PUACRE
IN +
P +
PIST Ab?A RHMA
AQU +
SAAL S M GL +LOJA F YPU ++ +
+
HYUN
+
ACRU MEVI

GURU
AVI CHMA
AGAL
+ +
+ +

DDI4DJI
MARA q
OSP
VIAE AGSP
"I ++

VIHA +
*
RUHI OS
I HISP
CSPP ASCO + + +
+
LY QU
VOL
+ F?
E
RUCA+ASP *LI +
?
A RTR
+
POAC
+
Eamm SOCU ANTR TAOF

VICU LUEC

TR GR
+
ASDI
+
LYCI
FEPR
POBI + DI+

CAFL + ROPS
EOF
+
THNO + + SOCA
+ICO + +
o?+

PRVU
BELE POPR
+ VIP R +
+ POSI
+ LYrP
DA SP +
HOPU + _y.DICA LERA.
+ ASSP-{,ANVI
HIPI
+ E
+
Axis 1
17
Discussion:
The results of this study allow us to draw a few conclusions about how the plant
communities of Lake Glenville have been impacted by development on the shorelines of
Lake Glenville. Based on the findings of this study there are significant differences
between the species assemblages within developed or experimental areas and those
within undeveloped or control areas. The results of the multivariate analysis combined
with the facts and figures derived from the field survey point to the dramatic differences
between the two types of sample plots.
From the results of the multivariate analysis we can infer how structure and
function of communities along the shoreline have been influenced from our analysis of
associations among species across plots. The cluster analysis suggests that plots can be
classified into groups based on the measure of dissimilarity among species. The NMS
and cluster analysis suggested the possibility of a number of groupings which were
further interpreted by the indicator species analysis. The indicator species analysis
eliminated four groupings due to single occurrences of species. The results of the
indicator species analysis reveal that for all three measured vegetation layers only two
groupings of plots are significant for each vegetation layer. This is interesting since plots
were categorized as one of two types, experimental or control.
The ISA showed that several species were perfect indicators of particular plots.
These species, although statistically insignificant as indicators, did however translate
directly to only being present in either the control or experimental plots. The results
clearly distinguished between the two plot categories. Based on the multivariate analysis
we can conclude that there is a distinct difference between species composition found on
developed and undeveloped plots.
With species abundance used as a measure of the impact of development it is
evident that areas that have been developed have been adversely modified. A difference
in the diversity of tree and shrub species in the developed plots from those in the
unmanipulated areas indicates a significant environmental impact. An observed increase
in herb diversity in the developed plots is expected due to the increased sunlight reaching
the forest floor following the removal of canopy. This increase in diversity may be
undesirable due to the presence of invasive exotics.
Invasive exotics can be a problem in re-establishment of natural communities.
For example, Multiflora Rose (Rosa multiflora) was observed in one of the releves and
becomes a major invasive. The potential also exists for Japanese Spirea (Spirea
japonica). Not yet observed in the study area, Oriental Bittersweet (Celastrus
orbiculatus), a species favored in plant arrangements, is becoming quite invasive
downstream on the Tuckasegee River. While these invasives may be somewhat effective
in helping curtail the erosion problems, they prove problematic for the viewshed that
adjacent land owners desire.
Aside from using abundance as a measure of impact one can also look to the
presence or absence of particular species as an indication of disturbances caused by
development. Based on the indicator species analysis the significant indicators for
developed and undeveloped sites provide further insight into how particular species relate
to the surveyed sites. Based on what we know about particular species responses and
habitat requirements these species make sense ecologically. Shoreline erosion involves at
least two forces: that from the wave action and overland wash from the slopes above.
18
Root masses of native trees, shrubs, and to a lesser extent, herbs, help control this erosion
from both processes. While the wave action is more significant in that it undercuts the
root masses, none-the-less the root mass counteracts erosion even after the roots and
plants connected to them have died. The experimental plots, having a smaller tree and
shrub root mass, are unable to retain the integrity of the shore nearly as well.
Projecting the processes that are taking place in the buffer zone, it is quite evident
that the removal of the woody vegetation, both in the form of living plants and woody
debris, will exacerbate the loss of shoreline integrity. If this process is to be curtailed, the
current interaction of landowners with the buffer zone vegetation must be altered. The
improvement of the management should be beneficial to both landowners in the future as
well as the reversal of the erosional rates that have occurred on Lake Glenville during the
past half-century.
References:
Cox, George W. 1985. Laboratory Manual of General Ecologv. Wm. C. Brown
Publishers, Dubuque, Iowa. Pp. 60-68.
Dufrene, M. & P. Legendre. 1997. Species assemblages and indicator
species: the need for a flexible asymmetrical approach. Ecological Monographs 67:345-
366.
McCune, Bruce. and Grace, J.B. 2002 Analysis of Ecological Communities. MjM
Software Designs, Glenden Beach, Oregon.
Schafale, M.. and A. Weakley. 1990. Classification of the natural communities of North
Carolina, third approximation. North Carolina Natural Heritage Program, Raleigh.
Weakley, Alan S. 2004. "Flora of the Carolinas, Virginia, and Georgia" (Working Draft)
University of North Carolina Herbarium, Chapel Hill .E-mail:
Weakley, Alan S. 1999. "Flora of the Carolinas and Virginia" (Working Draft).
University of North Carolina Herbarium. The Nature Conservancy, Chapel Hill, NC.
19
me e
erm
me
F me
✓A
'-
%.
4:rp }
me 11
y 4ttgv
St
el
' q ..^°'.
JJ ahe me m me.x A * yIL
.
9& Y A 'Y
- 1h�•} - - y6 r - e -
! s
e ,s y
el Afl
eve
" —
me
me
.2Nte
.r; r 'rem
% `. .
r V4 I�
ar
Am YAO
A me A
!.s owl,&ri
mA a
me -g
:r
a Ar
4
— -- ,
"' -me1� Arm
!A*a
AW :}I me
-4 �-
-- V - — 1�
— - >_ _ _
�. _
_ v _ -
-_
s_m _we
—mmAe me— -- .—""— — — —me

..+ ! �
�y -.- ��,.. ,r } � ��
�' � . �� ��
I` f �
7777 - j, L �.
1.y �' }fid � � Y
�:.
�^ �-
r r � -; _
,��;-
�-+
_ _ _ jo-�"_
.��

Fit
IF
4M %ifle
pe �pk
XTi OFF
- `
tt Y r` •,
F1 IF
tie
a
I
Fee
Fee,IrF
k.. -FF,
l� ..->_ IFl
�._". _ - _ _ _ _ s. 1...

�g,�rfS 'y. ,, -� r rq � a � �a'� as
'ter �,hii � �'- ��"t ,� � s w _ e� �� t 4 -a. � '�� .�''�'�" a azP z��.
�. ��, i w ink. � :�� � r .Y„� � K" � # '�% G,��F,,�`%;�*;'
��_ fax a ,� _ 2 ,�,�,.,,,-�- � s " r,.
t �nf � � '� f ,� li r
��� �� ,� � �o- �� z� ����
_ , y : _,.

AA i
YE fmm
,!
1444
j S
; - .'Im,§2Lx"
Y`Im.'+� p
4 .t
A If
1 AI
€f x
:i

.0*
All Ilk
90
s
{
f
s

JON ON
IN dr, lle
Ol
44 Lill - Ll
Lk
K 4 `
8 1 t v
}�ir 4-1 ...
eee Or
ir
f s
d
s _
a.
Oil
71. Id
lk 14
Kit
Ok
Of
OfOF
lqq
-_
_Nol
-NOl.,NN_
Oil N
,' Oa: .r

' ' y. Y �.
-, .y-_
..�� . �J��hi � hE� �y� _ _.�'i�� sc�i'I�aBS ��� . =�, 'u':�.

