Measuring Sustainability Using the U.S. Forest Inventory and Analysis Program
C.T. Scott1 and W.H. McWilliams1
1
Forest Inventory and Analysis, USDA Forest Service, 11 Campus Blvd., Suite 200,
Newtown Square, PA, USA. e-mail: ctscott@fs.fed.us
Introduction
At the direction of the U.S. Congress, the USDA Forest Service’s Forest Inventory and Analysis
(FIA) Program shifted from a periodic to an annual inventory of the nation’s forest in 1998. All
plots have been made permanent, and the remeasurement cycle now is 5, 7, or 10 years
depending on the region and a state’s contributions to funding. The shift to a panelized annual
survey with shorter remeasurement cycles is generating more current information and has
increased the accuracy of data collected on the status of and changes in the nation’s forest
resources. A more comprehensive inventory that includes forest-health indicators provides data
that address a number of Montreal Process Indicators (MPI) of forest sustainability.
FIA addresses sustainability at the national, regional and state levels. Nationally, FIA provides
data on 14 of the 67 MPI sustainability indicators and is working on 9 others (USDA Forest
Service 2004a). For example, FIA monitors forest areas by forest type, standing volume and
removals, erosion on forest soils, ozone (03) damage, forest biomass and carbon. Regionally, in
collaboration with federal and state partners, FIA monitors forest health, e.g., crown indicators
and mortality and their relationship with areas of pest outbreaks. Common regional and statelevel indicators of sustainability have been identified for use in managing forests and developing
policy. At the state level, FIA produces comprehensive reports that include data on volume by
species and diameter class, ownership productivity, disturbance, growth-to removals ratios,
stand structure, species composition, and regeneration.
Changes in FIA
Citing the need for more timely and comprehensive information, the U.S. Congress directed FIA
as part of the 1998 Farm Bill to make fundamental changes and develop a strategic plan to
implement these changes. As a result, FIA has shifted from a periodic survey to an annual
survey where plots systematic subsets of plots are measured in every state every year. FIA
also has stabilized the remeasurement cycle that varied by state and region to a 7-year cycle in
the East and a 10-year cycle in the West. States now have the flexibility to reduce the
prescribed cycle to 5 years. Twenty-two states currently provide resources to achieve 5-year
cycles.
In addition, FIA now:
1. Samples all forest land. Previously, many wilderness areas, “unproductive” forests, and
National Forest had been excluded.
2. Is changing from a regional to a national emphasis. The sampling frame, plot design,
and field protocols now are consistent nationally.
1
3. Has developed and is implementing standardized sampling and estimation procedures.
Examples include a field manual, data processing and estimation methods, a database,
and Internet access to the data and estimates.
4. Has expanded from a focus on timber to one that includes comprehensive monitoring of
forest ecosystems.
5. Works in partnership with the National Forest System, State and Private Forestry, and
state agencies. States contribute roughly $10 million (U.S.) annually.
6. Has shifted from producing occasional reports based on periodic inventories to
publishing comprehensive, analytical reports for each state every 5 years.
Three-Phase Design
Historically, different inventory methods were used by FIA in different regions. The new design
establishes a consistent sampling and estimation framework to be implemented nationally. The
selected design includes three Phases: stratification with satellite imagery, a ground-plot
network for traditional forest inventory, and a subset of the plot network augmented by foresthealth indicators.
Phase 1: Stratification
Sampling with stratification improves the precision of estimates by grouping similar plots into
strata. Until recently, this was accomplished using aerial photography. Today most FIA regions
use satellite imagery, and classification schemes are becoming more automated. Each region
retains the flexibility to choose the stratification attributes that are most appropriate for its
conditions, e.g., elevation in the mountainous Western United States.
Phase 2: Traditional Forest Inventory Plot Network
Phase 2 consists of FIA’s basic set of ground plots on which traditional forest and tree attributes
are measured by field crews. For the annual inventory, a hexagonal grid was superimposed
over the United States. Each cell covers 2,400 ha and grid centers are spaced about 5 km
apart. Existing plots from previous periodic designs were assigned to each hexagon to produce
a systematic sample across the country. If more that one existing plot fell within a cell, one was
chosen randomly and the rest were dropped. If there were none, a new plot was placed
randomly within the cell. These plots are permanent and scheduled to be remeasured on the
prescribed 5-, 7-, or 10-year cycle.
