Environmental Protection Environmental Effect Research September
Agency Laboratory 1998
Western Ecology Division
Corvallis, OR 97333
ssessment,
(FEIMAP)
Climate Change
Elevated CO2
Elevated 03
N-deposition
Nature Stressors
Climate
Soils
Management Practices
-------
EPA Report NHEERL-COR-
September 1998
Forest Ecosystem Indicators:
Monitoring, Assessment, Prediction
(FEIMAP)
by
S. Brown
C. Andersen, M. Johnson, A. Seidler
P. Beedlow, A. McKane, A. Solomon,
M. Cairns, P. Rygiewicz, D. Tingey,
B. Hogsett L. Watrud
National Health and Environmental Research Laboratory
Western Ecology Division
200 SW 35th St.
Corvallis, OR 97333
-------
Abstract
Ecological indicators for forests are
necessary for (1) predicting and! or
assessing the future response of forests
to anthropogenic stressors, (2) detecting
and quantifying changes and trends in
forest condition, (3) linking changes in
condition to likely stressors, and (4)
identifying early warning measures for
loss of integrity and sustainability of
ecological resources. The approach for
developing indicators includes: (I)
process-based models, (2) ecological
measures, (3) stress-response data
relationships, and methods, at scales
from ranging from populations to
landscapes. The approach includes
collection of climatic, edaphic and
ecological data from intensive field sites,
spatial data bases on land cover/land use
and elevation and controlled chamber
experiments to build and parameterize
site (e.g. biogeochemical cycling and
GAP) and landscape models (e.g. C, N,
and water quantity).
Initially, date will be collected along an
elevation gradient in the western
Cascades of Oregon that will be used to
parameterize the MBL-General
Ecosystem Model (GEM). This model
will be used to assess how changes e.g.,
in temperature, nitrogen deposition, CO 2
concentration, and soil moisture affect
biomass production and ecosystem
storage and cycling of C and N, and to
identify key ecosystem processes that
are sensitive to these stressors. Key
processes sensitive to climatic and
atmospheric stressors are those related to
C and N cycling in the rhizosphere or
plant/soil interface and to C and N
allocation and partitioning. Therefore,
we will develop and evaluate indicators
for rhizosphere processes (e.g. structure
and function soil food web),
ecophysiological processes (e.g. carbon
allocation to fine roots), and ecosystem
and landscape function. We will use
other sites in the western Cascade region
and Olympic National Park to validate
and verify the performance of the
models and potential indicators.
-------
Acknowledgement
The information in this document has been funded wholly by the U.S.
Environmental Protection Agency. It has been subjected to the Agency’s
peer and administrative review, and it has been approved for publication
as an EPA document. Mention of trade names or commercial products does
not constitute endorsement of recommendation for use.
-------
TABLE OF CONTENTS
1. INTRODUCTION .4
1.1. Research Needs within EPA Office of Research
Development 4
1.2. Current Status on Forest Indicators for Assessment and
Monitoring 5
2. RESEARCH APPROACH 7
2.1. Research Objectives 7
2.2. General Approach 8
3. RESEARCH IMPLEMENTATION 10
3.1. Field Sites 10
3.1.1. Intensive field sites 10
3.1.2. Extensive sites 15
3.1.3. Watersheds 17
3. 2. Modeling 18
3.2.1. Biogeochemical Cycling Modeling 19
Task 1: Modeling at the stand scale 19
Task 2: Modeling at the landscape scale 32
3.2.2. Forest Succession Modeling 35
Task 3: Forest succession model for stand to
landscape 35
3.3. Resource Utilization 43
3.3.1. Resource utilization-producers 45
Task 4: Carbon and Nitrogen Allocation 45
Task 5: Carbon and Nitrogen Partitioning 51
Task 6. Phenology
3.3.2. Resource utilization-consumers 59
Task 7: Carbon and Nitrogen Transformations 59
Task 8: Carbon and Nitrogen Losses 61
Task 9: Complexity and Function of the Soil Food Web 64
-------
4. LINKAGESTO OTHER WED PROJECTS .72
5. PROJECT MANAGEMENT AND QUALITY ASSURANCE
STATEMENT
6. REFERENCES 80
7. APPENDICES 103
3
-------
1. INTRODUCTION
1.1. Research Needs within
EPA Office of Research and
Development
The US Environmental Protection Agency
(EPA) is required to protect the
environment to ensure clean air and water
and an uncontaminated food supply.
Although many environmental problems
have diminished as a result of such Federal
actions, some have not and new problems
continue to arise. To deal with the new and
potential problems, the EPA has
established ecological protection as one of
its highest pnority research areas.
The EPA’s Office of Research and
Development (ORD) has substantially
changed its organization and operation to
strengthen the Agency’s science base and
improve the nation’s ability to effectively
respond to complex environmental
challenges (US EPA 1996) One of the
most Important changes is the explicit use
of the risk assessment and management
paradigms to shape and focus the Agency’s
research agenda. One of the key research
needs for improving ecological risk
management is to provide the scientific
tools necessary for cost-effective
management decisions by stakeholders on
the protection of ecological resources at
local, regional, and national scales. Such
scientific tools need to: identify the most
important ecological risks; perform risk
assessments, including problem
formulation, characterize stress-response
and exposure functions, and characterize
risks; identify and evaluate risk
management options; and venfy and
monitor measures or indicators of risk
reduction (US EPA 1996).
The Committee on Environment and
Natural Resources (CENR) calls for all
environmental agencies to merge efforts in
forming a national monitoring and research
network to achieve the national common
goal of understanding and managing
ecological systems for their continued and
sustained vitality, diversity, habitat, and
ability to provide goods, services, and
enjoyment for humans. In response to this
national goal, EPAJORD focuses its
research efforts through the Environmental
Monitonng and Assessment Program
(EMAP) to: (1) develop ecological
indicators that can be used to monitor
status and trends in ecosystems, (2) design
effective systems for monitoring status and
trends, and (3) integrate and synthesize
environmental data (US EPA 1997a).
Here we present a plan for a long-term
research project to support several of the
Agency’s research needs with reference to
forests. Specifically, the overall goal of this
plan is to develop and evaluate ecological
indicators that support (1) EMAP’s need
for forest indicators of ecological integrity
and sustainability (US EPA 1997c), (2) the
Global Change program’s need for
indicators of global change, and (3) risk
assessment and risk management protocols.
We consider ecological indicators to include
predictive models to identify potential
risks to global change and air pollutants,
stress-response functions, methods, and
measures and their behavior over time and
space. These indicators can be used to
4
-------
assess and m9n ltor change and trends in the
condition of forest resources, link changes
to likely stressors, predict response of
forests to future environmental change, and
identif ’ early warning signals.
1.2. Current Status on Forest
Indicators for Assessment
and Monitoring
In 1989, the EPA in collaboration with
other federal and state agencies and
research institutions, initiated the
Environmental Monitoring and Assessment
Program (EMAP). The EMAP is a national
program designed to assess the status and
trends of the Nation’s ecological resources,
includ ing forests, arid lands,
agroecosystems, wetlands, inland surface
waters, and estuaries (Hunsaker and
Carpenter 1990). The program was
mitiated as a long-term, interagency project
to monitor and evaluate the condition of
ecological resources, develop methods for
anticipating emerging problems before they
reach crisis proportions, and contribute
information to decisions on environmental
protection and management at regional and
national scales (Thornton et al. 1993).
At present, surveys of forests condition
are conducted by USDA Forest Service
through the Forest Health Monitoring
(FHM) program. The FHM program began
in 1990 in the Northeastern region of the
USA in partnership with EMAP. The
FHM program has four components:
detection monitoring, evaluation
monitoring, intensive site monitoring, and
research on monitoring techniques.
Detection monitoring is the most
developed component of the program. It is
composed of a nation-wide network of
long-term monitoring plots that have been
or are being established throughout most of
the forested areas of the US. The FHM
program also uses information from other
programs both within the Forest Service
(e.g., Forest Inventory and Analysis, fire
data, and forest pests and disease data) and
other agencies (e.g., EPA for air pollution
data and NOAA for weather data). A suite
of indicators, considered important in
assessing forest health, have been
developed (Lewis and Conkhng n.d.) and
are being monitored on the network of
FHM plots. Four groups of indicators are
being used (A. J. R. Gillespie, USFS,
Radnor, PA, pers. comm.): mensuration
(e.g., tree growth, mortality,
dendrochronology), crown condition (e.g.,
LAI, crown dieback, and branch
evaluation), tree damage (e g., leaf
coloration and stem damage), and ozone
sensitive plants (e.g., lichens and
bioindicator plants). The FHM program
uses these indicators to evaluate the current
forest condition and to determine if forest
health is static, improving, or degrading
over time.
Within the Pacific Northwest forests, a
FHM pilot study was established in 1994
(Campbell and Liegel 1996) using 251 -ha
sample plots located in forests of the
Cascade Range of Oregon and Washington.
Within these plots cover and species
composition of understory vegetation,
species composition and structure (e.g.,
size classes, tree density, basal area,
number of large live and dead trees) of the
overstory vegetation, crown condition (e.g.,
vigor, density, dieback), tree damage by
type (e.g., cankers, wounds, damaged
foliage), species richness of macro-lichens
on woody substrates, and abundance of
5
-------
songbirds in relation to forest habitat are
being measured (Campbell and Liegel
1996) Results from this pilot study have
provided limited baseline data on the
condition of the forests in the Cascades as
well as testing the FHM sampling
protocols for the region.
Several scientific concerns and challenges
have emerged during reviews of the EMAP
by the EPA Scientific Advisory Board, the
National Research Council (NRC 1995),
EPAJORD, and from recent research
findings at the EPA Western Ecology
Division (WED) of NHEERL relative to
forest indicator development (Table I).
The NRC (1995) concluded that a focused
research program on indicator development
is needed because the use of scientifically
defensible indicators should be the heart of
the EMAP.
The development of ecological mdicators
for forests represents a unique challenge.
For example, forests are dominated by
long-lived species, have simple to complex
species assemblages, contam multiple
vertical layers, composed of single- or
mixed-age individuals, grow in complex
terrains and environments, depend
primarily upon nutrient recycling rather
than external inputs, are limited by a
myriad of environmental factors that
change throughout the lifetime of the
forest, and are subject to an array of natural
(e.g., fire, pests) and anthropogenic (e.g.,
climate change, N deposition, tropospheric
ozone, UVB, management, chemical
application, exotic invasions) stressors of
different intensities and duration (Aber and
Melillo 1991). Further, the effects of
various anthropogenic stressors are likely
to differ depending upon the ecological
conditions under which forests grow For
example, ozone effects in arid regions are
different than in humid ones, and increased
N deposition is likely to affect N limited
forests differently than N saturated forests.
Despite the concerns that have emerged
from reviews of EMAP (Table 1), the
present suite of EMAPIFHM indicators
are useful for monitoring the status of
forests. However, they provide a signal late
in the response phase given the longevity
of trees; indicators are needed that occur
early enough during response to stress that
action can be taken. We believe that
indicators of this nature will be those that
are more direct measures of forest
processes or those structural components
that have fast turnovers, such as plant
resource utilization, fine root dynamics,
and the soil food web. The reason for
focusing on forests processes or
components with fast turnover is that they
are more sensitive to change, i.e., they
generally have faster response times
whereas major structural components have
slow turnover and response times (decades
or more).
6
-------
Table 1. Scientific concerns and challenges that have emerged from reviews of the EMAP
2. RESEARCH APPROACH
2.1. Research Objectives
The goal of this research project is to
develop, test, and evaluate ecological
indicators necessary to (1) predict or
assess future response of forest
ecosystems to anthropogenic stressors, (2)
detect and quantify status and trends in
forest ecosystem condition, (3) link
changes in condition to likely stressors, and
(4) identify early warning measures for loss
of integrity and sustainability of forest
ecosystems in the Pacific Northwest.
By ecological mdicator we mean a
measurement or group of measurements
that can be used to describe or predict the
condition and change of an ecosystem over
time, or of one or more of its critical
processes or components over time. Thus
indicators will include ecological measures,
stress-response functions, methods, and
process-based models, at scales from
populations to landscapes. Examples of
ecological indicators include stress-
response functions and predictions based
on simple regression models to more
complex process-based ecosystem or
landscape models; or they might include
simple measures of an ecological process,
surrogate measurements of an ecological
• Current indicators place heavy reliance on an epidemiological model which lacks the
predictive power to determine how changes in ecosystem properties (e.g., nutrient cycling,
nutrient loss, productivity, or biodiversity) will respond to stress.
• Current indicators are not sufficient to link changes to likely stressors nor identify early
warning signals.
• Current indicators cannot predict the future status of ecosystems under changed conditions.
• Current indicators do not include measures of the plantlsoil interface, which is critical in
maintaining ecosystem structure and function (Andersen and Rygiewicz 1996, Rygiewicz
and Ingham in press).
• Current indicators do not account for a plant’s ability to survive environmental stresses by
adjusting processes such as carbon allocation and partitioning, nutrient uptake, and water
use (Hogsett et al. 1985a, 1988 Andersen et al. 1991, 199Th).
• Current indicator development has not given priority to measurements that integrate limiting
factors (e.g., nutrient availability) over the growing season.
• EMAP has not developed an indicator program that adequately separates changes induced
by stress from natural cycles of disturbance from anthropogenic sources.
7
-------
process, and/or
measurements of
components.
2.2. General Approach
The research will focus on developing
ecologwal indicators that build on the
expertise of scientists at NHEERL-WED,
which includes: (1) process-based modehng
and extrapolation to landscape scales and
(2) experimental work on anthropogenic
stress-effects (e.g., ozone, and elevated
CO 2 concentrations and temperature) on
forest species, including C and N allocation
and partitioning, root dynamics, and
composition and function of the soil food
web. Results from previous and ongomg
research at WED suggest four potentially
fruitful areas of investigation for
developing ecological indicators: (I)
modeling biogeochemical cycles and
changing structure of forest stands
(Solomon and Leemans 1990, Solomon and
Bartlein 1992, McKane et al. 1997a,b), (2)
modeling and analyzing relations between
ecological processes and landscape pattern
(Cramer and Solomon 1993, Solomon and
Shugart 1993, Brown et al. 1993, 1994,
Brown and Gaston 1995), (3) measuring
processes of resource utilization by
producers (Andersen et al. 1991, 199Th;
Hogsett et al. 1985, 1988; Tingey et al.
1996, 1997), and (4) measuring resource
utilization by consumers, including soil
food web structure and function (Andersen
and Rygiewicz 1995; Porteous et a!. 1994,
Donegan et al. 1995, Seidler and
Fredrickson 1995, Widmer et a!. 1997a).
Through the combined use of field data,
remotely-sensed and other spatial data
bases, and simulation modeling we will
of produce measures, methods, stress-
or response functions, and models from which
indicators will be developed and evaluated
The ecological indicators will be derived
from various ecosystem components and
processes and represent a range of scales
from populations, to ecosystems, and to
the landscape level (Figure 2-1).
Two types of process-based models will
be included in this research: (1)
biogeochemical models with an emphasis
on C and N cycling and (2) forest
succession models to address changes in
stand structure (tree species composition
and age-size distributions) with succession
(Fig. 2-I). These models are driven by a
similar set of environmental factors and
will be used to identify key sensitive
ecosystem processes and components as
well as to assess and predict the effects of
changes in temperature and precipitation,
elevated atmospheric CO 2 concentrations,
air pollutants, N deposition, forest
management, and natural disturbance
regimes on Pacific Northwest forest
ecosystems.
Landscapes encompass a variety of
climatic, edaphic, and geomorphologic
factors all of which influence how
ecosystems respond to an individual or
suite of anthropogenic stressors. By
coupling site models with spatial data
bases of the biophysical factors, we plan to
investigate how anthropogenic stressors,
singly and in combination, affect the
functioning of landscapes (Fig. 2-1). In
particular, we plan to investigate how
different patterns of land use affect
ecological processes such as landscape-
level primary productivity, N losses, and
water quality. Furthermore, we also plan to
combinations
key processes
8
-------
investigate how other stressors such as N
deposition and climate change interact with
changes in land use to affect landscape-
level processes previously mentioned.
The field data collection phase will serve
three roles: (1) to parameterize, calibrate,
modify, and test the simulation models to
be used in this research, (2) to determine
the natural variation over time and space of
ecological processes that could serve as
indicators, and (3) to elucidate the
mechanisms that are involved in key
ecological processes. Similar measurements
of the same processes will also be collected
in on-site experimental facilities to allow us
to determine the response of the same
ecological processes under known
stressors.
Promising indicators for wider -scale
adoption for forest monitoring selected
from the research activities will be those
that have low spatial and temporal
variation and maintain a strong correlation
with changes in natural stress
Figure 2-1. Linkages among the various phases of the research plan and the outputs suitable
for assessing and monitoring forests. The task numbers and numbers in parentheses refer to
the sectional numbering scheme used in Section 3 where details for each component of the
research are given.
Site characterization (3.1)
•Climate
• Soils
• Disturbances
• Landscape pattern
Modeling
(3.2)
/
Blogeochemical cycling (3.2.1)
Field data collection
(3.3)
Tasici -
• Stand scale C and N
Task: 2
• Landscape modeling
Resource utilization (3.3)
Forest succession (3.2.2)
Producers (3.3.1)
Tasks: 44
•C& N allocation
• C & N partitioning
• Phenology
I
Task: 3
• age! size structure
• tree species composition
3
Consumers (3.3.2)
Tasks: 7.9
• C & N transformations
• C & N losses
• Soil food web
PRODUCTS:
• Models
• Stress-response functions
• Measures
• Methods
9
-------
factors in these complex environments.
These promising indicators will then be
subjected to sensitivity analyses. Here,
extensive field sites across climatic
gradients and long-term data from other
established research sites in representative
forests of the western Cascades will be
used to test the sensitivity of indicators to
changes in climate, land use/land
management, and pollutants. Finally,
indicators proven useful in specific areas
will be evaluated for regional application,
e.g., throughout forests of the Pacific
Northwest. Modelmg activities will focus
on validation to see if predictions hold
across the range of response and
environmental conditions, and ultimately to
evaluate indicator sensitivity. Indicator
selection will be based on comparative
sensitivity analyses across sites and
conditions and on EMAP selection criteria
and guidelines (US EPA 1997a).
3. RESEARCH
IMPLEMENTATION
The research will focus on developing
indicators for forest ecosystems. Most of
the research described here will be done in
field sites to determine the natural variation
over time and space of ecological
processes. This section contains a
description of the research field sites (3.1),
including data collection for characterizing
the sites. In this Project, several potential
types of indicators (Fig. 2-1) will be
evaluated and/or developed. They include:
• Modeling biogeochemical cycling at the
stand and landscape scale (3.2.1)
• Modeling forest succession processes
at stand to landscape scales (3.2.2)
• Measuring resource utilization (C and
N) by producers (3.3.1)
• Measuring resource utilization (C and
N) by consumers (3.3.2)
Similar modeling approaches and measures
of resource utilization have been or are
being investigated in three other projects at
WED: Effects of CO 2 and Climate Change
on Forest Trees (TERA I expenment),
Interactive Effects of 03 and CO 2 on the
Ponderosa Pine Plant/Litter/Soil System
(TERA H experiment), Effects of
tropospheric ozone on forest trees (Forest
Ozone project) (see Section 4. for further
details on linkages with other research
programs in WED).
3.1. Field Sites
The research activities described here will
take place in intensive field sites, extensive
field sites, and whole watersheds. Here we
describe the field sites, and types of data
that have been or will be collected to
characterize these study sites (Fig. 2-1).
These data will be used to interpret trends
in field data collection as well as to drive
the site and landscape models.
3.1.1. Intensive field sites
The intensive sites were established to
provide an understanding of ecosystem and
ecophysiological processes (e.g., nutrient
fluxes, net primary production) and to
provide data for model parametenzations.
This understanding is achieved by having a
number of investigators conduct a range of
studies and insuring that a common set of
environmental data is collected and made
available to all.
10
-------
The ecolog cal processes and forest
composition in the western Cascades varies
in response to several factors including:
climatic gradients and variability, soil type,
management activities, successional stage,
disturbances (e.g., fire and insects), and
intersite and interannual changes. To
capture the variability associated with
these factors, we have established and will
establish additional intensive field sites in
the Western Cascades. We will measure a
suite of parameters at our intensive sites
for a minimum of 5 yr to capture the
interannual variation in ecological processes
resulting from climatic vanability.
Our proposed indicators research project
will take advantage of established sites
when possible (Appendix B). For example,
existing research sites established at the H.
J. Andrews Expenmental Forest offer the
advantages of long-term data bases on
forest ecosystem response. However, we
believe it is also important to establish a
core set of sites that can be measured
and/or manipulated stnctly for the
purposes of the research proposed here.
There are two main reasons for this:
1) To be of optimum use as for our
project, most proposed measures
must be linked to a data set that
describes all of the major pools and
fluxes of C and N in the ecosystem,
as well as all of the environmental
conditions that may have a bearing
on the response of a measure. To
our knowledge, no other existing
sites outside of our current intensive
research sites meet this need. For
example, there are sites in Oregon
and Washington where research
projects are already underway to
describe ecosystem C budgets in
detail (e.g., OTTER and Wind River
Canopy Crane), however,
complimentary data on N budgets
are being given less emphasis.
Because of the importance of N in
limiting forest growth, it is essential
that C and N budgets be descnbed
together for the same sites, not
inferred from other nearby sites that
may or may not have similar soils,
vegetation or climate.
2) Activity or “wear and tear”
associated with intensive
measurements at a research site can
have adverse effects on ecosystem
processes. For example, foot traffic
can disrupt growth of fungal hyphae
and fine roots or alter rates of net N
mmeralization. Thus, a major
challenge in conducting our research
is to avoid the Heisenberg Principle,
whereby the act of measurement
altera the very thing that is to be
measured. By establishing sites that
we are responsible for managing, it
will be logistically easier to regulate
where foot traffic is allowed and
what areas are available for different
research activities. This becomes
very important in providing
researchers with a history of what
activities have taken place in any
given location, so that potential
problems can be avoided. Our goal
will be to facilitate collection of high
quality data while providing for the
long-term maintenance of the
research sites.
11
-------
Current field sites
Our current research sites include both low
and high elevation forests and clearcuts
located in the western Cascades along the
Highway 20 corridor between Sweet
Home, OR and the Santiam Pass (Fig. 3-1).
These sites differ in climate (e.g., growing,
temperature, precipitation form and
amount) and soil nutrient levels. In this
area many disturbances typical of the
western Cascades are at work including
flooding, insect and disease outbreaks, and
wind storms. However, fire and human
impacts (including logging and road
building) are the dominant disturbances
that have affected the conditions within
these forests. Historic fire return intervals
vary from 25 to 110 years for low-
intensity ground fires, to 100-200 years
and more for high-intensity stand-
replacement fires (Morrison and Swanson
1990).
The Falls Creek Site is located at an
elevation of 537 m in the western hemlock
(Tsuga heteropyhila) vegetation zone
(Franklin and Dyrness 1988) in the South
Santiam River drainage. The development
of dense, essentially even-aged stands of
Douglas fir is a common occurrence after a
wildfire or logging. These stands can be
dense enough to delay the establishment of
understory vegetation. Re-establishment of
the characteristic understory and invasion
of western hemlock occurs as mortality
begins to open up the overstory (Franklin
and Dyrness, 1988). Historic records reveal
a large stand-replacement fire (Moose
Mountain fire) in 1856 followed by a 100
year storm in 1861. However, free-ring
analyses show that the oldest Douglas firs
are 100 to 110 yr old. In 1969, there was a
commercial thinning. About 16 ha was
clear-cut at the Falls Creek Site in 1988,
broadcast burned in 1990, and replanted
Figure 3-1. Topographic map showing the locations of the current research sites in relation to
other points of interest.
12
-------
to Douglas fir in 1993. The area is
currently managed as elk wintering and
calving habitat; consequently, the area was
aerially seeded with legumes and grasses.
To maintain the forage, most of the clear-
cut, except where we are conducting our
studies, was fertilized (spring, 1994 and
1996).
We established research plots in both the
clear cut and the adjacent 140-year-old
forest in 1995. Douglas fir is the dominant
overstory tree with western hemlock and
vine maple occurring in the understory.
The current stand density for trees >20 cm
dbh is approximately 238 trees/ha. In 1995
we established a 75 m x 90 m study area in
the forest in which we sampled 12
randomly located 15 mx 15 m plots. Trees
range in diameter from 2 to 90 cm and up
to 60 m tall (Table 3-1). A vegetation
survey in 1996 found a total of 16 plant
genera (not including tree species), with a
mean percent cover in m 2 forest plots of
49%.
Vegetation studies in the clear cut in 1996
indicated the presence of at least 16 plant
genera, including the four genera that were
aerially seeded in 1990. Mean percent
cover in the low site m 2 clearcut plots was
99%.
Table 3-1. The composition and typical sizes of the trees on Falls Creek and Toad Creek Road
Site forest plots.
Plant species
Mean and max. height (m) Mean and max. diameter (cm)
Falls Creek Site
Douglas fir
49, 64
61, 91
Western hemlock
4, 36
4, 27
Toad Creek Road Site
Douglasfir
44,61
74,111
Western hemlock
7, 45
12, 76
Pacific silver fir
7, 29
12, 46
The Falls Creek Site is located on deep to
very deep well-drained soils derived from
colluvium, glacial till, and alluvium (Legard
and Meyer 1973). Bedrock materials
consist of andesites, basalt, tuffs and
breccias. This soil type is typically found
between about 460 and 1370 m. A
tentative classification of this soil is coarse-
loamy, mesic, Typic Hapludand
(Appendix A).
The western hemlock vegetation zone has a
wet, mild maritime climate (Franidin and
Dyrness 1988). As the Low Site lies on the
eastern side of this zone some distance
from the ocean, there can be large seasonal
variations in soil moisture and temperature.
13
-------
Precipitation occurs mamly during the
winter; summers are relatively dry and
account for only 6 to 9% of the total
precipitation (Franklin and Dyrness 1988).
Meteorological stations were established in
the clear cut in 1994 and in the forest in
1995. Table 3-2 shows the types of
environmental data collected in the clear
cut and forest locations and Table 3-3 gives
typical values for 1995.
Table 3-2. Types of meteorological and soils data collected at the Falls Creek and Toad Creek
Road Sites in both the clear cuts and late successional forests.
Parameters
.
Falls Creek Site
Toad Creek Road Site
Clear-cut Mature forest
Clear-cut Mature forest
Air temperature - C
Relative Humidity - %
Solar Radiation - PAR
Solar Radiation - Total radiation
Precipitation - mm/hr
Wind speed - rn/sec
Soil temperature - C (several depths)
Soil moisture - % (TDR)
Soil moisture - % (Reflectometer)
Snow depth
J
‘J
‘ .1
sJ
I
I
I
I
, I
I
I
. ,J
I
I
i
I
-
I
Data are collected as 1-hour averages except for soil moisture which is collected as 3-hour
averages or periodically depending on the measurement system.
Table 3-3. Typical climate conditions measured in the clear cut at the Falls Creek and Toad
Creek Road Sites for the calendar year 1995.
Mean annual Mean annual Mean annual Total annual Growing Maximum
air temperature relative daily total precipitation degree-days snow depth
(°C) humidity (%) PAR (jimol (mm) (cm)
m 2 day’)
Falls Creek Site
10.5 82 7246 1798 3656 NA
Toad Creek Road Site
7.6 76 6895 1947 2673 127
14
-------
The Toad Creek Road Site is located at an
elevation of 1220 m in the Pacific silver fir
(Abies amabalis) vegetation zone (Franklin
and Dyrness 1988) in the upper reaches of
the McKenzie River drainage. In this
vegetation zone, a typical successional
sequence begins with the establishment of
Douglas fir and/or noble fir (Abies
procera), followed by the shade tolerant
western hemlock and Pacific silver fir
which typically develop later under a
forest canopy (Franklin and Dyrness
1988). Up to 500 years following a
disturbance, a typical mixed stand includes
scattered, large Douglas fir, abundant but
smaller western hemlock, and abundant
seedlings, saplings and poles of Pacific
silver fir (Franklin and Dyrness, 1988). In
the high site forest, 28 plant genera
(excluding the trees species) were identified
in 1996, with a mean percent cover in m 2
plots of 78° .
Douglas-fir is the dominant tree species in
the Toad Creek Road Site forest with
western hemlock and Pacific silver fir
occumng in the understory. The dominant
Douglas firs are between 200-220 yr old
based on tree-ring counts. The even-aged
structure of the dominant Douglas-fir in
this forest is consistent with typical
patterns of natural regeneration following
high-intensity stand-replacement fires.
Current stand density of trees >20 cm dbh
is approximately 267 trees/ha. This forest
is classed as late successional/old growth.
In 1995 we established a 75m x 75 m area
in the forest in which we sampled 11
randomly located 15 in x 15 m plots. The
composition and sizes of the trees on these
plots is given in Table 3-1.
Following the 1991 clear cut of about 18
ha, the site was replanted in 1994 with a
mixture of Douglas-fir (15.3%) noble fir
(73.1%), grand fir (Abies grandis) (7.5%)
and western white pine (Pinus monticola)
(4.2%) at a density of 212 trees/ha. The
site was not burned following the clear cut
and most of the large woody debris was
left on the site. In 1996, a vegetation
survey found 28 plant genera with a mean
percent cover in m 2 plots of 42%.
The Toad Creek Road Site is located on
soils denved from volcanic ejecta and
glacial till overlying bedrock of hard
andesites and basalt (Legard and Meyer
1973). This soil type occurs at elevations
of 850 to 1250 m on glacially smoothed
lava flows. In general,, the soil has a fine
loam to loam texture with a medium
granular structure. The soil has been
classified as a coarse-loamy, mixed, frigid,
Typic Hapludand (Appendix A).
The Pacific silver fir vegetation zone is
wetter and cooler than the adjacent western
hemlock vegetation zone and receives
considerably more precipitation in the form
of snow. Winter snow packs of 1 to 3
meters are common (Franklin and Dyrness
1988). Meteorological Stations were
established in the clear-cut in 1994 and in
the forest in 1995. The types of
environmental data collected and typical
climate conditions at the High Site are
shown in Tables 3-2 and 3-3, respectively.
3.1.2. New Research Sites
To accomplish the scientific objectives of
the Project, we will establish additional
research sites. These sites will be selected
to: (1) meet one or more of the site
selection cnteria listed below, (2) maximize
15
-------
the opportunities for multiple investigators
to use the same sites and (3) be located
adjacent to or near a meterological station
operated by the Project or other group
(e.g., SNOTEL).
Critena for site selection: (1) provide
replication of the existing Falls Creek and
Toad Creek Road sites, (2) test specific
hypotheses and (3) provide data for model
development or testing.
Site Replication - A major constraint of
most ecological research is the lack of
replicated field sites. This lack makes
difficult the interpretation of variation
observed at the unreplicated sites. We
propose to establish new research sites to
serve as replicates of the existing sites
(Falls Creek and Toad Creek Road) and
also establish the variation in ecological
processes due to the intersite variability.
The replicates are Moose Mountain for
Falls Creek and Soapgrass Mountain for
Toad Creek Road. The Moose Mountain
site is at approximately the same elevation
and contains the same general vegetation as
the Falls Creek site (Douglas Fir, Vine
Maple, Oregon Grape and Salal). The land
type is also the same between the sites but
Moose Mountain has a southern exposure.
A battery of soil analyses will be obtained
to compare physical and chemical
composition with that of the existing Falls
Creek site. The Soapgrass Mountain site
also has a similar elevation, climate and
vegetation type as the Toad Creek site.
Initial measurements suggest that it may be
a wetter site with a high soil nitrogen level.
If detailed measures confirm the initial
observations, these differences will be
accounted for in comparing sites.
Test Specific Hypotheses - Several
ecophysiological processes are outlined in
tasks 4 through 6 for indicator
development. These studies are
appropriately addressed by testing specific
hypotheses. To achieve the conditons for
testing may require additional sites or the
use of existing sites.
For example, previous studies have
established baseline biological and nitrogen
fixation data for the Falls Creek site. A new
experimental site at Moose Mountain has
recently been established to carry out
intensive soil biological analyses and
nitrogen fixation studies to compare with
the existing data from Falls Creek. Also the
Moose Mountain site will be used to
establish a data set prior to, immediately
following and for about 5 years from when
the clearcut is established in the summer of
1998. Replanting with Douglas Fir will
take place within 12 months after the
clearcut is established. This unique
management scenario will allow us to
follow changes in the food web biota and
nitrogen fixation over a time sequence of a
major management perturbation
Within the Moose Mountain study area
there is a 41 acre clear cut (designated as
powder Regen III) scheduled for the
summer of 1998. A forested site, several
hundred feet from the clear cut on a similar
souti) facing slope, as has been identified as
a control. Meteorological stations will be
located in both the forested and clear cut
sites to monitor soil and litter temperature
and moisture, air temperature, precipitation
and wind speed. Thus, comparison of the
kinds of diversity of biological endpoints
with the Falls Creek site can be linked to
similarities or differences in the
16
-------
meteorologic 1 conditions as well as to
major management disturbances e.g. the
clear cut (see Subtask 9b for a detailed
explanation of the endpoints to be
examined).
A second example of site selection to test
specific hypotheses involves Toad Creek
Road and Soapgrass Mountain. The
hypothesis is that N retranslocation is
influenced by soil N levels. These two sites
were selected because that have similar
elevation, climate and vegetation
characteristics but differ in soil N. Total
soil N decreases 0.4% at Soapgrass to 0.1%
at Toad Creek Road.
