Forest Landscape Assessment Tool (FLAT)

Forest Landscape Assessment Tool (FLAT): Rapid Assessment for Land ManagementLisa Ciecko, David Kimmett, Jesse Saunders, Rachael Katz, Kathleen L. Wolf, Oliver Bazinet, Jeffrey Richardson, Weston Brinkley, and Dale J. Blahna

United States Department of Agriculture

Forest Service

Pacific Northwest Research Station

General Technical ReportPNW-GTR-941

September 2016

AuthorsLisa Ciecko is a plant ecologist, Seattle Parks and Recreation, 1600 S Dakota Street, Seattle, WA 98108; David Kimmett is a natural lands program/project manager, King County Parks and Recreation Division, King Street Center, 201 S Jackson St., Room 700, Seattle, WA 98104-3855; Jesse Saunders is a resource information forester, American Forest Management, Inc., 11415 NE 128th Street, Suite 110, Kirkland, WA 98034; Rachael Katz is an environmental planner, Tetra Tech, 19803 North Creek Parkway, Bothell, WA 98011; Kathleen L. Wolf is a research social scientist and Jeffrey Richardson is a postdoctoral research associ-ate, University of Washington, School of Environmental and Forest Sciences, Box 352100, Seattle, WA 98195; Oliver Bazinet is an environmental analyst, Seattle Parks and Recreation, 309 Pontius Ave. N, Seattle, WA 98109; Weston Brinkley is principal and owner, Street Sounds Ecology, LLC, Seattle, WA 98107; Dale J. Blahna is a research social scientist, U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station, 400 N 34th St., Seattle, WA 98103.

Cover: (left) field data collection in a King County, Washington, forest reserve area, photo courtesy of Forterra; (upper right) forested green spaces provide opportuni-ties for active living, and (lower right) field training for FLAT assessment volun-teers, photo courtesy of King County Department of Natural Resources and Parks.

In accordance with Federal civil rights law and U.S. Department of Agriculture (USDA) civil rights regulations and policies, the USDA, its Agencies, offices, and employees, and institutions participating in or administering USDA programs are prohibited from discriminating based on race, color, national origin, religion, sex, gender identity (including gender expression), sexual orientation, disability, age, marital status, family/parental status, income derived from a public assistance program, political beliefs, or reprisal or retaliation for prior civil rights activity, in any program or activity conducted or funded by USDA (not all bases apply to all programs). Remedies and complaint filing deadlines vary by program or incident.

Persons with disabilities who require alternative means of communication for program information (e.g., Braille, large print, audiotape, American Sign Language, etc.) should contact the responsible Agency or USDA’s TARGET Center at (202) 720-2600 (voice and TTY) or contact USDA through the Federal Relay Service at (800) 877-8339. Additionally, program information may be made available in languages other than English.

To file a program discrimination complaint, complete the USDA Program Discrimination Complaint Form, AD-3027, found online at http://www.ascr.usda.gov/complaint_filing_cust.html and at any USDA office or write a letter addressed to USDA and provide in the letter all of the information requested in the form. To request a copy of the complaint form, call (866) 632-9992. Submit your completed form or letter to USDA by: (1) mail: U.S. Department of Agriculture, Office of the Assistant Secretary for Civil Rights, 1400 Independence Avenue, SW, Washington, D.C. 20250-9410; (2) fax: (202) 690-7442; or (3) email: [email protected](link sends e-mail).

USDA is an equal opportunity provider, employer and lender.

Forest Landscape Assessment Tool (FLAT): Rapid Assessment for Land Management

Lisa Ciecko, David Kimmett, Jesse Saunders, Rachael Katz, Kathleen L. Wolf, Oliver Bazinet, Jeffrey Richardson, Weston Brinkley, and Dale J. Blahna

U.S. Department of AgricultureForest ServicePacific Northwest Research Station Portland, Oregon General Technical Report PNW-GTR-941 September 2016

AbstractCiecko, Lisa; Kimmett, David; Saunders, Jesse; Katz, Rachael; Wolf, Kathleen

L.; Bazinet, Oliver; Richardson, Jeffrey; Brinkley, Weston; Blahna, Dale J. 2016. Forest Landscape Assessment Tool (FLAT): rapid assessment for land management. Gen. Tech. Rep. PNW-GTR-941. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 51 p.

The Forest Landscape Assessment Tool (FLAT) is a set of procedures and tools used to rapidly determine forest ecological conditions and potential threats. FLAT enables planners and managers to understand baseline conditions, determine and prioritize restoration needs across a landscape system, and conduct ongo-ing monitoring to achieve land management goals. The rapid assessment process presents a cost-effective opportunity for landowners that include local governments, private owners, and nongovernmental organizations to use ecological data to guide decisionmaking and improve environmental outcomes on their lands. This report is an introduction to FLAT, providing an overview of its purpose, methods, and implications for land management in diverse regions. FLAT is executed in three sequential phases: Phase 1—Forest Cover Type Mapping, Phase 2—Field Assess-ment, and Phase 3—Management Prioritization. Overall, FLAT consists of onsite visual estimation (aided by remote sensing) of ecological conditions by a trained field team to produce a forest inventory. In addition to providing baseline data and a framework to prioritize actions, FLAT can be used as a monitoring tool to evaluate changing conditions and inform adjustments in management strategies and priori-ties. To illustrate FLAT implementation, the King County Department of Natural Resources and Parks case study details a pilot project conducted on 24,700 of the more than 26,000 ac of county lands. King County is using the results from FLAT to develop and implement forest stewardship plans and target efforts of its volunteer restoration program. Although the tool was initially used in lowland forests in the Puget Sound region, in concept, FLAT could be expanded and adapted for use in a wide variety of ecosystem types.

Keywords: Forest, assessment, stewardship, planning, ecology, baseline, monitoring, land management.

Executive Summary The Forest Landscape Assessment Tool (FLAT) is a set of procedures and tools designed to provide local government agencies, nongovernmental organizations, land managers, and private landowners with a rapid, systematic, flexible, and inex-pensive environmental evaluation. The technical information produced by FLAT provides a standardized baseline of ecological data.

This data can be used to evaluate forest parcels within the context of the broader landscape, providing information about key forest characteristics and potential threats. FLAT also provides a framework to identify high-ecological value, high-threat areas within a single site and across multiple holdings for poten-tial management action. The results from FLAT provide an essential starting point for developing either a systemwide stewardship plan or management plans for single parcels. In addition to providing baseline data and a framework to prioritize actions, FLAT could also be used as a monitoring tool to evaluate changing conditions and inform adjustments in management strategies and priorities.

FLAT was developed and piloted by the Green Cities Research Alliance. Key FLAT contributors include the USDA Forest Service Pacific Northwest Research Station, King County, Forterra, American Forest Management (formerly Interna-tional Forestry Consultants), and the University of Washington. Work on FLAT began in 2009 to provide a comprehensive forest resource analysis for King County, an important first step toward developing a long-term, systemwide forest steward-ship program on more than 26,000 ac of King County lands.

Based on principles of restoration and landscape ecology, as well as traditional forestry, FLAT is conducted in three broad sequential phases:

Phase 1—Forest Type Mapping: Aerial imagery and boundary data are used in a lab or office to divide a project area into management units (MUs), the unit of observation and measurement for the assessment. Data attributes are also developed during Phase 1 based on local conditions and assessment purposes (e.g., species composition, size and age classes, invasive species, tree-canopy vigor, etc).

Phase 2—Field Assessment: A trained field team visits the project area to collect estimates of each attribute for each MU. Such teams may include professionals, technicians, and volunteer stewards.

FLAT data provide the basis for forest stewardship or management plans.

Phase 3—Management Strategies and Prioritization: The data, which provide a snapshot of ecological conditions in the project area (within and across all MUs), can be used to classify or rank each MU. The assigned values can be viewed spa-tially to provide a mapped, visual representation of landscape conditions. These re-sults can then be used to prioritize where on-the-ground management actions would most improve ecological function and health, contributing to long-term sustainabil-ity of a forest area.

Owing to the success of local conservation efforts, property acquisitions, and various incentives to conserve open space, King County’s resource managers have an extensive inventory of resource lands, but have little condition information to guide management efforts. Seeking to learn more about these diverse holdings and inform management decisions, FLAT was implemented on 24,700 ac of the more than 26,000-ac system.

For the first time, King County managers now have baseline ecological infor-mation about all of their forested parks and natural areas. A key finding was the prominence of red alder (Alnus rubra Bong.) and bigleaf maple (Acer macrophyllum Pursh) on many parcels. Both are relatively short-lived species, and a high percent-age of these trees are 30 to 100 years old. These forests could benefit from active management and restoration, as the health decline of old trees signals the need for tree replacement and hazard management.

FLAT utilizes a straightforward rating scale of 1 to 9 to indicate a balance of species composition value (favoring larger native trees) and degree of health threat, particularly owing to invasive species cover. Overall, more than 5,000 ac of King County forest lands have a rating of 2 or 3, indicating high forest composition value and medium-to-high forest health threat values. In the case of McGarvey Park Open Space, as an example, a large portion of the forest cover is rated as 5 or 6, indicating that there is a medium composition value and medium-to-high presence of forest health threats. The ratings can become the basis for priority setting for both locations of management work, and for on-the-ground actions to conserve or restore forests.

King County is using FLAT results to develop forest stewardship plans for indi-vidual sites, as well as communicate and implement management priorities across the entire parks and open space system. In the future, FLAT may be conducted in recurring intervals to monitor progress and evaluate the effectiveness of restora-tion efforts. As a relatively new tool used in forested lands, FLAT has potential for further development, testing, and refinement.

FLAT can be used for both baseline and monitoring forest assessment.

Resource managers in other communities can learn from King County’s experi-ences and modify the FLAT methods to conduct an assessment that addresses local conditions and priorities. Although currently designed for use in lowland forests of the Puget Sound region, the tool’s basic framework and data variables could be adapted for a wide variety of ecosystem types. Future research could generate indicator and matrix tools that inform management prioritization for additional ecosystem types (such as pine forest or riparian systems). Practitioners may also be interested in applying FLAT as a practical method for general ecosystem monitor-ing focused on a particular ecosystem condition or outcome, such as biodiversity, wildlife habitat, or local effects of climate change.

Contents 1 Introduction 2 Background 2 Importance of Assessment for Decisionmaking and Adaptive Management 4 Overview of FLAT 5 Foundations of FLAT 8 Comparison to Existing Assessment Methods 12 FLAT Methodology 12 Project Considerations and Planning 14 Phase 1: Forest Cover Type Mapping 15 Phase 2: Field Assessment 16 Phase 3: Management Prioritization 19 King County Parks and Open Spaces Case Study 20 Project Process 26 Results 30 Management Implications 32 Conclusions 32 Importance in Land Management 33 FLAT Feasibility 34 Stewardship 35 Limitations 36 Next Steps 37 Acknowledgments 38 References 40 Appendix 1—How to Develop the Assessment Area and Management Units 45 Appendix 2—Necessary Data 47 Appendix 3—Equipment List 48 Appendix 4—King County Data Attributes 50 Appendix 5—Cost Analysis 51 Appendix 6—FLAT Field Manual

1


Introduction The Forest Landscape Assessment Tool (FLAT) is a set of processes and tools that produces an ecological assessment for environmental land use planning and man-agement. In its pilot use, FLAT has proven to be a flexible, systematic, and low-cost process for land managers and related decisionmakers to achieve a rapid ecological survey of a portfolio of diverse parcels and land holdings.

FLAT can provide landowners with straightforward ecological information regarding the composition and overall health of their forest lands, and to understand potential threats. It is a tool that can readily be used by such landowners as local governments, private owners, and nongovernmental organizations (NGOs) with minimal training and time invested to achieve useful results. The data and rating framework provided by FLAT better prepare land managers and planners to make strategic land management and restoration decisions.

Assessments can potentially be repeated over time as a practical monitoring program to observe forest changes and gauge the effectiveness of management actions. A longer view can help managers understand how and why conditions are changing on the ground and enable them to adjust programs accordingly.

Although the FLAT protocols presented here have been developed for use in lowland forests of the Puget Sound region located along an urban-to-rural landscape gradient, the tool could be adapted for land managers working in other ecosystems as well.

This report provides an overview of the FLAT approach. The following sec-tions will demonstrate its implementation and describe its data outputs, including assessment options. The “Background” section first reviews the importance of ecological assessment for effective land management, then describes FLAT, includ-ing its origins and function relative to other assessment methods. With this under-standing, the “FLAT Methodology” section provides details on important project considerations and how to use FLAT. Additional method details are provided in the appendices. Next, the “King County Parks Case Study” section presents results and insights from a pilot FLAT project, and is one example of how FLAT has been successfully applied to support stewardship goals. The report concludes with discussion of FLAT’s unique advantages, some key limitations, and potential next steps for management application, monitoring, and research.

FLAT is a flexible low-cost tool for rapid ecological field surveys.

2

GENERAL TECHNICAL REPORT PNW-GTR-941

BackgroundImportance of Assessment for Decisionmaking and Adaptive ManagementThe need for FLAT stems from the importance of having comprehensive, reliable, and unbiased data for decisionmaking. For instance, private firms recognize the importance of information to drive decisions. Private companies develop ways to access and purchase information to understand their clients and the markets they compete in, as well as their competitors. Having high-quality data is considered a defining element of rational strategic management and has been found to sig-nificantly influence decision effectiveness (Dean and Sharfman 1996). Access to good data for decision support is equally important for land use management and environmental planning.

An ecological assessment is an important component of the environmental planning process (fig. 1) (Randolph 2011). This is particularly true in situations in which initial data are minimal or lacking altogether. For example, designations of conservation easements (by NGOs and landowners) may happen opportunistically and result in public land managers having little information about lands for which they have become responsible. For many local agencies, data collection is low on the priority list owing to budget constraints and other pressing needs. In many

Social GoalsExamples:

recreation uses, equity, aesthetics

Social AssessmentExamples:

park ambassador program,user surveys

Ecological AssessmentExamples:

FLAT, FIA, REA

Ecological GoalsExamples:

ecosystem services, habitatprotection, sustainable harvest

Economic GoalsExamples:

revenue, employmentself-sufficiency

Economic AssessmentExamples:

cost-benefit analysis, valuationstudies, market analysis

ManagementDecision

Figure 1—Importance of assessment in the land-management decision process. FIA = ForestInventory andAnalysis, REA = Rapid Ecological Assessment.

3


cases, even if information exists, it may not be comparable in methodology, com-pleteness, or quality across different parcels. This lack of consistent information can make it difficult, if not impossible, to strategically prioritize sites for manage-ment actions or target resource allocations across a system.

For the forested landscapes considered by FLAT, management and plan-ning decisions often address multiple, nested landscape scales. They may start from large, more broadly defined areas and system goals (such as a watershed), then be translated to site-specific plans (Marsh 1978). Some form of assessment ideally takes place at each of these scales so that strategies can be developed to achieve goals:• Across a landscape system, by protecting specific, strategically located

lands with easements, conservation status, or acquisition,• That consider action alternatives, such as restoration, development,

or harvest,• Applied within a parcel or holding, for instance, by prioritizing certain

areas for restoration work, and• To address particular needs of any distinct system having a natural bound-

ary (such as a riparian corridor or wetland).

FLAT is designed primarily for use on the landscape scale to provide key eco-logical information for each of any number of sites within an ecological system or parcel network. The data can be aggregated and analyzed to support a management approach that is successful and sustainable over time. FLAT data can also serve as a starting point for more detailed site-specific plans or monitoring.

Specifically, FLAT can provide critical inputs in the adaptive management cycle (fig. 2). Its use can provide both upfront baseline information—assessment—and evidence of change over time—monitoring. Ongoing, systematic land assessments allow managers to recognize important environmental gaps and potential tipping points. Cost savings are possible because planned actions can both prevent emer-gency conditions and optimize possible revenue. Ongoing assessment also helps landowners and decisionmakers tell the story of their land and its variety of public and private benefits that are worthy of investment and protection.

Of course, assessment is useful only when it is carried out successfully. As discussed later, FLAT fills a need for a rapid, low-cost method that can be applied in a range of environmental contexts, and particularly in the high-pressure interface of natural and urban or suburban areas.

FLAT generates assessment data across a system of lands for land units within a parcel.

4


Overview of FLAT FLAT is an assessment technique that provides land managers and planners with ecological information to: • Make decisions at the broader, systemwide scale, and • Prioritize different parcels or areas within parcels for specific land uses

or actions.

Its advantage over more traditional forestry sampling, measurement, and extrapolation techniques is that it allows a rapid assessment of ecological condi-tions based on visual on-the-ground surveys of management units delineated from remote-sensing data. Although FLAT may not be as precise as more research-oriented sampling techniques, it efficiently provides accurate, practical descriptions of ecological conditions within and across a collection of sites. This “thumb-nail sketch” can be used to target where indepth assessment may be necessary.

Generally, FLAT consists of visual estimates of ecological conditions by trained individuals to produce a forest inventory. Although field teams should follow guidelines to calibrate their estimates, the assessment itself is generally qualitative and relies on carefully prepared estimates rather than precise measurements. FLAT is executed in three phases:

Phase 1: forest cover type mapping—Aerial imagery and boundary data are used to divide each parcel within a project area (such as King County, as described below) into polygons, and to delineate

Figure 2—Assessment-based adaptive management cycle.

Assess

Monitor

Definestrategy

Implement

Adjust

Evaluate

5


management units (MUs). This work is accomplished and recorded using geo-graphic information system (GIS) tools. Management units are the units of observa-tion and measurement for the onsite assessment.

Phase 2: field assessment—Trained field teams visit each MU within the project area to collect data for prede-termined attributes. As an example, collected attributes might include vegetative or built land covers, nonnative species in order of abundance, and tree age-class distribution. A condition rating is also recorded, based on a matrix and flow-chart analysis (described as Tree-iage). Data are collected for each MU and stored (using a GIS or other data management system) for each parcel within the project area. Field teams may also verify the boundaries of MUs in this phase.

Phase 3: management strategies and prioritization—The field procedures of FLAT provide a ranking of landscape conditions of the project area, both across numerous parcels and for subunits (the MUs) within a large parcel. Summary data and rankings are easily compared across the land management area. They can be used to establish management priorities for each MU, or aggregated to develop priorities at larger scales, such as across the entire open space or parks system.

FLAT provides a standardized baseline of ecological condition data. This information can be used to view each MU within the context of an entire land management system, as well as provide a starting point for developing a land-use or stewardship plan for particular parcels. Repeated over time, FLAT could serve as an effective monitoring tool for managers to review and then adapt management priorities and actions based on changing conditions.

This streamlined and systematic approach to ecological assessment applies principles of ecology and forestry to provide quality data that can inform land management priorities.

Foundations of FLATStarting in 2009, FLAT was developed and piloted by the King County Natural Resources and Parks Department, in collaboration with the U. S. Department of Agriculture (USDA) Forest Service and other partners. The resulting comprehen-sive resource analysis was an important first step toward developing a long-term, systemwide forest stewardship program on King County lands. FLAT incorporated the Tree-iage analysis approach that was initially developed in 2005 by the Green Seattle Partnership to prioritize restoration sites.

As its name suggests, the Tree-iage analysis is drawn from the medical practice of triage. Within the medical professions, triage emerged from the demands of war

6


casualties, where medical providers were faced with a scarcity of resources and needed to rapidly assess each patient’s condition, and, based on an established sys-tem or plan, determine the patient’s specific treatment or treatment priority (Iserson and Moskop 2007).

Just as a nurse or doctor in an emergency department uses a systematic check-list of symptoms or a set of criteria to rapidly prioritize patient care, FLAT provides a systematic checklist of indicators that determine an actionable score for a desig-nated forest area. A 9-point scoring matrix (fig. 3) combines observations of native tree and canopy composition with degree of invasive plant presence, particularly on the forest floor. Generally, low scores indicate lands that have good forest integrity and low invasive-species presence, thus merit management that will sustain their quality. High scores are indicative of highly altered forests combined with extensive invasive-species presence that require more resources to recover or restore.

Figure 3—Original Green Seattle Partnership Tree-iage matrix, including acreage per category in Seattle, Washington.

1Monitoring and

stewardship(41 ac)

4Planting(39 ac)

7Evaluation andmajor planting

(44 ac)

2Invasive plant

reduction(330 ac)

5Invasive plantreduction and

planting (442 ac)

8Invasive plantreduction andmajor planting

(380 ac)

3Major invasive plant reduction

(95 ac)

6Major invasive plant reduction and planting

(608 ac)

9Major invasive

plant reduction andmajor planting

(633 ac)

HIGH>25% native tree

canopy cover,>50% canopy cover

is evergreen

MEDIUM>25% native tree

canopy cover,<50% canopy cover

is evergreen

LOW<25% native tree

canopy cover

LOW<5% invasive cover

MEDIUM5 to 50%

invasive cover

HIGH>50% invasive cover

Threat

Tree

com

posi

tion

valu

e

7


The original Tree-iage matrix was enhanced for use in FLAT to serve more diverse needs and ecological conditions. Modified flowcharts accommodate additional ecosystem types such as wetlands. Additional data attributes include more indicators of forest health, adding new insight for management strategies. New attributes include stocking, crown closure estimates, and forest health indicators. Health indicator measures include low tree-canopy vigor, root rot, mistletoe, bare soils resulting from erosion, and the presence or lack of regenerating trees (canopy species less than 20 ft tall). Further, each visited stand is deemed “plantable” or “not plantable” based on whether site conditions are appropriate for tree seedling establishment.

Tree-iage was first used to evaluate forest condition of open spaces (of up to 100 ac) in highly urban areas in Seattle (fig. 3; Green Seattle Partnership 2006). The tool was exported to several other cities under the regional Green City Partnership model, including Everett, Kent, Tacoma, Kirkland, and Redmond. The development and use of FLAT in King County expanded use of the early tool for assessments of larger parcels (up to 2,500 ac). The FLAT team also prepared fieldwork protocols that are used for data-collection training of park staff and citizen volunteers.

The FLAT process borrows from a number of scientific disciplines. The axes and decision flowcharts of the Tree-iage matrix are based on principles of restora-tion ecology. Restoration ecology involves identifying prior or potential ecological conditions for a site that are then considered as goals or targets of the restoration effort and process. Such conditions (past and future) are shaped by a number of abi-otic factors such as climate, elevation, moisture and precipitation cycles, nutrients, water bodies, fire cycles, and soil or substrate (Clewell and Aronson 2007).

FLAT has been used exclusively within western Washington thus far. The clas-sification matrix and flowcharts therefore assume a desired condition of late-suc-cessional lowland native forests of the Puget Sound basin, characterized by mature conifer trees of mixed age classes and species, mixed with large deciduous trees. The matrix could be adjusted to acknowledge other Pacific Northwest ecosystems (such as shoreline, pine forest, or shrub-steppe) by specifying key ecological condi-tions and species indicators for the vertical and horizontal axes. To adapt FLAT for use in other ecosystems, goals of a desired condition or resource use would inform how the matrix and attributes were set up.

The FLAT process also incorporates knowledge from the field of landscape ecol-ogy, which makes explicit the importance of ecological diversity within the landscape (Turner 2005). When applied to land management, landscape ecology can provide insight on how bordering development, land use, or ecosystem types may influence restoration outcomes or ecosystem development within a particular management unit

FLAT is based on the best practices of foresty, restoration ecology, and landscape ecology.

8


(Turner 2005). One of the main products of FLAT is MU-specific information that can be viewed and queried in GIS together with information on neighboring MUs and other relevant spatial data or imagery. This integrated format allows managers to consider a landscape-scale perspective when making management decisions.

FLAT also includes aspects of traditional forestry field methods that were modified to include less data collection concerning timber evaluation. Associated data attributes in the FLAT include estimates of diameter at breast height (dbh) or size class, stocking, age class, and a number of forest health indicators as noted earlier (e.g., root rot, mistletoe, regenerating trees). This provides a critical starting point for developing silvicultural restoration actions that can promote forest stand recovery and long-term resilience. If a parcel is deemed to be suitable for harvest management, more detailed forest mensuration can be done.

Comparison to Existing Assessment MethodsIn recent years, many approaches to landscape or resource assessment have emerged. The following summary and table 1 provide a comparison of FLAT with other assessment programs.

Forest Inventory and Analysis Forest Monitoring Program (FIA)—Perhaps the most well-known forest inventory protocol is the one used by the Forest Service’s Forest Inventory and Analysis (FIA) program. The FIA consists of “a three-phase sample used to track status and trends in forest extent, cover, growth, mortality, removals, and overall health” for the entire United States. Data collection takes place through stratified random sampling, selecting one site for every 6,000 ac of forest (USDA FS 2014). The FIA has produced a dataset that enables analysis on a scale that is appropriate and useful for national and some statewide manage-ment and decisionmaking.

