United States       Office of Research and     EPA/620/R-94/028
 Environmental Protection   Development         November 1994
 Agency          Washington DC 20460
Forest Health
Monitoring

1991 Statistical
Summary
                                 >

                                /
 Environmental Monitoring and
 Assessment Program

-------
                                          EPA/620/R-94/028
                                          November 1994
FOREST  HEALTH  MONITORING
 1991  STATISTICAL  SUMMARY
                     Contract 68-DO-0106
                        Project Officer

                      Ralph Baumgardner
        Atmospheric Research and Exposure Assessment Laboratory
                U.S. Environmental Protection Agency
             Research Triangle Park, North Carolina 27711

                     Technical Director

                    Samuel A. Alexander

                U.S. Environmental Protection Agency
           Environmental Monitoring and Assessment Program
                        EMAP Center
                 Research Triangle Park, NC 27709

                           and

                     Program Manager

                     Joseph E. Barnard

                    U.S.D.A. Forest Service
                  U.S. Forest Service Laboratory
                Research Triangle Park, NC 27709
        Atmospheric Research and Exposure Assessment Laboratory - Research Triangle Park, NC 27711
                 Environmental Research Laboratory - Corvallis, OR 97333
              Environmental Monitoring Systems Laboratory - Las Vegas, NV 89119
                                              Printed on Recycled Paper

-------
FOREST HEALTH MONITORING

 1991  STATISTICAL SUMMARY
              Approved by
  Joseph E. Barnard          Samuel A. Alexander
  National Program Manager      Technical Director
  Forest Health Monitoring       Forest Health Monitoring

-------
                                                     Notice
             Funding for this research was provided by the Environmental Protection Agency to the U.S. Forest
             Service  through Interagency Agreement  Number DW12934170-5,  to  ManTech Environmental
             Technology, Inc. through Contract  Numbers  68-DO-0106 (RIP)  and 68-C8-0006  (Corvallis), to
             Lockheed Engineering & Sciences Company, Las Vegas, NV through Contract Number 68-CO-0049,
             to the University of Nevada-Las Vegas through Cooperative Agreement Number CR818526-01-0, to
             the Duke University through Cooperative Agreement Number CR-819634-01, to North Carolina State
             University through Cooperative Agreement Number 58-6645-0-002, to Statistical Consulting Service
             through P.O. Number 2B0270NASA, to the National Oceanic and Atmospheric Administration through
             Interagency Agreement, to Bionetics Corporation through Contract Number 68-03-3532. It has been
             subjected to the Agency's peer and administrative review, and it has been approved for publication as
             an EPA document.Mention of trade names or commercial products does not constitute endorsement
             or recommendation for use.

             Mention of trade names is for the information of the reader and does not constitute endorsement by
             the U.S. Government.

^           This  report represents data from one year of field operations  of the Environmental Monitoring  and
             Assessment Program  (EMAP).   Because the probability-based scietific design  used by EMAP
K            necesitates multiple years of sampling, there is uncertainty associated with these data. This uncertainty
             will decrease as the full power of the approach is realized.  Similarly, temporal changes and trends
«            cannot be reported, as these require multiple years of observation. Appropriate precautions should be
 {            exercised when using this information for policy, regulatory or legislative purposes.
^
v            Proper citation of this document is:
^                   Forest Health Monitoring. 1992. Forest Health Monitoring 1991 Statistical Summary. EPA/xxx/
^-                   /  .  U.S.  Environmental  Protection Agency, Washington, DC.
                                                        in

-------
                                       Forward

This document presents a statistical view of Forest Health Monitoring program data for 1991. It is a
limited summary of a few measurements in a few areas of the United States; its scope and content are
defined by the current (1991) stage of implementation and therefore it does not necessarily reflect how
data collection or analyses will be conducted when the program is fully implemented.  Because Forest
Health Monitoring is not fully implemented and because 1991 was the first year of data collection,
summaries such as this are necessarily incomplete.  However, responses to an early summary such
as this will help to define needed improvements to Forest Health Monitoring.

Forest Health Monitoring is an interagency program that has headquarters at the U.S. Forest Service,
Southeast Forest Experiment Station, Box 12254, Research Triangle Park,  NC 27709. Comments
concerning the program are encouraged and should be forwarded to the program management staff.

Program Manager:
       Joseph  E. Barnard
       U.S. Forest Service

Technical Director-
       Samuel A. Alexander
       U.S. Environmental Protection Agency

Deputy, Applications:
       Robert C. Loomis
       U.S. Forest Service

Deputy, Research:
       Kenneth W. Stotte
       U.S. Forest Service

Deputy Technical Director:
       Craig J. Palmer
       U.S. Environmental Protection Agency


National Program Management Coordinators:

       James Colby and James Francis, Bureau of Land Management
       Thomas Martin, Fish and Wildlife Service
       Joe Abrell, National Park Service
       Robert Smith, Soil Conservation Service
       Elizabeth Smith, Tennessee Valley Authority
The following individuals are recognized as the U. S. EPA Work Assignment Managers:

       Ralph Baumgardner, Atmospheric Research and Exposure Assessment Laboratory, Research
              Triangle Park, NC.
       Craig Palmer, Environmental Monitoring Systems Laboratory, Las Vegas, NV.
       Spencer Peterson, Environmental Research Laboratory, Corvallis, OR.
                                           IV

-------
                                Executive Summary

Introduction

        In response to the diverse and growing public concerns about potential human impacts on our
environment, the U.S. Forest Service and the U.S Environmental Protection Agency have initiated a
cooperative national program of Forest Health Monitoring (FHM). Although not established in all parts
of the nation, FHM has already provided valuable information about forest "health" in some areas. The
intention is to provide a many-sided view of forest health that will assist the public and other decision-
makers in setting priorities  and making informed choices aimed at reducing the ecological  risks of
human impacts.

        The purposes of this report are: to describe the current approach and activities; to  summarize
the data that were collected in  1991, and; to outline some new directions that are being explored for
possible implementation later. This report  is also a test of an assessment process by which data from
many sources will be brought together for analyses.  It is anticipated that suggestions received from
the readers will enable FHM to tailor future reports to better meet the needs of information users.  The
appendices to this report contain tables and  charts so that readers may explore particular items of
interest.  A description of the statistical procedures used to prepare this report is also provided.

Procedures

        There are now 925 plots in the FHM national network, of which 628 plots are forested. This
is about 16% (excluding Alaska) of the projected total number of forested plots that will be installed
nationwide over the next several years.   The installed plots are located  systematically, using  a
probability sampling design, throughout the forests of 12 states in the eastern United States.   During
the late summer of  1991,  these  plots  were visited by  trained field crews to make selected
measurements of forest health.  Over 45,000 trees and seedlings of more than 100 species in 10 major
forest types were measured by state and federal personnel.  Through a rigorous quality  assurance
program,  these data were  found  to meet  data quality objectives.   The  data and  supporting
documentation have been entered into the FHM databases.

       At each one-hectare forested plot location,  four fixed-radius  (7.32 meters) subplots were
established. Within these subplots, forest condition measurements were made of tree size, species,
frequency, damage, and crown condition.  Additional measurements, such as slope and aspect, were
made to characterize the forested plot locations. Forest condition measurements were analyzed by
using statistical procedures, and the results were expressed in tables, charts, and  cumulative
distribution functions. The cumulative distribution functions of the crown condition measurements were
further analyzed to estimate the proportions of the sampled populations with different health status.

        Data summaries and maps were also  prepared for three types of off-plot data: climate, forest
insects and diseases, and air pollution. For these off-plot data types, selected measurements that are
made by other monitoring programs were summarized in a way that will permit future analyses of the
possible determinants of forest  health as indicated by the on-plot measurements. It was not possible
to utilize the  FHM statistical estimators for the off-plot data types; summary procedures that were
appropriate for each particular off-plot database were used instead.

-------
Results and Discussion - On-Plot Data

       Although conclusions about regional forest health are tenuous when they are based on just a
few measurements taken in just one year, the following general findings emerged  from statistical
analyses of the 1991  on-plot data.

Tree Size and Stand Density

       Tree size and stand density measures are important for many analyses of forest health and
productivity.  This report summarizes stand density (expressed as basal area per unit area) and tree
frequency (number of stems per unit area) for selected subpopulations that were measured in 1991.
Results are presented in the form of cumulative distribution functions. These data will be the basis for
assessments of tree and stand growth after the plots are remeasured in later years.

Tree Crown Conditions

       Tree damage and crown condition was measured  on all trees > 12.7 cm dbh  on forested
subplots.  This report presents summary tables and cumulative distribution functions  of selected
measurements, aggregated by species, by forest type, and by crown group. Subpopulations of species,
forest types, and crown groups with a sufficient sample size were also delineated into categories of
optimal, nominal, subnominal,  and poor condition for  selected  measurements, based on expert
judgement or published "threshold" values for those categories.

       The measurement procedures are described in detail in the report. Briefly, the following crown
measurements were made.

       Crown transparency    A measure of the amount  of sunlight that passes through foliated
                             portions of a crown.
       Crown dieback        A measure of the mortality of relatively young branches that are in the
                             upper, sunlight-exposed portions of the crown.
       Crown position         A measure of the relative physical location of a tree crown in a stand
                             of trees.
       Crown density         A measure of the two-dimensional appearance of crown fullness and
                             symmetry.
       Crown ratio            A measure of the amount of an entire'tree bole that is foliated.

A Crown Defoliation Indicator (CDI) model was also used to describe overall crown condition as a
function of the crown  transparency and crown dieback measurements.

       The interpretation of tree crown condition  measurements depends not only on the ecological
significance of a particular measured value, but also on the perceived importances of relatively high or
relatively low values.  The  approach  taken  here  to setting the threshold values of condition  was
conservative. Where  possible, the setting  of threshold values considered expected differences among
species, size classes, crown positions, and typical habitat characteristics.

       A cursory evaluation of the cumulative distribution function analysis results indicated that most
of the tree species, all examined forest types, and most crown groups examined were in good condition.
It was recognized that since most of the population for species and crown groups were found to  be
nominal, that the thresholds  might be too  high. If this is true, then the percentages of the population
which were in a subnominal  condition should be of greater concern.
                                            VI

-------
        Research both off the Detection Monitoring plots and using data from the Detection Monitoring
plots is needed to set consistent and tailored threshold values for each species, forest type, and/or
crown group.
Tree Species Diversity

        This report quantifies the tree species diversity of the overstory community by using three well-
known diversity indices -- the number of species present, the natural exponent of Shannon's index, and
the reciprocal of Simpson's index.  Data for trees > 12.7 cm diameter at breast  height were used to
calculate the diversity indices.  Cumulative distribution functions of the indices were used to quantify
the 1991 regional status of overstory tree species diversity in the northeast and southeast United
States.

        The cumulative distribution functions  of the diversity indices were not significantly different
within either geographic region. Using the exponent of Shannon's index to illustrate the results for the
northeast region, cumulative probability values of 0.25, 0.50,  and 0.75 were obtained when  the index
was 2.2, 3.6, and 4.6, respectively.  The corresponding index values for the southeast region  were 1.6,
2.9, and 4.6, respectively.

        Although statistical  relationships do not imply cause  and  effect, some associations were
explored among forest characteristics and the values of the exponent  of Shannon's index. In the
southeast region, high overstory tree species diversity values were most closely associated with an oak
forest type, natural stand origin, saw limber size class, no recent disturbances, and more than one plot
condition code.  In contrast, low values were most closely associated with a pine forest type, planted
stand origin, seedling/sapling  size class, recent cutting disturbance, and one plot condition code.
Interpretations of these statistical associations were difficult  because of confounding  or correlations
among the  forest characteristics.

        In the northeast region, the possible associations of diversity values and forest characteristics
were  even  more difficult  to  investigate because there was  relatively little variation  in forest
characteristics. For example, the northern hardwoods forest type had  sample plots distributed over the
full range of diversity index values, but stand origins and size classes were the same for most of those
plots. Although few pine plots  were sampled in the northeast, their proportion decreased as diversity
values increased.  In contrast to the southeast region, recent disturbances were not associated with
diversity values in the northeast region.

        The cumulative distribution functions of the two geographic regions were most different at lower
diversity values.   The difference was not statistically significant,  and  was explainable by higher
proportions of  the pine forest  type  and planted stand origin in  the  southeast in comparison to the
northeast. The number of species was more sensitive than the exponent of Shannon's index to regional
differences.

Results and  Discussion -- Oil-Plot Data

        Three types of off-plot information were introduced into the FHM databases in 1991. Data from
the Forest Service (forest insects and diseases), the Environmental Protection Agency (ozone and wet
deposition), and the National Oceanic and Atmospheric Administration (precipitation, temperature, and
significant weather events) were summarized by FHM analysts. In  future reports, these and other
summaries will be used to explore possible regional associations between on-plot measurements and
particular stresses that are  of concern. The highlights of the auxiliary data summaries follow.


                                             vii

-------
 Major Forest Insects and Diseases (Eastern U.S.)

        A compilation of information from various State and Federal sources identified four problems
 of special interest.  A new disease (cause unknown) of blackgum has been found in the Appalachian
 mountains in three states. Thousands of cabbage palms have died from unknown causes along the
 Florida Gulf coast.  Populations of the black twig borer (an introduced ambrosia borer) are increasing
 and affecting a large number of tree species.  Dogwood anthracnose, first discovered in 1987, is
 expanding rapidly throughout the range of flowering dogwood.

        The report  summarizes the status of a large number of common forest insects and diseases.
 Detailed insect information is provided for the hemlock wooly adelgid, the hemlock loopers, the eastern
 spruce budworm, the southern pine beetle, and the gypsy moth.  Detailed information is also provided
 for fusiform rust, the littleleaf disease syndrome, the oak decline syndrome, and the beech  bark
 disease. Observations of other insects and diseases are organized according to forest types. Evidence
 of damage from weather events and ozone is mentioned.

 Climate

        The climate report summarizes selected climate conditions and events  which are known  from
 experience to impact forest health.  Data from a variety of sources  were summarized into databases
 and maps of precipitation, hurricane occurrence,  high wind events,  ice  storms, and late spring  hard
 freezes. The period from October, 1990 through September, 1991 is  summarized for the regions where
 FHM plots were measured in 1991.