Model Development - Parameterization,
calibration and testing of biogeochemical
and stand models requires a range of data
from a number of different sites. The
existing intensive sites (Falls Creek and
Toad Creek Road) have been used to
develop a parameter set for the
biogeochemical model. Additional sites will
be selected to: (1) estabLish the maximum
biomass which is needed to constrain the
model simulations, (2) provide
geographically extensive data to extend the
current biogeochemical model parameter set
to a regional parameter set using data from
the Cascade Center for Ecosystem
Management Permanent Study Plots and
(3) test the accuracy and reliability of the
regional biogeochemical model parameter
set. The model will be tested using new
data from a range of sites in the Olympic
National Park representing a range of
vegetation types, precipitation ranges and
elevations.
Additional Climate Stations - To provide
climate and soil data, we will also establish
additional weather monitoring sites
concurrent with the establishment of the
new research sites. In addition to our
monitoring sites, the US Department of
Agriculture, Natural Resources
Conservation Service maintains a number
of SNOTEL sites within the Santiam River
Basin and adjacent drainages where daily
temperature (maximum, minimum and
mean), precipitation, and snow water
equivalents are measured. If we locate some
of our new plots adjacent to the SNOTEL
sites we will use their data rather than
establishing our own sites.
3.1.3. Watersheds
Abundant spatial data bases on
subwatershed boundaries, seral stage,
climatic, soils, terrain (digital elevation
models [ DEMs] of 30 m to 500 m
resolution), ownership, harvest history,
and vegetation are available for most of the
Pacific Northwest region. These data bases
include products derived from remotely-
sensed imagery ranging from fine-scale
aenal photography and Landsat T M
satellite imagery with spatial resolutions of
30 m or less, up to more coarse scale
AVHRR imagery at resolutions of 1.1 km.
For example, Cohen et al. (1996) have used
Landsat TM imagery to map six forest
successional stages and harvest incidence
between 1972 and 1991 in a 1.2 x 106 ha
central Oregon Cascades landscape.
We propose to work at the watershed scale
initially within the 5.1 x lO ha South
Santiam River Basin. This is also where our
Low Site is located (Section 3.1.1). Situated
in one of the major drainages of the western
Oregon Cascades, the watershed has been
described by ownership, stream systems,
17
-------
current and historical seral stage, road
network, and fire history (Sweet Home
Ranger District 1995). The watershed is
composed of 10 subwatersheds ranging in
size from 1097 to 9630 ha and varying
from stand initiation to late-
successioriallold-growth seral stages.
Copies of the digital maps of these data
have been obtained. Further, the data of
Cohen et al. (1996) have been obtained.
3.2. Modeling
Models play a prominent role in risk
assessment because they are the primary
means for relating stressors to probable
effects, a conceptual basis for integrating
diverse measures into a self-consistent
framework, and for making meaningful
extrapolations across scales of time, space,
and biological organization (Suter 1993b,
Rapport 1992, Rastetter 1996). The
synergistic interactions in nature among the
various environmental driving forces, such
as temperature, precipitation, nutrient
inputs, topography, and soil moisture,
make it impossible to predict or assess
future response of ecosystems to
anthropogenic stressors, such as air
pollutants, climate change, and land use,
based on single-factor experiments alone
(Rastetter et al. 1991). Process-based
models such as biogeochemical cycling
models or forest succession models can
help improve such assessments by
providing a self-consistent synthesis of the
results of many experiments. The
synthesis provided by these models
includes the interactions among ecosystem
processes that give rise to the synergistic
responses to multiple factors.
The problem with models, however, is that
their long-term predictions are impossible
to test unambiguously except by allowing
enough time for the full ecosystem
response to develop. Unfortunately, when
potentially significant changes to
ecosystems must be assessed, time
becomes a luxury (Rastetter 1996).
Confidence in models, therefore, has to be
built through evaluation agamst
corroborating evidence. Such evaluations
may include: (1) comparisons against
short-term experiments (less than 10 yr),
(2) using space for time substitutions, (3)
using reconstruction of past responses, and
(4) compansons with other similar model
outputs (Rastetter 1996). Although these
methods can be used to evaluate models,
none can meet the crucial needs for rigorous
testing. This does not mean to downplay
the importance of models; their advantage
is that they can be used to synthesize all
the empirical evidence in a self consistent
set of interpretations (Rastetter 1996).
Thus models are a vital part of any nsk
assessment of the response of ecosystems
to anthropogenic stressors.
Two models will be used in this research
plan to integrate experimental and field data
for predicting effects of anthropogenic and
natural stressors, for developing stress-
response data bases, and for determining
sensitive processes or components that can
serve as potential early warning indicators
The two models, a biogeochemical model
(Section 3.2.1) and a forest succession
model (Section 3.2.2) focus on different
aspects of forests, and complement each
other as predictive tools. Indeed, the
dynamic forest processes which they
simulate are entirely different, although
several model output variables can be
18
-------
compared between the two models (e.g.,
stand biomass, stand leaf area, nutrient
content, etc.).
The biogeochenucal models focus on the
processes of carbon and nutrient utilization
(acquisition, allocation, and partitioning) in
plants and soils and how utilization
patterns are impacted over time by a suite
of anthropogenic stressors such as changes
in climate or atmospheric composition, and
by forest management In contrast, forest
succession models focus on year to year
stability and change in forest tree size
cl sses, age classes and species, as
controlled by establishment, growth and
mortality processes. The models mimic the
gradual replacement of fast-growing and
shade-intolerant trees by slower-growing
and shade tolerant species, generating the
time-ordered implications of this
successional process to shifts in
community-level carbon storage, soil
nutrient status, and vuEnerability to
anthropogenic stressors. The succession
models evaluate how forest stand structure
and composition are modulated over time
by changes in climate, atmospheric
chemistry, species availability and
management choices.
Both models will be used to simulate forest
dynamics at local scales, and at landscape
scales by linking each model to additional
spatial data bases. We also plan to integrate
(“hard wire”) the two kinds of models to
simulate and evaluate the importance of the
feedback processes between
biogeochemical and structurall
compositional attributes and processes,
although the tasks required to attain this
longer range goal cannot be defmed until
considerable progress has been obtained on
the separate model development paths
proposed below (Sections 3.2.1, 3.2.2).
The two models require unique suites of
parameters, and a common set of driving
variables. The data collected for site and
landscape characterization (Section 3.1)
will serve as some of the driving variables
for both models. In the following sections,
we present the approaches and data needs
to parametenze, simulate and evaluate each
model at the stand and landscape scale.
3.2.1 Blo geochemical cycling
modeling
Task 1: ModelIng at the stand
scale
Objectives and background
The basic objectives of this task are to
parametenze, evaluate, and verify a model
of biogeochemical cycling of C and N in a
forested ecosystem at the stand scale (Fig.
3-2) for use as an assessment and
predictive tool. The large number of
ecosystem processes and components that
may be affected by anthropogenic stressors
constrains the assessment of stressor-
effects by experimentation alone. An
alternative is to use process-based models
to predict how forest ecosystems will
respond to existing or projected scenarios
of stressors. Specifically, we will use a
biogeochemical cycling model to determine
the effects of natural and anthropogenic
stressors combined with future forest
management actions on PNW forests. For
example, the model can be used to
determine the relative risk of increased
temperature or N-deposition on forests
under different harvesting scenarios or to
assess the likely effects of forests on
various actions taken to control
19
-------
atmospheric CO 2 concentrations. The
model also directly supports risk
assessment by providing a tool that can be
used in the problem formulation and nsk
characterization steps. The experimental
data developed to parameterize the model
and the results of the model simulations
will provide stress-response functions (i.e.,
effects data) that are also needed to
conduct a risk assessment for PNW
forests.
Selection of biogeochemical
model
There are a number of models available that
simulate biogeochemical cycling in forest
ecosystems (Perruchoud and Fischlin
1995), including CASA (Potter et al. 1993),
CENTURY (Parton et al. 1987), FOREST-
BGC (Running 1994), LINKAGES (Pastor
and Post 1986), MBL-GEM (Rastetter et
al. 1991), and TEM (Raich et al. 1991).
While all of these models provide a
process-based view of ecosystem C and N
cycling, they differ with respect to model
structure (number of plant and soil
compartments), the incorporation of
particular processes, coupling with the
abiotic environment, and method of model
calibration. For our objectives, we require a
model that can address the responses of
forest ecosystems to changes in CO 2 .
climate, N deposition, and air pollutants,
and that these responses will encompass
enzymatic controls on C and N acquisition,
stoichiometric shifts in tissues, changes in
plant biomass allocation among tissues,
altered rates of organic matter turnover and
N mineralization, and ultimately a
redistribution of C and N between
vegetation and soils. Because it would be
extremely difficult to obtain sufficient fine-
scale data to characterize many of these
processes, we also require a model that can
be calibrated to mfer the needed
information from data that are more easily
obtained, namely, data collected at the
scales of ecosystems (e.g., net primary
production) and regions (e.g., vegetation C
stocks along temperature and precipitation
gradients).
Figure 3-2. Illustration of the various
steps and feedbacks that will be taken to
develop a verified biogeochemical model
for use as an assessment tool.
We have selected the Marine Biological
Laboratory General Ecosystem Model
(MBL-CIEM) (Rastetter et al. 1991) as our
primary working model because it most
closely meets the preceding requirements,
20
-------
i.e., it provides a process-based view of the
acquisition., allocation, and chemical
partitioning of C and N in plants and soils,
and it can be calibrated using ecosystem-
and regional-scale data. In addition, a
preliminary paranieterization of the model
has already been established for Pacific
Northwest forests (Appendix C).
Description of MBL-GEM
The MBL-GEM is a process-based model
of C-N interactions in terrestrial
ecosystems. Its structure (Fig. 3-3) is
described in detail in Rastetter et al. (1991).
The model is intended to be generally
applicable to most terrestrial ecosystems
and has been used in the past to analyze
the responses of temperate deciduous
forests, tropical evergreen forests, and
arctic tundra to changes in CO 2
concentration, temperature, N inputs,
irradiance, and soil moisture (Rastetter et
al. 1991, 1992a, 1997; McKaneet al. 1995,
1997a, 1997b).
The MBL-GEM simulates, at the stand
level, photosynthesis and N uptake by
plants, allocation of C and N to foliage,
stems, and fine roots, respiration in these
tissues, turnover of biomass through
litterfall, and decomposition of litter and
soil organic matter. The model currently
simulates responses to changes in
atmospheric C0 2 , temperature, soil
moisture, irradiance, and inorganic N inputs
to the ecosystem. Carbon dioxide is lost
from the ecosystem through plant and soil
respiration. Inorganic N losses are assumed
to be proportional to inorganic N
concentrations in soil. The model calculates
all changes on a monthly time-step. To be
consistent with the time-step, monthly
averages are used for all climate drivers.
Three major features of the model are
important to our application. First,
vegetation in the model acclimates to
changes in the environment to maintain a
nutritional balance between C and N
(flexible within specified C:N ranges).
Thus, environmental changes that stimulate
photosynthesis (e.g., increased CO 2 or
higher irradiance) result in an increase in
allocation of C and N to fine roots, thereby
stimulating N uptake. Similarly,
environmental changes that stimulate N
uptake (e.g., high inorganic soil N
concentration) increase allocation of C and
N to foliage, thereby stimulating C uptake.
A second important feature of the model is
that respiration rates of plant tissues are
proportional to the amount of
metabolically active N in those tissues
(Ryan 1991). Thus, increased N
availability increases productivity and
growth in the vegetation, and it also
increases the rate of plant respiration per
unit biomass if the N concentrations in
tissues increase. Finally, high N availability
in soil stimulates the rates of
decomposition for organic soil fractions
like cellulose that have high C:N ratios. In
effect, increases in N availability stimulate
decomposition by facilitating the
conversion of high C:N to low C:N
products. This is consistent with evidence
that increased N availability produces a
“printing” or accelerating effect on overall
rates of decomposition (Gill and Lavender
1983, Hunt et al. 1988).
21
-------
Co 2
Parameterization of MBL.GEM
The MBL-GEM is a lumped-parameter
ecosystem model, meaning that it is a
spatially averaged representation of
0
I
0
Figure 3-3. Schematic diagram of carbon and nitrogen cycles in the MBL-GEM (Rastetter et
al. 1991)
ecosystem processes (e.g., Cosby et al.
1985). Because of the non-linearity of the
process equations and because of the
intense sampling required to characterize
the spatial heterogeneity of most
22
-------
ecosystems, iumped-parameter models are
very difficult to parameterize from the
bottom up (Beven 1989). That is, it is
difficult to use fine-scale measurements of
individual processes like leaf-level
photosynthesis to parameterize the model
(Rastetter et al. 1992b). Instead, the
process equations in the model are “scaled”
by calibrating them to data that are
collected at the same scale at which the
model is to be applied (e.g., ecosystem-
level net primary production). The MBL-
GEM does this by inverting the process
equations so that the model parameters can
be either directly calculated or estimated by
least-squares from the system-level
calibration data (this is equivalent to what
Jarvis [ 1993] calls “optimization”). The
rate constants for each C and N flux within
the vegetation, the rate constants for
humus turnover, and the C:N ratio of
humus are calculated in this way.
This “top down” approach to
parameterizing MBL-GEM is similar to
the inverse modeling approaches used in
geophysical and global C cycle modeling,
where general circulation models and
observed atmospheric CO 2 concentrations
are used to estimate unmeasured CO 2
sources and sinks (e.g., Enting and
Mansbndge 1989, Cials et al. 1995).
Although this approach is not often used in
ecology it is extremely useful for our
purposes, and for ecosystem science in
general, because data for fine-scale
processes (e.g., leaf-level photosynthesis)
are often not available or are more difficult
to obtain than data for coarse-scale
processes (e.g., ecosystem-level net
primary production). McKane et al. (1995
and 1997a) discuss in detail how this
approach was used to parametenze the
MBL-GEM for tropical evergreen forest
and arctic tundra.
The data needed to parameterize MBL-
GEM (Table 3-4) describe the distribution
and fluxes of C and N among the various
vegetation and soil compartments and
concomitant changes in the environmental
drivers. Because our goal is to derive a
single parameterization of the model that is
generally applicable to forests throughout
the Pacific Northwest, we will obtain these
data for a number of sites representing a
range of age classes and climatic and
edaphic conditions (Section 3.1 Site
Characterization).
Methods used to collect
parameterizing MBL-GEM
data for
We will use the methods descnbed below
and in the various subtasks in Section 3.3
to obtain the vegetation and soil data from
the intensive field sites needed for model
parametenzation (Table 3-4). Two
categories of methods are described that
represent intensive measures and simpler
measures. The mtensive measures are most
accurate and will provide the primary data
for model parameterization. The simpler
measures will be developed in conjunction
with the intensive measures, then applied
to an extensive network of less intensively
studied sites to evaluate their use as
regional-scale indicators of ecosystem
response to change. These indicators may
be used for model parametenzation when a
measure is well-correlated to an intensive
measure or, alternatively, they may be used
for landscape-scale model validation when
a measure provides a relative index of
change among sites.
23
-------
C and N Stocks
Vegetation C and N stocks.
Stem C and N . Stem C and N stocks (above
and belowground woody tissues exclusive
of fine roots < 2 mm in diameter) will be
calculated by multiplying estimated stem
biomass by measured stem C and N
concentrations for each species. Total stem
biomass will be estimated allometrically
from measurements of tree diameter and
height. The live (sapwood) and dead
(heartwood) components of stem biomass
will also be determined by analysis of tree-
ring increment cores (intensive measure), or
by the allometric relationship of sapwood
area to leaf area index (simpler measure).
The optimum number and size of sample
quadrats at each field site will be
established using methods described by
Roberts et al. (1993). For our Cascade field
sites, we estimated total stem biomass
using a minimum of eleven 15 m x 15 m
plots randomly located within 0.6 ha
blocks.
Carbon and N concentrations of sapwood
and heartwood, and all other plant and soil
materials mentioned in this section, will be
determined using gas chromatography
methods (Carlo Erba) (LJSEPA SOP 3.01
Carbon/Nitrogen Analysis).
LeafCandN . Leaf C andN stocks will be
calculated by multiplying estimated leaf
biomass by the C and N concentrations of
live leaves collected at each site. Leaf
biomass will be estimated from LA! as
described under Subtask Ia. Estimated LA!
will be converted to leaf biomass on a stand
basis using the relationship between leaf
biomass and specific leaf area for live leaf
samples. Species’ contributions will be
weighted by their fraction of total stand
basal area.
Fine root C and N . Fine root (<2 mm) C
and N stocks will be calculated by
multiplying estimated fine root biomass by
measured C and N concentrations. Fine
root biomass will be estimated using
standard coring methods (Santantonio et al.
1977; see also Subtask 4a). At least ten 5-
cm diameter x 20-cm deep cores will need
to be collected per site (Vogt et al. 1981),
and additional coring to 30 cm will be made
for about half of these samples. Using
these intensive measures as a reference, we
will also investigate whether a simpler
measure, or indicator, of fine root biomass
can be developed using allometric
relationships to leaf biomass and/or
sapwood area.
Forest floor and soil C and N stocks
Coarse woody detritus C and N . Coarse
woody detritus includes standing dead
trees and fallen logs ? 10 cm diameter
(Harmon and Sexton 1996) and will be
sampled on the same 15 m x 15 m plots
used for live stems (see above). All coarse
woody detritus on each plot will be
measured for length, large and small end
diameters, and decomposition class (four
classes ranging from no decomposition to
highly decomposed). The C and N content
of coarse woody detritus will be
determined by calculating volume,
converting volume to dry weight based on
published wood densities by
decomposition class (Grier and Logan
1977), and multiplying dry weight by
measured C and N concentrations.
24
-------
Table 3-4. Data needed to parameterize the MBL-GEM. Data marked with an asterisk (net •N mineralization and soil respiration) are
not essential for parameterizing the model but are important as a check on the calibration procedure (the model calculates rates for
these processes to be consistent with mass balance requirements). The methods used to collect these data are classified into two
categories representing intensive/difficult measures and simpler measures.
Intensive/Difficult Measures
Simpler Measures (Indicators )
C & N Stocks (Mg fra )
Vegetation C & N
Stems (above and belowground wood)
Sapwood/Heartwood
Leaves
Fine roots (<2 mm)
SoilC& N
Soil Organic Matter
Litter layer (fine + coarse litter)
Mineral soil humus
Inorganic soil N (ammonium + nitrate)
C & N fractions (extractives, cellulose, lignin)
Leaf, wood, and fine root litter
Soil humus
C & N Fluxes (M ha’ yr’ )
•Net primaly production C & N
Stems (above- & belowground wood)
Leaves
Fine roots (<2 mm)
Litterfall C & N
Stems (aboveground wood)
Leaves
Vegetation N uptake
Leaf N retranslocation
N leaching Lysimeters
Allometrically from diameter & height
Tree increment core analysis
Calculate from LA! & leaf mass/area
Soil core analysis
Destructive quadrat samples
Soil core analysis
KCI extraction of soil cores
Chemical fractionation (subtask 5a)
Chemical fractionation (subtask 7a)
Tree-ring analysisfallometry
Litterfall traps
(I) Root turnover, (2) N budget
Annual total plot surveys
Litterfall traps
Calculate from N requirement of NPP
Subtask 4b
Allometncally from diameter only
Allometrically from LA!
Allometrically from stem diameter
Allometrically from leaf biomass
Estimate from depth & bulk density
Dendrometer bands
Needle age class analysis & allometiy
Annual line intercept surveys
‘ -Il
-------
Net N mineralization* Incubation of soil cores Resin bags
Soil respiration* Infrared gas analysis & Subtask 8a
Other Data Intensive/Difficult Measures Simpler Measures (Indicators )
Leaf area index (projected m 2 /m 2 ) Subtask 4a
Rooting depth Visual inspection of soil cores & pits
Environmental Data
Atmospheric CO 2 concentration Infrared gas analyzer
Photosynthetically active radiation Quantum sensors
Air temperature Temperature sensors
Soil temperature Buried temperature probes
Soil moisture Gravimetric & TDR methods
Atmospheric N deposition Dry & wet deposition collectors
-------
C and N Fluxes
Forest floor C and N . The forest floor has
been operationally defined for Pacific
Northwest forests as the 01 and 02
horizons plus woody matenal < 10 cm in
diameter (Harmon and Sexton, 1996).
Because of the high spatial heterogeneity of
forest floors in this region, about fifty 0.1
m 2 forest floor quadrats per site will need
to be sampled at each site. Forest floor C
and N content will be determined by
multiplying ash-free dry mass per unit area
by the concentrations of C and N. In
addition to these intensive measures, we
will investigate whether a simpler measure
of forest floor biomass can be developed
from depth and bulk density
measurements.
Mineral soil C and N . Mineral soil samples
will be collected at 20 cm increments to a
depth of 1 m from several randomly
located pits at each field site. Mineral soil
C and N content will be determined using
measured C and N concentrations, bulk
densities, and gravel and stone contents.
C & N proximate fractions in litter and
mineral soil humus (extrachves, cellulose,
lignin)
The chemical composition of soil organic
matter is an important control on its
decomposition and is represented in MBL-
GEM as proximate fractions of extractives,
cellulose, and lignin. The C and N content
of each of these fractions will be
determined for fresh plant litterfall (leaves
and wood), fine roots, and soil humus using
the procedure described by Ryan et al.
(1990 (see Section 3.3, Subtasks 5a and 7a
below).
Net primary production.
Stem NPP . Stem NPP includes the annual
increment of all above- and belowground
woody biomass. Whereas tree-ring analysis
can be used to estimate stem NPP for past
years, dendrometers (expandable metal
bands that track radial growth; Lassoie
1973 Keeland and Sharitz 1993; Telewski
and Lynch 1991) are used to estimate
current rates. For both methods, stem NPP
is calculated using published allometric
equations that relate stem biomass to
diameter and height (Means et al. 1994).
That is, stem NPP is the difference in total
stem biomass before and after accounting
for the measured diameter increment for 1
yr. We are estimating stem NPP for the
past 10 years at our existing field sites by
analyzing tree-ring samples for all trees
over 1 cm dbh on the 15 x 15 m plots
described above. Current monthly changes
in stem NPP will be estimated using
dendrometer bands attached at breast
height (1.37 m). For these estimates, trees
will be sampled by 20 cm diameter classes,
with the number of samples per class being
weighted according to the frequency
distribution of tree diameters.
Dendrometers provide a relatively simple
and easily deployable method for
measuring stem NPP across a number of
field sites.
Leaf NPP . Leaf NPP for closed-canopy
coniferous forests is approximately equal
to annual dead plus live leaf litterfall.
Because storm events may contribute to
interannual variations in live leaf litterfall,
leaf NPP estimates will be based on an
average of 2-3 years of litterfall data. Live
27
-------
and dead leaf litterfall will be determined
from monthly samples of total litterfall at
each site (see “Litterfall C & N” below).
Using these intensive measures as a
reference, we will also mvestigate whether
a simpler measure of leaf NPP can be
developed from the combination of leaf
biomass estimates and leaf (needle) age
class analysis.
Fine root NPP . We will estimate fine root
NPP using two primary methods. One
method is to divide measured fine root
biomass (see “Fine root C and N,” above)
by the turnover time of fine roots measured
in minirhizotrons. Minirhizotrons have
already been installed at the low and high
Cascade field sites, and new fine root
production and mortality (Tingey Ct al.,
1996, 1997) have been measured monthly
since the spring of 1995 and will continue
(Section 3.3, Subtask 6a). We will also
estimate fine root NPP using the N budget
method of Nadethoffer et al. (1985). That
is, by subtracting the N requirement of leaf
and stem NPP from annual net N
mineralization (see below), the remaining
amount of net N mineralization can be used
to calculate fine root NPP. Using both of
these methods should provide a means of
cross checking our data.
Litterfall C and N
Litterfall includes all aboveground fine
(leaves, cones, woody material < 10 cm in
diameter) and coarse (woody material> 10
cm in diameter) detritus production. Fine
litterfall will be measured using fifty
randomly located 0.3 in 2 screen-lined traps
at each of the forested sites (Grier and
Logan 1977). We will measure coarse
litterfall at our intensively studied field
sites by annually resampling the 15 m x 15
m coarse woody detritus plots (see above).
At less intensively studied sites, coarse
litterfall will be measured using the line
intercept surveys annually (Harmon and
Sexton 1996), a less accurate but much
faster method. The C and N content of all
fine and coarse litterfall components will be
determined by multiplying dry mass by
measured C and N concentration.
Vegetation N Uptake
Vegetation N uptake will be calculated as
the N requirement of above- and
belowground NPP. The basic calculation is
[ NPP x N concentration for leaves, wood,
and fine roots] minus leaf N
retranslocation.
N leaching losses
Leaching loses of ammonjum-N, nitrate-N,
and dissolved organic nitrogen (DON) will
be measured using porous cup lysimeters
buried at a depth of 1 m. Soil solutions will
be sampled at monthly intervals or as soil
moisture conditions dictate. The
concentrations of ammonium-N, nitrate-N
in solution will be analyzed
colorimetrically using autoanalyzer
methods (Technicon 1977, USEPA 1979).
DON will be determined as the difference
between total dissolved nitrogen (TDN)
and inorganic N, with TDN being
determined by persulfate oxidation
followed by colorinietric analysis for
nitrate-N (D’Elia et al. 1977, Currie et al.
1996). A hydrologic model parameterized
for the Cascade forests (Marks and Dozier
1992 and precipitation data collected at
the intensive sites will be used to estimate
total leachate volume so that total N losses
can be estimated.
28
-------
Net N mineralization
Net N mineralization will be measured
using in situ incubation of intact soil
samples within 20 cm deep PVC tubes
covered with thin polyethylene film (Hart
et al. 1994). The length of the incubation
period may be as short as a month or as
long as over winter depending upon
objectives and changing environmental
conditions. Soil samples taken before and
after incubation are extracted with 1 M
KCI and colorimetrically analyzed for
nitrate and ammonium (Technicon 1977,
US EPA 1979) (SOP # 3.10 A!pkem
Autoanalyzer). Net N mineralization is
estimated as the difference between the
amount of total inorganic-N (NO 3 + NH )
accumulated within the core after the
incubation period relative to the amount
present in the soil prior to incubation.
Ion-exchange resin bags (JER bags) buried a
few centimeters below the mineral soil
surface have been shown to be suited for
comparative studies of N availability
among sites or treatments (Binkley and
Matson 1983, Binkley 1984, Binkley et al.
1986, Giblin et al. 1994). The accumulation
of NH 4 ’ and NO 3 on the anion-cation
exchange resin is regulated by the processes
of mineralization, immobilization, plant
uptake, and transport; however the method
cannot distinguish the relative importance
of these processes. And, unlike the closed-
tube method, resin bags do not provide an
area! estimate of net N mineralization
because an unknown volume of soil is
sampled. The IER bag method has shown
to produce results that correlate highly
with other methods for measuring N-
mineralization, are less spatially variable
than other methods, and correlate with
NPP and fine root biomass (Binkley and
Matson 1983, Binldey et a!. 1986). We
believe that the strengths of the ER bag
method are particularly suited for the
development of an indicator for assessing
N availability or mineralization rates across
sites and years (e.g., Binidey and Hart
1989, Giblin et at. 1994).
We will prepare the resin bags and extract
the ions according to the methods described
in Binidey et al. (1986). [ ER resin bags will
be buried at about 5 cm beneath the top of
the mineral soil in each study site adjacent
to the locations used for measuring N
mineralization with soil cores. Bags will be
left in the field for up to a year (actual
length of time will be determined through a
series of preliminary experiments). The
ions will be extracted with 1M KCI, and
the ammonium and nitrate analyzed as
above.
Soil Respiration
Soil respiration is the total flux of CO 2
from the soil surface to the atmosphere and
includes both root respiration and microbial
respiration. Soil respiration rates can be
most accurately measured by infrared gas
analysis (Nay et al. 1994). We are
measuring soil respiration rates at our
intensive field sites by placing a portable
infrared gas analyzer (IRGA) over PVC
collars permanently installed in the soil
(Soil respiration SOP# 7.10 In situ Soil
Respiration (Li-Cor 6200): Field and
TERA). The IRGA-PVC collar system
gives reliable measures but is labor
intensive, especially for developing diurnal
and seasonal respiration patterns. An
automated system of measuring soil
respiration is also being developed (Section
29
-------
3.3, Subtask 8a), which will provide data
for cross-checking the results obtained
here.
Other data
obtained from USEPA CASTNet sites (US
EPA 1995) and the California Air
Resources Board (Blanchard et al., 1996).
Model evaluation and verification
Rooting depth . In addition to estimating
fine root biomass (see “Fine root C and N”
above), the maximum depth of fine roots
will be determined by examining deep soil
cores or pits excavated at each site.
Maximum rooting depth will be
operationally defined as the point above
which 95% of visible fine roots occur.
Environmental Data
The environmental drivers for MBL-GEM
include atmosphenc CO 2 concentration,
photosynthetically active radiation (PAR),
air and soil temperatures, soil moisture, and
N deposition. These data will be obtained
from several sources, including data from
the climate sensors installed at our
intensively studied sites (Section 3.1, Table
3-2), regional and site-specific data from
state (e.g., Oregon Climate Service,
University of Oregon) and U.S. Forest
Service, historical temperature and
precipitation data reconstructed from tree-
ring analyses, and future climate scenarios
predicted by general circulation models
(e.g., the Goddard Institute of Space
Studies GCM). Data for wet deposition of
N will be obtained from the National
Atmospheric Deposition Program (NADP)
site at the H.J. Andrews LTER established
in 1979. Few data are available for dry
deposition of N, however, a summary of
the data available for the western U.S.
indicates that thy deposition can be
approximated using a 1:1 relationship for
dry to wet deposition (Young et al. 1988).
Additional dry deposition data will be
We will use the parameterized MBL-GEM
as an assessment and predictive tool to
determine the effects of environmental
change on ecosystem C and N dynamics, to
identify additional potential sensitive
indicators of C and N dynamics, and to link
changes in the potential indicators
identified in Section 3.4 (below) to likely
stressors. For example, how will predicted
increases in C0 2 , temperature, and N
deposition during the next century affect
net primary production, soil respiration,
and N transformations and loss from forest
ecosystems? Or, how will periodic harest
of forestsaffect the long-term productivity
and sustainability of forest ecosystems
(see appendix C)? In these applications the
model will be used to predict long-term
changes that are outside the range of any
possible validation data set except, of
course, by waiting the requisite time for the
full ecosystem response to develop.
Obviously, waiting is ill-advised when one
must assess potentially significant changes
in the environment. Therefore, confidence
in these models must be built through the
accumulation of relatively weak
corroborating evidence and tests. Of the
four categories of evidence or tests that
Rastetter (1996) proposed for evaluating
models of ecosystem response to
environmental change (see above), only
three tests are useful for our purposes and
are described below. Although none of the
tests can be used as severe and crucial tests
of long-term (decades to centuries) model
predictions (Rastetter 1996), a conflict
30
-------
with any of these tests will lead to a re-
examination of the, model parameterization
or the model itself.
Tests using short-term experimental data:
Model predictions of short-term (seasonal
to less than a decade) ecosystem C and N
dynamics will be evaluated using the data
collected in this research plan that
describes seasonal and interannual
variations in photosynthesis, NPP,
litterfall, soil respiration, soil N
mineralization, and N leaching. While these
daia cannot address slow-responding
processes and feedbacks that influence
long-term ecosystem response, they are
extremely valuable for understanding the
transient responses of ecosystems to acute
disturbances such as harvesting, insect
outbreaks, fire, N deposition, etc. These
transient responses may be of primary
interest in some cases, for example, in
predictmg nutrient losses to “downstream”
ecosystems following forest harvest.
Space-for-time substitutions:
Model predictions of long-term ecosystem
C and N dynamics can be evaluated using
two types of space-for-time substitutions.
The first type of space-for-time
substitution uses chronosequences of plots
to examine successional changes in
ecosystem characteristics. In this way,
space (location of the plot) is substituted
for time (time since disturbance). We will
examine chronosequences of forest stands
in the Pacific Northwest following two
major types of disturbance, fire and
harvest, ‘to test how well the model
simulates post-disturbance recovery of
vegetation and soils.
A second type of space-for-time
substitution can be used to evaluate long-
term predictions of the effects of climate
on ecosystems. in this case, if the climate
at one location is expected to change so
that it resembles the climate at another
location, then the ecosystem characteristics
(e.g., C and N stocks) at the original
location might be expected to change so
that they resemble the present-day
characteristics of the second ecosystem
(Rastetter 1996). This type of space-for-
time substitution has been used to test the
equilibrium predictions of MBL-GEM for
projected increases in temperature across
the Amazon basin (McKane et al. 1995).
Similarly, we will test the predicted long-
term (equilibrium) adjustment of Pacific
Northwest forests to projected changes in
temperature and precipitation using data
describing C and N cycling in old-growth
forests at a number of sites throughout the
Pacific Northwest, for example, sites at the
H.J. Andrews Experimental Forest (Grier
and Logan 1977, Sollms Ct al. 1980) and the
Middle Santiam Wilderness (Fujimori et at.
1976).
Comparison with other models:
Although comparisons among different
models cannot substitute for tests against
real-world data, they can be used to
evaluate the relative importance of various
processes in determinmg long-term
responses to environmental change.
Agreement among models means that the
results of one model do not conflict with
the principles underlying the other models.