Because of its plot sampling design, FIA may not provide data at the scale that most local or regional landowners would need to make management decisions. Fur-ther, the extensive range of detailed measurements involved may be much too costly and time-consuming for managers with a limited budget to consider. As an exam-ple, FLAT data variables for King County numbered just 28, while FIA’s urban phase 2 protocol calls for more than a hundred variables to be carefully measured (USDA FS 2015). The advantage of FLAT compared to FIA is its lower cost, greater simplicity, and ability to provide information about each parcel and the management units within them. Nonetheless, in recent years, FIA has been extended to areas that are identified as urban. Urban FIA was launched in 2014, with Baltimore and Austin as the pilot cities. Data-collection protocols have incorporated i-Tree tools, and are being adapted to more directly reflect local community needs.

9


Tabl

e 1—

Sum

mar

y co

mpa

riso

n of

FLA

T to

oth

er a

sses

smen

t met

hods

Key

feat

ures

FLA

TFI

Ai-T

ree

Eco

RE

AN

RC

AG

eogr

aphi

c fo

cus

Urb

an/su

burb

an/ru

ral

Nat

iona

lU

rban

Nat

iona

lN

atio

nal P

ark

Syst

em

City

/regi

onal

/cou

nty

Inte

rnat

iona

l

Scal

eM

anag

emen

t uni

t (M

U)

App

roxi

mat

ely

one

site

per 6

,000

ac

acro

ss th

e U

nite

d St

ates

City

or l

and-

use

type

Varia

ble;

ofte

n la

rge-

scal

e la

ndsc

apes

Nat

iona

l par

k un

it (e

.g.,

Oly

mpi

c N

atio

nal P

ark)

Met

hods

GIS

fore

st c

over

map

ping

Fiel

d sa

mpl

e pl

ots e

stab

lishe

d fo

r ann

ual d

ata

colle

ctio

n

Stra

tified

fiel

d sa

mpl

ing

or in

vent

orie

sR

emot

e se

nsin

g or

pho

to

delin

eatio

n of

eco

logi

cal

units

Synt

hesi

s and

ana

lysi

s of

exi

stin

g da

ta

Fiel

d su

rvey

of M

Us

100+

var

iabl

esM

easu

rem

ents

focu

sed

on u

rban

fore

st st

ruct

ure

Fiel

d sa

mpl

ing

Dev

elop

men

t of

indi

cato

rs fo

r var

iety

of

eco

syst

em ty

pes

Estim

ates

of a

lim

ited

set o

f va

riabl

es fo

cuse

d on

fore

st

com

posit

ion

and

heal

th

char

acte

riza

tion

Det

aile

d m

easu

rem

ents

Mod

el e

stim

ates

ec

osys

tem

func

tions

and

ec

onom

ic v

alue

s bas

ed

on fo

rest

stru

ctur

e da

ta

Mea

sure

men

ts in

clud

e flo

ra a

nd fa

una

Hig

h-le

vel p

ark

man

agem

ent

prio

ritiz

atio

n

Expl

icit

man

agem

ent

prio

ritiz

atio

n fr

amew

ork

Dat

abas

e op

en fo

r va

riety

of a

naly

ses

U

niqu

e an

alys

es a

nd

wor

ksho

ps fo

r eac

h R

EA

to re

view

info

rmat

ion

and

set p

riorit

ies

FLA

T =

Fore

st L

ands

cape

Ass

essm

ent T

ool;

FIA

= F

ores

t Inv

ento

ry a

nd A

naly

sis;

REA

= R

apid

Eco

logi

cal A

sses

smen

t; N

RCA

= N

atur

al R

esou

rce

Con

ditio

n A

sses

smen

t;

GIS

= g

eogr

aphi

c in

form

atio

n sy

stem

.

10


I-Tree Eco—A widely-used inventory and assessment tool that has been used in cities across the world is i-Tree Eco, also developed and supported by the Forest Service. The i-Tree Eco tool is set up to measure and monitor urban forests in order to estimate ecosystem functions and economic values for any metropolitan area. These estimates include urban forest structure and associated ecosystem services, such as carbon storage and sequestration, and air pollution removal, as well as residential building energy effects, rainwater interception, and public health benefits (Nowak et al. 2008). When applied to metropolitan areas or counties, this tool uses random or stratified sampling.

The i-Tree Eco model provides an important baseline for systemwide informa-tion, especially within a city. However, much like FIA, these techniques do not readily describe the status of a specific site. An entire city, or the land-use types within, is the unit of analysis in i-Tree and results can be used to help identify systemwide goals and opportunities. Because FLAT’s unit of observation is a parcel (and even subunits within), it can be more readily used to create plans to meet place-based goals. Thus, although an i-Tree Eco assessment and analysis can be a useful complement to FLAT, its information is not tied as directly to site-specific management needs because of the difference in focus and scale of data collection. It should be noted that i-Tree Canopy and Landscape tools are applicable at the parcel scale, but do not address forest health and stand conditions.

The Nature Conservancy’s Rapid Ecological Assessment—The Rapid Ecological Assessment (REA) was developed by The Nature Conser-vancy to identify priority areas for conservation of biodiversity. In many ways, it is very similar to the FLAT process. In both, orthophotos or other remote-sensing technologies are used to classify and divide landscapes into ecological units, then field-based assessments are used to characterize the biota within these units. The purpose of REA is also to evaluate ecological conditions of specific units to support decisions about management priorities (Sayre et al. 2000).

There are a few key differences between REA and FLAT, however. Meth-odologically, REA field assessments (like FIA and i-Tree) consist of plot-based samples and interpolation of those measures across a much larger area. The REA also includes fauna explicitly in its sampling technique while FLAT focuses exclu-sively on flora. Unlike FLAT, REA was developed for use in lands more remote from human activity. The REA also involves a unique process of technical analyses and workshops in which scientists and managers review the information collected and decide on unit prioritization (Sayre et al. 2000). FLAT, on the other hand, has a predefined framework of matrices and flowcharts that can be used to readily translate assessment data into ecological management priorities.

11


National Park Service Natural Resource Condition Assessments—The National Park Service (NPS) Natural Resource Condition Assessment (NRCA) serves a similar purpose to FLAT. These reports synthesize preexisting scientific information on a particular park to support the development of management plans and identify what additional information is needed, as well as to develop priorities across the National Park System. Although some of the data may be collected in the field, the majority of report content is expected to come from “existing data from NPS and other sources” (USDI NPS 2009). The NRCA reliance on existing data is one important distinction between it and the FLAT method.

Another important difference stems from the diversity of holdings within the National Park System. The FLAT process can be customized based on a general understanding of the range of ecosystems that natural area managers or planners might encounter, as well as the management challenges that may be present. Within this range, specific indicators as well as flowchart and matrix analyses are created to generate information that is directly helpful to prioritize MUs and management activity. The NRCAs, on the other hand, are a step earlier in the process. The NRCAs are oriented toward identifying which indicators should be used to set priorities going forward for each park across a much wider range of habitats and conditions. The NRCAs might be used to define each park’s prioritization matrix and flowcharts for conducting FLAT. The differences between FLAT and the NRCAs are thus of process timing and scale.

Ecological Integrity Assessment of the Washington Natural Heritage Program—An Ecological Integrity Assessment (EIA) rates the current ecological integrity of an occurrence of a plant association or ecological system. NatureServe and the Natural Heritage Network have developed the EIA as an index of ecological integrity based on metrics of biotic and abiotic condition, size, and landscape context. Each metric is rated by comparing measured values with the expected values under relatively pristine conditions. The ratings are aggregated into a total score or a scorecard matrix. The EIA can be applied to multiple spatial scales (e.g., landscape-or site-scale) and with a variety of data types (e.g., GIS or field-based). The EIAs are developed for ecological systems using a three-level-metrics approach: remote sensing, rapid ground-based, intensive ground-based. In sum-mary, the EIA framework provides a standardized currency of ecosystem integrity across all terrestrial ecosystem types. This information can then be used for setting conservation priorities, identifying restoration strategies, and monitoring the effectiveness of conservation actions.

12


The EIA three-level-metrics approach may offer helpful concepts and methods for future updates of FLAT. The EIA is used to evaluate more pristine parcels for wildland conservation or management. FLAT is structured to enable land managers to integrate measures that reflect the local situations of urban and community forest systems. FLAT can be adjusted to reflect the “relative” value of natural parcels concerning desired attributes, whereas the EIA indicates “abso-lute” ecological values. Furthermore, although the EIA is intended to create a common currency of comparison between different ecosystem types, the metrics and methodologies employed in the field are detailed and ecosystem specific. FLAT, on the other hand, uses a consistent set of less detailed metrics in the field, without assuming prior knowledge of the ecosystem type. After FLAT is done, EIA could be used as a next step if more detailed assessment is needed within specific MUs.

Overall, FLAT fills an assessment role that traditional forestry assessments and the other methods discussed above do not address: • Uses methods that are simple and adaptable to project-specific goals.• Provides adequate, reliable, systematic, cost-effective, and local, site-based

information. • Informs decisions about where to initiate healthy forest management, stew-

ardship programs, restoration activities, or stand management for harvest. • Identifies where additional, more precise data may be needed.• May be used to monitor conditions and progress over time.

FLAT MethodologyThis section provides a summary of key project considerations and the basic meth-ods for the three phases of FLAT.

Project Considerations and PlanningBefore starting a FLAT project, managers should review and consider important factors for project planning, including the desired budget, staffing and training needs, equipment requirements, and a potential timeline.

Costs—The cost of executing FLAT will vary considerably depending on the size and nature of the assessment area. By far, the highest cost of the assessment will be the staff or contractor time needed to do fieldwork. Expenses will grow as the size of the assessment area, number of sites, and distances between sites increases. In addition, project planning, forest type mapping and MU delineation, as well as field supply purchases, contribute to the total cost. The cost per acre can be expected to

13


decrease as field crews and project coordinators become more experienced. Some of these relationships can be illustrated by comparing King County’s costs to those of municipalities that implemented the same or similar programs. Appendix 5 illus-trates the comparative costs in King County and the City of Everett where the same set of attributes data were collected.

Staffing and training— There are four main project responsibilities within the FLAT protocol: forest-cover-type mapping, field assessment, database design and management, and project management and planning. Those considering using FLAT should review what their in-house capabilities are for each of these activities, especially for more special-ized skills such as GIS and data processing, and decide what can be accomplished through training or seeking additional support. Training is most important for those doing field data collection and will help assure consistent and high-quality data. Training should include a walk-through of the protocols at a field site. In addition, it is helpful to pair new field staff with more experienced field staff, which helps to calibrate how estimates are done and improves consistency across field crews. The FLAT Field Manual provides guidance and detailed instructions that can support field staff training (see app. 6).

Equipment—The basic equipment needed to conduct a FLAT assessment include GIS, navigation devices, a field data entry system, plant identification resources, and measurement tools. The actual equipment that is needed or desired will depend on the size of the project and desired application of the collected data. For an equipment list and considerations, see appendix 3.

Timeline—In general, the FLAT timeline starts with a pre-field season to plan the project, pur-chase equipment as needed, train staff, and complete Phase 1 tasks (below). This is a critical time for FLAT projects to ensure that all components of a project are laid out adequately and that the plan for moving forward is understood by decisionmak-ers. The pre-field season is followed by one or more field seasons, most often but not necessarily during summer, to complete the field assessment (Phase 2). Fieldwork is the most time-consuming activity, so will differ based on condition and number of assessment sites. Lastly, post-fieldwork data management, analysis, and reporting should be estimated. Overall, the timeline will vary with the size and complexity of the project, and time requirements will decrease even for large projects as managers gain experience with FLAT implementation. To date, projects have ranged from a few months to a multiyear effort.

14


Phase 1: Forest Cover Type MappingDefining the project area— A FLAT project area is determined in several steps using a mapping system, management boundaries, and vegetation cover information. See appendix 1 for detailed procedures. Existing geospatial data are needed, including aerial imagery, parcel boundaries, and other management boundaries, to set up this process. For an explanation of necessary data, along with potential sources and considerations, see appendix 2.

Step 1: Determine which properties will be included in the assessment. This involves establishing initial site boundaries based on existing ownership and management boundaries.

Step 2: Designate land cover type, such as King County designations of forested, other ecosystems, natural (vegetation but <25 percent canopy cover), open water, hardscape, and cultural landscape.

Step 3: Delineate MUs based on forest species associations, geomorphic conditions, and land cover.

The resulting polygons will be the MUs, the unit of measure for the FLAT process. Each MU should be assigned a unique identifier to be used throughout the project. This is especially important on projects having multiple sites (e.g., dispersed parks or land holdings) that are then delineated into MUs. Keeping good records of MUs identifiers and associated data is critical to ensuring that FLAT information is easy to use during the Phase 3 analyses (discussed below), and for using FLAT to monitor changes over time.

Defining the data attributes—Field data collection is done using predefined attributes. The attributes are related to vegetation species, abundance, size and age classes, and identification of restoration opportunities. The list of attributes used during the King County pilot are included as appendix 4. Agencies or organizations using FLAT will have their own manage-ment priorities, so they may add a custom set of assessment attributes.

Defining the data attributes includes determining the level of measurement detail used during the field assessment. For example, forest overstory age classes may be divided into increments of 10 or 30 years. Selecting appropriate attributes and data-collection categories will require the expertise of people who are knowl-edgeable about the ecosystems being assessed, as well as those qualified to develop management priorities.

15


The attributes that are selected will shape the analysis that is done in Phase 3. Development of flowcharts and criteria that reflect potential management priorities should occur alongside data attribute selection. That is, managers should start the pro-cess of thinking through how their selected data attributes can be used to differentiate MUs and rank areas in order of urgency for management intervention. This can ensure that the most useful data for MUs classification are collected in the field (Phase 2). The actual flowcharts and criteria can then be refined and finalized as part of Phase 3.

Phase 2: Field AssessmentAssessment procedure—The field procedure entails visual estimates of the data attributes defined during Phase 1. Teams of two or more people visit each MU, walk through it, and record an estimate for each data attribute. The following summarizes key processes and considerations. The FLAT Field Manual, found in appendix 6, further describes the field data-collection methods.

Provide attribute estimates for the entire MU—The field team walks through each MU and records an estimated average attribute value for each of the variables se-lected in Phase 1. There may be patches of unusual species or conditions, but a best effort is made to estimate for the entire MU. Teams should be observant, perhaps even keeping notes of what they see. Judgments are made for each attribute and are entered into a data entry device before leaving the MU.

Use measurement tools to calibrate estimates—Some variables, such as dbh, crown closure, age, and height of regenerating trees (e.g., those less than 20 ft in height) can be measured with tools while in the field. It may be helpful to do this once or twice in an MU to calibrate the field team’s estimates. Determing the appro-priate level of accuracy is important, as excessive measurement will slow the rapid assessment process.

Alter MUs as needed—Once in the field and seeing on-the-ground conditions, the crew will consider whether the boundaries of the MUs should be altered, and if so, how. Pervasive differences in composition or age classes for large areas of the MU may necessitate either redrawing boundaries or splitting the MU into multiple, smaller MUs.

Record additional notes and variables—Field teams may find that something im-portant is in the MU that doesn’t necessarily fit into any of the attribute categories. Examples include unanticipated environmental or social hazards, such as homeless encampments or trail damage, which could supplement management decision crite-ria. This information can be captured in the “notes” field in the data entry tool.

16


Decide when to leave the road or trail system—Each attribute determination will apply to the entire MU, thus it is important that enough of the MU is observed by the field team. Sometimes this will require leaving trails and traveling on challeng-ing terrain. In other MUs, the view from a trail may be sufficient to make a judg-ment on most, if not all, of the attributes. Teams should be careful to recognize, and average into the MU estimate, any edge effect along trails where vegetation may have been influenced by trail activity or disturbance. If the area is large enough, the road or trail area may become a distinct MU. These decisions are left to the discre-tion of the field team; the desired speed of assessment must be balanced against the assessment’s accuracy.

Check for completeness—Sometimes MUs will be strangely shaped so that the ter-rain or trails will lead teams to travel in and out of a number of MUs. Data may be entered for each MU in order of discovery, but should be checked for completeness before leaving the MU.

Data management—Once the fieldwork is complete with MU attributes recorded using the data entry tools or field forms, the data should be checked and edited if needed, then entered into a database for analysis. The database type can be chosen according to the complexity of the project. For example, a small project could be managed with Microsoft Excel®, while a larger project with many different sites could be set up in Microsoft Access®, a server-based database management system, or a geo-database that includes the spatial information.1

The data from each field site can be compiled into one dataset across the project, then queried using the categories defined for each attribute. This system can help with data quality control, allowing managers to query data and check for discrepancies or abnormal values.

Phase 3: Management PrioritizationThe last phase of FLAT translates the results of the preceding mapping and field assessment efforts into valuable ecological information that can be readily under-stood by landowners, managers, and the general public. Once the summary and analysis processes are done, FLAT provides an advanced snapshot of conditions on the ground and evidence-based input toward determining land management goals and priorities.

1 The use of trade or firm names in this publication is for reader information and does not imply endorsement by the U.S. Department of Agriculture of any product or service.

17


Summary analysis—The data can be summarized to meet the specific needs of the user at various scales. FLAT provides the opportunity to summarize key characteristics about the parks or open space system, one project parcel within the system, or a subset of MUs chosen for a specific analysis.

Simple queries or formulas can be put to the database to report the number of MU acres representing attributes that were reported in the field data. These basic summaries can be reported as charts or maps (such as land-cover areas). Any attribute(s) can be displayed for the entire system or any subset within.

More specific analysis can be done with vegetation attributes, such as species composition, forest health indicators, invasive species, and so forth. Examples of summary analyses can range from a very basic percentage or total acreage of MUs containing a single vegetation attribute of choice, to a combination of attributes. For example, King County has produced summary statistics about the area of land occupied by different primary overstory species, the percentage of stocking classes found in young regenerating tree species, sites where a health threat indicator was identified, and a type categorization for each MU describing the primary species size and stocking.

There are numerous possible strategies for queries. One can query to show management subunits within the area of a larger parcel, or use one or more key attributes as criteria to identify individual MUs across the entire land base being assessed. The prioritization framework discussed below allows FLAT users to review their project area data and develop clear, ecological management priorities.

Prioritization analysis—Identifying and prioritizing areas in need of management is a key output of the FLAT process. Attributes assessed in the field are used to produce a qualitative value that combines two axes of a matrix. In King County, forest composition (y-axis) and forest threats (x-axis) were used (more detail is provided in the case study section that follows). The matrix combines the multi-attribute information to produce a classification value for each MU (e.g., fig. 4 is the matrix used for the King County project). This is the Tree-iage step of FLAT.

A flowchart can be used to determine MU values for the selected attributes. Generally, the resource composition flowchart and threat criteria are specific to each project’s conditions and priorities; the King County project is but one example (see fig. 8 on p. 25).

Resource composition values and threat values are determined, then combined using the matrix to produce classification values for each MU. This value represents how important taking action may be for a particular MU in relation to other MUs

18


within the site or larger system. The values suggest a ranking of priority for man-agement actions. For example, in figure 4, values 1 through 3 represent MUs with a tree composition that has high ecological value, and so are important to protect and maintain. Values 2 and 3 also represent the presence of a forest health threat and could be prioritized for restoration or maintenance. On the other end of the spec-trum, an MU with a value of 9 has a high threat presence and a lower quality tree composition, and therefore may not be as high a priority for management actions. A matrix value can be entered into the project database as an attribute for each MU and then mapped with GIS.

Based on the flexibility of this approach, other projects could use matrix tools that acknowledge project-specific ecosystems, threats, and management goals. As discussed in Phase 1, initial development of criteria and flowcharts should occur alongside data attribute development to ensure that desired data are collected in the field assessment. Then the flowchart(s) and criteria can be refined as needed to accurately reflect local conditions.

Threat

Valu

e

Fore

st c

ompo

sitio

n

Forest health

LOW

MED

IUM

HIG

H

LOW MEDIUM HIGH

1

4

7

2

5

8

3

6

9

Figure 4—Management unit classification matrix (Tree-iage) example, King County, Washington.

19


How the results of the matrix analysis are used in subsequent land management decisions is up to each FLAT user. FLAT provides an ecological input for land management decisionmaking. The combination of field data collection, flowchart processing, and classification of MUs using the matrix can be used to prioritize future management actions and monitoring. Resource managers will also consider other social and economic factors to create a strategic and feasible management approach for their lands. The following section details how FLAT was developed and tested in its pilot application in King County, an example of one way that FLAT has been used.

King County Parks and Open Spaces Case StudyThe King County Natural Resources and Parks Division, Parks and Recreation Division, deployed FLAT for a landscape-scale assessment of county parklands from 2010 to 2013. The implementation of FLAT in King County is an informa-tive case study that demonstrates the entire FLAT process from planning to final results. This case study presents the overall FLAT planning, implementation, and interpretation phases, as well as specific examples of park land assessment within King County.

The heterogeneity of forest lands within King County makes it an ideal case study. Although the west side of King County contains shoreline and pockets of urban forest within the large cities of Seattle and Bellevue, the east side consists predominantly of rural communities, agricultural lands, working forests, and wilderness. A long history of logging coupled with more recent rapid urbanization has created the diverse landscapes of today.

In 1989 and 2007, voters in King County approved measures that provided a combined total of more than $201 million toward open-space acquisition and improvement (Trust for Public Land 2012). In addition, King County (as part of its growth management policy) implemented an incentive program that allows 1 ac of land to be reassigned from rural to urban zoning if an associated 4 ac are dedicated to the county as permanent open space. The resulting rapid land acquisition has led to a diverse portfolio of more than 26,000 ac managed by King County. Holdings differ in terms of the level of previous use and surrounding development and the type of land cover, size, ecosystem type, and biotic composition. King County’s portfolio includes high-use active recreation parks, former agricultural lands and gravel mines, river floodplains, and working forests in the foothills of the Cascade Range, as well as Puget Sound coastline. More than 21,000 of the 26,000 ac are forested. Table 2 shows the number of acres and park sites per park classification.

FLAT was used to open the condition of nearley 25,000 acres of parks and open spaces, including nearly 200 parcels.

20


The large and rapid expansion of land holdings left King County managers with little information regarding forest composition, health, and diversity across the different park lands. As noted earlier, FLAT was developed and piloted in col-laboration with the Forest Service to provide a comprehensive forest assessment for King County Parks. Specific goals for the county included (1) assessing the timber resources and management needs within working forest lands, (2) assessing forest health and composition in forests managed for recreation and wildlife, (3) prioritiz-ing areas for forest stewardship and management, and (4) minimizing expenses.

Project Process Planning—King County received funding from the American Recovery and Reinvestment Act of 2009 (in coordination with the USDA Forest Service) to support job creation and training. These funds were used to support the implementation of FLAT over 3 years. The total budget for the project was about $194,000 and was used for FLAT development, field planning, fieldwork, database management, and data analysis. Planning occurred during 2009 and 2010, with fieldwork commencing in the sum-mer of 2010.

In total, 24,722 of the 26,000 ac of King County Parks were chosen for a FLAT assessment. These King County sites each contain a contiguous forest canopy com-ponent suited for a FLAT assessment. A FLAT assessment was performed on all of the multi-use, natural areas (lands managed to conserve and enhance ecological value and to accommodate passive recreational use), and working resource forest lands, which all are primarily heavily forested sites. Some recreation sites were assessed if they had a significant contiguous forest canopy. For example, Marymoor Park is a classic urban park with a traditional urban forest canopy of hundreds of single trees in a highly maintained park setting. Generally, these areas would be classified as “landscape” in the use of FLAT. However, Marymoor also has a significant riparian forest component along the river and lake that is classified as

Table 2—King County Parks distribution by designation, acreage, and number of sitesPark classification Area Number of parks

AcresWorking forest/resource 3,455 7Multi-use open space 12,812 41Natural areas 7,321 78Active recreation parks 2,472 67

Total 26,066 190

21


“forest” by FLAT guidelines. Regional multi-use trail corridors, such as the Burke Gilman Trail, were not assessed. Although most of these trails have a linear forest component, these were not considered to be a contiguous forest area, and would fall into the “landscape” category. Table 3 shows the land-cover-type results for the total King County FLAT project area across the different park designations.