        Highlights include severe tropical storms in the fall, a record  warm winter with low snowfall, an
 early spring, a severe spring ice storm, and a long and hot summer punctuated by Hurricane Bob.

 Air Pollution

        Maps of interpolated ozone data for low elevations in the eastern U.S. and maps and regional
 summary  statistics for selected ions in precipitation revealed the  general trends described below.
 Emissions data, dry deposition, and pollutants that are typically point source-oriented were not included
 in this report.

        A summary of ozone data for the years 1985-1989 was made using the seasonal (April through
 October) 'W126 index1 that places an emphasis on  peak ozone hourly average concentrations.  In
general, ozone concentrations vary among years, but over the five-year period there were two regions
that tended to give  higher values of the W126 index.  One region extends continuously along the  mid-
 Atlantic  seacoast and west to  the Appalachian Mountains.  The  second region  is smaller and is
 centered over the Ohio River valley near the Ohio-Indiana border.  Ozone data sources included the
 EPA Aerometric Information  Retrieval System, the National Dry Deposition Network, and the Mountain
 Cloud Chemistry Program.

        Data from the National Atmospheric Deposition Program were used to  prepare summaries of
 (precipitation-weighted) mean pH, sulfate deposition, nitrate deposition, and ammonium deposition for
 1990 (January through December).  Precipitation was most acidic, and sulfate and nitrate deposition
was highest, in a corridor that extended roughly from eastern Michigan and southern Indiana to  New
 York and southern  New England. Precipitation acidity and deposition of sulfate and nitrate generally
 decrease with distance away from this corridor.  This  general pattern is similar to that observed in
 earlier years. Nearly the entire eastern U.S. receives precipitation that is more acidic than 'normal
 rainfall'.  There was not a clear spatial pattern of ammonium deposition  over the study area.

                                            viii

-------
Conclusion

       This  Forest  Health  Monitoring Statistical Summary describes the current  approach and
activities, and summarizes the data that were collected in 1991. Looking to the future,  FHM plans the
following events and  improvements by 1994:

              Agreements will be reached with additional State and Federal agencies  to increase the
              scope of  FHM  participation, and to improve the efficiency of data collection and the
              depth of data analysis.

              The plot network will be expanded from 12 to 18-24 states, including states from the
              western and  north-central regions of the Nation.

              The core set of measurements made at each plot  location will be expanded so that
              more aspects of forest condition can be monitored.  New measurements must first
              pass a rigorous peer-review process, and candidates include measures of soil, wildlife
              habitat, and foliage chemistry.

              The collection of off-plot forest insect and disease information will be standardized.

              The set of off-plot databases will be augmented to include information from satellite
              sensors and  soil  surveys.

              The assessment capabilities will be developed further and made available to data
              analysts around the nation.  Annual statistical summaries will be prepared,  and the
              data will be made available for in-depth interpretive reports to address specific  forest
              health issues.
                                             IX

-------
                                 Table of Contents

Notice  	iii
Forward  	  iv
Executive Summary	v
List of Figures	 xii
List of Tables  	xv
Acronyms	xviii
Acknowledgements  	xx

       1.     Introduction 	  1-1
              The Challenge of Monitoring Forest Health	  1-3
              Current FHM Research	  1-7
              Looking to the Future	   1-12

       2.     Monitoring Methods 	  2-1
              Technical Objectives	  2-1
              Sampling Design and Measurements	  2-1
              Data Analysis	  2-8

       3.     Characterization of Stand Density and Number of Trees
              on FHM Plots	  3-1
              Introduction 	  3-1
              Objective	  3-1
              CDFs of Basal Area by Major Forest Group for the Northeast
              and Southeast	  3-1
              Method 	  3-1
              Figures	  3-2
              CDFs of Seedlings, Sapling, Trees and Stems for the Northeast
                      and Southeast	  3-6
              Method 	  3-6
              Seedlings  	  3-6
              Saplings 	  3-6
              Trees  	  3-6
              Stems (seedlings, saplings and trees combined)	  3-6

       4.     Crown Condition of Forest Trees on FHM Plots  	  4-1
              Introduction 	  4-1
              Crown Defoliation Indicator	  4-2
              Thresholds in CDFs 	  4-2
              Results 	  4-5
              Discussion	  4-9

       5.     Characterization of Regional Overstory Tree Species Diversity
              on FHM Plots	  5-1
              Background	  5-1
              Data Set Characteristics	  5-1
              Index Selection and Calculation	  5-1
              Analysis 	  5-4
              Results and Discussion	  5-4

-------
6.     Selected Climatic Data Summaries  	  6-1
       Introduction  	  6-1
       Synopsis	  6-1
       Precipitation	  6-1
       Hurricanes	  6-4
       High Wind Events	  6-4
       Ice Storms	  6-9
       Last Hard Spring Freeze (-2.2 C)	  6-9
       Data and Information Sources	  6-9

7.     Status of Major Forest Insects and Diseases in the Eastern
       United States, 1991  	  7-1
       Introduction  	  7-1
       New Problems of Special Interest  	  7-1
       Eastern White-Red-Jack Pine (includes Eastern Hemlock)	  7-1
       Eastern Spruce-Fir	  7-2
       Southern Pines  	  7-4
       Oak-Hickory	  7-5
       Northern Hardwoods	  7-16
       Aspen-Birch	  7-18
       Special Topics	  7-18
       Acknowledgements	  7-23

8.     Selected Air Quality and Deposition Data Summaries  	  8-1
       Ozone	  8-1
       Wet Deposition  	  8-8

9.     References  	  9-1

       Appendix A.    Statistical Design	  A-1

       Appendix B.    Quality Assurance Program for Forest Health
                      Monitoring for 1991  	  B-1

       Appendix C.    Nomenclature of Species in Text and Forest Type
                      and Species Codes	C-1

       Appendix D.    Conversion Factors for Common Measurement Units	D-1

       Appendix E.    Supplemental Tabular Summaries	  E-1
                                    XI

-------
                                    List of Figures




Figure 1-1.     Distribution of forests in the United States  	  1-4

Figure 1-2.     The 1990 and 1991 plot design and logistics studies	  1-8

Figure 1-3.     The 1992 regional projects  	  1-10

Figure 1-4.     Landscape pattern types for northern Georgia	  1-11

Figure 2-1.     The national grid of monitoring locations	  2-2

Figure 2-2.     The locations of sampling points that were classified (all or in part)
               as forested  	  2-4

Figure 2-3.     Typical layout of an FHM  plot 	  2-5

Figure 2-4.     Interpreting a CDF	  2-11

Figure 3-1.     Cumulative distribution function of (a) basal area for the pine group
               in the NE, and (b) basal area for the spruce-fir group in the NE	  3-3

Figure 3-2.     Cumulative distribution function of (a) basal area for the hardwoods
               and miscellaneous group in the NE, and (b) basal area for the pine
               group in the SE  	  3-4

Figure 3-3.     Cumulative distribution function of (a) basal area for the oak-pine
               group in the SE, and (b) basal area for the hardwoods and miscellaneous
               group in the SE  	  3-5

Figure 3-4.     Cumulative distribution function of (a) number of seedlings/hectare in
               the NE, and (b) number of seedlings/hectare in the SE 	  3-7

Figure 3-5.     Cumulative distribution function of (a) number of saplings/hectare in
               the NE, and number of saplings/hectare in the SE	  3-8

Figure 3-6.     Cumulative distribution function of (a) number of trees/hectare in
               the NE, and (b) number of trees/hectare in the SE 	  3-9

Figure 3-7.     Cumulative distribution function of (a) total number of stems/hectare
               in the NE, and (b) total number of stems/hectare in the SE  	  3-10

Figure 4-1.     The three established thresholds for the Crown Defoliation Indicator	  4-4

Figure 4-2.     Standard Federal Regions used for the analyses	  4-10
                                             xii

-------
Figure 5-1.     Forest Health Monitoring ecological assessment model for biotic diversity  ....  5-2

Figure 5-2.     Cumulative distribution function of (a) regional overstory tree
               diversity in the NE using species richness, and (b) regional
               overstory tree diversity in the SE using species richness	  5-5

Figure 5-3.     Cumulative distribution function of (a) regional overstory tree
               diversity in the NE using Simpson's Index, and (b) regional
               overstory tree diversity in the SE using Simpson's Index  	  5-6

Figure 5-4.     Cumulative distribution function of (a) regional overstory tree
               diversity in the NE using the Shannon-Weiner method, and (b) regional
               overstory tree diversity in the SE using the Shannon-Weiner method  	  5-7

Figure 6-1.     Deviation of annual climate division precipitation (October 1990 -
               September 1991) from 1961-1990 average conditions	  6-2

Figure 6-2.     Probability of a climate division receiving less precipitation than the
               October 1990 - September 1991 total	  6-3

Figure 6-3.     County frequency of hurricane contact, October 1984 - September 1991	  6-5

Figure 6-4.     County hurricane contact, October 1990 - September 1991  	  6-6

Figure 6-5.     County frequency of high wind events containing wind speeds in
               excess of 70 km/h, October 1984 - September 1991  	  6-7

Figure 6-6.     County frequency of high wind events containing wind speeds in
               excess of 70 km/h, October 1990 - September 1991	  6-8

Figure 6-7.     County frequency of ice storm events, October 1984 - September 1991  ....  6-10

Figure 6-8.     County frequency of ice storm events, October 1990 - September 1991  ....  6-11

Figure 6-9.     Deviation (days)  of 1991  date of last hard spring freeze (24-hour
               minimum temperature equal to or less than -2.2° C) from 30-year
               median conditions  	  6-12

Figure 7-1.     Hemlock wooly adelgid distribution in Virginia (Forest Pest
               Management, R-8)	  7-3

Figure 7-2.     Fusiform rust hazard for loblolly pine in 1991  (Forest Pest
               Management, R-8)	  7-7

Figure 7-3.     Fusiform rust hazard for slash pine in 1991 (Forest Pest
               Management, R-8)	  7-8

Figure 7-4.     Counties with outbreaks of southern pine beetle in 1991  (Forest
               Pest Management, R-8)  	  7-9

Figure 7-5.     Historic range of  littleleaf disease (Forest Pest Management, R-8)  	  7-10
                                            XIII

-------
Figure 7-6.     Gypsy moth defoliation, 1984-1991 (Forest Pest Management,
               R-8 and MA)  	  7-13

Figure 7-7.     Bottomland oak FIA plots affected by oak decline in the southern
               region (Forest Pest Management, R-8)	  7-14

Figure 7-8.     Upland oak FIA plots affected by oak decline in the southern region
               (Forest Pest Management, R-8)	  7-15

Figure 7-9.     Dogwood anthracnose  occurrence in the Eastern United States
               (Forest Pest Management, R-8 and NA)	  7-19

Figure 7-10.    Black gum disease occurrence (Forest Pest Management, R-8)	  7-21

Figure 8-1.     Map of kriged seasonal ozone W126 index at 10 ppm-h intervals for 1985  ... 8-2

Figure 8-2.     Map of kriged seasonal ozone W126 index at 10 ppm-h intervals for 1986  ... 8-3

Figure 8-3.     Map of kriged seasonal ozone W126 index at 10 ppm-h intervals for 1987  ... 8-4

Figure 8-4.     Map of kriged seasonal ozone W126 index at 10 ppm-h intervals for 1988  . . . 8-5

Figure 8-5.     Map of kriged seasonal ozone W126 index at 10 ppm-h intervals for 1989  . . . 8-6

Figure 8-6.     An example of a 'gradient' analysis showing one way that kriged ozone
               information across years can be aggregated	 8-7

Figure 8-7.     The 1990 annual levels of precipitation-weighted mean pH for the
               eastern United States	 8-9

Figure 8-8.     The 1990 annual levels of precipitation-weighted mean sulfate
               deposition for the eastern United States	  8-10

Figure 8-9.     The 1990 annual levels of precipitation-weighted mean nitrate
               deposition for the eastern United States	  8-11

Figure 8-10.    The 1990 annual levels of precipitation-weighted mean ammonium
               deposition for the eastern United States	  8-12
                                            XIV

-------
                                     List of Tables

Table 2-1.      Schedule of descriptive measurements and associated information
               collected for FHM plots in 1991  	  2-7

Table 2-2.      Schedule of mensuration measurements for FHM plots in 1991  	  2-7

Table 2-3.      Overstory crown condition and tree damage measurements made
               on large trees within subplots in 1991	  2-8

Table 3-1.      Basal areas associated with cumulative distribution probabilities
               of 0.25, 0.05, and 0.75 for Figures 3-1  to 3-3  	  3-2

Table 4-1.      CDF concern, optimal, and poor thresholds for the crown
               variables dieback, and transparency	  4-4

Table 4-2.      Population proportions for species in the East U.S. Region from
               CDF analysis   	  4-5

Table 4-3.      Population proportions for crown groups in the East U.S. Region
               from CDFanalysis	  4-6

Table 4-4.      Population proportions for species in Standard Federal Regions I and II
               from CDF analysis	  4-6

Table 4-5.      Population proportions for crown groups in Standard Federal Regions I and II
               from CDF analysis	  4-7

Table 4-6      Population proportions for species in Standard Federal Region III
               from CDF analysis	  4-7

Table 4-7.      Population proportions for crown groups in Standard Federal Region III
               from CDF analysis	  4-8

Table 4-8      Population proportions for species in Standard Federal Region IV
               from CDF analysis	  4-8

Table 4-9      Population proportions for crown groups in Standard Federal Region IV
               from CDF analysis	  4-9

Table 5-1.      Definition and calculation of plot-level values for overstory
               tree species richness and diversity	  5-3

Table 6-1.      Particularly  severe icing events in recent history	  6-9

Table 7-1.      Defoliation by eastern spruce budworm by state for 1990
               and 1991	  7-4

Table 7-2.      Area and percent of host type infected with fusiform rust  by
               state, 1991   	  7-6
                                             xv

-------
Table 7-3.

Table 7-4.


Table 7-5.

Table 7-6.

Table 7-7.

Table 8-1.

Table B-1.

Table B-2.


Table B-3.

Table E-1.

Table E-2.

Table E-3.


Table E-4.

Table E-5.


Table E-6.


Table E-7.


Table E-8.

Table E-9.

Table E-10.


Table E-11.