The degree of confidence obtained from
such agreement depends on how well the
underlying principles in each model have
been established and how independent
31
-------
these pnnciples are among the models
(Rastetter 1996). For example (Rastetter
1996), one model might be based on
biogeochemical principles of mass balance
and the interactions among C and N cycles
(e.g., MBL-GEM) and another might be
based on the principles of competition
among individual trees for light and soil
resources during succession (e.g., forest
succession models; Section 3.3.2). If the
two models agree on the accumulation rate
of carbon in the vegetation, then the
biogeochemical model has not conflicted
with the constraints of the succession
model, and the succession model has not
conflicted with the constraints of mass
balance and C-N interaction in the
biogeochemical model. Alternatively, when
the model results are not in agreement,
important controls on ecosystem response
not included in one or all of the models
may be identified (Ryan et al. 1996). We
will compare the predictions of MBL-
GEM against those for a forest succession
model (Section 3.3.2), as well as other
biogeochemical models (e g., CENTURY).
We anticipate that these comparisons will
be extremely useful in guiding further
development of the MBL-GEM.
Task 2: Modeling at the
landscape scale
The main goals of this task are to: (I) link
the MBL-GEM and a coupled energy and
water balance model (Daley et al. 1994,
Marks and Dozier 1992) to common
spatial data on land cover/land use, terrain
structure, and soil characteristics to
determine the effects of differences in
landscape pattern on model outputs at
watershed scales, (2) use the product of
goal 1 to predict the effects of
anthropogenic stressors on forest
landscapes of different pattern, and (3)
develop relationships between forest
landscape spatial patterns and ecological
processes. This research task will examine
those ecological processes whose changes
are most likely to be influenced by changes
in spatial and temporal patterns of forested
ecosystem landscapes (i.e., changes in land
use/land cover) and by other anthropogenic
stressors such as change in climate and air
pollutants. Initially these ecological
processes will include water quality, net
primary production, and C and N balances.
The distribution and pattern of natural and
human dominated systems in landscapes
influence ecological processes which in turn
affect production of goods and services
that landscapes provide humans. Changes
in pattern of landscapes can significantly
affect their response to other
anthropogenic stressors such as air
pollution and global change, thus affecting
the input and output of matenals (e.g.,
water quantity and quality [ O’Neill et al
1977], soil and sediments), regional
productivity, or biodiversity. Certain
configurations of ecosystems are likely to
make some landscapes more vulnerable to
stressors than others. Although the
linkages between landscape structure and
function have been subjects of interest in
ecological research for some time (Bormann
and Likens 1979), recent advances in the
developmg field of landscape ecology has
provided a basis for studying relationships
among landscape structure, pattern,
function, and change (Forman and Godron
1986; O’Neill et al. 1988; Turner 1989).
Advances in remote sensing and
subsequent data analysis in a GIS have
facilitated analyses of major ecological
32
-------
processes in relation to landscape pattern
and composition. Indicators reflecting
changes in land cover and/or land use are
likely to be useful in monitoring large-scale
changes in the function of forested
ecosystems if relations can be documented.
Methods
Linking the MBL-GEM to
landscape-scale spatial data bases
Initially, we plan to use the South Santiam
drainage (10 sub-watersheds of different
land use/land cover patterns and climatic
gradients) as a test area for this objective,
followed by a scaling up to the Western
Cascades. As described above (section
3.1.2), digital data for land cover/land use
and terrain structure are available for the
entire South Santiam system.
The energy and water balance model
(Marks and Dozier 1992) provides daily
outputs of precipitation, snowmelt,
evaporation, surface and soil temperature,
incident and net radiation, runoff, soil
moisture, and available moisture. These
outputs and the land cover/land use data
for the South Santiam sub-watersheds will
serve as input drivers to the MBL-GEM.
Soil C and N content and texture are also
important input drivers of the MBL-GEM.
We plan to use the Willamette National
Forest Soil Resouce Inventory as the
starting point for this work, and collect
additional soil samples for quantification of
soil C, N, and texture as needed. A
parameterized MBL-GEM will then be
simulated, pixel by pixel, for each land-
cover / land-use class in the sub-watersheds
using the appropriate climatic and soil
moisture regimes generated by the energy
and water balance model and soil C and N
status from the soil surveys for the same
pixels. The MBL-GEM-simulated outputs
of NPP, N losses, and C balances will be
summed by sub-watershed. These
simulated outputs will be used to
investigate relationships between landscape
pattern and ecological processes as
descnbed next.
Predicting the effects of
ant hropogenic stressors on forest
landscapes of different pattern
This objective will build on the previous
one, in that the landscape-scale MBL-
GEM will be simulated under a range of
possible scenarios of climate change,
change in atmospheric composition, and
increased N deposition. Changes in the
outputs of NP?, N losses, and C balances
for each subwatershed under these different
scenarios will be compared to indices of
landscape pattern (described below) to
investigate how the pattern influences the
vulnerability of landscapes to further
change.
Development of relationships
between forest landscape spatial
patterns and ecological processes
The ability to quantify landscape pattern
and structure is requisite to understanding
landscape function and change over time.
For this purpose we will use a versatile
public domain computer program,
FRAGSTATS, which produces a
comprehensive array of landscape pattern
metrics (McGarigal and Marks 1995). The
software is automated for use with either
vector or raster images on either Unix or
PC platforms. We will examine a variety of
possible indicators of landscape spatial
pattern of different land cover/land use
classes, including various area measures,
33
-------
patch density, size and variability, as well
as edge, shape, core area, diversity,
contagion, nearest neighbor and
interspersion metrics. We will also examine
percent distribution of forests in the
various seral stages as a measure of
landscape vegetation structure.
We will use the sub-watersheds of the
South Santiam watershed and their 1991
land cover/land use data (Section 3.1.2).
The map data will be analyzed with
FRAGSTATS to quantify landscape
pattern in the sub-watersheds, detecting
differences based on land-use, land cover,
and forest management practices. The
metrics of landscape pattern will then be
related to MBL-GEM outputs described
above to determine watershed-level
functional responses to climatic and
edaphic factors, forest management
practices, and other stressors.
To understand how spatial patterns in the
South Santiani subwatersheds influence
water quality measures (nutrients, other
chemical ions, and microbiological
contaminants), we will institute a regular
water sampling program for the
subwatersheds near their mouths over a
penod of at least 3 yr. Using field sampling
methods adapted from the EMAP Surface
Waters Program (US EPA 199Th), we will
collect grab samples monthly and analyze
for nitrate, ammonium, dissolved organic
nitrogen, dissolved organic carbon, sulfate,
chloride, iron, calcium, magnesium,
potassium, sodium, turbidity,
conductivity, p1-I, alkalinity, and presence
of various microorganisms. To calculate
mass movement of the various chemical
species, stream flow will be measured.
Monthly, seasonal, base-flow, and peak-
flow bulk movement of nutrients and other
elements in stream water will be calculated
as the product of water concentrations and
discharge. The water quality values will be
statistically compared with landscape
pattern metrics as detailed above.
Water quality impacts of forest
management activity have been
demonstrated under drastically different
conditions, such as comparing paired
watersheds where one is intact old-growth
forest and another is entirely clear-cut (e.g.,
Sollins et a!. 1981). Much less has been
reported on the impacts on water quality
when more subtle differences in forest
management are compared. In order to
understand what differences in management
practices and landscape patterns are
necessary to cause significant changes in
water quality parameters, we will
investigate nested South Santiani sub-
watersheds at different spatial scales. Five
sub-watersheds of near-equal size (— 5 000
ha) will be sampled. In addition, paired
sub-watersheds of approximately 500, 50
and 5 ha will be monitored. The smaller
sub-watersheds will provide greater
opportunities for observing greater
differences in forest management,
landscape pattern and seral stage
distribution.
Monitoring will be accomplished by
obtaiping continuous records of stream
discharge and automating collection of as
much water quality data as feasible. Weirs
may be installed with depth sensors to aid
in estimating instantaneous discharge.
Electronic sensors and data loggers will be
installed for monitoring parameters such as
water temperature, conductivity, turbidity,
dissolved oxygen, and pH. Anions, cations,
34
-------
dissolved inqrganic and organic nitrogen,
dissolved organic carbon, ANC, total
phosphorus, and several microbiological
measures will be determined for grab
samples collected less frequently (e.g., bi-
weekly).
Combining the water quality data with
metrics of landscape pattern and seral stage
distribution will enable us to answer the
question of how much difference is
necessary to detect changes in water
quality. Conversely, we will be able to
estimate the extent of management
disturbance required to produce adverse
impacts on forest streams.
3.2.2. Forest succession modeling
Task 3: Forest succession model
for stand to landscape
Objectives and background
The objective of this task is to enhance the
applicability of forest succession models in
montane areas generally, and in the Pacific
Northwest (PNW) specifically for use in
forest indicator research and assessments.
The goal is to develop, test and use models
reliable enough to be integral components
of forest indicators of ecosystem
vulnerability to environmental change .
From the suite of potential field data
collection and model building activities
which can enhance forest succession model
validity (Bugmann and Solomon, 1997), we
chose as the first task, the creation of a
model of tree mortality, both as a forest
indicator in its own right, and for
incorporation into a montane forest
succession model. This task requires
documenting specific processes important
in the mortality of trees. Our second task is
to increase the accuracy of the method used
to estimate environmental limits of trees
species, which underlies the basic climate
response of the forest succession models.
This task requires documenting fine-scale
distributions of climate variables and tree
species geography.
Forest structure obviously will respond to
directional shifts in environmental
variables. Natality and mortality rates of
individual tree species will change. Gaps in
forest canopies, and loss of whole forests,
are likely to result 1 ecause death of tree
populations requires only a few years
while establishment arid regrowth of
replacement trees requires several decades
to centuries (Solomon, 1986; Kirschbaum
and Fischlin 1996). This lag in replacement
of dying trees should be statistically
detectable at regional scales. However, it
will be more difficult to produce definitive
measures of related changes that are
specific to any given locality, such as the
Cascades study sites of this proposal. The
competitive relationships among adjacent
trees in a stand will be modified as e.g.
warming increases growth of some and
decreases it or leaves it unaffected in
others. Eventually, certain species will be
outcompeted entirely and will disappear
from individual forest stands, local
watersheds, then regions. The trees with
the poorest success under chronic climate
stress are likely to be late-successional
species, that is, those with the longest life
cycles, with the slowest growth rates, and
with the greatest shade tolerance (Solomon
and Leemans 1990, Solomon et al. 1993), a
set of properties which also charactenze
tree species capable of the slowest
responses to climate change.
35
-------
in addition to different tree species,
different stages of tree life cycles also are
differentially vulnerable to environmental
variations. The middle-aged trees present in
a forest canopy are most resilient (Lonmer
and Frelich 1984, Peet and Cbnstiansen
1987). Those approaching senescence are
much more vulnerable, and newly-
germinated seedlings are the most
vulnerable of all (Harcombe 1987). Hence,
the establishment phase of late-
successional species should provide the
most sensitive indicators of forest structure
response to chronic environmental change.
Yet, the establishment of tree seedlings is
also the most variable. Western hemlock, a
late-successional tree species characteristic
of the Pacific Northwest, may produce 2 X
107 seeds per ha each year which may
reduce to 2 X 104 saplings 20 years later
(Packee 1990). However, fewer than 2 X
102 trees reach maturity, even in the
absence of stand-replacement disturbances.
Moreover, establishment “pulses” at tree
range boundaries may be decades to
centuries apart (Savage et al. 1996,
Arseneault and Payette 1997).
Documenting the sensitivity of seedling
establishment to changing climate would
require many thousands of samples over
many years in many locations for each tree
species. Therefore, this project will focus
upon the age-dependent mortality of late-
successional trees as the most appropriate
indicator of directional changes in forest
structure.
Although tree mortality is less vulnerable
than seedling establishment to
environmental variations, it has certain
advantages as an indicator of long-term
directional changes in forest structure and
function. Mortality is an unambiguous
state which varies in response to stress and
age (Waring, 1987). With seeding
establishment, it defines long term forest
dynamics (Harcombe, 1987), making
inclusion of mortality considerations
critical to the application of most forest
mdicators. Mortality rates under differing
past environmental conditions can be
objectively defined (e.g., Henry and Swan,
1974) and departures from them by other
mortality events and trends can be
measured and assigned statistical
probabilities of occurrence. Large-scale
mortality (die-back) is an obvious
condition which possesses considerable
political importance, and which sometimes
can be measured with inexpensive remote
sensing techniques, permitting rapid
assessment of the statistical characteristics
needed to attribute cause to the mortality
rate changes. When included in forest
succession models, mechanistic mortality
models can permit calculation of forest
community-level tree deaths related to
environmental change.
Strong advantages are also found in forest
succession models themselves for forest
indicator research. Particularly pertinent is
their ability to capture and predict the
year-to-year changes to be expected in
forest structural characteristics (e.g.,
changing size and age distributions of
individuals from each species), and to do so
as a function of yearly variations in
temperature, soil moisture, C0 2 , ozone,
and other (changing) environmental
properties. Their weaknesses for our
purposes includes their non-mechanistic
treatment of stress-induced mortality, the
inaccurate species-climate correlations
permitted by available environmental data
in montane areas, and their inability to
36
-------
predict the dynamics on any specific plot
or set of plots without very detailed
mformation on plot history (e.g., Solomon
1988). These weaknesses will be treated in
the research proposed below.
Nature offorest succession models
Forest succession models (also referred to
as gap or stand models) mimic the
dynamics of tree establishment, growth and
mortality by multiple species of differing
ages in a gap created by the death of a
dominant tree in an otherwise continuous
forest canopy (Fig. 3-4). There, the models
simulate mterspecific competition for
sunlight, water and nutrients based on
mdividual species differences in shade
tolerance, drought tolerance, and nutrient
requirements. They simulate the vertical
characteristics of tree density on a
(usually) circular plot of specified size and
they calculate the amount of light which
reaches each vertical level as a function of
the leaf areas above that level (Fig. 3-4,
center). They assume that the maximum
dimensions of each tree species (maximum
diameter, height and age), previously
measured in the field, are also the maximum
dimensions each species could reach under
ideal environmental conditions (light,
warmth, soil moisture, and nutrients). Each
simulated year, climate conditions (Fig. 3-
4, right margin) are calculated as random
variables from documented means and
standard deviations, and reduce growth of
trees on the plot from the growth maxima
which rarely if ever occur either in nature
or in the models.
Both tree establishment and mortality of
established trees, of interest here, are
treated as stochastic processes (Fig. 3-4,
left margin). Mortality, for example, is
modeled as a constant probability of death
such that only 2% of established trees
reach their maximum known age. A second
stochastic process provides that diameter
growth below a threshold at any age
produces an enhanced mortality
probability (usually a 1 in 3 chance of
mortality during the subsequent 10 years),
which is most likely in the youngest (i.e.,
smallest diameter increment) trees and in
the oldest (i.e., slowest growing) trees.
These stochastic rules may simulate much
more rapid elimination of trees under
changing climate, than rates of tree loss
under the actual mechanisms which control
tree mortality. This problem becomes a
critical model flaw if tree mortality rates
are to be reliable indicators of changing
forest condition.
Climate constraints on mortality, as well as
on growth and reproduction, are defined by
the coincidence of mapped boundaries of
tree species with mapped climate variables.
From these overlays, one defmes such
extreme climate parameters as the minimum
temperature of the coldest month or the
maximum annual days of soil moisture
below wilting point, found within the
geographic range of each species. Recent
reviews of the capability of forest
succession models to represent probable
forest responses to future climate change
(e.g., Bonan and Sirois 1992, Loehle and
LeBlanc 1996, Schenk 1996) condemn this
underlying methodology. The cntics
suggest that accidents of Holocene tree
migration have produced a current
geographic range of species (“realized
niche,” more correctly “realized range”)
which is smaller than the geographic range
the species potentially could occupy
37
-------
Figure 3-4. Conceptual diagram of the forest succession model ForClirn 2.9, after adaptation of
a generic temperate-zone model (Bugmann
Northwest (Bugmann and Solomon 1997).
(“fundamental niche” or range), introducing
a false climate-growth correlation into
succession models.
This criticism may (or may not) be
appropriate in flat or gently rolling terrain.
However, Tsukada (1982a, 1982b) has
showed that species ranges in mountainous
terrain have not been subject to delayed or
obstructed migration paths because of their
close (up or downslope) proximity to
suitable habitat under even rapid past
climate changes. Hence, the montane
and Solomon 1995) for forests in the Pacific
forests of interest in the PNW should be
much less prone to the realized versus
fundamental range problem.
However, implementation of the climate-
species correlations in the PNW encounters
more direct methodological restrictions.
Growth of montane tree species at their
boundaries, where extreme climate
parameters are documented, is restricted to
a few suitable habitats (e.g., cold north or
warm south-facing slopes, cool, damp
valley bottoms), rather than to average
\j/
EXTRINSIC LIMITS
\ \ \ \% \ \
\\\\ \\\
AGE-DEPENDENT
MORTALITY
MOISTURE
I
E.* DEPB4DfNT MO TALJTY
* CREASED
O TALflY
wTh
DEcREASED GROWTH
UIABUINMENT
LJMATE
5ff E
CONDITiONS
UGNT
LEVELS
SPEcIES AVALJRLj
OF
FULL SUNLIGHT
38
-------
conditions which are descnbed by regional
climate measures. Similarly, climatic
gradients are so steep in mountainous
terrain that the entire range of climate
values found within a tree species’ range
can be telescoped within the few
kilometers between adjacent climate
stations at different elevations. This scale
problem is the specific focus of proposed
research described below.
FORCLIMforert succession model
Gap models have been applied in Pacific
Northwest forests for at least 15 years
(Hemstrom and Adams 1982, Dale and
Hemstrom 1984, Kercher and Axelrod
1984, Urban Ct al. 1993, Burton and
Cumniing 1995). Each model version varies
in capability and validity, none being well-
suited to the problem of regional indicators,
even in the absence of the weaknesses
discussed above. Variables and parameters
in the ZELIG model (Urban et al. 1993)
have been calibrated to work for the H. J.
Andrews LTER Site (ZELIG2.PNW;
Stephen Garman, pers. comm. 1997), but
documentation has not been published to
date, and the value of one site to represent
the PNW is not great.
The ForClim model (Bugmann 1994)
simulated forests of Europe and eastern
North America (Bugmann and Solomon
1995) with equal accuracy. It was modified
(V2.9) to incorporate the peculiarities of
forests of the Pacific Northwest (Bugmann
and Solomon 1997). ForClim 2.9 has been
tested for its ability to reproduce stand
biomass and species composition on a
transect of 27 sites from the Oregon Coast,
eastward across the Coast Range, the
Willamette Valley, the Cascades, and into
the cold desert near Bend, Oregon. The
model also has been tested against tree size
distributions and species composition at
three elevations (500 m, 1000 m, 1400 m)
in the H. J. Andrews LTER Site. These
tests reveal that ForClim reproduces PNW
forest dynamics more accurately than other
available forest succession models, and that
the ability of the model to reproduce
eastern North American and European
forest composition and biomass also is
improved by the PNW modifications.
The new modeling developments on which
we can focus our necessarily limited
resources include the current the lack of
mechanistic stress-related mortality
processes, and the inability of the model to
relate species geography to the very steep
temperature and moisture gradients found
in the western mountains. The former
problem (mortality processes) will require
field data and literature synthesis on
natural mortality rates. The latter problem
(species geography) will require additional
data on geographical distributions and
associated environmental conditions under
which individual tree species are found.
Some of these data are available by careful
examination of the published literature, and
some will require field excursions to record
previously undocumented occurences. The
methods used to reduce both of these
weaknesses will coincidentally permit
appLication of the models to specific
locations, by generating the historical data
(tree-ring data) on mortality with which to
replace the succession model’s stochastic
estimator of these data.
39
-------
Methods
The data we will collect to improve the
FORCLIM model to the point that it more
accurately reproduces the spatial
distributions and temporal sequences of
species composition, density and siz&age
distributions in the Pacific Northwest are
in two groups: one group is required to
develop the mechanistic mortality model
and a second group is required to improve
the way m which species’ environmental
limits are derived. The following
information describes the research activities
we propose to implement to reach these
specific goals.
Data for constructing a mechanistic
mortality simulator
In addition to use of the scientific literature
to create a mortality simulator, we will
collect data sets on mortality in the PNW
to parameterize and test the simulator.
Three methods will be applied. One
approach is to examine stand
remeasurement data from permanent plots,
where available. These data frequently
include measurements of the mortality
since the previous survey, and may extend
over several decades, from samples
collected at regular or irregular intervals.
Harcombe (1986) utilized such records
from seven remeasurements during 50
years at 130-year old forests from Cascade
Head Experimental Forest in the Coast
Range near Otis, Oregon to define temporal
sequences of mortality, and their sources,
during stand development there. We will
determine the availability of appropriate
remeasurement plot data in the Pacific
Northwest forests of Oregon, Washington
and Idaho, and will define mortality rates
either by our analysis of these data, or by
encouraging mortality analyses by those
with responsibility for the data
A second means to examine the rates and
changes in mortality is by reconstruction of
forest site histories (e.g., Henry and Swan
1974, Franklin et al. 1981). The approach
is based on the slow rates of deterioration
of fallen dead trees in the Pacific
Northwest (Harmon, Ct al. 1986) combined
with the ability to crossdate the outer
growth rings of long-dead trees with the
inner rings of recently dead or living trees
(e.g., Smiley and Stokes 1968, Cook and
Kariukstis 1990). The primary method
consists of developing crossdated
dendrochronologies, mapped for all the
dead standing and downed trees in several
stands or study areas, and the calculation
of the mean and variance of mortality rates
for the areas represented by the several
data sets (Franklin et al. 1981). We
propose to reconstruct mortality histories
of several stands in the Toad Creek Road
and Falls Creek intensive study sites,
where some tree ring chronologies have
already been collected and processed. In
addition to our documenting of long-term
mortality rates for development of a
mortality simulator, we will use the
resulting measurements to provide
important background mortality and tree
growth information for paraineterizing
stand mortality in the forest succession
simulation verification exercises discussed
above. The mortality measurements and
site histories will also support the other
above- and below-ground research at the
intensively-studied sites. Because of the
labor-intensive nature of these field data
collections, we expect to generate data from
only three or four 0.2 ha plots at each of
the two sites.
40
-------
A third, very different means to calculate
mortality rates, is to measure the change in
growth rates recorded by the last decades
of annual growth rings in dead trees which
occur in larger diebacks. This• approach
establishes species-specific growth
variations which may predict imminent
mortality (Schwiengruber et al. 1986,
LeBlanc et al. 1987), and which can then be
used to hypothesize specific mortality
cause and effect. These regional
dendrochronologies provide a spatial
contrast to the site-specific forest site
histories described immediately above. We
propose to implement this approach by
developing dendrochronologies (2 cores
from each of 20-40 trees per site) at 10-20
sites in Cascades forest stands which have
died back (e.g., the Blue Mountain dieback;
the Santiam Pass dieback), and in adjacent
still-living stands of the same species, in
order to document the growth patterns in
the decade or so immediately preceding the
diebacks (e.g., Swetnam 1987). Although
this research activity is considerably more
exploratory than other work proposed
here, we believe its potential for defining
productive mortality indicators is quite
high.
Data for enhancing accuracy of
species-environment relationships.
A fundamental problem to implementing
gap models in the Pacific Northwest
involves the difficulty of defining
environmental tolerances of species from
their co-occurring spatial distributions in
montane regions of very steep
environmental gradients. Distributions both
of tree species and of climate variables are
mapped too coarsely to permit accurate
estimation of the tolerance values. We
propose to resolve the problem of steep
elevational gradients through use of fine-
resolution digital elevation models
(DEMs). The DEMs will be used to
disaggregate the range of topographic
values represented in spatial units of the
finest-resolution climate data base
available. These data will define a fine-scale
set of climate data required to drive the
stand models (monthly growing degree
days, temperature of coldest month, tn-
monthly soil moisture and snow status),
and will be matched with field observations
of species distributions in the areas of their
geographic ranges where they reach critical
limits.
Because local temperature and soil
moisture depend on elevation, exposure,
and topographic position, these physical
variables will be critical targets to calculate
and map from a DEM (500 m resolution)
within each grid square (VEMAP data
containing 4 km x 4 km pixels) represented
in the climate data base (Dodson and
Marks 1997.). Distributions of topographic
variables will be transformed to proxy
climate variables using topography-climate
relationships (e.g., solar insolation and sun
angle for temperature versus topographic
position; known seasonal lapse rates and
humidity for temperature versus elevation;
PRISM model for moisture versus
elevation [ Daly et al. 1994]).
In addition to the fine-scale mapping of
proxy climate variables, we will also
document the relationships between
climate and actual local tree distributions
(approximately 50 species are currently
found in the Oregon Cascades or are
appropriate for survival there under global
climate change scenarios). Review of the
literature, visits with local forest experts,
41
-------
as well as field inspections will be
conducted in regions where tree species
reach a critical geographic limit. We
envision several two-week excursions in
which we visit multiple locales
representing east-west and north-south
range limits of the most common of the 50
species. The objective will be to determine
the frequency and fidelity with which
species are segregated in specific
topographic situations (e.g., on north-
f cmg slopes and in deep valleys at the
southernmost edge of their range; on south-
facing slopes and on calcareous soils at the
north edge of their range; etc.) in these non-
optimal areas.
The proxy climate data and the local
distributional data will be used to define
the actual range of environmental
conditions in which each species can
survive within its current geographic range.
The proximity of widely diffenng
conditions in montane areas should have
permitted most available niches to fill
during the past several thousand years
since inception of the current climate
(Tsukada 1982a, 1982b). Hence, the
realized range (current range limits) of these
montane species is probably little different
from the fundamental (potential) range
beyond which climate actually limits
species’ presence.
The relationship between climate and
distributions of 50 tree species, generated
by the foregoing, will be independently
tested in part by examining relationships
between tree ring indices and climate at 20
to 100 sites scattered throughout the
geographic ranges of two of the most
important tree species of the Pacific
Northwest: coastal Douglas fir
(Pseudotsuga menziesii var menziesiz), and
Pacific western yellow pine (Pinus
ponderosa var. ponderosa). Douglas fir
occupies cool, mesic sites along the Pacific
coast some 2200 km from central British
Columbia to central California. The Pacific
variety of western yellow pine occupies
warm, dry sites from southern British
Columbia almost to Mexico. The
elevational ranges of the two species at any
given location overlap, with dominance
determined where fire favors ponderosa
pine and fire suppression favors Douglas
fir. The analysis will require us to collect
tree ring chronologies from the entire
latitudinal N-S range of the two species, as
well as from selected low to high elevation
transects, each chronology containing
paired sites consisting of climatically
stressed and non-stressed trees.
We will identify old-growth stands of
trees, and adjacent weather stations with
long station histories, the set of which
represent the range of conditions
throughout the geographic ranges of
Douglas fir and yellow pine. Tree ring
indices will be developed, based both on
tree ring widths and on late-wood density,
and will be used to create response surfaces
to characterize the response of tree rings to
individual climate variables as they vary
with elevation and latitude (e.g., Cook and
Cole 1991, Thompson Ct al. 1998). The
response surfaces in turn will provide a
measure of tree growth response to climate
independent of the DEM surface analyses.
This will be particularly valuable as a
means to testing our assumption of
equivalence between the realized and
fundamental range of these two species.
42
-------
The tree-ring analyses proposed here have
not been used for this purpose in the past,
and hence, may not provide the model
verification required for indicator-level
reliability. However, if we can demonstrate
the technique to be effective, we expect to
collect and analyze data on additional tree
species important in the Pacific Northwest
stand models. In any case, the development
of a permanent tree-ring analysis capability
at WED will coincidentally lead to
development of stand histories at the
Cascades research sites. Such histories will
be. a critical component of our analyses
comparability of research results from
these sites.
3.3. Resource utilization
The previous section (3.2) described
modeling approaches that will be evaluated
and/or developed for assessing impacts and
predicting changes in forested ecosystems
from natural and anthropogenic stressors
and management practices. This section
will focus on measurements of C and N
utilization by trees (3.3.1) and metabolism
of C and N in the forest litter and soil
(3.3.2). The Resource Utilization Tasks
will serve several functions including:
• Providing data for model
parameterization, calibration, evaluation,
and modification
• Providing fundamental data on key
ecosystem processes and their response
to various stressors
• Evaluating the suitability of various
ecosystem processes as indicators to
assess the current condition of forested
ecosystems
• Developing indicators
ecophysiological processes
complexity and function of
foodweb to assess the current
forested ecosystems.
These tasks will focus on determining the
natural spatial and temporal variability of
key. ecophysiological processes and the
complexity and function of the soil
foodweb (Fig. 3-5) at intensive and
extensive field sites. Field sites will be
selected that are appropriate for testing
specific hypotheses regarding resource
utilization (see section 3.1.2). For example,
retranslocation of N and nitrogen use
efficiency as a function of site nutrient
status is being investigated in tree species.
To accomplish this, three sites are being
used that have differing nitrogen
availability. Two of the sites differ fourfold
in N, and are at the same elevation and
approximately the same climate. A third
site has higher N availability but is at a
lower elevation and a differing growing
season length. Specific hypotheses and the
use of field sites with appropriate climate,
nutrient, etc gradients will enable
interpretation of variance m the measures
of carbon and nitrogen allocation.
Many of the measurements descnbed in this
section will also be made in the experimental
facilities at WED where we will measure the
same ecological processes and components
under known stressors (see Section 4).
of
and
the soil
status of
43
-------
5a. Secondary metaboiltes
compounds
5b. Levels of secondary metabolites
and carbohydrates f
Task 6
Phenology
Task 9
Complexity and function of soil foodweb
I 9a. Mycorrhlzal symblonts
b. Soil foodweb blota
Figure 3-5. Identification of the research tasks to measure spatial and temporal variability of resource
utilization (C and N) by producers and consumers.
Task 4
C& N allocation
4a. Leaf/fine roots
4b. N retransloctlon
4c. Root respiration
4d. Reproduction
Task 5
C & N partitioning
Task 8
C& N losses
8a. Effux of soil carbon
6a. Fine root dynamics
44
-------
3.3.1 Resource utilization-
producers
Task 4: Carbon and Nitrogen
Allocation
Subtask 4a. Leaf area/fine root biomass
ratios as measure of C allocation
Objectives and background
The main goal of this task is to provide
fundamental data on C and N allocation
between leaves and fine roots, including
determining the interannual variation in the
LAI/fine root biomass ratio. The ratio will
be compared to climate records and data on
soil moisture, temperature, and C and N
content to determine the sensitivity of the
ratio to vanous environmental and edaphic
factors. This task will also contribute
several key types of data needed for the
biogeochemical model (Section 3.2.1). Light
extinction data will be used to estimate LAI
for the stands and to provide information
on light extinction important for calibrating
MBL-GEM. The fine root biomass data
will contribute to estimating the C and N
pools of the fine roots and also will be used
to determine the rooting depth of the stand.
Leaves and fine roots play significant roles
in resource acquisition. Consequently, the
allocation of C and N to these organs to
maintain a balance between C and water
and nutrient acquisition is important for a
The ratio between leaves and fine roots
(biomass or area) has been used to
investigate the impacts of various
treatments on C allocation. In ponderosa
pine, N fertilization initially increased the
LAI/fine root biomass ratio. However, this
plant to grow, reproduce, and to permit it
to adapt to changing environmental and
edaphic conditions. For example, C fixation
occurs in leaves but its rate depends on
their N content and thus on N uptake via
root activity. Because the chemical
composition of a plant can vary only
within narrow limits, a proper balance
between the functioning of leaves and fine
roots must occur. The functional balance
concept (Brouwer 1962, 1983) implies that
fine roots and leaves continually adjust
their relative resource allocation patterns as
resources change. For example, Lou et al.
(1994) proposed that the fine root fraction
would be stable when the N supply
matched the photosynthetic supply but if
the CO 2 stimulation of photosynthesis was
larger than shoot growth (limited by
nutrient shortage) the excess
photosynthate would stimulate root
growth.
To study the allocation of C and N, we will
follow the recommendation of Kôrner
(1994) and use a three compartment
(needles, woody tissue and fine roots)
model to describe carbon allocation and
functional relationships in plants. The
three compartment model is the same as
that used in the MBL-GEM model
(Rastetter et al. 1991) which is being used
in the biogeochemical cycling modeling
portion of the research (3.2.1).
effect decreased as the plants grew,
possibly because the relative abundance of
N to plant size decreased (Tingey Ct al.
1996). Pregitzer et al. (1995) also found N
fertilization increased the LAlJfine root
ratio in poplar. Both Norby et al. (1992)
and Körner (1994) reported that elevated
CO 2 increased fine root mass but had no
45
-------
Fine root biomass
effect on leaf mass leading to a decrease in
leaf mass/fine root mass ratio. However, in
ponderosa pine, the LAJJfine root length
ratio was unchanged by elevated CO 2
indicating that the relationship between
nutrient absorbing surfaces and
photosynthetic surfaces was not changed
by C02 exposure (Tingey et al., 1996);
Pregitzer et al. (1995) found a similar trend
for poplar.
Methods
Leaf Area Index (L41) and leaf
biom ass
LA! will be estimated from measures of
light extinction made below the canopy
(Pierce and Running 1988, Gholz et al.
1991). The light extinction method is: (1)
site specific and (2) provides information
on light extinction important for calibrating
MBL-GEM.
