&EPA
United States	National Health and Environmental EPA/600/R-98/181
Environmental	Effects Laboratory	August 1999
Protection Agency	Corvallis, OR 97333
Development and Application of
Assessment Protocols for Determining
the Ecological Condition of Wetlands
in the Juniata River Watershed
Environmental Monitoring and
Assessment Program

-------
%

-------
EPA/600/R-98/181
August 1999
Development and Application of Assessment Protocols
for Determining the Ecological Condition of Wetlands
in the Juniata River Watershed
by:
Robert P. Brooks, Denice Heller Wardrop, and Jennifer K. Perot
Penn State Cooperative Wetlands Center
The Pennsylvania State University
i	Environmental Resources Research Institute
University Park, PA 16802
Agreement Number: CR826662-01-0
Project Officer:
Mary E. Kentula
Western Ecology Division
National Health and Environmental Effects Laboratory
Corvallis, OR 97333
National Health and Environmental Effects Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711

-------
DISCLAIMER
This project has been funded by the U.S. Environmental Protection Agency and
conducted through assistance agreement number CR826662-01-0 to the Penn State Cooperative
Wetlands Center. This document has been subjected to the Agency's peer and administrative
review and approved for publication. The official endorsement of the Agency should not be
inferred. Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
This document should be cited as:
Brooks, Robert P., Denice Heller Wardrop, and Jennifer K. Perot. 1999. Development and
Application of Assessment Protocols for Determining the Ecological Condition of Wetlands in
the Juniata Watershed. EPA/600/R-98/181. U.S. Environmental Protection Agency, Western
Division, National Health and Environmental Effects Laboratory, Corvallis, Oregon.
11

-------
This study will contribute to the development of a means to accurately, efficiently, and fairly assess a wetland's
condition in the context of the surrounding watershed that can then be used to implement protective and restorative
strategies that are appropriate for both the individual wetland and the watershed. This has been one of the primary goals
of research and outreach efforts conducted by the Penn State Cooperative Wetlands Center (CWC) since 1993, and will
guide their approach to monitoring and assessing wetlands in the Juniata watershed in central Pennsylvania. The
objectives for the study are:
1)	To determine and report on the ecological condition of wetlands in the Juniata River watershed using a series of
assessment tools.
a)	Develop a preliminary assessment of wetland abundance on two sub-watersheds in the Juniata River
watershed. Our experience with applying NWI digital data and other remotely-sensed data for inventorying
wetlands in the unglaciated portion of Pennsylvania has shown that these sources do not include the majority of
wetlands occurring in the watershed. To effectively sample wetlands in the Juniata, a better estimate of their
abundance and general location is necessary (i.e., a Level 1 inventory is not adequate). To help remedy this
situation, we are developing a process for deriving a best estimate of wetland acreage from a combined set of
GIS databases and a series of decision rules (Level 2 inventory). Acreage will be expressed as an estimate of
total wetland acreage in each subwatershed, with zones of high, moderate, and low probability of significant
wetland acreage identified on a map
b)	Verify and calibrate the inventory process on two subwatersheds in the Juniata, before the process is applied to
the entire watershed, including ground reconnaissance. During the reconnaissance, a cursory inspection of
wetland stressors will be performed, resulting in a preliminary indication of condition (Level 2 assessment).
c)	Conduct an inventory of wetland acreage and an assessment of condition for the entire Juniata River watershed.
i The inventory of the entire watershed will be based on the results of the work done to accomplish Objectives
la and lb. Condition will be expressed in terms of HGM functions and HGM type. For example, condition
might be expressed as: "Thirty percent of depressional wetlands in the Juniata watershed are exhibiting only a
moderate degradation of the long-term storage of surface water function." Condition will be assessed by
applying the HGM functional assessment models at a set of wetlands selected by probability-based sampling.
The verified inventory and map of acreage zones, and application of HGM functional assessment models
constitute a Level 3 assessment.
2)	Evaluate the feasibility of integrating a series of bioindicators into the wetland condition assessments for the two
sub-watersheds.
3)	Evaluate the feasibility of using citizen volunteers to apply the wetland monitoring protocols throughout the Juniata
River watershed.
Mary Kentula
541-754-4478

-------
TABLE OF CONTENTS
LIST OF FIGURES	iv
LIST OF TABLES	iv
BACKGROUND	1
PROJECT DESCRIPTION	10
WETLAND INVENTORY	10
ASSESSMENT CONDITION	14
OBJECTIVES	15
APPROACH AND METHODS	17
USE OF REFERENCE WETLANDS			17
< .	OBJECTIVE 1A. DEVELOP PRELIMINARY ESTIMATE OF WETLAND
ABUNDANCE ON TWO SUB-WATERSHEDS USING GIS	18
OBJECTIVE IB. VERIFY AND CALIBRATE THE PRELIMINARY
ESTIMATE OF WETLAND ABUNDANCE FOR THE TWO SUB-
WATERSHEDS	19
OBJECTIVE 1C. CONDUCT AN ASSESSMENT OF WETLAND
ABUNDANCE AND CONDITION IN THE ENTIRE
JUNIATA WATERSHED	20
OBJECTIVE 2. EVALUATE THE FEASIBILITY OF USING BIOINDICATORS
IN ASSESSING CONDITION	21
Plant Community Assessment	21
Avian Community and Landscape Pattern Assessment	23
Macroinvertebrate Community Assessment	24
OBJECTIVE 3. EVALUATE THE FEASIBILITY OF WORKING WITH
CITIZEN VOLUNTEERS	25
iii

-------
GENERAL PROJECT INFORMATION	26
PERSONNEL ASSIGNMENTS	26
TIMETABLE AND PRODUCTS FOR THE PROPOSED WORK	27
LITERATURE CITED	28
LIST OF FIGURES
Figure 1. W3ATER-Wetlands, Wildlife, and Watershed Assessment Techniques
for Evaluation and Restoration	3
Figure 2. Key for hydrogeomophic classification of wetlands into classes
and subclasses in Pennsylvania (Cole et al. 1997). Underlined items are
HGM subclasses	5 & 6
Figure 3. Integration of wetland inventory, assessment, and restoration	12
LIST OF TABLES
Table 1. Reference wetlands sampled in 1993 (1-22), 1994 (23-38), 1995 (39-51),
1997 (52-63, and 1998 (64-70)	2
Table 2. Variable and Functional Assessment Models in Development	7 & 8 & 9
Table 3. PCM Structue	11

-------
BACKGROUND
All wetlands are not equal in their ecological functions or societal values, thus, we should
not treat them as such. If we do, the result will be mediocrity in the way we protect wetland
resources overall. A means is needed to accurately, efficiently, and fairly assess a wetland's
condition in the context of the surrounding watershed, and then use that assessment to implement
protective and restorative strategies that are appropriate for both the individual wetland and the
watershed. This has been one of the primary goals of research and outreach efforts conducted by
the Penn State Cooperative Wetlands Center (CWC) since 1993, and will guide our approach to
monitoring and assessing wetlands in the Juniata watershed.
From 1993 to the present, the CWC has studied representative wetlands across the
Commonwealth of Pennsylvania (Table 1). The goal of the original research project was to
develop and evaluate a series of tools to be used by regulatory and non-regulatory staff to assess
wetlands by characterizing their current conditions, potential functions, and restoration potential
ih a watershed context. This was accomplished and the results are briefly summarized below.
The work from 1993-1996 was conducted primarily under Service Purchase Contract #275178
from the Pennsylvania Department of Environmental Protection (PADEP) and Federal Contract
#CD993282-01 from the U.S. Environmental Protection Agency (USEPA), Region HI. The
work on reference wetlands is continuing under a Water and Watersheds contract through
NSF/USEPA and a State Wetlands Protection Grant through PADEP and USEPA-Region IH.
These assessment tools have direct applications to this study of the Juniata River watershed. A
list of the assessment tools relevant to the proposed project is provided below. Their integration
with new approaches that will be developed during the current work is outlined later in this
research plan:
•	Developed W3ATER, a watershed assessment approach for application throughout
Pennsylvania and surrounding states (Figure 1).
•	Developed a Hydrogeomorphic (HGM) Classification Key for Pennsylvania's inland
freshwater wetlands (Figure 2, Cole et al. 1997).
1