Table E-12.
Southern Pine Beetle Outbreak Area 1990-1991 	  7-6
Area and number of countries within the historic range of
littleleaf disease	
7-11
Area of defoliation by gypsy moth by state, 1990 and 1991  	  7-11

Incidence and impact of oak decline in the Southeast, 1991	  7-12

Area affected by Dogwood Anthracnose in the Southeast 1988-91  	  7-20

Median values of annual summary statistics,  1979-1990  	  8-13

FHM measurements subject to 1991  QA requirements   	  B-2

Format for quality assurance for implementation measurement
documentation for QAPjP  	  B-3

Personnel responsibilities in 1991  FHM monitoring QA activities	  B-4

Number of subplots by state and land use 	  E-2

Number of subplots by state and major forest type  	  E-3

Number of forested subplots by major forest type and stand
size class  	  E-4

Number of forested subplots by major forest type and stand origin	  E-5

Number of forested subplots by major forest type and utilization
disturbance  	  E-6
Number of forested subplots by major forest type and care
disturbance  	
 E-7
Number of forested subplots by major forest type and natural
disturbance  	  E-8

Number of subplots by major forest type and terrain position  	  E-9

Number of subplots by major forest type and elevation class  	  E-10

Number of live trees greater than 12.7 cm DBH by species and
crown position on subplots  	  E-11

Number of trees tallied by species and size class (seedlings,
saplings, and large trees > 12.7 cm DBH) 	  E-14

Number of live trees > 12.7 cm DBH  tallied on subplots, by
species and crown ratio class 	  E-18
                                            XVI

-------
Table E-13.     Number of live trees > 12.7 cm DBH tallied on subplots, by
               species and crown density class  	  E-21

Table E-14.     Number of live trees > 12.7 cm DBH tallied on subplots, by
               species and crown dieback class	  E-24

Table E-15.     Number of live trees > 12.7 cm DBH tallied on subplots, by
               species and crown transparency class	  E-27

Table E-16.     Number of live trees > 2.5 cm DBH tallied on microplots, by
               species and crown ratio class 	  E-30

Table E-17.     Frequency of visible tree damage types on all live trees greater
               than 12.7 cm DBH, by species  	  E-32
Table E-18.    Numbers of seedlings and saplings tallied by species and vigor
              class on microplots	
E-41
                                           XVII

-------
                                    Acronyms

AIRS         -      Aerometric Information Retrieval System (U.S. EPA)
CDF          -      cumulative distribution function
DBH          -      diameter at breast height (1.37 meters)
DQO         -      data quality objective
EMAP        -      Environmental Monitoring and Assessment Program
EPA          -      U.S. Environmental Protection Agency
EQO         -      ecosystem-level data quality objective
FIA           -      Forest Inventory and Analysis
FHM          -      Forest Health Monitoring
CIS           -      Geographic Information System
ha            -      hectare
IQO           -      indicator-level data quality objective
LPT           -      landscape pattern type
m            -      meter
MGCP        -      Mountain Cloud Chemistry Program
MQO         -      measurement quality objective
NADP         -      National Atmospheric Deposition Program
NASF         -      National Association of State Foresters
NE           -      northeast region of the United States
NDDN        -      National Dry Deposition Network
NE           -      Northeast region
PDR          -      Portable Data Recorder
QA           -      Quality Assurance
QAPjP        -      Quality Assurance Project Plan

                                         xviii

-------
QC




RQO




SAP




SAMAB




SE




SOP




USDA




VCR
Quality Control



resource-level data quality objectives



Society of American Foresters



Southern Appalachian Man and Biosphere program



Southeast region of the United States



standard operating procedure



United States Department of Agriculture



visual crown rating
                                          XIX

-------
                               Acknowledgements
Forest Health Monitoring is an interagency program including the Forest Service, the Environmental
Protection Agency, the Tennessee Valley Authority, the U.S. Soil Conservation Service, the National
Park Service, the Bureau of Land Management, the Fish and Wildlife Service, the National Association
of State Foresters, and individual State forestry authorities.

This document was conceived and prepared by the members of the Forest Health Monitoring national
assessment group. The  following lists the current membership of the group.
Bob Anderson
Chuck Barnett
Bill BechtoW
Bill Burkman
Jerry Byers
David Cassell
Steve Cline
Barbara Conkling
Ellen Cooler
Terry Droessler
Beth Eastman
Karl Hermann
Eric Hyatt
Chuck Liff
Bob Loomis
Craig Palmer
Kurt Riitters
Victoria Rogers
Doug Shadwick
Luther Smith
Tom Stockton
Ken Stolte
Larry Truppi
Jim Wickham
USDA Forest Service, Asheville, NC
USDA Forest Service, Radnor, PA
USDA Forest Service, Asheville, NC
USDA Forest Service, Radnor, PA
Lockheed Engineering & Sciences Co., Las Vegas, NV
ManTech Environmental Technology, Inc.,  Corvallis, OR
ManTech Environmental Technology, Inc.,  Corvallis, OR
North Carolina State University, Raleigh, NC
U. S. Environmental Protection Agency, Research Triangle Park, NC
ManTech Environmental Technology, Inc.,  Corvallis, OR
North Carolina State University, Raleigh NC
ManTech Environmental Technology, Inc.,  Research Triangle Park, NC
US Environmental Protection Agency, Research Triangle Park, NC
University of Nevada-Las Vegas, Las Vegas, NV
USDA Forest Service, Research Triangle Park, NC
U. S. Environmental Protection Agency, Las Vegas, NV
ManTech Environmental Technology, Inc.,  Research Triangle Park, NC
Lockheed Engineering & Sciences Co., Las Vegas, NV
ManTech Environmental Technology, Inc.,  Research Triangle Park, NC
ManTech Environmental Technology, Inc.,  Research Triangle Park, NC
Duke University, Durham, NC
USDA Forest Service, Research Triangle Park, NC
U. S. Environmental Protection Agency, Research Triangle Park, NC
Bionetics Corp., Warrenton, VA
The following members of the Forest Health Monitoring Information Management group are recognized:
David Dickson, Andy Gillespie, Doug Griffith, Larry Royer, Mark Rubey, and Art Walmsley.

The authors thank the external reviewers of this document, Dr. Daniel Binkley and Dr. George Furnival,
for their time in reading the document and their constructive comments.

The authors gratefully acknowledge the additional review comments from Roger Blair, David Cassell,
Gardner Evans, Tony Olson, Don Stevens, and John Vissage.

Appreciation goes to Kurt Riitters and Barbara Conkling for their contributions as editors.

Appreciation also goes to the following persons: Elizabeth Eastman for her assistance with editing, Karl
Hermann and Tom Stockton for their graphics assistance, and Cheryl Simmons for her word processing
assistance.
                                           xx

-------
    1.  Introduction

           In 1990, the U.S. Forest Service and
    the U.S  Environmental  Protection Agency
    initiated a  cooperative national program of
    Forest  Health Monitoring,  known  as  FHM.
    Participants include the National Association of
    State   Foresters,   individual  State forestry
    agencies, the Tennessee Valley Authority, the
    Soil Conservation Service, the Bureau of Land
    Management, the Fish and  Wildlife Service,
    and the National Park Service. Although FHM
    is not established in all parts of the nation, it
    has  already  provided  valuable  information
    about forest health in some areas.  Today, as
    more  Federal and  State  organizations are
    joining the program, FHM is becoming a focal
    point for many efforts to  understand  forest
    health and to assess the impacts of natural and
    man-made  stresses.   In  response to the
    diverse  and growing public  concerns  about
    human  impacts  on  our  environment,  the
    objective of FHM is to give a  many-sided view
    of  forest health that  will assist the public in
setting priorities and making informed choices
aimed at reducing the environmental risks of
human activities.

       The Forest Health Monitoring program
is designed to provide information  to  help
protect, manage,  and  use forest  resources
wisely. But monitoring alone can only provide
information about  the status and  trends  of
forest  health,  and that  is  why  FHM  also
supports   more  intensive  research  that  is
intended to discover specific cause  and effect
relationships.  When these relationships are
well-understood, then specific actions may be
taken to  mitigate  or  prevent  impacts,  and
monitoring can again be  used to measure the
effectiveness of those actions.  Some stresses
which operate at  regional  scales can cause
subtle and slow, yet widespread, changes in
forest health that can only be detected by using
data  from  a  long-term   and  large-scale
monitoring program such as FHM.   For the
sake  of   efficiency,  FHM  is   designed  to
complement and utilize the results of ongoing
                     ; iPA% Environmental Monitoring ant* Assessment Program
       The Environmental Monitoring and Assessment Program (EMAP) is an interagency program to
monitor the condition of the nation's ecological resources. EMAP is the foundation of EPA's comprehensive
ecological risk assessment program and it utilizes forest condition information from the Forest Health
Monitoring  program.  Other interagency efforts will supply monitoring data for lakes, streams, wetlands,
agricultural lands, deserts, and  estuaries.  Together,  these data bases will permit scientists to make
comprehensive analyses of ecological resource conditions at regional scales.

       In a sense, EMAP is a 'consortium without walls' where members of the consortium share common
objectives for ecological monitoring. The varied efforts of the agencies in the consortium, made necessary
because of the variety of ecological resources in the nation, are tied together by EMAP's common statistical
sampling design and  approach to regional assessments.  Because  EMAP represents  a substantial
commitment of resources, it has become a focal point for both scientific discussions and agency planning
as many groups work to deploy an integrated ecological monitoring network. EMAP is fostering interagency
cooperation and the development of tools and procedures for regional ecological risk assessments. EMAP
also permits the EPA to target its own resources where its scientists can make the highest contribution to
the overall  effort.

For more information, contact:
       Director,  Environmental Monitoring and Assessment Program
       U.S. Environmental Protection Agency, RD-680
       401 M Street, SW
       Washington, DC 20460
                                             1-1

-------
                                           ***
State Forestry Agencies and the National Association of State Foresters (NASF) -- through its Forest Health
Committee -- are key partners for planning and implementing Forest Health Monitoring. Contact the NASF
at (202) 624-5415 for more information on state forestry activities.

The USDA Forest Service has several programs which relate to FHM. Some of these currently participate
in FHM and others may contribute or use FHM information in the future.  Information about the following
programs can be obtained from the Forest Service National Headquarters Office at (202) 205-1760.

       Forest Inventory, Economics, and Recreation Research - Forest Inventory and Analysis Program
       Forest Pest Management
       Watershed and Air Management - Air, Water, and Soils Programs
       Timber Management ~ Timber Inventory Program
       Resources Program and  Assessment
       Land Management Planning
                                           ***
environmental monitoring and research that is
carried  out by other  Federal and  State
organizations.  In these ways, FHM  makes a
unique   and   valuable   contribution  to
comprehensive  assessments  of the  health
status and trends of the nation's forests.

       Forest   Health  Monitoring   is  being
improved during its  initial stages based on
suggestions received from information  users
and  environmental scientists.   The  scientific
community  has  been involved in the planning
and initial deployment of the program. It is now
time to  present  FHM  in  some  detail to
the public.  The main purposes of this report
are:  1) to describe the current approach and
activities; 2) to summarize the limited amount
of data that has been collected to date; and 3)
to outline some new directions that are being
explored.  This report is itself part of a test of
an assessment process by which data from
many sources will be brought together for
analyses.   As sufficient  data  are  available,
FHM will produce a series of annual 'statistical'
reports  and  periodic  'interpretive'  reports.
Improvements to the assessment process will
be based on reactions to the tables and charts
that appear here as prototypes.
                                             1-2

-------
           This report is  organized as  follows.
    This introductory  section  provides  a brief
    overview of forest health monitoring. The next
    section describes briefly the methods used by
    FHM to collect data on a network of plots. This
    is followed by a brief characterization of the
    plots in the network at the end of 1991.  The
    final sections provide statistical summaries and
    some interpretations of the  data that were
    collected  in  1991, and summaries  of the
    auxiliary data bases that were made available
    in  1991.   Detailed tables and charts,  and
    additional   methodological  information,  are
    appended to this report.

    The Challenge of Monitoring Forest
    Health

           Forests are an  important part of the
    American  economy,  culture,  and  ecology.
    Overall,  about one-third  (296  million hectares)
    of the United States is forested (Figure 1-1).
    Of this total, about 200 million  hectares help to
support forest products-related industries that
employ over 18 million people and contribute
over 5% of the gross national product.  Forests
everywhere provide important wildlife  habitat,
watershed    protection,   and    recreational
opportunities.  These amenities are difficult to
evaluate in economic terms, but their value is
at  least  substantial  and  in  some  cases
priceless.   Forests occur  in  a  variety  of
environments  ranging  from near-deserts  to
wetlands, and they are a part of many national
parks, wilderness areas, and community parks
and  greenways.   Some  forests are virtually
untouched,  while  others  are  completely
artificial. Forest lands are owned by the public,
by  businesses,  and  by   private  individuals.
Nearly  everyone  has had some  personal
experience with forests that helps to shape his
or her expectations and therefore  perceptions
of  what  constitutes  'good  forest   health'.
American forests  are an integral part  of a
global ecosystem that permits and sustains the
richness and productivity of life on our planet.
                                 What is'normal'
       Your doctor can monitor your general health status with a few measurements such as temperature
and blood pressure. These simple tests are inexpensive and non-specific and therefore, they efficiently
screen a broad range of possible health problems. The test results are taken to be indicators of health
because 'normal' values are well-known and interpretations are not subject to much debate.  Similarly, it
is believed that a few key measurements will efficiently and effectively detect changes in forest condition.
But unlike human health standards, forest health standards are not well-defined.  The measurements can
be interpreted in different ways because 'normal' forest health means different things to different people.
Different indicators can yield different results if only because they measure different components of forest
health.  Furthermore, if populations of forests are normally more diverse than populations of people, it is
possible  that different standards will apply to different situations.  Finally, foresters consider broad
definitions of wellness that go beyond the normality of physiological processes.