To accommodate the requirement for an annual inventory, the grid for each state was divided
systematically into panels that extend across each state. The number of panels in a given state
is based on the desired length of a measurement cycle. Because each panel covers the entire
population of interest, population estimates can be obtained from a single panel or any number
of consecutive panels within a measurement cycle. With 15,000 to 20,000 forest plots
measured annually, regional and national trends will be available on an annual basis.
The ground plots that make up FIA’s Phase 2 plots network are clusters of four points. Each
point in the cluster is surrounded by a 7.3-m, fixed radius subplot where trees 12.7-cm or larger
in diameter at breast height are measured. All four subplots total about 1/15 ha. Regions can
2
use a larger diameter limit for the 18.0-m-radius macroplot. Each subplot contains a 2.1-m,
fixed-radius microplot where seedlings and saplings less than 12.7 cm dbh are tallied.
Microplots are offset from subplot centers to minimize trampling. To analyze various domains of
interests, plots are mapped by condition class. Condition classes are defined by a series of
discrete variables – land use, forest type, stand size, regeneration status, tree density, stand
origin, ownership group, and disturbance history. FIA regions began switching to this design in
1995 based on the design developed by the Forest Health Monitoring (FHM) program in 1990.
Phase 3: Forest-Health Indicators
The 1998 Farm Bill mandated the integration of the FIA and FHM plot networks. FHM plots
were incorporated systematically into the new grid as FIA’s Phase 3 subset of plots. This
subset constitutes 1/16 of the total number of Phase 2 ground plots. The Phase 3 grid is
relatively sparse because these plots are expensive and many of the attributes associated with
forest-health indicator must be measured during the growing season.
Phase 3 plots include all of the features and attributes of Phase 2 plots but also additional
protocols associated with the sampling of six “indicators” of forest health: crowns, down woody
materials, soils, lichens, ozone and understory vegetation. Crown density, dieback, and
transparency are measured on all sample trees. Coarse and fine down woody materials are
sampled along three 7.3-m transects radiating from each subplot center. Litter and soil samples
are taken between plot center and the other subplots. Lichen species and abundance are
recorded within a 36.6-m radius of plot center. Ozone exposure is estimated by evaluating
damage to sensitive species within open field conditions. The vegetation indicator will be fully
implemented nationally in 2006. Individual species are recorded on three 1-m² quadrats within
each subplot. Cover by height class also is recorded, by species, within each subplot.
Implementation Status
A strategic plan was developed in conjunction with the 1998 Farm Bill to guide the
implementation of the new requirements. New states have been added each year. Currently,
all but five states and interior Alaska have implemented the new design.
National technical teams have been established to ensure that FIA collects, stores, compiles,
and reports forest-resource information consistently across the United States. The teams have:
1. Published national sampling design and estimation methods for Phase 2 (Bechtold
and Patterson, 2005).
2. Developed national field manuals for Phases 2 and 3 data collection protocols
(USDA Forest Service, 2004b, c).
3. Developed data collection, storage, retrieval, processing, and reporting systems
(http://fia.fs.fed.us/tools-data/)
Several publications that describe standardized estimation and analysis methods for the Phase
3 forest-health indicators are being prepared.
The new design includes a rigorous Quality Assurance/Quality Control (QA/QC) program that
incorporates three checks of field data. Hot checks consist of on-site interaction between field
crews and QA/QC specialists during data collection. Cold checks are remeasurements of
3
recently completed plots by QA/QC staff. Data from the original crews are available for review
by the QA/QC staff in the field. Crews receive scores and additional feedback based on the
quality of their data. Blind Checks are remeasurements taken when data from the original crew
are not available to the QA/QC staff. Data from Blind Checks are used to produce estimates of
repeatability for FIA’s national QA reports.
Measuring Sustainability
Data from each of the three Phases can be used to address a variety of sustainability issues at
various scales.