A Licor light bar (1 m) will be used to
measure light extmction at permanently
marked locations along a transect in the
forest at each site. Measurements in the
clearcuts will serve as the reference for top-
of-canopy light levels. All light
measurements will be made within one
hour of solar noon near the time of the
summer solstice to minimize the effect of
shadows from tree stems. Beer’s Law will
be used to estimate leaf area from the light
extinction data (Jarvis and Leverenz 1983).
Leaf area will be then converted to leaf
biomass on a stand basis using measures of
the specific leaf area (g dry weight m 2 leaf
area). The light extinction method is
preferred over allometric methods of
estimating leaf biomass because it is site
specific.
Fine root biomass will be estimated using
standard coring methods (Santantonio et al
1977). At least 20 locations will be
sampled per site using cores 5 cm in
diameter and 30 to 100 cm deep. Soil and
roots will be separated using wet sieving
methods. Roots will be separated into
coarse (>2 mm) and fine root ( 2 mm)
fractions, dried and weighed. Core samples
will be collected at 3 mo intervals to
determine seasonal changes in fine root
biomass.
C and N allocation between leaf and
fine root pools
Measurement of leaf tissue C and N pools
is described in section 3.2.1 (Leaf C and
N). Oven-dried and weighed (corrected to
an ash-free basis) roots will be ground to
pass a 40 mesh screen., and analyzed for C
and N concentration using a flash-
combustion gas chromatography method
(Carlo Erba Standard Operating Procedure
(SOP) 3.01 Carbon/Nitrogen Elemental
Analysis). Carbon and N pools will be
calculated by multiplying fine root biomass
by measured C and N concentrations.
Subtask 4b: Nitrogen and phosphorus
rezranslocation and resorption
efficiency/proficiency
Objectives and background
The objectives of this subtask are to
determine (1) the potential resorption of N
and P (maximal withdrawal of each nutrient
from senescing foliage), the resorption
efficiency (percent reduction of nutrient
between green leaves and senesced leaves),
and the resorption proficiency (measured
level of nutrient in senesced foliage) in 1-2
46
-------
dominant tree species over a period of 3 yr
and across a charactenzed gradient of
growing conditions; and (2) values needed
to parametenze MBL-GEM (see section
3.2.1).
Nutrient resorption, and in particular N
resorption from senescing leaves before
abscission takes place, is one of the most
important processes for plants to conserve
nutrients. Because senesced, falling leaves
account for >70% of litter, the efficient
retranslocation of nutrients from leaves
into storage tissues in stem or root is an
essential process in both the individual and
the ecosystem. A number of factors can
cause realized resorption of nutrients to be
less than potential resorption in some
years, including water availability
(Escuedero et al. 1992), timing of
abscission (Killingbeck et al. 1990) and
shade (Chapin and Moilanen 1991).
Anthropogenic stresses, such as
tropospheric ozone, could also affect the
resorption process. For example, ozone
causes premature leaf fall in several trees
species (e.g., Keller 1988, Wiltshire et al.
1993, Pell et al. 1995, Weber et al. 1997). If
ozone-induced abscission proceeds before
the resorption process is complete, this
could disrupt nutrient conservation in
individuals and possibly cause changes in
nutrient availability from litter.
Millard (1996) summarized several
nitrogen budget studies in evergreen and
deciduous trees showing that internal
cycling of N (i.e., N conservation) is a
major source of N providing up to 90% of
this nutrient used in seasonal growth. In
addition, as trees get larger their N uptake
rate decreases but their N storage capacity
and reliance on internal nitrogen cycling
increases (Millard 1996). Disruption of the
N resorption process reduces plant fitness
in subsequent years. May and Killingbeck
(1992) demonstrated substantial reductions
in foliar biomass (4 1%), radial stem growth
(54%) and fruit production (90%) in a 3
year study of oak when nutrient resorption
was blocked. Late season drought causes
early leaf abscission in deciduous trees and
disruption of normal nutrient resorption,
and is a contributing factor to reduced stem
growth in mature trees (Killingbeck et al.
1990) and a selective pressure in changing
community species composition.
Killmgbeck (1996) distinguished the
measured differences in resorption
efficiencies (percent reduction of a nutrient
between green and senesced leaves) as
differences in resorption potential of the
species or temporal differences in realized
resorption. The measurement of the
potential resorption (the maximum
withdrawal of nutrients from foliage) is not
possible directly; however, estimates of the
degree to which realized resorption
approaches potential resorption in
individual species can be achieved by
developing a knowledge of the levels to
which the species can reduce nutrients in
senescing leaves (resorption proficiency)
(Killingbeck 1996). This task proposes to
take the multifaceted approach suggested
by Killingbeck (1996) to develop a measure
evaluating N cycling in different site
conditions and over time as a potential
indicator of tree (individual to population)
condition or as an input variable into an
indicator of forested ecosystem condition.
47
-------
Methods’
Subi ask 4c: Root respiration
Measurements of concentrations of N, P
and Ca in presenescence and senesced
needles of ponderosa pine, Douglas-fir and
western hemlock will be made on
composited needle samples at selected
dates after determining the time period
when concentrations of these nutrients are
relatively stable (maximum and constant)
for each species. Samples of presenescence
senesced needles will be collected from a
range of diameter classes for each of the
three species.
Needle area and dry weights will be
measured to calculate nutrient content.
Content is calculated as the product of
specific leaf mass (g cm 2 of needle) and N
or P concentration (jig gm or ng gm 1 ).
Total N will be measured on dried and
ground needle samples using the Carlo Erba
C and N analyzer (SOP 3.01). Total P and
Ca will be measured on dned, ground and
digested tissue using an ICP
spectrophotometer (William 1984). The
content of Ca is determined for comparison
because Ca is not resorbed.
Resorption efficiency will be calculated for
each tree in each sampling period as the
difference in N or P content (jig cm 2 of
needle) between presenescent and senesced
needles, divided by presenescent needle N
or P content. Temporal changes in N, P,
and Ca concentration will be determined
and related to changes in estimated needle
mass.
Objectives and background
The goal of this subtask is to examine (I)
the use of stable C and 02 isotope ratios in
quantifying specific root respiration, as
well as area-specific microbial activity, and
(2) the utility of such a measure as an
indicator of C dynamics in evaluation of
forest condition.
Previous studies have found that soil CO 2
flux increases from soils containing plants
exposed to ozone stress, and the increase
occurs before any changes in growth are
observed, suggesting that a change in soil C
flux may be an early indicator of stress
(Andersen and Scagel 1997; Scagel and
Andersen 1997). One critical component of
soil CO 2 flux is root respiration. Root
respiration is important both for
understanding plant response to stress and
for characterizing system level C fluxes.
Because C tiux from soil occurs both from
plant roots and decomposition of soil
organic matter, it is important to
discnminate between the two. Isotopic C
and 02 ratios offer a means to discriminate
between the two C flux sources because
each has a unique isotopic signature (Lin et
al. 1998). We can quantify the source of
the CO 2 released from a known volume of
soil by knowing the isotopic characteristics
of the C and 02 sources. This approach
allows us to quantify specific root
respiration (per am root), as well as area
specific microbial activity. The subtask
will compliment the isotopic work being
conducted to deconvolute soil-system CO 2
fluxes (see subtask 8a).
48
-------
Two approaches are envisioned that will
use stable carbon isotope ratios. The first
will involve characterizing the öCl3 ratio in
root total nonstructural carbohydrates
(TNC) and in root structural components.
By knowing isotopic ratios of C in TNC
and structural root components, we will be
able to better understand the conditions
under which the carbon was fixed and later
transported to the root. While not a stress-
specific indicator, it does provide an
mtegrated measure of conditions present
during carbon acquisition. The second
approach will be to measure the various
components of soil respiration. Knowing
the C and 02 isotopic signature of each soil
component will allow us to calculate the
contribution of each component of the soil
flux to the total öCl3 signature of the CO 2
collected from the soil surface, including
fine root respiration (this will be
accomplished under subtask 8a).
Methods
Characterizing the ôC13 ratio in root
TNC:
During the first year, large lateral roots will
be identified and traced to donor tree
(initially Douglas firs of several size
classes). Soil will be removed from around
a portion of the root, and samples will be
taken. Roots in three different diameter
classes will be sampled: > 2 mm, 5-20 mm,
and> 10 cm. Large diameter roots (>10
cm) will be cored with a 1 cm diameter
increment corer rather than by excavation.
Samples will be frozen on dry ice for
transport to the laboratoiy, . and
subsequently stored frozen until they are
lyophihzed (Andersen et al. 1991). Freeze-
dried samples will then be ground, and
analyzed for TNC (Wilson et al. 1995) and
carbon isotopic signatures.
Characterizing the 8C13 ratio in soil
plugs:
A trench will be dug approximately 3-5 m
from the base of each of 6 mature trees, in
an area where root proliferation is evident.
To obtain a baseline value, soil-surface CO 2
flux will be measured in undisturbed soil
adjacent to the trench using a Licor 6200
(Andersen et al. 1997a) and using gas
collection apparatus. A horizontal plate
will then be driven into the soil, parallel to
the soil surface, at a depth of
approximately 30 cm. Using the same
location as baseline values, a PVC pipe (15
cm diameter) will be driven into the soil
surface directly above the plate until it
reaches the plate at a depth of
approximately 30 cm. This process will
result in isolatmg an intact cylinder of soil.
Soil surface CO 2 will again be measured
using a Licor and the gas collection device.
After CO 2 collection, the intact plug will be
removed from the soil and taken to the lab
for analysis. The soil plug will be separated
into litter, roots, soil organic matter, and
mineral soil, and dried to quantify the mass
of each component. Dried soil components
will be analyzed to determine their carbon
isotopic signature. Knowing the carbon
isotopic signature of each soil component
will allow us to calculate the contribution
of each component of the soil to the total
6C 13 signature of the CO 2 collected from
the soil surface. Using mass balance
equations (Lin et al. 1998), the contribution
of each component of the soil flux to the
total respiratory flux will be obtainable,
including fine root respiration. By
comparing values before and after cylinder
49
-------
insertion, it will be possible to evaluate
whether significant changes in CO 2 flux
resulted from wound response of severed
roots in the soil column.
Subtask 4d: Reproduction
Objectives and background
The main goals of this sub-task are to (1)
Quantify seed viability, as a measure of
reproductive potential, for dominant tree
species, and (2) quantify stoichiometric
relationships of C and N allocation to
wood, needles, and seeds. Quality of
reproductive tissues plays a significant role
in forest resource allocation and the
perpetuation of tree stands and
communities from one generation to the
next The life of a seed is a complex senes
of events from the initiation of fruiting
through the dispersal of the mature seed
and germination (Krugman et al. 1974).
Fruit, cone, and seed production and
viability are affected by various
combinations of physiological factors,
weather, insects, diseases, predation by
birds and mammals, and anthropogenic
stressors (Bums and Honkala 1990; Harper
1977). We believe that an understanding of
seed quality, and relative allocation of C
and N to tree wood, needles, and seeds
(Waring and Schlesinger 1985) has
potential as a sensitive indicator of forest
condition.
Seed viability
Tree climbers will collect samples of ripe
cones, as well as associated branches and
needles, from the three species in August
and September each year at the Falls Creek
and Toad Creek Road sites. Twelve
Douglas fir cones per tree will be collected
from different heights and sides of teen
randomly selected mature (of seed-bearing
age) trees at each field site. Branch and
needle tissue will also be collected at two
of the 12 cone collection locations on each
sampled tree. Sixteen western hemlock
cones per tree will be collected from four
mature trees at each site. Branch and needle
tissue will also be collected at two of the
16 cone collection locations on each
sampled tree. Thirty-two Pacific silver fir
cones per tree will be collected from eight
mature trees at the Toad Creek Road site
only. Branch and needle tissue will also be
collected at two of the 32 cone collection
locations on each sampled tree. Seeds will
be tested by a recognized tree seed research
laboratory (1 e., USDA Forest Service’s
Dorena Tree Improvement Center) by both
X-ray contrast (Moore 1969) and
germination procedures for reproductive
capacity.
Carbon and nitrogen allocation
C and N concentrations will be determined
on oven-dried subsamples of the seeds,
branches, and needles collected as described
above. The tissues will be ground to pass a
40-mesh screen and analyzed for C and N
using a flash-combustion gas
chromatography method (Carlo Erba SOP
3.0.1 Carbon/Nitrogen Elemental Analysis).
50
-------
• Task 5: Carbon and Nitrogen
Partitioning
Sublask 5a: C partitioning to secondary
metabolizes and structural compounds
Objectives and background
The main goals of this sub-task are to (1)
quantify the partitioning of C into
secondary metabolites and structural
compounds in leaf and fine root tissues of
the dominant trees to determine how the
quantities vary by forest successional
state, climatic factors, and by season and
(2) use the results to contnbute to the
parameterization and verification of the
MBL-GEM (Section 3.2.1).
The C fixed by plants is generally allocated
for growth, reproduction, storage, and
defense (Waring and Schlesinger 1985,
Tuomi et a!. 1988). Major determinants of
plant tissue quality are an array of chemical
and structural compounds, i.e., secondary
metabolites, lignin, and cellulose (Coley et
a!. 1985, Lindroth et al. 1993) that serve as
defenses against herbivory and pathogenic
infection (Coley et al. 1985, Waring and
Schlesinger 1985). Secondary metabolites
and structural compounds m both above
and below ground tissues also influence the
quality of the organic matter inputs to the
soils, and subsequently its decomposition
(Swift et a!. 1979, Homer et a!. 1988) and
quantity of soil organic matter. The nature
and quantity of secondary metabolites in
plant tissues are determined by the
availability and utilization of resources in
the local environment (Coley et a!. 1985,
Lindroth et al. 1993), which in turn are
affected by other stressors such as air
pollutants and global change (Lawler et al.
1997).
Research shows that polyphenols (e.g.,
tannins, terpenes) and fiber in leaves
significantly increases with increasing leaf
lifetimes on the tree. The types of available
resources in an environment also place
constraints on the types of secondary
metabolites plants make. For example, in
N-limited environments one would expect
higher concentrations of C-based chemicals
such as tannins or phenols than N-based
chemicals such as alkaloids (Coley et al.
1985, Lindroth et al. 1993). Recent studies
have shown that conditions in the
environment that increase the C:N ratio in
leaves, such as low soil N or increased
atmospheric C0 2 , significantly increased
the amount of C partitioned to secondary
metabolites (Lindroth et al. 1993, Lawler et
al. 1997). However, to what extent
environmental stressors that reduce C
fixation affect the concentration of
secondary metabolites and structural
compounds is unknown. Further, increased
atmospheric N deposition has the potential
to alter C:N ratios of leaf tissues (Asner Ct
al. 1997) and thus affect the quantity and
nature of the secondary metabohtes.
We propose that the concentrations of
secondary metabolites and structural
compounds have the potential to serve as
indicators of forest condition. Any change
in the quantity and composition of leaf
tissue is likely to change a forest’s
susceptibility to herbivory and pathogenic
infection. Moreover, changes in the
quantity and composition of secondary
metabolites and structural compounds of
both leaf and fine root tissues are likely to
affect the structure and function of the soil
food web, which in turn will likely affect
organic matter decomposition and nutrient
mineralization.
51
-------
Methods
Samples of leaf and fine root tissue will be
collected from the dominant trees in the
intensive study sites. Tissue samples will
be collected four times a year — winter,
spring, summer, and fall months — for at
least 3 yr. Once a seasonal pattern has been
established in the intensive sites, tissue
samples will also be collected from the
extensive sites in subsequent years for
further evaluation of this potential
indicator. The timing of leaf and root
sampling will be coordinated with other
sampling schemes to maximize tissue
collection efficiency. Leaf samples will be
collected from about ten different diameter
trees at each site, and separated into new
and old needles. Samples of fine roots will
be collected m coordination with subtask
Ia (above).
All tissue samples will be dried at 50 C and
ground at 0.85 mm mesh size (Ryan et al.
1990). We will use the methodology
recommended by Ryan et al. (1990) for
determining the concentrations of
secondary metabolites, ligmn, and cellulose.
This methodology is a sequential treatment
of the sample using the wood products
techniques for polar (simple sugars, and
polyphenols) and non-polar (fats, oils, and
waxes) extractives, which we assume are
surrogates for secondary metabolites, and
the forage fiber techniques for cellulose and
lignin (Ryan et al. 1990). The polar
extractives will be further analyzed for
simple sugars by a colonmetric method
(DuBois et al. 1956) and for phenols by
the Folin-Denis colorimetric method (Allen
etal. 1974).
Subtask Sb: Levels of secondary
metabolites and carbohydrates
Objectives and background
The main goals of this subtask are to
determine (1) the amount of within plant,
within community and between
community variability in nonstructural
carbohydrates and secondary metabolites
and (2) if the level of nonstructural
carbohydrates and secondary metabohtes
varies in response to stress.
Secondary metabolites are formed from
photosynthate that is not used for
construction or maintenance, and one
primary function appears to be protection
against predation by animals and
microorganisms (Waring and Schlesinger
1985). The synthesis of secondary
metabolites may be costly to the plant,
requiring a steady flow of precursors from
primary metabolism, enzymes and energy
nch co-factors (ATP, NADPH, etc.)
(Harbome 1993) unless they are being used
for storage. This cost has led to at least
three different but not mutually exclusive
theories on their evolution (Tuomi 1992).
1) The optimal defense theory assumes
that there is a limited amount of resources
that a plant can devote to defense and that
there are alternative demands for these
limited resources (Rhoades, 1979). The
theory concerns evolutionary trade offs
between growth and defense (Tuorni 1992).
The theory predicts that 1) commitment to
defense should decrease in the absence of
enemies and increase when plants are
subjected to a high risk of attack, 2) plants
should evolve defenses in inverse
proportion to the cost of defense and 3)
52
-------
plant parts most valuable in terms of
fitness should be most effectively
defended. Therefore; environmentally
stressed plants should be less well
defended against pathogens and herbivores
than unstressed plants.
2) The carbon-nutrient balance (CBN)
theory suggests that carbon-based
metabolites will be positively correlated
with the carbon-nutrient balance and
conversely that nitrogen-based metabolites
will be negatively correlated with this
balance (Bryant et al. 1983). Bryant et al.
(1983) suggested that phenotypic
responses of secondary metabolism are
governed by a carbon-nutrient balance.
Moderate nutrient stress should in fact
enhance carbon-based defenses. CBN
theory emphasizes site differences in
resource availability as a factor in
differences in secondary metabolite within
a species. Therefore, moderate nutrient
stress should enhance carbon-based
defenses.
3) The growth-differentiation balance
(0DB) theory suggests the existence of a
physiological trade-off between growth and
differentiation with the latter term
mcludmg secondary metabolism synthesis
(Tuomi et al. 1990). The theory has
perennial plants divided into two groups;
1) growth dominated plants with rapid
growth, poor chemical defense but with a
highly inducible resistance system, and 2)
differentiation-dominated plants with a
slow growth rate, well defended with high
levels of toxin but with poorly developed
inducible resistance (Harbome 1993). 0DB
theory emphasizes temporal variation in
resource availability. Rapidly growing
plants should be characterized by lower
and more plastic defense levels than slow-
growing plants. The theory is based on
three assumptions; 1) that genetic
correlations between growth and defense
are negative, 2) that growth is more
sensitive than photosynthesis than to
specific environmental stresses, and 3) that
growth dominated plants have more plastic
defenses than differentiation dominated
plants (Tuomi 1992).
There is limited experimental evidence in
support of these theories. However,
Waring and Pitman(1985) were able to
demonstrate that lodgepole pines
susceptibility to mountain pine beetle
decreased with the lessening of
intraspecific competition or the increasing
the availability of nitrogen. This work
supports the optimal defense theory. It
suggests that the lowest priority was the
allocation of photosynthate to protective
chemicals and therefore a plants first
response to environmental or abionc stress
should be a shift of carbon allocation away
from secondary metabolism. Other work
by Reichardt et al. (1991) tends to support
the CBN model when using carbon based
secondary metabolites with low potential
turnover rates. The lack of consensus and
expenmental evidence on the response of
secondary metabolites to stress suggests
there is a need for further investigation.
In addition, plants may respond to
different stresses including ozone
(Matyssek er al. 1992), herbivory (Ayers
1984) and disease (Farrar 1992) by the
accumulation of nonstructural carbohydrate
in their leaves or stems. Therefore, it is
plausible that high levels of leaf
nonstructural carbohydrate and low levels
of constitutive protective chemicals could
53
-------
be an early indicator or predictor of an
ecosystem in decline. In some cases,
especially were the majority of the
protective chemicals contain nitrogen,
changes in the C/N ratio in leaf tissue may
be a simple but sensitive indicator.
Methods
In order to answer the first goal of
quantifying the amount of variability at
several scales on the levels of nonstructural
carbohydrates and secondary metabolites
in plants, several field sites along the
Sanatiam corridor will be sampled.
Suggested sites would include the current
Falls Creek, Soap Grass, and Toad Creek
sites. Plant species common to the sites
will be used to determine nonstructural
carbohydrates and secondary metabolites
in the roots and leaves. Plant species of
interest are Douglas-fir (Pseudotsuga
menziesii), western hemlock (Tsuga
heterophylla), Oregon grape (berberis
nervosa), salal (Gaultheria shallon) and
bracken fern (Pteridium aquilinum).
Douglas-fir and western hemlock were
chosen because they are common dominant
forest trees in the Pacific Northwest. The
other three species are common understory
or pioneer species following forest
disturbance. The species are taxonomically
diverse and have different secondary
metabolic compounds. The results will
determine the between site variability of
similar communities composed of the same
species.
Sampling of tissues will coincide with
changes in plant phenology. This will
determine the amount of seasonal
variability within and between species and
sites. Trees will be sampled prior to bud
break 1 during the expansion of new growth,
fully expanded growth and after bud set.
Needle samples will be taken by one year
age classes up to three years at three height
levels in the canopy and in all four cardinal
directions. Herbaceous plants will be
sampled as new growth emerges, when
fully expanded and prior to tissue death in
the fall. On plants with over wmtenng
leaves tissue samples will be taken pnor to
bud break, also. Root samples will be
harvested from each plant at the same time
as foliar samples are taken. Initially five
plants of each species per site will be
sampled in order to determine within
community variability.
All tissue samples will be analyzed for
non-structural carbohydrates using High
Performance Anion-Exchange
Chromatography. Samples will be
analysized for total carbon and total
nitrogen using flash combustion
methodology. Trees will be analyzed for
condensed . tannins (l-Iagerman 1987;
Jullcunnen-Tiitto 1985), bracken fern for
cyanogens (Jones 1988), and salal and
Oregon grape for alkaloids (Keeler 1975)
Once the amount of within plant, within
community and between community
variability has been characterized, the same
plant species will be used to address the
second goal of detemiining what effects
stress has on the level of nonstructural
carbohydrates and secondary metabolites
in plants. Differences in response to
nitrogen levels will be determined between
the Toad Creek (low soil nitrogen) and
Soap Creek (high soil nitrogen) sites.
Differences in light quantity between
shaded and unshaded bracken plants in the
clear cuts at the Falls Creek and Toad
54
-------
Creek sites. Eighty percent shade cloth will
cover some of the fern plants while others
will remain in full sun. The bracken ferns
are commonly connected by rhizomes.
Areas around each fern sample area will be
trenched below the rhizome level and back
filled after sheet metal or plastic sheeting
has been place in the trench for a root
barrier. Some plants of salal and Oregon
grape will be protected from herbivoty and
pathogens by spraying with pesticides
using a back pack sprayer or fenced from
large herbivores. Salal and Oregon grape
plants not sprayed with the pesticide will
be sprayed with an equal volume of water.
The amount and timing of tissue sampling
will be determined from the results of the
first goal. The types of analysis will be the
same as descnbed under the first goal.
Subi ask Sc: Root total nonstructural
carbohydrates (TNC)
Objectives and background
The goals of this subtask are to determine
(1) the seasonal patterns of INC
accumulation and use; 2) the optimum time
to measure concentrations; and 3) the
appropriate root diameter class to measure.
Root carbohydrates may be useful as an
indicator of current and past tree “status”
because TNC represents an important
reservoir of energy for the tree during times
of stress (Waring and Schlesinger 1985).
Trees store carbohydrates throughout the
plant to maintain respiratory pools and to
provide energy for new growth in the
spring. Any stress that reduces the plants
ability to store TNC may reduce the plants
ability to deal with stressors such as
insects or fungal pathogens, poor nutrition,
or anthropogenic in origin (Johnson 1989,
Tingey and Andersen 1991).
To establish root TNC as a useful
indicator, historical and seasonal trends
need to be well documented for each forest
species of interest. In addition, tree age
affects TNC status and, therefore,
sampling across a range of age classes will
be required to provide information on the
spatial and temporal variability of this
potential indicator.
Methods
During the first year, large lateral roots will
be identified and traced to donor trees. Soil
will be removed from around a portion of
the root, and samples will be taken. Roots
in three different diameter classes will be
sampled: <2 mm, 5-20 mm, and > 10 cm.
Large diameter roots (>10 cm) will be cored
with a 1 cm diameter increment corer rather
than by excavation. Samples will be frozen
on dry ice for transport to the laboratory,
and subsequently stored at -80 C until
they are lyophilized (Andersen Ct a!.
1991). Freeze-dried samples will then be
ground and analyzed for starch and TNC
content (Wilson et al. 1995).
Root samples will initially be collected
from a subsample of trees at 6-week
intervals to identify the appropriate
sampling times, e.g., sample periods when
tree-to-tree variability is minimal
(seasonally synchronous). After initial
studies, samples will be taken on a larger
number of trees and samples will likely be
restricted to late fall after the first frost,
early spnng prior to bud flush, and mid
summer, after full shoot expansion.
Additional sample periods may be
55
-------
necessary depending on the results of the
initial studies.
Task 6. Phenology
Subsask 6a: Fine fool dynamics
Objectives and background
The objective of this subtask is to employ
minirhizotrons to track individual roots or
root segments over their lifespan to
determine (1) optimum sampling frequency
for quantifying fine root dynamics and (2)
fine root production and mortality over
time and environmental gradients. Data on
fine root dynamics will also contribute to
the parameterization of the MBL-GEM.
Under normal conditions, the allocation of
photosynthate to the various plant
compartments (leaves, stems, roots,
reproductive tissues) proceeds at set rates,
priorities, and seasons. If plants are
stressed at a level beyond which the plant
can accommodate, this pattern of
carbohydrate allocation is altered. It has
been shown that ozone exposure causes
carbohydrates usually destined for both
fine root growth and large root storage to
be retained in the leaf instead for
maintenance and repair (Hogsett et al.
1985, Andersen Ct al. 1991, 1997b).
Stressors also impact C allocation to fine
roots leading to alterations in fine root
growth and life span. Carbon transport to
roots and new root growth is reduced in
ozone treated ponderosa pine seedlings
(Andersen and Rygiewicz 1995; Andersen
et al. 1991, 199Th). Although we have not
yet tested the hypothesis that ozone
increases root mortality and turnover, soil
organic matter is greater in ozone than
control treatments, which is consistent
with this hypothesis. In contrast, elevated
CO 2 increases carbon allocation below
ground resulting in increased fine root area
density (Tingey et al. 1996) and fine root
life span (Tingey et at. 1997). This
sensitivity of fine roots to various
anthropogenic stressors suggests that fine
root dynamics have the potential to serve
as early warning indicators of forest
condition.
In most plant communities, root systems
are poorly understood and data from
natural plant communities are hmited
(Hendrick and Pregitzer 1996).
Minirhizotrons are well suited to study
natural plant communities as they are
relatively small and permit the continuous,
in situ, non-destructive monitoring of fine
root processes over long time periods. A
unique strength of the minirhizotron
approach is the ability to study individual
roots or root segments over their life-
spans. As individual roots are tracked the
fine root production and fine root mortality
is directly observed. Processes like root
production, elongation and mortality are
measured separately rather than inferred
indirectly using mass-balance procedures
on soil cores (Hendrick and Pregitzer
1996). Only a few studies have directly
measured root production and mortality as
istinci processes which is a sigmflcant
limitation in forest ecological studies
(Hendrick and Pregitzer 1996). Although
minrhizotion systems are useful for
montoring fine root production and
turmver. they are difficult to use for
est nating root bioimss (Upduurch 1987,
Merrill and Upchurch 1994). However,
mmwhizotzon data have been converted to
root lengh densities whith can be used to
estimate bioimis if speafic root lengh is
56
-------
known (Upthurch 1987, Menill and
Upchurch 1994).
Methods
Basic minirhizotron approach
Minirhizotron tubes (5 cm inside diameter,
fitted with a water-tight PVC plug on the
soil end) will be installed at an angle 450
from vertical; extending 1.4 in into the soil
(1 m vertical depth). The aboveground
portion of each tube is painted to exclude
light and covered with a closed-cell foam
rubber and a PVC cap to keep out moisture
and minimize heat exchange between the
tube and the air. The application of
minirhizotron technology has been
descnbed in detail by Brown and Upchurch
(1987). Minirhizotron images will be
recorded periodically on S-VHS tape using
a minirhizotron camera (Bartz Technology
Company, 650 Aurora Ave., Santa
Barbara, CA, U.S.A. 93109) (Tingey et al.
1995, 1996). The camera is remote
focusing, with a white light source and
equipped with an indexing handle that
locks into position in an index hole in each
minirhizotron tube. The indexing handle
(Johnson et al. 1997) has a ratchet
advancing mechanism and regularly spaced
detents to reliably advance the camera from
one field of view (frame) to the next. The
indexing handle system insures that the
camera is returned to the same position in
each tube and travels along the same
vlewmg line each time images are collected.
Root images will be recorded on the
uppermost surface of the minirhizoiron
tubes beginning at the bottom of the tubes.
In this application the minirhizotron
camera has a field of view of about 1.8 cm 2
(1.1 cm high by 1.6 cm wide). Based on
tube length and video camera field of view a
continuous soil strip representing about
10% of the total surface area of a
minirhizotron tube is sampled.
The minirhizotron video images will be
analyzed using “MSU-ROOTS”, an
interactive PC-based software program
(Hendrick and Pregitzer 1992, 1996). For
analysis, the video images are displayed on
a video monitor; the combination of image
collection and display on the video monitor
magnifies the images approximately 25
times. The software allows the user to
review all images and use a mouse to trace
various root features (length and diameter)
and annotate mycorrhizae and fungal
hyphae occurrence. For roots which
branch, each branch is tracked as a separate
root segment. Each digitized root is
assigned a developmental class (SOP 6.02,
Minirhizotron Image Data Extraction
Using the MSU-ROOTS Software System)
All roots are “new” at the first observation
while roots that disappear between
samplings are classified as “missing. Roots
are separated into coarse (> 2 mm
diameter) and two classes of fine roots (
1mm and 1 to 2mm).
The occurrence of mycorrhizal fungi are
inferred by the presence of monopodal,
bifurcated or highly branched root tips.
Fungal hyphae will also be noted but it is
not possible to distinguish among various
types of fungi from the hyphae observed in
the images.
Several types of data will be obtained from
the mimrhizotron images (Hendrick and
Pregitzer 1992, Majdi 1996): (1)
occurrence of fine ( 2 mm diameter) roots,
ectomycorrhizae, and fungal hyphae,
measured as the proportion of images
57
-------
containing these structures (Tmgey et al.
1995, 1996); (2) length and diameter of
individual fine root segments (Tmgey et a!.
1996), and (3) lifespan and mortality of
individual fine root segments and
mycorrhizae (Johnson et al. 1997;
Rygiewicz et al. 1997) (see also Subtask 9a
below).
Sample frequency to assess fine root
production and mortality
Minirhizotrons are well suited for
measuring fine root production and
mortality (e.g., Hendrick and Pregitzer
1966, Majdi 1996). However, the
appropriate sample frequency has not been
determined and the frequency may vary
among species. To estimate total tine root
production it is important not to “miss”
any production dunng the interval between
image collections. This is especially a
problem if new roots are produced, die and
disappear between image collection
intervals. In ponderosa pine, median
lifespan of tine roots can be as short as 50
days during the summer when the soils are
warm (Johnson ci al. 1997). Given such a
short median lifespan for ponderosa pine
and asswning that the median lifespan for
Douglas-fir is similar, it will be necessary
to collect images every 2 to 4 weeks to
estimate the production and mortality rates
reliably for Douglas fir. The key feature of
this research is to determine the best
sampling frequency to determine “total”
fine root production.
We propose to place minirhizotron tubes
around several Douglas-fir trees and collect
images at least twice a week for a year. The
images will be digitized and then the data
base subsainpled at different time intervals
to establish the minimum sample frequency
required to capture fine root production
and mortality. Soil temperature and soil
moisture will be measured over the course
of the experiment. Air temperature, solar
radiation and precipitation are available
from an adjacent meteorological tower.
Field measurement of fine root
production and mortality
Little data exist showing the total
production of fine roots and their mortality
rates, under natural conditions, at the stand
level. The sample collection frequency will
be based on the previous study that
established the optimum frequency to
quantify all new fine root production. To
determine fine root production and
mortality rates, minirhizotron tubes
(25/site) will be placed at the High and
Low intensive field sites. The
minirhizotron tubes will be co-located with
htterfall traps (section 3.2.1), N-
mineralization studies (section 3.2.1), and
soil cores collected to estimate fine root
standing crop (subtask 4a). As the various
samples are co-located we will compare
various methods of estimating new root
production and turn-over (section 3.2.1).