Table 2—King County Parks distribution by designation, acreage, and number of sitesPark classification Area Number of parks

AcresWorking forest/resource 3,455 7Multi-use open space 12,812 41Natural areas 7,321 78Active recreation parks 2,472 67

Total 26,066 190Table 3—Acres of land-cover types in entire King County FLAT project area

Landcover type AreaAcres

Forested 21,058Natural 2,024Water 639Hardscape 48Landscape 953

Total 24,722

Phase 1: forest cover type mapping—Several years of orthophotos were used in developing the MUs, including 2009 true color imagery having 0.5 ft per pixel resolution collected specifically for King County; 2009 and 2001 USDA Natural Aerial Imagery Program (NAIP) imagery with 1 m per pixel resolution; and online public access oblique angle imagery. Multiple datasets were used to reduce errors resulting from shadows, parallax, photo mosaicing, and varying light conditions. In addition, raster files developed from 2003 LiDAR data collected over King County were used for assessing tree height in GIS.

Site boundaries were provided by the King County GIS Center in Esri shapefile format and included, at a minimum, a unique facilities identification number and a site name. Sites that were smaller than the minimum mapping size, or where the land cover appeared to be homogenous in the imagery, were classified as a single MU. Sites that contained multiple MUs were first delineated based on general land cover type. The land cover type was classified as forested, natural (vegetation with <25 percent forest canopy cover), water, hardscape, or landscape. Small and isolated areas of landscape and hardscape were combined with surrounding cover types, as it was not the intent of this project to track such features. The remaining MUs categorized as forest and natural were then viewed with the aerial imagery to further delineate polygons containing similar vegetation types. Color, texture, tree shadows, and crown shape were used to determine MU edge breaks. Because these characteristics can appear differently depending on the imagery used, several

22


available image sources were reviewed, as listed above. These provided views having different lighting, color balance, and resolution. A canopy height raster file generated from King County LiDAR ground and surface models was also used to provide guidance on stand heights.

McGarvey Park Open Space serves as an example of how Phase 1 was carried out in King County. The park was acquired by King County in 2000 through the 4:1 program, and is composed of 401 ac of forest (fig. 5). Photo interpretation was used to classify the park into 35 MUs as seen in figure 6, guided by differences in color and texture. For instance, MU 8 is a much lighter shade of green than the surround-ing MUs.

Figure 5—Boundaries of McGarvey Park Open Space.

23


Phase 2: field assessment—Specific forest measurements and observations were selected to meet the specific King County objectives of assessing timber resources and assessing the integrity and health of forest ecosystems. Attributes such as overstory tree diameter, stock-ing, and age were collected to evaluate the timber resource within the MU, while attributes such as native tree composition, invasive species presence, and forest health were collected to evaluate ecological integrity. Attributes were separated into broad categories that could be quickly estimated in the field. A full list of data attributes used in the King County FLAT is in appendix 4.

The ideal crew size was found to be two people (fig. 7). The learning curve for the fieldwork was minor, but it was essential that all field personnel be able to identify native plants and common nonnative invasive plants and noxious weeds.

Figure 6—McGarvey Park Open Space management unit delineation.

24


The crews carried both hardcopy orthophotos and digital versions on field global positioning system (GPS) units. Crews surveyed MUs that were clustered geograph-ically in the same day. To begin surveying an MU, crews would drive to the closest road access point. Some of the larger sites would require using multiple access points, while other sites had trail networks that allowed crews to easily move from MU to MU. Depending on the size of the MUs and the level of forest heterogeneity, a crew was able to survey between 3 and 20 MUs in one day. It was also important for crews to maintain their “ocular” calibration by measuring tree diameters and ages at least once on every site. Many of the forests in King County were clearcut in the last 100 years, and it was relatively reliable to assign most trees to the 50- to 99-year-old age class, but there are clusters of both younger and older trees. The FLAT Field Manual (app. 6) provides further information about the process.

The field campaign took three summer seasons to complete (2010–2012). Factors that affected the speed of the field crews were (1) heterogeneity of parks—parks with many separate MUs in a small geographic area took longer to survey than parks with a single forest type; (2) size of the park—larger parks required less driv-ing time to get from MU to MU; and (3) distance between parks—parks within a small geographic extent required less driving time. The forests of multiple MUs also tended to be more similar, resulting in less time spent calibrating ocular estimates.

Figure 7—King County field staff Jack Simonson and Brett Roberts consult management unit maps.

Dav

id K

imm

ett

25


Phase 3: management prioritization—Variables were selected to construct the matrix (fig. 4), which includes forest composition on the x-axis and forest health threats on the y-axis. The forest com-position values were defined prior to the fieldwork using forest ecology principles. In the field, teams determined a high, medium, or low forest composition value for each MU using a flowchart (fig. 8). This flowchart was designed to appropriately rate important landscapes that may not necessarily support a full forest canopy, particularly wetlands.

The y-axis of the Tree-iage matrix for King County includes low, medium, and high threat values. In the Green Seattle Partnership analysis noted above, the extent of invasive species cover was used as the criterion for determining threat values (i.e., low = <5 percent invasive cover; medium = 5 to 50 percent invasive cover; high = >50 percent invasive cover). In this King County case study, managers used additional multiple criteria to describe forest health threats, including the presence of root rot, mechanical tree failure, low tree vigor, presence of mistletoe, bare soil (as an indicator of disturbance), and an “other” category that allowed field teams to record additional observations. Each health indicator was recorded as a yes/no

Figure 8—Forest composition flowchart for King County management units.

All sites

<25% native tree canopy cover

>25% native tree canopy cover

0% of canopy is conifer

and/ormadrone

Unable to support>50%

conifer or madrone

cover

Able to support>50%

conifer or madrone

cover

Able to support

1 to 50% conifer or madrone

cover

Able to support>50%

conifer or madrone

cover

Capabilityto support

canopy

Modifiedtree

compositionvalue

Conifer/madrone

cover

Canopycover

>50% ofcanopy is

conifer and/ormadrone

Unable to support>50%

conifer or madrone

cover

Able to support>50%

conifer or madrone

cover

Unable to support

conifer or madrone

cover

Unable to support

conifer or madrone

cover

1 to 50% ofcanopy is

conifer and/ormadrone

HIGH

LOW

MEDIUM

26


observation. To determine if an MU had a low, medium, or high threat value, the number of observed health threat indicators were tallied back at the office. Low is interpreted as no observed health threats, medium as one observed threat, and high as two or more observed health threats (table 4). Other projects may develop a tiered approach, creating a decision tree akin to the forest composition flowchart to differentiate between low, medium, and high threat levels incorporating multiple types of indicators.

Table 4—Methodology used to select management unit threat value

Threat value Observed unhealthy forest valuesLow 0Medium 1High 2 or greater

ResultsAn ArcGIS® geodatabase was assembled from the field data. It was used to produce maps, conduct database queries, and generate summary statistics.

Summary analysis of forest characteristics—The forest variables collected using FLAT helped King County managers to understand baseline information about all of their forested parklands for the first time. Each of the forest attributes collected (app. 4) can be summarized across King County. For instance, the age class structure of the forest can be quickly viewed in figure 9. These data show that red alder (Alnus rubra bong.), bigleaf maple (Acer macrophyllum), and Douglas-fir (Pseudotsuga menziesii Pursh) dominate forest cover. Red alder and bigleaf maple are relatively short-lived species, and the large numbers of trees in age classes 2 (30 to 49 years old) and 3 (50 to 99 years old) suggest that these forests are in need of management and restoration.

Adding a third variable, stocking density, allows for creating type calls that can be useful for timber harvest planning and management (fig. 10). A type call is an integration of the major species (table 5), its size class, and the stocking class of that MU. Size class contains four categories: 1 = 0 to 5 inches dbh, 2 = 6 to 10 inches dbh, 3 = 11 to 20 inches dbh, and 4 = >21 inches dbh. Stocking also has four categories: 0 = <10 percent crown closure, 1 = 10 to 39 percent crown closure, 2 = 40 to 69 percent crown closure, and 3 = >70 percent crown closure. The FLAT data describes a hardwood-dominated forest structure at McGarvey Park (fig. 11).

27


Figure 9—The predominant overstory species and age class across all management units surveyed in King County using FLAT.

Acr

es

7,000

6,000

5,000

4,000

3,000

2,000

1,000

0

Age class 1 (0–29 years)

Age class 2 (30–49 years)Age class 3 (50–99 years)Age class 4 (100+ years)

Dominant overstory species

Alnus r

ubra

Acer m

acrop

hyllu

m

Pseud

otsug

a men

ziesii

Tsug

a hete

rophy

lla

Populu

s bala

miphera

Thuja

plica

ta

Abies a

mabilis

Picea s

itche

nsis

Fraxinu

s lati

folia

Betula

papy

rifera

Salix s

pp.

Arbutus

men

ziesii

Table 5—Major overstory speciesAcronym Scientific name Common nameCW Populus balsamifera L. ssp. tricocarpa (Torr. & A. Gray ex Hook.) Brayshaw Black cottonwoodRA Alnus rubra Bong. Red alderDF Pseudotsuga menziesii (Mirb.) Franco Douglas-firBM Acer macrophyllum Pursh Bigleaf mapleWH Tsuga heterophylla (Raf.) Sarg. Western hemlockWI Salix sp. Willow speciesSF Abies amabalis (Douglas ex Loudon) Douglas ex Forbes Pacific silver firRC Thuja plicata Donn ex D. Don Western red cedar

28


Prioritization analysis—Because both variables of forest composition and forest health represent important concerns in King County’s open space management, the matrix analysis immedi-ately suggests management priorities (with #1 indicating the highest quality forest condition and #9 the lowest). The number of acres of each prioritization value is shown in figure 12, and these results mapped across King County are shown in figure 13.

The matrix values for McGarvey Park are shown in figure 14. Note that a large proportion of McGarvey Park is classified as 5 and 6, indicating that there is a medium composition value with forest health threats present. For example, an MU having forest composition keyed out to medium, and having health threats of both root rot and low vigor (keyed to high in table 4) would be placed in the matrix of figure 13 as a “6.” At McGarvey Park, this translates to MUs that contain dominant red alder and bigleaf maple that are near the end of their productive lives, and have at least one area of conifers infected with root rot. The matrix value of an MU can inform decisions about the priority of action based on management goals.

Figure 10—Type calls identified within the King County FLAT that comprised more than 1 percent of the project by area. Note: The type call includes an acronym for the major species, the first digit represents the size class, and the second digit represents the stocking class.

0

2.0

4.0

6.0

8.0

10.0

12.0

RA33DF43

BM33RA23

DF33BM42

BM32RA32

WH33

DF23BM43

WH43

DF42RA22

RA13CW43

CW33

SF23DF32

RC43W

I21RA21

SF33CW42

Perc

enta

ge o

f all

man

agem

ent

unit

s

Type call

29


Figure 12—Priority matrix for King County parklands including number of acres of each ranked category.

Figure 11—Type calls identified within McGarvey Park Open Space by acres. Note: The type call includes an acronym for the major species, the first digit represents the size class, and the second digit represents the stocking class.

0

20

40

60

80

100

120

140

160

BM33BM32

WH33

DF33W

H22RC32

RC41DF42

Acr

es

Type call

Threat

Valu

e

Fore

st c

ompo

sitio

n

Forest health

LOW

MED

IUM

HIG

H

LOW MEDIUM HIGH

1

4

7

2

5

8

3

6

9

3,795 3,784 1,384

7,107 4,853 802

1,066 209 75

30


Management ImplicationsFLAT results were used immediately. For instance, in 2012, FLAT data guided forest management decisions for McGarvey Park, where old, dying, and diseased red alder and bigleaf maples were harvested for revenue. These areas were replanted with mixed conifers to meet forest restoration and future timber harvest goals. As FLAT data are updated during future monitoring phases, a shift should be seen in composition from older hardwood to vigorous conifers that return greater benefits. Note that to complete a timber sale, a supplemental timber cruise is required to establish the quantity of trees to be removed and subsequent value. More

Figure 13—Distribution of management units classified by Tree-iage matrix values for all King County parks.

31


information about how FLAT results were implemented can be seen in the McGar-vey Park Open Space Stewardship Plan (King County 2011).

FLAT results are of value across the county’s holdings, and will inform man-agement decisions for years to come. With many recent acquisitions of forest land, King County Parks is using FLAT results as baseline forest data for forest steward-ship strategies. For the 3,000 ac of designated working forests in King County, FLAT will support the creation or update of required forest stewardship plans. Of the other lands in the multi-use and ecological designations that do not require

Figure 14—Distribution of management units classified by Tree-iage tree composition and health threat values for McGarvey Park Open Space.

32


formal stewardship plans, FLAT will provide data to help King County identify specific MUs that need further attention because of forest health threats. Based on the field data collected through FLAT, King County has started to conduct addi-tional site-specific analyses, including actions that target areas of root rot. With the data now in hand, King County managers can continue to assess their lands and designate planning resources strategically.

ConclusionsThe purpose of FLAT is to provide a systematic survey of forested lands that supports the prioritization of future management actions in a landscape system. This information can be used widely to implement landscape-level planning, unit-specific management, and ecological knowledge building.

Importance in Land Management King County and the other municipalities that have implemented FLAT and FLAT-precursors have used the resulting data in a number of ways. Their primary use has been in the planning and budgeting for stewardship and restoration programs. A baseline understanding of the status of each MU gives natural resource managers the ability to estimate the costs of future action as well as determine where and what more detailed or site-specific assessments may need to take place. The rapid assessment process can also alert them to threats or land conditions they may not have anticipated or suspected.

The FLAT data can also be used in conjunction with other spatial data or infor-mation for decisionmaking. As an example, figure 15 shows how the Green Red-mond Partnership used assessment data along with social information to prioritize restoration activities (Green Redmond Partnership 2009). Using GIS, FLAT data can be combined with demographic, ecological, or environmental data to investi-gate trends and identify relationships across a landscape or public lands system.

FLAT was designed to purposefully support adaptive management. FLAT can be iteratively applied and combined with other assessments. Also, as repeated FLAT assessments are conducted and management is carried out, the repeated measures can provide managers with a way to observe the effectiveness of management actions. This application of FLAT has not yet been done in King County. In prin-ciple, by supporting longer term observations of patterns of ecosystem response, FLAT can provide more than just a snapshot survey. As FLAT is a practical and fairly low-cost tool, it may be redone on a periodic basis (perhaps every 5 years) and become a monitoring tool for management assessment and improvement. Because it is straightforward to learn and do, a variety of participants can be involved in a data-collection cycle, from staff to volunteers to local stewardship groups.

Local governments have used FLAT data to plan their forest stewardship and restoration programs.

33


FLAT FeasibilitySimplicity, flexibility, and low cost make FLAT an appropriate tool for many users. By assessing a limited set of variables and applying methods, including field protocols, that are easily teachable, FLAT makes forest assessment doable across a wide spectrum of landowners. Where forest condition information is lacking or piecemeal, FLAT can fill the gap and provide consistency without requiring signifi-cant time, training, or large equipment purchases.

In addition, the FLAT framework is inherently flexible for use on a wide range of land ownerships, from small and contained to large and complex. For any given project, there is also flexibility in how detailed to make the field assessment, match-ing procedures to the size of the forest and intended use of the data.

The simplicity and flexibility of FLAT allows it to be implemented within a tight budget. This is a crucial factor that makes FLAT feasible for cash-strapped local government, private, and NGO landowners. In addition to the reduced time, training, and equipment that normally make assessment more expensive, costs can also be kept low by using the growing interest in citizen science. Where appropri-ate, projects can support community involvement by using volunteers to conduct fieldwork. Project planning should then include training costs and field checks for data reliability.

Currenthigh-value

forestcomposition?

Expressedcommunity

importance orcritical area?

Geographicdistribution?

Priority site:Create work plan; begin

restoration andmaintenance

Yes Yes Yes

Volunteer interest or available Forest Steward?

Not a priority site at this time

No No

No

NoYes

Figure 15—Green Redmond Partnership decision tree for prioritizing restoration sites.

34


StewardshipThe landscape assessment of King County’s parks and open spaces revealed that there are natural systems that in some instances are robust and healthy and, in other cases, are in decline or face substantial threats. Ongoing forest resource management is needed across all these conditions, both to restore compromised landscapes and sustain those that are healthy or moving to an improved condition. Forest and landscape management is an intensive activity, involving strategic planning, technical and scientific input, and on-the-ground actions. To carry out these activities, a resource management agency can use its own staff or contracted consultants to do land-based work. Yet fiscal constraints in local government and environmental resource organizations can limit their capacity to address ecosystem needs and recovery.

In the face of limited and declining fiscal and technical resources for ecosystem management, communities and agencies must consider new solutions to restore and sustain natural systems, particularly in urban settings (Wolf et al. 2013). Engage-ment of people and social systems, from individuals to organizations, is another stewardship option (Wolf 2012). Environmental stewardship is an increasingly com-mon response to the decline of natural systems in and around built environments (Romolini et al. 2012). Recent research in multiple U.S. cities suggests that steward-ship may be an effective and viable strategy for ecosystem management (Svendsen and Campbell 2008), particularly in urbanized areas.

There is a great deal of variability in the organizational and administrative structures that support stewardship efforts. At one level, programs and projects may be activated by landscape-scale policies and associated regulations that are promulgated by community planners, agency officials, and policy decisionmakers (Brunckhorst 2002). Other stewardship groups are composed of citizens organized to address a defined ecosystem condition that has direct personal consequences, acting for change through place-based projects and resources. Some groups are formally self-organized and have 501(c)(3) status, some are informal organizations without legal status, and some are membership organizations facilitated by a public entity (Brinkley et al. 2010).

No matter who participates in stewardship, and whatever their motivations, better landscape health outcomes can be promoted using the information provided by a landscape assessment like FLAT. For instance, having full knowledge of parcel conditions can help a stewardship coordinator within an agency to prioritize stew-ardship programs across a landscape system, to ensure that the greatest work effort is assigned to places having the greatest need. In addition, those nonprofit orga-nizations that host stewardship projects can use an assessment to assign volunteer

Stewardship organizations and volunteers can assist with FLAT data collection, contributing to better management of urban forests.

35


work parties within their activity area, as well as communicate to the public about the high-health parcels that may deserve conservation status. Finally, a “friends of” group of citizen stewards can use assessment information to identify the most important project sites within a park or open space within their community, so that they achieve satisfying outcomes from their self-sufficient efforts. In all these situ-ations, periodic reassessment can help the agency, organization, or group to better understand how their efforts contribute to better ecological health.

Limitations FLAT is a useful tool for planning, budgeting, stewardship, and performance measurement. However, its application does have limitations that are important to consider. The qualitative nature of the field assessment makes the FLAT unsuitable as evidence of compliance with environmental standards or the creation of envi-ronmental impact statements under state or federal law. FLAT would also not be sufficient to develop a sustainable harvest plan, though it can serve as a first step to get a picture of resources on the land.

Another limitation of FLAT is the coarse-grain scale of the data. Management Unit sizes may limit use of FLAT as a tool for understanding fine-scale ecological structures and processes. Most MUs will be larger than the level at which some plant interactions take place. Attempts to scale the MU down to small land areas will greatly increase the number of MUs, increasing cost and time of the assess-ment. FLAT, therefore, does not take into account small pockets or sites within the unit. In theory, a unit with a number of concentrated pockets of invasive species would appear the same as a unit within which a similar proportion of invasive species are found throughout a more dispersed area. Although a classification of heterogeneity could be added, a more detailed assessment will likely be required to understand site-specific details and evaluate what type of action may be neces-sary within an MU. The purpose of FLAT is therefore to determine where these detailed assessments may need to take place.

While FLAT is limited in spatial scale, it is also limited in its scale of possible indicators. As more indicators are added, the speed and simplicity of the assess-ment—two of the primary advantages—will be reduced. For FLAT to retain these qualities, it may be necessary to exclude certain variables or indicators that may be useful in specific management situations. The current FLAT protocols and analysis are a reflection of both the priorities of King County as a pilot user and the need for FLAT to retain its rapid deployment character. Project managers will have to carefully weigh the tradeoffs of including more indicators based on their specific circumstances.

36


In addition, the current FLAT protocols do not include wildlife. If managers have a specific species or cohort of species in mind, attributes could be added that indicate habitat suitability or presence.

Next StepsFLAT is a relatively new tool, and as such, has potential for future development, testing, and refinement. Formal testing would enhance the credibility of FLAT. Although it has already been informally tested through application to the satisfac-tion of practitioners, neither its reliability nor accuracy has been systematically evaluated. A reliability test might entail examining whether individuals with similar training and experience using FLAT on the same MU would produce similar results. An accuracy test might compare the results of a FLAT assessment for particular MUs with evaluations that involve random sampling and more traditional forest ecological data collection within those MUs.

As it currently exists, FLAT is designed for use in the lowland forests of the Puget Sound basin and, to a limited degree, in wetlands. Expanding its use to other ecosystem types in other parts of the Pacific Northwest, the Nation, or the world would require revising the measures variables. In principle, however, the basic framework should work for any number of ecosystem types. As discussed in the “FLAT Methodology” section, the system is oriented to a project-specific selection of prioritized data attributes and criteria, in order to best achieve local management goals. Expanding FLAT’s ecosystem range is thus a matter of degree, requiring additional work to develop appropriate data attributes and field test new implemen-tations. Concerns about different land scales and vegetation associations would have to be addressed.

One way that this extension of FLAT might be facilitated would be the develop-ment of a publicly available clearinghouse that catalogs indicators, classification schemes for each indicator, and prioritization systems that have been developed for various ecosystem types. Practitioners planning a FLAT project could then devise a system built on the experiences of others, just as those in the Pacific Northwest can use the King County example as a starting point.

Lastly, future research may explore applying FLAT as a practical method of monitoring ecosystem changes not necessarily tied to natural resource management goals. For example, decisionmakers interested in more general information regard-ing emerging local effects of climate change, or changes in habitat quantity or quality, or shifts in biodiversity, or any number of other topics, may adapt FLAT to focus on such system-level questions.

Although used for forest assessment in this report, FLAT can be adapted for assessment of every ecosystem type in the Pacific Northwest region.

37


AcknowledgmentsThis work was developed by the Green Cities Research Alliance (GCRA). In 2009, GCRA was initiated by the USDA Forest Service, Pacific Northwest Research Station, to build a program of research about urban ecosystems in the Puget Sound region. The GCRA is an integrated social-ecological research program that engages the social and biophysical sciences to meet the practical needs and concerns of local organizations and agencies. It also is an effort to coordinate science and community partners within the Pacific Northwest region and to link investigations to other U.S. urban areas. The goal of this collaboration is to increase the knowledge necessary to build healthy, sustainable urban environments. The GCRA pairs scientists with practitioners and local decisionmakers to co-design and implement research efforts that provide relevant and practical information. Major collaborators include the University of Washington, King County, Forterra, and the City of Seattle. For more information, visit http://www.fs.fed.us/pnw/research/gcra.

The GCRA’s work on FLAT has been particularly informed and supported by the Green Seattle Partnership, with special thanks to Mark Mead for his efforts on developing the Tree-iage analysis tool. The Green Everett Partnership, Forterra, and American Forest Management have also been instrumental in sharing the FLAT analysis process and database. Additional thanks and acknowledgement goes to Ara Erickson (Forterra, now with Weyerhaeuser Company), Kim Frappier (Forterra, now with the City of Mercer Island), Bill Loeber (King County), and Ted Hitzroth (American Forest Management) for their help in developing FLAT. Project field implementation was provided by the King County Department of Natural Resources and Parks, led by Brett Roberts and Jack Simonson.

Major funding for this project was provided by the American Recovery and Reinvestment Act of 2009 (ARRA) and the USDA Forest Service Pacific Northwest Research Station.

38


ReferencesBrinkley, W.; Wolf, K.L.; Blahna, D.J. 2010. Stewardship footprints and potential

ecosystem recovery: preliminary data for Seattle and Puget Sound. In: Laband, D., ed. Emerging issues along urban/rural interfaces III: linking science and society (Proceedings). Atlanta, GA: 24–30. http://www.naturewithin.info/CivicEco/InterfacesIII%20Prcdngs_GCRA.Dec2010.pdf. (December 17, 2015).

Brunckhorst, D. 2002. Institutions to sustain ecological and social systems. Ecological Management and Restoration. 3(2): 108–116.

Clewell, A.F.; Aronson, J. 2007. Ecological restoration: principles, values, and structure of an emerging profession. Washington, DC: Island Press. 232 p.

Dean, J.W., Jr.; Sharfman, M.P. 1996. Does decision process matter? A study of strategic decision-making effectiveness. Academy of Management Journal. 39(2): 368–396.

Green Redmond Partnership. 2009. 20-year forest management plan. Redmond, WA: City of Redmond. http://forterra.org/wp-content/uploads/2015/05/FINAL_GRP_20_YP_InDesign_FORWEB.pdf. (December 17, 2015).

Green Seattle Partnership. 2006. 20-year strategic plan. Seattle, WA: Cascade Land Conservancy and Seattle Parks and Recreation. http://greenseattle.org/ wp-content/uploads/2015/04/GSP_20YrPlan5.1.06.pdf. (April 25, 2016).