Phase 1
Geostatistical methods are being applied in different ways to develop a variety of spatial
products. Byproducts of Phase 1 stratification are forest/nonforest maps which are useful in
assessing forest fragmentation. FIA data also are being used to develop more detailed maps of
land use and land cover. Over time, these maps can be used to assess changes in land
use/cover. While FIA has produced tabular estimates of area by forest type for decades, 250-m
MODIS imagery now is being used to develop national forest-type maps.
Phase 2
FIA ground plots have been the primary source of forest statistics in the United States for 75
years. Beginning in the 1960’s, calculations of components of change from remeasured plots
added a new dimension to the inventory. The emphasis on monitoring versus current inventory
has continued to increase. In addition to volume, current status and change estimation can be
expressed as biomass and carbon. Changes affecting carbon pools are of particular interest.
FIA also assesses stand structure and species composition and their trends. By monitoring tree
damage, we can identify relationships between damage and mortality.
Phase 3
FIA collects a variety of forest-health attributes that are used to assess forest sustainability.
Tree-crown attributes are used as early indicators of health problems. Lichen diversity is used
to monitor air quality over time. Some lichen species serve as indicators of old-growth
conditions. Field and lab chemical analyses of soils are used to assess factors such as erosion,
acidity, calcium, organic matter, and carbon. Measures of litter and duff are combined with those
of down woody materials to estimate above-ground carbon pools, fuelbed loading, and wildlife
habitat. Data on ozone damage help relate ozone exposure data and drought conditions to
potential damage to vegetation. Sampling all vascular plants aids in monitoring plant
biodiversity and assessing nonnative invasive species.
Since the late 1940’s, FIA has monitored primary wood-using mills throughout the nation to
monitor timber removals. These studies have been augmented by periodic studies of active
4
logging operations to estimate the source of wood in the forest used to deliver primary forest
products such as sawlogs, pulpwood, veneer logs and other material to the mills.
Since the mid 1970’s FIA has conducted studies of private forest-land owners to determine their
goals and objectives. These studies also are being annualized to provide valuable trend data
on the 10 million private forest owners from whose lands more than half of the nation’s forest
products are derived annually.
Future Direction
FIA continues to work in several areas to enhance its ability to assess sustainability and make
the data useful to forest managers, scientists, policymakers, and other citizens. It is developing
statistical methods to extrapolate the Phase 3 information to Phase 2 plots (and Phase 1) to
better relate and localize the valuable Phase 3 information. Similarly, FIA is developing
techniques for scaling up research results associated with air pollution and climate change.
This also assists in efforts to create nationwide maps for a variety of attributes. Techniques are
being developed to localize estimates (small area estimation). As the U.S. forest inventory
evolves, there is considerable interest in integrating other metrics and other populations of
interest, for example, expanding to other treed lands in urban areas, farm fencerows, and
riparian zones. FIA’s strategic plan for the next 5 years will address these and other issues.
Reference
Bechtold, W.A. and Patterson, P.L. (eds.) (2005) The enhanced forest inventory and analysis
program—national sampling design and estimation procedures, Gen. Tech. Rep. SRS-80.
Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern Research Station. 85 p.
U.S. Department of Agriculture, Forest Service (2004a) National report on sustainable
forests – 2003, FS-766 Washington, DC: U.S. Department of Agriculture, Forest Service. 139 p.
U.S. Department of Agriculture, Forest Service. (2004b) Forest inventory and analysis
national core field guide, volume 1: field data collection procedures for phase 2 plots, version
2.0. Internal report file at USDA, Forest Service, Forest Inventory and Analysis, 201 14th St.,
Washington, D.C., 20250, or http://fia.fs.fed.us/library/field-guides-methods-proc/
U.S. Department of Agriculture, Forest Service. (2004c) Forest inventory and analysis
national core field guide, volume 2: field data collection procedures for phase 3 plots, version
2.0. Internal report file at USDA Forest Service, Forest Inventory and Analysis, 201 14th St.,
Washington, D.C., 20250, or http://fia.fs.fed.us/library/field-guides-methods-proc/
U.S. Department of Agriculture, Forest Service and Northeastern Forest Planners
Association (2002) Sourcebook on criteria and indicators of forest sustainability in the
Northeastern Area, NA-TP-03-02 Newtown Square, PA, U.S. Department of Agriculture, Forest
Service, Northeastern Area State and Private Forestry.
5