III AI A fT
v�
t I I W��nljoiPoe IJ:`„
be
i 9
PIL ell le IV'
AA
le
VI-
ma
s X
t�IV r.0 e XL I
10
Vv
I Ile
Ile lk
eV
-
We L
J.IF
IF 4
ec
, ,�;, ,� 3 a,� y .iY� PYA Q ,fi
�tel IV,
ml 41 IIIV -r I,VI
s`k4 Axa
le, I+c¢
-LwIYVa: V, .t .
PO
x >
I ell
}
4 b In
IL x. 54
> i @
> N
l3TC��lig
iYLe
ole
3 xY
21 11
.Pe,.LZ

7''-'6pre
^k
on
F OR, r"4`' nom/nice :,�,;. Ito. On ` J•^.I ler,
7 f
or
Id 0
i
_�- £ ^� L of
eSI t
�.ere
s�,•a�lefor, N,`�+...ati s+5., 3� -�csfi1 [, t,'
off .1 offer re onto Pool I
1 `+ ry;
ee
el neer
fee
Iry
or ev: In Or
roof I On 1.4
Ol ell
. , e'nor t leer .113
rel oreL y dor
Aw
lona
IV em Iol O!"Or
Iler
of
.'C It
fino
r^or Ire �k
leek
I l:�zI M'k *L 0*
1
or
forefif. le e 16, Onorr.
rIn
In
Ol
SO n
e+ -
"•T — _
. _. ••`�'^�s--...-.�. -rte, h "'`' --
..
reeler
we I
..-._ - 4In I
-
fee efem re �fe... �.

OFF
� FFF,I91
P4 I
IF
IF
I IF
IF
_ ?INa 1.
dlIF
r
11IFa f `U, t
FIFA
Ile 0
�r l
. x
IF
IF
IF
IF
I Fir
IF,
O M _ ,.
_ a
-a
--

"n. FIFE FIX FF
IF
i 4r
IF
IF m
m 4�ilftl wt
s f 4
IF
F.,rFF IF
61
IF I
I Fe
40F.e III
IF
,y ...
�IF
,.IFIF
IF
ILL>
1 IFIF
-`- IF -- -
IF
IF
IF IF
IF
IF
IF IF
IF
IF IF
_�� — -FFFI,_ _ _

IF I
?eFeFei
P
t
if !Fee
Fee 1
Fee r
eF Or IF Feel I
IF u * Zr s ZZIt
Fee
CT a
Foef�� 14 er Fie
He Ile
'r
Fee, 1
a
'Fee IhPAd IF
:r tr
Fee V' -�
J
Fell,xP IF IF r
Fe
r Fee
�-
Ile
s,81 MCSIF
.
,. FY3,A�'•.
,.IF
_
_
-
Fees
1
OF�
_
,IF
Ile I
_
-
_.
-- -
-
—
--
_
_
- y
-
-_
_
_
_
_
_ :_
-
_ - _
_
-_ -
_
_
Fee
I
_
Fee
_
—_
IF.-
II--
- --
_
a_-
_ _-

e y
7th
6116
let 0
6 let
AW
y
6^.
yA
r _
4.
Y _

a a
�-Ile,
;.i
r,� ..s
wilt. rix
IT e
00
;'IN_ •:.,

Ct
,�
Lae

v ..
fes.,
�. � �i�+ Eli• _ r°'-. .. i�i.f - moi' �'R!h_ t.. .�_.��'3 -�� -

a'
MAN
.A �., fes,
r.
ft.
•
r

a r
Rkill
IlkkI
wr
dee
ol
el
:.
pkill
Illipi,y ' b t PIAze:
jo
kill i
kI
or
le
kill
Phi
kil
TLI
s
+.
t
qp koo
Al
dI
kill
__ __.
kil

FIN
1.
vv
Nit IF
IF
IN t f
� .
FINI
Fa
AI
yI
IF
IF IF
IN
{ ;�n; .
IF
It ?1 .fir. 4 I�a> r For b
IF
a
IN
FIFO
IF
IF
NiFM
IN
IF IF NFFFN+I
IF IN
IF
I IF
"IFF" IF ­FFFIIII
FIVII IV
IFFIFFIN
FIV
e
VNNm
—
VON FIFF FINN
-VAI "Imp IMINIFFIL
— _ _ _ IF- _ _ _

eytIF
'V,
}1. 7 °
PIP FiF. IF
I I Pr.
n "
PIP
IF
IF I
1
a1IFPIP: 11 q IF
i}
If
i.10 I
I IF
PIP: 9
IF
IF IF,
j
���P

+p
w

lots if
�� •

,y
� ,;+•e'er r '
_ss.+, - _ _ — �- _� _ .�.-it• _tet-..►�- 'T::
- 3

ac
K
Y }}�
l ,
,..
am
bbb
T
y
�R
a.
r.
a at
Iq
al
am av
"a le
�t,+I am
amax
F�
�- >�- - -
ma
I rami; �-
.„. --
IT
v. N '
3
maw
,yam— -- _
IW It.
—_ - -
mam 11
am
Ipm
mar b: WI
aw,maI II
am.

t
Vii*'.? . � k � � � :" a+ • �-,pr; � ��� ^��� t �; 4 VON
w, 3Y i
1 21
Im
77 mv
I WWI
ftb
tow c
_ _ . - •.'•;.+w'" ,�._ �, ter„

T IV
kvla'S
1. "c _
' h'4
3
i
'VI
sf1 ,
IV
I IR
s�
Atel+
.�IV,41
e�
#! VV
LL
Iz
14
x

'I

9
XF
lot
%oil
!y _
- ,..
LV
.� ' R
tt
a
tp

4t
r

ell
IF
�SL two
:�
ee
10
00
� r
�v
_ - —�Ott

Alf
Cyd V`�§cx. r<�
T
'^- '� .
p I �I w _
ICA
a{..
e, o�-Vn
iII
Vill
u
%
did I,
leddo
ld
did
MW 1 41
YL
_.
- - =-- _=_ - - - - - -_ - _ " _ - - _ -- _
- -
_ _. -
- -
_mmIdd- w = _ _
- - _ �s� - ` -_�f- - _ - -

i _ c11
IF
FF
FI
el
Ff
tF
IF d
tdl.Ll
Nat �y!,^.' rte.OF
I t IF I
FFFa+IF
41
.{
Ak IF m IF
Fir Fri k
IF
hFIC IF r
^y �p9 d� elf of _
f ,x I y� x�
�. `� PrP
3:.
IFI—
- _ _ �c
For 0
I IF
-- r�FIF_
FV r
AFF I
IL
IF
IF
r'
_ _ _ _ _ .>

If FiN NNNFN We
IQ
Y -vim}✓ ,� �N�y
N;7 OF
ell
Itt7iznn IF. It
AFN
IN
�Ff ys P, ckFeeIQ -
r..r i*R `r �Z--F f' 3 ue' ¢
tell,I N,�'-,-�'•- I ' h
—11
AQ I on
1% 1 6%ell QIa Inf Y,. ^�'I INre 11 IQ I eNT IN, 1
NFL NA�1€lipQ� `X
7
I IQ IN 11 te, OF FF
t FQ N4
,xIt I F7
IN I
IN
' .' ,. -
Feletr
�r IN'�" <
IN
IN g V
0
IN
It I
—rtIN I �y 4
INF e, IF
IN
,yk+ 'iYr
ye 7�Fenn
te
137L I— I Few
1. New
FN
Fell I Isk6w, . FIN . Feel INr
I IN III
IF
I IN Inner
IF
— - _ —
r
i-,, Fell_

IT
+ i
Lo
a fy(r;
LL
WA
a
y N

k_
LLL
w ztp it
# d:
to
titt
SWt
tt
it
-24
1 a cc :tt�'. s id�
fir
It J;tz% Joiiiv Ilk It
Aw
It, zt
'. ✓,\§;tilt
IFRIt
It
4 it
It
y
tl
I
..rl o low i
MR
-�_-_�__--
`�

s ..,•. Jam-�+, .. .. _ ,rte_ zit.
a Y
_

do
"} f .
vtIleAd do
«R ^1I N1 do fix, k r:4
do
IL Ited
may' a OW! I ' s a' a
g
A odd do 'Ir Ad3 vg 4i�
I4led I do
. s.
r� » u ti
IV
t -
Id14
,Add Id F f ✓'" 11 Add,5'q
1 50
Ad
14
dr
Al
d,
C
07 1
Idw
r� Ile }
i ^r.�
Idd
- ~ 1
1 yArd. z< a r yp.
s
s -
dor I
t
do o Ledde— deded
_
III All
Id
do -
odd

al►► .r, ••. I. SIL
av
t, z
+ws j � "'tilt► . � r a - `i
"y�� ®,yam •. s.� 1'
ve
r � �

"xarm
xir'. F l r
14 1
err
;S fat
dill _off
dI
3 5r, NP
a
44
ie
a4 KI''
IIIIP
Aj
dew
5 S.�
m�O

• ,� "';o- y .t; a J:
tip
�s tIN
y
�4 �;; C�
Ip
W.
ViC
rr
{, k
\_ st WII . e � ;
�iIV
4zoVr
.. r_ -- 41
A V, IV
It

WA3
l
lj,js,

yrs

W,, 11,da
w ._
•.�
1 ^ i:` 9 yrIW J r.
9
mw_ 'offkak
r.
4 kip
km
lid 0
j/
v1.;

` ,''` dot
UllAy.�,
L Ll R- - �,
11yyR
411

mv R
'•;; a>.F
Ito

VOW
LZ
s�.r �•.-• 4 `

all
id r ; wr ay
a s,
wl
JIM. Awl wo0o
ll ow�
Of
f`
lit lip f. ISLA
g-,
If
-_o...
' ` ,

A +'ef
may.' s
�It
,.
If
lo
111
40,
4 lot
41
ot
--�
�' ,•^/�, ° r ?� ysl�+,� it
16
zls
d
40.
' 3
Oat
� 2 ,
.tea• i.. .. -1-u* +r idyl'+�' w .'i:J:,=<'�4 .,,s� 3... ..�

I, r Ile 91W w toe We
All
r x+16 — 'Rt _ jw%
API
e
WI
S t.M1M'{t.,I me i r
.Lb-
'1Ilk
Ilea
x lob
me
t _
{ �Y
F i
S
leple
Ile•Joe
le Vol
p
0 One,
pe
or
e Apo
or e Ile
Ar be
too
Ile 'YO
x y� t
ri;low � : h
r Ole
I Ie
el Iler mob s ,.
if1r�►PIlIb
s
oimb"_. `A tior
be b We A he
IS
AoI
the p

Y P
PAO to
to
ONO
It
y� M yp
to ei
o,t to
o
`!I

,�
l�
-
��
_

F.

Ilk
it
Jelk F° akk
• ,�,
IWb
a s x
- 'f �� i+..t i �►
'4 N .. f �i�. _
. V• i
17, ii,ok
all _
i iL
oil
40
Milk
Y
�'w %

ELI
It,
Ir
+x + y t mI
11FIII It. It
AL
It
-3'.'}'�+ ,z•` 4*

����yy....
41

'%
rim
t-
q 4
f

Fm
S r ;bib*
�si 3•;0 Win,..
4
'Po, V
WI
IYY
VV

a�TIPr
t
i
.os.Rn6u.: y.:.y

o,
Available online at www.sciencedirect.co
NON S0
SCIENCE DIRECT*
ELSEVIER Ecological Economics 49 (2004) 19-30
ANALYSIS
ECOLOGICAL
ECONOMICS
www.elsevier.com/locate/ecolecon
Contingent valuation, net marginal benefits, and the scale of
riparian ecosystem restoration
Thomas P. Holmes a,,*, John C. Bergstrom b, Eric Huszare, Susan B. Kaskd, Fritz Orr IIIe
a USDA Forest Service, Southern Research Station, P.O. Box 12254, Research Triangle Park, NC 27701, USA
bDepartment of Agricultural and Applied Economics, The University of Georgia, 208 Conner Hall, Athens, GA 30602, USA
USDA Animal and Plant Health Inspection Service, Policy Analysis and Development, 4700 River Road Unit 119,
Riverdale, MD 20737-1238, USA
d Warren Wilson College, CPO 6125, PO Box 9000, Asheville, NC 28815, USA
eRabun Gap-Nacoochee School, 339 Nacoochee Drive, Rabun Gap, GA 30568, USA
Received 17 November 2002; received in revised form 5 September 2003; accepted 20 October 2003
Available online 8 April 2004
Abstract
A study was undertaken to estimate the benefits and costs of riparian restoration projects along the Little Tennessee River in
western North Carolina. Restoration benefits were described in terms of five indicators of ecosystem services: abundance of
game fish, water clarity, wildlife habitat, allowable water uses, and ecosystem naturalness. A sequence of dichotomous choice
contingent valuation questions were presented to local residents to assess household willingness to pay increased county sales
taxes for differing amounts of riparian restoration. Results showed that the benefits of ecosystem restoration were a non-linear
function of restoration scale and the benefits of full restoration were super -additive. We estimated the costs of riparian
restoration activities by collecting and analyzing data from 35 projects in the study area. After adjusting our estimated valuation
function for socio-economic characteristics of the local population, the benefit/cost ratio for riparian restoration ranged from
4.03 (for 2 miles of restoration) to 15.65 (for 6 miles of restoration). Riparian restoration in this watershed is therefore an
economically feasible investment of public funds at all measured spatial scales.
© 2004 Elsevier B.V. All rights reserved.
Keywords: Riparian restoration; Contingent valuation; Super -additivity; Complements in valuation
1. The need for marginal economic analysis of
ecosystem restoration
Ecological systems provide an array of benefits to
humans that are not generally accounted for in
* Corresponding author. Tel.: [phone]; fax: +1-919-
549-4047.
E-mail address: tholmesgfs.fed.us (T.P. Holmes).
0921-8009/$ - see front matter © 2004 Elsevier B.V. All rights reserved.
doi: 10. 10 1 6/j.ecolecon.2003.10.015
market transactions. Consequently, economic activi-
ties can degrade ecological systems and valuable
ecosystem services may be underprovided or entirely
lost. If ecosystems are resilient to changes caused by
degradation, it may be possible to restore ecosystem
services either to some pre-existing level or to a
level that is commensurate with the demands of the
current human population. However, because eco-
system restoration is still highly experimental and