The data on fine root production and
mortality will also be used to determine
seasonal patterns of fine root formation
and loss for the forest stands. These data
will be linked with seasonal stem growth
patterns obtained from dendrometers (see
section 3.2.1) placed on individual trees
(covering the species present) at each site
to determine the temporal relationship
between fine root dynamics and stem
growth.
58
-------
3.3.2. Resource utilization-
consumers
Task 7: C and N Transformations
Subtask 7a: Soil organic matter fractions
Objectives and background
The goal of this subtask is to characterize
the forms and chemistry of C fractions in a
range of forested soils. Comparisons
between the soils at the intensive forest
sites (section 3.1) and the adjacent clear-
cuts will be used to identify changes in
SOM (0 horizons and mineral soil) due to
management and to develop SOM-based
indicators. Soils are an important
component of forested ecosystems having
vital roles in supplying water and nutrients
for plant growth and maintenance and in
ecosystem stability and productivity
(Agren et al. 1996). Forests generally have
a fairly tight nutrient cycle that relies upon
decomposition of plant denved organic
matter to supply nutrients. Perturbations
in decomposition that upset or interrupt
the nutrient cycle can affect the stability of
the system. Stressors such as elevated CO 2
and/or management may affect soil
processes, such as decomposition, to that
extent that the whole system is affected.
The character and composition of SOM
fractions may provide valuable information
about the system stability and the effects
of stressors on the trajectory of system
stability (e.g., decreasing SOM may signal
decreasing system stability).
The amount of soil organic matter in any
soil is a function of: climate, relief, parent
material, time, and quantity and quality of
organic inputs. Natural and anthropogenic
factors that affect any of these SOM
forming factors will affect the quantity and
quality of SOM (Oades 1988). Because
SOM can have mean residence times of up
to hundreds to thousands of years (Post et
al. 1982), its characterization may provide
long-lived metrics of ecosystem status and
indicators of ecosystem change.
Methods
Basic soil and litter chemistry
Litter and mineral soil samples will be
collected with a corer four times per year,
to capture seasonal variability, in the forest
and adjacent clearcut at the intensive (Falls
Creek and Toad Creek Road) field sites.
Samples will be collected from five
randomly selected subplots of the intensive
field sites. Soil samples will be collected
from all horizons. Litter layer samples will
be separated into two horizons, the Oi
(slightly decomposed) and Oa (highly
decomposed). Each time samples are
collected their bulk density and moisture
content will be measured.
The concentrations of nutrients will be
measured in the soil and litter samples.
Field moist samples will be used for pH,
extractable N and extractable cations
(Nutrient Analysis SOP#3.02 Macro-
nutrient analysLs by suppressed ion
chromatography with conductivity
detection; Alpkem Autoanalyzer SOP#
3.10; and ICP Analysis SOP#3.04 version
2.00). Extractable S and P will be run on
fresh mineral soils (Soil, Litter and Plant
Tissue Preparation EP#01). Freeze-dried
samples will be used for total C and N
(Carbon and Nitrogen Elemental Analysis
SOP#3.01), and a total elemental analysis
(ICP Analysis SOP#3.04 version 2 00).
59
-------
Ratios of SOM to various nutrients may
provide useful system indicators.
Soil organic matter quality
For this study quality refers to the ease
with which SOM is decomposed. Rapidly
decomposed SOM serves as a desirable
substrate for decomposers and is high
quality. Recalcitrant SUM is an undesirable
substrate and consequently poor quality,
and more likely to have a much longer
residence time.
Samples from the uppermost soil and litter
collected as described above will be used
for the studies described below. We
propose using two measures of soil and
litter quality. One will be based upon the
methods given m Cambardella and Elliott
(1992) who developed a physical
separation method that links measurable
fractions of SUM to kinetically-defined
pools. This method includes dispersion,
size fractionation, density separation and
chemical analysis. Second we will use
incubation studies to characterize the CO 2
production, 02 consumption, N-
mineralization and decomposition of soil
and litter samples (Nadelhoffer 1990;
Catricala et al. 1995). Soil and litter
samples will be incubated under optimal
moisture and temperature conditions to
facilitate rapid decomposition. During the
incubation, production of CO 2 and
consumption of 02 will be measured.
These data will be used to investigate the
kinetics of decomposition. Also, during the
incubations the soil and litter samples will
be subsampled and total C, N and
extractable N will be measured. These data
will provide information about changes in
the quantity of soil and litter C and N pool
and the timing of N mineralization. These
studies will provide information on the
spatial and temporal differences in SOM
quality and the differences due to
management ( e.g., cut versus uncut). This
information will lend itself to the
development of SOM-based indicators of
forest condition.
Soil organic matter characterization
Classical SOM characterization methods
use assorted extracting solutions to
separate it into operationally defined
organic fractions (e.g., base extractable
humic acids, Stevenson et al 1989). We
propose using solid-state cross-
polarization/magic-angle-spinning
NMR (CP/MAS ‘ 3 C NMR) to charactenze
the composition of SOM. This technique
has been described by Baldock et al. (1992)
and can be used to examine the
composition of SOM. Because the
CP/MAS 13 C NMR technique can be used
to observe subtle changes in the
composition of SOM we will use this
technique to characterize the C in a subset
of the soil and litter samples; if the results
appear to be promising more samples will
be analyzed. Instrument time permitting,
we intend to integrate CPIMAS ‘ 3 C NMR
characterization with our incubation
studies and with the SOM separation
methods described by Cambardella and
Elliott (1992 and 1994) and by Baldock Ct
al. (1992). These results will provide
information on SOM processing. Other
techniques, such as those described by
Ryan et al. (1990) may also be employed
to facilitate integrating this SOM research
with the MBL-GEM component (section
3.2.1).
60
-------
Task 8: C and N Losses
Subtask 8a: Efflux of soil carbon
Objectives and background
The specific research objectives of this
subtask are to: (1) quantify the diurnal and
seasonal changes in soil C efflux, as C0 2 ,
using a newly-developed surface in situ
measurement device; (2) quantify the
easonal changes in the relative
contributions of root, litter decomposition,
and SOM oxidation to total soil efflux; (3)
calculate the direction in the net change of
SOM; and (4) use output from these
results to contribute to the evaluation and
verification of the MBL-GEM (section
3.2.1). The soil-effiux fractionation method
and a newly-developed technique for
automatically and semi-continuously
monitoring soil efflux will be used to
address these objectives.
The direction of change in long-term
functioning of a terrestrial ecosystem,
especially of temperate forests, relies
partly on changes in the amount of soil
organic matter (SOM) (Tilman 1997).
Specifically, forest functioning is related to
the balance between accrual of SOM when
the system exists either in the undisturbed
condition or while it is recovering from a
disturbance, relative to SOM losses due to
disturbance. If the system can not
replenish lost SOM before disturbance
reoccurs, the trajectory is negative.
Similarly, if a chronic stress causes SOM
to be depleted faster than it can be
replenished, ecosystem health may be
compromised. Conversely, if the chronic
stress depletes SOM at a rate less than the
SOM replenishment rate, long-term
functioning may not be affected.
Observing a net change in SOM can be
difficult because it may entail measuring
small changes in a large pool. In our work
in the teracosms on the effects of elevated
CO 2 and climate change on a Douglas-
fir/soil system, we developed a method to
partition soil litter CO 2 efilux into its
component sources of root respiration,
litter decomposition and oxidation of SOM
(Lin et al. 1998). This method relies on the
simultaneous determination of the natural
abundance of ‘ 3 C and/or ISO in the soil
efflux, and of the C and water in the litter
and soil. By using a mixing model, soil CO 2
efflux is partitioned into its components by
using a 2-end-member linear model for the
5180 value of C0 2 , and a 3-end-member
triangular model for the 6’ 3 C value of the
efflux (Lin et al., 1998). Thus, efflux of C
from the SOM pool can be determined. Net
flux (influx - efflux) of SOM C is calculated
after determining influx of C to the SOM
pool using a similar approach relying on
‘ 5 N and ‘ 3 C (analyses and model
development are on-going for the work
being done in the teracosms).
Methods
Determining the relative contributions of
each C source (root, litter decomposition,
and SOM oxidation) to total soil efflux
involves the following two experimental
techniques: 1) quantifying and capturing
soil CO 2 effiux for subsequent isotopic
analysis, and 2) collecting and processing
soil, root, and soil water samples for
isotopic analysis. Details for the method to
determine influx of C to the SOM pool
using N and ‘ 3 C are still under evaluation
from samples collected in the TERA 1
experiment.
61
-------
Soil CO 2 efflux estimation
Soil efflux will be estimated using a newly-
developed technique for automatically and
semi-continuously monitonng soil efflux.
The linear in situ soil-litter respirometer
(LISR) was developed to estimate soil
efflux automatically, semi-continuously,
and non-destructively. The sample area of
a single LISR is 1 cm wide by 1 m long,
thereby providing an integrated measure of
soil efflux. Effiux estimates are recorded
automatically and semi-continuously using
a Campbell Scientific CR1OX datalogger.
Placing the LISR on the soil-litter surface
minimizes disturbing roots or mycorrhizal
hyphae located in the litter layer or mineral
soil.
Prior to using the LISR we will evaluate it
under field conditions to see if it has a
“footprint” effect that could bias the
results of CO 2 release rates. At present, the
relative and absolute accuracy and
precision of soil efflux estimates obtained
by the LISR and the LICOR Soil
Respiration Chamber are being evaluated
using a controlled-environment test
chamber. Soil efflux estimates derived by
the LISR are within 5% of values obtained
using the LICOR headspace chamber. Low
labor requirements for its operation, and
high portability make the LISR a useful
tool for measuring soil efflux in a variety of
ecosystems. Assessing spatial variability in
soil efflux is accomplished by hooking
several LISRs to the datalogger equipment.
Automated operation of the LISR readily
permits collection of data describing the
diurnal and seasonal patterns in soil efflux.
The LISR can be modified to readily permit
soil gas samples to be collected for stable
isotope analysis.
A single estimate of soil efflux will be
estimated once every hour at five sampling
locations over a 48-hour penod. These
estimates will be made once every four
weeks for an entire calendar year (i.e., 13
sample periods) at the intensive (Falls
Creek and Toad Creek Road) field sites.
The five sampling locations will be
positioned adjacent to any litterfall traps,
nitrogen mineralization tubes,
minirhizotron access tube locations, and
soil moisture and soil temperature sensors
installed so that soil processes can be more
fully descnbed and incorporated into any
statistical inferences made. We will also
attempt to measure wintertime soil efihix,
thereby more fully understanding the
contribution of soil processes to the annual
ecosystem C cycle. Prior to snowfall, five
LISRs will be placed on the ground and
their gas sample plumbing lengthened for
wintertime measurements. At each sample
period when snow is present, another LISR
will be placed directly over the LISR
positioned on the ground. This will allow
near-simultaneous estimates of soil efflux
at the soil surface and at the snow-
atmosphere interface.
Soil respiration partitioning
The dual isotope (180 and ‘ 3 C ) method is
based on the differences in natural
abundance of ‘ 3 C among the three C
sources, and differences in natural
abundance of 180 between litter water and
soil water. We demonstrated the success of
the method in the TERA I experiment
where the source of CO 2 for the newly-
formed C (i.e., C accrued during the climate
exposure treatments) was a tank CO 2 of
geologic origin (i.e., highly depleted of ‘ 3 C).
We expect thaf the differences in natural
62
-------
abundance of the isotopes among the
sources of soil efflux at our field sites will
be sufficient so that the method will work.
It is the differences in the respective 8’ 3 C
and 6180 values among the component
pools for the two isotopes that will
determine if the method will work. For
example, in the field, the 6’ C value now
for recently-fixed CO 2 is likely not too
different from that of CO 2 fixed in the past.
From our preliminary data of respective
differences in the field, there are sometimes
sufficient differences which are dependent
on climatic conditions. We will undertake
an expanded analysis of this work initially
to confirm our expectations.
Although the use of stable isotopes of C
and 0 individually has been used
successfully to measure a variety of
ecosystem processes (cf., Keeling et al.
1979, Ehleringer et al. 1986, Peterson and
Fry 1987, Farquhar et at. 1993), the
simultaneous application of these isotopes
to study root vs. microbial fractionation
has not been accomplished until recently
(Lin et at. 1998). The relative contributions
of root and microbial respiration to total
soil efflu.x wilt be estimated by measuring
the ratios ER] of ‘ 3 C/’ 2 C and 1801160 in
roots, soil water, and forest litter. These
isotopic ratios are usually expressed in 8
units relative to a standard [ Pee Dee
Belemnite or Standard Mean Ocean Water
(SMOW)] where 8’ 3 C or 8180 =
((R ,JR )- 1)1000.
The C isotopic ratio of soil efflux is
determined by the 6’ 3 C of its C sources
which are derived primarily from roots and
litter. Litter, comprised of tissue with C
isotopic ratios reflecting initial fixation of
atmospheric CO 2 and re-fixation of soil-
respired CO 2 (soil efflux). Therefore, a C
isotopic ratio gradient exists, vertically and
horizontally in the forest canopy. An
accurate assessment of the 6’ 3 C across the
presumed vertical gradient of 6’ 3 C in the
titter needs to be made and a representative
value needs to be determined. It is assumed
that the vertical gradient of titter 81 3 C also
exists due to seasonal variation in canopy
photosynthesis. Because root respiration is
not subject to these seasonal re-fixation
processes, its 8’ 3 C should be different from
the litter.
Surface 6 180 values of soil water are higher
than values deeper in the soil profile due to
a higher evaporation rate of the lighter
oxygen isotope (Craig and Gordon 1965).
As CO 2 is produced, the oxygen isotope of
CO 2 comes into equilibrium with the
oxygen isotope of the soil water at the site
of CO 2 production. In other words, the
isotopic ratio of soil efflux depends on the
8180 value of the soil water at the site
where the CO 2 is produced since soil efflux
quickly comes into equilibnum with soil
water through a carbonic anhydrase-
mediated reaction (Hesterberg and
Siegenthaler 1991). Consequently, one is
able to distinguish whether the C in the soil
efflux originated from SOM oxidation or
from litter decomposition.
The following sampling protocol wilt be
followed:
A.ir Soil efflux will be collected by
shunting gas collected by a LISR into a 2
liter vessel. A single sample will be
collected from each LISR once during each
sample period (n = 13). Respired CO 2 gas
samples will be collected at approximately
63
-------
the same time as soil, root and litter; and
soil H 2 0 samples are collected (see below).
The soil efflux sample will be cryogenically
extracted and then the isotopes analyzed
using WED’s Finnigan delta S Isotope
Ratio Mass Spectometer (IRMS). It is
critical to extract the C from the CO 2
within 4 hours of collection to avoid
altering the C isotope signature through
reactions with the water vapor present in
the gas sample.
Soil, roots, and litter : Three soil cores, 2.54
cm diameter, will be collected within 5 cm
of each of five LISR’s during each sampling
period (n=13). Soil cores will be collected
at equal intervals (i.e., 25, 50, and 75 cm)
along the 1 m long LISR once each sample
period. Soil cores will be collected at
approximately the same time that soil
efflux samples are collected. All soil cores
will be separated into litter, and if possible,
shallow (A-horizon) and deep root zone
(B- and C-horizons) components and the
components placed m a dry ice container
for transport to WED. Each soil
component (A, B, and C horizons) will be
further divided into root and soil fractions
at WED.
A 50 mg sub-sample from the shallow (A
horizon) and deep (B and C horizon) root
zone soil sample will be dried and ground
for isotopic analysis. ö’ 3 C values for each
component will be determined by analysis
in the WED IRMS. Water samples from
litter, and shallow and deep roots will be
cryogenically extracted using a vacuum line
extractor. A 1 cc water sample of each
component will also be analyzed for
isotopic ratios, using a modified CO 2
equilibrium method described by Socki et
al. (1992), and the WED IR.MS.
Task 9: Complexity and Function of
the Soil Food Web
Subiask 9a. Mycorrhizal symbionts
Objectives and background
The main objectives of this task are to: (1)
calculate lifetimes of mycorrhizae. i.e., time
between formation and disappearance to
determine turnover, (2) measure carbon
allocation to extraradical hyphae, (3)
calculate colonization levels of root tips (%
mycorrhizal tips), and 4) estimate natural
variation in the community of fungi
forming mycorrhizae. Mycorrhizae, by
acquiring nutrients for their hosts and
acting as carbon sinks, are important for
plant carbon allocation (Rygiewicz and
Andersen 1994), plant nutritional status
(Rygiewicz et al. 1984a, b; Bledsoe and
Rygiewicz 1986), nutrient cychng (Harley
and Smith 1983), and the regeneration of
plants and, hence, the sustainability of
forest ecosystems (Perry et al. 1989). We
propose therefore that measures of the
structure and process rates of the
mycorrhizal community have the potential
to serve as forest indicators.
Mycorrhizal fungi are among the first soil
biot.a to receive C from plants, and to
acquire nutrients and water taken up by
plants. Consequently, subsequent release
of rootlmycorrhizal C as exudates and
turnover, supports soil foodweb activity
and many “downstream” processes in
terrestrial ecosystems. Process rates among
mycorrhizae and mycorrhizal fungi are
highly variable in vitro, and also
presumably in the field (e.g., Harley and
Smith 1983, Zhu et al. 1988, Wagner et al.
1989). It is important, therefore, to
evaluate rates while considering the
64
-------
community structure of the fungi forming
the symbiosis. Stressors may influence,
directly or indirectly, the capabilities of
mycorrhizal fungi to colonize roots, and
acquire nutrients and other resources.
(Andersen and Rygiewicz 1991). When
scaled to the ecosystem, changes in the
community structure of mycorrhizae may
greatly alter resource balances. For
example, Fogel and Hunt (1983) found that
up to 70% of net primaiy productivity in a
Douglas-fir ecosystem was invested in
growth and maintenance of roots and
mycorrhizae.
How anthropogenic stressors . acting on
forest ecosystems affect community
structure and processes of mycorrhizal
fungi is not well known. Stressors due to
air pollution or global change can increase,
decrease or have no effect on C allocation
belowground with subsequent similar
changes in mycorrhizal fungi (Andersen
and Rygiewicz 1991, 1995). Responses to
stressors may result in larger or smaller
root systems with different allometric
relationships among root sizes (e.g.,
changed proportions of mycorrhizae, and
non-mycorrhizal fine, intermediate and
coarse roots). Changes in C allocated
belowground also may cause changes in
colonization rates by mycorrhizal fungi and
development of exiraradical hyphae.
Considering the potential variability in C
requirements of fungal species, it is
prudent to consider both the dynamics and
demographics of mycorrhizal fungi in
ecosystem indicator development.
Methods
Lifetimes,
ext ra radical
colonization levels
These will be determined using the same
minirhizotron tubes descnbed in subtask
6b. The dates of formation and
disappearance of each mycorrhizal tip
encountered in the tubes will be recorded
and used to calculate lifetimes. If a large
enough sample population is encountered,
cohort analyses (monthly, seasonal, annual,
etc.) for lifetimes will be done. Total
number of non-mycorrhizal and
mycorrhizal root tips will be recorded to
calculate colonization rates. For each
minirhizotron frame containing
mycorrhizae, an estimate will be made of
the percent of the screen containing hyphae
that appear in some way associated with a
mycorrhiza. This is an imperfect approach,
but it will allow us to estimate C allocation
to extraradical hyphae in situ. Independent
estimates of mycorrhizal colonization will
be determined using root samples obtained
from the soil cores taken periodically from
the intensive field sites in coordination
with the sampling scheme described in
subtask 4a and 6.
Natural variation in
structure of fungi
mycorrhizae
The structure of the mycorrhizal fungal
community will be assessed at the
individual mycorrhizal tip level. Soil cores
will be taken periodically from the
intensive field sites in coordination with
the sampling scheme descnbed above
(Subtask 4a and 6b). Results from the
TERA I project indicate that overall
seasonal variation in the community
carbon allocation
hyphae,
to
and
community
forming
65
-------
structure of qiycorrhizae on Douglas-fir is
low (i.e., comparing spring samples, taken
when soils are at field moisture capacity
and cool, with fall samples, taken when soil
moisture is very low and temperatures are
high). We do not have seasonal results for a
more mature forest stand, so we are unable
to suggest an exact sampling schedule. We
intend to sample more frequently during
the first year, and then adjust the coring
schedule accordingly.
The structure of the mycorrhizal
community will be assessed both
phenotypically (categonzing mycorrhizae
into morphotypes based on gross
morphology), and genotypically (using
molecular methods). We have become
proficient at categorizing mycorrhizal tips
of Douglas-fir seedlings of the TERA I
project. After the tips are assigned to
morphotypes, a selection of replicate tips
per morphotype is subjected to
polymerase chain reaction (PCR)-RFLP
analysis to determine the fidelity of the
initial categorizations. Additional PCR-
RFLPs are produced in any morphotype
where genetic variation occurs. If
necessary, the entire morphotype
community is then re-categorized according
to the molecular analyses. Ontogenetic
processes and certain stressors (such as N
loading) may interfere with using only
gross morphological features to analyze the
community since morphology may be
affected.
We found that the ribosomal DNA primer
pair generally recommended and used to
produce PCR-RFLPs of fungal DNA from
ectomycorrhizae does not work for
Douglas-fir (both the host and mycobiont
DNA are amplified). We identified other
ribosomal DNA sequences having high
homology with basidiomycetous DNA to
construct primers for Douglas-fir
mycorrhizae. We constructed two nested
primer pairs which are used in nested-PCR
amplifications. Using two primer pairs,
rather than one pair, increases the
specificity for the mycobiont DNA. We
tested the two new primer pairs with a
limited number of Douglas-fir mycorrhizae
and obtained mycobiont amplification
without host DNA amplification. The PCR
reaction is being optimized.
Until we can estimate natural variation in
the mycorrhizal community structure,
temporally and spatially, we will not be
able to distinguish natural cycles due to
season or succession from changes induced
by anthropogenic stressors. The work
proposed here in this task, along with work
completed or ongoing in the TERA 1 and
TERA II experiments, will yield an
analysis of how the host-symbiont
association integrates changes in C and N
allocation belowground. Changes in
community structure, coupled with
changes in other measures developed in this
plan may prove useful as indicators.
Subtask 9b. Soil foodweb biotd
Objectives and background
The objectives of this sub-task are to (I)
investigate the composition and complexity
of soil foodweb organisms including
microarthropods, nematodes, bacteria, and
fungi present in litter, soil, and the
rhizosphere as a function of forest
successional state and season, (2) measure
the molecular diversity and activity of
associative Nrflxing bacteria, fluorescent
pseudomonads, and symbiotic and free
66
-------
living fungi, and (3) measure key
physiological and metabolic soil microbial
community activities associated with the
above parameters. To address these
objectives, we are taking a diverse
methodological approach. This involves the
use of established techniques (population
biology) as well as the application of newer
methodologies such as community
metabolic analyses (Biolog) and DNA
fingerprinting techniques to relate foodweb
composition to important functional
aspects of forest responses to stressors.
We propose that these structure and
functionality parameters have the potential
to serve as indicators of C and N
processing in forest soils.
The quantity and quality of herbaceous and
woody plant materials and litter may be
expected to vary with different
environmental and edaphic conditions,
plant community composition,
successional stage, and anthropogenic
stressors. We hypothesize that these
variations in organic inputs in turn will
affect the diversity, population sizes, and
rates of nutrient and C utilization by litter
and soil foodweb biota, and hence the rate
of C and N cycling (Fig. 3-6). The model
shows the linkages between initial
fragmentation, detritivores, and trophic
roles of invertebrates with other foodweb
microbiota and plant tissues, with the
subsequent metabolic roles of microbes
(bacteria and fungi), in ecosystem level
processing of aboveground and
belowground plant inputs.
We assume that plants exposed to
anthropogenic stressors will have altered
patterns of C and N allocation. Differences
in the composition of living plant tissues
(shoots and roots), litterfall and root
exudates will bring about changes in the
population size, diversity and functions of
foodweb biota in litter and soil. Changes in
the population size, diversity and biomass
of foodweb components are proposed to
affect the rates, amounts and types of
microbial metabolism of both C and N
compounds. Resultant differences in C and
N metabolism will then impact the amount
of CO 2 released from soil to the
atmosphere and the amount of N available
for plant uptake. The amounts of CO 2
released and N available for plant uptake
would in turn be expected to potentially
impact plant productivity for example.
Furthermore, once altered, the foodweb
microbiota may induce a feedback that
could further affect the nutritional status of
the plantlrhizosphere/soil complex.
Methods
Samples will be collected from the
mtensive and extensive field sites described
above (Section 3.1). Samples will be
collected from 2 m x 2 m subplots. Litter
will first be removed, using gloved hands to
prevent cross contamination, from a 20-cm
diameter circular area for the forested plots
and a 60-cm circle for clear-cut plots to
have sufficient biomass. Soil samples will
then be taken with 3-cm diameter
aluminum cores to depths of 0-10 cm and
10-20 cm; this is where the highest
densities of foodweb populations aiv
located. Contents of the cores will be
separated into the two depths, mixed by
hand, and subsamples taken to satisfy all
the biological measurements described
67
-------
foodweb structure and composition
.,o—$ons s
omm
DNA fing.rprints
Rates/amounts N metabolism
Nitrogen fixation
N mineralization
N Immobilization
Nltr lf lcation
Denitrifleation
Volatilization
Nitrooen untake - c•. N k ’s Sc
Rates/amounts C metabolism
Metabolic profiles
Respiration
Decomposition
Leaching
/
Figure 3-6. Conceptual model showing the role of
metabolism responses to anthropogenic stressors.
below. Over time, samples will be taken
diagonally across the 4 m subplots to
maximize the space available for coring
within the designated plots. Data on soil
foodweb structure and composition will be
correlated with several of the field
measures gathered from the work desciibed
above (section 3.2.1, Task 1) or from
additional measures described here.
the soil foodweb in mediating soil C and N
Composition and complexity of
foodweb organisms:
Bacterial antifungal counts
Samples of litter and soil will be diluted in
sterile buffer, shaken, and spread plated
onto selective media as described in
Donegan et aL. (1995) to determine viable
populations of bacteria, fungi, and spore-
forming bacteria (USEPA SOP 7.16m
Microbial Plating).
umm [
Changes in plant chemical composition and blomass
In response to anthropogenic stressors
C
Abov.g ound C and N
quality and quantity
1
11
Carbon
Dioxide
I
B&owground C and N
quality and quantity
Utterfall Root exudates Root turnover
I
I
#7
41 afld fitter
respiration
68
-------
Fungi found to increase or decrease with
the seasons will be grown in pure culture
for subsequent taxonomic, phylogenetic
and functional characterization based on
microscopic morphology (Raper and
Fennell 1965, Domsch et al. 1980, Stone
1993, USEPA SOP 7.14 Fungal Colony
Growth and Hyphal Morphology on Agar
Media) and DNA fingerprinting (see
below). Colony morphology richness will
be determined to identify shifts in clonal
and species composition associated with a
given season and successional stage. The
Shannon-Weaver index will be used to
mathematically evaluate fungal richness in
the various ecological situations. By using
cultural biotyping, colony morphotyping,
and DNA flngerpnnting patterns we will
index fungi so that their potential as
indicators may be developed, evaluated,
and defined.
Nematode and microarthopod extraction,
identification, and enumeration
Studies in a diverse range of soils have
shown that soil fauna (nematodes,
microarthropods and also protozoa)
mediate approximately 15% of the C and
30% of the N turnover in the soil
(Anderson 1995). Because nematodes and
microarthropods have high species richness
and a complete representation of trophic
groups (fungivores, bacterivores,
herbivores, omnivores, and predators),
measuring their numbers and species can
provide sensitive detection of changes in
many processes of forest ecosystems.
Changes in the population dynamics and
structure of nematodes and
microarthropods have been used as
bioindicators of the effects of several
environmental stressors (Parmelee 1995,
Larink 1997, Zunke and Periy 1997).
To determine the composition and
population complexity of soil and litter
invertebrates, samples of soil and litter will
be weighed and placed into modified
Baermann funnels and extracted as
described by Donegan et al. (1997)
(USEPA SOP 7.04). Total numbers of
nematodes will be counted and expressed
as the number per g of dry weight of soil or
litter. For extraction of microarthropods,
soil and litter samples will be placed into
modified Tullgren funnels in beakers
containing water (Donegan et al. 1997).
Microarthropods will be collected and
identified to class (Diploploda, Chilopoda,
or Symphytla) or to order (Acari, Protura,
Collembola, Diiplura, and order of insect)
according to Borror et al. (1992). The total
number of each class or order will be
expressed as the number per g dry weight
of soil or titter. In key samples, about 30-
50 nematodes per sample will be randomly
picked and transferred to slides for
identification to trophic groups according
to Yeates et al. (1993) and Donegan et al.
(1997).
Molecular measurements
Structural and firnctional analyses
associative nitrogen fixers
of
Use of DNA oligonucleotides as markers is
facilitating in situ detection, enumeration,
and determination of metabolic activities of
both culturable and non-culturable
microbes found in natural habitats (Seidler
and Fredrickson 1995, Ueda et al 1995).
Using modified niji-! primer sequences, we
have found DNA fingerprints that indicate
that certain N 2 -fixing organisms are unique
69
-------
to litter in mature forests, but are not
detectable in litter from nearby clearcuts
(Widmer et al. 1997a). We will use the nt/i-f
primer to document the kinds and
molecular phylogenies of organisms
present in the litter and belowground that
contribute to N cycling in forests.
Acetylene reduction measurements will
estimate amounts of N fixed and provide
functional evidence to support the
molecular studies on nzJi-f biodiversity.
DNA fingerprint for N fixation
Information identifying the molecular
heterogeneity and phylogenies of N 2 -flxing
organisms will be achieved by polymerase
chain reaction amplification (PCR) of the
nt/H gene following extraction of bulk
DNA from litter and soil using methods
that we have developed previously
(Porteous et al. 1994, Porteous et a!. 1997,
Widmer et al. 1996, USEPA SOP 7.15m
Bulk Soil DNA). The nt/H gene codes for a
subunit of the dimeric nitrogenase enzyme
and its presence is associated with N 2 -
fixing activity. Primers for PCR
amplification have been synthesized and
function well in detecting all phylogenetic
groups of N 2 -fixing microorganisms. The
reaction conditions for amplification have
been described (Widmer et a!. 1997a).
Purified plasmid libraries of characteristic
nt/H PCR products from selected field
samples will be constructed and
transformed into Escherichia coli. Plasmid
DNA extracts from the E. co /i strains will
be characterized to validate R.FLP patterns
and representative clones for each pattern
will be subjected to sequence analysis by
the Center of Gene Research and
Biotechnology, Oregon State University.
Phylogenetic analyses of these sequences
will be conducted by alignment with those
retrieved from Gen Bank using the multiple
alignment routine of MacDNASIS pro
v3.2. Further details for cluster analysis
and construction of phylogenetic trees have
been described (Ueda et al. 1995, Widmer
etal. 1997a).
Associative N 2 - fixation through acetylene
reduction
To follow both spatially and temporally
the influence of seasons and litter quantity
and quality on the kinetics of N 2 -flxation
and to supplement molecular
characterization of nt/H populations
therein, we will measure mtrogenase
activity of various types of plant litter.
Standard techniques will be used to detect
in situ associative N 2 -flxmg activity based
on reduction of acetylene to ethylene
(Seidler et al. 1972, Bormann et al, 1993).
We will compare rates of acetylene
reduction/gm of litter or root for each field
plot of interest. Appropnate controls will
be included to measure any indigenous
ethylene production produced by plant or
microbial sources in our samples. Analyses
will compare seasons, litter quality and
chemistry, meteorological and elevation
conditions, and successional stage with the
molecular nt/i-f patterns and with the
kinetics of N 2 -flxation at the various sites.
A comprehensive assessment of the
physical, chemical, and biological
associations that impact N 2 -fixation in
forests will be derived through
comparisons with other data gathered in
this project.
70
-------
Use of molecular tools to evaluate
phylogeny and diversity of plant-associated
Pseudomonas and fungal species
In addition to structural genes such as n4fl1 ,
l6S nbosomal genes have been used by
molecular ecologists to study phylogeny
and to map the distribution and occurrence
of specific bacterial groups through in situ
analyses. We have developed new DNA
primer sequences specific for fluorescent
Pseudomonas species (Widmer et al.
199Th). Certain strains of this group are
rhizobacteria that may promote plant
growth, e. g., by inhibiting activities of
plant pathogenic fungi. We will use the
Pseudomonas primers to determine the
kinds and distributions of fluorescent
Pseudomonas species present in litter and
rhizosphere samples. In addition, molecular
methods for determining the presence,
diversity, and functionality of mycorrhizal,
saprophytic, and plant pathogenic fungi are
available that will be evaluated for their
utility to study changes in the litter and
rhizosphere of forests.
DNA fingerprint patterns (RFLPs) will be
generated from amplified 16S rRNA
sequences specific for fluorescent
Pseudomonas species using bulk DNA
obtained from lifter, soil, and rhizoplane
samples. Amplified 16s rDNA sequences
will be digested to generate DNA
fingerprints that are unique to the various
species of fluorescent pseudomonads
(Widmer et al. 199Th). DNA fingerprint
patterns from litter will be compared with
soil and rhtzosphere samples taken directly
below the litter. These data will provide
new information about the origins of
ecologically significant groups of
rhizosphere bacteria in forest ecosystems
and their response to management. Similar
tools will also be developed for fungi (see
next paragraph). These results will provide
some of the first molecular measurements
to document possible ecological linkages
between aboveground litter and forest
management strategies and belowground
plant root colonization by beneficial
bacteria. If the populations in litter and soil
are consistently different from those on the
roots, we will determine whether plant
rhizoplane populations are dictated by the
plant species, are influenced by
aboveground management, or are influenced
by other, unidentified processes.