       The FHM program  relies on the research results of forest scientists everywhere to  help define
biological expectations of normality. But lacking a complete  biological understanding,  FHM can still use
the accumulated monitoring data to identify statistical patterns that indicate relative forest health.  The
assessment process will permit people to apply different value judgements to the measurements, further
defining what is meant by a normal forest condition. Meanwhile, interpretations are still necessary because
most people want to know whether or not the forests of the nation are healthy.  FHM uses the experiences
of scientists who have studied forest health for many years to guide data interpretation.  All interpretations
will be made stronger as the operational definitions  of  normality become more rigorous and generally
accepted.
                                            ***
                                               1-3

-------
                              Alllkt
                                                                  Hoilii
Figure 1-1.  Distribution of forests in the United States (digitized by S. Azevedo).
                                               1-4

-------
        How can  FHM best help to resolve
current  and potential threats to forest health?
The public discussion of environmental threats
tends  to  be  issue-oriented:  global climate
changes, acid rain, deforestation, and non-point
source  pollution are current examples.  The
FHM program can contribute to this discussion
in at least two ways.  First, FHM can provide
objective  statistical information  that can  be
used  to evaluate  real and  perceived threats
from many viewpoints, including the regulatory
or management viewpoints of the agencies that
sponsor FHM,   Second, FHM can contribute
thoughtful and understandable, issue-oriented
interpretations that show a scientific basis  for
why forest health is as it appears to be.

        Forest health monitoring is not a simple
task.   Forests  are continually exposed  to a
changing  array  of  natural and  man-made
stresses.   For this reason, it  is  not usually
possible to point to a single factor that is the
sole determinant  of  forest health.   Forest
function may be viewed as a  complex set of
interactions among physical factors such  as
weather or disturbance, chemical factors such
as  nutrient or pollution levels,  and  biological
factors such as growth or competition.  Often,
the specific response of a forest to some stress
will  depend   very   much  on   the   local
circumstances.  Against this backdrop of poorly
understood and interacting factors, the regional
status and trends of forest health can be subtle
and difficult to identify.

        The  multi-tiered FHM approach  to
regional   monitoring   directs the  immediate
objective  to identifying those forest regions
where conditions are poor or are changing,  for
better or  worse.  On  FHM plots,  a  set  of
'leading indicators' is used to classify the health
status of forests.  Because  the indicators are
not  diagnostic,  the   specific  reasons   for
changing conditions may not be obvious at first.
Instead, the  indicators  collectively represent
many dimensions of forest health and they are
generally responsive to many types of stresses.
The indicators are measured at locations that
are  selected  by statistical  procedures  to
systematically represent populations of forests
on  a  regional basis.  The off-plot data bases
may also be used to quantify status and trends
in some cases.

        When  the  measurements   indicate
changes in forest health, follow-up  analyses
are  made in an  attempt to  associate  the
changes with particular classes  or  types of
stresses.  Forest Health Monitoring plot data in
combination with auxiliary data are  used for
this task.  The resulting picture of forest health
is sharper and guides the next steps of the
assessment   process.     Sometimes,   the
exploratory  analysis may conclude  that  the
observed changes are 'normal',  or perhaps not
important   enough   to  warrant   further
investigation.   At  other times,  exploratory
analysis may suggest research needed to fully
understand the observed changes. In the rare
cases  when  FHM measurements  include
diagnostic   tests  of  specific  cause-effect
relationships,  the  exploratory  analysis  can
determine the effectiveness of health mitigation
or stress reduction actions.

        The multi-tiered approach to monitoring
preserves options  by not focusing on specific
cause-effect relationships at the beginning.  For
example, if FHM set out to monitor the effects
of a particular stress such as  global climate
changes,  then the  effects  of  other equally
important  stresses  could  be  missed entirely.
Furthermore, an assessment of the possible
effects  of global climate changes on forests
would not be able to separate the effects of
any other interacting stresses.  In other words,
forest health monitoring must  not be issue-
oriented   even    though   forest   health
assessments may  be.  In another example, if
FHM focused  measurements  on only  one
attribute of the forest, for example biodiversity,
then an effect on another forest attribute, say
productivity,   could   be   missed,   and
assessments would then be unable to address
the full  range of public concerns.  In  practice,
the basic  FHM design goal is  to focus later
research  by  distinguishing  among  effects
caused  by broad classes of stresses such as
weather, land management, and air pollution.

        Another challenge for FHM comes from
the realization that regional forest health is not
just the sum of the individual forest  tracts or
                                               1-5

-------
species.  The health of a regional  population
depends in part on the  spatial  context and
functional arrangement of  forests.   Hunters
know that  some grouse  species use aspen
forests of different  age classes for food  or
shelter;  the health  of  the grouse  population
depends on a 'healthy'  spatial arrangement of
a population of aspen forests.   In another
example, a fire that occurs in a remote forest
can have different  social  consequences  than
one that occurs in a municipal watershed. The
historical context is also important. Consider,
for example,  that   this century has seen the
virtual  disappearance of the once-dominant
American chestnut tree from the eastern forest
(caused by an  imported  fungus), and yet to
many  people  the  eastern  forest  appears
'healthy* today.   This does not  imply that
species extinctions are a healthy occurrence,
but rather that interpretations of current health
are always contextual.

       Finally,  there  is  the challenge  of
actually doing forest health  monitoring on  a
national scale.  There will  never be enough
funds  to  deploy   all  possible   monitoring
       Historically, the EPA's science programs have focused primarily on single pollutants, and on their
chemical compositions and impacts on a single medium (e.g., forests).  Assessments have been made
primarily  in terms of concentrations of that pollutant, and its ultimate transport and fate.  Although this
single-medium, stressor-based approach allows us to set regulatory standards for specific pollutants, it may
not be the most effective approach for problems like global warming, stratospheric ozone depletion, habitat
destruction, loss of biodiversity, tropical deforestation, and extinction of amphibian and neotropical migrant
bird species.

       One alternative recommended by the EPA Science Advisory Board is an environmental health-
oriented,  assessment-driven, ecological monitoring and assessment program which can tell us with known
confidence the actual condition of our environmental resources.  Scientifically rigorous information in a
multi-media context would allow decision-makers to focus resources where they could make the greatest
difference.   Though  somewhat  simplified, this  approach to achieving the "greatest  bang for the
environmental buck" is generally known as risk-based decision making.

       Risk-based decision-making uses the relatively new science of  ecological risk assessment. The
beginnings of  the EPA ecological  risk assessment program exist  in the Environmental Monitoring and
Assessment Program  (EMAP), and in the Forest Health Monitoring program which supplies the forestry
component of EMAP.  The  goal of EMAP is to use long-term environmental  monitoring  data  for
consolidated assessments that will 'make a difference'.  The Environmental Monitoring and Assessment
Program  is intended to allow EPA to manage for environmental results as called for by EPA's Administrator
William Reilly.  In EPA, EMAP is one of  several innovative techniques such as risk-based  strategic
planning, regional comparative risk, and ecological risk assessment programs that will be used, to assist
Agency decision-makers in solving environmental problems.
For more information, contact:
       Project Officer for EMAP-lntegration and Assessment
       U.S. Environmental Protection Agency
       Atmospheric Research and Exposure Assessment Laboratory MD-75
       Research Triangle Park, NC 27711
                                           ***
                                              1-6

-------
procedures, or to measure every acre of forest
land, but a statistical framework and a higher
measurement efficiency mean that fewer trade-
offs  have  to  be  made.   The  operational
problems of logistics, quality assurance, and
data management are not trivial for a national
network  of plots.   Effective  forest health
monitoring requires a substantial investment in
a  plot  network and in  a  cadre  of forest
monitors.  The approach taken by FHM is to
utilize existing resources whenever possible,
and to test procedures in medium-scale pilot
tests before deployment.  Much attention has
also been  given to  optimizing  the  statistical
designs.     Uniform  training  and  quality
assurance programs ensure  that the data are
comparable and of  sufficient quality  for the
intended use.  The data are managed almost
entirely  by  electronic means, from hand-held
computers  used  for data  collection to  an
interagency network of large  computers  for
data  analysis  and  assessments.    Some
features of  the implementation of FHM will be
highlighted  in later sections of this report.

       Despite the  difficulties, forest health
monitoring  is  an activity that  needs to  be
started now. The full value of such monitoring
will be realized when the network has  been
operating  for  several  decades,  but useful
information can be obtained  in the short term
also.   Imagine how data from decades  of
monitoring could be used today, a time when
regional  forest   health   assessments   are
sometimes  rendered anecdotal by a  lack  of
statistical baseline information.   The nation
recently spent  more than one hundred million
dollars to assess acid rain impacts in forests ~
without  the benefit  of  baseline  data  that
monitoring  could  provide.    The   National
Academy of Sciences, the President's Council
on  Environmental  Quality,   and  the  EPA
Administrator's Science Advisory Board have all
concluded that  knowledge of regional status
and trends will greatly  improve this kind  of
assessment. The legacy of our efforts today is
a generation of Americans better-equipped to
make  informed  decisions  that  affect  the
environment.
Current FHM Research

        Scientific monitoring is not a matter of
just looking at a few trees here and there.  The
FHM program has been started with a sound
scientific basis, and a wide range of research
and  development  projects are underway to
improve monitoring.   The current  monitoring
procedures emphasize trees because they are
an important element of forests and because
most  monitoring  experience  is with  trees.
When  peer-review  panels  agree,  and when
logistically   feasible,  new  monitoring  and
assessment procedures can be deployed.  The
panels represent many scientific disciplines and
have  been  selected by, for example,  the
National Academy  of Sciences,  the  EPA's
Science Advisory  Board,  and the American
Statistical Association.  Every major plan  and
report produced by FHM is peer-reviewed by
scientists  from  the  relevant   disciplines.
Change is  inevitable because  science  will
progress, and so  FHM has been designed to
accommodate new information.

        One  exciting research opportunity is
provided by the national FHM network of  plot
locations. Whereas most ecological research
is confined  to a  few  plots, or  to  single
watersheds, the FHM plot network is being
used to test regional-scale hypotheses  that
could not be addressed in most other studies.
This type of research complements work that is
conducted by others at experimental forests
and watersheds around  the nation.   Forest
Health Monitoring supports research to improve
the set of monitoring measurements, and to
increase the sampling and logistical efficiency
of monitoring operations.

        The Plot Design and Logistics Study
done in 1990 was a test  of the  basic  plot
design  and monitoring procedures  in eastern
forest  types  (Figure  1-2).   The  primary
objectives  were  to  optimize the  sampling
designs and logistical procedures that are now
used.   Also included were  tests  of a  few
measurements  that are not  yet routine, for
example,  soil  chemistry   and    vegetation
                                          1-7

-------
                                                             IO
                                                             CO

                                                             'o


                                                             "5
                                                             re
                                                             CD
                                                             t
                                                             cn
                                                             01


                                                             i
                                                             co

                                                             o
                                                             01
                                                             cn
                                                             T—

                                                             

                                                             U.
1-8

-------
                     Landscape patterns: Research to improve monitoring.
        When the earth's surface is viewed from an airplane, a 'landscape' can be represented as a
heterogeneous composition of ecosystems and land uses. Ecologists are learning to classify these patterns
in many ways. EMAP-Landscape Characterization and FHM are testing the use of landscape pattern types
(LPTs) as a means to analyze forest data measured on ground plots in a landscape context.

        The LPTs are mapped by visual interpretation of satellite images according to the land cover types
(e.g., forest, agriculture)present and the pattern that these cover types form. The land cover categories
used to map LPTs include forest, clearcut, agriculture, wetland, water, residential, and urban. The pattern
categories are based on the dominance or evenness of the land cover categories.  Like a ground-based
measurement, the dynamics of LPTs can indicate large-scale changes in forest condition. The LPTs can
also be used to organize summaries of the FHM ground plot  data.

        Figure 1-4 shows LPTs for northern Georgia.  Landscape pattern type mapping and analysis is part
of current research designed to bring landscape analysis into FHM assessments.

For more information, contact:
        Project Officer for EMAP-Landscape Characterization
        U.S. Environmental Protection Agency
        Environmental Photographic Interpretation Center
        Building 166, Bicher Road, Vint Hill Farms Station
        Warrenton, VA22186
                                           ***
structure.  The Georgia Demonstration and the
Western Pilot projects done  in  1991  were
similar tests that extended results to different
ecological regions and more types of forests
(Figure 1-2).

        Apart from optimizing data collection
after   a  measurement  is  decided   upon,
improving efficiency also means optimizing the
set of measurements that are made. A suite of
measurements  is needed that characterizes
forest health, broadly defined, and that signals
changes in forest health from the wide variety
of possible causes. Ideally, the suite is chosen
such  that  analyses  also give  clues  about
plausible causes of change, and hence suggest
the next  logical  steps  in the  assessment
process.  Starting from several conceptual
models of forest  health, FHM scientists are
conducting regional demonstration projects to
test the efficiency and efficacy of alternate sets
of measurements.
       The Southeast Loblolly/Shortleaf Pine
Demonstration project (1992  - 1993)  is one
such test of a set of measurements,  in this
case for a  single forest type over a large
geographic region (Figure 1-3). The Southern
Appalachian Man  and  Biosphere (SAMAB)
program in cooperation with FHM is conducting
the SAMAB Demonstration  project (1992  -
1993) to make the same kind of test in multiple
forest types in a somewhat smaller geographic
region (Figure 1-3). In California and Colorado,
the Western Pilot project (1992) is also the first
step towards  implementing  a basic  set  of
measurements in western forests  (Figure 1-3).

       Where appropriate, FHM also sponsors
research  projects that are not based solely on
the regional plot  network.  The  interagency
approach and  FHM peer-review  process are
used, and teams of researchers are formed to
collaborate on parallel studies in their local
areas.  For  example, to develop methods  to
measure forest canopy light transmittance,  a
                                              1-9

-------
                                                      o
                                                      92.
                                                      e
                                                      Q.
                                                      "5
                                                      .2
                                                      O>
                                                      CM
                                                      I
                                                      i
                                                      CO
                                                      2
                                                 «

                                             O  O
                                          o
                                          k.
                                         a.
                                          qj  03   O
1-10

-------
     matrix
Forest matrix with patch
Dominated by forest dearcutting
Residentid dominated
Urban dominated
Water
Dominated by Agriculture
Agriculture mosaic
Clouds
 Figure 1-4.  Landscape pattern types for northern Georgia.
                                   1-11

-------
team is made up  of researchers from the
Forest Service, the Tennessee Valley Authority,
the  University  of  Wisconsin-Madison,  the
Michigan  Technological  University, and the
Environmental Protection Agency. Other teams
are  being  formed  to  develop  field   and
laboratory methods for soil and foliage analysis,
and  to  learn  to use  satellite  sensors  in
conjunction  with  high-resolution photography
and the global positioning system to augment
ground-based measurements.  In this fashion,
FHM gains the benefits of cost-sharing and the
wisdom of researchers  who are experts for
particular new measurements.  It is prudent to
explore the latest technology at smaller scales
than would be required for regional-scale tests.