20 T.P. Holmes e2 al. / Ecological Economics 49 (2004) 19-30
can be quite costly to implement, it is not immedi-
ately obvious which ecosystems deserve priority for
restoration or, once specific ecosystems are chosen
for restoration projects, just how far restoration
activities should proceed.
During the past decade, the federal government has
become increasingly concerned with protecting eco-
system integrity and federal funds have been provided
for ecosystem restoration activities. Although policy
decisions might be made on the basis of noneconom-
ic as well as economic criteria, analysis of the benefits
and costs of restoration projects provides policy
makers with information by which they can gauge
the efficiency of public investments. The costing of
restoration activities is conceptually straightforward,
although the acquisition and analysis of such data is
complicated by its relative scarcity. On the benefits
side of the ledger, multiple human services are pro-
vided by riparian ecosystems, and the value of many
services can only be measured using non-market
valuation methods.
The existence of multiple, interconnected, non-
market ecological services presents significant chal-
lenges to researchers seeking unbiased estimates of
ecosystem values (Turner et al. 2002). Macroeco-
nomic analysis of the net benefits associated with
marginal changes in ecosystem services, from a
clearly defined baseline to a new condition resulting
from the imposition o
ceptual foundation for
et al. 2002). However, ecosystem services are
intrinsically connected and conventional valuation
methods might produce piecemeal, incomplete esti-
mates of the benefits of restoration (Bockstael et al.
2000).
The objectives of this study were to develop and
test a general methodology for valuing the restora-
tion of a set of ecosystem services and to compare
the economic benefit of riparian ecosystem restora-
tion with its cost. Based on detailed conversations
with stream ecologists and local citizens, a set of
relevant ecosystem services associated with riparian
restoration was identified. Duriig these conversa-
tions, it became apparent that the scale of restoration
could be used as a summary measure for the
provision of ecosystem services. That is, because
the overall biological condition of a river basin is
negatively linked to the degree of human influence
(Karr and Chu 1999), the scale of restoration activ-
ities that mitigate human impacts provides a conve-
nient meta-indicator that can be linked with the
supply of ecosystem services. A contingent valuation
survey was developed to estimate the benefits asso-
ciated with the provision of different levels of
ecosystem services, and data provided by the US
Natural Resources Conservation Service were used
to analyze the cost of restoration projects in the
study watershed. This process allowed us to com-
pare the costs and benefits of watershed restoration
at different spatial scales.
2. History of the Little Tennessee River ecosystem
fn this paper, we conduct a benefit-cost analysis of
restoration activities along the Little Tennessee River
(hereafter LTR) located in the southern Appalachian
Mountains. The Little Tennessee River (LTR) origi-
nates in Rabun County, GA; it flows north into North
Carolina before terminating at Fontana Darn, just
south of the Great Smoky Mountains. The LTR basin
contains about. 100,000 ha of mountainous terrain of
which 49% is part of the Nantahala National Forest,
37% is in privately held forest, and the remainder
(14%) is developed.
Historically, the LTR watershed was within the
f a policy; provides a con- homeland of the Cherokee Nation. After European
empirical analysis (Balmford settlement, the region supported logging, agriculture
and mining industries. During the late 1940s, the
Tennessee Valley Authority began to address the
sediment loads in the LTR and grasses were planted
on steep slopes to reduce soil erosion. Subsequently,
land use shifted as fanners began increasing livestock
production and many farmers cleared their land to-
ward the riverbank to maximize output.
Most recently, tourism, recreation and the draw of
living in an aesthetically pleasing environment has led
to rapid population growth and an iicrease ii the
number of people who visit the area. In the last 20
years, the population in Macon County, NC (our study
area), has doubled, leading to concerns about the
future health of the watershed and the ecosystem
services the watershed provides.
The majority (51%) of land within the watershed is
privately owned and private land use decisions have
had a major impact on ecosystem structure and
T.P. Holmes et al. /Ecological Economics 49 (2004) 19-30
function (Wear and Bolstad 1998). Non-point pollu-
tion from agricultural activities (such as watering
cattle in streams) and development (housing and
commercial development along streams and creeks)
threaten the ecological integrity of the watershed.
Economic activities have introduced increased levels
of sediment, nutrients, fecal coliform bacteria, toxic
chemicals, oil, grease, and road salt into the river
system.
3. Prior riparian restoration activities and costs in
the LTR watershed
A restoration program for the LTR watershed was
initiated in 1995 and 59 projects had been completed
by 2001. A total of 54 projects have set aside 45,118
ft, or 8.5 miles, of riparian buffer. This activity
consists of planting trees and grasses to stabilize the
riverbank. On 14 projects, fences were installed to
prohibit livestock from entering the river. And on five
projects, alternative water systems were developed for
watering livestock.
Only 35 of the riparian buffer projects had suffi-
cient cost data available to estimate project costs.' We
estimated that riparian buffers without fencing cost,
on average, $0.98/ft (based on data from 29 projects).
With fencing, average costs were $3.1341 (based on
data from 6 projects).
Another restoration activity in the LTR watershed
involved rebuilding eroding stream banks with
revetments. Revetments consist of large tree
branches or logs that are anchored to the stream
bank with cables. Of the 54 projects, 45 landowners
restored 15,321 ft (or 2.90 miles) of stream bank
using revetments; 34 landowners used trees from
their own property and the other projects brought in
trees from off-site. Revetments are typically quite
costly to construct. The average cost of revetments
where on-site trees were available for construction
was $15.50/ft. If on-site trees were not available,
the average cost of revetments was $20.33/ft.
To permit comparison of the costs and benefits of
restoration, it was necessary to make some assump-
tions about the typical mix of restoration activities to
construct a representative scenario. Based on docu-
1 All costs and benefits are standardized to the year 2000.
21
mented restoration data, it was assumed that, for
every mile of riparian buffer that, is established, 0.34
miles of revetment, is installed (2.9 miles of revet-
ment/8.5 miles of buffer). Using a weighted average
cost estimate (on-site trees and without on-site trees)
of $16.371ft for constructing revetments, this trans-
lates into $5.56 per "representative" foot of resto-
ration ($16.37*0.34). Next, it was assumed that
fencing is installed for 46% of the length of riparian
buffers (46°X0 of the total length of riparian buffers
was fenced in our project data). The average cost of
establishing a riparian buffer in a "representative"
restoration would then be $2.06/ft. The average cost
per foot establishing a representative mix of resto-
ration activities would be $7.62,1ft. (calculated as the
sum $5.56+$2.06).
Cost sharing is provided through the Natural
Resources Conservation Service to landowners de-
siring to create riparian buffers (with or without
fencing) or install revetments on their land. The
NRCS program funds 75% of the cost while the
landowners must provide the other 25%. If land-
owners contribute their own trees to a revetment
project, then their cost share falls to 10% of that
project. The private benefit to landowners who
decide to enter into a project with the NRCS can
be presumed to equal or exceed the dollar amount of
their cost share agreement. Under the cost-share
program, then, landowners must pay 25% of the
cost, or $1.91/ft in our example. Thus, the public
benefits must equal or exceed 75% of the cost, or
$5.72/ft, for public investment in a representative
mix of restoration activities to be economically
feasible.
The upper LTR watershed is approximately 20
miles in length. Although approximately 8.5 miles
along the river have received restoration treatments,
many segments along the river still need to be
restored. In consultation with local experts, it was
determined that restoration of 6 additional miles of
river would constitute complete restoration (not all
stretches of the river require restoration).
4. Issues in the valuation of freshwater ecosystems
Economic valuation of ecosystems is complicated
by the fact that ecosystems are characterized by
22 T.P. Holmes e2 al. / Ecological Economics 49 (2004) 19-30
multiple, interdependent services that possibly ex-
hibit complex dynamics and discontinuous change
around critical thresholds (Limburg et al. 2002).
Faced with this complexity, marginal economic
valuation of ecosystems has typically proceeded
via simplification. fn a review of published research
on the valuation of freshwater ecosystems from
1971 to 1997 (30 studies), Nilson and Carpenter
(1999) found that most studies focused on a specific