We will also use molecular tools to generate
community DNA fingerprints that
represent flingal species present in
rhizoplane, soil, and litter samples. Pnmer
pairs will be identified, based on small
subunit (18S) nuclear rDNA sequences, to
obtain community fingerprints that can be
used to screen for changes in fungal
diversity that occur belowground in
forests. We will use the same field samples
as descnbed above as sources of DNA.
Using standard cloning and sequencing
methods as described above, sequence
information will be used to identify and
elucidate phylogenetic relationships (White
et al. 1990, Egger 1994, Edel Ct al. 1995,
Claasen et al. 1996, Perotto et al. 1996,
O’Donnell et al. 1997) and functional roles
(Kreuzmger et al. 1996, Howe et al. 1997,
Perotto et al. 1997) of both free-living and
mycorrhizal fungal species. DNA
fingerprints will also be generated from
isolated pure cultures, most likely
primarily of culturable Deuteromycete
species, which will be obtained from field
samples. This will allow us to ascertain and
document the ongins of culturable fungi
71
-------
present in cQmmunity fingerprint patterns
obtained from bulk DNA extracted from
environmental samples.
Community metabolic profiling of microbes
by the Biolog technique
The Biolog approach (Garland and Mills
1991), initially developed to identify pure
cultures of bacteria based on patterns of
utilization of 95 different C and N
substrates, has recently been successfully
used to characterize microbial communities
(Ellis et a!. 1995, Garland 1996a,1996b,
Grayston and Campbell 1996). The
technique has also been used to evaluate
the effects of land management, agricultural
practices, and natural disturbances on
microbial communities (Zak Ct a!. 1994,
Bossio and Scow 1995, Willig et al. 1996).
We will use the Biolog approach to
identify potential changes in metabolic
profiles or substrate utilization patterns
that may occur over time (1) under
different land management practices and (2)
with plant communities of different
successional stages in the Oregon Cascades.
The types of changes we hope to find are
in substrate utilization profiles of given
compounds or groups of compounds (e. g.,
sugars, sugar alcohols, ammo acids,
aromatics or xenobiotics), which are
represented on the Biolog plates.
Community-level metabolic fingerprints of
the rhizosphere microbial communities
(Garland, 1996) will be generated using
Biolog GN microplates (Biolog, Inc.,
Hayward, CA) (USEPA SOP 7.1 7m Biolog
Community Analysis). Three replicate
extracts will be prepared for each sample
type, incubated, then read with a
microplate reader (Molecular Devices, Inc.,
Sunnyvale, CA; Di Giovanni et al. 1997).
We will treat and process the rhizoplane
and rhizosphere samples for bacteria
according to the methods of Di Giovanni et
a!. (1997). Substrate utilization data will be
analyzed by principal component analysis
(PCA). Analysis of variance of PCA scores
will be used to determine whether
statistically significant differences occur
between samples.
Substrate induced respiration (SIR)
Ibiomass
The size and activity of the soil microbial
biomass needs to be measured to fully
understand nutrient fluxes and
biogeochemical transformations in natural
ecosystems (Horwath and Paul 1994). We
will measure substrate induced respiration
(SIR) on selected samples of litter and soil
at strategic sites associated with the other
samples being intensively analyzed as
described in this subtask (USEPA SOP
7.19m Substrate Induced Respiration).
Samples will be analyzed with and without
glucose additions using standard methods
(Anderson and Domsch 1978). We will
also conduct SIR analyses with selected
antibiotic inhibitors to measure the
separate contributions of bacteria and fungi
to the total respiration in unique types of
litters and soils (Anderson and Domsch
1975, Beare et al. 1990).
4. INTEGRATION WITH
OTHER WED PROJECTS
The Forest Ecosystem Indicators Project is
one of several in the Terrestrial Plant
Ecology Branch of WED, and its
coordination with other projects in the
Branch greatly enhances its potential for
success (Fig. 4-1). The indicator project’s
72
-------
initial emphasis is field research in the
Cascade forests, consequently, climatic,
edaphic and management practices are the
principle stressors for which data will be
collected. The ability to test models and
indicators against additional anthropogenic
stressors of interest to EPA will
significantly improve the utility to the
Agency of the resultant models and
indicators for use as both assessment and
predictions tools and increase the reliability
of indicators of condition that are
developed. Consequently, measures,
stress-response functions and models from
other research projects being conducted in
the Terrestrial Ecology Branch will be used
to strengthen the indicator project by
providing specific data on the response of
trees and forest ecosystems to known
exposures of specific stressors (Fig. 4-1).
To investigate how the measures and
models developed in the Forest Ecosystem
Indicators Project may respond to a range
of other environmental stressors (increased
temperature, elevated atmospheric CO 2
concentrations, and ozone), we will rely on
the integration with three other projects in
the branch (Stress response projects—Fig.
4-1). Many of the same measures and
models will be investigated by the same
scientists in other stress-response
experiments.
Stress responses
Natural conditions of
climate/soil gradients &
management actions
Effects of CO 2 and Climate Change on
Forest Trees (TERA I experiment) - The
project assessed the effects of elevated
CO 2 (ambient and ambient +200 j mol mol
and temperature (ambient and ambient +4
C) on a Douglas-fir (plant and soil) system
over a four-year period. The expcrimental
research was conducted in the Terrestrial
Ecophysiological Research Area (TERA)
which is a set of sun-lit controlled
environment chambers located in Corvallis.
TERA I
C02 sad temperature
TERMI
C02 and 03
N
7
Forest Indicator
project ,
Forest ozone
effects
DISPro
(Ia-house)
Figure 4 -1. Relation between the forest indicator projects and other research projects in the
Terrestrial Ecology Branch of WED.
73
-------
The trees were subjected to the typical
wet-dry seasonal soil moisture cycles and
relied on soil biological processing of litter
for nutrients The experiment was initiated
in June 1993 and was completed in August
1997. The effects of elevated CO 2 and
temperature were assessed on individual
rhizosphere and canopy processes as well
as developmg system budgets for C and N.
System carbon, nitrogen and water fluxes
were measured using the mass balance
approach -- by accounting for inputs and
losses from the terracosms, as well as
internal differentiation between plant and
litter/soil flux components. Nondestructive
measures were made of needle, shoot, and
bud development. Minirhizotron tubes and
a minirhizotron camera system were used
to monitor fine root dynamics and soil
cores were collected to follow changes in
fine root biomass. Soil and litter samples
were also collected to follow soil flora and
fauna. A key objective TERA I to collect
the necessary data to parametenze
TREGRO (an individual plant growth
model) and the MBL-General Ecosystem
Model [ GEM] (an ecosystem level
biogeochemical model).
interactive Effects of 03 and CO 2 on the
Ponderosa Pine Plant/Litter/Soil System
(TERA II experiment) - The project will
assess the interactive effects of
tropospheric ozone (03) and elevated
(C0 2 ) on critical indicators of ecosystem
function: carbon (C), nitrogen (N), and
water (H 2 0) cycling. The study will test
three hypotheses: (1) elevated 03 decreases
C, N and H 2 0 cycling rates; (2) elevated
CO 2 increases C, N, and decreases H 2 0
cycling rates; and (3) elevated CO 2
eliminates negative effects of 03 on C and
N cycling rates. The study was initiated in
the terracosms in the Spring of 1998 usmg
ponderosa pine (an 03 sensitive and CO 2
responsive species) seedlings growing in a
reconstructed ponderosa pine soil and litter
layer from the east side of the Oregon
Cascade Mountains. The research is an
integrated study with experimental and
modeling components. The experimental
research addresses: (1) system gas
exchange; (2) plant phenology, allometry
and carbon allocation; (3) litter and
soilJrhizosphere ecology; (4) litter and soil
chemical and physical properties; and (5)
system materials budgets, pools and fluxes
The modeling research will use the MBL-
GEM to evaluate effects on C and N
cycling, and TREGRO, to study the
potential impact of increased 03 and CO 2
on photosynthesis, respiration, C
accumulation, and C allocation. This
research will provide unique infonnation on
the responses of ecosystem functions
related to the interactions of 03 and CO 2 .
The research will also provide a
complementary set of measurements for
many of the parameters (plant, soil, and
litter) that will be taken in the field as part
of the forest indicators project. Because the
measurements in the terracosms will be
under clearly defined stress conditions they
will highlight the field indicators most
useful for detection of environmental
stress.
Effects of tropospheric ozone on forest trees
(Forest Ozone project) - The project is
developing a biological database to support
a secondary National Ambient Air Quality
Standard (NAAQS) for ozone, which
includes: (1) developing an understanding
of the nature and extent of ozone’s effect
on forest trees, and (2) developing a
meaningful index of ozone exposure that
74
-------
can be used to protect forests. The research
is conducted in the open-top chamber
facility located at Corvallis. The research
foci include: (1) seedling/small tree
response to ozone in open-top chambers,
(2) extrapolation of responses measured in
small trees to larger trees, and (3)
assessment of ozone exposure in forests
across the country and estimating the
effects of ozone on those forests. Two
important components of the experimental
studies on small trees are: (1) physiological
and growth responses to multiple stresses
(e.g.. ozone and nitrogen limitation), and
(2) characterizing below ground responses
of plants to ozone, including root
physiology and rhizosphere activity of
free-living organisms. Much of this
information is bemg used directly or
indirectly to inform simulations models, in
particular TREGRO, as a means of
projecting the responses seen in
expenmental studies across time and space.
The ability to provide stress-response
functions and incorporate specific stress
data into the models provides a unique
advantage that other groups developing
ecosystem indicator do not enjoy. This
integration of stress-response studies with
field studies of climatic and edaphic
stressors and management practices is a
unique strength of the forest indicator
project.
Regional Validation of MBL- (GEM)
General Ecosystem Model (Demonstration
Index Size Project -DISPro) Project
The primary objective is to develop a
parameterization of MBL-GEM that can
be used as a risk assessment tool for
ecosystems in the Olympic National Park
and the Pacific Northwest in general.
Specifically, we will use MBL-GEM to:
• Assess and predict future
responses of forest ecosystems to
natural and anthropogenic stressors,
including changes in temperature,
precipitation, cloudiness (light),
C0 2 , ozone, and N deposition
• Link changes in condition to likely
stressors
• Identify efficient and sensitive
indicators (early warning measures)
for loss of ecosystem integrity and
sustainability.
Models play a prominent role in ecological
risk assessments because they are the
primary means for relating stressors to
probable effects, and for making meaningful
extrapolations across scales of time, space,
and biological organization. Models are
particularly important for risk assessments
at the scale of ecosystems because it is
exceedingly difficult to experimentally
isolate the interactive effects of natural
environmental driving forces (temperature,
precipitation, cloudiness, etc.) and
anthropogenic stressors (e.g., air
pollutants, climate change, land use).
Process-based models that simulate
biogeochern.ical cycles or forest succession,
for xample, can help improve such
assessments by providing a self-consistent
synthesis of the results of many
experiments. The synthesis provided by
these models includes the interactions
among ecosystem processes that give rise
to the synergistic responses to multiple
factors. Under DISPro, we propose to use
75
-------
the Marine Biological Laboratory’s General
Ecosystem Model (MBL-GEM; Rastetter
et a! 1991) to assess and predict how
natural and arithropogenic stressors may
affect the health and sustainability of
ecosystems in the Olympic National Park.
The MBL-GEM is intended to be generally
applicable to most terrestrial ecosystems
and has been used in the past to analyze
the biogeochemical responses of temperate
deciduous forests, tropical evergreen
forests, and arctic tundra to changes in
atmospheric CO 2 concentration, N
deposition, temperature, irradiance
(cloudiness), and soil moisture (Rastetter et
al 1991, 1992, 1997; McKane et al. 1995,
1997a, 1997b). In addition, the EPA has
recently parameterized MBL-GEM for
forest ecosystems located along an
elevational gradient in the South Santiani
watershed in the western Cascades of
Oregon, as a first step toward the
development of a regionally robust
parametenzation for Pacific Northwest
forests (McKane and Tingey 1997) To
evaluate the regional applicability of MBL-
GEM to Pacific Northwest forests, we
require sites that are geographically and
ecologically distant from the sites being
used to parameterize the model, i.e., several
sites in Oregon including the South Santiam
watershed, Cascade Head and H.J.
Andrews Experimental Forest. The
Olympic National Park in western
Washington and the eastern Cascades of
Oregon are ideal sites for model evaluation
because they occur in areas that have very
different climates, soils, andlor vegetation
than the sites used to parameterize MBL-
GEM. The range of conditions provided by
these sites provide a good opportunity for
developing a regionally robust
biogeochemical model for the Pacific
Northwest. The data from the Olympic
National Park will provide an important
test of the reliability of the regional model
parametenzation as it contains a larger
range of conditions (e.g., precipitation,
growing season, elevation, etc.) than for
which the model was parameterized.
To accomplish the regional validation
objectives we propose to carry out two
primary activities under the DISPr0
Project:
• Collect biogeochemical and
meteorological data at four
relatively pnstine sites in the
Olympic National Park and at a site
on the east side of the Oregon
Cascades preliminary to submitting
a proposal to work in Kings
Canyon/Sequoia or Glacier National
Park(s) to evaluate the regional
applicability of MBL-GEM. The
data and data collection methods as
descnbed in Section 3.2.1 (Task 1),
will be used with this project.
• Make additional improvements to
MBL-GEM to better address risk
assessment issues of interest to the
EPA.
In addition to addressing DISPro goals, the
research activities proposed here will
contribute to a number of other nsk
assessment activities at the EPA, including
the Forest Ecosystem Indicators Program
and TER.A I and II. All of these projects
share the goal of making spatially explicit
predictions of how environmental stressors
will effect the health and sustainability of
forest ecosystems throughout the Pacific
76
-------
Northwest. A major product of the DISPro
and Forest indicators Program will be to
produce a well-validated, single parameter
set of MBL-GEM that can be applied with
confidence to all major forest types, soils
and climatic conditions within the Pacific
Northwest.
5. PROJECT MANAGEMENT
AND QUALITY ASSURANCE
STATEMENT
Management
This Project is an integrated effort with
contributions from many different
organizations and individuals from various
disciplines. Maj or responsibilities for
management of the Project within the
Terrestnal Plant Ecology Branch, of the
Western Ecology Division are given below:
Branch Chief
Ensures that research addresses needs
of EPA ORD Programs (EMAP,
Global Change, Tropospheric
Ozone); and themes of the Western
Ecology Division (rhizosphere
processes, extrapolation)
• Allocates funds and EPA FTEs to
Projects according to ORD Program
needs
• Ensures that all publications and
presentations meet the requirements
of the Laboratory and Agency.
Project Leader
• Represents and manages Project
within Branch
• Ensures that research addresses
project hypotheses and objectives.
• Peer review manuscripts, and
interacts with people within and
outside the Agency (e.g.,
International Geosphere-Biosphere
Programme, Global Change and
Terrestrial Ecosystems Program)
• Allocates funds according to Project
needs in terms of on-site contractors,
staff from EPA’s Senior
Environmental Employment Program
(SEE), post-docs (National Research
Council), undergraduate or graduate
students (National Network for
Environmental Management Studies,
NNEMS), other contracts or
agreements, and equipment and
supplies
• Ensures that
presentations
of the Project
• Facilitates scientific and
administrative responsibilities within
Project including regular staff
meetings. The meetings include
seminars and discussions where
participants will discuss critical
issues including the relationship
between the experimental and
modeling research such as model
assumptions and predictions.
all publications and
meet the requirements
77
-------
Principal Jhvestigators
Task Leaders
• Represent and manage Tasks within
Project
• Ensure that research addresses Task
objectives.
• Determine staff and fundmg needs
(EPA staff, contractors, SEEs,
NNEMS, Post-docs) and oversees
them within Task
Others
• Oversee specific task measurements,
statistics
• Other EPA scientists, on site
contractor Work Plan Manager, NRC
Postdocs
Project Scientirts
• Responsible for day-to-day
implementation of research plan.
Recommend equipment and supplies
needs, make direct measurements,
process and evaluate data, maintain
facilities and equipment, contribute to
and/or provide publications and
presentations
• EPA, contract staff, SEEs, NRC,
NNEMS, other affiliations
Admintctrative Assi tantc
• Record and distribute
Investigator meeting notes
Principal
• Facilitate preparation of publications
and presentations
• Maintain common Procite literature
database
Quality Assurance
All data collected as part of this Project
must meet EPA requirements regardmg
Quality Assurance (QA). This involves
preparation and approval of a Quality
Assurance Project Plan (QAPP),
Standard Operating Procedures (SOPs),
and other documentation according to the
QAPP, monitoring research activities to
make sure the QAPP is adhered to, and
reporting indicators of data quality to
management Scientists in this Project
have the responsibilities regarding QA
given below.
Branch Chief
• Ensures that all technical outputs
meet the QA requirements of the
Division and Agency.
Project Leader
• Is responsible for overall QA
performance and coordination of the
Project. This includes primary
responsibility for QAPP and
approval of specific SOPs. Works
with PIs to evaluate and maintain
quality of data in Project. Interacts
with Division QA staff in terms of
official audits.
• Facilitate processing of purchases
78
-------
Principal Investigators
Project Scientists
• Are responsible for carrying out
entire or portions of the Project
Research Tasks and insuring the
quality of the results generated from
the tasks. This includes contributing
to the QAPP and primary
responsibility for evaluating the
quality of data measured under
specific SOPs.
• Assist in writing SOPs and have
primary responsibilities for
implementing SOPs as they perform
measurements.
Branch Administrative Assistant
• Assist the Project Leader in managing
QA documents.
79
-------
6. REFERENCES
Abbas, J. D., B. A. D. Hetnck., and J. E.
Jurgenson. 1996. Isolate specific
detection of mycorrhizal fungi using
genome specific primer pairs. Mycologia
88(6)939-946.
Aber, J. D., and J. M. Melillo. 1991.
Terrestrial ecosystems. Saunders College
Publishing, Vhiladelphia, PA.
Aber, J.D., B.T. Botkin, and J.M. Melillo.
1979. Predicting the effects of different
harvesting regmies on productivity and
yield in northern hardwoods. Canadian
Journal of Forest Research 9:10-14.
Agren, G.I., M U.F. Kirschbaum, D.W.
Johnson and E. Bosatta. 1996
Ecosystem physiology: soil organic
matter. Pages x-x n A.!. Breymeyer,
D.O. Hall, J.M. Melillo and G.I. Agren
(eds.), Global Change: Effects on
Coniferous Forests and Grasslands. 1996
SCOPE, John Wiley and Sons.
Allen, S. E., H. M. Grimshaw, J
Parkinson, and C. Quarmby. 1974.
Chemical analysis of ecological materials.
Blackwell Scientific, Oxford.
Andersen, C. P. and P. T. Rygiewicz.
1991. Stress interactions and mycorrhizal
plant response: Understanding carbon
allocation priorities. Environmental
Pollution 73:217-244.
Andersen, C. P. and P. T. Rygiewicz.
1995. Allocation of carbon in
mycorrhizal Pinur ponderosa seedlings
exposed to ozone. The New Phytologist
131:471-480.
Andersen, C.P. and P.1. Rygiewicz. 1996
Understanding plant-soil relationships
using controlled environment facilities.
Advances in Space Research, in press.
Andersen, C.P. and C.F. Scagel. 1997.
Nutrient availability alters below-ground
respiration of ozone exposed ponderosa
pine. Tree Physiology 17:377-387.
Andersen, C.P., C.C. Lipp, and C.F.
Scagel. 1997a. The effects of ozone on
CO 2 flax from soils with ponderosa pine
and blue Wildrye in competition
Abstract, 1997 Annual Meeting,
Ecological Society of America,
Albuquerque, NM, August 11-14.
Andersen, C.P., R. Wilson, M. Plocher,
and W.E. Hogsett. 1997b. The carry-over
effects of ozone on root growth and
carbohydrate concentrations in
ponderosa pine. Tree Physiology 17:805-
811.
Andersen, C.P., W.E. Hogsett, R
Wessling, and M. Plocher. 1991. Ozone
decreases spring root growth and root
carbohydrate content in ponderosa pine
the year following exposure. Canadian
Journal of Forest Research 21:1288-
129 1.
Anderson, J. M. 1995. Soil Organisms as
Engineers: Microsite and Modulation of
Macroscale Processes. Pages 94-106 in
C. G. Jones, and J. H. Lawton (eds.),
Linking Species and Ecosystems.
Chapman and Ha1l, New York., NY.
80
-------
Anderson, J.P., and K.H. Domsch. 1975.
Measurement of bacterial and flingal
contributions to respiration of selected
agricultural and forest soils. Canadian
Journal of Microbiology 21 314-322.
Anderson, J.P., and K.H. Domsch. 1978. A
physiological method for the quantitative
measurement of microbial biomass in soil.
Soil Biology & Biochemistry 10:215-221.
Anderson, RV., Coleman, D.C., and C.V.
Cole. 1981. Effects of saprotrophic
grazing on net mineralization. Ecological
Bulletin 33 :201-216.
AOSA (Association of Official Seed
Analysis). 1981. Rules for testing seeds.
Journal of Seed Technology 6(2):l-125.
Arseneault, D. and S. Payette. 1997.
Reconstruction of millennial forest
dynamics from tree remains in a subarctic
tree line peatland. Ecology 78:1873-1883.
Asner, G. P., T. R. Seastedt, and A. R.
Townsend. 1997. The decoupling of
terrestrial carbon and nitrogen cycles.
Bioscience 47:226-234.
Baldock, J.A., J.M. Oades, A.G. Waters,
X. Peng, A.M. Vassallo, and MA.
Wilson. 1992. Aspects of the chemical
structure of soil organic matter materials
as revealed by solid-state ‘ 3 C NMR
spectroscopy. Biogeochemistry 16:1-42.
Beare, H.H., C.L. Neely, D.C. Coleman
and W.L. Hargrove. 1990. A substrate-
induced respiration (SIR) method for
measurement of fungal and bacterial
biomass on plant residues. Soil Biology
& Biochemistry 22:585-594.
Belcher, E.W. 1968. Use of soft X-ray in
tree seed testing and research. Pages 74-
96 in Proceedings of the Southeastern
Forest Radiography Workshop 1968. U.
S. Department of Agriculture, Forest
Service, Atlanta, GA.
Beven, K. 1989. Changing ideas in
hydrology -- the case of physically-based
models. Journal of Hydrology 105:157-
172.
Binkley, D. 1984. Ion exchange resin bags:
factors affecting estiatmnes of nitrogen
availability. Soil Science Society of
America Journal 48:1181-1184.
Binidey, D. and S. C. Hart. 1989. The
components of nitrogen availablity
assessments in forest soils. Advances in
Soil Science 10:57-112.
Binidey, D. and P. Matson. 1983. Ion
exchange resin bag method for assessing
forest soil nitrogen availability. Soil
Science Society of America Journal
47: 1050-1052
Binkley, D. J. Aber, J. Pastor, and K.
Nadelhoffer. 1986. Nitrogen availability
in some Wisconsin forests: comparisons
of resin bags and on-site incubations.
Biology and Fertility Soils 2:77-82.
Blanchard C.L., H. Michaels and S.
Tannenbaum. 1996 . Regional Estimates
of Acid Deposition Fluxes in California
for 1985-1994. California Environmental
Protection Agency, Air Resources Board,
Sacramento CA, Contract No. 93-332.
Bledsoe, C. S., and P. T. Rygiewicz. 1986.
Ectomycorrhizas affect ionic balance
during ammomun, uptake by Douglas-fir
81
-------
roots. The New Phytologist 102.271-
283.
Bochner, B. 1989. “Breathprints” at the
microbial level. ASM News 55(10):536-
539.
Bonan, GB. and L. Sirois. 1992. Air
temperature, tree growth, and the
northern and southern range limits to
Picea rnariana. Journal of Vegetation
Science 3:495-506.
Bonnann, F.H. and G.E. Likens. 1979.
•Pattern and Process in a Forested
Ecosystem. Springer-Verlag, New York
Bormann, T. B., F. H. Bormann, W. B
Bowden, R. S. Peirce, S. P. Hamburg, D.
Wang, M C. Snyder, C. Y. Li, and R. C.
Ingersoll. 1993. Rapid N 2 fixation in
pines, alder, and locust: evidence from
the sandbox ecosystem study. Ecological
Society of America. 74:583-598.
Borror, D.J., C.A. Triplehorn and N.F.
Johnson. 1992. An Introduction to the
Study of Insects, 6th edn. Harcourt Brace
College Publishers, New York, NY, 875
pp
Bossio, D. A., and K. M. Scow. 1995.
Impact of carbon and flooding on the
metabolic diversity of microbial
communities in soils. Applied and
Environmental Microbiology 1(1 1):4043-
4050.
Botkin, D.B., J.F. Janak, and J. R. Wallis.
1972. Some ecological consequences of a
computer model of forest growth. Journal
of Ecology 60:849-872.
Bowen, 0. D. 1980. Contemporary
Microbial Ecology In D. C. Ellwood, J.
W. Hedger, M. J. Latham, J. M. Lynch,
and J. H. Slater (eds.), Contemporary
Microbial Ecology. Academic Press, New
York.
Bowen, G. D., and A. D. Rovira. 1991.
The rhizosphere, the hidden half of the
hidden hali Pages 641-649 in Y. Waisel,
and U. Kafkafi (eds.), Plant Roots-The
Hidden Half. Marcel Dekker, New York.
Brouwer, R. 1962. Nutritive influences on
the distribution of dry matter in the
plant. Netherlands Journal Agricultural
Science 10:399-408.
Brouwer, R. 1983. Functional equilibrium
sense or nonsense? Netherlands Journal
Agricultural Science 31:335-348.
Brown D. A. and D. R. Upchurch. 1987.
Minirhizotrons: A summary of methods
and mstruments in current use. Pages 15-
20 rn H. M. Taylor (ed.), Minirhizotron
Observation Tubes: Methods and
• Applications for Measuring Rhizosphere
Dynamics. American Society of
Agronomy Special Publication 50.
American Society of Agronomy, Crop
Science Society of America and Soil
Science Society of America, Madison,
q.
Brown 1 S. and G. Gaston. 1995. Use of
forest inventories and geographic
information systems to estimate biomass
density of tropical forests: application to
tropical Africa. Environmental
Monitoring and Assessment 38 (2-
3): 157-168.
82
-------
Brown, S., L., Iverson, and A. E. Lugo.
1994. Land use and biomass changes of
forests in Peninsular Malaysia during
1972-82: use of GIS analysis. Chapter 4
in V. H. Dale (ed.), Effects of land use
change on atmospheric CO 2
concentrations: Southeast Asia as a case
study. Springer Verlag, NY.
Brown, S., L. R. Iverson, A. Prasad, and D.
Liu. 1993. Geographic distribution of
carbon in biomass and soils of tropical
Asian forests. Geocarto International
8(4):45-59.
Bryant, J.P., F.S. Chapin, Ill,, and DR.
Klein. 1983. Carbon /nutrient balance of
boreal plants in relation to vertebrate
herbivory. Oikos 40: 357-368.
Bugrnann, H. 1994. On the ecology of
mountamous forests in a changing
climate: A simulation study. Ph.D.
Thesis no. 10638, Swiss Federal Institute
of Technology Zunch, Switzerland.
Bugmann, H. K. M. and A. M. Solomon.
1995. The use of a European forest
model in North America: A study of
ecosystem response to climate gradients.
Journal of Biogeography 22:477-484.
Bugrnann, I-i. K. M. and A. M. Solomon,
A.M. 1997. Toward a unified gap model
for global temperate forests. Ms.
submitted.
Burns, R. M. and B .H. Honkala. 1990.
Silvics of North America. Volume 1,
Conifers. Agncultural Handbook 654. U.
S. Department of Agriculture, Forest
Service, Washington, DC.
Burton, P. J. and S. G. Cumming. 1995.
Potential effects of climatic change on
some western Canadian forests, based on
phenological enhancements to a patch
model of forest succession. Water, Air
and Soil Pollution 82:401-414.
Cainbardella, C. A. and E. T. Elliott. 1992.
Particulate soil organic-matter changes
across a grassland cultivation sequence.
Soil Science Society of America Journal
56:777-783.
Cambardella, C. A. and E. T. Elliott. 1994.
Carbon and nitrogen dynamics of soil
organic matter fractions from cultivated
grassland soils. Soil Science Society of
America Journal 58:123-130.
Campbell, S. and L. Liegel. 1996.
Disturbance and forest health in Oregon
and Washington. General Technical
Report PNW-GTR-381, USDA Forest
Service, Pacific Northwest Research
Station, Portland, OR, 105 pp.
Catricala, C.E., K.M. Newkirk, and J.M.
Melillo. 1995. Effects of temperature on
soil respiration and N-availability for
arctic and temperature zone soils.
Agronomy Abstracts 315, American
Society of Agronomy, Madison, WI.
Chen, J., iF. Franklin, and l.A. Spies.
1995. Growing-season microclimatic
gradients from clear-cut edges into old-
growth Douglas-fir forests. Ecological
Applications 5:74-86.
Ciais, P., P. P. Tans, J. W. C. White, M.
Trolier, R. J. Francey, J. A. Berry, D. R.
Randall, P. J. Sellers, J. G. Collatz, and
D. S. Schimel. 1995. Partitioning of ocean
and land uptake of CO 2 as inferred by
83
-------
8’ 3 C measurements from
Climate Monitormg and
Laboratory Global Air
Network. Journal of
Research 1 00(D3):505 1-5070.
Claassen, V. P., R. 3. Zasoski and B. M.
Tyler. 1996. A method for direct soil
extraction and PCR amplification of
endomycorrhiza l fungal DNA.
Mycorrhiza 6:447-450.
Cohen, W. B., M. E. Harmon, D. 0.
Wallina and M Fiorella. 1996. Two
decades of carbon flux from forests of the
Pacific Northwest. BioScience 46(11):
836-844
Cohen, W. B , T. A. Spies and M. Fiorella.
1995. Estimating the age and structure of
forests in a multi-ownership landscape of
western Oregon, U.S.A. International
Journal of Remote Sensing 16:721-746.
Coley, P. D., J. P. Bryant, and F. S
Chapin III 1985. Resource availability
and plant antiherbivore defense. Science
230:895-899
Cook, E. R. and 3. Cole. 1991. On
predicting the response of forests in
eastern North America to future climatic
change. Climatic Change 19:71-282.
Cook, E. R. and L. A. Kairiukstis. 1990.
Methods of dendrochronolgy:
Applications in the environmental
sciences. Kiuwer Academic Publishers,
Dordrecht, the Netherlands.
Cosby, B.J., G.M. Hornberger, and J.N.
Galloway. 1985. Modeling the effects of
acid deposition: assessment of a lumped
parameter model of soil water and
streamwater chemistry. Water Resources
Research 21:51-63.
Craig, H. and L. Gordon. 1965. Deuterium
and oxygen-18 variation in the ocean and
marine atmosphere. Pages 9-30 in
Proceedings of the Conference on Stable
Isotopes in Oceanographic Studies and
Paleotemperature. Laboratory of
Geology and National Science. Pisa,
Italy.
Cramer, W. P. and A. M. Solomon. 1993.
Climatic classification and future
distribution of global agricultural land.
Climate Research 3:97-110.
Curl, E. A., and B. Truelove. 1986. The
Rhizosphere. Springer-Verlag, New York.
Currie, W. S., J. D. Aber, W. H
McDowell, R. D. Boone, and A. H.
Magill. 1996. Vertical transport of
dissolved organic C and N under long-
term N amendments in pine and
hardwood forests. Biogeochemistry
35:471-505.
D’Elia, C.F., P.A. Steudler and N. Corwin.
1977. Determination of total nitrogen in
aqueous samples using . persulfate
digestion. Limnology Oceanography
22:760-764.
Dale, V.H. and M. Hemstrøm. 1984.
CLIMACS: A computer model of forest
stand development for western Oregon
and Washington. Res. Paper PNW-327,
Pacific Forest and Range Experiment
Station, U.S. Department of Agriculture,
Forest Service, 60 pp.
the NOAA
Diagnostics
Sampling
Geophysical
84
-------
Daley, C., R. P. Neilson, and D. L.
Phillips. 1994. A statistical-topographic
model for mapping climatologwal
precipitation over mountainous terrain.
Journal of Applied Meteorology 33:140-
158.
de Bruin, F. J. 1992. Use of repetitive
(repetitive extragenic palindromic and
enterobacterial repetitive intergenenc
consensus) sequences and the
polymerase chain reaction to fingerprint
the genomes of Rhizobium meliloti
isolates and other soil bacteria. Applied
and Environmental Microbiology
58(7):2 180-2187.
Di Giovanni, G. D., L. S. Watrud, R. J.
Seidler, and F. Widmer. 1997.
Characterization of microbial populations
by metabolic and molecular
fingerprinting. Submitted.
Dilly, 0., and J.-C. Munch. 1996.
Microbial biomass content, basal
respiration and enzyme activities during
the course of decomposition of leaf litter
in a black alder (Alnu.s glutinosa (I))
Gaertn.) forest. Soil Biology and
Biochemistry 28(8): 1073-1081.
Dodson, R. and D. Marks. 1997. Daily air
temperature interpolated at high spatial
resolution over a large mountainous
region. Climate Research. 8:1-20
Dommergues, Y. R., and S. V. Krupa. 1978
Interactions Between Non pathogenic
Soil Microorganisms and Plants. Elsevier
Publishing Co, Amsterdam.