-------
Table 1. Reference wetlands sampled in 1993 (1-22), 1994 (23-38), 1995 (39-51),
1997 (52-63), and 1998 (64-70).
SITE#
SITE NAME
SITE#
SITE NAME
1
BESP - PFO
36
Decker Pond
2
BESP-PEM
37
Peck's Pond
3
Bald Eagle Creek
38
Twin Ponds-PGC
4
LFC Dam
39
Little Sewickley Creek
5
McCall Dam
40
Little Sewickley Creek 2
6
Sand Spring
41
North Park
7*
Canoe Creek
42
Bar an Estates
8
Duncansville
43
1-80 SS
9
PSU Airport
44
Black Forest
10*
Whipple Dam SP
45
Spruce Swamp
11
Tofitrees
46
Long Pond PFO
12*
Mothersbaugh
47
Long Pond PEM
13
Clark's Trail
48
Mid State Upper
14
LFC - PFO
49
Brandywine Flood Plain
15
CPA Lumber
50
Mid State Middle
16
Old Greentown Rd
51
Donut Hole
17
Lakeville Hunt Clb
52
Tadpole
18
Buffalo Run
53
Nittany B&B Headwater Floodplain
19*
Rothrock St. For
54
Wardrop's
20
Black Moshannon
55*
Swamp White Oak
21
Marsh Creek-PEM
56
Farm 12
22
Marsh Creek-PFO
57
Thompson Run
23*
Shaver's Creek
58
Lock Haven
24*
McGuire Rd.
59
Nittany B&B Riparian Depression
25
Windy Hill Farms
60*
Laurel Run
26
Water Authority
61
Schneider Farm
27
WDC - Gaging Sta.
62
Flatbrookville
28
Millbrook Marsh
63
Shimer's Run
29
Colyer Lake
64
State College High School
30
PFBC - Spr. Creek
65*
Juniata Valley High School
31
Cedar Run
66
Tyrone Area High School
32
Fravel
67
Cumberland Valley High School
33
Lee's Gap
68
Nine Mile Run - Trailer
34*
Stone Valley
69
Nine Mile Run - Slope
35*
Davis
70
Bald Eagle Area High School
* = Reference wetland located in Juniata Watershed
2

-------
Figure 1. W3ATER—Wetlands, Wildlife, and Watershed Assessment Techniques for
Evaluation and Restoration
General Objective
No net reduction in ecological integrity and function of resources
1
r
Assess watershed to identify
potential areas
(proactive)
Identify specific problems in
the watershed on the ground
(reactive)
Evaluate permit
application in context
of watershed (reactive)
*
r
Select appropriate Level 1, 2 or 3
inventory, condition, and restoration protocol
5
Prioritize sections bv "triage" criteria:
Probable action level:
3. ecological integrity intact

a. protected
continue protection, no permit
b. not protected
seek protection, condition permit
2.. moderately disturbed

a. high restoration potential
seek restoration, condition permit
b. low restoration potential
postpone restoration, condition permit
1. severely disturbed

a. high restoration potential
seek restoration, grant permit
b. low restoration potential
postpone restoration, grant permit
Consider restoration and mitigation options based on risk assessment:
Assess probability of achieving predicted functional change based on,
a.	availability of technological solutions
b.	degree of reversibility
c.	short-term (days to months) vs. long-term (years to decades)
realization of results
Assess costs of no action or costs of implementing the project based on,
a.	threats to public health, safety, or welfare
b.	chronic degradation of ecological integrity
c.	likelihood of implementation based on volunteer, incentive, or
regulatory solutions and funding
d.	comparative economic costs among options
1
r
Implementation phase:

1. Notify cooperators and partners

2. Develop design and implementation plans
3. Secure financial resources and schedule actions
' i
r
Evaluation phase:

1. Compare observed outcome with predicted outcome
2. Compare restored condition with initial condition
3. Implement further action as needed

Repeat process iteratively as needed
3

-------
•	Established a set of 70 naturally occurring reference wetlands (Table 1) for long-term
studies and intensively monitored the set for use as a benchmark for wetland mitigation
designs and impact analyses (Brooks et al. 1996 and unpublished data). Reference wetlands,
as defined by the CWC, do not consist only of wetlands in a pristine or unimpacted condition.
Their use a benchmark in various types of studies has dictated that they span a range of
condition, ranging from pristine to heavily impacted. At this time, 11 reference wetlands
from this set are located in the Juniata watershed.
•	Completed the development of HGM Functional Assessment Models for four wetland
subclasses in the Ridge and Valley Province - headwater floodplain, mainstem floodplain,
riparian depression, slope - which are typical of the Juniata River watershed. Models were
peer-reviewed during a workshop in 1997 and are currently in the process of being calibrated.
Calibration of models for depressions, slopes, headwater and mainstem floodplains will be
completed by the end of 1999. It should be noted that calibration requires characterizing
wetlands across a condition gradient, i.e., both the best condition and the worst condition
' • must be sampled to determine the non-impacted and impacted endpoints of any variable
value. Table 2 provides a list of functions and their associated variables for the HGM
models. Development and calibration of models for remaining HGM types of importance
will be completed by 2000.
•	Developed a standard monitoring protocol for wetland studies (Brooks et al. 1996).
Recently, the protocol was modified into a Rapid Assessment Protocol suitable for use by
diverse groups such as agency personnel and high school students.
•	Completed Synoptic Watershed Maps and landscape analyses for four sample watersheds in
Pennsylvania, including Shaver's Creek within the Juniata watershed. Comparable work will
be done on at least two sub-basins in the Juniata watershed during 1999; one in cooperation
with PADEP wetland biologists, and one as part of a ecological/socioeconomic impact and
restoration study of acid mine drainage affected watershed (Aughwick Creek, Huntingdon
Co.).
4

-------
Figure 2. Key for hydrogeomorphic classification of wetlands into classes and subclasses in
Pennsylvania (Cole et al. 1997). Underlined items are HGM subclasses.
1. Wetland associated with a stream or river	 Floodplain or depression 2
1. Wetland not associated with a stream or river	 Fringing, slope, or depression 14
2. Wetland located within defined banks
or channel of stream or river	 Floodplain in-stream
2. Wetland does not occur within defined banks
or channel of stream or river) 	3
3. Equivalent stream order is 1st or 2nd order 	Floodplain, headwater (H) 4
3. Equivalent stream order is 3rd or larger	Floodplain, mainstem (M) 9
4. Wetland is impounded	 Headwater Impoundment (HI) 5
4. Wetland is not impounded 	6
5. Wetland impounded by beaver activities	 Beaver, HI
5. Wetland impounded by human activities	 Human, HI
6. Wetland has evidence of recent flooding	 Headwater floodplain
6. Wetland has no evidence of recent flooding			7
t	7. Wetland located on a topographic slope
with unidirectional flow of water	 Slope
7. Wetland located in a topographic depression	Depression, headwater (H) 8
8. Wetland located in a topographic
depression with discernable inlets or
outlets where primary source
is groundwater	 Riparian depression (H)
8. Wetland located in a topographic
depression with discernable inlets or
outlets and with organic soil 	 Organic depression (H)
8. Wetland located in a topographic
depression with discernable inlets
and outlets and where primary
sources of water are overland
flow or interflow 	 Surface water depression (H)
9. Wetland is impounded	 Mainstem impoundment (MI) 10
9. Wetland is not impounded 		11
10. Wetland impounded by beaver activities	 Beaver, MI
10. Wetland impounded by human activities	 Human, MI
11. Wetland has evidence of frequent flooding	 Mainstem floodplain
11. Wetland has no evidence of frequent flooding	 12
5