indicator of water quality such as water clarity or
the fi-equency of noxious algal blooms. While these
studies have made important contributions by dem-
onstrating that freshwater ecosystems have econom-
ic value, particularly non-use value, they only
provide partial benefit estimates because they are
based on an incomplete list of potentially valuable
services.
Turner (1999) defines Total Economic Value
(TEV) as the sum of all use and non-use values
provided by an ecosystem.' One approach, then, is
to obtain values for each of the services provided by
ecosystem protection or restoration. Recognizing the
multi-dimensional nature of water quality, the US
Environmental Protection Agency developed a six-
dimensional characterization of the benefits provided
by freshwater systems (USEPA, 1994): aquatic life
support (providing habitat for fish and other aquatic
organisms), fish consumption (fish do not pose a
human health risk), drinking water supply (eater is
safe to drink with conventional treatment), primary
contact recreation-swimming (no adverse- health
effects from occasional contact), secondary contact
recreation (no adverse health effects from activities
such as canoeing), and agricultural use (suitable for
irrigation or watering livestock). Using this scheme,
states are requested to report the percentage of lakes,
rivers and streams that meet five levels of water
quality, ranging from "good, fully supporting" to
"poor, not attainable" for each of the water quality
dimensions.
Magat et al. (2000) developed a study for valu-
mg water quality based on a simplified version of
the EPA monitoring structure. Using the method of
paired comparisons, they found that swimmable
water quality accounted for the greatest proportion
of overall benefit, followed by quality of the aquatic
environment, and finally by fishable water quality.
They also noted that people were willing to pay a
disproportionately high premium for water quality
improvements in areas that they would never use,
suggesting that non-use values are an important
benefit provided by enhanced water quality.
An alternative, holistic approach to ecosystem
valuation was reported by Loomis et al. (2000),
who used the contingent valuation method to evaluate
the benefits of restoring a portion of the Platte River
watershed. The approach used in this paper described
the current level of provision of four ecosystem
services: dilution of wastewater, natural purification
of water, erosion control, and habitat for fish and
wildlife. Specific mechanisms for restoring ecosystem
services were then described, followed by a referen-
dum WTP question asking respondents whether or
not they would vote in favor of a specific restoration
program. Using estimated water leasing costs and
farmland easement costs necessary to implement the
program, benefit/cost ratios varied between 1.4:1 and
5.22:1 depending on whether those refusing to be
interviewed had a zero value or not.
Zhongmin et al. (2003) estimated the benefits and
costs of restoring ecosystem services in the Hei
River basin in China using a holistic approach to
valuation, similar to the method used by Loomis et
al. (2000). Five ecosystem services were listed that
ecosystem restoration could provide: control soil
erosion and reduce sand storms, provide habitat for
wildlife, natural purification of water, dilution of
wastewater, and limit land salinization. Results of
the in-person interviews indicated that over 90
percent of the respondents were willing to pay a
positive amount for ecosystem restoration. However,
the amount that the general public was willing to
pay was found to be substantially less than the
estimated costs of restoration.
The decision of whether to value a set of ecosystem
services holistically, as is done using the contingent
T.P. Holmes et al. /Ecological Economics 49 (2004) 19-30
valuation method, or whether to focus valuation on
trade-offs between specific services using an attribute-
based stated preference method, depends on the goals
of a study. If management actions can differentially
affect the provision of iri&,idual ecosystem services,
then information on value trade-offs between ecosys-
tem services can be estimated using attribute-based
methods (e.g., Holmes and Adamowicz 2003). How-
ever, if ecosystem services are highly correlated in
production, then contingent valuation is probably
more appropriate.3
5. Ecosystem valuation survey design
Hoehn et al. (2003) recognized that the economic
value of freshwater ecosystems is derived from the
services they provide, and stressed the importance of
linking ecosystem science with ecosystem services in
conducting stated choice experiments. For our study,
conferred with a team of economists conferred with
at team of ecologists from the USDA Forest Service
Coweeta Hydrologic Laboratory to discuss the set of
ecosystem services that have been impacted by land
uses in the LTR watershed and the particular resto-
ration activities that were being undertaken to ad-
dress riparian ecosystem degradation. In these
sessions, concern was expressed both about the
current agricultural practice of watering cattle in
the LTR and its tributaries and about the impact of
residential and commercial development along
streams in the river basin. Review and input on the
relationships between ecosystem services and resto-
ration activities in the LTR watershed were also
obtained in focus group sessions with ecologists in
the Institute of Ecology at the University of Georgia,
and through discussions with the Little Tennessee
River Association and members of the Macon Coun-
ty Soil and 'h'ater Conservation District.
Based on these conversations, several ecosystem
services (and indicator variables for each service)
' Attribute-based stated preference methods rely on indepen-
dent variation in attributes (e.g., ecosystem services) to estimate
attribute values. Correlated attributes introduce the statistical
problem of multicollinearity. For a discussion of this issue, see
Holmes and Adaniowicz (2003).
23
were identified: (1) habitat for fish (abundance of
game fish), (2) habitat for wildlife (wildlife habitat
in buffer zones), (3) erosion control and water
purification (water clarity), (4) recreational uses
(allowable water uses), and (5) ecosystem integrity
(index of ecosystem naturalness). Generalized cate-
gories representing the level of provision of each
indicator were assigned to represent low, moderate
or high levels of provision of these services. This
technique is a modification of the "good-poor"
categorical scale used by the USEPA (1994), and
was used to obviate problems associated with char-
acterizing an exact change in ecosystem services that
could be expected to obtain from the implementation
of specific riparian restoration activities.
Marginal ecosystem values may vary depending
on the scale (scope) of ecosystem restoration. If a
single restoration project is not effective in enhanc-
ing the overall level of ecosystem services, the
derived economic benefits will probably be low. In
contrast, the value of multiple projects that do in
fact enhance the overall provision of ecosystem
services may be greater than the sum of the benefits
provided by individual projects valued in isolation.
That is, if restoration projects are what Madden
(1991) calls R-complements, then benefits might
be "super-additive".4
This perspective is important to recognize because
restoration projects are often conducted piecemeal,
using the logic that some restoration is better than
none, and that it is important to "start somewhere"
with available funding. However, from an economic
perspective, it is possible that, for geographically
isolated projects, the costs of restoration exceed the
benefits. This result becomes increasingly likely if
restoration is expensive and if extensive restoration is
required to change the overall level of services pro-
vided by an ecosystem.
To test the hypothesis that program scale has an
impact on marginal economic benefits, it was neces-
sary to link indicators of ecosystem services with
" A useful discussion of additivity effects and scale (scope) can
be found in Hanemann (1994). In terms of the economic
terminology defining scale effects, variation of the scale of
restoration within-subjects is referred to as an internal scale (scope)
test.
24 T.P. Holmes et at. / Ecological Economics 49 (2044) 19-30
Table 1
Overview of hypothetical Little Tennessee River riparian restoration programs used in the iterative contingent valuation experiment
Current situation Program I Program 2 Program 3 Program 4
Indicator of No small streams All small streams All small streams All small streams All small streams
ecosystem protected by BMPs+ protected by BMPs+ protected by BMPs+ protected by BMPs+ protected by BMPs+
service no new river no new river 2 miles of new 4 miles of new river 6 miles of new river
restoration restoration river restoration restoration restoration
Gagne fish Low Low Low Low High
Water clarity Low Low Moderate Moderate High
Wildlife habitat Low Moderate Moderate High High
in buffer zones
Allowable Low Moderate Moderate Moderate High
water uses
Index of ecosystem Low Low Moderate High High
naturalness
scale (Table 1).5 Indicator levels were also provided
for the status quo scenario, and for a base program
consisting of best management practices at construc-
tion sites and along private roads in order to prevent
sediment from entering tributaries to the LTR.6
Respondents were asked to vote on four different