Iserson, K.V.; Moskop, J.C. 2007. Triage in medicine, part I: concept, history, and types. Annals of Emergency Medicine. 49(3): 275–281. doi:10.1016/j.annemergmed.2006.05.019.

King County. 2011. McGarvey Park Open Space forest stewardship plan. Seattle, WA: King County Department of Natural Resources and Parks, Water and Land Resources Division. http://your.kingcounty.gov/dnrp/library/2011/kcr2230.pdf. (December 17, 2015).

Marsh, W.M. 1978. Environmental analysis: for land use and site planning. New York: McGraw-Hill, Inc. 292 p.

Nowak, D.J.; Crane, D.E.; Stevens, J.C.; Hoehn, R.E.; Walton, J.T. 2008. A ground-based method of assessing urban forest structure and ecosystem services. Aboriculture & Urban Forestry. 34(6): 347–358.

Randolph, J. 2011. Environmental land use planning and management. Washington, DC: Island Press. 776 p.

39


Romolini, M.; Brinkley, W.; Wolf, K.L. 2012. What is urban environmental stewardship? constructing a practitioner-derived framework. Res. Note PNW-RN-566. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 41 p.

Sayre, R.; Roca, E.; Sedaghatkish, G.; Young, B.; Keel, S.; Roca, R.; Sheppard, S. 2000. Nature in focus: rapid ecological assessment. Washington, DC: Island Press. 202 p.

Svendsen, E.S.; Campbell, L.K. 2008. Urban ecological stewardship: understanding the structure, function and network of community-based urban land management. Cities and the Environment. 1(1): 5.

Trust for Public Land. 2012. TPL LandVote database. https://tpl.quickbase.com/db/bbqna2qct?a=dbpage&pageID=10. (December 17, 2015).

Turner, M.G. 2005. Landscape ecology in North America: past, present, and future. Ecology. 86(8): 1967–1974.

U.S. Department of Agriculture, Forest Service [USDA FS]. 2014. The forest inventory and analysis phase 3 indicators database 6.0: description and user guide. Washington, DC. 203 p. http://www.fia.fs.fed.us/library/database-documentation/. (December 17, 2015).

U.S. Department of Agriculture, Forest Service [USDA FS]. 2015. Urban field data collection procedures. The Forest Inventory and Analysis database: database description and user guide for urban data. Washington, DC. 464 p. http://www.fia.fs.fed.us/library/database-documentation/urban/current/FIADB_user%20guide_urban_10_2015.pdf. (December 17, 2015).

U.S. Department of the Interior, National Park Service [USDI NPS]. 2009. NPS natural resource condition assessments (NRCAs): standards and guidelines. Washington DC. https://www.nature.nps.gov/water/nrca/assets/docs/NRCA_Standards_and_Guidelines_Sept2009.pdf. (December 17, 2015).

Wolf, K.L. 2012. The changing importance of ecosystem services across the landscape gradient. In: Laband, D.N.; Lockaby, B.G.; Zipperer, W., eds. Urban-rural interfaces: linking people and nature. Madison, WI: American Society of Agronomy/Soil Science Society of America/Crop Science Society of America: 127–146.

Wolf, K.L.; Blahna, D.; Brinkley, W.; Romolini, M. 2013. Environmental stewardship footprint research: linking human agency and ecosystem health in the Puget Sound Region. Urban Ecosystems. 16: 13–32.

40


Appendix 1: How to Develop the Assessment Area and Management UnitsThe forest-cover-type mapping process defines the boundaries and area for each management unit (MU). This process establishes boundaries based on existing ownership, management goals, and vegetation and land-coverage typing from aerial or satellite photos. This section describes the steps of this process.

STEP 1: Determine Which Properties Will Be Included in the Assessment The project area should be defined in this stage based on relevant management programs and resource information needs (see fig. 16 for an example). It may help to consider if different sites will be managed by different crews or be located in different districts that will affect how management decisions are made, and may change desired assessment boundaries. Thought should also be given to regulatory designations such as wetland or riparian areas that may limit or specify manage-ment actions. Each site should have a unique name and number for identification throughout the project. For small projects, there may only be one site, having a single project area name and boundary.

STEP 2: Designate Land Cover TypeOnce the assessment sites have been chosen and identified by property ownership or management blocks, the next step is to use aerial or satellite imagery to review, then delineate similar conditions within the predesignated site or project areas. The general purpose of this photo-typing is to identify zones based on the major differ-ences in land or forest cover.

Sites should be divided into units that are given one of a number of broad land cover designations. The land cover designations for the King County pilot project are included below in table 6. Some Forest Landscape Assessment Tool (FLAT) users may wish to expand these designations to include types such as wetlands or shoreline, given the particular characteristics of the system they are evaluating.

The boundaries between land cover types should be based on breaks visible in the aerial or satellite photos. It is not necessary to precisely measure the for-est canopy cover to differentiate between, for example, a forested or natural area designation. An estimate should be fine in these cases because the typing will be verified during field data collection (see the FLAT Field Manual in app. 6). The designations can be done largely based on viewing the photos, but some may choose to use software or tools like Feature Analyst.

When in doubt about a land cover type, a convention should be established and applied consistently throughout the process. In some cases, it may make sense to

41


create a new category that indicates the call will be made in the field assessment process. The conventions will likely change based on the team and their priorities, thus it is essential to make assumptions explicit so that users of the information later on will take that into consideration as they make decisions.

A threshold should be set for the minimum size of a land cover designation. In King County, significant and distinct nonforest areas were delineated down to 0.1 ac. It is not the intent of FLAT to map paved paths, tennis courts, or other facilities within site areas. These smaller developed features can be lumped into surrounding landscape or hardscape MUs.

Figure 16—Forest Landscpe Assessment Tool assessment boundary for Maury Island Marine Park, a project site in King County, Washington.

42


STEP 3: Delineate Management Units Based on Vegetation Differences.With property information and land cover differences delineated, the next step is to further divide a site according to vegetation differences (fig. 17). This applies specifically to land cover designations that fall within the purview of FLAT (i.e., not open water, hardscape, or landscape).

The forested and natural cover typing of Step 2 should be reviewed using aerial imagery to delineate polygons within that contain similar vegetation types. Areas with clearly different species, structural, or age composition should be placed into different MUs. Color, texture, tree shadows, and crown shape can be used to determine MU breaks. Because these characteristics can appear differently on different imagery, use of several image sources is recommended to provide views with varied lighting, color balance, and resolution (see app. 2 for further discussion of necessary data). This vegetation typing requires the most skill and experience with photo-typing and boundary delineation. In addition, if data are available and project staff have the needed expertise, remote-sensing ground and surface datasets (e.g., LiDAR) may be used to generate a canopy height layer for use in a geographic information system (GIS) to assist in differentiating stand heights.

The MU size will vary based on the amount of contiguous similar vegetation coverage as well as thresholds decided on by the FLAT project manager. As an example, King County established a minimum MU size of 5 ac, as it was deter-mined that further subdivisions of MUs would add too significantly to the total number of MUs and time necessary to complete field data collection. This is a judg-ment call—what you may gain in precision, you may sacrifice in time. In smaller communities, cities, or systems of parcel holdings, it may make sense for MUs to be much smaller.

Table 6—King County Parks Forest Landscape Assessment Tool land cover designations

Type Acronym Definition Forested FOR ≥25 percent of the area covered by forest canopy

Natural NAT Any natural vegetation that has <25 percent forest canopy cover

Open water WAT Open water without woody vegetation

Hardscape HS Impervious surface such as parking or buildings

Landscape LS Areas that are currently landscaped or have been mechanically maintained within the past year

43


Data format—If you are using digital GIS software to delineate management units, an attribute field should be created within the polygon layer so that each MU polygon feature has a unique identifying number. This unique identifier should reference both the site number and the MU number within the site. Table 7 shows an example of this as the FID_MU where the first four digits are from the site name and the two digits after the dash specify the MU within the site. It is important to track MUs in this precise manner as fieldwork may call for merging or splitting MUs, resulting in new polygons within a site that require careful data management updates. The MU

Figure 17—Delineated management units within Maury Island Marine Park, King County, Washington.

44


should also have attribute fields for its land cover designation and site, as well as for any important administrative or legal boundaries. This georeferenced attribute table will form the basis of data entry for fieldwork, and into future assessments. Table 7 is an illustration of what this attribute table might look like at the end of this process.

Again, for small-scale projects, there may be only one “site,” a single property, and, in that case, MU identifiers can be a simple sequential numbering system.

Without GIS software, the MU identifying number should be written directly onto the hardcopy aerial photos used to draw the MU boundaries. The same table should be created in a digital form such as Microsoft Excel or Access.

Table 7—A sample attribute table after type mapping and data collectionFLAT_MUFID_MU SITE_NAME ACRE LAND_COV ASPECT SLOPE AGE_CLASS OVR1_SPC OVR1_SIZE4601-01 Sugarloaf

Mountain Forest

10.6 FOR 72 54 2 PSME 3

4601-02 Sugarloaf Mountain Forest

273.6 FOR 165 34 1 ALRU 1

4632-01 Mirrormont Park

4.9 FOR 215 4 3 ALRU 3


4.2 FOR 215 5 3 PSME 4


1.7 LS 259 7

4634-01 Fall City Natural Area

2 WAT 196 2

4634-02 Fall City Natural Area

3.9 NAT 194 0 1 ALRU 2

45


Appendix 2: Necessary Data Aerial ImageryAerial photographs or satellite imagery are important sources of information for initial management unit (MU) delineation and mapping. There are several consider-ations that will help determine your best source of data: • If you are using geographic information system (GIS) software, it is

important that the imagery used be geometrically corrected, or “orthorec-tified,” and geographically referenced so that boundaries drawn using the photos will correspond to the correct places and distances on the surface of the earth.

• Resolution should be high enough to distinguish between different cover-age types, and color makes this interpretation much easier. If tools such as Feature Analyst are being used to aid in the photo typing, there may also be resolution thresholds or requirements.

• Photos should be taken during the summer, when vegetation from both deciduous and evergreen trees and shrubs are visible, as vegetation types will be used to define MU boundaries. In some cases, having multiple image sets at different times of the year will be helpful.

For the King County pilot project, several sets of aerial imagery were used, in part to be able to compare conditions. These included: • King County 2009 natural color ortho photos in 0.5 to 1 ft resolution.

These data displayed significant parallax along mosaic edges, which made typing larger areas more difficult.

• Arc® Online imagery. Arc images can change without notice and there is no control over color balance, and the radiometric settings are more likely to have rectification errors. Overall, there are fewer paral-lax and mosaic issues.

• U. S. Department of Agriculture National Aerial Imagery Program (NAIP) 1 m orthophotos.

ParcelsParcel data are important for delineating MUs. Aerial and satellite imagery can be used to type by natural conditions (such as vegetation cover, or surface water elements), but parcel boundaries are often the basis of management programs. Because the Forest Landscape Assessment Tool (FLAT) is a tool intended to sup-port management decisions, MUs should be informed by ownership and manage-ment jurisdictions. In most cases, parcel data will be the source of legal boundary delineation. Some considerations:

46


• Parcel data should include attributes that indicate ownership or manage-ment goals for each feature. At the very least, it is important to determine if a parcel is managed by the organization conducting the FLAT assessment.

• Note that assessor parcel GIS layers often display approximate boundaries. Locations do not always match imagery and other layers. This is usually adequate for the level of mapping used for FLAT.

Other Management Boundaries• Sometimes there may be other management program boundaries that

should be taken into consideration when creating MUs.• If there are existing stewardship plans in effect or additional information

about a specific area, it may make sense to classify MUs to accommodate preexisting documents or delineations.

• Management systems will often be divided into subdistricts or areas, each having their own programs or assigned management staff. This may also be important information for dividing MUs and should be considered along with the parcel and ownership data.

• Legal designations that limit management actions or have additional regula-tory requirements, such as riparian zones or wetlands, should also be indi-cated when delineating MUs.

47


Appendix 3—Equipment List Traditional forestry tools and equipment are used in a Forest Landscape Assessment Tool (FLAT) assessment (table 8) to help field crews calibrate their visual estimates. Depending on the ecosystem type, management goals, and complexity of other FLAT projects, this list can be modified to incorporate additional or different tools. The most expensive piece of equipment is the global positioning system (GPS) unit. In lieu of a costly GPS unit, most smartphones have GPS capabilities and can be programmed with a background map. If a GPS unit or smartphone is unavailable or not desired, a FLAT assessment can be implemented with a hardcopy paper map and data sheet.

Table 8—King County, Washington, equipment listItem Considerations King County pilot projectGeographic information system (GIS) Having expensive GIS software such as

ArcGIS® in which to track and query information is convenient for conducting FLAT but probably not necessary

ArcGIS 10.0

Free open source tools such as GrassGIS or even free “cloud-based” tools such as Google Earth® or Google Fusion Tables

High-resolution orthophotos, maps and handwritten tables could qualify as a GIS if necessary

Paper map Boundaries of each management unit should be overlaid and labeled on the map

Yes

Camera Photos of forest conditions are very helpful Yes

Compass Good for orientation YesGlobal positioning system (GPS) A smartphone could also be used in some cases

instead of a GPS unitTrimble Geo XT

Data entry tool Microsoft Office® is compatible on most smartphones as well

Trimble Geo XT® with Microsoft Windows Mobile®

Paper data forms Good for backup in case electronic devices are not working

Yes

Plant identification reference books Very important that field crews are able to identify native and nonnative plants

Pojar and MacKinnon (2004)

FLAT Field Manual This provides a quick reference to the data-collection procedures

Yes (early draft)

Diameter tape For estimate calibration YesClinometer For estimate calibration YesIncrement Borer For estimate calibration YesDensiometer For estimate calibration NoFLAT = Forest Landscape Assessment Tool.

48


Appendix 4: King County Data Attributes Table 9 displays the data attributes included in the King County Parks pilot project. Attributes were chosen that described forest characteristics considered most rel-evant for setting land management priorities and subsequent development of land management plans for forested parks and resource lands in the county.

Table 9—King County, Washington, data attributes (continued)Data attribute Data field Explanation/criteria Site name SITE NAME GIS identifierManagement unit number HMU_NO GIS identifierDate of data collection DATEAssessors initials CREWLand cover LANDCOV Forested FOR ≥25 percent forest canopy Natural Area NAT <25 percent forest canopy Open Water WAT No woody vegetation Hardscape HS Buildings, parking Landscaped LS Landscaped, mechanically maintainedAge class AGECLASS Category 1 1 0 to 29 years Category 2 2 30 to 49 years Category 3 3 50 to 99 years Category 4 4 100+ yearsOverstory species OVR1_SPC Overstory species, most abundant dominant or

codominant >20 ftOverstory size OVR1_SIZE Overstory dbh size classCategory 1 1 0 to 5 inches dbhCategory 2 2 6 to 10 inches dbhCategory 3 3 11 to 20 inches dbhCategory 4 4 21+ inches dbhSecond overstory species OVR2_SPC Second overstory species, in order of abundance

codominant >20 ftSecond overstory size OVR2_SIZE Overstory dbh size class, see size class chart aboveThird overstory species OVR3_SPC Third overstory species, if present, in order of abundance

codominant >20 ftThird overstory size OVR3_SIZE Overstory dbh size class; see size class chart aboveStocking STOCKING Crown closure estimate, as viewed directly aboveCategory 0 0 Less than 10 percent crown closure Category 1 1 10 to 39 percent crown closure Category 2 2 40 to 69 percent crown closureCategory 3 3 70+ percent crown closure Management unit composition HMU_CMP

49


Table 9—King County, Washington, data attributes (continued)High composition H >50 percent conifer/madrone; or

≤50 percent conifer/madrone with no capacity for restoration (includes wetlands)

Medium composition M 1 to 50 percent conifer/madrone with capacity to support restoration to H; or

<25 percent native cover with capacity to restore up to 50 percent conifer

Low composition L < 25 percent native cover with capacity for full restoration planting; or

No conifer/madrone with capacity for full restorationLow vigor LOW VIGOR Conifer: live crown ≤40 percent? Y or N

Hardwood decline: top dieback or snags ≥5 percent? Y or N

Mechanical tree failure FAILURE Mechanical tree failure in ≥1 percent of MU, Y or N (e.g., windthrow, landslide)

Root rot ROOT ROT Root rot pockets present? Y or NMistletoe MISTLETOE Mistletoe present? Y or NBare soil BARE SOIL ≥1% bare soil present from recent disturbance, erosion,

etc.? Y or NOther OTHER Present in ≥1 percent of MU? Y or N (*note in

comments*)Regeneration species RGN1_SPC Regeneration species <20 ft height, in order of abundanceSecond regeneration species RGN2_SPC Regeneration species <20 ft height HT, in order of

abundanceRegeneration stocking class RGN_TPACategory 1 1 0 to 49 TPA (>30 ft spacing)Category 2 2 50-149 TPA (between 30 and 16 ft spacing)Category 3 3 150+ TPA (<16 ft spacing)Plantable space PLANTABLE Suitable growing space for restoration planting? Y or NNative shrub and herb species GRD1_SPC Native shrubs and herbs, most abundantNative shrub and herb species GRD2_SPC Second native shrubs and herbs in order of abundanceInvasive species INV1_SPC Nonnative species, most abundantInvasive species INV2_SPC Second nonnative species in order of abundanceInvasive species INV3_SPC Third nonnative species in order of abundanceInvasive species INV4_SPC Fourth nonnative species in order of abundanceInvasive species INV5_SPC Fifth nonnative species in order of abundanceTotal invasive cover INVCOV Total invasive cover High cover H > 50 percent Medium cover M 5 to 50 percent Low cover L <5 percentNotes NOTES Unique site conditions and other dominant trees present

50


Table 10—Comparative cost estimates of implementing Forest Landscape Assessment Tool (FLAT) in the city of Everett and in King County, Washington

Assessment attributes King County parks Everett parksTotal project area (acres) 24,724 605Total time Three summer field seasons Two weeksCost in dollars: Staff 111,000 6,000 Consultant 72,000 15,000 Travel 7,400 200 Supplies/materials 3,600 200 Total cost 194,000 21,400

Appendix 5—Cost AnalysisTable 10 compares the estimated costs associated with carrying out the FLAT assessment in two park systems, those of King County and the City of Everett, Washington. Staff costs include project management and fieldwork, while consul-tant costs include project development and phototyping, as well fieldwork during the first field season. For the King County Forest Landscape Assessment Tool (FLAT), the staff spent an estimated 2,880 hours implementing the project. The difference in supplies and materials costs can be explained by King County’s fieldwork startup costs, whereas the Everett project made use of equipment owned by consultants.

51


Appendix 6— Forest Landscape Assessment Tool (FLAT) Field ManualThis field manual was prepared by the project partners to provide a concise proce-dures manual for reference in the field, and to provide information for field crew training. Readers can cross reference the procedures described in this technical report with this appendix. In addition, a .pdf file of the manual can be downloaded at http://www.naturewithin.info/UF/FLAT_Field_Manual.pdf.

FLAT

Fie

ld M

anu

al

The

Fore

st L

ands

cape

As

sess

men

t Too

l

GREEN

CITIES

RESEARCH

ALLIANCE

Dep

artm

ent o

f Nat

ural

Res

ourc

es a

nd P

arks

Park

s and

Rec

reat

ion

Div

isio

n

Ack

now

led

gem

ents

The

Gre

en

Cit

ies

Res

earc

h

All

ian

ce

was

in

itia

ted

by

the

USD

A F

ore

st S

ervi

ce P

acif

ic

No

rth

wes

t R

esea

rch

Sta

tio

n i

n 2

009

to b

uild

p

rog

ram

of r

esea

rch

ab

ou

t urb

an e

cosy

stem

in

the

Pug

et S

ou

nd

reg

ion

. GC

RA

pai

rs s

cien

tist

s w

ith

pra

ctit

ion

ers

and

loca

l dec

isio

n m

aker

s to

co

-des

ign

an

d im

ple

men

t res

earc

h e

ffo

rts

that

p

rovi

de

rele

van

t an

d p

ract

ical

in

form

atio

n.

The

Fore

st L

and

scap

e A

sses

smen

t To

ol

was

d

evel

op

ed b

y th

e fo

llow

ing

par

tner

s:

K

ing

Co

un

ty P

arks

an

d N

atu

ral R

eso

urc

es

U

SDA

Fo

rest

Ser

vice

PN

W R

esea

rch

Sta

tio

n

A

mer

ican

Fo

rest

Man

agem

ent

Fo

rter

ra

U

niv

ersi

ty o

f Was

hin

gto

n

For

mo

re in

form

atio

n, v

isit

w

ww

.fs.fe

d.u

s/p

nw

/res

earc

h/g

cra

Fun

din

g f

or

this

pro

ject

was

pro

vid

ed b

y th

e 20

09 U

nit

ed S

tate

s A

mer

ican

Rei

nve

stm

ent

and

Rec

ove

ry A

ct (

AR

RA

) co

ord

inat

ed b

y th

e U

SDA

Fo

rest

Ser

vice

Pac

ific

No

rth

wes

t (P

NW

) R

esea

rch

Sta

tio

n.

In a

ccor

danc

e w

ith F

eder

al la

w a

nd U

.S. D

epar

tmen

t of A

gric

ultu

re p

olic

y, th

is in

stitu

tion

is

proh

ibite

d fr

om d

iscr

imin

atio

n on

the

basi

s of

race

, col

or, n

atio

nal o

rigin

, sex

, age

or d

isab

ility

. U

SDA

is a

n eq

ual o

ppor

tuni

ty p

rovi

der a

nd e

mpl

oyer

.

Co

nte

nts

Proj

ect B

ackg

roun

d

2

Sum

mar

y of

the

FLA

T Pr

oces

s

3

Get

ting

Sta

rted

4

Qua

lity

Con

trol

Met

hods

6

Fore

st C

over

Typ

e M

appi

ng

7

Und

erst

andi

ng M

anag

emen

t Uni

t Del

inea

tion

9

Intr

oduc

tion

to F

ield

Pro

cedu

res

10

Dat

a C

olle

ctio

n D

efin

itio

ns a

nd P

roce

dure

s

12

App

endi

x

24

Refe

renc

es

39

Not

es

40

Fron

t cov

er p

hoto

cre

dit:

King

Cou

nty

Park

s and

Nat

ural

Res

ourc

es, R

ing

Hill

For

est

Firs

t edi

tion

- Dec

embe

r 201

3

3

Su

mm

ary

of

the

FLA

T P

roce

ss

At i

ts c

ore,

FLA

T co

nsis

ts o

f vis

ual e

stim

ates

of e

colo

gica

l con

ditio

ns b

y tr

aine

d st

aff to

pro

duce

a fo

rest

inve

ntor

y. W

hile

fiel

d te

ams

may

mak

e so

me

mea

sure

men

ts to

cal

ibra

te th

eir e

stim

ates

, the

ass

essm

ent i

tsel

f is

gene

rally

qua

litat

ive

and

relie

s on

car

eful

ly p

repa

red

estim

ates

rath

er th

an

prec

ise

mea

sure

men

ts.

FLA

T is

exe

cute

d in

thre

e br

oad,

seq

uent

ial p

hase

s. T

hese

incl

ude:

Ph

ase

1 –

Fore

st C

over

Typ

e M

appi

ngA

eria

l im

ager

y an

d bo

unda

ry d

ata

of th

e la

nds

unde

r con

side

ratio

n ar

e us

ed

to d

ivid

e th

e ar

ea in

to p

olyg

ons

and

delin

eate

Man

agem

ent U

nits

(MU

s).

This

wor

k is

acc

ompl

ishe

d an

d re

cord

ed u

sing

Geo

grap

hic

Info

rmat

ion

Syst

em (G

IS) t

ools

. The

se M

Us

beco

me

the

unit

of o

bser

vatio

n an

d m

easu

rem

ent f

or th

e on

-site

ass

essm

ent.

Phas

e 2

- Fie

ld A

sses

smen

tTr

aine

d fie

ld te

ams

visi

t eac

h M

U w

ithin

the

proj

ect a

rea

to c

olle

ct

pred

eter

min

ed a

ttrib

utes

. As

an e

xam

ple,

col

lect

ed a

ttrib

utes

mig

ht in

clud

e la

nd c

over

, non

-nat

ive

spec

ies

in o

rder

of a

bund

ance

, and

tree

age

cla

ss

dist

ribut

ion.

Dat

a is

col

lect

ed fo

r MU

s an

d st

ored

(usi

ng G

IS o

r oth

er d

ata

man

agem

ent s

yste

m) f

or a

ll su

rvey

ed p

arce

ls.