-------
Figure 2 (cont.). Key for hydrogeomorphic classification of wetlands into classes and subclasses in
Pennsylvania (Cole et al. 1997). Underlined items are HGM subclasses.
12. Wetland located on a topographic slope
with unidirectional flow of water	 Slope
12. Wetland located in a topographic
depression	 Depression, mainstem (M) 13
13. Wetland located in a topographic
depression with discernable inlets
or outlets and where primary source
is ground-water	 Riparian depression (M)
13. Wetland located in a topographic
depression with discernable inlets
or outlets and with organic soil	 Organic depression (M)
13. Wetland located in a topographic
depression with discernable inlets
or outlets and where primary
sources of water are overland
or interflow	 Surface water depression (M)
14. Wetland associated with a lake, reservoir, or large pond 	:	Fringing
14. Wetland not associated with a lake, reservoir, or large pond	 15
i
15. Wetland located on a topographic slope with
unidirectional flow of water	 Slope
15. Wetland located in a topographic depression
without discernable surface water
inlets or outlets	 Isolated depression (I) 16
16. Wetland located in a topographic
depression without discernable surface water
inlets or outlets and with organic soil	 Organic depression (I)
16. Wetland located in a topographic
depression without discernable surface water
inlets or outlets where primary sources of
water are overland flow or interflow	 Surface water depression (I)
6

-------
Table 2. Variable and Functional Assessment Models in Development
Variable
Variable
Acronym
HGM Model Function
Site Characteristics
Slope of wetland surface area
Vslope
Flbl,F6,F7
Macrotopographic relief
Vmacro
F2,F8b
Presence of outlets for macrotopographic depressions
within floodplain
Vmacro-out
F8
Microtopographic complexity of wetland surface
Vmicro
Fla,Flbl,F6,F7,F8,F9a
% cover of bare ground surface per standard area
Vbaregnd
F9a
Manning's roughness
Vroughness
Fla,Flbl,F7,F8
Representation of the general shape and orientation of
the wetland as it relates to the flow path
Vshape
Flbl
Estimated mean depth of standing water during
storage event
Vdepth
Flb2
Wetland surface area available for short term storage
Vsurface area
Fla,Flbl,Flb2
Ponded surface area available for short term storage
Vpondedsurface
area
Flb2
Above-ground volume available for storage
Vsurfacevolume
Flb2
Belpw-ground volume available for storage
Vsubsurface
volume
Flb2
Depth of restricted area
Vdepthra
Flb2
% total wetland surface area affected by physical
features such as culverts, ditches, etc.
Vdisturb
F9a
Presence of disturbance to groundwater flow or
discharge
Vgndwater
F4
Plant Community
Presence of each of four vertical strata: canopy,
sapling, shrub, & herbaceous
Vstrata
F9a
Dominant species by subclass or plant community
Vspcomp
F9b
Distribution of sizeclass values for all strata
Vsizeclass
F9a
Presence of propagules of dominant species in each
stratum
Vregen
F9b
Proportion of dominance of non-native species or
aggressive/invasive native species
Vexotic
F9b
Herbaceous Vegetation
% cover of persistent herbaceous vegetation per
standard area
Vperherb
Fla,Flbl,F5,F7,F8
Woody Vegetation
Basal area of standing wood per standard area
Vbtree
Fla,Flbl,F7,F8
Basal area of live standing wood per standard area
Vbtreelive
F5
Basal area of dead standing wood per standard area
Vsnags
F5,F8,F10
Density of standing wood per standard area
Vdtree
Fla,Flbl,F7,F8,F9a
7

-------
Table 2 (cont.). Variable and Functional Assessment Models in Development
Variable
Variable
Acronym
HGM Model Function
Woody Vegetation (cont.)
Density and sizeclass distribution of saplings per
standard area
Vsapling
F9a
Total volume of shrub cover per standard area
Vshrub
Fla,Flbl,F5,F7,F8
Amount of coarse woody debris per standard area
Vcwd
Fla,Flbl,F5,F7,F8,F10
Amount of fine woody debris per standard area
Vfwd
F5,F8,F10
Soil Characteristics
Depth of OE and 01 horizons
Vdepthoe,01
F10
Amount of soil organic matter
Vsorgm
F5,F6,F8
Presence of redoxymorphic concentrations in upper
part of soil profile
Vredox
F5
Soil texture
Vtex
F6
Permeability of the most restrictive layer present in
the upper meter
Vperm
F6,F7
Presence of evidence of anaerobic activity
Vanaerobic
F8
Landscape Characteristics
Categorical ranking of landscape characteristics
Vlandscape
F12
Width of buffer zone surrounding wetland separating
it from agricultural or developed land use
Vbuff
F12
% agricultural cover in 1-km radius circle
Vagcov
F12
Degree of aquatic connectivity in 1-km radius circle
Vaqcon
F12
% forest cover in 1-km radius circle
Vforcov
F12
% open water cover in 1-km radius circle
Vowcov
F12
% open urban cover in 1-km radius circle
Vurbcov
F12
8

-------
Table 2 (cont.). Variable and Functional Assessment Models in Development
Fla - ENERGY DISSIPATION/SHORT-TERM SURFACE WATER DETENTION
Applicable
Subclasses:	Headwater floodplain and mainstem floodplain	
Flbl - ENERGY DISSIPATION
Applicable
Subclasses:	Slope wetlands	
Flb2 - SHORT-TERM SURFACE WATER DETENTION/STORAGE
Applicable
Subclasses:	Slope wetlands	
F2 - LONG-TERM SURFACE WATER STORAGE
Applicable	Headwater floodplain, mainstem floodplain, and alluvial riparian
Subclasses:	depression (slope subclass)		
F4 - INTERCEPTION OF GROUNDWATER FLOW OR DISCHARGE
Applicable
Subclasses: 	Depressions and slopes (see discussion)	
F5 - CYCLING OF REDOX-SENSITIVE COMPOUNDS
Applicable	Depressions, headwater floodplain and mainstem floodplain, slope,
Subclasses:	and impoundments	
F6 - SOLUTE ADSORPTION CAPACITY
Applicable	Depressions, headwater floodplain and mainstem floodplain, slopes,
Subclasses:	and impoundments	
F7- RETENTION OF INORGANIC PARTICULATES
(assumes retention of organic matter considered elsewhere)
Applicable	Depressions, headwater floodplain and mainstem floodplain, slopes,
Subclasses:	and impoundments	
F8a - EXPORT OF ORGANIC PARTICULATES
F8b - EXPORT OF DISSOLVED ORGANIC MATTER
Applicable	Depressions, headwater floodplain and mainstem floodplain, slopes,
Subclasses:	and impoundments	
F9-F12 BIODIVERSITY/HABITAT FUNCTIONS
Applicable	Depressions, headwater floodplain and mainstem floodplain, slopes,
Subclasses:	and impoundments		
F9 PLANT COMMUNITY STRUCTURE AND COMPOSITION	
F10 DETRITUS		
F11 VERTEBRATE COMMUNITY STRUCTURE AND COMPOSITION	
F12 MAINTENANCE OF LANDSCAPE SCALE BIODIVERSITY
9