Looking to the Future

        Forest  Health Monitoring  plans the
following events and improvements by 1994.

• Agreements will be  reached with additional
  State and Federal agencies to increase the
  scope of FHM participation and to improve
  the efficiency of data collection and the depth
  of data analysis.

• The plot network will be expanded from 12 to
  18-24 states, including  states from the
  western and north-central regbns of the
  nation.

• The core set of measurements made at each
  plot location will be  expanded so that more
  aspects of forest condition can be monitored.
  New measurements must first pass a
  rigorous peer-review process; candidates
  include measures of soil, wildlife habitat, and
  foliage chemistry.

• The collection of off-plot forest pest
  information will be standardized.

• The set of off-plot auxiliary data bases will be
  augmented to include information from
  satellite sensors and soil surveys.

• The assessment capabilities will be
  developed further and made available to data
  analysts around the nation. Annual statistical
summaries will be prepared and the data will
be made available for in-depth interpretive
reports to address specific forest  health
issues.
                                          1-12

-------
    2.  Monitoring Methods

           The objective of this chapter is to give
    a brief overview of the forest health monitoring
    procedures that  were  used  in  1991.    A
    summary will be  given of the basic sampling
    methods, measurements, and data  analysis
    procedures.    Later chapters  will consider
    particular classes of measurements in some
    detail.   Other  references  for sampling and
    statistical analysis include Appendix A of this
    document, Overton et al. (1990), and Stevens
    et al. (in  preparation).  Monitoring operations
    and measurements are also described in  a
    series of methods guides (Conkling and Byers
    1991;  Byers  and   Van   Remortel   1991;
    Chojnacky 1991;  Conkling and Byers 1992).

    Technical Objectives

           Three technical objectives determine
    the methods of forest health monitoring. The
    first  objective  is to  estimate   the  status,
                                                    and  change over  time, of a set of  regional
                                                    forest condition  and stress indicators.  The
                                                    second   is  to   explore   some   possible
                                                    associations  between  regional  trends  of
                                                    condition and stresses. The third is to provide
                                                    statistical and interpretive summaries of the
                                                    observations that are made.

                                                    Sampling Design and Measurements

                                                           The technical  objectives stress  the
                                                    need for statistical rigor at the regional scale of
                                                    the plot network. The statistical theme is also
                                                    emphasized for  the measurements that  are
                                                    made at every plot location and for analysis of
                                                    the data.

                                                    The  Plot Network

                                                           The plot  network  is  defined  by a
                                                    regular grid of sampling points that is laid out
                                                    over the entire nation (Figure 2-1).  There are
                                                    about 12,500  sampling points on the grid,
1
                      Forest Bsafch Indicator Development and Evaluation
       Experience has shown that there is no single 'best' way to measure the health of a particular tract
of forest. The main reason for the differences is the lack of consensus about what 'forest health' means.
For example, some view forest health in terms of long-term soil productivity while others see it in terms of
aesthetic features. These valid differences of perspective can lead to very different sets of measurements
for monitoring. Even with a consensus view, however, there would still be alternatives for making particular
measurements.  For example, tree crown density can be gauged by eye or by instrument, and instruments
can be located in space or on the ground.  Indicator development and evaluation refers to the process of
sorting out the options and choosing the measurements and the methods best suited to regional monitoring.

       There is a defined procedure for identifying candidate measurements and moving  them through
a series of tests prior to their acceptance  or rejection.   This  mechanism ensures that the selected
measurements represent a good  cross-section of scientific and social perspectives about  forest health.
More measurements are possible than can ever be implemented.  It follows that measurement testing and
implementation should also be aimed at improving the efficiency of the overall measurement effort so that
assessments can be  made using as  many definitions of health  as  is  possible  within  the logistical
constraints.  Inevitable advances in biological understanding and technological capabilities will lead to
changes in the details of measurements over time.  The indicator development and evaluation strategy
ensures that  the changes do not disrupt the continuity or value  of any  long-term records  that have
accumulated.

Contact the  FHM Technical Director or Program Manager for information on indicator evaluation and
selection.
                                           ***
                                              2-1

-------
                                               s
                                               o
                                               I
                                                o
                                                .•s;
                                                o

                                                1
                                                o>
                                                .g
                                                  o
                                                 OJ


                                                  £


                                                 .S>
                                                 U-
2-2

-------
excluding Alaska, equally spaced about 27 km
apart.  There is a sampling rule for choosing a
random plot location in the vicinity of each grid
point (Appendix A).  Once defined, each plot
location is fixed. In a given year, plot locations
are  classified  as  forested  or  non-forested
(Chojnacky 1991). Non-forested plot locations
are noted, but no field measurements are made
at those locations in that year (Figure 2-2).  A
plot is established  (or remeasured) at every
forested location and the  plot further defines
the sampling rules for particular measurements
(see below).

        This  type  of  systematic  sampling
design has proven to be extremely efficient for
large-scale forest inventory and is expected to
be equally efficient for monitoring forest health.
The design provides an unbiased sample of the
nation's forests at any point in time, because
different types  of  forests  are  sampled  in
proportion to their  actual occurrence  and
distribution at that time. Land-use conversions
from forested to non-forested, or vice versa,
are expected and will not disrupt the  design.
The relative simplicity of this design ensures
that  it  will  be   possible  to  use  more
sophisticated statistical  methods  during data
analysis.  For example,  post-stratification can
be used to  group plots of different types for
more in-depth analyses, or spatial statistical
methods can be applied to estimate trends over
an entire region.  The particular triangular grid
design   also   facilitates  the  use    of
interpenetrating   sampling  methods   and
increasing the  numbers of plots  in specific
areas  without  losing the strengths  of  the
design. Other approaches that are not based
on a grid were evaluated and rejected because
they do not  provide  such a combination  of
flexibility and efficiency.

A Typical FHM Plot

        An FHM plot defines the sampling rules
that are used at a  given  location (Chojnacky
1991). The FHM plot is presently defined as a
cluster of four   subplots  that are contained
within a one-hectare area, centered on the plot
center (Figure 2-3). All measurements made in
1991 were made within these four subplots. In
the future, other measurements may be made
outside the four subplots, but they will still be
referenced to the same plot center to preserve
the overall design. It is also possible to define
sub-subplots  within  an  FHM  plot.    This
procedure is commonly used for measurements
of small trees and seedlings,  to reduce the
numbers that  would otherwise have to be
measured on the larger subplots.

        The FHM plot is now being  used  in
most forest  types.    In the  future, some
alterations  will be necessary in forests that
have exceptionally large trees which occupy a
high percentage of the plot area (redwood and
sequoia forests), and in  forests that naturally
have a very low density of trees (pinyon-juniper
forests).   Consistency is highly desirable to
maintain comparability of data among regions
and forest types, but the flexible plot design
permits modifications to deal with these special
situations.

       Although  the   plot  locations  are
considered to  be permanent,  there are no
physical indications that would permit a casual
observer to identify an FHM plot on the ground.
This is important in  maintaining an unbiased
sample and the courtesy of the landowner.
Landowners who are reluctant to have plots on
their land have the opportunity to refuse the
entry of field  crews.   These  locations are
considered 'non-respondents' for the current
year, and   landowner  permission  is again
requested at the next scheduled measurement.

Measurements in 1991

        It is beyond the scope of this chapter
to  describe  the  measurements  in  detail.
Appropriate  references  for  methods include
Chojnacky  (1991) and  Conkling and Byers
(1992).  The former reference applies strictly to
the 1991 procedures, and the latter is a more
comprehensive edition of procedures that is
being used for the 1992  field measurements.
Subsequent chapters of this  report also define
measurements   as   necessary.      Some
measurements and observations are made for
purely  descriptive purposes  and others are
made to characterize the health of the forest at
a plot location.   Still other measurements can
                                          2-3

-------
Figure 2-2.  The locations of sampling points that were classified (all or in part) as forested are
            shown.
                                            2-4

-------
Distance between points is  120  ft.

Azimuth  1-2 360*
Azimuth  1-3 120*
Azimuth  1-4 240*
 Subplot
 24.0' radius (7.32 m)
Annular Ptet
58.9' radius (17.95 m)
                             ' Micro-plot
                              6.8' radius (2.1 m)
                              12' @  90' az. from
                              subplot center (3.66 m)
                     Figure 2-3.  The FHM field plot design.
                                   2-5

-------
                                             , «**> Ct^y Assurance
       A large program such as Forest Health Monitoring requires a substantial amount of coordination
to ensure that all field operations are completed in a timely and high-quality fashion.  Our field crew
candidates come from many agencies with various backgrounds, and each must complete a  rigorous
training session to become certified 'forest health monitors.'  Every year, the actual field work is preceded
by inter-regional, pre-training workshops to ensure that methods are being taught consistently across all
regions.  In 1991, the formal training sessions were held in Durham, NH and Asheville, NC.

       At the training sessions, the final logistical preparations are also made and the field equipment is
distributed to the crews. To the field logistics specialists, this is just one more step in a year-round process
of procurement,  maintenance, inventory, and planning. All logistical arrangements and personnel are
coordinated at the national and regional levels to ensure consistency and efficiency of the field operations
and staffing. A logistics plan specifies 'how the job will get done', and it is safe to say that without logistical
support, very little could be accomplished.

       Quality assurance (QA) is not just  a 'buzz-word' for the Forest Health Monitoring  program.  A QA
program is  required by several of the participating agencies for all data collection activities, and there are
very specific guidelines to folbw.  But QA is also more than just an agency requirement.  Without QA, the
data that are collected would be of unknown quality and therefore much less useful for  assessments of
forest health. Few rf any forest monitors think that their data are of poor quality, and they welcome a QA
program to document just how good they are!

Contact the FHM Technical Director or Program Manager for information on training,  logistics, and quality
assurance.
                                           ***
                        Poiiabte Data i$ft&»«i#B* Sp«<>d Field Operations
       Portable data recorders (PDRs) are the compact and rugged field'computers that Forest Health
Monitoring crews use to record the observations and measurements they make.  An interagency team of
computer programmers, data analysts, and crew persons collaborated to design, test, and deploy PDRs
for data capture.

       From the start, FHM has used PDRs because they speed up data collection and processing, and
because they improve the quality of the data. No longer is data processing deferred until the end of the
field season; every several days the field data are transmitted to a larger computer where it is verified and
entered into the data bases.  At the same time, crew status and location, equipment needs, and weekly
plans are transmitted so that  field operations can be fine-tuned during the season.  The use of PDRs
improves data quality by providing real-time verification. This ensures that only valid data codes are used
and that all the measurements are made on each tree or plot.  In some cases, apparently extreme values
are brought to the attention of the operator for on-site verification.

Contact the FHM Technical Director or Program Manager for information on portable data recorders and
information management.
                                            ***


                                              2-6

-------
be used either as indicators or as classification
variables depending on the particular analysis.

       It is convenient to group the descriptive
measurements and  observations into  three
categories depending on their level of detail at
a given plot location (Table 2-1). The plot-level
category consists of measurements of the plot
location as a whole. The subplot-level category
has measurements that are made at the center
of each of  the four subplots.  The microplot-
level category has measurements that apply to
the sub-subplots, the microplots, where small
trees  are   measured.     The   descriptive
measurements are used to  characterize the
sample of plots.

       Each subplot may be made up of very
different sub-areas.  For example, a subplot
may include  both forested and non-forested
land uses.  This within-plot variation must be
accounted  for when  aggregating the data for
analyses.  In the field, the transitions between
land uses,  forest  types, stand origins,  stand
sizes, and  past disturbances are  mapped for
each subplot.  In the following chapters, these
recognizable  categories are  referred  to as
condition categories (Chojnacky 1991).
Table 2-1.      Schedule   of  descriptive
measurements  and  associated  information
collected for FHM plots in 1991. See text for
definition of information categories.
Plot-level information
  Plot number
  State
  County
  Old plot status
  (Aerial) photo year
  Date of measurement
  Elevation
  Terrain position
  Crew identification

Subplot-level information
  Land use class
  Forest type class
  Stand origin class
  Stand size class
  Past disturbances
  Slope correction
  Slope percent
  Aspect
  Microrelief
  Subplot map

Microplot-level information
  Land use class
  Forest type class
  Stand origin class
  Stand size class
  Past disturbances
       A set of mensuration measurements
(Table 2-2)  is made partly  for descriptive
purposes and partly to estimate  growth and
related indicators following remeasurements in
later years.  In 1991, all  large trees (stem
diameter > 12.7 cm) within each subplot, and
all saplings (stem  diameter between  2.5 cm
and 12.7 cm) within each microplot were tallied
separately.  Seedlings (stem height > 30.5 cm
and  stem diameter  < 2.5  cm)  within  each
microplot were tallied as frequency by species,
condition class, and crown class.  The large
tree  mensuration data were also used for an
analysis of tree species diversity (see Chapter
5).
Table 2-2.      Schedule   of   mensuration
measurements for FHM plots in 1991.  See text
for explanation of tree categories.
Large trees and Saplings
 Tree number
 Location on the plot
 Tree history (e.g., live, ingrowth, cut)
 Species
 Stem diameter
 Condition class
 Crown class

Seedlings
 The frequency of seedlings is tallied by:
       Tree species
       Condition class
       Crown class
                                           2-7

-------
        A final set of crown classification and
tree damage measurements  were  made on
large trees  (Table  2-3).   In  addition to the
measurements   shown  for  large  trees  on
subplots  in  Table  2-3,  the  frequency of
seedlings by vigor class,  and the vigor class,
damage,  and crown  ratio of  saplings  was
measured  on   microplots.     This  set  of
measurements is intended to partially represent
forest health in  terms of  visual evidence of
stress  responses and damage,  especially to
overstory trees. The majority of measurements
are not diagnostic indicators but rather help to
characterize overall tree condition. The crown
classification  measurements  are reported in
Chapter 4.
Table 2-3.      Overstory crown condition and
tree  damage measurements  made on large
trees within subplots in 1991.  See text for
measurements of seedlings and saplings.
Crown measurements
 Crown ratio
 Crown diameter width
 Crown density
 Crown dieback
 Foliage transparency

Damage measurements
 Up to three 'significant' (in terms of obvious
 impact on tree health) damages to the tree
 are noted,  and the following information is
 recorded for each:

  Condition  (e.g., dead, open wound,
    deformed, watersprouts)
  Location on the sample tree
  Probable cause (only if obvious; e.g.,
    insect, fire, logging)
Data Analysis

       The data were summarized in several
ways for this report.  A statistical summary of
most measurements is based upon cumulative
frequency distribution methods described below
(see also  Appendix  A).   The  methods  of
indicator analysis  in  this  report are  under
continuous review as part of the FHM indicator
development and selection process. Reviewers
of  this  report  have  suggested alternative
methods of data analysis for  several of the
indicators, which may improve the interpretation
of the  results.  The FHM team has accepted
these comments  and  will be studying these
suggestions for data analysis of  existing  and
new indicators. Tabular  summaries were also
prepared in some cases.   Finally,  indicator
leaders sometimes use innovative techniques
to highlight particular aspects of their analyses.