Domsch, K. H., W. Gains, and T-H.
Anderson. 1980. Compediurn of soil
fungi. Academic Press, London.
Donegan, K. K., R. J. Seidler, V.J. Fieland,
D. L. Schaller, C. J. Palm, L. M. Ganio,
D.M. Cardwell, and Y. Stemberger. 1997.
Decomposition of genetically engineered
tobacco under field conditions:
Persistence of the proteinase inhibitor I
product and effects on soil microbial
respiration and protozoa, nematode, and
microarthropod populations Journal
Applied Ecology 34:767-777
Donegan, K. K., C. J. Palm, V. J. Fieland,
L. A. Porteous, L. M. Ganio, D. L.
Schaller, L. Q. Bucao, and R. J. Seidler.
1995. Changes in levels, species and
DNA fingerprints of soil microorganisms
associated with cotton expressing the
Bacillus thuringiensis var. kurstaki
endotoxin. Applied Soil Ecology 2:111-
124
DuBois, M., K. A. Giles, J. K. Harrington,
P. A. Rebers, and F. Smith. 1956.
Colorimetric method for determination of
sugars and related substances. Analytical
Chemistry 28:350-356.
Edel, V., C. Steinberg, I. Avelange, G.
Laguere, and C. Alabouvette. 1995
Comparison of three molecular methods
for the characterization of Fusarium
oxysporum strains. Phytopathology
85:579-585.
Egger, K.N. 1994.
ectomycorrhizal
Canadian Journal
S 1422.
Ehieringer, J.R., C.B. Field, Z.F. Lin, and
C.Y. Kuo. 1986. Leaf carbon isotope and
mmeral composition in subtropical plants
along an irradience dine. Oecologia
70:520-526.
Molecular analysis of
fungal communities.
of Botany 73:Sl415-
85
-------
Ellis, R. J., I. P. Thompson, and M. J.
Bailey. 1995. Metabolic profiling as a
means of characterizing plant-associated
microbial communities. FEMS
Microbiology Ecology 16(1):9-17.
Enting, I.G., and J.V. Mansbndge. 1989.
Seasonal sources and sinks of
atmospheric C0 2 : Direct inversion of
filtered data. Tellus 41B:1 11-126.
Escuedero et al 1992
Farrar, J.F. 1992. Beyond photosynthesis
the translocation and respiration of
diseased leaves. IN P.O. Ayres, ed.,
Pests and Pathogens: plant responses to
foliar attack Bios Scientific Publishers,
Oxford.
Farquhar, G.D., J Lloyd, J.A. Taylor, L.B.
Flanagan, J.P. Syvertsen, K.T. Hubrick,
S C. Wong, and J.R. Ehlennger. 1993.
Vegetation effects on isotope
composition of oxygen in atmospheric
CO 2 Nature 363 439-443.
Finlay, R. D. 1985. Interactions between
soil micro-arthropods and
endomycorrhizal associations of higher
plants. Pages 319-331 in A. H. Fitter, D.
Atkinson, D .J. Read, and M. B. Usher
(eds.), Ecological interactions in soil.
Blackwell Scientific Publications, Oxford.
Fogel, R., and G. Hunt. 1983. Contribution
of mycorrhizae and soil fungi to nutrient
cycling in a Douglas-fir ecosystem.
Canadian Journal of Forest Research
13:219-232.
Forman. R.T.T. and M. Godron. 1986.
Landscape Ecology. John Wiley & Sons,
New York.
Franklin, J. F., K. Cromack, W. Denison,
A. McKee, C. Maser, J. Sedell, F.
Swanson and G. Juday. 1981. Ecological
characteristics of old-growth Douglas-fir
forests. United States Department of
Agriculture, Forest Service, Pacific
Northwest Forest and Range Expenment
Station Geneneral Technical Report
PNW-118. 48p.
Franklin, J.F. and C.T. Dyrness. 1988.
Page 452 n Natural Vegetation of Oregon
and Washington. Oregon State
University, Corvallis, OR
Fujimori, T., S. Kawanabe, H. Saito, C.C.
Grier, and T. Shidei. 1976. Biomass and
primary production in forest of three
major vegetation zones of the
northwestern United States. Journal of
the Japanese Forestry Society 58:360-
373.
Garland, J. L. 1996. Patterns of potential C
source utilization by rhizosphere
communities. Soil Biology and
Biochemistry 28:223-230.
Garland, J. L., and A. L. Mills. 1991.
Classification and characterization of
heterotrophic microbial communities on
the basis of patterns of community-level
sole-carbon-source-utilization. Applied
and Environmental Microbiology
57:2351-2359.
Garland, J. L., and A. L. Mills. 1994. A
community-level physiological approach
for studying microbial communities.
Pages 77-83 in K. Ritz, I Dighton, and
K. Oilier (eds.), Beyond the biomass:
compositional and functional analysis of
soil microbial, communities. John Wiley
& Sons Ltd, Chichester, UK.
86
-------
Gholz, H.L., S.A. Vogel, W.P. Cropper,
Jr., K. McKelvey, and K.C. Ewel. 1991.
Dynamics of canopy structure and light
interception in Pinus e!lioztii stands,
North Florida. Ecological Monographs
6:33-5 1.
Giblin, A.E., J. Laundre, K. Nadeihoffer,
G. Shaver. 1994. Measuring nutrient
availability in arctic soils using ion-
exchange resins: a field test. Soil Science
Society of America Journal 58:1154-1162
Gill, R.S., and D.P. Lavender. 1983. Urea
fertilization and foliar nutrient
composition of western hemlock (Tsuga
heterophylla, (Raf.) Sarc.). Forest
Ecology and Management 6:333-34 1.
Grayston, S. J., and C. D. Campbell. 1996.
Functional biodiversity of microbial
communities in the rhizospheres of
hybnd larch (Larix eurolepis) and Sitka
spruce (Picea sitchensis). Tree
Physiology 16:1031-1038.
Gner, C.C., and ItS. Logan. 1977. Old-
growth Pseudotsuga menz:esiz
communities of a western Oregon
watershed: biomass distribution and
production budgets. Ecological
Monographs 47:373-400.
Hagerman, A.E. 1987. Radial diffusion
method for determining tannin in plant
extracts. Journal of Chemical Ecology
13:437-449.
Harborne, J.B. 1993. Ecological
Biochemistry. Academic Press. New
York, NY.
Harcombe, P. A. 1986. Stand development
in a 130-year-old spruce-hemlock forest
based on age structure and 50 years of
mortality data. Forest Ecology and
Management 14:41-58.
Harcombe, P. A. 1987. Tree life tables.
Bioscience 37:557-568.
Harley, J. L., and S. E. Smith. 1983.
Mycorrhizal Symbiosis. Academic Press,
London.
Harmon, M. E., J. F. Franklin, F. J.
Swanson, P. Sollms, S. V. Gregory, J. D.
Lattin, N. H. Anderson, S. P. CIme, N.
G. Aumen, J. R. Sedell, G. W.
Lienkaemper, K. Cromack and K. W.
Cummins. 1986. Ecology of coarse
woody debris in temperate ecosystems.
Advances in Ecological Research 15:133-
302.
Harmon, M.E. and J. Sexton. 1996.
Guidelines for m easurements of Woody
Debris in Forest Ecosystems. Publication
No. 20. U.S. LTER Network
Office:University of Washington, Seattle,
WA, USA. 73 pp.
Harper, J.L. 1977. Population biology of
plants. Academic Press, London, UK.
Hart, S.C., J.M. Stark, E.A. Davidson and
M.K. Firestone. 1994. Nitrogen
mineralization, immobilization, and
mthfication. Pages 985-1018 in Methods
of Soil Analysis, Part 2, Microbiological
and Biochemical Properties--SSSA Book
Series, no. 5. Soil Science Society of
America, Madison, WI.
Hemstrom, M. and Adams, V. D. 1982.
Modeling long-term forest succession in
the Pacific Northwest. Pages 14-23 in J.
E. Means (ed.), Forest Succession and
87
-------
Stand Development Research in the
Northwest Forest Research Laboratory,
Oregon State University, Corvallis OR
Pollutants: Proceedmgs
International Conference.
Applied Science, New York.
from an
Elsevier
Hendrick R. L. and K. S. Pregitzer. 1992.
The demography of fine roots in a
northern hardwood forest. Ecology
73:1094-1104
Hendrick R. L. and K. S. Pregitzer. 1993.
Patterns of fine root mortality in two
sugar maple forests Nature 361:59-61.
Hendrick R. L. and K. S. Pregitzer. 1996.
Apphcations of minirhizotrons to
understand root function in forests and
other natural ecosystems. Plant and Soil
185:293-304
Hefty, J. D. and J M. A. Swan. 1974.
Reconstructing forest history from live
and dead plant matenal-An approach to
the study of forest succession m
southwest New Hampshire. Ecology
55:772-783.
Hesterberg, R. and U Siegenthaler. 1991
Production and stable isotopic
composition of CO 2 in a soil near Bern,
Switzerland. Tellus 43B:197-205.
Hogsett, W.E., M.C. Plocher, V. Wildman,
D.T. Tmgey and J.P. Bennett. 1985.
Growth response of two varieties of
slash pme to chronic ozone exposure.
Canadian Journal of Botany 63:2369-
2376.
Hogsett, W.E., D.T. Tingey and E. H. Lee.
1988. Ozone exposure indices: Concepts
for development and evaluation of their
use. Pages 107-138 n W.W. Heck, O.C.
Taylor and D.T. Tingey (eds.),
Assessment of Crop Loss from Air
Homer, J. D., J. R. Gosz, and R. M. Cates.
1988. The role of carbon-based
secondary metabolites in decomposition
in tenestrial ecosystems. The American
Naturalist 132:869-883.
Horwath, W. R. and E. A. Paul. 1994.
Microbial biomass. Pages 753-773 n J.
M. Bigham (ed.), Methods of soil
analysis, Part 2 Microbiological and
biochemical properties. Soil Science
Society of Amenca, Madison, WI.
Howe, R., R. L. Evans, and S. W.
Ketteridge. 1997. Copper-binding
proteins in ectomycorrhizal fungi. New
Phytology 135:123-131.
Hughes, J. W., and T. J. Fahey. 1994.
Litterfall dynamics and ecosystem
recovery during forest development.
Forest Ecology and Management 63:181 -
198.
Hunsaker, C. T. and D. E. Carpenter (eds.).
1990. Ecological indicators for the
Environmental Monitoring and
Assessment Program. EPA 600/3-90/-6-.
UE EPA, ORD, Research Triangle Park,
NC.
Hunt, H.W., ER. Ingham, D.C. Coleman,
E.T. Elliott, and C.P.P. Reid. 1988.
Nitrogen limitation of production and
decomposition in a prairie, mountain
meadow, and pine forest. Ecology
69:1009-1016.
88
-------
Jarvis, P.G. 1993. Prospects for bottom-up
models. Pages 115-126 in JR. Ehieringer
and C.B. Field (eds.), Scaling
Physiological Processes: Leaf to Globe.
Academic Press, Inc., San Diego.
Jarvis, P.G. and J.W. Leverenz. 1983.
Productivity of temperature, deciduous
and envergreen forests. Pages 234-280 in
O.L. Lange, P.S. Nobel, C.B. Osmond
and H. Ziegler (eds.), Physiological Plant
Ecology IV. Ecosystem Processes:
Mineral Cycling, Productivity and Man’s
Influence Encyclopedia of Plant
Physiology I 2D. Springer-Verlag, Berlin,,
Germany.
Johnson M G., D.T. Tingey, M.J. Storm,
L.M. Ganio, and D.L. Phillips. 1997.
Effects of elevated CO 2 and N
fertilization on the lifespan of Pinus
ponderosa fine rootspages 370-373. in
H.E. Flores, Lynch, J.P., Eissenstat, D.
(eds.) Radical Biology: Advances and
Perspectives on the function of Plant
Roots
Johnson, A.H. 1989. Decline of red spruce
in the Northern Appalachians:
Determining if air pollution is an
important factor. Pages 91-104 in
Biologic Markers of Air Pollution Stress
and Damage in Forests. National
Academy Press, Washington, DC.
Jones, D.A. 1988. Cyanogenesis in animal-
plant interactions. fl44 Everd, D and J.B.
Harborne, eds. Cyanide copounds in
biology. John Wiley, Chichester.
Julkunen-Titto, R. 1985. Phenolic
constituents in leaves of nothem willows:
Methods for the analysis of certam
phenolics. J. Agric. Food Chem. 33:213-
217.
Keeland, B.D. and R.R. Sharitz. 1993.
Accuracy of tree growth measurements
using dendrometer bands. Canadian
Journal of Forest Research 24:2454-
2457.
Keeler, R.F. 1975. Toxins and teratogens of
higher plants. Lloydia 3 8:56-86.
Keeling. C, W. Mook, and P. Tans. 1979.
Recent trends in i 3 C/l 2 C ratio of
atmospheric carbon dioxide. Nature
277:21-123.
Keller, 1. 1988. Growth and premature leaf
fall in American aspen as biomdications
for ozone. Environmental Pollution
52:183-192.
Kercher, J.R. and M.C. Axelrod. 1984. A
process model of fire ecology and
succession in a nuxed-conifer forest.
Ecology 65:1725-1742.
Kienast, F. 1987. FORECE - A forest
succession model for southern central
Europe. Oak Ridge National Laboratory,
Oak Ridge, Tennessee, ORNL/TM-
10575, 69 pp.
Killingbeck, K. T., J. D. May and S.
Nyman. 1990. Foliar senescence in an
aspen (Populus tremuloides) clone: the
response of element resorption to
interramet variation and timing of
abscission. Canadian Journal of Forest
Research 20: 1156-1164.
89
-------
Killmgbeck, ‘K.T. 1996. Nutrients in
senesced leaves: keys to the search for
potential resorption and resorption
proficiency. Ecology: 1716-1727.
Kirschbaum, M.U.F. and A. Fischlm.
1996. Climate change Impacts on forests.
Pages 95-129 in R.T. Watson, M. C.
Zinyowera, and R. H. Moss (eds.),
Climate Change 1995: Impacts,
Adaptations and Mitigation of Climate
Change. Cambndge University Press,
Cambridge UK.
Kbmer, C. 1994. Biomass fractionation in
plants: A recommendation of definitions
based on plant function. Pages 173-185
in J. Roy and E. Gamier (eds.), A Whole
Plant Perspective on Carbon-Nitrogen
Interactions SPB Academic Publishing,
The Hague, The Netherlands.
Körner, C. and J.A. Arnone III. 1992.
Responses to elevated carbon dioxide in
artificial tropical ecosystems. Science
257 1672-1675
Kreuzinger, N., R. Podeu, F. Gruber, F.
Göbl, and C. P. Kubicek. 1996.
Identification of some ectomycorrhizat
basidiomycetes by PCR amplification of
their gpd (Glyceraldehyde-3-phosphate
dehydrogenase) genes. Applied and
Environmental Microbiology 62:3432-
3438.
Krugman, S.L., W.I. Stein and D.M.
Schmitt. 1974. Seed biology. Pages 5-40
n U.S. Department of Agriculture Forest
Service. Seeds of woody plants in the
United States. Agricultural handbook No.
450. USDA FS, Washington, DC, 883
pp.
Lanfranco, L., P. Wyss, C. Marzachi, and
P. Bonfante. 1995. Generation of RAPD-
PCR primers for the identification of
isolates of Glomus mosseae, an
arbuscular mycorrhizal fungus Molecular
Ecology 4:61-68.
Larink, 0. 1997. Springtails and Mites:
Important Knots in the Food Web of
Soil. Pages 225-264 in Fauna in Soil
Ecosystems - Recycling Processes,
Nutrient Fluxes, and Agricultural
Production. Marcel Dekker, Inc. New
York, NY.
Lassoie, J.R. 1973. Diurnal dimensional
fluctuations m a Douglas-fir stem in
response to tree water status. Forest
Science 19:251-255.
Lawler, I. R., W. J. Foley, I. E. Woodrow,
and S. J. Cork. 1997. The effects of
elevated CO 2 atmospheres on the
nutritional quality of Eucalyptus foliage
and its interaction with soil nutrient and
light availability. Oecologia 109:59-68.
LeBlanc, D. C., D. J. Raynal, E. H. White
and E. I-I. Ketchledge. 1987.
Characterization of historical growth
patterns in declining red spruce trees. pp.
360-371 IN Jacoby, G. C. and J.
Hornbeck, eds., Proceedings of the
International Symposium on Ecological
Aspects of Tree-Ring Analysis. CONF-
8608144. National Technical Information
Service. Springfield VA.
Legard, H.A., and L.C. Meyer. 1973.
Willamette National Forest Soil Resource
Inventory, USDA Forest Service, Pacific
Northwest Region, Willamette National
Forest Supervisor’s Office, Eugene, OR.
90
-------
Lewis, T. E. and B. L. Coniding (eds.). n.d.
Forest Health Monitoring. Indicators of
forest health: mdicator development
research. EMAP Center, EPA/ORD,
Washington, DC.
Lieth, H., (ed.). 1974. Phenology and
Seasonality Modeling. Ecological Studies
Vol. 8. Springer-Verlag, New York. 444p.
Likens, G.E., F.H. Bormann, and N.M.
Johnson. 1969. Nitrification: importance
to nutrient losses from a cutover forested
ecosystem. Science 163:1205-1206.
Likens, G.E., F.H. Bormann, N.M.
Johnson, W.W. Fisher, and R.S. Pierce.
1970. Effects of forest cutting and
herbicide treatment on nutrient budgets in
the Hubbard Brook ecosystem m New
Hampshire. Ecological Monographs
40:23-47.
Li ii, 0., J. R. Ehieringer, P. T. Rygiewicz,
M. 0. Johnson, and D. T. Tingey. 1998.
Effects of elevated CO 2 and temperature
on different components of soil CO 2
efflux in Douglas-fir terracosms. Global
Change Biology, in press.
Lindroth, R. L., K. K. Kinney, and C. L.
Platz. 1993. Responses of deciduous
trees to elevated atmosphenc C0 2 :
productivity, phytochemistry, and insect
performance. Ecology 74:763-777.
Loehie, C. and D. LeBlanc. 1996.
based assessments of climate
effects on forests: a critical
Ecological Modelling 90:1-31.
saccha rum). Bulletin of the Torrey
Botanical Club 111:193-199.
Lou, Y., C.B. Field and H.A. Mooney.
1994. Predicting responses of
photosynthesis and root fraction to
elevated [ C0 2 ]: interactions among
carbon, nitrogen and growth. Plant Cell
and Environment. 17:1195-1204.
Lussenhop, J. 1992. Mechanisms of
microarthropod-microbial interactions in
soil. Pages 1-33 M. Begon, and A. H.
Fitter (eds.), Advances in Ecological
Research. Academic Press, San Diego.
Lynch, J. M., and J. M. Whipps. 1991.
Substrate flow in the rhizosphere. Pages
15-24 in D. L. Keister, and P. B. Cregan
(eds.), The rhizosphere and plant growth.
Kuuwer Academic Publisher, Dordrecht.
Majdi, H. 1996. Root sampling methods -
applications and limitations on the
minrhizotron technique. Plant and Soil
185 :255-258.
Marks, D. and J. Dozier. 1992. Climate
and energy exchange at the snow surface
in the alpine region of the Sierra Nevada:
2. Snow cover energy balance. Water
Resources Research 28.3043-3054.
Matyssek, R., M.S. Gunthardt-Goerg, M.
Saurer, and T. Keller. 1992. Seasonal
growth, ‘ 3 C in leaves and stem, and
phloem structure of birch (Betula
pendula) under low ozone
concentrations. Trees 6:69-76.
McCool, P. M., and J. A. Menge. 1983.
Influence of ozone on carbon partitioning
in tomato: Potential role of carbon flow
in regulation of the mycorrhizal
Model-
change
review.
Lorimer, C. G
simulation
distributions
and
of
of
L. E. Frelich. 1984. A
equilibrium diameter
sugar maple (Acer
91
-------
symbiosis tunder conditions of stress.
The New Phytologist 94:241-247.
McGarigal, K and B.J. Marks. 1995.
FRAGSTATS: spatial pattern analysis
program for quantifying landscape
structure. General Technical Report
PNW-GTR-35 1. USDA Forest Service,
Pacific Northwest Research Station,
Portland, OR.
McKane, R.B., E.B. Rastetter, G.R.
Shaver, K.J. Nadeihoffer, A.E. Giblin,
J.A. Laundre, F.S. Chapin Ill. 1997a.
Climatic effects on tundra carbon storage
inferred from experimental data and a
model. Ecology 781 170-1187.
McKane, R.B., E.B. Rastetter, G.R.
Shaver, K.J. Nadeihoffer, A.E. Gibim,
J.A. Laundre, F.S. Chapin III. 1997b.
Reconstruction and analysis of historical
changes in carbon storage in arctic tundra.
Ecology 78:1188-1198.
McKane, R.B. and D.T. Tingey. 1997.
Analysis of climatic and edaphic controls
on carbon storage in Douglas-fir forest
ecosystems. Bulletin of the Ecological
Society of Amenca 78(4):144.
McKane, R.B., E.B. Rastetter, J.M.
Melillo, G.R. Shaver, C.S. Hopkinson,
and D.N. Fernandes. 1995. Effects of
global change on carbon storage in
tropical forests of South America. Global
Biogeochemical Cycles 9:329-350.
McQuaker, N.R., P.D. Kiuckner and G.N.
Chang. 1979. Calibration of an
inductively coupled plasma-atomic
emission spectrometer for the analysis of
environmental materials. Analytical
Chemistry 5 1:888-895.
Means, J.E., H.A. Hansen, G.J. Koerper,
P.B. Alaback, and M.W. Klopsch. 1994.
Software for computing plant biomass--
BIOPAK users guide. General Technical
Report PNW-GTR-340. Portland, OR:
U.S. Dept. of Agnculture, Forest Service,
Pacific Northwest Research Station. 184
pp.
Memil, S.D and D. R. Upchurch. 1994.
Converting root numbers observed at
mmirhizotrons to equivalent root length
density. Soil Society of America Journal
58:1061- 1067.
Millard, P. 1996. Ecophysiology of the
internal cycling of nitrogen for tree
growth. Journal Plant Nutrition and Soil
Science 159: 1-10.
Miller, H. J., 0. Henken, and J. A. van
Veen. 1989. Variation and composition of
bacterial populations in the rhizospheres
of maize, wheat, and grass cultivars.
Canadian Journal of Microbiology
35:656-660.
Miller, R. M., and J. D. Jastrow. 1992.
The application of VA mycorrhizae to
ecosystem restoration and reclamation.
Pages 438-467 in M. F. Allen (ed.),
Mycorrhizal functioning. Chapman &
Hall, New York.
Monsi, M. Z. Uchijima, and T. Oikawa.
1973. Structure of foliage canopies and
photosynthesis. Annual Review of
Ecology and Systematics 4:30 1-327.
Moore, R.P. 1969. History supporting
tetrazolium seed testing. Proceedings of
the International Seed Testing
Association 34:233-242.
92
-------
Morrison, P.H. and F.J. Swanson. 1990.
Fire history and pattern in a Cascade
Range landscape. General Technical.
Report PNW-GTR-254. USDA Forest
Service, Pacific Northwest Research
Station, Portland, OR.
Nadeihoffer, K.J. 1990. Microlysimeter for
measuring nitrogen mineralization and
microbial respiration in aerobic soil
incubations. Soil Science Society of
America Journal 54:411-415.
Nadeihoffer, K.J., J.D. Aber, and J.M.
Melillo. 1985. Fme root production in
relation to total net primary production
along a nitrogen availability gradient in
temperate forests: a new hypothesis.
Ecology 66:1377-1390.
National Research Council. 1995. Review
of EPA’s Environmental Monitoring and
Assessment Program: overall evaluation.
Washington, DC.
Nay, S.M., K.G. Mattson, and B.T.
Bormann. 1994. Biases of chamber
methods for measuring soil CO 2 efflux
demonstrated with a laboratory
apparatus. Ecology 75:2460-2463.
Norby, R.J., C.A. Gunderson, S.D.
Wullschleger, E.G. O’Neill and M.K.
McCracken. 1992. Productivity and
compensatory responses of yellow-
poplar trees in elevated CO 2 . Nature
357:322-324.
O’Donnell, K., E. Cigelnik, N .S. Weber,
and J. M. Trappe. 1997. Phylogenetic
relationships among ascomycetous
truffles and the true and false morels
inferred from 18S and 28S nbosomal
DNA sequence analysis. Mycologia
89(1):48-65.
O’Neill, R.V., B.S. Ausmus, D.R. Jackson,
R.I. Van Hook,, P. Van Vons, C.
Washburn, and A.P. Watson. 1977.
Monitormg terrestrial ecosystems by
analysis of nutrient export. Water, Air,
and Soil Pollution 8:27 1-277.
O’Neill, R.V., J.R. Krummel. R.H.
Gardner, G. Sugihara, B. Jackson, D.L.
DeAngelis, B.T. Mime, M.G. Turner, B.
Zygmunt, S.W. Christensen, V.H. Dale
and R.L. Graham. 1988. Indices of
landscape pattern. Landscape Ecology
1(3): 153-162.
Oades, J.M. 1988. The retention of organic
matter in soils. Biogeochemistry 5:35-70.
Ohtonen, R., and A. M. Markkola. 1989.
Effect of local air pollution on the
sporophore production of mycorrhizal
fungi, mycorrhizae and microbial activity
in Scots pine forests. Meddelelser fra
Norsk Institut for Skogforskning 42:121 -
131.
Ohtonen, R., P. Lahdesmàki, and A. M.
Markkola. 1994. Cellulose activity in
forest humus along an industrial pollution
gradient in Oulu, Northern Finland. Soil
Biology & Biochemistry 26:97-101.
Oliver, C.D. and B.C. Larson. 1990. Forest
Stand Dynamics. McGraw-Hill, Inc.,
New York, N.Y.
Packee, E. C. 1990. Tsuga heterophylla
(Raf.) Sarg., Western Hemlock. Pages
613-622 in R.M. Burns and B. H.
Honkala (eds.). Silvics of North Amenca
Vol. 1, Conifers. USDA Handbook. 654,
93
-------
Forest Service, U.S. Department of
Agnculture, Washington DC.
Parmelee, R. W. 1995. Soil Fauna: Linking
Different Levels of the Ecological
Hierarchy. Pages 107-ll6mC .G. Jones,
and J. H. Lawton (eds.), Linking Species
and Ecosystems. Chapman and Hall.
New York. NY.
Parton, W.J., D.S. Schirnel, C.V. Cole, and
D.S. Ojima. 1987. Analysis of factors
controlling soil organic matter levels in
Great Plains grasslands. Soil Science
society of America Journal 51:1173-
1179.
Parton, Wi., J.W.B. Stewart and C.V.
Cole. 1988. Dynamics of C, N, P and S in
grassland soils: a model. Biogeochemistry
5: 109-131.
Pastor, I. and W.M. Post. 1988. Response
of northern forests to C0 2 -induced
climate change. Nature 334:55-58.
Pastor, J. and W.M. Post. 1986. Influence
of climate, soil moisture, and succession
on forest carbon and nitrogen cycles
Biogeochemistry 2:3-27.
Peet, R. K. and N. L. Christiansen. 1987.
Competition and tree death. Bioscience
37:586-595.
Pell, E.J., J.P. Sinn and C. Vinten Johansen.
1995. Nitrogen supply as a limiting
factor determining the sensistivity of
Populus fremuloides Michs. to ozone
stress. New Phytologist 130:437-446.
Perotto, S., E. Actis-Permo, J. Penigini,
and P. Bonfante. 1996. Molecular
diversity of fungi from ericoid
mycorrhizal roots. Molecular Ecology
5:123-131.
Perotto, S., J. D. Coisson, I. Perugini, V.
Cometti, and P. Bonfante. 1997.
Production of pectin-degrading enzymes
by encoid mycorrhizal fungi. New
Phytology 135:151-162.
Perruchoud, DO. and A. Fischlm. 1995.
The response of the carbon cycle in
undisturbed forest ecosystems to climate
change: a review of plant-soil models.
Journal of Biogeography 22:759-774.
Perry, D. A., M. P. Amaranthus, J. G.
Borchers, S. L. Borchers, and R. E.
Brainerd. 1989. Bootstrapping in
Ecosystems. Bioscience 39:230-237.
Peterson, B. and B. Fry. 1987. Stable
isotopes in ecosystem studies. Annual
Review of Ecology and Systematics 8:51 -
81.
Pierce, L.L., and SW. Running. 1988
Rapid estimation of coniferous forest leaf
area index using a portable integration
radiometer. Ecology 69:1762-1767.
Porteous, L. A., J. L. Armstrong, R. I.
Seidler, and L. S. Watrud. 1994. An
effective method to extract DNA from
environmental samples for polymerase
chain reaction amplification and DNA
fingerprint analysis. Current
Microbiology 29:30 1-307.
Porteous, L. A., R. J. Seidler, and L. S.
Watrud. 1997. An improved method for
purifying DNA from soil for polymerase
chain reaction amplification molecular
ecology applications. Molecular Ecology.
In Press.
94
-------
Post, W.M., W.R. Emanuel, P.J.Zinke and
A.G. Stangenberger. 1982. Soil carbon
pools and world hfe zones. Nature
298:156-159.
Potter, C.S., J.T. Randerson, C.B. Field,
P.A. Matson, P.M. Vitousek, H.A.
Mooney, and S.A. Klooster. 1993.
Terrestrial ecosystem production: a
procees model based on global satellite
and surface data. Biogeochemical Cycles
7:811-841.
Pregitzer, K.S., D.R. Zak, P.S. Curtis, ME.
Kubiske, J.A. Teen and C.S. Vogel. 1995.
Atmospheric C0 2 , soil nitrogen and
turnover of fine roots. New Physiologist
129:579-585.
Prentice, I.C., Cramer, W., Hamson, S.P.,
Leemans, R., Monserud, R.A. &
Solomon, A.M. 1992. A global biome
model based on plant physiology and
dominance, soil properties and climate.
Journal of Biogeography 19117-134.
Prentice, I.C., M.T. Sykes, and W. P.
Cramer. 1993. A simulation model for the
transient effects of climate change on
forest landscapes. Ecological Modellmg
65:51-70.
Raich, J.W., ES. Rastetter, J.M. Melillo,
D.W. Kicklighter, P.A. Steudler, B.J.
Peterson, A.L. Grace, B Moore III, and
C.J. Vorosmarty. 1991. Potential net
primary productivity in South America:
Application of a global model. Ecological
Applications 1:399-429.
Raper, K. B., and D. I. Fennell. 1965. The
genus Aspergillus. The Williams &
Wilkins Company, Baltimore.
Rapport, D. J. 1992. Evolution of
indicators of ecosystem health. Pages
121-134 in D. H. McKenzie, D. E.
Hyatt, and V. J. McDonald (eds.),
Ecological Indicators, Vol. 1. Elsevier
Science Publishers, Ltd., England.
Rastetter, E.B. 1996. Validating models of
ecosystem response to global change.
BioScience 46:190-198.
Rastetter, E.B., M.G. Ryan, G.R. Shaver,
J.M. Melillo, K.J. Nadelhoffer, J.E.
Hobbie, and J.D. Aber. 1991. A general
biogeochemical model describing the
responses of the C and N cycles in
terrestrial ecosystems to changes in C0 2 ,
climate, and N deposition. Tree
Physiology 9:101-126.
Rastetter, E.B., R.B. McKane, G.R.
Shaver, and J.M. Melillo. 1992. Changes
in C storage by terrestrial ecosystems:
How C-N interactions restrict responses
to CO 2 and temperature. Water, Air, and
Soil Pollution 64:327-344.
Rastetter, E.B., R.B. McKane, G R.
Shaver, K.J. Nadethoffer, and A.E.
Giblin. 1997. Analysis of C0 2 ,
temperature, and moisture effects on C
storage in Alaskan arctic tundra using a
general ecosystem model. Pages 437-451
in W.C. Oechel, T. Callaghan, T.
Giomanov, J. I. Holten, B. Maxwell, 0.
Molau, and B. Sveingjornsson (eds.),
Global change and arctic terrestrial
ecosystems. Springer-Verlag, New York..
Rastetter, E.B., A.W. King, B.J. Cosby,
G.M. Hornberger, RV. O’Neill, J.E.
Hobbie. 1992b. Aggregating fine-scale
knowledge to model coarser-scale
95
-------
attributes of ecosystems. Ecological
Applications 2:55-70.
Reichardt, P.B., F.S. Chapin, III, J.P.
Bryant, B.R. Mattes, and T.P. Clausen.
1991. Carbon/nutrient balance as a
predictor of plant defense in Alaskan
balsam poplar: potential importance of
metabolic turnover. Oecologia 88:401-
406.
Roberts, M.J., S.P. Long, L.L. Tieszen, and
C.L. Beadle. 1993. Measurement of plant
biomass and net pnmary production of
herbaceous vegetation. Pages 1-21 in
D.O. Hall, J.M.O. Scurlock, H.R. Boihar-
Nordenkampf, R.C. Leegood and S.P.
Long (eds.), Photosynthesis and
Production in a changing environment.
Chapman and Hall, London.
Rhoades, D. 1979. Evolution of plant
chemical defense against herbivores. IN
G.A. Rosenthal and D.H. Janzen, eds.,
Herbiores: their interaction with
secondary plant metabolites. Academic
Press. New York.