-------
•	Derived a set of Performance Criteria Matrices (PCMs) from studies of reference wetlands
for establishing reference standards on hydrology, soils, sediments, vegetation, and wildlife
habitat for mitigation projects and for assessing wetland condition. The PCMs describe
standard conditions in wetlands both by HGM type and by level of condition. The matrix
structure is illustrated in Table 3.
•	Developed a Wildlife Community Habitat Profile to facilitate wildlife assessments among
different wetland types based on habitat potential (Brooks and Prosser 1995).
PROJECT DESCRIPTION
For decades, a great deal of attention has been focused on the Chesapeake Bay ecosystem
and the threats to the ecological health of this valuable natural resource. The emphasis, however,
has been on the portions of the watershed nearest the estuary proper. Only recently have major
projects included the headwater regions of the ecosystem, such as the Juniata River watershed.
i Much of the CWC's work over the last five years has been to develop a cost-effective
approach to gathering and synthesizing information needed in wetland decision making. The
approach involves the integration of information to support three aspects of decision making-
inventory, assessment of condition, and determination of restoration potential (if applicable).
Inventory and assessment will be employed in this study and are described below. The entire
approach is illustrated in Figure 3. The aspects of decision making are considered sequentially,
and each step in the process involves a series of tools developed by the CWC that have been
tested in wetlands and watersheds across the state. Each step requires a different level of effort.
Whether one goes on to the next step in the process and to a greater level of effort depends on the
outcome of the previous effort and the quality of information required.
WETLAND INVENTORY
Assessment of watershed condition, based on wetland abundance and condition, in a
majority of the watersheds in the Northeast is impossible or inaccurate without a reliable
10

-------
Table 3. PCM Structure.
HGM Class
Disturbance Level
%Organic Matter 5cm
%SiIt 5cm
Isolated Depression
Pristine (n=5)
Moderate (n-2)
Severe (n=3)
55.5+/-30.4%
38.2+/-4.8%
8.2+/-2.8%
16.3+/-7.1%
18.2+/-2.8%
47.7+/-2.9%
Riparian Depression
Pristine (n=19)
Moderate (n=5)
25.4+/-15.6%
12.0+/-3.5%
37.4+/-12.2%
52.9+/-13.6%
Headwater Floodplain
Pristine (n=2)
Moderate (n=9)
Severe (n=8)
58.7+/-2.9%
8.7+/-2.3%
8.2+/-3.1%
32.4+/-11.4%
37.0+/-12.9%
48.3+/-20.5%
Mainstem Floodplain
Pristine (n=5)
Moderate (n=6)
Severe (n=10)
4.7+/-1.3%
6.9+1-0.6%
8.8+/-5.3%
27.1+/-14.3%
50.9+/-6.2%
37.0+/-20.2%
Slope
Pristine (n=27)
Moderate (n=16)
Severe (n=5)
27.2+/-23.5%
9.6+/-3.1%
9.9+/-1.7%
34.9+/-10.4%
35.6+/-9.1%
63.3+/-4.9%
Headwater
Impoundments
Pristine (n=23)
Moderate (n=4)
27.9+/-24.2%
10.8+/-3.4%
37.9+/-13.4%
51.1+/-10.9%
i

-------
Figure 3. Integration of wetland inventory, assessment, and restoration
INVENTORY	CONDITION
RESTORATION
LEVEL 1
LEVEL 2
LEVEL 3
Decision
Rules
Stressor
checklist
Calibrated HGM
functional assessment
v * models ^
Apply stressor checklist
Utilize existing resources (NWI)
Generate improved
inventory map
Map of abundance zones with
verified inventory
Performance criteria matrices
provide restoration standards
Map landuse in watershed;
calculate preliminary
landscape measures
Synoptic map of restoration
potential (existing wetlands,
landuse, roads & streams)
Develop and apply landscape-
based approach to obtain
abundance map
Map depicting abundance zones,
verified inventory, and probable
condition
Apply HGM functional
assessment models to
probability based sampling
locations
Map depicting overlay of
wetland abundance zones,
levels of potential threat,
and landuse, roads & streams
12

-------
inventory of wetland area. Wetlands in the unglaciated portion of Pennsylvania are believed to
encompass only 3-5% of the landscape and they are relatively small in area. Our experience has
shown that National Wetland Inventory (NWI) quads for Pennsylvania underestimate the
occurrence of wetlands by nearly 100%, and any wetland assessment of an entire watershed will
not include the majority of wetlands occurring in the watershed if based on NWI information. To
help remedy this situation, we are developing a best estimate of wetland occurrence derived from
a combined set of Geographic Information System (GIS) databases and a series of decision rules.
The inventory methodology is composed of three levels of effort and is described below.
LEVEL 1 - Gather best available mapped information from NWI and other sources.
LEVEL 2 - If NWI-based information is deemed to be insufficient, landscape-based decision
rules are developed to identify relative areas of high, moderate, and low probability of wetland
acreage. This process ultimately requires ground reconnaissance to locate and classify wetlands
from a probability-based sample, resulting in verification and calibration of the approach.
Calibration provides an estimate of total acreage of wetland area associated with each zone (high,
moderate, and low probability of wetland acreage), ultimately leading to an estimate of total
wetland acreage in the watershed. It is important to note that this procedure results in zones of
relative wetland acreage, with zones of high, moderate, and low probability of significant
wetland acreage displayed on a map; it does not locate individual wetlands over the entire
watershed. However, some specific wetlands are identified and mapped during the ground
reconnaissance, although the numbers of such wetlands are limited. Level II results in a map
with the following items: 1) all NWI wetlands, 2) zones of high, medium, and low probability
of wetland acreage, and 3) all "new" wetlands (i.e., those not indicated on the NWI map)
discovered during ground reconnaisance activities.
LEVEL 3 -Additional ground reconnaisance events may occur during a range of watershed-
related activities, not necessarily related to the construction of an inventory per se. For example,
during a condition assessment of the watershed, field activities may identify additional wetlands
not present on the Level II map. Any additional wetlands (i.e., those not indicated on the Level II
13

-------
map) discovered during these ground reconnaisance activities are therefore added to the Level II
map, and the resulting product is termed a Level IE map. A Level HI map may constantly evolve,
as additional wetlands are encountered and verified during ongoing watershed assessment and
planning activities.
ASSESSMENT OF CONDITION
A primary question is always, "What is the condition (eventually, level of impairment) of
a wetland?" The CWC has developed a triage approach to wetland assessment (see description
of W3ATER in Figure 1), which utilizes three levels of condition, i.e., intact ecologically,
moderately disturbed, and severely disturbed. These three levels of condition can be established
with varying levels of certainty, i.e., confidence intervals of varying widths. Not all decision-
making requires fine-scale information on wetland condition; the requirements of decision-
making may dictate the level of condition assessment required. In order to provide a range of
condition assessment options, the CWC has developed a three-tiered approach. Each tier in the
process of assessing condition requires a different level of effort. Whether one goes on to the
next step and a greater level of effort depends on the outcome of the previous effort and the
quality of information required.
LEVEL 1 - As a screening tool, prepare a watershed map utilizing the synoptic approach and the
best, available inventory information (provided by Level 1 of the inventory process, as described
above). The synoptic approach is documented in USEPA, 1992, and uses readily-available GIS
data layers to produce statewide maps that rank portions of the landscape according to a set of
landscape variables, or indices. The maps and indices are intended to provide regulators with a
measure of the landscape condition of an area and a relative rating of cumulative impacts
between areas. The indices are determined by the user, and, thus, may reflect the user's priorities
and needs. For example, if a map depicting the loss of flood storage function is desired, the
synoptic approach would combine GIS layers containing information on wetland loss and
hydrologic loading. At a minimum, a synoptic map should characterize land use patterns of
broad areas of the watershed and present this information on a map of Level I inventory
wetlands. It is anticipated that an update of GIS land cover data layers would occur about every
14