        At the first stage of data analysis, the
emphasis  is  on  providing  objective  data
summaries that will be valuable for looking at
forest health from a variety of perspectives.
The techniques for these purely statistical
summaries were peer-reviewed  (during  the
design of FHM and EMAP) before they were
used in this report. The value of this rather
conservative  approach is that there will be
much more consistency over time and among
regions, in comparison to an ad hoc approach
every year.  Once a reader  learns  how to
interpret the summary of one  measurement,
then it  is  a simple  matter to interpret  the
summary of every measurement, every year.
The drawback is that, at  this stage of program
development, only a limited set  of statistical
summary procedures have been agreed upon.

        In  later  stages  of  analysis,  the
measurements are interpreted from more than
just a statistical point of  view.  Here is where
value judgements become extremely important.
It  is recognized  that  not  everyone's  value
judgement  is exactly  the same,  and  so not
everyone will necessarily agree with any single
interpretation.  Where interpretations are made
in this report, the analysts state carefully their
assumptions and the  scientific basis of their
interpretation.

        Rather than being of a purely statistical
or purely interpretive  style,  this  report is  a
hybrid.   In  time, two types of  reports will be
produced when the amount of data becomes
sufficient. For this report, the basic statistical
summaries appear because their preparation
                                              2-8

-------
    forced the annual process of data verification,
    validation, and data base documentation to a
    timely conclusion.  Individual indicator leaders
    interpreted their particular results based upon
    available  data  and  accepted  interpretive
    techniques. This middle ground was chosen
    because we know in advance that we cannot
    fully interpret all  the measurements, or forest
    health in  general,  when the program has not
    been fully implemented.  On the other hand,
    the more  common measurements should  be
    interpreted in ways that have been acceptable
    historically. The intention is to provide the most
    information possible from the limited data, and
    at the same time, to acknowledge differences
    of  opinion and  preserve options for other
    analyses and conclusions.

           This report was subject to  a review
    process  to validate the methods that were
used.  As experience grows  with  particular
measurements, it may be possible to routinely
include  interpretive   procedures that were
originally  tenuous  but which  have  become
more accepted over time.

       This first report makes few attempts to
analyze many measurements at one time. This
is mainly because there are not many accepted
methods for doing so at present. There is also
not  a  sufficient breadth  of measurements
available for comprehensive, biological model-
based analyses  of forest conditions.  Where
feasible, measurements have been condensed
into  index values  which  were  statistically
summarized and interpreted. The choice of an
index is itself an interpretation, and  so these
indices were also subject to the peer review
process.
                       HOW to Interpret Cumulative Distribution Functions
       The cumulative distribution function (CDF) graphical technique that is used extensively in this report
is unfamiliar to many readers.  For the casual reader, this section will describe how to interpret a CDF.
Appendix A supplies additional details for those who require a more rigorous mathematical treatment. The
effort to learn the technique is modest, and the reward is substantial;  many monitoring programs around
the world use this simple, powerful, and informative graphical technique for presenting data summaries.
Parts of the following description follow closely the discussion in Linthurst et al. (1986).

       A CDF is better than a tabular summary for presenting an objective view of FHM data. Specifically,
'poor1 forest health is not unequivocally or universally defined (see sidebar on  "What is Meant by 'Good'
Forest Health?"). The use of a CDF permits the reader to choose a 'critical value' of a measurement, and
to see what  proportion  of  the  sampled  population is  estimated to fall below the threshold for that
measurement.  Tabular  summaries summarize data according to a critical value defined by the data
analyst, but the reader is unable to see  how an interpretation could change if a different value was chosen.
If needed, a tabular summary with confidence  intervals can be  prepared from CDFs for any particular
critical values.

       Each CDF  is identified by the variable X (a measurement  or an index, on the horizontal axis) and
by the subset of plots  in the population being examined.  For example, X could be a crown rating score
and a subset of the population could be all oak trees found in oak-pine forests.  To find the estimated
proportion of the population that falls below some critical value of X, use this procedure (see Figure 2-3):
1.     Find the chosen  critical value (X) on the horizontal axis.
2.     Draw a line straight  up to meet the solid line that is charted on the CDF.
3.     Draw a line straight  left to meet the vertical axis.
4.     Read the estimated  proportion  where the line meets the vertical axis.
5.     The proportion is the fraction of the population that is estimated to have a value of X that is less
       than the critical value.
                                              2-9

-------
        In the example, a critical value of 17 in step 1 leads to the interpretation that about 47% of the
population has a value of X less than 17.  Different critical values (in step 1) will yield different population
estimates (in step 4). The critical value is really a criterion by which 'poor health' will be gauged.  By
experimenting with critical values, the reader may see how data interpretations change with the criterion
chosen to signify 'poor health1.  For  example,  a different critical value of 10 in step  1  leads to the
interpretation that about 38% of the population is in 'poor health' (this example is not shown in Figure 2-3).

        There are some  important things to remember when using the above procedure.   First, this
graphical technique is suitable for rough estimates, but precise computation requires the use of data base
and appropriate algorithms (see Appendix A). Second, the CDF cannot be used (strictly) to get estimates
of proportions greater than some value of X.  Instead, a new chart (of 1 - F(X)) must be prepared for that
purpose, again using the  data base and appropriate algorithms.

        The CDFs in this  report have solid and dashed lines charted.  The solid line yields the population
estimate as described above. The dashed lines yield the lower and upper 90% confidence limits for the
population estimate.  A good statistical text and Appendix A should be consulted for a discussion of what
the confidence limits mean.

        In summary, the CDF technique is a flexible format for presenting an unbiased view of FHM data.
It permits any reader or analyst to impose a value judgement of what 'poor" health means, and to see what
fraction of the population is estimated to be in 'poor" health  according to that criterion.  The CDF does not
change the data in any way , it simply presents it in a non-tabular fashion.  Despite the simplicity of the
approach, some readers will never be comfortable with the CDF technique. For these readers, only the
possibility of making one's own interpretation will be lost;  it  is always possible to accept (or argue) the
interpretations that other analysts will make in later chapters or in other reports.
                                            ***
                                              2-10

-------
     Population:  Oak species in oak-pine forests
F(x)
                        10
20
30
40
50
                            Hypothetical Crown Rating Score
                      Figure 2-4.  Interpreting a CDF.
                                     2-11

-------
3.   Characterization of Stand
    Density and Number of
    Trees on FHM Plots

    T. D. Droessler

Introduction

       The mensuration statistical  summary
presents  baseline estimates of basal area for
trees 12.7 cm DBH and larger and numbers of
stems/ha  for seedlings,  saplings, and trees,
both separately and combined.  Tables of the
number of subplots by state, land use, major
forest type, stand size, stand origin,  elevation,
natural and anthropogenic disturbance history,
terrestrial  position, elevation  class,  subplot
characteristics, and size class of tally trees are
also provided (Appendix E).  Point estimates of
mensurational characteristics can be used to
help  interpret  other  measurements.    For
example,  the  probability  of  a forest  pest
infestation is frequently a function of the density
(basal  area or otherwise) of combinations of
tree species.

Objective

       The objective is to present cumulative
distribution  functions   (CDFs)   with   90%
confidence bounds of basal area per hectare by
major forest  group over the subplots,  and
numbers  of seedlings, saplings, and trees per
hectare over the microplots, separately  and
combined,  for  the   Northeast  (NE)   and
Southeast  (SE)  United   States.     Other
characteristics of the subplot/microplot tally are
presented in tables.   Minimal  commentary
regarding the CDFs or tables is provided.

CDFs  of Basal Area by  Forest
Group  for  the Northeast  and
Southeast

Method

       The data from  the 12 states was
divided into Northeast and Southeast states.
The  Northeast  states  were  Maine,  New
Hampshire, Vermont, Massachusetts, Rhode
Island, and Connecticut. The Southeast states
were   New   Jersey,  Delaware,  Maryland,
Virginia, Georgia, and Alabama.

       The Northeast and Southeast data sets
were  divided  into major  forest groups.  The
major forest groups were partitioned as follows:
the pine group consisted of the white, red, and
jack, longleaf-slash, and loblolly-shortleaf pine
groups; the  spruce-fir group consisted  ol
spruce   and  fir;  and   the   hardwoods/
miscellaneous group took in the remaining
categories. The species groups were originally
further subdivided (e.g., the longleaf-slash pine
group), but the sample sizes were too small to
be meaningful.

       Only  forested  land  use  portions  of
subplots (wholly or partially) were included in
the calculations. Each subplot was considered
a sampling unit.  Basal area by forest group
was first  calculated by  subplot  and  then
expanded to a hectare basis before calculating
the mean for the plot.   If  basal  area  of a
particular forest group was present on only one
of the four subplots, the basal  area values for
that forest group on the other three  subplots
were included as zeros in the calculation of the
mean  basal  area.   The calculations were
conducted as follows.

       First the basal area (m2) was calculated
by forest group for each subplot. The condition
number was  used for subsetting to  a forest
group. Next the forest group basal area was
expanded to a per hectare value by multiplying
by the subplot per hectare  expansion factor
(approximately 59.48).

       Next,  a weighted mean  basal  area
(m2/ha) was calculated by forest group over the
four subplots.   The  weight  used  in the
numerator was the fractional area  of each
forest group  for  each subplot  calculated by
summing the fractional area percents (area
percent divided by 100) by condition  number.
A weight of 1.0 was  used if the whole subplot
had a forested land use and was of a particular
forest group.  It was possible for a plot to have
zero basal area  if  it  was  classified as  a
                                         3-1

-------
forested land use with a forest type identified,
but there were no trees greater than 12.7 cm
DBH present.

       The denominator in the weighted mean
was the total area  over the four subplots
classified as forested land use (maximum of
4.0). The sum of the weights in the numerator
for a  particular  forest group  equalled  the
denominator only if  all forested area on  all
subplots was of a particular forest group.
       The mean basal area for a forest group
represented an observation for calculating a
CDF. The CDFs can be used to determine the
cumulative probability of the relevant forested
population  that has a  basal area equal to or
less than a specified level.

Figures

       Figures 3-1 through 3-3 present CDFs
with 90% confidence  bounds  of basal area
(m2/ha)   by  major forest  groups  for   the
Northeast and Southeast states. The maximum
basal area ranged from 0 to 60 m*/ha for any
forest group in the Northeast and Southeast.

       Figure 3-1 shows CDFs of basal area
in m2/ha for the pine group and the spruce-fir
      group in the Northeast. The 90% confidence
      region for the pine group CDF is large because
      the sample size used to calculate the CDF was
      24 hexes.

             Figure 3-2 shows CDFs of basal area
      in m2/ha for the hardwoods and miscellaneous
      group in the Northeast and for the pine group
      in the Southeast.

             Figure 3-3 shows CDFs of basal area
      in m2/ha for the oak-pine  group  and the
      hardwoods  and miscellaneous group  in the
      Southeast.

             As an example of how the CDFs can
      be used, consider the basal area associated
      with specified cumulative distribution functions
      (Table 3-1).

             Looking at NE Pine, 75 percent of the
      population  have a basal area equal to or less
      than 31 m2/ha.  For SE Pine, 75 percent of the
      population have a basal area equal to less than
      12  m2/ha.    The  interpretation of  these
      differences   must   be   based   on  other
      characteristics  of the  population  which  is
      beyond the scope of this example.
Table 3-1.      Basal areas associated with cumulative distribution probabilities of 0.25, 0.50, and 0.75 for
               Figures 3-1 to 3-3.
Figure
                                                Cumulative Probability
0.25
0.50
0.75
1 (a) NE Pine
1 (b) NE Spruce-fir
2 (a) NE Hardwoods + miscellaneous
2(b) SEPine
3 (a) SE Oak-pine
3 (b) SE Hardwoods + miscellaneous
18.
7.
3.
0.
1.5
2.
27.
18.
13.
3.
6.5
10.
31.
30.
22.
12.
15.
21.
                                             3-2

-------
       (a)
          0.0
             0.0
10.0       20.0       30.0       40.0       50.0
60.0
       (b)
                                    Basal Area ( sq m / ha )
             0.0
10.0       20.0       30.0       40.0       50.0       60.0
                                    Basal Area (sq m / ha
Figure 3-1.   Cumulative distribution function of (a) basal area for the pine group in the NE, and
            (b) basal area for the spruce-fir group in the NE.

                                          3-3

-------
       (a)
         1.0 T
         0.8 -•
  Rx),
         0.2
         O.Q
            0.0        10.0       20.0      30.0      40.0

                                   Basal Area ( sq m / ha )
       (ft)
         1.0 T
         QiS -•
  F(x)
         0.4 -•
         0.2 -,
         0.0
    	1	1	1	1—

0.0        10.0      20.0       30.0       40.0

                       Basal Area  (sq  m / ha )
50.0       60.0
            -t
                                                                  50.0       60.0
Figure 3-2.  Cumulative  distribution  function of  (a)  basal  area  for  the  hardwoods and
            miscellaneous group in the NE, and (b) basal area for the pine group in the SE.
                                          3-4

-------
       (a)
         1.0 T
            0.0
10.0       20.0       30.0       40.0

            Basal Area ( sq m / ha )
       (b)
         1.0 T
  F(x)   0.6-•
         0.0
            0.0        10.0      20.0      30.0      40.0

                                   Basal Area  (sq m / ha )
50.0       60.0
                                            50.0
           60.0
Figure 3-3.   Cumulative distribution function of (a) basal area for the oak-pine group in the SE,
            and (b) basal area for the hardwoods and miscellaneous group in the SE.