programs. In consultation with local experts, it was
determined that complete restoration could be accom-
plished by installing riparian restoration proj ects along
an additional 6 miles of the riverbank. A base program
mandating BIVIPs along tributaries of the LTR was
designated, along with three alternative levels of river
restoration (2, 4 and 6 miles of new restoration).
6. A computerized survey instrument
A computerized survey instrument was developed
to facilitate communication of information about the
sources of riparian ecosystem degradation in the LTR
' The scientific basis for linking ecosystem services with scale
ofrestoration is extremely limited and was therefore based on expert
judgment. We let the overall provision of ecosystem services
increase in a roughly linear fashion with scale up to 4 miles of
restoration. That is, letting "low"=1 point, "medium"=2 points
of restoration, the change in aggregate ecosystem services doubles
from the level provided at 4 miles of restoration.
6 BMPs include activities such as construction of drop
structures (e g., weirs) to minimize soil movement down slopes.
BIV1P activities would be paid for by the private sector.
watershed, the various riparian restoration and protec-
tion activities that could be implemented to address
the problem, and ecosystem services that would be
enhanced by the watershed programs. This format
allowed us to make extensive use of photographs
and diagrams depicting restoration activities. Land
use maps were included to depict land use change in
the study area, and showed the proximity of economic
development to the LTR and its tributaries. Although
the use of a computerized instrument may eliminate
the potential for bias that is sometimes induced by in-
person surveys, it might discriminate against people
who are less computer literate, such as older people or
people with lower incomes.
The computerized instrument provided us with the
opportunity to customize the bidding structure in the
iterative referendum voting scenarios. Bid amounts
for the 4-mile and 6-mile restoration programs were
conditional on the response to the prior referendum
question. A YES response to the 2-mile or 4-mile
restoration referendum questions led to a higher bid
amount for the subsequent program, whereas a NO
response led to an identical bid amount for the
subsequent program. All respondents were presented
with bids for each of the four programs. The initial
bids were randomly selected from the amounts $1, $5,
$10, $50 or $75. Bid amounts for the 6-mile restora-
tion program ranged from $1 (resulting from a string
of prior NO responses) to $500 (resulting from a
string of prior YES responses).
The valuation questions asked the respondent to
consider a vote to approve or reject specific man-
T.P. Holmes et al. /Ecological Economics 49 (2004) 19-30
Table 2
statistics for Macon County, North Carolina.
'?y
Data source Median Median Age Males per Bachelor's degree Property along
income ($) (years) 100 females or higher (00/0) LTR (%)
Sample 45,000 47 82 72 52
2000 Census 28,696 45.2 92.1 13.2 15'
'Based on the estimated number of households within 200 in of the LTR and its major tributaries using aerial photographs available at
"terrasewer-usa.com" and data on building density bi the LTR Basin (Wear and Bolstad 1998).
agement programs for the Little Tennessee River
watershed. The management program would be one
of the alternative riparian ecosystem programs
shown in Table 1. The scenarios stated that if the
respondent agreed to support the program, payment
would be collected through an increase in the local
(county) sales tax that would be implemented over
a 10-year period. It was also stated that a resto-
ration program would be implemented only if a
majority of county residents voted in favor of it.
Finally, respondents were asked to consider their
current expenses before answering the referendum
questions.
7. Citizen focus groups and valuation panels
Four focus group sessions were conducted in the
study area to facilitate the development of the com-
puterized survey instrument.8 Two major concerns
with the survey instrument emerged from the citizen
focus group sessions. First, some people found it
difficult to distinguish between the different programs,
and recommended presenting a matrix showing the
level of ecosystem services provided by each pro-
gram. This structure, which facilitates a comparison of
the ecosystem services across programs, was devel-
oped and used in the final survey (Table 1).9 In
addition, in order to familiarize respondents with the
different. programs, people were asked to rate, on a 1
' The 10-year time frame was established based on the
assumption that, once restoration activities were implemented, the
flow of enhanced ecosystem services would continue unimpaired
for 10 years.
S Focus group participants were provided with a $25 incentive
for their time.
9 For each WTP question, respondents were shown the level of
ecosystem services associated with the status quo and with the
restoration program they were voting on.
to 7 scale, each of the programs prior to being asked
the WTP questions.to
A second concern expressed in the focus groups
was the construct of our payment vehicle. Initially, we
included State income tax as the payment vehicle.
After concern was expressed by focus group partic-
ipants that State taxes should not be increased to
support a local initiative, we altered our payment
vehicle to represent an increase in the local sales
tax," It. was noted that the county had recently passed
an increase in the sales tax and that some people were
reluctant to vote for further tax increases.
The citizen valuation panel was a non-probability
sample made up of recruits from local civic organ-
izations. Although we did not use a formal "quota"
sample, where quotas are defined over specific socio-
economic variables, an attempt was made to recruit a
diverse set of citizens to make up the panel. 12 Then,
once the valuation function was estimated, population
values for stratification categories such age, gender,
and income were inserted in the valuation function to
predict WTP for the local population.
Each individual who participated in the final
survey received a $40 incentive payment. The sur-
vey panels were held in the study area using
10 Mean values for the desirability ratings for programs I
through 4 were 3.07, 4.01, 4.82 and 5. 715, respectively. These values
increase monotonically, indicating that people were sensitive to the
level of ecosystem services provided by each program.
11 In the southern U.S. where the study area is located, local
sales taxes must be approved in a public referendum and are a
common and familiar means of financing local public goods and
services.
12 Harrison and Lesley (1996) state "If the goal ofthe sampling
exercise is indeed to generate a good valuation function for the
purpose of predicting population responses, then it does not follow
that probability sampling is the best thing to do. Instead, one should
try to ensure sample variation in all of the explanatory variables that
will be used to predict the population mean, even if this means
generating a greater number of responses for certain stratification
categories than is found in the population" (p. 83).
26 T.P. Holmes et al. / Ecological Economics 49 (2004) 19-30
computer labs at Franklin High School and South-
western College. Ninety-six people completed the
computerized interviews (this represents about. 0.7%
of the households nn the County). A comparison of
socio-economic characteristics of the sample and the
County (based on the 2000 Census) showed that the
income and education of the sample were higher
than the values for the population (Table 2). This
result is not uncommon for probability samples. It
was found that the age and gender characteristics of
our sample were quite close to the population
values. Finally, our sample included a larger propor-
tion of people who owned property along the LTR
than occurs in the general population.
8. Statistical analysis
Binary responses to the referendum questions were
analyzed using a random utility model. For each of the
different programs shown in Table 1, respondents (i)
were asked if they would vote to support the LTR
watershed program at the stated bid amount. The
probability of voting YES can be expressed as
Pr[v(zJ, v - tj) + sysv(zY) + sio] (1)
where 0 is the normal cumulative distribution func-
tion, it is the parameter estimate on the log-bid
amount and a is either the estimated constant (if no
other explanatory variables are included in the equa-
tion) or the "grand" constant, which is computed as
the sum of the estimated constant plus the product of
the other explanatory variables times their mean
values.
Hanemann (1984) advocated the use of median
WIT as a measure of economic welfare. While the
mean WTP has been shown to be very sensitive to
small changes in the right tail of the WTP distribution,
the median is much more robust to these effects. 13 in
addition, median WTP indicates the amount at which
50% of the sample would vote for a particular
referendum. This is in keeping with our survey
structure, where we reminded people that the condi-
tions of a referendum would only take effect if at least
one-half of the population voted in favor of it.
Consequently, we use the median as a conservative
estimate of WTP. As shown by Hanemann and Kan-
ninen (1999) median WTP can be computed from the
parameter estimates in Eq (3):