-------
five years. The map can then be used to see if significant or particularly sensitive wetland
acreage is located in proximity to a land use considered to have a high potential for impact (i.e.,
with potential for impact on wetland functions). Areas of present impact, potential impact, and
no probable impact can be approximated, and used to prioritize watershed activities.
LEVEL 2 - If the existing inventory is judged to be insufficient for the level of decision-making
desired, landscape-based decision rules are developed and applied to provide an improved
estimate of wetland acreage. This information is provided by Level 2 of the inventory process (as
described above). Wetland acreage is expressed as an estimate of total wetland acreage in the
watershed, with zones of high, moderate, and low probability of significant wetland acreage
identified on a map. This process ultimately requires ground reconnaissance to locate and classify
wetlands from a probability-based sample. During this ground reconnaissance, a preliminary
assessment of condition is also performed, utilizing a simple checklist to identify probable
stressors. This level of assessment provides both an estimate of wetland acreage and level of
potential threat with wide confidence intervals.
LEVEL 3 - If assessments at Levels I or II detect potential problems, a more detailed ground-
based assessment to assess condition and diagnose specific stressors (about one half day per
wetland) can be performed. If HGM functional models are chosen to serve this purpose,
condition can be expressed in terms of HGM functions and HGM types. For example, condition
would be expressed as: "Thirty percent of depressional wetlands in the Juniata watershed are
exhibiting moderate degradation of the long-term storage of surface water function."
OBJECTIVES
Our objectives for the study are:
1) To determine and report on the ecological condition of wetlands in the Juniata River
watershed using a series of assessment tools.
a) Develop a preliminary assessment of wetland abundance on two sub-watersheds in the
Juniata River watershed. Our experience with applying NWI digital data and other
remotely-sensed data for inventorying wetlands in the unglaciated portion of
15

-------
Pennsylvania has shown that these sources do not include the majority of wetlands
occurring in the watershed. To effectively sample wetlands in the Juniata, a better
estimate of their abundance and general location is necessary (i.e., a Level 1 inventory is
not adequate). To help remedy this situation, we are developing a process for deriving a
best estimate of wetland acreage from a combined set of GIS databases and a series of
decision rules (Level 2 inventory). Acreage will be expressed as an estimate of total
wetland acreage in each subwatershed, with zones of high, moderate, and low probability
of significant wetland acreage identified on a map.
b)	Verify and calibrate the inventory process on two subwatersheds in the Juniata, before the
process is applied to the entire watershed, including ground reconnaissance. During the
reconnaissance, a cursory inspection of wetland stressors will be performed, resulting in a
preliminary indication of condition (Level 2 assessment).
c)	Conduct an inventory of wetland acreage and an assessment of condition for the entire
Juniata River watershed. The inventory of the entire watershed will be based on the
' results of the work done to accomplish Objectives la and lb. Condition will be expressed
in terms of HGM functions and HGM type. For example, condition might be expressed
as: "Thirty percent of depressional wetlands in the Juniata watershed are exhibiting only
a moderate degradation of the long-term storage of surface water function." Condition
will be assessed by applying the HGM functional assessment models at a set of wetlands
selected by probability-based sampling. The verified inventory and map of acreage
zones, and application of HGM functional assessment models constitute a Level 3
assessment.
2)	Evaluate the feasibility of integrating a series of bioindicators into the wetland condition
assessments for the two sub-watersheds.
3)	Evaluate the feasibility of using citizen volunteers to apply the wetland monitoring protocols
throughout the Juniata River watershed.
16

-------
APPROACH AND METHODS
USE OF REFERENCE WETLANDS
As stated previously, the CWC has intensively studied 70 reference wetlands in
Pennsylvania spanning both a variety of HGM subclasses and a disturbance gradient. The
majority of these sites are located in the Ridge and Valley Province. Eleven reference wetlands
from the set are located in the Juniata River watershed, including one wetland being monitored in
cooperation with the Juniata Valley High School in Alexandria. It is important to reiterate that
the use of the term "reference" applies to the entire collection of wetlands that span a disturbance
gradient, while other investigators may use the term "reference" to imply only pristine
conditions. The characterization of wetlands across a disturbance gradient is an intentional
characteristic of the reference collection's architecture. We must know not only the level of
function a wetland of a given type may achieve (assumed to be in a pristine, unimpacted
landscape), but also the level of functioning that is attainable in an impacted landscape. The data
from the reference collection provides two major products: Performance Criteria Matrices
(PCMs) and a means to calibrate the HGM functional assessment models. PCMs establish
reference standards on hydrology, soils, sediments, vegetation, and wildlife habitat for mitigation
projects and for assessing wetland condition. The PCMs describe standard conditions in
wetlands both by HGM type and level of condition. The matrix structure is illustrated in Table 3.
The original PCMs (Bishel-Machung et al. 1996, Brooks et al. 1996) are being continuously
updated as new data become available. Calibration of the HGM functional assessment models
utilized some of the PCM data, although some variables contained in the models were never
measured during the initial characterization of the reference set. To address this deficiency,
many of the reference wetlands were re-sampled during 1998 using our Rapid Assessment
Procedures (RAPs) to ensure that all members of the reference set were assessed for all potential
functions as described by the HGM models for the Ridge and Valley. These data are being used
to calibrate the HGM functional models for the Ridge and Valley Province.
17

-------
OBJECTIVE 1A. DEVELOP PRELIMINARY ESTIMATE OF WETLAND
ABUNDANCE ON TWO SUB-WATERSHEDS USING GIS.
Wetlands in the unglaciated portion of Pennsylvania are believed to encompass only 3-
5% of the landscape and they are relatively small in area. Based on our experience, National
Wetlands Inventory (NWI) quads for Pennsylvania underestimate the total acreage of wetlands
by nearly 100%. Any wetland trends assessment of the entire watershed will not include the
majority of wetlands occurring in the watershed. To help remedy this situation, we will develop
a best estimate of wetland abundance derived from a combined set of GIS databases and a series
of decision rules.
Two sub-watersheds in the Juniata River watershed will be selected for detailed
investigations. Although we have some candidate watersheds in mind (e.g., portions of the
Spruce Creek sub-watershed), the selection of these watersheds will be made in consultation with
USEPA, PADEP, and local community leaders. The intent will be to select two watersheds that
are typical of the geologic and land use diversity found in the Juniata River watershed.
1 Each of the sub-watersheds selected will be portrayed with digital data from satellite
imagery that characterize land cover and land use. For example, USEPA's Multi-Resolution
Land Cover (MRLC) and classified satellite imagery for Pennsylvania (Terrabyte from the
Pennsylvania GAP Project) are both at 30-m pixel resolution with overlays of 1:24,000 scale
stream and road data digitized by the Pennsylvania Department of Transportation (PennDOT).
Again, in our experience, wetlands are poorly recognized in this database. We have tried using
on-screen identification of known wetlands in an effort to identify appropriate spectral
signatures. However, the lack of unique vegetation patterns in most wetland types of the
unglaciated portions of the Commonwealth make this task difficult for all sites except those with
significant amounts of open water and/or aquatic beds, i.e., the wetter sites. Thus, we will use
other relevant data such as stream data, watershed boundaries, surficial geology (Pennsylvania
Geologic Survey, Map 51), elevation/slope/aspect, soils, the Federal Emergency management
Agency's (FEMA) floodplain maps, within a GIS to develop decision rules. The rules will then
be used to predict the probability of wetland abundance in three categorical zones: high,
moderate, and low.
18

-------
OBJECTIVE IB. VERIFY AND CALIBRATE THE PRELIMINARY ESTIMATE OF
WETLAND ABUNDANCE FOR THE TWO SUB-WATERSHEDS.
We will use an EMAP-style probability-based sampling approach to verify and calibrate
our preliminary wetland abundance estimate in the targeted two sub-watersheds (Stevens, 1997).
The EMAP sampling points will be randomly stratified into high, moderate and low probability
of being associated with wetlands. Initially, the EMAP sampling points located in the moderate
and low probability areas will be identified on aerial photographs. DEP staff and interns will
summarize the wetland area within a 1 -km strip on the photo that is centered on the sampling
point and oriented on a randomly-selected compass direction. If it is determined that not all
wetlands or wetland area can be identified using the aerial photographs, the sampling point will
be visited in the field by DEP staff and interns. At the sampling point, the DEP staff and interns
will inventory wetlands within the 1-km strip previously identified on the aerial photos to obtain
an estimate of wetland abundance. They will also perform a cursory inspection of stressors to the
wetland. CWC staff and volunteers will visit each sampling point located in the high probability
area and inventory the wetlands within the 1-km strip. The results of the inventory process
(estimates of abundance) in all three categories will be used to verify the GIS probability map
and to calibrate the decision rules.
In addition, a stressor checklist (Level 2 condition assessment) will be completed for each
wetland identified in the field. The checklist is composed of a set of indicators used to identify
probable stressors, such as sedimentation, hydrologic modifications, habitat fragmentation, and
acidification (Adamus and Brandt 1990). The purpose of the indicators is to allow agency
biologists and trained volunteers to rapidly identify the stressors affecting individual wetlands,
stream reaches, and the surrounding landscape. Wherever feasible, there will be both field and
landscape versions of each indicator. Some stressors, such as habitat fragmentation and
sedimentation, must be assessed both from the synoptic watershed maps and from ground
reconnaissance. An example of one field indicator for one stressor - sedimentation - might be
observations of potential pathways for sediments such culverts, ditches, or exposed earth around
the edge of a wetland. For hydrologic modification one field indicator might be evidence of
dying trees in a flooded wetland.
19