                                         3-5

-------
CDFs  of  Seedlings,  Saplings,
Trees   and   Stems   for   the
Northeast and Southeast

Method

       The hex data from the 12 states were
divided into Northeast and Southeast states in
the same way as was done for basal area
above.  The seedling and  sapling  count for
each microplot was expanded to a per hectare
basis  and a  mean  calculated  over  the
microplots on the hex (maximum of four). The
mean  count  for a  hex  was used as an
observation to calculate the CDF.  Microplots
with zero seedlings or saplings were included
in  the mean  as  long as they had a forested
land  use classification.    Although  it was
possible to have more than  a single condition
class on a microplot, only two microplots (out of
more  than 2000) were  split.   Therefore,
condition class was ignored. Tree count was
calculated as a weighted mean  over the
subplots  using  the forested  area  on  each
subplot  as  a  weight.    Stem  count was
calculated as the sum of the seedlings and
saplings  on the microplot and trees on the
subplot,   using  the  appropriate  expansion
functions to bring the  estimates to a hectare
basis.

       Figures   3-4   through   3-7  show
seedlings, saplings,  trees  and  stems  per
hectare for the Northeast and Southeast states.

Seedlings

       Figure 3-4 presents CDFs of seedlings
per hectare for the Northeast and Southeast.
The mean number of seedlings per hectare is
greater in the Northeast  than the Southeast
throughout the range of seedling density.

Saplings

       Figure 3-5 presents CDFs of saplings
per hectare for the Northeast and Southeast.
The mean number of saplings  per hectare is
greater in the Northeast  than the Southeast
throughout the range of sapling density.
Trees

       Figure 3-6 presents CDFs of trees per
hectare for the Northeast and Southeast. The
mean number of trees per hectare is greater in
the Northeast than the Southeast throughout
the range of tree density.

Stems (seedlings, saplings and trees
combined)

       Figure 3-7 presents CDFs of stems per
hectare  for  the  Northeast  and Southeast.
Consistent with Figures 3-4 through  3-6, the
mean number of stems per hectare is greater
in the Northeast than the Southeast throughout
the range of stem density.
                                        3-6

-------
       (a)
   F(x)
         1.0 T
         0.8 -•
         0.6 -
         0.4 -•
         0.2 -•
         0.0
       (b)
         1.0 T
                            20000            40000            60000

                              Number of Seedlings per Hectare
                            20000           40000            60000

                              Number of Seedlings per Hectare
Figure 3-4.   Cumulative distribution function of (a) number of seedlings/hectare in the NE, and (b)
            number of seedlings/hectare in the SE.
                                        3-7

-------
        (a)
          1.0



          0.8



   F(x)    0.6


          0.4
          0.2 -
          0.0
         ^ r
      f r
                        2000       4000        6000        8000

                               Number of Saplings per Hectare
       (b)
          1.0 T
          0.8 -
   F(x)    0.6-
          0.4 -•
         0.2 -
         0.0
             0
2000       4000        6000        8000

       Number of Saplings per Hectare
                                                 10000
10000
Figure 3-5.  Cumulative distribution function of (a) number of saplings/hectare in the NE, and (b)
           number of saplings/hectare in the SE.

                                        3-8

-------
       (a)
         1.0 T
         0.0
                     200      400      600      800     1000

                                Number of Trees per Hectare
                                           1200    1400
       (b)
         1.0 T
         0.8 -•
   F(x)   0-6 -•
         0.4 -•
         0.2 -
         0.0
-t-
                                            -f-
-\
             0       200      400      600      800     1000    1200     1400

                                Number of Trees per Hectare
Figure 3-6.  Cumulative distribution function of (a) number of trees/hectare in the NE, and (b)
           number of trees/hectare in the SE.
                                        3-9

-------
       (a)
         1.0 T
         0.0
                           20000           40000           60000


                            Total Number of Stems per Hectare
	1	

 20000
                                     -i-
                                            40000
60000
                            Total Number of Stems per Hectare
Figure 3-7.  Cumulative distribution function of (a) total number of stems/hectare in the NE, and
           (b) total number of stems/hectare in the SE.

                                       3-10

-------
4.   Crown Condition  of Forest
Trees on FHM Plots


K. Stolte, R. Anderson,  W.  Burkman,
T. Stockton and R. Binns

Introduction

       The crown portion of trees has always
been an indicator  of forest tree health.  Long
before scientists attempted to add the scientific
process  and  rigor  to  the  estimation   or
measurement of crown variables, humans have
used tree crown condition as an instantaneous
visual estimation of the vigor of individual trees,
tree  stands, and sometimes  entire  watersheds.
With the possible exception of root pathologies,
healthy tree crowns can be equated with good
condition of trees.  Conversely, unhealthy tree
crowns are clear indicators of threats to tree and
stand productivity, biodiversity, and sustainability.
Additionally,  when natural  or  anthropogenic
stresses  are affecting a forest, the first signs of
deterioration  are often  observed  in  the  tree
crowns.   Since tree crowns  form  the  basic
structural architecture of a forest ecosystem, they
directly affect the composition,  processes, and
vigor of the   understory  floral   and  faunal
components of the forest.

       Morphological  determinations of crown
and bole condition were made on all forest trees
with diameters at breast height greater than 12.7
cm.  Measurement and estimation of variables to
describe  tree  crown condition have  provided
reliable information on the growth characteristics
of trees.   In the eastern United States, crown
diameter has  been related  to  the size  of
hardwood  trees  (Francis  1986;  Sprinz  and
Burkhart  1987), and crown  density  has  been
related to the growth of loblolly pines (Grano
1957; Anderson and Belanger 1991; Anderson et
al. 1992; Belanger and Anderson 1991).  In the
western United States, crown  ratio and crown
density  have   been related  to  growth  and
survivorship of conifers  (Dolph 1988).  Other
crown variables, such as dieback, transparency,
and density, can be related to insect defoliation
and subsequent growth and survivorship effects
on both conifers and hardwoods (Kulman 1971;
Schmitt et al. 1984).

        Crown transparency refers to the amount
of sunlight  that  passes through  the  foliated
portions of the tree crowns ignoring holes in the
tree   crown  caused  by  lack  of  branches.
Transparency   partially  overlaps   with  crown
density; the difference is  in the emphasis on
foliage.  For example, a tree could have only one
branch with thick foliage on  that branch.  This
tree would score  poorly  for crown density (and
live crown ratio), but would score well for crown
transparency (i.e., it would have  a low crown
transparency). Crown transparency is estimated
in five percent classes, with a range of 0-100%.
Trees  with  a  low crown transparency  score
increase the potential for carbon  fixation and
nutrient storage, and indicate a lack of defoliating
agents that attack new and older foliage.

       Crown dieback refers to the mortality of
relatively new branches (less  than 2.5 cm in
diameter) that are in the upper, sunlight-exposed
portions of the crown. The premise is that these
branches have died from some stress other than
normal  branch  mortality  which  occurs  from
shading.  On open-grown trees, crown dieback
can occur on most of the crown.  On codominant
trees, crown dieback is confined primarily to the
top and upper sides of  the  crown that  are
exposed to  direct sunlight.  Crown dieback is
estimated in five percent classes, with a range of
0-100%.  Trees with a low crown dieback score
increase the potential for carbon fixation and
nutrient storage, and indicate a lack of defoliating
agents that attack new foliage.

       Crown  position  refers  to  the  relative
position of the tree crown in a stand of trees.
The five crown positions often utilized to  stratify
crowns  are: open grown (crown  exposed  to
sunlight on  all sides);   dominant  (tree crown
above the general height of the stand; some of
the sides  of  the crown  in  direct  sunlight);
codominant (tree crown about the same height as
the general  height of the  stand; top of crown
exposed to  direct sunlight);  intermediate  (tree
crown generally below height of the stand; very
top of crown may be exposed to  some direct
sunlight); and suppressed (tree crown far below
                                             4-1

-------
the general height of the stand; tree crown may
never be exposed to direct sunlight). Only grown,
dominant, and codominant trees are used in the
analysis of crown data.

       Crown  density  is  the two-dimensional
appearance   of  crown  fullness  when  ideally
viewed against the sky.  It estimates how much
of the sky is blocked from view by the upper tree
bole,  branches,  foliage, and any  reproductive
structures (e.g., cones). Crown diameter is the
measured average of the crown's widest part and
the 90° perpendicular axis to the widest part of
the crown. It is measured by standing under the
estimated drip line of the crown at the  two axes
described.

       Crown  ratio  is  the  percentage  of  the
entire tree bole, from the ground to the top of the
tree,  that supports   living,   foliated  canopy
contributing to the vigor and growth of the tree.
It is length of  the bole that is foliated (X), divided
by  the total  length  of  the bole  (Y).   These
variables  have been shown  to  be important
measures of growth and survivorship, and will be
represented  by appropriate indicators  in  future
statistical summaries.

       Crown  assessments  directly  address
productivity, sustainability,  aesthetics, and more
indirectly biodiversity.  Contamination is partially
addressed by crown ratio, density, and dieback.
Gaseous pollutants are known to accelerate leaf
senescence,  reduce  production of new foliage,
and cause mortality of lower or upper branches
(Stolte et  al.  1992).   Air pollution can produce
thinner crowns (density) that are reduced in size
(ratio)  and   that may  have dead  branches
(dieback) in susceptible species and individuals.
Forest extent  is not  addressed  by   crown
assessments.   Deterioration of  tree crowns
affects the sustainability of  forests.   Nutrient
cycles, light regimes, and climatic habitats  are
altered.
Crown Defoliation Indicator

       The   Crown   Defoliation   Indicator
developed for the FHM 1992 Statistical Summary
is an additive index.  This index and  its  CDF
thresholds discussed below  have evolved from
some of the initial  work on crown indices by
Anderson et al. (1992).  This  indicator addresses
the  defoliation  of  the  tree  crown  (Crown
Defoliation Index [GDI], composed of dieback and
transparency). The crown variables transparency
and  dieback were aggregated into a  plot-level
index (indicator) (Equation 1), which reflects the
nature of the tree crown  response to stressors.
These variables reflect the tree foliage condition,
both within the crown (transparency) and in the
sun-exposed outer crown foliage  (dieback).  At
this point in time, the variables in the GDI are not
scaled by species and have equal weights:
GDI = (Transparency + Dieback) / 2
(1)
       The focus of the FHM sampling design is
regional assessments of forest health, focused on
aggregating plot-level  data  into a single datum
point for  analyses using cumulative distribution
functions.  The  data were post-stratified  for
analysis in  three ways: by species, forest type,
and crown group (species  with similar crown
characteristics such as long  needle pines [Millers
et al.  1992]).   Cumulative distribution functions
were based upon plot-level averages which were
weighted using basal area.   Each stratum must
contain enough plots to calculate CDFs. Based
on consultation with the EMAP Statistical Team
(Cassell and Hazard pers. comrn.), 50 plots has
been chosen  as the minimum  for calculation of
CDFs in this report.


Thresholds  in CDFs

       Determining the points in a variable's
population  distribution that  are considered as
anthropogenic valuation-based  thresholds  is a
key component of assessing forest health.   In
general,  thresholds delineate a population  into
acceptable (nominal and optimal subclasses) and
undesirable (subnominal and poor subclasses)
categories  and should be based on reasonable
and  consistent criteria for  all species,  forest
types, or condition classes. There are several
possible   approaches  for   determining   CDF
thresholds, but   in   all  cases,  any  criteria
developed  should address  a  relevant societal
                                              4-2

-------
value.  Two  relevant  approaches  for setting
thresholds for crown variables with respect to
forest health are:

1. Tree growth/survivorship relationship to crown
   condition  (addresses  productivity   and
   biodiversity):

•  Based on  experimental data  relating  tree
   height or basal area growth to the condition of
   the tree crown.

•  Based on expert opinion ( >10 years exper-
   ience in research/monitoring tree crowns and
   tree vigor).

2. Visual  appearance of the tree (addresses
   aesthetics):

•  Public perception of healthy trees.

       Choosing  the method  for  delineating
thresholds  in CDFs depends  on the intended
assessment or endpoint requirements of the user
information.  Industry may be interested in forest
productivity; thresholds based  on growth would
be the most meaningful.   Alternately, the public
may be primarily concerned with aesthetics and
thresholds based on the aesthetic appearance of
the crown may  be more  meaningful.  Another
group, such as  foresters, may  be interested in
both growth and the visual appearance of the tree
crowns.  A limitation to both approaches is  the
lack of data about the relationship between crown
condition and growth or appearance that may be
currently  available  from  the   literature.     In
comparing  thresholds  set for productivity  and
those set for aesthetics, it must be recognized
that  large differences in the  perception of forest
health may occur  depending on  the type  of
threshold  used.   For most  species  and forest
types, it is likely that thresholds for aesthetics will
be more conservative than thresholds for growth;
the visual appearance of the tree may become
undesirable  well in  advance of any negative,
discernible growth  effects.   In general,  it  is
surmised that aesthetic thresholds will be more
rigorous than growth thresholds.  In this report
the thresholds  were  set  based on known  or
suspected relationships between crown condition
and tree growth  and/or survival  (Anderson et al.
1992; Belanger and Anderson 1991).
        Ultimately FHM would like to have data
about the growth-crown condition relationships for
all common species  (identified in the frequency
distribution  analyses),  since  forest growth is
probably a better estimate of resiliency than is
visual appearance (aesthetics).  The thresholds
based on growth effects may be determined by:

•  Specific  growth  studies   in  which  crown
   condition estimates were  compared to tree
   growth rates or tree DBH size.

•  Existing  models  of  crown  condition  and
   growth; values set  for variables in those
   models.

•  Peer-reviewed literature.

•  Estimates by scientists weighted by years of
   experience  working  with  the  species  of
   concern.