Rovira, A. D. 1991. Rhizosphere research -
85 years of progress and frustration.
Pages 3-13 in D. L. Keister, and P. B.
Cregan (eds.), The rhizosphere and plant
growth. Kiuwer Academic Publishers,
Dordrecht.
Running, SW. 1994. Testing FOREST-
BGC ecosystem process simulations
across a climatic gradient in Oregon.
Ecological Applications 4:238-247.
Running, S.W. and J.C. Coughian. 1988. A
general model of forest ecosystem
processes for regional applications. I.
Hydrologic balance, canopy gas exchange
and primary production processes.
Ecological Modeling 42:125-154
Ryan, M. G., J. M. Melillo, and A. Ricca.
1990. A comparison of methods for
determining proximate carbon fractions of
forest litter. Canadian Journal of Forest
Research 20:166-171.
Ryan, M.R. 1991. The effects of climate
change on plant respiration. Ecological
Applications 1:157-167.
Ryan, M.R., E.R. Hunt, Jr., G.l. Agren,
A.D. Friend, W.E. Pulliam, S.E. Pulliam,
S.E. Linder, R.E. McMurtrie, J.D. Aber,
E.B. Rastetter and R.J. Raison. 1996.
Comparing models of ecosystem function
for temperate comfer forests. 1. Model
description and validation, in Al.
Breymeyer, DO. Hall, 3. Melillo and
G.I. Agren (eds.), Global change: Effects
on coniferous forests and grasslands.
John Wiley and Sons, London pp. 313-
362.
Rygiewicz, P. T., and C. P. Andersen.
1994. Mycorrhtzae alter quality and
quantity of carbon allocated below
ground. Nature 369:58-60.
Rygiewicz, P. T., C. S. Bledsoe, and R. J.
Zasoski. 1984a. Effects of
ectomycorrhizae and solution pH on
[ ‘ 5 NJ ammonium uptake by coniferous
seedlings. Canadian Journal of Forest
Research 14:885-892.
Rygiewicz, P. T., C. S. Bledsoe, and R. J.
Zasoski. 1984b. Effects of
ectomycorrhizae and solution pH on
[ ‘ 5 N] nitrate uptake by coniferous
seedlings. Canadian Journal of Forest
Research 14:893-899.
96
-------
Rygiewicz, P. T., M. G. Johnson, L. M.
Ganio, D. T. Tingey, and M. J. Storm.
1997. Lifetime and temporal occurrence
of ectomycorrhizae on ponderosa pine
(Pinusponderosa Laws.) seedlings grown
under varied atmospheric CO 2 and
nitrogen levels. Plant and Soil 189:275-
287.
Rygiewicz, P.T. and E.R. Ingham. 1997.
Soil biology and ecology. Encyclopedia
of Environmental Sciences. Encyclopedia
of Earth Sciences Series. Van Nostrand
Reinhold, NY. In Press.
Santantonio, D., R.K. Hermann, and W.S.
Overton. 1977. Root biomass studies in
forest ecosystems. Pedobiologia 17:1-31.
Savage, M., P. M. Brown and J. Feddema.
1996. The role of climate in a pine forest
regeneration pulse in the southwestern
United States. Ecoscience 3:310-318.
Scagel, C.F., and C.P. Andersen. 1997.
Seasonal changes in root and soil
respiration of ozone exposed ponderosa
pine grown in different substrates. New
Phytologist 136:627-643.
Schenk, H.J. 1996. Modeling the effects of
temperature on growth and persistence of
tree species: A cntical review of tree
population models. Ecological Modelling
92: 1-32.
Schweingruber, F. H., H. Albrecht, M.
Beck, J. Hessel, K. Joos, D. Keller, R.
Kontic, K. Lange, M. Niederer, C.
Nippel, S. Spang, A. Spinnier, B. Steiner,
and A. Winlder-Seifert. 1986. Abrupte
Zuwachsschwankungen in
Jahrringabfolgen als oekologische
Indikatoren. Dendrochronologia 4:124-
183.
Seidler, R. J., and J. K. Fredrickson. 1995.
Introduction to the special issue on
molecular microbial ecology. Molecular
Ecology 4:533-534.
Seidler, R. J., P. E. Aho, P. N. Raju, and H.
J. Evans. 1972. Nitrogen fixation by
bacterial isolates from decay in living
white fir trees (Abies concolor ((Gord.
and Glend.)) Lindi). Journal of General
Microbiology 73:413-416.
Shugart, H.H. 1984. A theory of forest
dynamics. The ecological implications of
forest succession models. Spnnger, New
York. 278 pp.
Shugart, H.H. and D.C. West. 1980. Forest
Succession models. BioScience 30:308-
313.
Smiley, T. L. and M. A. Stokes. 1968. Tree
Ring Dating. Univ. Arizona Press.
Tucson AZ.
Socki, R.A., H.R. Karlsson, and E.K.
Gibson, Jr. 1992. Extraction technique
for the determination of oxygen-18 in
water using pre-evacuated glass vials.
Annals of Chemistry 64: 829-83 1.
Sollins, P., K. Cromack, Jr., F.M.
McConson, R.H. Waring and R.D. Harr.
1981. Changes in nitrogen cycling at an
old-growth Douglas-fir site after
disturbance. Journal of Environmental
Quality 10: 37-42.
97
-------
Solims, P., C. Grier, F.M. McCorison,
K. Cromack Jr, R. Fogel, R. Frednksen.
1980. The internal element cycles of an
old-growth Douglas-fir ecosystem m
western Oregon. Ecological Monographs.
50: 26 1-285.
Solomon, A. M. 1988. Use of stand models
at varying spatial scales to simulate
forest responses to environmental
changes. Pages 46-58 in R. S. Seymour
and W. B. Leak (eds.), Proceedings of the
New England Growth and Yield
Workshop, Miscellaneous Report 325,
Maine Agriculture Experiment Station,
Orono, ME.
Solomon, A.M. and P.J. Bartlein. 1992.
Past and future climate change response
by mixed deciduous-coniferous forest
ecosystems in northern Michigan.
Canadian Journal of Forest Solomon, A.
M. and H. H. Shugart, Jr. 1993.
Vegetation Dynamics and Global Change
Chapman and Hall, New York NY.
Solomon, A. M., and R. Leemans. 1990.
Climatic change and landscape ecological
response: Issues and analysis. Pages 293-
316 in M. M. Boer and R. S. de Groot
(eds.). Landscape Ecological Impact of
Climatic Change. lOS Press, Amsterdam.
Solomon, A. M., I. C. Prentice, R. Leemans
and W. P. Cramer. 1993. The interaction
of climate and land use in future
terrestrial carbon storage and release.
Water, Air, and Soil Pollution 70:595-
614.
Solomon, A. M. and D. C. West. 1993.
Evaluation of stand growth models for
predicting afforestation success during
climatic warming at the northern limit of
forests. Pages 167-188 in R. Wheelon
(ed). Forest Development in Cold
Regions. Proceedings, NATO Advanced
Research Workshop. Plenum, New York
NY.
Stevenson, F.J., E. Elliott, C.V. Cole, J.
Ingram, J.M. Oades, C. Preston and P.J.
Sollins. 1989. Methodologies for
assessing the quantity and quality of soil
organic matter. P?ges 173-199 in D.C.
Coleman, J.M. Oades, and G. Uehara
(eds.), Dynamics of Soil Organic Matter
in Tropical Ecosystems NifTAL Project,
University of Hawaii Press, Honolulu,
Hawaii.
Stone, J. 1993. Identification of fungi
associated with transgenic potatoes at a
release site. Report to US EPA.
Suter H, G. W. 1993b. Ecological risk
assessment. Lewis Publishers, Chelsea,
MI.
Sweet Home Ranger District, Willamette
National Forest, USDA Forest Service.
1995. South Santiam Watershed
Analysis. The Synopsis. April 1995.
USDA Forest Service. Sweet Home, OR.
Sweinam, T. W. 1987. Western spruce
budworm outbreaks in northern New
Mexico: tree-ring evidence of occurrence
and radial growth impacts from 1700 to
1983. pp. 130-141 IN Jacoby, G. C. and
J. Hombeck, eds., Proceedmgs of the
International Symposium on Ecological
Aspects of Tree-Ring Analysis. CONF-
98
-------
8608144. National Technical Information
Service. Springfield VA.
Swift M. J., 0. W. Heal, and J.
Anderson. 1979. Decomposition
terrestrial ecosystems. University
California Press, Berkeley, CA.
Sylvia, D. M., and S. E. Williams. 1992.
Vesicular-arbuscular mycorrhizae and
environmental stress. Pages 101-124 in
G.J. Bethienfalvay arid R.G. Lindem ian
(eds.), Mycorrhizae in sustainable
agriculture. American Society of
Agronomy, Inc., Madison, WI.
Technicon Industrial Method No. 158-71.
1977. Technicon Industrial Systems.
Tarrytown, New York, 10591.
Telewski, F.W. and A.M. Lynch. 1991.
Measuring growth and development of
stems. Pages 503-555 in J.P. Lassoie and
T.M. Hinckiey (eds.), Techniques and
Approaches in Forest Tree
Ecophysiology. CRC Press Boca Raton,
Florida.
Thompson, R. S., S Hostetler, P. J.
Bartlein, and K. H. Anderson. 1998. A
strategy for assessing potential future
changes in climate, hydrology, and
vegetation in the western United States.
United States Geological Survey Circular
1153, 20 p.
Thornton, K. W., D. E. Hyatt, and C. B.
Chapman. 1993. Environmental
monitoring and assessment program
guide. EPA/6201R-93/012, US EPA,
Office of Research and Development,
Washington, DC.
Tilman., D. 1997. Biodiversity and the
productivity, stability and sustainability
of ecosystems. in Ecosystem services:
Societal dependence on natural
ecosystems. Island Press. Washington,
DC.
Tingey, D.T., and C.P. Andersen. 1991.
The physiological basis of differential
plant sensitivity to changes in
atmospheric quality. Pages 209-234 in
G.E. Taylor, Jr., M.T. Clegg, and L.F.
Pitetka (eds.), Ecological Genetics,
Terrestrial Vegetation and Anthropogenic
Changes in the Atmosphere. Springer-
Verlag, New York.
Tingey D. T., M. G. Johnson D. L.
Phillips, and M. J. Storm. 1995. Effects
of elevated CO 2 and nitrogen on
ponderosa fine roots and associated
fungal components. Journal of
Biogeochemistry 22:281-287.
Tingey, D.T., M.G. Johnson, D L.
Phillips, M.J. Storm and J.T Ball. 1997.
Effects of elevated CO 2 and nitrogen on
the fine root dynamics and fungal growth
in seedling Pinus ponderosa.
Environmental Experimental 37:78-83
Tingey, D.T., D.L. Phillips, M.G.
Johnson, D.W. Johnson and J.T Ball.
1996. Effects of elevated CO 2 and
nitrogen on the synchrony of shoot and
root growth in ponderosa pine. Tree
Physiology 16:905-914.
Tingey, D.T., S. Raba, K.D. Rodecap and
J.J. Wagner. 1982. Vermiculite, a source
of metals for Arabidopsis thahana.
Journal of theAmerican Society for
Horticultural Science 107:465-468.
M.
in
of
99
-------
Tsukada, M. 1982a. Late Quaternary
development of the Fagus forest in the
Japanese Archipelago. Japanese Journal
of Ecology 32:113-118.
Tsukada, M. 1 982b. Cryptomeriajaponica:
Glacial refugia and late-glacial and
postglacial migration. Ecology 63:1091-
1105.
Tuomi, J. 1992. Toward integration of
plant defense theories. Tree 7:365-367.
Tuomi, J., P. Niemalà, F. S. Chapin III, J.
P. Bryant, and S. Sirm. 1988. Defensive
responses of trees in relation to their
carbon/nutrient balance. Pages 57-72 in
W. J. Mattson, J. Levieux, and C.
Bernard-Dagan (eds.), Mechanisms of
woody plant defenses against insects.
Search for pattern. Springer-Verlag, New
York. NY.
Tuomi, J., P. Niemela and S. Siren. 1990.
The panglossian paradigm and delayed
inducible accumulation of foliar phenolics
in mountain birch Oikos 59:399-410.
Turner, M.G. 1989. Landscape ecology:
the effect of pattern on process. Annual
Review of Ecology and Systematics 20:
171-197.
Ueda, 1., Y. Suga, N. Yahiro, and T.
Matsuguchi. 1995. Remarkable N-fixing
bacterial diversity detected in rice roots
by molecular evolutionary analysis of
nifil gene sequences. Journal of
Bacteriology 177:1414-1417.
Upchurch, D.R. 1987. Conversion of
mimrhizotron-root intersections to root
length density. Pages 15-20 H. M.
Taylor (ed.), Mmirhizotron Observation
Tubes: Methods and Applications for
Measuring Rhizosphere Dynamics.
American Society of Agronomy Special
Publication 50. American Society of
Agronomy, Crop Science Society of
America and Soil Science Society of
America, Madison, WI.
Urban, D.L., M. E. Harmon, and C. B.
Halpern. 1993. Potential response of
pacific northwestern forests to climatic
change, effects of stand age and initial
composition. Climatic Change 23:247-
266.
US Department of Agriculture Forest
Service. 1974. Seeds of woody plants in
the United States. Agricultural handbook
No. 450. USDA FS, Washington, DC.
883 pp.
US EPA Science Advisory Board
Environmental Futures Committee. 1995.
Beyond the honzon: using foresight to
protect the environmental future. EPA-
SAB-EC-95-007, Washington, DC.
US EPA. 1979. Method No. 353.2 in
Methods for chemical analysis of water
and wastes. United States Environmental
Protection Agency, Office of Research
and Development. Cincinnati, Ohio.
Report No. EPA-600/4-79-020 March
1979.
US EPA. 1996. Slrategic plan for the
Office of Research and Development.
EPAJ600IR-96/059, Washington, DC.
US EPA. 1997a. Evaluation guidelines for
ecological indicators. ORD Draft report.
US EPA. 199Th. Field operations and
methods forrn measuring the ecological
100
-------
conditions of wadeable and nonwadeable
streams. June 1997. U.S. Environmental
Protection Agency, Corvallis, OR.
US EPA. 1997c. Ecological indicators
researchstrategy. ORD, Draft Report
Vitousek, P.M., and RW. Howarth. 1991.
Nitrogen limitation on land and in the sea:
How can it occur? Biogeochemistiy
13:87-115.
Vitousek, P.M., J.R. Gosz, C.C. Grier,
J.M. Melillo, and W.A. Remers. 1982. A
comparative analysis of potential
mtrification and nitrate mobility in forest
ecosystems. Ecological Monographs
52:155-177.
Vogt, K.A., R.L. Edmonds, and C.C. Gner.
1981. Seasonal changes in biomass and
vertical distribution of mycorrhizal and
fibrous-textured comfer fine roots in 23-
and 180-year-old subalpine Abies
amabilis stands. Canadian Journal of
Forest Research 11:223-229.
Vogt, K.A., D.J. Vogt E.E. Moore, W.
Littke, C.C. Grier, and L. Leney. 1985.
Estimating Douglas-fir fine root biomass
and production from living bark and
starch. Canadian Journal of Forest
Research 15:177-179.
Wagner, F., G. Gay, and J. C. Debaud.
1989. Genetic variation of nitrate
reductase activity in mono- and
dikaryotic populations of the
ectomycorrhizal fungus, Hebeloma
cy1indrospon m Romagnési. The New
Phytologist 113:259-264.
Waring, RH. and G.B. Pitman. 1985.
Modifying Lodgepole pine stands to
change susceptibility to mountain pine
beetle attack. Ecology 66:889-897.
Waring, it H. and W. H. Schlesinger.
Forest Ecosystems: Concepts
Management. Academic Press,
Orlando, FL.
White, T. J., T. Bruns, S. Lee, and J.
Taylor. 1990. Amplification and direct
sequencing of fungal ribosomal RNA
genes or phylogenetics. Pages 3 15-322
n M.A. Innis, D.H. Gelfand, J.J. Srirsky
and T.J. White (eds.), PCR Protocols: A
guide to methods and applications.
Academic Press, Inc.
Widmer, F., L. A. Porteous, K. K.
Donegan, and R. J. Seidiler. 1997a.
Detection of n fFf gene pool complexity
and distribution in a forest soil
ecosystem. Abstracts, American Society
for Microbiology, Miami, FL. Paper N-
180, P.410.
Widmer, F., R. J. Seidler, L. A. Porteous,
L.S. Watrud, and G. Di Giovanni. 199Th.
A highly selective PCR protocol for
detecting 16S rRNA genes of the genus
Pseudomonas (sensu stricto) in
environmental samples. Applied and
Environmental Microbiology, In Press.
Williams, S. 1984. Official methods of
analysis of the Association of Official
Analytical Chemists. 14th ed.
Association of Official Analytical
Chemists, Inc. Arlington, VA.
Willig, M. R., D. L. Moorhead, S. B. Cox,
and J.C Zak. 1996. Functional diversity
of soil bacterial communities in the
Tabonuco Forest: interaction of
1985.
and
Inc.,
101
-------
anthropogenic and natural disturbance.
Biotropica 28(4a):47 1-483.
Wilson, R., A. Cataldo, and C.P. Andersen.
1995. Determination of total
nonstructural carbohydrates in tree
species by high-performance anion
exchange chromatography with pulsed
amperometric detection Canadian
Journal of Forest Research 25:2022-
2028.
Wiltshire, J.J., C.J. Wnght, M.H.
Unsworth, and J. Craigon. 1993. The
effects of ozone episodes on autumn leaf
fall in apple. New Phytologist 124:433-
437.
Yeates, F. W., R. Bongers, R.G.M. Dc
Goede, D.W. Freckman, and S.S.
Georgeiva. 1993. Feeding habits in soil
nematode families and genera--an outline
for soil ecologists Journal of Nematology
25:315-331.
Young, iA. and C.G. Young. 1992. Seeds
of woody plants in North America.
Disocondes Press, Portland, OR. 407 pp.
Young, J.R., E.C. Ellis, and G.M. Hidy.
1988. Deposition of air-borne acidifiers
in the western environment. Journal of
Environmental Quality 17.1-26.
Zak, J. C., M. R. Willig, D. L. Moorhead,
and H 0. Wildman. 1994. Functional
diversity of microbial communities a
quantitative approach. Soil Biology and
Biochemistry 26(9):1 101-1108
Zhu, H., K. 0. Higginbotham, and B. P.
Danick. 1988. Intraspecific genetic
variability of isozymes in the
ectomycorrhizal fungus Suillus
lomenlosus. Canadian Journal of Botany
66588-594.
Zhu, W., and 3. G. Enrenfeld. 1996. The
effects of mycorrhizal roots on litter
decomposition, soil biota, and nutrients
in a spodosolic soil. Plant and Soil
179.109-118.
Zunke, U., and R. N. Perry. 1997.
Nematodes: Harmful and Beneficial
Organisms. Pages 85-133 0. Benckiser
(ed.), Fauna in Soil Ecosystems -
Recycling Processes, Nutrient Fluxes,
and Agricultural Production. Marcel
Dekker, Inc. New York. NY.
102
-------
APPENDIX A
Description of soil profiles
at the Intensive Sites in the
western Cascades.
Falls Creek Site
The soil pit was located on a well-
drained bench with a slope of about 2 -
6%. Subrounded stones within the soil
profile mdicate glacial transport. The
weathering rinds on the stones are
approximately 3 mm thick. The land
surface of this soil may be as old as
800,000 years (Doug Shank, pers.
comm.). Soil development is fairly well
expressed with strong granular A and
upper B horizons with moderate organic
matter content. The texture of the soil is
mostly silt loam with particles of very
coarse sand and very fine gravel. Table
A-i describes the soil profile for the pit
in the clear-cut.
Toad Creek Road Site
The description of the soils derived from
a profile in the clear cut at the High Site
is given in Table A-2.
Table A-i. Soils description for the Falls Creek Site.
Horizon
designation
Depth
(upper-lower)
(cm)
Bulk
density
(g cm
Description
(based upon a description by J. Kern, April 1993)
Oi
2 - 0
Fibnc material, organic layer, few fine and medium
roots
A
0 - 8
0.86
Loam to coarse silt loam with some sand, weak fine
and medium subangular blocky and medium to strong
granular structure, very dark brown, many fine and
medium roots
BA
8 - 13
Silt loam with very fine gravel, very dark grayish
brown and dark brown, strong fine and medium
granular structure, many medium roots
Bwl
13 - 33
0.96
Silt loam with very fine gravel, very dark grayish
brown and dark brown, moderate fine and medium
granular structure, common medium and few coarse
roots
Bw2
33 - 55
0.96
Silt loam with very fine gravel, dark brown, weak
fine and medium granular structure, very few medium
roots
C
55+
Silt loam, clark brown, massive structure, very few
fine roots
103
-------
Table A-2 S’oils description for the Toad Creek Road Site.
Horizon
Depth
Bulk
Description
designation
(upper-lower)
(cm)
density
(g cni 3 )
(based upon a descriptions by J. Kern, April 1993
and D. Lammers, summer 1994)
Oi 6 - 4 Needles and twigs
Oa 4 - 0 Decomposed needles and twigs
Al 0 - 13 0.85 Sandy loam, weak medium subangular blocky
structure, dark reddish brown color, few fine to
medium roots
A2 13 - 38 0.82 Sandy loam, weak medium subangular blocky
structure, dark reddish brown color, many very fine
roots
AB 38 - 52 0.78 Sandy loam, weak medium subangular blocky
structure, dark reddish brown color, common very
fine to coarse roots
Bwl 52 - 92 0.78 Gravely, sandy loam, weak medium subangular blocky
structure, dark reddish brown color, few fine to coarse
roots
Bw2 92 - 112 Cobbelly, sandy loam, weak medium angular blocky
structure, dark reddish brown color, very few fine and
medium roots
Bw3 112 - 136 Gravely, loam, weak fine single grain structure, dark
brown color, few fine and medium roots
Bw4 136 - 166 ExtTemely gravely, loam, weak fine single grain
structure, dark brown color, very few fine roots
104
-------
APPENDIX B
Description of the extensive
field sites
There are a number of existing research
sites in the Pacific Northwest Forests
operated by a number of different agencies
and universities that may be suitable for
out needs. Where possible, our Research
Project will take advantage of established
research sites. Existing sites can reduce our
costs of data collection if they have the
appropriate data and have the appropriate
ecological characteristics.
The H.J. Andrews LTER - The Andrews
LTER is located in the western Cascades in
the McKenzie River Basin. Elevation in the
Andrews forest ranges from 410 to 1630
m. At the main meteorological station at
430 m elevation, mean monthly
temperature ranges from 1 (January) to 18
C (July). Average annual precipitation
varies from 2300 mm at the base of the
Andrews to 3550 mm at upper elevations.
Forest types in the Andrews are typical of
those described above for the intensive
sites, i.e., Douglas fir, western hemlock,
and western red cedar dominate at the
lower elevations, with shifts to Pacific
silver fir and noble fir at higher elevations.
About 40% of the Andrews is in old-
growth stands (dominant trees> 400 yr-
old), about 20% in mature stands that
originated from wildfires (100 to 140 yr-
old), and the remainder in stands of various
ages, composition and stocking densities
resulting from clearcutting and shelterwood
cutting since 1950. The main theme of
ecosystem research in the Andrews is to
understand and contrast the effects of land
use, natural disturbances, and climate
change on key ecosystem properties of
hydrology, C and N cycling, biological
diversity, albedo, trace gas exchange, and
site productivity, topics that mesh well
with some of the research proposed here.
Cascade Center for Ecosystem
Management Permanent Study Plots - The
Cascade Center for Ecosystem
Management oversees an extensive
network of over 100 permanent vegetation
plots in Oregon, Washington and other
western states. Ongoing measurements are
being conducted on the dynamics of tree
and understory species in young, mature
and old-growth forests. These long-term
data descnbe the processes of succession,
tree mortality, canopy gap formation,
biomass accumulation, dynamics of coarse
woody debris, and timber growth. We are
working with Drs. M. Harmon and S.
Acker of Oregon State University at a
subset of these sites to obtain
geographically extensive data that can be
used to develop a regionally robust
parameterization of the biogeochemical
model. The specific sites that we are
studying include the Cascade Head
Experimental Forest near the Oregon
Coast, the Hi. Andrews Experimental
Forest and the Middle Santiam Research
Natural Area in the west Cascades, and the
Metolius Research Natural Area in the east
Cascades. The west-to-east transect
represented by these sites captures a wide
range of climatic and edaphic conditions.
For example, the east end of the transect
(Metolius) has a mean annual temperature
3 °C cooler than the west end (Cascade
Head), receives 1/4 as much precipitation,
and has substantially less total soil
105
-------
nitrogen. As a result of these differences,
there are significant differences in
stemwood production and biomass
accumulation across this transect. At
Cascade Head,, we will study two plots, a
150-yr-old Douglas-fir/western hemlock
stand and a 150-yr-old sitka spruce stand,
both of which are highly productive. Al
H.J. Andrews, we will use two mature
Douglas-fir/western hemlock stands. The
Middle Santiani RNA study plot is in an
old-growth (>500 yr) stand of Douglas fir
that has one of the highest standing
biomasses reported for the Pacific
Northwest. The Metolius RNA study plot
consists of slow-growing, old-growth
Ponderosa Pine. in addition to the existing
long-term biomass data, total soil carbon
and nitrogen stocks have recently been
measured for all of these sites. We will also
collect litterfall data for at least one year at
the Cascade Head and Metolius sites.
Olympic National Park - The University of
Washington has been conducting research
on C and N cycling in the National Park for
more than a decade. These plots in addition
to other plots being established by EPA,
Olympic National Park and USGS will
provide data on C and N cycling under a
range of vegetation and climate conditions.
The Park offers a unique opportunity to
utilize a number of gradients to test model
performance and also specific hypotheses.
The influence of a range of vegetation
types on model performance will use data
East Twin Creek (Hoh River) and the
Quinault River (Sitka spruce, Douglas-fir
hemlock) and Hurricane Ridge (Subapline
fir). Also the model will be tested using
sites differing in precipitation: High
Precipitation, more than 4 meters annually
(East Twin Creek (Hoh River) and the
Quinault River); Low Precipitation, less
than 2 meters annually (Deer Park).
Long-Term Ecosystem Productivity -
Long-term research sites for the USFS
LTEP are located in the Siskiyou,
Willamette, and Siuslaw National Forests
in Oregon and in the Wenatchee National
Forest and on Washington state land in the
Olympic National Forest in Washington.
Many of these forests are generally of a
similar type as our intensive study sites,
but grow under different climatic and
edaphic conditions. The major scientific
questions being addressed by research in
these areas is how species composition and
organic matter affect a wide range of
ecosystem patterns, processes, and
ecosystem productivity over the long term
(200 yr; B. Bormann, pers. comm). These
questions will be addressed by the use of
replicated stand-scale experimental
manipulations which represent a range of
possible forest management practices,
including likely organic matter removals
and establishment of different successional
plant assemblages.
Small-Plot Manipulations - In addition to
the stand-scale manipulations, small-plot
mampulations are being installed in the
Willamette LTEP research site (B.
Bormann, pers. comm.). The major focus
of the small-plot studies is to examine how
mulched organic matter originating from a
wide variety of different plant species and
parts will change soil properties and
growth of Douglas fir seedlings over a 5-10
yr period. These experimental small-plot
sites will provide ideal locations to test
hypotheses for the development of
indicators related to C and N utilization by
producers and consumers.
106
-------
Forest Health Monitoring Program - The
USFS and State Forestry departments are
establishing permanent monitoring plots
under the FHM program as described
above (Campbell and Liegel 1996). We plan
to evaluate sensitivity and variability of
potential ecological tools resulting from our
research in areas adjacent to the FHM
plots. This will enable us to seek
relationships between our process-based
indicators and the FHM program’s
structure-based indicators.
Wind River Canopy Crane - The Wind
River Canopy Crane Facility (WRCCF) is
located in a research natural area (RNA) of
the Gifford Pinchot National Forest, just
north of the Columbia River Gorge in
southern Washington. The site is
cooperatively managed by the College of
Forest Resources of the University of
Washington, the USFS PNW Research
Station, and the Gifford Pinchot National
Forest Wind River Ranger District. The
site is about 330 m in elevation and
exemplifies the old-growth Douglas-fir and
western hemlock forests (trees are
estimated to be about 460 yr-old) which
once covered much of the western
Cascades. The climate is typical of the
western Cascades with about 2500 mm
precipitation per year, 2330 mm of
snowfall per year, and a mean annual
temperature of 8.7 C. Most precipitation
falls between October and May, creating a
summer drought condition. The soils are
sandy barns developed in volcanic tephra
over lahar and basalt bedrock. This site is
unique as the crane allows access to the
canopy at about 60 m above the forest
floor in an area of about 2.3 ha. The site
collects a full suite of micrometeorobogical
data and has collected baseline data on
forest structure.
Ponderosa Pine Research Sites - Several
field sites are located in the ponderosa pine
region of central and eastern Oregon that
we can use for this study. These include
sites in (1) Pringle Falls Experimental
Forest, (2) a Research Natural Area of the
Deschutes National Forest near Camp
Sherman (west of Sisters), and (3) near
Black Butte. The Pringle Falls
Experimental Forest, south of Bend, OR,
was established in 1931 as a field site to
study silviculture and related topics in the
conifer forests east of the Cascade
Mountains, including studies on growth,
insect damage, etc. The site near Camp
Sherman was established to conduct a
stand gas exchange expenment, the major
goal being to detennine the exchange of
CO 2 (including separation of fluxes from
understory, soil, and tree foliage) and water
vapor between the atmosphere and a
ponderosa pine forest using eddy
correlation techniques (M. Unsworth and
colleagues). This site is well instrumented
for climatic variables and has a tower for
canopy access. At the site near Black
Butte, experiments into the response of
large trees to water limitations due to
environmental and hydraulic transport
capacity in the ponderosa pine forest are
being conducted by B Bond and M Ryan
(OSU and USFS).
107
-------
APPENDIX C
Application of MBL-GEM to
the H. J. Andrews Forest
Data collected to date
Data that have been collected to date
include a detailed characterization of C and
N cycling in the low and high elevation
field site forests and clearcuts (see Section
3.1.). Essentially all of the data listed in
Table 3.3 (Section 3.2.1) have been
measured for the Toad Creek Road and
Falls Creek Cascade field sites. Most of the
fluxes have been measured for more than
two years and will continue to be measured
for a period of at least three years. Climate
data (Section 3.1) have been recorded
continuously for at least a full annual cycle
so that seasonal changes in carbon and
nitrogen cycling can be more meaningfully
interpreted. We are currently using the field
data to calibrate MBL-GEM and have
developed a preliminary parameterization
of the model. This parameterization will be
improved as additional data from these and
additional sites become available.
Application of MBL-GEM: an
example
As an illustration of how MBL-GEM can
be used to assess short- and long-term
effects of disturbance on ecosystem C and
N dynamics, we describe here a simulation
of the effects of harvest on an old-growth
Douglas-fir ecosystem. For this example,
we parameterized the model for a 450 yr-
old Douglas-fir forest at the H.J. Andr ws
Experimental Forest in Oregon (Grier and
Logan 1977, Sollins et al. 1980). This
parameterization was then used to simulate
the effects of three consecutive harvests
spaced 60 years apart. At each harvest,
50% of living biomass was removed as
forest products, 40% of the pre-harvest
biomass was converted to logging slash and
added to the detritus pool, and living
biomass was reduced to 10% of the pre-
harvest biomass. After each harvest the
model was run for 60 years assuming no
change in present-day climate.
The model results suggest that this harvest
regime appreciably reduces ecosystem C
(Fig. C-la) and N (Fig. C-ib) stocks, a
result attributable to both the removal of
forest products and to increased rates of
soil respiration (Fig. C-Ic) and N leaching
(Fig. C-id). The recovery of vegetation C
during forest regrowth steadily declines for
consecutive harvests. Figure C-2
summarizes the simulated cumulative
effects of the three harvests, showmg that
after 180 years total ecosystem C
decreased by almost 60%, while decreases
in ecosystem N, NPP and net N
mineralization approached 20%.
These results suggest that frequent and
intense harvests may have a substantial
impact on ecosystem productivity and
sustainability, however, much work
remains to be done to constrain and test the
model. Our purpose here is to illustrate the
use of a process-based model for
synthesizing information on ecosystem
processes and on how those processes
interact over time. It is because of this
synthesis that process-based models
surpass any other method of projecting
responses to environmental change
108
-------
to
S
I ’ _NPP
SoURuon
0 3o
Cut
40 SO SO tOO 120 I40 ISO ISO 300
Cut Cut
Years
-I ._.-...-. Vsgstalc, -
0Dct _ 2 4 0 10 .0I;I,0 ” ”O 200
0
g
03
—. NOON Mm
t—1
,7
02
.
•
0 ?20 40
Cut
SOT SO 100 120T 140 ISO I tO 200
Cut Cut
YeaTs
Figure C-i. MBL-GEM simulation of 3 consesecutive clear cuts. Parameterized for the H.J.
Andrews LTER Site.
Figure C-2. Summarizes the simulated cumulative effects of the three harvests, showing that
after 180 years total ecosystem C.
109
(I ,
a
$
U
IS
I
001
Cumulative Effects of Three Consecutive Harvests
(Harvest Interval =60 yr, Total wood production over 180 yr = 287 MgC/ha)
100
80
-20
-40
0)
-60
-80
0
°‘ -100
Mineralization
------- |