-------
OBJECTIVE 1C. CONDUCT AN ASSESSMENT OF WETLAND ABUNDANCE AND
CONDITION IN THE ENTIRE JUNIATA WATERSHED.
We will use the EMAP-style probability-based sampling approach (Stevens 1997), tested
in Objective lb, to characterize the wetland abundance and condition in the entire Juniata. For
this objective, the EMAP sampling points will be randomly located in only areas of high
probability of wetland occurrence. Since, at this time, it is not known how many points should
be sampled, we will assume about 150 will be sufficient (and probably a maximum number). The
inventory process described in Objective lb will be performed at the sampling points to obtain an
estimate of wetland abundance.
Once wetland area at each sampling point has been inventoried, the wetlands will be
weighted by their area, and then one wetland within the 1-km strip will be randomly selected to
perform the condition assessment. At each selected wetland, the Rapid Assessment Procedures
(RAPs) and an alternative protocol, which requires less plant identification, will be performed.
The Rapid Assessment Procedures (RAPs) were developed for the Adopt-a-Wetland Program of
Pennsylvania's High Schools and for collecting calibration data for our HGM models (Brooks
and Wardrop in prep.). The results of both the RAPs and alternative protocol will be compared to
see if the alternative protocol provides adequate information for decision making.
Assuming about 150 points will be sampled, we estimate that each site can be monitored
with our Rapid Assessment Procedures (RAPs) (Level 3 condition assessment) in about 3 hours
in the field with a two-person team, for a total of about 600 hours of actual sampling time.
Where access to a site is not allowed, an alternative point will be selected and assessed.
The data collected in the field on wetland condition will be used to develop an index of
wetland condition. The final form of the index is not known at this time. A potential model,
however, can be found in US EPA's "Surf-Your-Watershed" web site (www.epa.gov/surf7iwi)
where an "Index of Watershed Integrity" (IWI) can be, generated. There are two categories for
the IWI, one of condition and one of vulnerability. The former consists of characteristics, much
like those measured by the RAPs for individual wetlands. The latter represent stressors similar to
the ones measured by during the landscape assessments. So perhaps, a similar index to wetland
integrity for an entire watershed might be created and displayed on the same web page with the
IWI.
20

-------
OBJECTIVE 2. EVALUATE THE FEASIBILITY OF USING BIOINDICATORS IN
ASSESSING CONDITION.
The CWC has extensive experience in the development and testing of biological,
chemical, and physical indicators for use in assessing wetland condition, e.g., plants (Goslee et
al. 1997), soils (Bishel-Machung et al. 1996, Stauffer and Brooks 1997), sediments (Wardrop and
Brooks 1998), hydrology (Cole et al. 1997, CWC unpublished data), water quality (Babb et al.
1997, CWC unpublished data), birds (Croonquist and Brooks 1993, O'Connell et al. 1998,
Gaudette 1998), amphibians (Brooks et al. 1996, CWC unpublished data), and
macroinvertebrates (Bennett, CWC in progress). Also, work has been conducted on a watershed
basis at the landscape scale (Brooks et al. 1996, Miller et al. 1997, Wardrop 1997, O'Connell et
al. 1998). Brooks and Wardrop are participants in US EPA's Biological Assessment of Wetlands
Working Group (BAWWG), so the principals in the Juniata study will remain current with regard
to recommended bioindicators and methods.
For this project, we plan to test the use of several bioindicators in conjunction with the
RAPs. This work will be conducted during the work in the field to verify and calibrate the
estimates of wetland abundance in the two sub-watersheds (objective 2a). At this time, we plan
to collect plant (dominant species) data, at a minimum, which can be easily collected by trained
volunteers. Pending the results of our work in progress on birds, wetland macroinvertebrates and
streamside salamanders, we may add these components. A brief discussion of the approach used
for each of these indicators presented below.
Plant Community Assessment
Indicators can generally be thought of as measurable variables that are directly or
indirectly related to parameters of interest. When indicators are intended to infer a measure of
biological function, they are termed bioindicators. Attempts to compile exhaustive lists of
potential bioindicators have been attempted elsewhere, and a short list has been prepared by the
USEPA (Adamus and Brandt 1990). Potential responses of a wetland to stressors are many, and
involve plant, animal, and microbial communities. While not all plant species are highly sensitive
to disturbance, the immobility of the plant community, its amenity to remote sensing techniques,
and easily recognized signs of stress make it preferable for an initial study of disturbance effects.
21

-------
Previous work at the CWC studied the impact of one stressor (sedimentation) on the plant
community, and investigated the potential utility of plant community measures as indicators of
wetland disturbance (Wardrop and Brooks, 1998). Responses did occur at the level of individual
species, and species can be categorized as sediment tolerant, moderately tolerant, slightly
tolerant, and sediment intolerant based on their association with environments of varying
magnitudes of sedimentation. In general, species that were categorized as sediment tolerant or
moderately intolerant increased in percent cover (dominance) over a gradient of increasing
sediment accumulation. Mean percent cover, when plotted versus sediment accumulation,
provides a stressor-impact curve for an individual species.
The RAP developed by the CWC contains a comprehensive plant community sampling
methodology, which has been used in a variety of project types. Three sizes of plots are used to
record various measures of the plant community: a 1 m plot, a circular plot with a radius of 3 m,
and a circular plot with a radius of 11.6 m. The activities in each plot are:
1 m2 Plot
•	Percent cover to the nearest 5% for dominant species (up to 5 herbaceous species).
3 m-radius Plot
•	Species richness (i.e., number of species present)
•	Percent aerial cover of downed leaf and small woody material (less than 1 cm in diameter)
•	Height and circular projection of cover (crown) for all shrubs
11.6 m-radius Plot
•	Basal area, by species of trees and estimates of crown closure
•	Estimates of percent herbaceous cover
•	Number of occurrences of downed woody material
This protocol has been used with a variety of sampling personnel, including high school
students, and has been shown to be fairly robust if the sampling team is properly trained.
22