Phas

e 3

- Man

agem

ent P

rior

itiz

atio

nTh

e fie

ld p

roce

dure

s of

FLA

T pr

ovid

e a

rank

ing

of p

arce

l con

ditio

ns, b

oth

acro

ss n

umer

ous

parc

els

and

for s

ubun

its w

ithin

a la

rge

parc

el. S

umm

ary

data

can

be

used

to e

stab

lish

man

agem

ent p

riorit

ies

for e

ach

MU

, and

ag

greg

ated

to d

evel

op p

riorit

ies

at la

rger

sca

les,

suc

h as

the

park

or p

ark

syst

em-s

cale

(as

pres

ente

d in

the

King

Cou

nty

exam

ple

late

r in

this

repo

rt).

This

prio

ritiz

atio

n ap

proa

ch is

bas

ed o

n a

mat

rix a

nd fl

ow-c

hart

ana

lysi

s (c

alle

d Tr

ee-ia

ge) t

hat c

lass

ifies

eac

h M

U u

sing

ratin

gs o

f lan

dsca

pe q

ualit

y th

at a

re e

asily

com

pare

d ac

ross

land

man

agem

ent u

nits

. The

info

rmat

ion

prod

uced

by

the

FLA

T pr

ovid

es a

sta

ndar

dize

d ba

selin

e of

eco

logi

cal d

ata

for a

var

iety

of l

ands

cape

type

s. T

his

info

rmat

ion

can

be u

sed

to v

iew

eac

h M

U w

ithin

the

cont

ext o

f an

entir

e la

nd m

anag

emen

t sys

tem

, as

wel

l as

prov

ide

a st

artin

g po

int f

or d

evel

opin

g a

land

-use

or s

tew

ards

hip

plan

fo

r par

ticul

ar p

arce

ls. R

epea

ted

over

tim

e, F

LAT

serv

es a

s an

effe

ctiv

e m

onito

ring

tool

for m

anag

ers

to re

view

and

then

ada

pt m

anag

emen

t pr

iorit

ies

and

actio

ns b

ased

on

chan

ging

con

ditio

ns.

This

str

aigh

tfor

war

d, s

yste

mat

ic a

ppro

ach

to e

colo

gica

l ass

essm

ent

appl

ies

prin

cipl

es o

f eco

logy

and

fore

stry

to p

rovi

de q

ualit

y da

ta a

nd h

elp

dete

rmin

e la

nd m

anag

emen

t prio

ritie

s. T

he fo

llow

ing

disc

usse

s th

e or

igin

s an

d ba

sis

of F

LAT.

2 2

Pro

ject

Bac

kgro

un

d

The

Fore

st L

ands

cape

Ass

essm

ent T

ool (

FLA

T) w

as d

evel

oped

by

the

Gre

en

Citie

s Re

sear

ch A

llian

ce a

nd c

oord

inat

ed b

y th

e U

SDA

For

est S

ervi

ce P

acifi

c N

orth

wes

t Res

earc

h St

atio

n, in

par

tner

ship

with

Kin

g Co

unty

, For

terr

a an

d th

e U

nive

rsity

of W

ashi

ngto

n.

FLA

T w

as b

uilt

on th

e fr

amew

ork

of th

e Tr

ee-ia

ge to

ol, i

nitia

lly d

evel

oped

in

200

5 by

Gre

en S

eatt

le P

artn

ersh

ip s

taff.

Sim

ilar t

o m

edic

al tr

iage

, the

ap

proa

ch a

llow

s la

nd m

anag

ers

to ra

pidl

y as

sess

land

scap

e co

nditi

ons,

then

pr

iorit

ize

rest

orat

ion

activ

ities

. Usi

ng h

igh,

med

ium

, or l

ow v

alue

s fo

r bot

h fo

rest

can

opy

com

posi

tion

and

fore

st h

ealth

thre

ats,

eac

h M

anag

emen

t U

nit (

MU

) is

assi

gned

one

of n

ine

desc

riptiv

e ca

tego

ries.

The

mod

el a

ssum

es

that

with

out d

istu

rban

ce, n

atur

al a

reas

wou

ld b

e do

min

ated

by

mat

ure,

ev

ergr

een

coni

fer t

rees

, suc

h as

wes

tern

hem

lock

and

Dou

glas

-fir,

with

a

med

ium

- to

high

-den

sity

can

opy,

mix

ed a

ge-c

lass

es, a

nd s

peci

es d

iver

sity

. Th

ese

high

-qua

lity

fore

st s

tand

s, la

ckin

g in

vasi

ve s

peci

es, r

epre

sent

a ty

pica

l Pa

cific

Nor

thw

est l

owla

nd fo

rest

- th

e re

fere

nce

for t

he tr

ee-ia

ge a

naly

sis.

Base

d on

inpu

t fro

m K

ing

Coun

ty D

epar

tmen

t of N

atur

al R

esou

rces

and

Pa

rks,

Am

eric

an F

ores

t Man

agem

ent (

AFM

, for

mer

ly In

tern

atio

nal F

ores

try

Cons

ulta

nts,

Inc.

), an

d Fo

rter

ra, F

LAT

was

dev

elop

ed to

furt

her s

uppo

rt

fore

st m

anag

emen

t nee

ds. M

ore

attr

ibut

es w

ere

adde

d to

refin

e ou

tput

s an

d ne

w fl

owch

arts

cre

ated

to a

ccom

mod

ate

alte

rnat

ive

habi

tats

. The

to

ol w

as a

lso

mod

ified

to in

clud

e a

fore

st h

ealth

val

ue, p

rese

ntin

g a

new

dim

ensi

on to

the

tree

-iage

mat

rix a

naly

sis.

The

add

ition

of t

ype

call

info

rmat

ion

give

s ea

ch M

U a

spe

cies

-sto

ckin

g-de

nsity

cod

e th

at c

an b

e us

ed to

str

atify

MU

's fo

r fut

ure

man

agem

ent.

Thi

s ne

w to

ol c

aptu

res

info

rmat

ion

esse

ntia

l to

deve

lopi

ng m

anag

emen

t str

ateg

ies

for i

ndiv

idua

l M

anag

emen

t Uni

ts, a

s w

ell a

s fo

r dra

ftin

g m

anag

emen

t pla

ns a

t the

par

k or

fo

rest

sca

le.

The

FLA

T to

ol w

as p

ilote

d du

ring

2010

-201

2 fie

ld s

easo

ns a

t 149

site

s co

mpo

sed

of 1

,457

MU

’s, c

over

ing

appr

oxim

atel

y 24

,700

acr

es o

f Kin

g Co

unty

par

klan

ds. F

LAT

was

an

impo

rtan

t firs

t ste

p in

dev

elop

ing

a lo

ng

term

, sys

tem

-wid

e fo

rest

ste

war

dshi

p pr

ogra

m. D

eter

min

ing

the

cond

ition

an

d he

alth

of a

ll fo

rest

land

s w

ill h

elp

guid

e Ki

ng C

ount

y in

mak

ing

criti

cal

fore

st m

anag

emen

t dec

isio

ns.

This

man

ual w

as d

evel

oped

to re

cord

the

FLA

T pr

oces

s an

d pr

otoc

ols

for

futu

re u

se a

t Kin

g Co

unty

Par

ks, a

s w

ell a

s fo

r lan

d m

anag

ers

inte

rest

ed in

an

inno

vativ

e ra

pid

asse

ssm

ent t

ool.

4

Ge

ttin

g S

tart

ed

Con

sid

erat

ion

s

Staff

Whe

n po

ssib

le, fi

eld

asse

ssm

ents

sho

uld

be c

ondu

cted

by

two

or m

ore

peop

le b

oth

as a

saf

ety

prec

autio

n an

d fo

r qua

lity

cont

rol r

easo

ns. S

ince

th

e FL

AT

asse

ssm

ent r

elie

s on

ocu

lar e

stim

ates

of e

colo

gica

l con

ditio

ns, i

t is

oft

en h

elpf

ul to

hav

e an

othe

r set

of e

yes

and

a co

mpa

nion

to c

ompa

re

findi

ngs.

Seas

onFo

r the

dat

a co

llect

ion

com

pone

nt o

f the

FLA

T, fi

eld

asse

ssm

ents

sho

uld

take

pla

ce d

urin

g th

e pe

riod

whe

n fo

liage

is m

ost v

isib

le a

nd v

igor

ous.

Thi

s w

ill a

llow

fiel

d te

ams

to p

ositi

vely

iden

tify

plan

t spe

cies

and

mak

e th

e be

st

eval

uatio

n of

site

con

ditio

ns.

Tool

s an

d M

ater

ials

Befo

re le

avin

g on

a fi

eld

asse

ssm

ent,

a te

am s

houl

d be

pro

perly

equ

ippe

d w

ith r

equi

red

data

col

lect

ion

tool

s as

wel

l as

the

’11

esse

ntia

ls’ p

artic

ular

ly

whe

n ou

t in

the

field

for a

full

day.

A

ll fie

ld te

ams

shou

ld h

ave:

•Dat

aen

tryto

ols

Han

d-he

ld e

lect

roni

c da

ta re

cord

er o

r fiel

d da

ta s

heet

s•

Nav

igat

ionde

vice

s

GPS

Map

(inc

ludi

ng o

verla

y of

the

MU

bou

ndar

ies)

Co

mpa

ss•

Plan

tide

ntifica

tion

reso

urce

s•

Camer

a•

Tree

and

can

opymea

sure

men

ttoo

ls

Reco

mm

ende

d fo

r tra

inin

g an

d ca

libra

tion

of o

cula

r est

imat

es b

ut n

ot

requ

ired:

In

crem

ent b

orer

D

iam

eter

tape

D

ensi

tom

eter

(Moo

seho

rn)

Cl

inom

eter

5

THE

“11”

ESS

ENTI

ALS

for

com

fort

an

d sa

fety

in th

e fie

ld

1.

Sun

pro

tect

ion

(su

ng

lass

es, l

ip b

alm

, an

d s

un

scre

en)

2.

Bug

rep

elle

nt

3.

Pro

per

clo

thin

g a

nd

fo

otw

ear

to d

eal w

ith

har

sh t

erra

in

or

incl

emat

e w

eath

er s

uch

as

rain

gea

r, w

ater

pro

of

hik

ing

/wo

rk b

oo

ts,

gai

ters

, an

d i

nsu

lati

on

lik

e g

love

s,

hat

s, a

nd

jack

ets.

4.

Firs

t ai

d s

up

plie

s5.

U

tilit

y kn

ife o

r m

uli-

too

ls (

e.g

. Lea

ther

man

, Sw

iss

arm

y kn

ife)

6.

Foo

d (p

lus

an e

xtra

day

’s s

up

ply

)7.

Lo

ts o

f Wat

er! (

plu

s an

ext

ra d

ay’s

su

pp

ly)

8.

Hea

dla

mp

or

illu

min

atio

n s

ou

rce

9.

Fire

(mat

ches

or

ligh

ter

in w

ater

pro

of c

on

tain

er10

. Em

erg

ency

sh

elte

r (t

ent,

tarp

, biv

y, o

r refl

ecti

ve b

lan

ket)

11.

Co

mm

un

icat

ion

dev

ice

like

cell

ph

on

e o

r tw

o w

ay r

adio

Phot

o cr

edit:

Lis

a Ci

ecko

7

Fore

st C

ove

r T

ype

Map

pin

g

Fore

st C

over

Typ

e M

appi

ng is

the

proc

ess

of d

ivid

ing

a pa

rcel

of l

and

into

are

as o

f sim

ilar l

andc

over

and

veg

etat

ion/

fore

st ty

pes.

The

pro

pert

y yo

u ar

e to

ass

ess

has

been

“for

est-

type

d” u

sing

aer

ial p

hoto

grap

hy in

to

Man

agem

ent U

nits

(MU

s).

Ow

ner

ship

an

d M

anag

emen

t Bou

nd

arie

sEa

ch p

rope

rty

or p

arce

l is

first

del

inea

ted

by p

rope

rty

owne

rshi

p an

d m

an-

agem

ent b

ound

ary.

Nex

t, or

thop

hoto

grap

hs o

r oth

er a

eria

l im

ager

y ar

e us

ed

to g

roup

and

del

inea

te t

he la

nd in

to fi

ve b

road

land

cove

r cl

assi

ficat

ions

. Th

ese

are

fore

sted

, nat

ural

, ope

n w

ater

, har

dsca

pe, a

nd la

ndsc

ape.

Veg

etat

ion

Fea

ture

sW

ith p

rope

rty

info

rmat

ion

and

land

cove

r diff

eren

ces

delin

eate

d, th

e ne

xt

step

is to

refin

e fo

rest

ed a

nd n

atur

al s

ites

acco

rdin

g to

thei

r veg

etat

ion

feat

ures

. Lar

ge a

reas

with

cle

arly

diff

eren

t spe

cies

, str

uctu

ral f

eatu

res,

or

age

com

posi

tion

are

plac

ed in

to d

iffer

ent M

anag

emen

t Uni

ts.

LID

AR

data

can

als

o be

pro

cess

ed to

dis

play

can

opy

heig

hts.

Thi

s is

use

ful i

n in

terp

retin

g st

and

boun

darie

s, e

spec

ially

whe

n st

ereo

aer

ial i

mag

ery

is n

ot

avai

labl

e.

NO

TE: A

ll M

U b

ound

arie

s ar

e dr

awn

dire

ctly

into

the

GIS

so

that

they

do

not

need

to b

e di

gitiz

ed p

ost d

ata

colle

ctio

n.

Phot

o cr

edit:

Kin

g Co

unty

Par

ks a

nd N

atur

al R

esou

rces

Ba

ss L

ake

Nat

ural

Are

a

6

Qu

alit

y C

on

tro

l M

eth

od

s

Qua

lity

cont

rol c

onsi

sts

of a

ny p

roce

dure

s us

ed to

“cal

ibra

te” o

r rev

iew

fiel

d as

sess

men

ts.

Beca

use

of th

e qu

alita

tive

natu

re o

f rap

id a

sses

smen

t, it

is

expe

cted

that

indi

vidu

als

will

hav

e sl

ight

ly d

iffer

ent i

nter

pret

atio

ns o

f eac

h at

trib

ute.

The

goa

l is

to e

ncou

rage

sim

ilar o

r con

sist

ent i

nter

pret

atio

n an

d as

sess

men

ts.

To t

est

con

sist

ency

of

fiel

d d

ata

emp

loy

the

follo

win

g q

ualit

y co

ntr

ol

pro

ced

ures

:

Pre

-ass

essm

ent

Trai

nin

g•

Perf

orm

the

FLA

T on

a k

now

n M

U th

at h

as a

lread

y be

en a

sses

sed

and

com

pare

you

r find

ings

.

•Ch

oose

one

or m

ore

MU

s an

d ha

ve th

em a

sses

sed

by tw

o se

para

te

team

s un

der t

he s

uper

visi

on o

f a c

rew

lead

er fa

mili

ar w

ith F

LAT.

If

th

e te

ams’

att

ribut

e m

easu

res

diffe

r sig

nific

antly

, it s

ugge

sts

that

som

e ca

libra

tion

may

be

nece

ssar

y.

Dai

ly -

on

go

ing

•Se

lf ch

eck

each

day

. Ex

ampl

e –

Do

an o

cula

r est

imat

e of

dia

met

er-a

t-br

east

hei

ght.

Aft

er w

ritin

g do

wn

your

ans

wer

, tak

e th

e m

easu

rem

ents

us

ing

a db

h ta

pe a

nd c

ompa

re y

our e

stim

ates

. Thi

s ca

n be

don

e fo

r ot

her d

ata

varia

bles

usi

ng o

ther

cal

ibra

tion

devi

ces

such

as

tree

age

us

ing

a tr

ee c

orer

.

Phot

o cr

edit:

Kim

Fra

ppie

r

8

LAN

DC

OV

ER D

ESIG

NA

TIO

NS

MU

s are

ass

igne

d on

e of

five

bro

ad la

nd c

over

des

igna

tions

:

Fore

sted

(FO

R)≥

25%

of

the

area

cov

ered

by

fore

st c

anop

y

Nat

ural

(NA

T)na

tura

l veg

etat

ion

that

has

< 2

5% fo

rest

ca

nopy

cov

er

Ope

n W

ater

(W

AT)

op

en w

ater

with

out w

oody

veg

etat

ion

Har

dsca

pe (H

S)im

perv

ious

surf

ace

such

as p

arki

ng o

r bui

ldin

gs

Land

scap

e (L

S)la

ndsc

aped

or h

ave

been

mec

hani

cally

m

aint

aine

d w

ithin

the

last

yea

r.

The

Trut

h ab

out G

roun

d Tr

uthi

ngTh

e M

Us

are

initi

ally

del

inea

ted

with

out t

he b

enefi

t of g

roun

d tr

uthi

ng.

Fiel

d te

ams

are

resp

onsi

ble

for fi

eld

verifi

catio

n an

d sh

ould

adj

ust

boun

darie

s an

d la

nd c

over

type

s if

need

ed.

Som

e is

sues

that

may

be

enco

unte

red

incl

ude:

•La

ndsc

apin

g un

der c

anop

y •

Chan

ges

that

hav

e oc

curr

ed s

ince

the

phot

o da

te, a

nd

•A

reas

bei

ng re

stor

ed to

a n

atur

al c

ondi

tion

•

Inac

cura

cies

in in

terp

reta

tion

due

to d

eep

shad

ows,

par

alla

x, a

nd p

hoto

m

osai

c bo

unda

ries

on a

eria

l im

ager

y

Tran

smis

sion

line

cor

ridor

s w

ere

assi

gned

a “N

atur

al” d

esig

natio

n w

hen

type

d in

the

GIS

, but

can

be

give

n a

field

des

igna

tion

of “L

ands

cape

” due

to

vege

tatio

n m

aint

enan

ce in

thos

e ar

eas.

In a

reas

of l

ow c

anop

y an

d sh

rubs

, typ

ical

ly s

een

in w

et a

reas

, MU

s w

ere

assi

gned

a “F

ores

t” c

all o

ver a

“Nat

ural

” cal

l. T

hese

are

as m

ust h

ave

mor

e th

an 2

5% c

anop

y co

ver.

Fie

ld T

ech

nic

ian

s m

ake

th

e f

inal

cal

l!

9

Und

erst

andi

ng M

anag

emen

t Uni

t Del

inea

tion

Bo

un

dar

ies

Man

agem

ent U

nits

wer

e de

linea

ted

base

d on

fore

st c

ompo

sitio

n so

that

he

tero

gene

ity w

ithin

par

ks o

r par

cels

cou

ld b

e ac

coun

ted

for.

MU

s do

not

cr

oss

“adm

inis

trat

ive”

bou

ndar

ies,

eve

n if

the

cove

r typ

e is

iden

tical

on

both

si

des.

Size

The

min

imum

siz

e fo

r an

MU

is 5

acr

es, u

nles

s th

e ty

pe is

sur

roun

ded

by

dist

inct

non

-for

est t

ypes

or p

rope

rty

boun

darie

s. S

igni

fican

t and

dis

tinct

no

n-fo

rest

are

as m

ay b

e de

linea

ted

dow

n to

0.1

acr

e. I

t is

not t

he in

tent

of

this

ass

essm

ent t

o m

ap p

aved

pat

hs, t

enni

s co

urts

, or o

ther

faci

litie

s w

ithin

la

ndsc

aped

are

as. T

hese

sm

alle

r dev

elop

ed fe

atur

es a

re g

roup

ed w

ith

surr

ound

ing

land

scap

e or

har

dsca

pe M

Us.

Lab

els

Each

par

k or

par

cel h

as a

uni

que

FID

_MU

iden

tifier

. FID

refe

rs to

the

park

’s

Faci

lity

Iden

tifica

tion

num

ber.

The

MU

refe

rs to

the

poly

gon

num

ber w

ithin

th

at p

ark.

Thi

s id

entifi

er is

spe

cific

to e

ach

poly

gon

for e

ase

of id

entifi

catio

n,

sort

ing,

and

ana

lysi

s of

dat

a. F

or e

xam

ple,

the

Duv

all P

ark

FID

is 2

598,

and

th

ere

are

five

MU

’s in

that

par

k. T

he F

ID_M

U id

entifi

ers

are

2598

-01,

259

8-02

, 259

8-03

, 259

8-04

, and

259

8-05

.

Imag

e cr

edit:

Kin

g Co

unty

Par

ks a

nd N

atur

al R

esou

rces

M

aury

Isla

nd M

Us

11

Chec

kfo

rco

mplete

nes

sSo

met

imes

MU

s w

ill b

e st

rang

ely

shap

ed s

o th

at th

e te

rrai

n or

trai

ls w

ill le

ad

team

s tr

avel

ing

in a

nd o

ut o

f diff

eren

t MU

s. D

ata

may

be

ente

red

for e

ach

MU

in o

rder

of d

isco

very

but

sho

uld

be c

heck

ed fo

r com

plet

enes

s be

fore

le

avin

g th

e M

U.

When

tole

aveth

ero

adortrails

yste

mEa

ch a

ttrib

ute

dete

rmin

atio

n w

ill a

pply

to th

e en

tire

MU

, thu

s it

is im

port

ant

that

eno

ugh

of th

e M

U is

see

n by

the

field

team

. Som

etim

es th

is w

ill re

quire

le

avin

g tr

ails

and

trav

elin

g on

cha

lleng

ing

terr

ain.

In o

ther

MU

s th

e vi

ew

from

a tr

ail m

ay b

e su

ffici

ent t

o m

ake

a ju

dgm

ent o

n m

ost,

if no

t all,

of t

he

attr

ibut

es. T

eam

s sh

ould

be

care

ful t

o re

cogn

ize,

and

ave

rage

into

the

MU

es

timat

e, a

ny e

dge

effec

t alo

ng tr

ails

whe

re v

eget

atio

n m

ay h

ave

been

in

fluen

ced

by tr

ail a

ctiv

ity o

r dis

turb

ance

. A g

ood

ques

tion

to a

sk b

efor

e bu

shw

hack

ing

thro

ugh

a si

te m

ay b

e “H

ow m

uch

grea

ter u

nder

stan

ding

of

the

MU

will

I ga

in fr

om th

e tim

e it

will

take

me

to d

o th

is?”

Pre

par

ing

th

e G

PS

un

it o

r d

ata

form

s fo

r fi

eld

ass

essm

ents

•

GIS

sta

ff o

r p

roje

ct l

ead

s sh

ou

ld p

re-l

oad

bac

kgro

un

d i

mag

es o

f ve

get

atio

n o

verl

ayed

wit

h t

he

MU

map

s in

acc

ord

ance

wit

h t

he

har

dw

are

and

so

ftw

are

in u

se.

•

Elec

tro

nic

dat

a co

llect

ion

so

ftw

are

or

spre

adsh

eet

sho

uld

be

pre

-lo

aded

wit

h t

he

init

ial M

U L

and

cove

r co

des

. If u

sin

g a

GPS

wit

h t

he

bac

kgro

un

d lo

aded

, th

e im

age

file

nam

e sh

ou

ld c

orr

esp

on

d t

o t

he

map

nu

mb

er.

•

Info

rmat

ion

on

the

MU

lan

dco

ver d

esig

nat

ion

and

bo

un

dar

ies

mu

st

be

incl

ud

ed e

ith

er o

n t

he

map

, wit

hin

th

e d

ata

entr

y sy

stem

, or

on

field

form

s.

Fiel

d te

ams

will

then

be

able

to g

rou

nd

tru

th th

e la

nd

cove

r des

ign

atio

n as

wel

l as

kno

w w

hen

th

ey h

ave

ente

red

or

exit

ed t

he

MU

.

Sin

ce d

iffer

ent

GPS

un

its

hav

e ve

ry d

iffer

ent

inst

ruct

ion

s, t

his

pro

cess

w

ill b

e d

evic

e sp

ecifi

c.

10

Intr

od

uct

ion

to

Fie

ld P

roce

du

res

The

field

pro

cedu

res

enta

il oc

ular

est

imat

es o

f pre

dete

rmin

ed fo

rest

qu

aliti

es.

Fiel

d te

ams

of o

ne o

r mor

e pe

ople

vis

it ea

ch M

U, w

alk

thro

ugh

it, a

nd re

cord

an

aver

age

attr

ibut

e va

lue

for e

ach

of th

e va

riabl

es li

sted

in

this

fiel

d m

anua

l or d

ata

entr

y sy

stem

. The

fiel

d m

anua

l con

tain

s de

finiti

ons

for e

ach

varia

ble

to b

e as

sess

ed.

As

field

team

s w

alk

thro

ugh

the

MU

, the

y sh

ould

be

obse

rvan

t and

kee

p a

men

tal c

atal

og o

f wha

t the

y se

e.