WTPmedian = exp (4)
where v is indirect utility, j is a vector of ecosystem
services for program j, z° is a vector of ecosystem
services for the status quo, y is income, tj is the tax
payment for program j ands is a random error term.
Eq. (1) can be re-written as:
Pr[.w>_a,o - ey] = Pr[w>q]. (2)
If it is assumed that it is normally distributed, Eq. (2)
can be estimated using a probit model.
It is popular in the valuation literature to specify
the WTP function as lognormally distributed. Similar
to Bishop and Heberlein (1979), we used a logarith-
mic transformation of the bid amount in our statistical
model. This model, which constrains WTP to be non-
negative, can be shown to provide a utility-theoretic
estimate of WTP (Hanemann and Kanninen 1999). If
the random component of utility e is randomly dis-
tributed, and if n and WTP are lognormal, then the
probability of a YES vote is
Pr[vote yes] = 0(a - yln(bid)) (3)
By including socio-economic characteristics in the
model specification, WTP values for the sample can
be estimated using sample means to compute the
grand constant a in Eq. (4). Alternatively, WTP values
for the population can be estimated by computing the
grand mean using population values.
Ecosystem services may be a non-linear function of
the scale of restoration activities (A/11LES). Thus, in
order to estimate the response of WTP to changes in
restoration scale, a quadratic form was used (MILES 2).
Once the response surface is estimated, WTP for
varying degrees of restoration can be computed. This
is accomplished by adj ustnng the term representing the
product of the parameter estimates on MILES and
" One concern in dichotomous choice CV studies has been the
presence of "fat tails" in the WTP distribution. This effect might be
due to the propensity for some respondents to answer YES to a.
IA'TP question irrespective of the price. The identification of yea-
saying in dichotomous choice CV typically relies on split-samples
(Holmes and Kramer 1995. Use of median, rather than mean, WTP
acts to obviate this potential problem.
T.P. Holmes et al. /Ecological Economics 49 (2004) 19-30
MILES2 and the number of miles restored in the
computation of the grand constant.
The iterative sequence of valuation questions used
in our survey design suggested the use of a panel
model to conduct the analysis. In particular, an error-
components model was used to control for individual
effects that might persist across iterations of the
experiment and which contribute to the overall vari-
ance in responses. to an error-components model, the
error term is comprised of a permanent component a;
that captures idiosyncratic behavior of the individual i,
and a transitory random shock vij (llsiao 1986):
s? = as + V# (5)
The idea behind Eq. (5) is that two identical
individuals may differ systematically in their propen-
sity to choose identical policy options due to idiosyn-
cratic preferences. If the parameter a is treated as
randomly distributed across the population, a random
effects model can be estimated (Greene 2000). In this
model, an idiosyncratic component in the error term
introduces autocorrelation in the responses. The cor-
relation coefficient p is equal to the ratio of the
variance of the permanent component to overall
variance:
0?2
6)
P _ v2" + ffa
where, n1 dichotomous choice models, it is typically
assumed that o2, = 1. Thus, the value of p increases as
the variance of the idiosyncratic component increases
relative to the variance of the random component.
9. Statistical results
Standard and random effects versions of the statis-
tical model were estimated (Table 3). A likelihood
ratio test showed that the random effects model was
statistically superior to the standard probit model (X2
statistic= 58.41, significant at > 0.01 level). The
correlation coefficient (p) in the random effects model
was significantly different than zero at greater than the
0.01 level and the magnitude of p suggests that
preferences among respondents were heterogeneous
after controlling for the effects of the explanatory
variables.
27
Table 3
Parameter estimates from simple and random effects probit models
of willingness to pay for local riparian restoration
Variables Standard probit Random effects probit
Constant - 1982 (1.444) -3.135 (3.774)
In(Bid) -0.199*** (0.054) -0.539*** (0.150)
MILES -0.283*** (0.107) -0.454*** (0.171)
MILESZ 0.060*** (0.017) 0.098*** (0.027)
In(Income) 0.178 (0.124) 0.322 (0.324)
Female 0.077 (0.149) 0.031 (0.395)
Age 0.011 * (0.006) 0.021 (0.016)
College 0.402** 0.624 (0.482)
Property -0.591*** (0.145) - 1.063*** (0385)
P - 0.641 *** (0.087)
log Likelihood -229.396 -200.192
McFadden R' 0.11 0.22
Observations 384 384
Standard errors in parentheses.
* Significance at the 0.10 level.
**Significance at the 0.05 level.
***Significance at the 0.01 level
The sales tax parameter In(BID) was negative and
significant at the 0.01 level in both regression models.
As anticipated, an increase in the sales tax amount
decreased the probability of voting YES for riparian
restoration. In the standard probit model, whether or
not the respondent had a COLLEGE degree was
positive and significant at the 0.02 level, AGE of
the respondent was positive and significant at the 0.07
level, and log(INCOME) was positive and significant
at the 0.15 level in explaining variation in the refer-
enda votes. The statistical significance of these vari-
ables decreased in the random effects model.
The parameter estimate on the variable indicating
whether or not respondents owned PROPERTY
along the LTR or its tributaries was negative and
significant at the 0.01 level in both model specifi-
cations. This result may reflect actual or anticipated
expenditures for riparian restoration by people living
along the LTR or its tributaries, or opportunity costs
associated with land use restrictions in riparian
buffers. Because restoration costs accrue to people
participating in restoration programs, their WTP for
new programs via higher sales taxes would presum-
ably be less than WTP by people not facing those
expenditures.
The scale of restoration, as measured by linear
and quadratic terms describing MILES of restora-
tion, was found to be statistically significant at the
28 T.P. Holmes et al. / Ecological Economics 49 (2004) 19-30
0.01 level in both regression models. The WTP
response surface, as a function of the scale of
restoration, was therefore non-linear (Table 4).
Respondents to our survey demonstrated a positive
WTP amount for the implementation of Best Man-
agement Practices along tributaries to the LTR. A
downward shift in WTP was observed for the
program that would restore 2 additional miles of
river (in addition to the BIVIP program). This finding
is consistent with comments in the focus groups of
some people who declared they favored total resto-
ration of the watershed and disliked the idea of
piecemeal, partial restoration. A small marginal
increase in benefits was observed for the 4 mile
restoration program. Notably, a very large increase
in marginal benefits was observed for the 6-mile
(total) restoration program, and suggests that riparian
restoration projects are super-additive in valuation.
14
Table 4
Annual economic benefits (median WTP), calculated at sample and
population means, for riparian restoration in the Little Tennessee
River watershed
I4lodel used Partial Partial Partial Full
for calculation" program program program program
benefit (BMP (BMP+2 (BMP+4 (BMP+6
category only) [$J miles) [$1 miles) [$] miles) [$J
Probst
Per household 5.66 1.09 2.30 53.76
benefits,
sample means
Per household 3.62 0.69 1.47 34.34
benefits,
population
means
County benefits, 72,608 13,954 29,551 689,652
sample means
County benefits, 46,375 8912 18,875 440,486
population
means
10. Comparing benefits and costs of riparian
ecosystem restoration
Annual median WTP values were estimated using
the values for the socio-economic variables comput-
ed fiom our sample and using the population values
for Macon County as reported in the 2000 Census
(Table 4). In both statistical models, WTP values
estimated using Census data were less than WTP
values estimated using sample means. Population
adjusted valuation functions derived from the statis-
tically superior random effects model were used to
compare ecosystem restoration benefits and costs.
Using the estimates reported in Table 4, present
values for a 10-year stream of annual benefits were
14 The proportions of people voting for programs I through 4
were 0.28, 0.34, 0.28, and 0.55, respectively. Note that the
proportion of people voting for the 6-mile restoration program
(0.55) was nearly double the proportion voting for the 4-mile
restoration program (0.28). If benefits were linear in the level of
ecosystem services, WTP should double between the 4-mile and 6-
mile programs (see footnote 5) However, as is shown in (Table 4),
per household WTP more than doubles-a result that is consistent
with super-additivity. For example, household WTP computed using
the statistically superior random effe is probit model, and using
population (rather than sample) means, increases from $4.42 to
$27.26-approximately a 6-fold increase. Thus, WTP is super-
additive in both miles of restoration and in ecosystem services.
Random effects probit
Per household 8.97 3.48 5.73 40.89
benefits,
sample
means
Per household 6.91 2.68 4.42 27.26
benefits,
population
means
County benefits, 115,092 44,672 73,539 524,559
sample means
County benefits, 88,659 34,412 56,649 349,705
population
means