-------
Avian Community and Landscape Pattern Assessment
Additional indicators of landscape condition are useful and relevant to this study because
of the relationship between watershed-wide landscape condition and the condition of wetlands in
the Juniata. Bird communities provide one type of regional indicator of landscape condition, and
their use is easily justified. Due to their mobility, birds may respond to a wide range of stressors
affecting both terrestrial and aquatic habitats. Predictions regarding bird community responses to
changes in land cover and connectivity are based on readily-available life history information,
and have proven to be reliable (e.g., Croonquist and Brooks 1991). Although census data are
usually site-specific, they can be aggregated at least to a landscape scale (multiple km2), and
perhaps to an ecoregion. Trends in songbird populations are reported both regionally and
nationally, and their suitability as a regional indicator is currently being tested (O'Connell et al.
1998).
We are engaged in a separate project to examine changes in bird communities across
landscapes in the Mid-Atlantic Highlands Area (MAHA)(0'Connell et al. 1998). A Bird
Community Index (BCI) that is responsive to changing landscape patterns has been developed.
Data were collected in 58 plots during 1995 and for 68 plots in 1996 centered on random points
of the EMAP hexagonal grid. In addition, we have bird data from 34 reference wetlands and
associated upland plots in the Ridge and Valley Province from 1994, and a similar set of data
from 60 other wetlands collected in 1995 (Gaudette 1998). Numerous points from both of these
studies were located in the Juniata watershed. These data are being correlated with landscape
metrics developed from 1-km diameter circles. Results from these studies show that response
guilds of the bird community vary predictably as the landscape matrix shifts from predominantly
forest to a mixed mosaic of patches (Gaudette 1998, O'Connell et al. 1998). At least five
categories of landscape configuration have been identified, with corresponding responses by bird
guilds.
Measurement of bird communities is relatively simple, a volunteer data collection
network is in place, and historic databases exist. This information could be used in conjunction
with on-site avian censuses conducted by knowledgeable volunteers, as a coarse indicator of
landscape condition within each watershed. The Juniata Audubon Chapter is quite active and
competent, so at least a modest pool of potential volunteers is available. Avian communities will
23

-------
be assessed using standard 10-minute point counts (i.e., morning census period under suitable
weather conditions). Point counts will be conducted three times during the breeding season at a
minimum of 10 points per stream reach. Birds detected by sound or sight within a 50-m radius
plot adjacent to the stream will be recorded. Plots will be at least 150 m apart. Habitat
characteristics at point counts will follow those used by O'Connell et al. (1998) for plots the
EMAP Bird Landscape Study. If avian community data becomes available for the Juniata, it
could be applied to the existing BCI as a means of assessing landscape condition around selected
wetlands. Data for wetland-dependent species could be applied to a wetland bird IBI (proposed
for development in late 1999) to evaluate the condition of wetlands in the Juniata basin.
Macroinvertebrate Community Assessment
Aquatic invertebrate communities are known to change in response to a variety of
stressors (Adamus and Brandt 1990, Brooks et al. 1991, Hicks 1995). A significant effort has
been made to integrate chemical, biological, and physical parameters for assessing the ecological
ihtegrity of streams (e.g., USEPA 1991), resulting in satisfactory predictions of the health and
condition. Considerably less effort has been directed towards wetlands. Use of
macroinvertebrates as an indicator will depend on the level of taxonomic detail needed for the
Invertebrate Community Index (ICI) being developed by Bennett and Brooks for wetlands under
separate funding. If feasible, we will aggregate species into easily identifiable groups and
response guilds to simplify the ICI.
There are no standard methods recommended for sampling macroinvertebrates in
wetlands. In previous studies, we have investigated the utility of several techniques, including
submergence traps, emergence traps, benthic grab samples, benthic cores, and sweep nets
(Brooks et al. 1991, Brooks and Prosser, unpublished). In still waters having an open water
column, submergents, or emergents, we will use a D-net, swept in a 1-m arc 10 times. Benthic
cores (5-10 cm in diameter and depth) will be taken in wetlands with standing water, saturated
soils, or seasonally saturated soils (Kentula et al. 1992, Hicks 1995). For sweeps and benthic
cores, three samples will be taken in representative habitats and pooled for sorting and analysis.
All samples will be rinsed through a No. 35 mesh (500-micron) screen. The remaining
material will be distributed evenly in a light-colored pan and the macroinvertebrates removed.
24

-------
Specimens will be preserved in alcohol before being identified (or afterwards if sorting occurs
immediately). The level of identification will generally be to order. Further identification to
family, genus, and species may be required for some taxa. Voucher specimens will be kept for
reference. The primary identification guides used will be Thorp and Covich (1991) and Merritt
and Cummins (1996).
OBJECTIVE 3. EVALUATE THE FEASIBILITY OF WORKING WITH CITIZEN
VOLUNTEERS.
We will contact leaders of the communities and conservation groups within the watershed
to discuss the objectives of the proposed study, discuss opportunities for collaboration and
sharing data, and request their assistance in the completion of this work. Four conservation
organizations are identified in USEPA's "Surf Your Watershed" site for the Juniata, although
others exist. In addition, we will request to work explicitly with the County Conservation
Districts, County Cooperative Extension Offices, PADEP's Southcentral Regional Office,
Pennsylvania Game Commission's Southcentral Regional Office (we have worked previously
with Willis Sneath, the Regional Director), and other interested parties. These community
outreach efforts will be organized by our Research Assistant - Jennifer Perot, and coordinated
with the Juniata Monitoring Coordinator for the project. A web site should be established to
communicate the progress of the study and to provide a location for displaying data and
information. One possible location for summarized data and maps is USEPA's "Surf Your
Watershed" site (www.epa.gov/surf7iwi). Our queries to this site have found it to be very useful
for both passive and interactive inquiries about the watershed.
We will test the suitability of the condition assessment protocol for trained volunteers
during the initial 1999 field season. Field team leaders from the Southern Alleghenies
Conservancy (SAC) will accompany CWC personnel during condition assessments of at least 10
wetlands. The protocol will be open to evolution during that time, with input from the SAC
personnel on its appropriateness for implementation by volunteers. In addition, a formal test of
two versions of vegetation sampling will occur, and the results will be used to finalize the
protocol for the year 2000 field season (with accompanying QA plans (USEPA, 1996)).
25

-------
GENERAL PROJECT INFORMATION
PERSONNEL ASSIGNMENTS
Robert P. Brooks, Ph.D. -- Principal Investigator (PI): Dr. Brooks has over 20 years of
experience as a wildlife biologist and wetland scientist. Currently, he is Professor of Wildlife
and Wetlands Ecology and Director of the Penn State Cooperative Wetlands Center. He has
experience in managing multi-scale projects. He recently completed a 3-year statewide study of
reference wetlands, is the PI for the EMAP Bird Landscape study in the MAHA, and Co-PI for a
multi-year Water and Watersheds study cooperatively funded through the National Science
Foundation and USEPA . He has extensive expertise regarding the ecology and conservation of
wetland, stream, and riparian components of watersheds, but is also familiar with terrestrial
habitats, land use planning, and landscape analysis. Dr. Brooks will serve as Project Director,
and in that role will oversee the work of others on the project, including the GIS analyses. He
will also guide and participate in the development of the wetland trends analysis for the total
\vatershed. He will work with the team members to compile, analyze, and interpret the project's
data in preparation for submittal of reports.
Denice Heller Wardrop, PE, Ph.D. — Co-PI and Project Manager : - Dr. Wardrop has over 20
years of experience in environmental sciences, the ecology of wetland and aquatic systems, risk
assessment, and the fate and transport of sediment. She is currently a Research Associate with
the Penn State Cooperative Wetlands Center. She has extensive experience in project
management, both technical and administrative. She recently participated in a 3-yr statewide
study of reference wetlands, and completed her dissertation on the occurrence and impact of
sedimentation in central Pennsylvania wetlands. Dr. Wardrop will serve as Project Manager, and
will be responsible for preparation of reports and submittals. She will also work with Dr. Brooks
to compile, analyze, and interpret project data.
Jennifer K. Perot — Research Assistant: Ms. Perot has over seven years of experience in aquatic
ecology, use of GIS, and risk assessment. She is currently a Research Assistant with the Penn
State Cooperative Wetlands Center. She has recently used GIS to classify watersheds in the
Lower Peninsula of Michigan and the Illinois River Basin.
26