Thin

gs

to k

eep

in m

ind

wh

ile c

on

du

ctin

g t

he

asse

ssm

ent:

Pro

videat

trib

ute

estim

ates

forth

een

tire

MU

Ther

e m

ay b

e pa

tche

s of

cer

tain

spe

cies

or c

ondi

tions

, but

a b

est e

ffort

sh

ould

be

mad

e to

est

imat

e fo

r the

ent

ire M

U.

This

is w

hy m

any

of th

e at

trib

ute

valu

es w

ill b

e en

tere

d up

on le

avin

g th

e M

U.

Mea

sure

men

tto

olscan

beuse

dtocalib

rate

youres

ti-

mat

esSo

me

attr

ibut

es s

uch

as D

BH, c

row

n cl

osur

e, a

ge, a

nd re

gene

ratio

n tr

ees

(tre

es le

ss th

an 2

0ft i

n he

ight

) can

be

mea

sure

d w

ith to

ols

in th

e fie

ld.

It m

ay b

e he

lpfu

l to

do th

is o

nce

or tw

ice

on a

n M

U to

cal

ibra

te e

stim

atio

ns.

How

ever

, exc

essi

ve m

easu

rem

ent t

akin

g in

the

field

will

slo

w d

own

the

rapi

d as

sess

men

t pro

cess

.

DotheMUboundar

iesnee

dtobealte

red?

Tosplit

orlump?

The

field

team

sho

uld

ask

them

selv

es w

heth

er o

r not

the

boun

darie

s of

the

MU

s sh

ould

be

alte

red,

and

if s

o, h

ow.

Ther

e ar

e of

ten

patc

hes

of d

iffer

ent

vege

tatio

n w

ithin

an

MU

whi

ch is

exp

ecte

d, b

ut p

erva

sive

diff

eren

ces

in c

ompo

sitio

n or

age

cla

sses

for l

arge

are

as o

f the

MU

may

nec

essi

tate

re

draw

ing

boun

darie

s. T

his

may

invo

lve

com

bini

ng o

r lum

ping

sm

alle

r MU

s in

to o

ne la

rger

one

or s

plitt

ing

an M

U in

to m

ultip

le, s

mal

ler M

Us.

AdditionalN

ote

san

dO

bse

rvat

ions

Fiel

d te

ams

may

find

that

ther

e is

som

ethi

ng im

port

ant o

f not

e in

the

MU

th

at d

oesn

’t ne

cess

arily

fit i

nto

any

of th

e at

trib

ute

cate

gorie

s. F

or th

is

reas

on it

is g

ood

to in

clud

e ad

ditio

nal i

nfor

mat

ion

in th

e “n

otes

” fiel

d of

th

e da

ta e

ntry

tool

or d

ata

colle

ctio

n sh

eet.

Exa

mpl

es o

f add

ition

al n

otes

in

clud

e: a

ny n

ew tr

ee o

r pla

nt s

peci

es, p

reva

lenc

e of

a 4

th o

vers

tory

tree

, un

usua

l site

or s

oil c

ondi

tions

, rec

ent d

istu

rban

ces,

and

any

rece

nt o

r on

goin

g la

nd m

anag

emen

t act

iviti

es.

12

Dat

a Co

llect

ion

Defi

niti

ons

and

Proc

edur

es

LAN

DC

OV

ERLa

ndco

ver i

s in

itial

ly a

ssig

ned

in th

e offi

ce b

y th

e G

IS p

rofe

ssio

nal a

nd

need

s to

be

field

ver

ified

.

Ther

e ar

e fiv

e la

ndco

ver c

ateg

orie

s us

ed fo

r the

Kin

g Co

unty

FLA

T as

sess

men

t: Fo

rest

ed, N

atur

al, W

ater

, Har

dsca

pe, a

nd L

ands

cape

. Usi

ng

your

fiel

d m

aps

with

MU

bou

ndar

ies,

ver

ify a

nd re

cord

the

MU

s la

ndco

ver

clas

sific

atio

n.

Ver

ify

and

rec

ord

lan

dco

ver

usi

ng

th

e fo

llow

ing

ca

teg

ori

es

ASP

ECT

AN

D S

LOP

EA

spec

t and

slo

pe a

re p

roce

ssed

in th

e offi

ce th

roug

h av

aila

ble

digi

tal

elev

atio

n m

odel

dat

a in

GIS

for e

ach

MU

. Thi

s is

esp

ecia

lly u

sefu

l for

larg

er

MU

’s w

here

asp

ect a

nd s

lope

var

y th

roug

hout

the

unit.

If a

pro

ject

cal

ls fo

r fie

ld d

eter

min

atio

n th

en p

roce

ed a

s fo

llow

s:

•U

se a

com

pass

to d

eter

min

e th

e pr

edom

inan

t dire

ctio

n of

the

slop

e on

th

e si

te.

•A

spec

t is

ofte

n de

scrib

ed a

s th

e di

rect

ion

in w

hich

wat

er fl

ows

off a

site

•Co

de a

s fo

llow

s: N

, NE,

E, S

E, S

, SW

, W, N

W o

r flat

. •

Det

aile

d in

stru

ctio

ns o

n us

e of

a c

ompa

ss c

an b

e fo

und

in A

ppen

dix

E.

Fore

sted

(FO

R)≥

25%

of t

he a

rea

cove

red

by fo

rest

can

opy

Nat

ural

(NA

T)na

tura

l veg

etat

ion

that

has

< 2

5% fo

rest

can

opy

cove

r

Ope

n W

ater

(W

AT)

op

en w

ater

with

out w

oody

veg

etat

ion

Har

dsca

pe (H

S)im

perv

ious

sur

face

suc

h as

par

king

or

build

ings

Land

scap

e (L

S)la

ndsc

aped

or

have

bee

n m

echa

nica

lly m

aint

aine

d w

ithin

the

last

yea

r.

13

OV

ERST

OR

YO

vers

tory

refe

rs to

tree

s w

hose

folia

ge fo

rms

the

uppe

rmos

t cro

wn

cove

r or

cano

py o

f a fo

rest

sta

nd.

The

fore

st a

sses

smen

t tea

m w

ill c

hara

cter

ize

dom

inan

t or c

o-do

min

ant

over

stor

y tr

ee s

peci

es in

ord

er o

f abu

ndan

ce fo

und

in th

e M

U in

clud

ing

the

size

cla

ss o

f eac

h sp

ecie

s. A

ge c

lass

info

rmat

ion

will

onl

y be

col

lect

ed fo

r the

do

min

ant o

vers

tory

spe

cies

.

The

first

, sec

ond,

and

third

ord

er o

f abu

ndan

ce s

houl

d be

det

erm

ined

by

the

two

dim

ensi

onal

are

a th

at th

e sp

ecie

s w

ould

occ

upy

if lo

okin

g at

the

aeria

l im

age.

Fie

ld te

ams

can

use

the

aeria

l im

ager

y to

hel

p id

entif

y th

e pr

esen

ce in

the

MU

.

Co

llect

th

e fo

llow

ing

ove

rsto

ry d

ata

Ove

rsto

rySpe

cies

1:T

he m

ost a

bund

ant d

omin

ant o

r co-

dom

inan

t ov

erst

ory

spec

ies

grea

ter t

han

20 fe

et in

hei

ght.

•Sp

ecie

s co

de (N

ote:

if th

e tr

ee is

not

list

ed in

the

plan

t lis

t in

Appe

ndix

B,

prov

ide

the

4 di

git s

peci

es c

ode,

com

mon

, and

scie

ntifi

c na

mes

in th

e no

tes

sect

ion)

•A

ge C

lass

•Si

ze C

lass

Ove

rsto

rySpe

cies

2and

3:T

he s

econ

d an

d th

ird m

ost a

bund

ant c

o-do

min

ant o

vers

tory

spe

cies

gre

ater

than

20

feet

in h

eigh

t.

•Sp

ecie

s co

de (N

ote:

if th

e tr

ee is

not

list

ed in

the

plan

t lis

t in

Appe

ndix

B,

prov

ide

the

4 di

git s

peci

es c

ode,

com

mon

, and

scie

ntifi

c na

mes

in th

e no

tes

sect

ion)

•Si

ze C

lass

for e

ach

spec

ies.

Dom

inan

t ver

sus

Co-

dom

inan

t

Do

min

ance

is a

rel

ativ

e d

esig

nat

ion

of t

ree

cro

wn

s an

d is

al

so r

efer

red

to

as

cro

wn

cla

ss. D

om

inan

t tr

ees

are

tho

se w

ith

cro

wn

s ab

ove

th

e g

ener

al le

vel o

f th

e ca

no

py.

Co

-do

min

ance

re

fers

to

tre

es w

ho

se c

row

ns

form

th

e g

ener

al le

vel o

f th

e ca

no

py.

15

SIZ

E C

LASS

Seco

nd, r

ecor

d th

e si

ze c

lass

or a

vera

ge d

iam

eter

at b

reas

t hei

ght (

DBH

) for

ea

ch d

omin

ant o

r co-

dom

inan

t tre

e lis

ted

abov

e. D

BH is

the

diam

eter

of a

tr

ee a

t 4.5

feet

abo

ve th

e gr

ound

on

the

uphi

ll si

de o

f the

tree

. Cal

ibra

tion

tree

s ca

n be

cho

sen

from

ave

rage

def

ect-

free

tree

s in

the

MU

, kee

ping

in

min

d th

e ta

rget

cla

sses

list

ed in

the

tabl

e be

low

. For

mor

e in

form

atio

n on

m

easu

ring

tree

dia

met

er, s

ee A

ppen

dix

C.

Rec

ord

Siz

e C

lass

usi

ng

th

e fo

llow

ing

cat

ego

ries

Size

Cla

ss C

ode

DBH

in in

ches

10

-5”

26

- 10”

311

- 20

”

421

”+

Phot

o cr

edit:

Lis

a Ci

ecko

14

AG

E C

LASS

Age

cla

ss is

an

estim

ate

of th

e ag

e ra

nge

of th

e do

min

ant o

vers

tory

tree

s (s

ee b

elow

). A

ge c

lass

es p

rovi

de m

anag

ers

with

an

over

view

of w

hich

M

Us

have

the

oppo

rtun

ity fo

r sta

nd im

prov

emen

t ope

ratio

ns, h

arve

st, o

r pr

eser

vatio

n op

port

uniti

es.

It is

impo

rtan

t to

calib

rate

est

imat

es o

f tre

e ag

e us

ing

an in

crem

ent b

orer

. Th

is s

houl

d be

don

e at

the

begi

nnin

g of

the

asse

ssm

ent p

roce

ss to

cal

ibra

te

your

ocu

lar e

stim

ates

and

then

spo

t che

ck y

ours

elf t

hrou

ghou

t the

fiel

d as

sess

men

t. Se

e A

ppen

dix

F fo

r ins

truc

tions

on

how

to u

se a

n in

crem

ent

bore

r.

Exte

rnal

indi

cato

rs o

f the

ass

ocia

ted

age

rang

e ca

n be

use

d to

est

imat

e ag

e cl

ass.

The

se in

clud

e tr

ee s

ize

and

grow

th c

hara

cter

istic

s (e

.g. b

ranc

h w

horls

on

Dou

glas

fir)

, bar

k ap

pear

ance

, nea

rby

cut s

tum

ps o

r fal

len

tree

s.

How

ever

, soi

l and

site

qua

lity

dete

rmin

e ac

tual

gro

wth

rate

s, s

o th

is s

houl

d be

take

n in

to c

onsi

dera

tion

whe

n as

sess

ing

the

age

clas

s of

the

stan

d or

m

anag

emen

t uni

t.

Rec

ord

th

e av

erag

e A

ge

Cla

ss a

cro

ss t

he

MU

u

sin

g t

he

follo

win

g c

ateg

ori

es Co

deA

ge R

ange

(y

ears

)Ca

tego

ryD

escr

iptio

n

10-

29Pr

e-m

erch

anta

ble

Stan

d of

com

mer

cial

spe

cies

that

ha

ve n

ot y

et g

row

n la

rge

enou

gh

to b

e sa

leab

le

230

-49

Subm

erch

anta

ble

Stan

d ju

st c

omin

g in

to m

axim

um

valu

e, b

ut n

ot re

ady

for h

arve

st.

350

-99

Mer

chan

tabl

eSt

and

has

grow

n la

rge

enou

gh

to b

e sa

leab

le a

nd is

read

y fo

r ha

rves

t and

repl

antin

g

410

0 +

Mat

ure

Pote

ntia

l for

futu

re o

ld g

row

th

char

acte

ristic

s. M

ay a

im to

im

prov

e he

alth

and

reta

in m

atur

e tr

ees.

Test

You

rsel

f!

Rem

embe

r tha

t ext

erna

l ind

icat

ors

do n

ot a

lway

s pr

ovid

e su

ffici

ent

info

rmat

ion

to d

eter

min

e th

e ag

e cl

ass

of a

tree

, so

age

corin

g si

mila

r tre

es

can

be d

one

to c

alib

rate

you

r est

imat

es.

Age

by

eye,

then

cor

e to

con

firm

! H

ow c

lose

was

you

r est

imat

e?

16

STO

CK

ING

Fi

eld

team

s w

ill b

e us

ing

mea

sure

s of

can

opy

cove

r to

est

imat

e st

ocki

ng in

th

e M

Us.

Can

opy

cove

r is

a v

ertic

al m

easu

re o

f the

can

opy

(dom

inan

t an

d co

-dom

inan

t tre

e cr

owns

) as

wou

ld b

e se

en o

n an

aer

ial p

hoto

grap

h or

look

-in

g up

ver

tical

ly fr

om o

ne p

oint

on

the

grou

nd. M

easu

res

of c

over

ass

ess

the

pres

ence

or

abse

nce

of c

anop

y ve

rtic

ally

abo

ve a

sam

ple

of p

oint

s ac

ross

a

defin

itive

are

a of

fore

st a

nd h

elp

asse

ss fo

rest

stru

ctur

e (J

enni

ngs e

t al,

1999

).

To m

easu

re c

anop

y co

ver,

stan

d in

one

loca

tion,

and

ass

ess

the

cano

py

dire

ctly

ove

rhea

d (s

ee F

igur

e 1)

. Fie

ld s

taff

can

also

com

pare

on

the

grou

nd

mea

sure

men

ts w

ith th

e or

thop

hoto

of t

he s

ite to

aid

in e

stim

atin

g co

ver.

In o

rder

to c

aptu

re th

e va

riabi

lity

acro

ss la

rge

MU

s, fi

eld

crew

s m

ust t

ake

this

read

ing

at d

iffer

ent p

oint

s ac

ross

the

MU

and

then

ave

rage

thos

e co

nditi

ons.

To c

alib

rate

ocu

lar e

stim

ates

, can

opy

cove

r can

be

mea

sure

d us

ing

inst

rum

ents

suc

h as

a d

ensi

tom

eter

, som

etim

es c

alle

d a

“moo

se h

orn.

” The

de

nsito

met

er p

rovi

des

a po

int m

easu

re o

f can

opy

cove

r. S

ee A

ppen

dix

D

for a

dditi

onal

info

rmat

ion

abou

t den

sito

met

ers.

Figu

re 1

. Can

opy

cove

r mea

sure

d fr

om o

ne p

oint

on

the

grou

nd. I

mag

e cr

edit:

Kor

hone

n et

al,

2006

17

Rec

ord

sto

ckin

g u

sing

the

follo

win

g c

ateg

orie

s

Stoc

king

Cod

eCa

nopy

Cov

er R

ange

in P

erce

nt

0Le

ss th

an 1

0% c

anop

y co

ver

110

- 39

% c

anop

y co

ver

240

- 69

% c

anop

y co

ver

3G

reat

er th

an 7

0% c

anop

y co

ver

Wh

at is

sto

ckin

g?

•

Sto

ckin

g (

rela

ted

to

sta

nd

den

sity

) is

a m

easu

re o

f th

e cr

owd

ing

of t

rees

in a

sta

nd o

r the

are

a oc

cup

ied

by

tree

s re

lati

ve t

o a

n o

pti

mu

m o

r d

esir

ed l

evel

of

den

sity

th

at

sup

por

ts g

row

th fo

r tim

ber

man

agem

ent.

•

Stoc

king

can

be

exp

ress

ed q

uant

itat

ivel

y as

the

bas

al a

rea

or v

olum

e of

tree

s p

er a

cre.

•

The

FLA

T m

easu

res

stoc

king

in re

lativ

e te

rms

usin

g ca

nop

y co

ver

esti

mat

es a

s an

in

dic

atio

n o

f h

ow

cro

wd

ed t

ree

crow

ns a

re w

ithi

n a

stan

d.

•

Oft

en s

tock

ing

is

des

crib

ed i

n r

elat

ive

term

s, s

uch

as

par

tial

ly s

tock

ed, a

deq

uate

ly s

tock

ed, o

r ove

rsto

cked

.

MU

Tre

e C

ano

py

Co

mp

osi

tio

n

Each

MU

is a

ssig

ned

a va

lue

(Hig

h, M

ediu

m, o

r Low

) for

tree

can

opy

com

posi

tion,

bas

ed o

n pe

rcen

t nat

ive

tree

can

opy

cove

r, an

d pe

rcen

t of

can

opy

cove

r mad

e up

by

ever

gree

ns a

nd/o

r mad

rone

s. T

ree

Cano

py

Com

posi

tion

is o

ne o

f the

var

iabl

es (a

long

with

inva

sive

thre

at c

over

) use

d to

det

erm

ine

the

Tree

-iage

Cat

egor

y of

eac

h M

U.

Rec

ord

Tre

e C

ano

py

Co

mp

osi

tio

n a

s H

igh

, Med

ium

, or

Lo

w b

ased

on

th

e flo

w c

har

t o

n p

age

18

19

Ove

rsto

ry F

ore

st H

ealt

h T

hre

at In

dic

ato

rs

Fore

st h

ealth

thre

at in

dica

tors

refe

r to

attr

ibut

es in

dica

tive

of p

oor t

ree

heal

th a

nd c

anop

y de

clin

e. T

hese

att

ribut

es n

egat

ivel

y aff

ect t

he lo

ng-

term

sus

tain

abili

ty o

f the

fore

st c

anop

y an

d di

rect

ly a

ffect

man

agem

ent

stra

tegi

es. T

he in

dica

tors

mus

t be

pres

ent i

n 1%

or m

ore

of th

e M

U to

trig

ger

reco

rdin

g a

“Yes

.” If

not o

bser

ved

in th

e w

alkt

hrou

gh it

will

be

reco

rded

as

“No.

”

Re

cord

the

follo

win

g O

vers

tory

For

est H

ealt

h Th

reat

Indi

cato

rs

Low

vig

orD

eter

min

e lo

w v

igor

by

asse

ssin

g th

e tr

ee’s

live

cro

wn

ratio

. Cro

wn

ratio

is a

mea

sure

of t

he le

ngth

of

a tr

ee’s

live

cro

wn

rela

tive

to to

tal

tree

hei

ght.

Reco

rd a

“yes

” cal

l if c

onife

rs h

ave

a liv

e cr

own

of 4

0% o

r les

s of

the

tota

l hei

ght o

f tha

t tre

e. H

ardw

ood

decl

ine

in th

e fo

rm o

f sna

gs o

r to

p di

ebac

k of

5%

or g

reat

er a

lso

rece

ive

a “Y

es” c

all.

Failu

re

Mec

hani

cal t

ree

failu

re re

fers

to th

e br

eaka

ge o

f tre

e tr

unks

and

bra

nche

s an

d th

e up

root

ing

of tr

ees

caus

ed b

y fa

ctor

s su

ch a

s la

ndsl

ides

, ice

and

sn

ow d

amag

e, h

igh

win

ds, o

ld a

ge, p

aras

ites

or d

isea

se.

Reco

rd “Y

es” i

f fo

und

in 1

% o

r mor

e of

the

MU

. Ro

ot R

ot

A fu

ngal

root

infe

ctio

n th

at a

ttac

ks th

e liv

e an

d de

ad ro

ots

of s

ome

coni

fers

. D

ougl

as-fi

r is

high

ly s

usce

ptib

le a

long

with

true

firs

suc

h as

gra

nd fi

r. W

este

rn re

dced

ar is

resi

stan

t to

infe

ctio

n an

d ha

rdw

ood

tree

s ar

e im

mun

e.

For m

ore

info

rmat

ion

on ro

ot ro

t dis

ease

go

to:

http

://ex

t.nrs

.wsu

.edu

/for

estr

yext

/for

esth

ealth

/not

es/la

min

ated

root

rot.h

tm

Roo

t rot

sym

pto

ms

to lo

ok fo

r:•

Pock

ets

of s

tand

ing

dead

tree

s•

A c

lear

ing

with

man

y tr

ees

falle

n to

the

grou

nd•

Stun

ted

root

bal

ls o

n fa

llen

tree

s•

A y

ello

w a

nd th

inni

ng tr

ee c

row

n

(Ove

rsto

ry fo

rest

hea

lth

thre

at in

dica

tors

con

tinu

ed o

n pg

20)

W

hat

is 1

% o

f an

MU

?

For a

10

acre

par

cel,

1%

= 1

/10

of a

n ac

re

= 4

,356

sq

uare

feet

= 6

6 x

66 fe

et

18

Can

opy

Com

pos

itio

n Fl

ow C

hart

20

Mis

tlet

oe

Mis

tleto

e is

a p

aras

itic

plan

t tha

t cau

ses

grow

th

redu

ctio

n an

d de

form

ities

on

tree

s. I

nfec

ted

tree

s w

ill

prod

uce

“witc

hes

broo

ms”

w

hich

are

abn

orm

al

grow

ths

of s

mal

l tw

igs.

Sy

mpt

oms

of in

fect

ions

on

bra

nche

s in

clud

e a

spin

dle-

shap

ed, s

wol

len

appe

aran

ce. O

n tr

unks

, in

fect

ions

may

cau

se th

e tr

ee to

sw

ell t

o tw

ice

its

orig

inal

dia

met

er. T

he

mos

t com

mon

spe

cies

that

in

fect

s co

nife

rs in

Pug

et

Soun

d fo

rest

s is

Dw

arf

mis

tleto

e (A

rceu

thob

ium

sp

.). F

or m

ore

info

rmat

ion

go to

: htt

p://

ext.n

rs.

wsu

.edu

/for

estr

yext

/fo

rest

heal

th/n

otes

/dw

arfm

istle

toe.

Fiel

d st

aff d

o no

t nee

d to

iden

tify

spec

ific

spec

ies

of m

istle

toe,

but

sim

ply

note

its

pres

ence

or a

bsen

ce.

If m

istle

toe

is o

bser

ved,

a “Y

es” c

all i

s gi

ven.

A

lso,

incl

ude

whi

ch tr

ee s

peci

es h

ave

been

infe

cted

in th

e no

tes

field

of

your

dat

a co

llect

ion

shee

t or G

PS u

nit.

Bare

Soi

l1%

or m

ore

of th

e M

U is

dev

oid

of v

eget

atio

n an

d ha

s ex

pose

d ba

re s

oil d

ue

to u

nsta

ble

soils

and

/or r

ecen

t dis

turb

ance

, lan

dslid

e et

c. O

ther

If

“Yes

” is

reco

rded

, det

ails

mus

t be

prov

ided

in th

e “n

otes

” fiel

d of

the

data

sh

eet o

r GPS

uni

t. Ex

ampl

es o

f add

ition

al fo

rest

hea

lth c

once

rns

incl

ude

but

are

not l

imite

d to

:

•Be

ar d

amag

e•

Beet

le d

amag

e•

Brow

n cu

bica

l but

t rot

•La

rge

gap

in c

ente

r of M

U –

pot

entia

l roo

t rot

dam

age

Phot

o cr

edit:

Jess

e Sa

unde

rs

21

Reg

ener

atio

n S

pec

ies

Rege

nera

tion

spec

ies

refe

r to

over

stor

y ca

nopy

spe

cies

und

er 2

0 fe

et ta

ll.

Thes

e ar

e th

e tr

ees

that

will

bec

ome

dom

inan

t and

co-

dom

inan

t spe

cies

as

the

fore

st m

atur

es.

Rec

ord

th

e tw

o m

ost

ab

un

dan

t o

vers

tory

tre

e sp

ecie

s

un

der

20

feet

tal

l

Reg

ener

atio

n S

pec

ies

Sto

ckin

g C

lass

Reco

rd th

e co

mbi

ned

estim

ated

sto

ckin

g cl

ass

of th

e tw

o m

ost a

bund

ant

rege

nera

tion

tree

spe

cies

und

er 2

0 fe

et in

the

MU

. Th

is is

mea

sure

d in

tree

s pe

r acr

e (T

PA).

Rec

ord

sto

ckin

g c

lass

usi

ng

th

e fo

llow

ing

co

des

:

Stoc

king

Cod

eTr

ees

per a

cre

Refe

renc

e

10-

49 T

PA

> 30

ft. x

30

ft.