computed." The present value of public benefits
generated by full restoration (BMPs plus 6 miles of
riparian restoration) was estimated to be $2,835,373.
This is equivalent to benefits of $89.50/ft ($472,560/
mile) of restoration, or a benefit./cost ratio of 15.65
(recall that the public cost associated with a repre-
sentative mix of activities was estimated to be $5.72/
ft). The present value of benefits generated by BMPs
plus 2 miles (or 4 miles) of restoration was estimat-
ed to be $243,732 (or $401,645). This translates into
benefits of $23.08/ft of restoration (or $19.02/ft of
" Recall that people were asked to vote on programs that
would increase local sales tax by a given amount for the next 10
years. The discount rate used in the calculations was 0.05.
T.P. Holmes et al. /Ecological Economics 49 (2004) 19-30
restoration), leading to benefit/cost ratios of 4.03 and
3.33, respectively.
The range of benefiticost ratios in our study,
3.33-15.65, spans the estimate of 5.22 reported by
Loomis et al. (2000) for restoring a 45-mile section
of the Platte River. Although the household benefits
of restoring the Platte River ($252/household/year)
were larger than the household benefits of restoring
the Little Tennessee River, the household benefits
per mile were quite similar ($5.604household/mile for
Platte River restoration vs. $4.54,-'household/mile for
full restoration of the LTR). The relatively high
benefit/cost ratio estimated for fall restoration of
the LTR relative to the Platte River, therefore,
appears to be due to the fact that the public share
of restoration costs per mile were considerably lower
for the LTR ($30,202/mile) relative to the Platte
River ($298,4441/mile). 16
11. Conclusions
Scale is an important factor in conducting benefit/
cost analyses of ecosystem restoration projects. In this
study, respondents were willing to pay a premium for
total restoration of the LTR ecosystem relative to more
modest restoration levels, and the benefits of ecosys-
tem restoration were super-additive in the sense that
the value of total restoration was greater than the sure
of benefits measured for partial restoration programs.
In turn, this result showed a strong preference for
programs that fully restored the level of ecosystem
services.
Of particular interest, it was found that the
benefits of partial restoration projects exceeded their
costs. Thus, the philosophy held by some stream
ecologists that partial restoration should proceed
with available funds even if fanding is not available
for total restoration proved to be economically
feasible in this case. This result is partially due to
the relatively low costs associated with ecosystem
restoration in this watershed.
Future research on the economics of ecosystem
restoration is clearly needed. Among the greatest
" The major restoration cost on the Platte River was the
potential purchase of conservation reserve program farmland
easements.
29
challenges facing ecological economists is the ability
to discern and articulate the linkages between ecosys-
tem science and the things that people value. In this
study, a carefully developed characterization of a set
of ecosystem services was developed, and ecosystem
services were linked with the scale of restoration.
Although this procedure facilitated the survey
respondents' understanding of the issues, much
remains to be done to improve methods for commu-
nicating complex ecological dynamics in the context
of economic valuation studies.
Although the results here showed that respondents
were sensitive to the (internal) scale of ecosystem
restoration in a sequence of valuation questions, more
rigorous (external) scale tests could be conducted by
eliciting and comparing WTP values for different
subsets of respondents faced with restoration choices
at different spatial scales. In addition, it would be
useful to investigate how WTP is affected by the
number of restoration programs. Because the research
reported here was based on restoration projects for a
single watershed, it is not clear how the value of
restoring a particular watershed might, be influenced
by the restoration of other ecosystems. It is possible
that different ecosystems are valued as complements
or substitutes, although very little is known about
value interactions in ecosystem studies. 17
Finally, human populations living in many differ-
ent and diverse watersheds may benefit from riparian
restoration activities. Future research needs to be
conducted to discern within which watershed restora-
tion activities could be justified using a benefit/cost
criterion and what scale of restoration provides the
greatest net social benefits.
Acknowledgements
This research was made possible by joint funds
provided by the US Environmental Protection Agen-
cy; the USDA Forest Service Southern Appalachian
Ecosystem Management Project, and the Georgia
Agricultural Experiinent Station. The authors would
like to acknowledge the support and advice provided
" For example, Hailu et al. (2000) found that multiple
programs for protecting old-growth forests, prairie grasslands and
mountain streams were perceived as complements by respondents.
30 T.P Holmes ,t ar / F, ological rr ononnes 49 12904) 19-30
by Jim Vose (Coweeta Hydrological Laboratory;
USDA Forest Service), Dave Louis (Bent Creek
Experimental Forest, USDA Forest Service), Chis
Geron (U.S. E.P.A.), Doug Johnson {Natural Resour-
ces Conservation Service), Jamtc Johnson (Little
J'elmessee River Watershed Association), and John
Loomis (Colorado State University).
References
Balmtord, A., Prmcp A, Coopen R, Costanza, k , Farber, S., GIC(ol,
RE,, Jenkins, M, Jelleriss, P., Iessamy, V, Madden, I, Munro,
K., Myers, N, Naeem, S., Paavola, L, Rayment, M, Rossendo,
S, Roup_hgarlen, I., Tnunpev, K, Tamer R.K., 2692. Econom-
ic reasons for conserving wild ranee. Science 297, 950-953.
Bishop, R-C, Heherlein, 'I'A, 19^9. Measuring values ofextra-
marketgoods: are indirect measures biased? Am. 7. Agrie. Econ.
61. 926--930.
Bockstael, N.E., Freeman HE A.M., Kopp, R.I., Portney, P.R.,
Smith, V.K. 2009. On measuring economic values for nature.
Emviron. Sci. Technol. 3z 1384-1389.
t rccne, W, 2000. Econometric Analysis. Pnmticc-Hall, Upper Sad-
cte River, Ni.
Haile, A., Adamowicz, W.L., Boxall, P.C., 2000. Commlennents,
suhstihaes, b_tdget coxstrairls and valuation. Erotica. Resour.
Econ. 16, 51--68.
Hanemann, W AT, 1984. W dfsre evaluations in contingent valua-
tion experiments with discrete responses. Am. I. Agrie,. Econ.
66, 33.2-1,11,
Hanemann, W.M., 1994. Valuing tine. environment through contin-
gent valuation. I. Econ. Perspeet. 8, 19-43.
Hanemann, WM, Karminen, B, 1999. The statistical analysis of
disc etc-response CV data In: Bateman, [ 1 , Willis, K G (Eds.),
Mdum-' Environmental Preferences: Theory and Practice of the
Contingent Valuation Method in the U.S, E.U, and Developing
Countries. Oxford Univ. Press. Now York.
Harrison, G.W, Leslie, J.C, 1996. Must contingent valuation sur-
veys cost so much? L Environ. Econ. Manag. 31, 79-95.
Hoe}m:, J.P., Lapi, E, Kanlouitz, M.D., 2003. Untying a Lancashi-
an bundle: valung, ecosystems and ecosystem services fm' wet-
land mitigation. I. Environ. Manag 68, 263-272.
Holmes, TP, Adamowicz, W.L, 2003. Atamate--based methods. In
Champ, P.A., Boyle, Kl, Brown, T.C. (Eds.), A Prirner on Non-
market Valuation. Kluwer Acadeanic PuLilishing, Dordrecht.
HoImes, 'I'.V, Kramer; RA, 1995. An independent sample test of
yea-saying and starting point bias in dichotomous-choice con-
tingent valuation, 7. Env nun. Aeon. Manag. 28, 121-132.
Hsiao, C., 1986. Analysis of Panel Data. Carnmidge Univ Press,
New York.
Karr, JR., Chu, E.W, 1999 Restoring Life in Ranni ng Waters
Better Biological Monitoring. Island Press, Washington, DC.
Limburg, K.E., O'Neill, R. V, Costa va, R, Farber, S, 2002. Co n-
plex systems and valuation. pool. Econ. 41, 409-420.
Loomis, I, Kent, P, Strange, L., Faasch, K, Covieh, A, 2900.
Measuring the. total econonnic value of restoring ecosystem serv-
ices iu an unpaired river hash., resells form a cortmboR valu-
ation survey. Fool. Econ. 33, 103-117.
Madden, P, 1991, A generalima(m of Hiclsian q suhstitares and
complements with application to demand rationing. Econometi-
ica 59, 1497-1508.
Magat, W, Huber, J, Viseasi, WK., Bell, I, 2000. An itmative
choice approach to valuing clean lakes, rivers, and strearns.
I. Risk iinceraor.. 21, 7-43.
'Miner; R-k, 1999.11e place of economic values in environmental
valuation. In Bateman, IS, Willis, K G (Ads ), Valung' Envi-
mmnental Preferences. Theory and Practice of the Contingent
Valuation Method in the U.S., E.U., and Developing Courhies.
Oxford Univ. Press, New York.
Turner; R.K., Paavola, L, Come} P, Farb-r, S, Iessamy; V, Gem'-
giou, S., 2002. Valuing Nature-Lessons Learned and Future
Research Directions. CSFRGE Working Paper FEW 02-051
University of East Anglia, Norwich, UK.
United States Enviromnertal Protection Agency, 1994. National
Water Quality Inventory, 1992 Report to Congress, EPA 841-
R-94-001.
Wear, D.N, BoNtad, P., 1998. Land-use changes in Southern Ap-
palachian landscapes: spatial analysis and forecast c-valuation.
Ecosystems 1 `75-594.
Wnlsor:, M.2, Carpenter, S.R., 1999. Economic valuation of fresh-
water eeoevetem services in the United States: 1971-1947.
Each Anph 9, 772-744.
Lhongmin, X. C oodong C, 71iligiang Z, hhiyong S, I oomfs, L,
2003. Applying contingent valuation in China to measure the
total economic value of restoring ecosystem services in Ajina
region. Ecol. Econ. 44, 345-358.