-------
GIS Research Assistant - This staff person will be responsible for compiling databases and
conducting landscape analyses using the GIS resources of Penn State's Office of Remote Sensing
of Earth Resources (ORSER). This person will be supervised by Barry Evans of ORSER. He
currently manages the GIS database and all task orders requested by state agencies in
Pennsylvania.
TIMETABLE AND PRODUCTS FOR THE PROPOSED WORK:
January 1999 - Selection of subwatersheds and initiation of GIS assessments of subwatersheds.
Finalization of decision rules for preliminary inventory.
Spring 1999- Submission of final Study Plan and Quality Assurance Project Plan (includes RAP
and QA/QC procedures)
June 1999 - Reconnaissance of subwatersheds for both inventory verification and condition
assessment protocol testing
September 1999 - Compilation of field data; refinement of inventory and condition protocol
January 2000 - Selection of watershed sites on final inventory map
June 2000 - Reconnaissance of randomly-selected wetlands in watershed
September 2000 - Compilation of field data
January 2001- Begin preparation of final report
27

-------
LITERATURE CITED
Adamus, P. R., and K. Brandt. 1990. Impacts on quality of inland wetlands of the United States:
A survey of indicators, techniques, and application of community-level biomonitoring data. U.
S. Environ. Prot. Agency, Environ. Res. Lab., Corvallis, OR. EPA/600/3-90/073.
Babb, J. S., C. A. Cole, R. P. Brooks, and A. W. Rose. 1997. Hydrogeomorphology, watershed
geology, and water quality of wetlands in central Pennsylvania. J. PA Acad. Sci. 71(1 ):21 -28.
Bishel-Machung, L., R. P. Brooks, S. S. Yates, and K. L. Hoover. 1996. Soil properties of
reference wetlands and wetland creation projects in Pennsylvania. Wetlands 16(4):532-541.
Brooks, R. P. 1990. Wetlands and deepwater habitats in Pennsylvania. Pages 71-79 in S. K.
Majumdar, E. W. Miller, and R. R. Parizek (eds.). Water Resources in Pennsylvania:
Availability, Quality and Management. Pennsylvania Academy of Science, Easton, PA. 580+xiii
pp.
Brooks, R. P., C. A. Cole, D. H. Wardrop, L. Bishel-Machung, D. J. Prosser, D. A. Campbell,
and M. T. Gaudette. 1996. Wetlands, wildlife, and watershed assessment techniques for
evaluation and restoration (W3ATER). Vol. 1, 2A, and 2B, Rep. No. 96-2, Penn State Coop.
Wetlands Ctr., University Park, PA. 782pp.
Brooks, R. P., M. J. Croonquist, E. T. D'Silva, J. E. Gallagher, and D. E. Arnold. 1991.
Selection of biological indicators for integrating assessments of wetland, stream, and riparian
habitats. Biological Criteria: Research and Regulation. U.S. Environ. Prot. Agency, Office of
Water, EPA-440/5-91-005, Washington, DC. 171pp.
Brooks, R. P., and D. J. Prosser. 1995. Habitat suitability index models and wildlife community
habitat profiles for use in Pennsylvania wetlands. Penn State Coop. Wetl. Ctr., Rep. No. 95-1,
University Park, PA. 27pp.
Cole, C. A., R. P. Brooks, D. H. Wardrop. 1997. Wetland hydrology as a function of
hydrogeomorphic (HGM) class. Wetlands 17(4):456-467.
Croonquist, M. J., and R. P. Brooks. Effects of habitat disturbance on bird communities in
riparian corridors. J. Soil Water Conserv. 48(l):65-70. 1993. (Co-authored 50%, supervised
research)
Day, R. L., P. L. Richards, and R. P. Brooks. 1997. Chesapeake Bay riprain forest buffer
inventory. Final Report, Chesapeake Bay Program Office, Annapolis, MD. 113pp.+app.
Gaudette, M. T. 1998. Modeling wetland songbird community integrity in central Pennsylvania.
Ph.D. Thesis. Wildlife and Fisheries Science. Pennsylvania State University, University Park,
PA.
28

-------
U.S. Environmental Protection Agency. 1991. Biological criteria: research and regulation.
Proc. Symp., 12-13 December 1991, Arlington, VA. Washington, DC. 171pp.
Wardrop, D. H., and R. P. Brooks. 1998. The occurrence and impact of sedimentation in central
Pennsylvania wetlands. Environ. Monit. Assmt. In press.
Wardrop, D.H. 1997. The Occurrence and Impact of Sedimentation on Central Pennsylvania
Wetlands Ph.D. Thesis. Ecology. Pennsylvania State University, University Park, PA. 209 pp.
30

-------
NHEERL-COR-2351A
TECHNICAL REPORT DATA
(Please read instructions on the reverse before completing)
NHEERL-COR-897R
1. REPORT NO.
EPA/600/R-98/181
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE: Development and application of assessment protocols for
determining the ecological condition of wetlands in the Juniata River Watershed.
5. REPORT DATE
6. PERFORMING ORGANIZATION
CODE
7. AUTHOR(S) Robert P. Brooks, Denice Heller Wardrop, Jennifer K. Perot
8. PERFORMING ORGANIZATION REPORT
NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Penn State Cooperative Wetlands Center
The Pennsylvania State Universty
Environmental Resources Research Institute
University Park, PA 16802
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
US EPA ENVIRONMENTAL RESEARCH LABORATORY
200 SW 35th Street
Corvallis, OR 97333
1 3. TYPE OF REPORT AND PERIOD
COVERED
14. SPONSORING AGENCY CODE
EPA/600/02
15. SUPPLEMENTARY NOTES:
16. ABSTRACT: This study will contribute to the development of a means to accurately, efficiently, and fairly assess a wetland's condition in
the context of the surrounding watershed that can then be used to implement protective and restorative strategies that are appropriate for both
the individual wetland and the watershed. This has been one of the primary goals of research and outreach efforts conducted by the Penn State
Cooperative Wetlands Center (CWC) since 1993, and will guide their approach to monitoring and assessing wetlands in the Juniata watershed in
central Pennsylvania. The objectives of the study are:
1.	To determine and report on the ecological condition of wetlands in the Juniata River watershed using a series of assessment tools.
a.	Develop a preliminary assessment of wetland abundance on two sub watersheds in the Juniata River Watershed. Our experience with
applying NWI digital data and other remotely-sensed data for inventorying wetlands in the unglaciated portion of Pennsylvania has
shown that these sources do not include the majority of wetlands occurring in the watershed. To effectively sample wetlands in the
Juniata, a better estimate of their abundance and general location is necessary (i.e., a Level 1 inventory is not adequate). To help remedy
this situation, we are developing a process for deriving a best estimate of wetland acreage from a combined set of GIS databases and a
series of decision rules (Level 2 inventory). Acreage will be expressed as an estimate of total wetland acreage in each subwatershed,
with zones of high, moderate, and low probability of significant wetland acreage identifies on a map.
b.	Verify and calibrate the inventory process on two subwatersheds in the Juniata, before the process is applied to the entire watershed,
including ground reconnaissance. During the reconnaissance, a cursory inspection of wetland stressors will be performed, resulting in a
preliminary indication of condition (Level 2 assessment).
c.	Conduct an inventory of wetland acreage and an assessment of condition for the entire Juniata River watershed. The inventory of the
entire watershed will be based on the results of the work done to accomplish Objectives 1a and lb. Condition will be expressed in terms
of HGM functions and HGM type. For example, condition might be expressed as: "Thirty percent of depressional wetlands in the Juniata
watershed are exhibiting only a moderate degradation of the long-term storage of surface water function." Condition will be assessed by
applying the HGM functional assessment models at a set of wetlands selected by probability-based sampling. The verified inventory and
map of acreage zones, and application of HGM functional assessment models constitute a Level 3 assessment.
2.	Evaluate the feasibility of integrating a series of bioindicators into the wetland condition assessments for the two sub-
watersheds.
3.	Evaluate the feasibility of using citizens volunteers to apply the wetland monitoring protocols throughout the Juniata River
watershed.
17.
KEY WORDS AND DOCUMENT ANALYSIS

-------
m
a. DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED
TERMS
c. COSATI Field/Group
Wetlands, Juiata Watershed, Assessment


18. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report)
21. NO. OF PAGES: 30
20. SECURITY CLASS (This page)
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE

-------