250

-149

TPA

Betw

een

30 ft

. and

16

ft.

spac

ing

315

0 +

TPA

<

16 ft

. x 1

6 ft

.

Is t

hat

a S

HR

UB

or

REG

ENER

ATI

ON

TR

EE?

Refe

r to

th

e FL

AT

pla

nt

gu

ide

in A

pp

end

ix B

if y

ou

hav

e q

ues

tio

ns

abo

ut

wh

eth

er a

sp

ecifi

c sp

ecie

s is

co

nsi

der

ed a

tre

e (a

nd

th

eref

ore

co

nsi

der

ed p

art

of t

he

ove

rsto

ry) o

r a

shru

b.

Exam

ple:

Will

ow s

peci

es a

re c

onsi

dere

d tr

ees

w

here

as V

ine

map

le is

a s

hrub

.

23

Inva

sive

Sp

ecie

s Ea

ch M

U is

ass

igne

d a

valu

e (H

igh,

Med

ium

, or

Low

) for

inva

sive

cov

er th

reat

, bas

ed o

n pe

rcen

t inv

asiv

e pl

ant c

over

. Rem

embe

r th

at th

is is

a q

ualit

ativ

e as

sess

men

t to

refle

ct

perc

ent c

over

acr

oss

the

MU

. Q

uant

itativ

e m

etho

ds u

sing

a tr

anse

ct li

ne o

r qua

drat

are

no

t req

uire

d. In

vasi

ve p

lant

cov

er is

one

of t

he

varia

bles

(alo

ng w

ith tr

ee c

anop

y co

ver)

use

d to

det

erm

ine

the

Tree

-iage

Cat

egor

y of

eac

h M

U.

If th

e in

vasi

ve c

over

thre

at o

bse

rvat

ion

is u

ncer

tain

and

you

are

uns

ure

whe

ther

the

MU

sho

uld

be

cod

ed a

s M

ediu

m o

r Hig

h, b

e co

nser

vati

ve a

nd c

ode

the

MU

as

Hig

h.

Ass

ign

eac

h M

U o

ne

of t

he

follo

win

g in

vasi

ve c

ove

r-

th

reat

val

ues

HIG

HM

Us

with

mor

e th

an 5

0% in

vasi

ve

spec

ies

cove

r

MED

IUM

MU

s w

ith b

etw

een

5% a

nd 5

0% in

vasi

ve

spec

ies

cove

r

LOW

MU

s w

ith le

ss th

an 5

% in

vasi

ve s

peci

es

cove

r

R

eco

rd t

he

spec

ies

cod

e o

f all

inva

sive

sp

ecie

s p

rese

nt

in

ord

er o

f ab

un

dan

ce

Reco

rd a

ll no

n-na

tive

inva

sive

spe

cies

pre

sent

in th

e M

U. T

hese

incl

ude

herb

aceo

us s

peci

es a

s w

ell a

s in

vasi

ve tr

ee s

peci

es s

uch

as E

nglis

h ho

lly.

Thes

e m

ust b

e re

cord

ed in

ord

er o

f abu

ndan

ce. S

ee A

ppen

dix

B fo

r a li

st o

f th

e m

ost c

omm

on in

vasi

ve s

peci

es fo

und

in P

uget

Sou

nd lo

wla

nd fo

rest

s.

W

hat

if I

obse

rve

a sp

ecie

s n

ot in

clu

ded

in th

e FL

AT

pla

nt l

ist?

Ente

r the

4-d

igit

sp

ecie

s co

de

and

reco

rd th

e co

mm

on a

nd s

cien

tific

na

me

in th

e no

tes

sect

ion.

Thi

s ap

plie

s to

ove

rsto

ry tr

ees

and

all

nati

ve

gro

und

and

non

-nat

ive

inva

sive

sp

ecie

s.

Phot

o cr

edit:

For

terr

a

22

Pla

nta

ble

This

att

ribut

e pr

ovid

es in

form

atio

n re

gard

ing

pres

ence

of a

vaila

ble

grow

ing

spac

e to

pro

mot

e tr

ee s

eedl

ing

rest

orat

ion

to h

elp

shad

e ou

t pot

entia

l in

vasi

ve s

peci

es. P

lant

able

are

as m

ay b

e ob

serv

ed a

s ba

re s

oil f

ollo

win

g cu

rren

t or f

utur

e in

vasi

ve re

mov

al, o

r lar

ge o

peni

ngs

in fo

rest

cov

er

resu

lting

from

ext

ensi

ve ro

ot ro

t or w

ind

failu

res.

Rec

ord

wh

eth

er a

n M

U is

“p

lan

tab

le”

usi

ng

th

e

fo

llow

ing

cri

teri

a

A “y

es” c

all i

ndic

ates

a p

riorit

y ar

ea fo

r pla

ntin

g th

at is

exp

osed

to fu

ll su

n an

d m

ay h

ave

bare

soi

l whe

re in

vasi

ve s

peci

es c

ould

est

ablis

h or

exp

and

if th

e ar

ea is

not

pla

nted

to n

ativ

e sp

ecie

s in

the

shor

t ter

m.

A “n

o” c

all i

ndic

ates

that

the

maj

ority

of g

row

ing

spac

e is

alre

ady

occu

pied

by

nat

ive

spec

ies

and

does

not

war

rant

imm

edia

te a

ctio

n.

Gro

un

d S

pec

ies

Gro

und

spec

ies

are

nativ

e he

rbs

and

shru

bs in

the

unde

rsto

ry o

f the

fore

st.

Thes

e in

clud

e bu

t are

not

lim

ited

to a

ll un

ders

tory

sh

rubs

and

her

bs fo

und

in th

e pl

ant l

ists

foun

d in

App

endi

x B.

Not

e th

at g

roun

d sp

ecie

s do

not

incl

ude

sapl

ings

or

rege

nera

tion

tree

s of

can

opy

spec

ies.

Use

the

four

lett

er

spec

ies

code

to re

pres

ent t

he

spec

ies.

Ex

ampl

e: P

olys

tichu

m

mun

itum

(PO

MU

)

Rec

ord

th

e tw

o m

ost

ab

un

dan

t g

rou

nd

co

ver

spec

ies

pre

sen

t in

th

e M

U

Gro

und

spec

ies m

ust b

e lis

ted

in o

rder

of a

bund

ance

:

•G

roun

d sp

ecie

s 1

– Pr

imar

y or

mos

t abu

ndan

t•

Gro

und

spec

ies

2 –

Seco

ndar

y or

sec

ond

mos

t abu

ndan

t. Phot

o cr

edit:

For

terr

a

24

Ap

pe

nd

ix

App

endi

x A

: Tre

e-ia

ge M

atri

x A

naly

sis

Iden

tifyi

ng a

nd p

riorit

izin

g ar

eas

in n

eed

of m

anag

emen

t is

a ke

y ou

tput

of

the

FLA

T pr

oces

s. A

ttrib

utes

ass

esse

d in

the

field

are

use

d to

pro

duce

a

qual

itativ

e va

lue

for t

wo

axes

of a

mat

rix. I

n Ki

ng C

ount

y, fo

rest

com

posi

tion

(y-a

xis)

and

fore

st th

reat

s (x

-axi

s) w

ere

used

. The

mat

rix c

ombi

nes

the

attr

ibut

e in

form

atio

n to

pro

duce

a c

lass

ifica

tion

valu

e fo

r eac

h M

U.

In th

e fig

ure

at th

e rig

ht,

valu

es 1

-3 re

pres

ent M

Us

with

a

tree

com

posi

tion

that

has

hi

gh e

colo

gica

l val

ue, a

nd s

o ar

e im

port

ant t

o pr

otec

t and

m

aint

ain.

Val

ues

2 an

d 3

also

re

pres

ent t

he p

rese

nce

of a

fo

rest

hea

lth th

reat

and

cou

ld

be p

riorit

ized

for r

esto

ratio

n or

m

aint

enan

ce. O

n th

e ot

her e

nd o

f th

e sp

ectr

um, a

MU

with

a v

alue

of

9 h

as a

hig

h th

reat

pre

senc

e an

d a

low

er tr

ee c

ompo

sitio

n,

and

ther

efor

e m

ay n

ot b

e no

t be

as h

igh

a pr

iorit

y fo

r man

agem

ent

actio

ns.

How

the

resu

lts o

f the

mat

rix a

naly

sis

are

used

in s

ubse

quen

t lan

d m

anag

emen

t dec

isio

ns is

up

to e

ach

FLA

T us

er. F

LAT

prov

ides

an

ecol

ogic

al

inpu

t for

land

man

agem

ent d

ecis

ion-

mak

ing.

The

com

bina

tion

of fi

eld

data

co

llect

ion,

flow

cha

rt p

roce

ssin

g, a

nd c

lass

ifica

tion

of M

Us

usin

g th

e m

atrix

ca

n be

use

d to

prio

ritiz

e fu

ture

man

agem

ent a

ctio

ns a

nd m

onito

ring.

25

App

endi

x B:

Com

mon

Pla

nt S

peci

es L

ist

TREE

S

Cod

e

Spec

ies

Scie

ntifi

c N

ame

A

CMA

Bigl

eaf m

aple

Acer

mac

roph

yllu

m

PR

EM

Bi

tter

che

rry

Pr

unus

em

argi

nata

POBA

Blac

k co

tton

woo

d

Popu

lus b

alsa

mife

ra

CR

DO

Blac

k ha

wth

orn

Cr

atae

gus d

ougl

asii

RH

PU

Ca

scar

a

Rh

amnu

s pur

shia

na

PS

ME

D

ougl

as fi

r

Pseu

dots

uga

men

zies

ii

ACG

L

Dou

glas

map

le

Ac

er g

labr

um

Q

UG

A

G

arry

oak

Que

rcus

gar

ryan

a

ABG

R

Gra

nd fi

r

Ab

ies g

rand

is

SAH

O

H

ooke

r’s w

illow

Salix

hoo

keria

na

ABP

R

Nob

le fi

r

Ab

ies p

roce

ra

FRLA

Ore

gon

ash

Fr

axin

us la

tifol

ia

MA

FU

Pa

cific

cra

bapp

le

M

alus

fusc

a

CON

U

Pa

cific

dog

woo

d

Corn

us n

utta

llii

A

RME

Pa

cific

mad

rone

Arbu

tus m

enzi

esii

A

BAM

Paci

fic s

ilver

fir

Ab

ies a

mab

alis

TABR

Paci

fic y

ew

Ta

xus b

revi

folia

SA

LU

Pa

cific

will

ow

Sa

lix lu

cida

BE

PA

Pa

per b

irch

Be

tula

pap

yife

ra

POTR

Qua

king

asp

en

Po

pulu

s tre

mul

oide

s

ALR

U

Re

d al

der

Al

nus r

ubra

PI

CO

Sh

ore

pine

Pinu

s con

tort

a

ALC

R

Sitk

a al

der

Al

nus c

rispa

ssp.

sinu

ata

PI

SI

Si

tka

spru

ce

Pi

cea

sitc

hens

is

SASI

Sitk

a w

illow

Salix

sitc

hens

is

ACC

I

Vine

map

le

Ac

er c

ircin

atum

TS

HE

W

este

rn h

emlo

ck

Ts

uga

hete

roph

ylla

THPL

Wes

tern

redc

edar

Thuj

a pl

icat

a

PIM

O

W

este

rn w

hite

pin

e Pi

nus m

ontic

ola

Q

UA

L

Whi

te o

ak

Q

uerc

us a

lba

27

HER

BACE

OUS

C

ode

Sp

ecie

s

Sc

ient

ific

Nam

e

FR

CH

Be

ach

stra

wbe

rry

Fr

agar

ia c

hilo

ensi

s

DIF

O

Bl

eedi

ng h

eart

Dic

entr

a fo

rmos

a

SIID

Blue

-eye

d gr

ass

Si

syrin

chiu

m id

ahoe

nsis

PT

AQ

Brac

ken

fern

Pter

idiu

m a

quili

num

CO

UN

Bunc

hber

ry

Co

rnus

una

lasc

hken

sis

TR

IF

Cl

over

spe

cies

Trifo

lium

sp.

EQ

AR

Co

mm

on h

orse

tail

Eq

uise

tum

arv

ense

H

ELA

Cow

par

snip

Her

acle

um la

natu

m

BLSP

Dee

r fer

n

Bl

echn

um sp

ican

t

CID

O

D

ougl

as w

ater

-hem

lock

Ci

cuta

dou

glas

ii

MA

DI

Fa

lse

lily-

of-t

he v

alle

y M

aian

them

um d

ilata

tum

SM

RA

Fa

lse

Solo

mon

’s s

eal

Smila

cina

race

mos

a

EPA

U

Fi

rew

eed

Epilo

bium

aug

ustif

oliu

m

TI

TR

Fo

amflo

wer

Tiar

ella

trifo

liata

D

IPU

Foxg

love

Dig

italis

pur

pure

a

TEG

R

Frin

gecu

p

Telli

ma

gran

diflo

ra

ALA

C

Hoo

ker’s

oni

on

Al

lium

acu

min

atum

FAH

E

Insi

de-o

ut fl

ower

Vanc

ouve

ria h

exan

dra

A

RUV

Ki

nnik

inni

ck

Ar

ctos

taph

ylos

uva

-urs

i

ATF

I

Lady

fern

Athy

rium

filix

-fem

ina

PO

GL

Li

coric

e fe

rn

Po

lypo

dium

gly

cyrr

hiza

AD

PE

M

aide

nhai

r fer

n

Adia

ntum

ped

atum

CL

PE

M

iner

s Le

ttuc

e

Clay

toni

a pe

rfolia

ta

GYD

R

Oak

fern

Gym

noca

rpiu

m d

ryop

teris

IR

TE

O

rego

n iri

s

Iris t

enax

O

XOR

O

xalis

; woo

d so

rrel

O

xalis

ore

gana

D

IFO

Paci

fic b

leed

ing

hear

t D

icen

tra

form

osa

H

YTE

Pa

cific

wat

erle

af

H

ydro

phyl

lum

tenu

ipes

TO

DI

Pa

cific

poi

son-

oak

To

xicod

endr

on d

iversi

lobu

m

PEFR

Palm

ate

colts

foot

Peta

site

s pal

mat

us

CO

MA

Pois

on h

emlo

ck

Co

nium

mac

ulat

um

SAXI

Saxi

frag

e sp

ecie

s

Saxi

fraga

sp.

A

NM

A

Pe

arly

eve

rlast

ing

An

apha

lis m

arga

ritac

ea

TOM

E

Pigg

y-ba

ck p

lant

Tolm

iea

men

zies

ii

GO

OB

Ra

ttle

snak

e pl

anta

in

Goo

dyer

a ob

long

ifolia

CO

SC

Sc

oule

r’s c

oryd

alis

Co

ryda

lis sc

oule

ri

LYA

M

Sk

unk

cabb

age

Ly

sich

iton

amer

ican

um

TRLA

Star

flow

er

Tr

ient

alis

latif

olia

VI

GL

St

ream

vio

let

Vi

ola

glab

ella

U

RDI

St

ingi

ng n

ettle

Urt

ica

dioi

ca

POM

U

Sw

ord

fern

Poly

stic

hum

mun

itum

LI

BO

Tw

inflo

wer

Line

ae b

orea

lis

ACT

R

Vani

lla le

af

Ac

hlys

trip

hylla

26

SHRU

BS

C

ode

Sp

ecie

s

Sc

ient

ific

Nam

e

RO

GY

Ba

ldhi

p ro

se

Ro

sa g

ymno

carp

a

CO

CO

Be

aked

haz

elnu

t

Cory

lus c

ornu

ta

RU

LE

Bl

ackc

ap

Ru

bus l

euco

derm

is

RILA

Blac

k go

oseb

erry

Ribe

s lac

ustr

e

VAM

E

Blac

k hu

ckle

berr

y

Vacc

iniu

m m

embr

anac

eum

SA

CA

Bl

ue e

lder

berr

y

Sam

bucu

s cae

rule

a

KAM

I

Bog

laur

el

Ka

lmia

mic

roph

ylla

A

NPO

Bog

rose

mar

y

Andr

omed

a po

lifol

ia

OPH

O

D

evil’

s cl

ub

O

plop

anax

hor

ridus

MA

NE

D

ull O

rego

n gr

ape

M

ahon

ia n

ervo

sa

VAO

V

Ever

gree

n hu

ckle

berry

Va

ccin

ium

ova

tum

MEF

E

Fals

e az

alea

Men

zies

ia fe

rrug

inea

A

RDI

G

oat’s

bea

rd

Ar

uncu

s dio

icus

A

RCO

Hai

ry m

anza

nita

Arct

osta

phyl

os c

olum

bian

a

SP

DO

Har

dhac

k; s

pira

ea

Sp

iraea

dou

glas

ii

VIED

Hig

hbus

h cr

anbe

rry

Vibu

rnum

edu

le

O

ECE

In

dian

plu

m

O

emla

ria c

eras

iform

is

LEG

R

Labr

ador

tea

Le

dum

gro

enla

ndic

um

PH

LE

M

ock

oran

ge

Ph

ilade

lphu

s lew

isii

RON

U

N

ootk

a ro

se

Ro

sa n

utka

na

H

OD

I

Oce

ansp

ray

H

olod

iscu

s dis

colo

r

PYM

Y

Ore

gon

boxw

ood

Pa

chis

tima

myr

sini

tes

PH

CA

Pa

cific

nin

ebar

k

Phys

ocar

pus c

apita

tus

RH

MA

Paci

fic rh

odod

endr

on

Rhod

oden

dron

mac

roph

yllu

m

M

YCA

Paci

fic w

ax m

yrtle

M

yric

a ca

lifor

nica

SA

RA

Re

d el

derb

erry

Sam

bucu

s rac

emos

a

VAPA

Red

huck

lebe

rry

Va

ccin

ium

par

vifo

lium

CE

SA

Re

d st

em c

eono

thus

Ce

anot

hus s

angu

ineu

m

CO

SE

Re

d os

ier d

ogw

ood

Corn

us se

ricea

RISA

Red-

flow

erin

g cur

rant

Ri

bes s

angu

ineu

m

GA

SH

Sa

lal

Gau

lther

ia sh

allo

n

RUSP

Salm

onbe

rry

Ru

bus s

pect

abili

s

AM

AL

Se

rvic

eber

ry

Am

elan

chie

r aln

ifolia

SOSI

Sitk

a m

ount

ain

ash

Sorb

us si

tche

nsis

SY

AL

Sn

owbe

rry

Sy

mph

oric

arpo

s alb

us

RIBR

Stin

k cu

rran

t

Ribe

s bra

cteo

sum

RO

PI

Sw

amp

rose

Rosa

pis

ocar

pa

MYG

A

Sw

eet g

ale

M

yric

a ga

le

MA

AQ

Tall

Ore

gon

grap

e

Mah

onia

aqu

ifoliu

m

RUPA

Thim

bleb

erry

Rubu

s par

viflo

rus

LO

IN

Tw

inbe

rry

Lo

nice

ra in

volu

crat

a

SPBE

Whi

te s

pire

a

Spire

a be

tulif

olia

var

.luci

da

28

HER

BACE

OUSCO

NT.

C

ode

Sp

ecie

s

Sc

ient

ific

Nam

e

A

QFO

Wes

tern

col

umbi

ne

Aqui

legi

a fo

rmos

a

TRO

V

Wes

tern

trill

ium

Trill

ium

ova

tum

A

SCA

Wild

gin

ger

As

orum

cau

datu

m

FRVE

Wild

stra

wbe

rry

Fr

agar

ia v

irgin

iana

FR

VE

W

oodl

and

stra

wbe

rry

Frag

aria

ves

ca

ACM

I

Yarr

ow

Achi

llea

mill

efol

ium

M

IGU

Yello

w m

onke

y-flo

wer

M

imul

us g

utta

tus

GRA

SS-LIK

E

G

RASS

Nat

ive

gras

ses

SC

AM

Am

eric

an b

ulru

sh

Sc

irpus

am

eric

anus

TY

LA

Ca

ttai

l

Ty

pha

latif

olia

JU

EN

D

agge

r-le

aved

rush

Ju

ncus

ens

ifoliu

s

CAD

E

Dew

ey’s

sed

ge

Ca

rex

dew

eyan

a

ELM

O

D

uneg

rass

Elym

us m

ollis

SC

AC

H

ards

tem

bul

lrush

Sc

irpus

acu

tus

FE

ID

Id

aho

fesc

ue

Fe

stuc

a id

ahoe

nsis

CA

LY

Ly

ngby

e’s

sedg

e

Care

x ly

ngby

ei

CA

OB

Sl

ough

sed

ge

Ca

rex

obnu

pta

SC

MI

Sm

all-f

ruite

d bu

llrus

h Sc

irpus

mic

roca

rpus

CA

PA

Th

ick

head

ed s

edge

Ca

rex

pach

ysta

chya

D

ECE

Tu

fted

hai

rgra

ss

D

esch

amps

ia c

espi

tosa

W

EFE

W

este

rn fe

scue

Fest

uca

occi

dent

alis

VIN

ES

LOH

I

Hai

ry h

ones

uckl

e

Loni

cera

his

pidu

la

LO

CI

O

rang

e ho

neys

uckl

e Lo

nice

ra c

ilios

a

RUU

R

Trai

ling

blac

kber

ry

Rubu

s urs

inus

29

NON-N

ATIVEH

ERBS

ANDSHRU

BS

Code

Sp

ecie

s

Sc

ient

ific

Nam

e

Wee

d Cl

ass

SOD

U

Bitt

ersw

eet n

ight

shad

e So

lanu

m d

ulca

mar

a W

eed

of c

once

rn *

*PO

BO

Bohe

mia

n kn

otw

eed

Poly

gonu

m x

boh

emic

um

Non

-reg

ulat

ed *

**CI

VU

Bull

this

tle

Ci

rsiu

m v

ulga

re

N

on-r

egul

ated

BUD

A

Butt

erfly

bus

h

Bu

ddle

ia d

avid

ii

Non

-reg

ulat

edCI

AR

Cana

da th

istle

Cirs

ium

arv

ense

Non

-reg

ulat

edCL

VI

Clem

atis

; Old

man

’s b

eard

Cl

emat

is v

italb

a

Non

-reg

ulat

ed

VIM

I Co

mm

on p

eriw

inkl

e Vi

nca

min

or

N

one

TAVU

Co

mm

on ta

nsy

Ta

nace

tum

vul

gare

N

on-r

egul

ated

D

IFU

Co

mm

on te

asel

Dip

sacu

s ful

lonu

m

Non

-reg

ulat

edRA

RE

Cree

ping

but

terc

up

Ranu

ncul

us re

pens

W

eed

of c

once

rnH

EHE

Engl

ish

ivy

H

eder

a he

lix

N

on-r

egul

ated

MYS

P Eu

rasi

an w

ater

milf

oil

Myr

ioph

yllu

m sp

icat

um

Non

-reg

ulat

edRU

LA

Ever

gree

n bl

ackb

erry

Ru

bus l

acin

iatu

s

Non

-reg

ulat

edLY

VU

Gar

den

loos

estr

ife

Lysi

mac

hia

vulg

aris

Cl

ass

B*A

LPE

Gar

lic m

usta

rd

Al

liaria

pet

iola

ta

Cl

ass

A*

HEM

A

Gia

nt h

ogw

eed

H

erac

leum

man

tega

zzia

num

Cla

ss A

HIE

R H

awkw

eeds

Hie

raci

um sp

.

Clas

s B

GER

O

Her

b Ro

bert

Ger

aniu

m ro

bert

ianu

m

Non

-reg

ulat

edRU

AR

Him

alay

an b

lack

berr

y Ru

bus a

rmen

iacu

s

Non

-reg

ulat

edPL

CU

Japa

nese

kno

twee

d Po

lygo

num

cus

pida

tum

N

on-r

egul

ated

CA

SE

Mor

ning

Glo

ry; H

edge

/Fie

ld b

indw

eed

Cal

yste

gia

sepi

um

Wee

d of

con

cern

LEVU

O

xeye

dai

sy

Le

ucan

them

um v

ulga

re

Non

-reg

ulat

edCO

MA

Po

ison

hem

lock

Coni

um m

acul

atum

N

on-r

egul

ated

IMG

L Po

licem

an’s

hel

mit

Im

patie

ns g

land

ulife

ra

Clas

s B

LYSA

Pu

rple

loos

estr

ife

Ly

thru

m sa

licar

ia

Cl

ass B

PHA

R Re

ed-c

anar

y gr

ass

Ph

alar

is a

rund

inac

ea

Non

-reg

ulat

edCY

SC

Scot

ch b

room

Cytis

us sc

opar

ius

N

on-r

egul

ated

POH

Y Sm

artw

eed;

wat

er-p

eppe

r Po

lygo

num

hyd

ropi

pero

ides

N

one

SPA

L Sm

ooth

cor

dgra

ss

Spar

tina

alte

rnifl

ora

Clas

s ACE

ST

Spot

ted

knap

wee

d

Cent

aure

a st

oebe

Clas

s B

PO

RE

Sulfu

r cin

quef

oil

Po

tent

illa

rect

a

Cl

ass

BSE

JA

Tans

y ra

gwor

t

Sene

cio

jaco

baea

Clas

s B

LAG

A

Yello

w a

rcha

ngel

Lam

ium

gal

eobd

olon

N

on-r

egul

ated

IRPS

Ye

llow

flag

iris

Iris p

seud

acor

us

N

on-r

egul

ated

* W

ashi

ngto

n St

ate

Wee

d Cl

ass A

and

B -

cont

rol r

equi

red

in K

ing

Coun

ty**

Wee

d of

con

cern

: app

lies t

o Ki

ng C

ount

y on

ly; N

o st

ate

clas

sific

atio

n; c

ontr

ol

reco

mm

ende

d bu

t not

requ

ired

in K

ing

Coun

ty**

* Non

-reg

ulat

ed C

lass

B a

nd C

nox

ious

wee

ds; c

ontr

ol re

com

men

ded

but n

ot re

quire

d in

Ki

ng C

ount

y

3131

App

endi

x C:

How

To

Mea

sure

Tre

e D

iam

eter

-at-

Brea

st H

eigh

t

The

stan

dard

pro

toco

l for

mea

surin

g di

amet

er a

t bre

ast h

eigh

t is

as fo

llow

s:

•U

sing

a d

iam

eter

tape

, mea

sure

the

diam

eter

at b

reas

t hei

ght t

o th

e ne

ares

t inc

hes.

•Br

east

hei

ght i

s co

nsid

ered

to b

e 4.

5 fe

et a

bove

the

grou

nd o

n th

e up

hill

side

of t

he tr

ee.

•Fo

r tre

es w

ith

swel

lings

, bum

ps, d

epre

ssio

ns, a

nd b

ranc

hes

at D

BH,

mea

sure

dia

met

er a

bove

the

irre

gula

rity

whe

re th

e tr

unk

is n

o lo

nger

af

fect

ed.

Imag

e cr

edit:

ww

w.w

oodl

ands

tew

ards

hip.

org

Imag

e cr

edit:

ww

w.w

oodl

ands

tew

ards

hip.

org

30

NO

N-N

ATI

VE

TREE

S

Code

Sp

ecie

s

Sc

ient

ific

Nam

e

Wee

d Cl

ass

M

ALU

S A

pple

M

alus

sp.

N

one

CRM

O

Com

mon

haw

thor

n Cr

atae

gus d

ougl

asii

Wee

d of

con

cern

CUPP

Cy

pres

s

Cu

pres

sus s

p.

N

one

ROPS

Bl

ack

locu

st

Ro

bini

a ps

eudo

acac

ia L

. N

one

PRL

A

Engl

ish

laur

el

Pr

unus

laur

ocer

asus

W

eed

of c

once

rnIL

AQ

En

glis

h ho

lly

Ile

x aq

uifo

lium

Wee

d of

con

cern

SOA

U

Euro

pean

mou

ntai

n as

h So

rbus

auc

upar

ia

W

eed

of c

once

rnPO

NI

Lom

bard

y po

plar

Popu

lus n

igra

L.

N

one

ACR

U

Red

map

le

Ac

er ru

brum

Non

eA

IAL

Tree

of h

eave

n

Aila

nthu

s alti

ssim

a

Non

-reg

ulat

ed

PRU

N

Wild

plu

m

Pr

unus

sp.

N

one

ORN

AM

O

ther

har

dwoo

d or

nam

enta

ls

Non

e

* W

ashi

ngto

n St

ate

Wee

d Cl

ass A

and

B -

cont

rol r

equi

red

in K

ing

Coun

ty**

Wee

d of

con

cern

: app

lies t

o Ki

ng C

ount

y on

ly; N

o st

ate

clas

sific

atio

n; c

ontr

ol

reco

mm

ende

d bu

t not

requ

ired

in K

ing

Coun

ty**

* Non

-reg

ulat

ed C

lass

B a

nd C

nox

ious

wee

ds; c

ontr

ol re

com

men

ded

but n

ot re

quire

d in

Ki

ng C

ount

y

32

App

endi

x D

: H

ow T

o U

se A

Den

sito

met

er (M

oose

horn

)

Den

sito

met

ers

are

smal

l sig

htin

g in

stru

men

ts w

ith c

ross

hairs

and

a b

ubbl

e le

vel t

hat a

llow

s th

e ob

serv

er to

det

erm

ine

whe

ther

can

opy

is p

rese

nt d

irect

ly

over

head

.

Sinc

e th

e de

nsito

met

er m

easu

res

cano

py c

over

(pre

senc

e or

abs

ence

) at a

si

ngle

poi

nt, m

ultip

le s

ampl

e po

ints

mus

t be

mea

sure

d to

obt

ain

a ca

nopy

co

ver e

stim

ate.

Usu

ally

, sam

ple

poin

ts a

re s

pace

d al

ong

a tr

anse

ct o

r ar

rang

ed in

a g

rid p

atte

rn to

obt

ain

an e

stim

ate

for a

spe

cifie

d ar

ea.

Not

e, b

ecau

se F

LAT

is a

rapi

d as

sess

men

t an

d da

ta is

not

col

lect

ed

quan

titat

ivel

y, fi

eld

staff

sho

uld

only

use

a d

ensi

tom

eter

(moo

se h

orn)

as

a qu

ality

con

trol

mea

sure

or t

rain

ing

exer

cise

to c

alib

rate

est

imat

es.

For a

dditi

onal

info

rmat

ion

go to

: htt

p://

ww

w.fo

rest

ry-s

uppl

iers

.com

/pro

duct

_pa

ges/

View

_Cat

alog

_Pag

e.as

p?m

i=65

121&

title

=GRS

+Den

sito

met

er

Imag

e cr

edit:

w

ww

.fore

stry

-sup

plie

rs.c

om

33

App

endi

x E:

How

To

Use

A C

ompa

ss

Impo

rtan

t Voc

abul

ary

Mag

netic

Nor

thIt

is im

port

ant t

o kn

ow th

at m

agne

tic n

orth

is a

lway

s m

ovin

g an

d di

ffers

at

diff

eren

t loc

atio

ns o

n th

e ea

rth.

Com

pass

nee

dles

alw

ays

poin

t tow

ard

mag

netic

nor

th.

True

Nor

thTr

ue n

orth

is th

e ge

ogra

phic

nor

th.

It do

es n

ot m

ove,

it is

the

fixed

loca

tion

on th

e ea

rth

whe

re th

e N

orth

Pol

e is

loca

ted.

Dec

linat

ion

This

refe

rs to

the

diffe

renc

e in

deg

rees

bet

wee

n m

agne

tic n

orth

and

true

no

rth.

Her

e in

Sea

ttle

in 2

012

the

decl

inat

ion

is a

bout

17

degr

ees

East

. Th

is c

an c

hang

e ov

er ti

me

and

loca

tion.

Man

y co

mpa

sses

, inc

ludi

ng th

ose

prov

ided

in th

is m

onito

ring

prog

ram

, are

adj

uste

d fo

r dec

linat

ion.

If u

sing

yo

ur o

wn

com

pass

, it i

s re

com

men

ded

to h

ave

a co

mpa

ss th

at a

llow

s yo

u to

ad

just

for t

he d

eclin

atio

n so

you

r rea

ding

is a

ccur

ate

and

no m

ath

is n

eede

d.

Anat

omy

of a

Com

pass

Imag

e cr

edit:

ww

w.o

ffro

ad-e

d.co

m

35

App

endi

x F:

Use

of I

ncre

men

t Bor

ers

to D

eter

min

e Tr

ee A

ge

Tree

cor

es c

an b

e us

ed to

det

erm

ine

the

aver

age

age

of a

fore

st s

tand

or fi

nd a

n ex

act

age

of a

sin

gle

tree

. An

incr

emen

t bor

er is

the

leas

t inv

asiv

e m

etho

d us

ed to

cou

nt tr

ee ri

ngs.

Th

is in

volv

es ta

king

a s

mal

l (0.

2 in

ch d

iam

eter

) st

raw

-like

sam

ple

from

the

bark

to th

e pi

th

of a

tree

. Tho

ugh

the

hole

is s

mal

l, it

can

still

in

trod

uce

dise

ase

into

the

trun

k.

1.

Ass

embl

e th

e in

crem

ent b

orer

by

atta

chin

g th

e bi

t to

the

hand

le; s

et th

e ex

trac

tor a

side

.

2.

Dril

l in

tow

ard

the

cent

er o

f the

tree

at b

reas

t hei

ght (

4 ½

ft.).

3.

Whe

n yo

u ha

ve g

one

far e

noug

h to

reac

h th

e ce

nter

of t

he tr

ee, i

nser

t th

e ex

trac

tor a

nd re

vers

e th

e bi

t one

full

turn

.

4.

Slow

ly re

mov

e th

e ex

trac

tor.

If th

e in

crem

ent c

ore

does

not

com

e ou

t the

fir

st ti

me,

try

agai

n.

5.

Onc

e th

e co

re is

out

, set

it a

nd th

e ex

trac

tor a

side

.

6.

Imm

edia

tely

rem

ove

the

incr

emen

t bor

er fr

om th

e tr

ee, b

efor

e it

beco

mes

stu

ck.

7.

Onc

e th

e in

crem

ent b

orer

is b

ack

out o

f the

tree

, cou

nt th

e rin

gs o

n th

e in

crem

ent c

ore

to d

eter

min

e br

east

hei

ght a

ge. A

dd s

ever

al y

ears

to

estim

ate

tota

l age

.

For t

he d

etai

led

expl

anat

ion

on in

crem

ent b

orer

use

, go

to:

http

://fo

rest

andr

ange

.org

/Virt

ual%

20Cr

uise

r%20

Vest

/ le

sson

s/le

sson

_06/

Less

on_6

_PD

F.pd

f

Phot

o cr

edit:

ww

w.re

dorb

it.co

m

Phot

o cr

edit:

ww

w.fo

rest

ry-s

uppl

iers

.com

34

How

to d

eter

min

e ca

rdin

al d

irect

ions

1.

Stan

ding

at p

lot c

ente

r, ho

ld

your

com

pass

in y

our h

and

so th

at th

e ba

sepl

ate

is le

vel

and

the

lid is

ope

n at

abo

ut 6

0 de

gree

s. H

old

it ou

t in

fron

t of

you

with

arm

ext

ende

d ha

lf w

ay

and

the

com

pass

at e

ye le

vel.

Yo

u w

ill b

e lo

okin

g in

to th

e m

irror

, not

dire

ctly

at t

he d

ial.

2.

Turn

the

com

pass

dia

l unt

il th

e di

rect

ion

you

wan

t to

go is

lo

cate

d at

the

top

of th

e di

al.

For e

xam

ple,

to g

o ea

st, “

E” w

ould

be

loca

ted

at th

e to

p of

the

dial

.

3.

Onc

e yo

u ha

ve th

e di

rect

ion

of tr

avel

det

erm

ined

, rot

ate

your

bod

y un

til

the

red

orie

ntin

g ar

row

on

the

com

pass

hou

sing

(out

line

of a

n ar

row

) lin

es u

p w

ith th

e m

agne

tic n

eedl

e (t

he fl

oatin

g re

d m

agne

tic a

rrow

).

4.

Use

the

line

of s

ight

(tria

ngul

ar n

otch

in to

p of

com

pass

lid)

as

your

poi

nt

of d

irect

ion.

Hav

e yo

ur m

onito

ring

team

mem

ber s

tand

at t

he e

dge

of

the

plot

and

gui

de th

em le

ft/r

ight

to a

lign

with

you

r car

dina

l dire

ctio

n re

adin

g. O

nce

in th

e rig

ht p

ositi

on, h

ang

flagg

ing

to m

ark

the

plot

edg

e at

the

card

inal

dire

ctio

n. R

epea

t ste

ps fo

r eac

h of

the

card

inal

dire

ctio

ns.

How

to ta

ke a

bea

ring

1.

Hol

d yo

ur c

ompa

ss in

you

r han

d so

that

the

base

plat

e is

leve

l and

the

lid

is o

pen

at a

bout

60

degr

ees

Hol

d it

out i

n fr

ont o

f you

with

arm

ext

ende

d ha

lf w

ay a

nd th

e co

mpa

ss a

t eye

leve

l. Yo

u w

ill b

e lo

okin

g in

to th

e m

irror

, no

t dire

ctly

at t

he d

ial.

2.

Turn

you

r bod

y un

til th

e de

sire

d ta

rget

(e.g

. a D

ougl

as fi

r) is

in s

ight

. U

se

the

tria

ngul

ar n

otch

in th

e lid

of t

he c

ompa

ss a

s yo

ur c

ente

r of s

ight

, ai

min

g it

at th

e ce

nter

of t

he tr

ee.

3.

Mak

e su

re th

at th

e si

ghtin

g lin

e in

the

mirr

or ru

ns th

roug

h th

e m

iddl

e of

th

e ca

psul

e vi

ew.

4.

Rota

te th

e co

mpa

ss d

ial u

ntil

the

red

orie

ntin

g ar

row

on

the

com

pass

ho

usin

g (o

utlin

e of

an

arro

w) l

ines

up

with

the

mag

netic

nee

dle

(the

flo

atin

g re

d m

agne

tic a

rrow

).

5.

You

can

read

the

bear

ing

in d

egre

es.

Read

from

the

top

of th

e co

mpa

ss

(whe

re th

e co

mpa

ss a

nd m

irror

ed li

d m

eet)

, the

re is

a s

mal

l ind

icat

or

line

(dire

ctio

n of

trav

el a

rrow

). N

ote:

do

not g

ive

card

inal

dire

ctio

ns; g

ive

exac

t deg

rees

(ex.

76

degr

ees)

.

36

Dat

a A

ttri

bu

teD

ata

Fiel

dEx

pla

nat

ion

Site

Nam

eSI

TE N

AM

EG

IS id

enti

fier

Man

agem

ent U

nit N

um-

ber

MU

_NO

GIS

iden

tifie

r

Dat

e of

dat

a co

llect

ion

DA

TE

Ass

esso

rs in

itia

lsC

REW

Land

cove

rLA

ND

CO

V

Fore

sted

FOR

≥ 2

5% fo

rest

can

opy

Nat

ural

Are

aN

AT

< 2

5% fo

rest

can

opy

Op

en W

ater

WA

TN

o w

ood

y ve

get

atio

n

Har

dsc

ape

HS

Build

ing

s, p

arki

ng

Lans

dsc

aped

LS

Land

scap

ed, m

echa

nica

lly

mai

ntai

ned

Ag

e C

lass

AG

ECLA

SS

cate

gor

y 1

10-

29 y

ears

cate

gor

y 2

230

-49

year

s

cate

gor

y 3

350

-99

year

s

cate

gor

y 4

410

0 +

yea

rs

Ove

rsto

ry S

pec

ies

OV

R1_S

PCO

vers

tory

sp

ecie

s, m

ost a

bun

-d

ant d

omin

ant o

r cod

omin

ant

>20

ft)

Ove

rsto

ry S

ize

OV

R1_S

IZE

Ove

rsto

ry D

BH s

ize

clas

s

cate

gor

y 1

10

-5" D

BH

cate

gor

y 2

26

- 10"

DBH

cate

gor

y 3

311

- 20

" DBH

cate

gor

y 4

421

"+ D

BH

Seco

nd O

vers

tory

Sp

e-ci

es

OV

R2_S

PC2n

d o

vers

tory

sp

ecie

s, in

ord

er

of a

bun

dan

ce c

odom

inan

t >

20ft

App

endi

x G

: FLA

T D

efini

tions

At-

A-G

lanc

e

37

Dat

a A

ttri

bu

teD

ata

Fiel

dEx

pla

nat

ion

Seco

nd O

vers

tory

Siz

eO

VR2

_SIZ

EO

vers

tory

DBH

siz

e cl

ass,

see

si

ze c

lass

cha

rt a

bov

e

Thir

d O

vers

tory

Sp

ecie

sO

VR3

_SPC

3rd

ove

rsto

ry s

pec

ies,

if

pre

sent

, in

ord

er o

f ab

und

ance

co

dom

inan

t >

20f

t

Thir

d O

vers

tory

Siz

eO

VR3

_SIZ

EO

vers

tory

DBH

siz

e cl

ass,

see

si

ze c

lass

cha

rt a

bov

e

Stoc

king

ST

OC

KIN

GC

anop

y co

ver e

stim

ate,

as

view

ed d

irec

tly

abov

e

cate

gor

y 0

0Le

ss th

an 1

0% c

anop

y co

ver

cate

gor

y 1

110

- 39

% c

anop

y co

ver

cate

gor

y 2

240

- 69

% c

anop

y co

ver

cate

gor

y 3

370

% +

can

opy

cove

r

Man

agem

ent U

nit C

om-

pos

itio

n M

U_C

MP

Hig

h C

omp

osit

ion

H

> 5

0% c

onife

r/m

adro

ne O

R

≤50

% c

onife

r/m

adro

ne w

ith

no c

apac

ity

for r

esto

rati

on (i

n-cl

udes

wet

land

s)

Med

ium

Com

pos

itio

n M

1-50

% c

onife

r/m

adro

ne w

ith

cap

acit

y to

sup

por

t res

tora

tion

to

H O

R

<25

% n

ativ

e co

ver w

ith

cap

acit

y to

rest

ore

up to

50%

co

nife

r

Low

Com

pos

itio

nL

< 2

5% n

ativ

e co

ver w

ith

cap

ac-

ity

for f

ull r

esto

rati

on p

lant

ing

O

R

No

coni

fer/

mad

rone

wit

h ca

-p

acit

y fo

r ful

l res

tora

tion

39

Dat

a A

ttri

bu

teD

ata

Fiel

dEx

pla

nat

ion

Nat

ive

Shru

bs

and

Her

bs

Spec

ies

GRD

1_SP

CN

ativ

e sh

rub

s an

d h

erb

s, m

ost

abun

dan

t

Nat

ive

Shru

bs

and

Her

bs

Spec

ies

GRD

2_SP

CSe

cond

nat

ive

shru

bs

and

he

rbs

in o

rder

of a

bun

dan

ce

Inva

sive

Sp

ecie

s IN

V1_

SPC

Non

-nat

ive

spec

ies,

mos

t ab

un-

dan

t

Inva

sive

Sp

ecie

s IN

V2_

SPC

Seco

nd n

on-n

ativ

e sp

ecie

s in

or

der

of a

bun

dan

ce

Inva

sive

Sp

ecie

s IN

V3_

SPC

Thir

d n

on-n

ativ

e sp

ecie

s in

or

der

of a

bun

dan

ce

Inva

sive

Sp

ecie

s IN

V4_

SPC

Four

th n

on-n

ativ

e sp

ecie

s in

or

der

of a

bun

dan

ce

Inva

sive

Sp

ecie

s IN

V5_

SPC

Fift

h no

n-na

tive

sp

ecie

s in

or

der

of a

bun

dan

ce

Tota

l Inv

asiv

e C

over

INV

CO

VTo

tal I

nvas

ive

Cov

er

Hig

h co

ver

H>

50%

Med

ium

cov

erM

5% -

50%

Low

cov

erL

<5%

Not

esN

OTE

SU

niq

ue s

ite

cond

itio

ns a

nd

othe

r dom

inan

t tre

es p

rese

nt.

38

Dat

a A

ttri

bu

teD

ata

Fiel

dEx

pla

nat

ion

Low

Vig

orLO

W V

IGO

RC

onife

r: Li

ve C

row

n ≤

40%

, Y o

r N

; Har

dw

ood

dec

line:

Top

Die

-b

ack

or S

nag

s ≥

5%

, Y o

r N

Mec

hani

cal T

ree

Failu

reFA

ILU

REM

echa

nica

l Tre

e Fa

ilure

in ≥

1%

of

MU

, Y

or N

) ex.

: win

dth

row

, la

ndsl

ide

Root

Rot

ROO

T RO

TRo

ot R

ot P

ocke

ts p

rese

nt,

Y or

N

Mis

tlet

oeM

ISTL

ETO

EM

istl

etoe

pre

sent

, Y o

r N

Bare

Soi

lBA

RE S

OIL

≥1%

Bar

e So

il p

rese

nt fr

om re

-ce

nt d

istu

rban

ce, e

rosi

on, e

tc.

Y or

N

Oth

er

OTH

ERPr

esen

t in

≥ 1

% o

f MU

, Y

or N

*N

ote

in c

omm

ents

req

uire

d

Reg

ener

atio

n Sp

ecie

s RG

N1_

SPC

Reg

ener

atio

n sp

ecie

s <

20FT

H

T, in

ord

er o

f ab

und

ance

Seco

nd R

egen

erat

ion

Spec

ies

RGN

2_SP

CRe

gen

erat

ion

spec

ies

<20

FT

HT,

in o

rder

of a

bun

dan

ce

Reg

ener

atio

n St

ocki

ng

Cla

ssRG

N_T

PA

cate

gor

y 1

10-

49 T

PA (>

30

ft s

pac

ing

)

cate

gor

y 2

250

-149

TPA

(bet

wee

n 30

and

16

ft s

pac

ing

)

cate

gor

y 3

315

0+ T

PA (

< 1

6 ft

sp

acin

g)

Plan

tab

le S

pac

ePL

AN

TABL

ESu

itab

le g

row

ing

sp

ace

for r

es-

tora

tion

pla

ntin

g?

Y or

N

40

Re

fere

nce

s

Arn

ey, J

.D.,

K.S.

Miln

er, a

nd B

.L. K

lein

henz

. 200

8. B

iom

etric

s of

For

est I

nven

tory

, Fo

rest

Gro

wth

, and

For

est P

lann

ing.

Tec

hnic

al R

epor

t No.

12.

For

est B

iom

etric

s Re

sear

ch In

stitu

te.

Jenn

ings

S.B

., N

.D. B

row

n, a

nd D

. She

il. 1

999.

“Ass

essi

ng fo

rest

can

opie

s an

d un

ders

tory

illu

min

atio

n: c

anop

y cl

osur

e, c

anop

y co

ver a

nd o

ther

mea

sure

s.”

Fore

stry

, Vol

72,

No.

1 pp

59-

73.

Korh

onen

, L.,

K.T.

Kor

hone

n, M

. Rau

tiain

en, a

nd P

. Ste

nber

g. 2

006.

“Est

imat

ion

of fo

rest

can

opy

cove

r: A

com

paris

on o

f fiel

d m

easu

rem

ent t

echn

ique

s.” S

ilva

Fenn

ica

40(4

): 57

7–58

8.

Wol

f, K.

et a

l. 20

13 [I

n Pr

ess]

. For

est L

ands

cape

Ass

essm

ent T

ool G

ener

al

Tech

nica

l Rep

ort.

USD

A F

ores

t Ser

vice

PN

W R

esea

rch

Stat

ion.

Mon

itorin

g D

ata

Colle

ctio

n M

etho

ds. 2

013.

Dev

elop

ed b

y th

e G

reen

City

Pa

rtne

rshi

ps fo

r the

Reg

iona

l Sta

ndar

dize

d M

onito

ring

Proj

ect.

Ben

Mea

dow

s. U

RL: w

ww

.ben

mea

dow

s.co

m

The

Nat

iona

l Lea

rnin

g Ce

nter

for P

rivat

e Fo

rest

and

Ran

ge L

ando

wne

rs.

URL

: ww

w.fo

rest

andr

ange

.org

/Virt

ual%

20Cr

uise

r%20

Vest

/less

ons/

less

on_0

6/Le

sson

_6_P

DF.

pdf

Woo

dlan

d St

ewar

dshi

p O

nlin

e Re

sour

ce. 2

011.

Uni

vers

ity o

f Min

neso

ta.

URL

: ww

w.w

oodl

ands

tew

ards

hip.

org

No

tes

Pacific Northwest Research Station

Website http://www.fs.fed.us/pnw/Telephone (503) 808–2592Publication requests (503) 808–2138FAX (503) 808–2130E-mail [email protected] address Publications Distribution Pacific Northwest Research Station P.O. Box 3890 Portland, OR 97208–3890

U.S. Department of Agriculture Pacific Northwest Research Station 1220 SW 3rd Ave. P.O. Box 3890 Portland, OR 97208–3890

Official Business Penalty for Private Use, $300

Forest Landscape Assessment Tool (FLAT)

Documents