        In this report three thresholds have been
used  in the CDFs:  poor, concern,  and optimal
(Figure  4-1).    Since  each  threshold value
intersects the CDF and the confidence bounds for
the CDF, each threshold also defines a proportion
of the population in  a specific category and  the
confidence range for that proportion (Figure 4-1).
A  brief  discussion   of  the  purpose of  each
threshold  and  its  effects on the  population
follows:

•  Concern Thresholds-A concern threshold is
   the nominal-subnominal  break point at which
   one  becomes concerned  about the values
   measured or estimated for a variable.  It is the
   threshold  based  on  growth data or  expert
   opinion, at which the tree(s) may begin to be
   stressed to  a degree that is  detrimental to
   growth or survival.  Delineation of a concern
   threshold divides the population into nominal
   and subnominal portions.

•  Optimal Thresholds-An  optimal threshold is
   the breakpoint which indicates a  portion of
   the nominal  population that is considered to
   be in superior condition.  On plots considered
   optimal, the condition of the crown variable is
   outstanding  relative  to  values that  can  be
   expected  for  individuals  or  stands  of a
   species,   forest   type,  or   crown  group.
                                              4-3

-------
      020
      OX
      000
                    Delotcmon indicator
                      Lotlolyflne

                       f
         • ssx^
            400
                   800    1200
                   Index value
   Figure 4-1. The three established thresholds
           for the Crown Defoliation Indicator.
      Delineation  of   an  optimal   threshold
      highlights an  optimal  proportion of the
      nominal population.

   •  Poor Thresholds-The poor threshold is the
      breakpoint   which   delineates   the
      subnominal portion of the population that is
      considered to be in serious condition.  The
      trees on these plots are considered either
      to be susceptible to mortality as a direct
      result  of  a particular  stress  or to  be
      exceptionally vulnerable to other  forest
      stresses because of that particular stress.
      Delineation of a poor threshold highlights
      the poor  proportion of  the  subnominal
      population.

        The degree of uncertainty in  delineating
the population  further depends  on  where the
thresholds are placed in relation to the curve of
the CDF. If a threshold occurs where the slope
of the CDF curve is steepest, then the  uncertainty
in delineating the  population segments will be
increased (i.e., the confidence bounds on the
proportion of the population are wider).

        The values selected for CDF thresholds
in this report are derived from the visual crown
rating (VCR) classification model developed by
Anderson et al. (1992).  This  model is based on
relationships   between  crown  condition  and
growth, derived from the available literature and
expert opinion on crown condition relationships to
tree growth and mortality for a limited number of
species. The determination of tree crown health
of other species is based on  expert opinion of
foresters with many years of experience working
with the species found in the eastern forests, and
their evaluation of  crown health reflects what is
normal  for each species.  They recognize that
some species normally have less dense crowns,
small live  crown  ratios,  etc.   Although  the
thresholds proposed for the  CDFs  in this report
are not solely based on functional relationships
between crown condition and tree growth/survival,
they  do  broadly  reflect  those  relationships.
Thresholds  set  to  delineate the  subnominal
proportion of the population are intended to infer
reduced growth for those  proportions of  the
species, forest type, or  crown group population
under consideration.

Table 4-1.   CDF  concern,  optimal, and poor
            thresholds for the crown variables
            dieback and   transparency.    For
            dieback and transparency, only one
            value  represents all species, forest
            types, and crown groups.
CDF
Crown
Variable
Transparency
Dieback
Thresholds(%
Optimal
20
5
)
Concern
45
20

Poor
70
30
      We decided to use a single value for each
threshold  since  these values are  intended  to
represent  the mode for a range  of values for a
variety of  species. We did not feel we  had
sufficient data to calculate any confidence bounds
for  the  threshold  values  and  question  the
appropriateness   of  confidence  bounds when
these values are intended  to be generic  and
represent a wide range of species. We recognize
the need to obtain much more information about
the relationship between tree crown condition and
tree growth (shoots and roots) and survivability.
By  setting thresholds, we were  able to make
                                               4-4

-------
inferences concerning the status of the forests in
12 eastern states based on the 1991 data and
will be able  to track changes in the status  of
these forest over time.

Results

      The CDFs  for the Crown Defoliation Index
were generated for the most common species,
forest  types,  and  crown  groups  based  on
frequency analysis of occurrence on Detection
Monitoring plots (Tables 4-2 through  4-9). The
Standard Federal Regions are shown in Figure
4-2.
      A cursory evaluation of the CDF analysis
results indicates that most  tree  species,  all
examined forest types, and most crown groups
examined  are  in  good  condition,  with  high
percentages of  the population in a nominal or
optimal classification.  There are some species
that do not appear in the following tables even
though they are on more than 50 plots because
we could not estimate the proportions in a given
class  with known confidence.  No species or
crown groups examined have high percentages of
the population in the subnominal classification. In
most  cases  the  entire  population  could be
TABLE 4-2. POPULATION PROPORTIONS FOR SPECIES IN THE EAST U.S. REGION FROM CDF
   ANALYSIS. Numbers represent the percentages of the population classified as poor, subnominal,
   nominal, and optimal. Numbers in  parenthesis  are the lower  and upper bounds of the CDF 90%
   confidence interval.
Species
Balsam Fir
Eastern White
Pine
Virginia Pine
Northern
White Cedar
Red Maple
Yellow Birch
Paper Birch
American
Beech
White Ash
Yellow Poplar
Black Cherry
White Oak
Scarlet Oak
Southern Red
Oak
Northern Red
Oak
Optimal
0.80 (0.68 0.92)
0.73 (0.60 0.86)
0.72 (0.55 0.89)
0.56(0.400.71)
0.75 (0.68 0.82)
0.87 (0.73 0.97)
0.54 (0.43 0.66)
0.73 (0.60 0.86)
0.63 (0.47 0.78)
0.92 (0.81 0.99)
0.66 (0.50 0.82)
0.83 (0.72 0.94)
0.65(0.500.81)
0.88 (0.73 0.98)
0.66 (0.53 0.79)
Nominal
0.99 (0.84 0.99)
0.99(0.91 1.00)
0.98 (0.77 0.99)
0.99 (0.89 1 .00)
0.98 (0.90 1 .00)
0.99 (0.87 1 .00)
0.98 (0.83 0.99)
0.96 (0.81 0.99)
0.96 (0.77 0.99)
0.97 (0.85 0.99)
0.94 (0.75 0.99)
0.99 (0.88 1 .00)
0.98(0.871.00)
0.99 (0.88 1 .00)
0.97 (0.82 0.99)
Subnominal
0.01 (0.01 0.16)
0.01 (0.00 0.09)
0.03 (0.01 0.23)
0.01 (0.000.11)
0.02(0.000.10)
0.01 (0.000.13)
0.02(0.01 0.17)
0.04(0.01 0.19)
0.04 (0.01 0.23)
0.03 (0.01 0.15)
0.06 (0.01 0.25)
0.01 (0.000.12)
0.02(0.000.13)
0.01 (0.000.12)
0.03(0.01 0.18)
Poor
*
*
*
*
0.00 (0.00 0.07)
0.01 (0.00 0.09)
0.01 (0.00 0.08)
0.00 (0.00 0.05)
*
*
*
*
*
*
0.01 (0.000.16)
 This threshold is outside the numerical range of data collected
                                            4-5

-------
TABLE 4-3. POPULATION PROPORTIONS FOR CROWN GROUPS IN THE EAST U.S. REGION FROM
   CDF ANALYSIS. Numbers represent the percentages of the population classified as poor, subnominal,
   nominal, and optimal.   Numbers in parenthesis are the lower and upper bounds of the CDF 90%
   confidence interval.
  Crown Group     Optimal            Nominal           Subnominal        Poor
  Pine-Short
  Needles          0.70 (0.54 0.87)     0.98 (0.78 0.99)    0.02 (0.01 0.22)

  Spruce-Fir        0.90 (0.78 0.98)     0.99 (0.86 0.99)    0.01 (0.01 0.14)

  Cedar-Juniper     0.71(0.590.84)     0.98(0.830.99)    0.02(0.010.17)

  Hardwoods-
  Closed Canopy
  Large Leaves     0.81 (0.75 0.87)     0.99 (0.92 1.00)    0.01 (0.00 0.08)

  Hardwoods-
  Open Canopy
  Large Leaves     0.92 (0.81 0.99)     0.97 (0.85 0.99)    0.03 (0.01 0.15)

  Hardwoods-
  Small Leaves     0.74(0.670.80)     0.98(0.911.00)    0.02(0.000.09)    0.00(0.000.03)

  Hardwoods-
  Compound
  Leaves           0.85(0.750.95)     0.99(0.881.00)    0.01(0.000.12)    0.00(0.000.04)

  Hardwoods-
  Oaks            0.80 (0.73 0.87)     0.99 (0.91 1.00)    0.01 (0.00 0.09)

  Miscellaneous     0.72 (0.64 0.80)     0.96 (0.87 0.99)    0.04 (0.01 0.13)    0.01 (0.00 0.10)
* This threshold is outside the numerical range of data collected

TABLE 4-4. POPULATION PROPORTIONS FOR SPECIES IN STANDARD FEDERAL REGIONS I AND
   II FROM CDF ANALYSIS. Numbers  represent the percentages of the population classified as poor,
   subnominal, nominal, and optimal.  Numbers in parenthesis are the lower and upper bounds of the CDF
   90% confidence interval.
Species
Eastern White
Pine
Balsam Fir
Red Maple
Paper Birch
American
Beech
White Ash
Yellow Birch
* This threshold
Optimal
0.68
0.80
0.64
0.54
0.60
0.65
0.87
(0.55
(0.68
(0.55
(0.43
(0.46
(0.48
(0.73
is outside the
0.81)
0.92)
0.74)
0.66)
0.74)
0.81)
0.97)
numerical
Nominal
0.99 (0.91
0.99 (0.84
0.97 (0.86
0.98 (0.83
0.94 (0.77
0.95 (0.76
0.99 (0.87
Subnominal
1.00)
0.99)
0.99)
0.99)
0.99)
0.99)
1.00)
0.01
0.01
0.03
0.02
0.06
0.05
0.01
(0.00 0
(0.01 0
(0.01 0
(0.01 0
(0.01 0
(0.01 0
(0.00 0
09)
16)
14)
17)
23)
24)
13)
Poor
*
*
0.01
0.01
0.00
*
0.01



(0.00
(0.00
(0.00

(0.00



0.10)
0.08)
0.06)

0.09)
range of data collected
                                           4-6

-------
TABLE 4-5. POPULATION PROPORTIONS FOR CROWN GROUPS IN STANDARD FEDERAL REGIONS
   I AND II FROM CDF ANALYSIS.  Numbers represent the percentages of the population classified as
   poor, subnominal, and optimal.  Numbers in parenthesis are the lower and upper bounds of the CDF
   90% confidence interval.
Crown Groups
Spruce-Fir
Cedar-Juniper
Hardwoods-
Closed Canopy
Large Leaves.
Hardwoods-
Small Leaves.
Miscellaneous
Optimal
0.90 (0.78 0.98)
0.56 (0.40 0.72)

0.71 (0.62 0.80)

0.60 (0.52 0.69)
0.61 (0.51 0.70)
Nominal
0.99 (0.86 0.99)
0.99 (0.89 1 .00)

0.98(0.88 1.00)

0.96 (0.86 0.99)
0.94 (0.83 0.99)
Subnominal
0.01 (0.01 0.14)
0.01 (0.000.11)

0.02 (0.00 0.12)

0.04 (0.01 0.14)
0.06(0.01 0.17)
Poor
*
*

*

0.00 (0.00 0.05)
0.01 (0.000.12)
* This threshold is outside the numerical range of data collected
TABLE 4-6. POPULATION PROPORTIONS FOR SPECIES IN STANDARD FEDERAL REGION III FROM
   CDF ANALYSIS. Numbers represent the percentages of the population classified as poor, subnominal,
   nominal, and optimal.  Numbers in parenthesis are the lower and upper bounds of the CDF 90%
   confidence interval.
 Species        Optimal           Nominal           Subnominal         Poor
 Red Maple      0.76(0.620.91)    0.99(0.871.00)     0.01(0.000.13)

 Yellow Poplar    0.92 (0.74 0.99)    0.97 (0.78 0.99)     0.03 (0.01 0.22)
* This threshold is outside the numerical range of data collected
                                         4-7

-------
TABLE 4-7. POPULATION PROPORTIONS FOR CROWN GROUPS IN STANDARD FEDERAL REGION
    III  FROM CDF ANALYSIS. Numbers represent the percentages of the population classified as poor,
    subnominal, nominal, and optimal. Numbers in parenthesis are the lower and upper bounds of the CDF
    90% confidence interval.
 Crown Groups     Optimal           Nominal           Subnominal        Poor
 Hardwoods-
 Oaks            0.62(0.500.74)    0.99(0.901.00)     0.01(0.000.10)

 Hardwoods-
 Closed Canopy
 Large Leaves     0.79(0.660.91)    0.99(0.921.00)     0.01(0.000.08)

 Hardwoods-
 Open Canopy
 Large Leaves     0.93(0.760.99)    0.98(0.790.99)     0.02(0.010.21)

 Hardwoods-
 Small Leaves     0.75(0.620.89)    1.00(0.971.00)     0.00(0.000.03)

 Hardwoods-
 Compound
 Leaves           0.72(0.570.86)    0.98(0.820.99)     0.02(0.010.18)    0.00(0.000.06)
 This threshold is outside the numerical range of data collected
TABLE 4-8. POPULATION PROPORTIONS FOR SPECIES IN STANDARD FEDERAL REGION IV FROM
   CDF ANALYSIS. Numbers represent the percentages of the population classified as poor, subnominal,
   nominal, and optimal.  Numbers in parenthesis are the lower and upper bounds of the CDF 90%
   confidential interval.
Species
Yellow Poplar
White Oak
Southern Red
Oak
Optimal
0.93 (0.77 0.99)
0.96 (0.80 0.99)
0.95 (0.75 0.99)
Nominal
0.98 (0.81 0.99)
0.99 (0.91 1 .00)
0.98 (0.86 1 .00)
Subnominal
0.02 (0.01 0.19)
0.01 (0.00 0.09)
0.02 (0.00 0.14)
Poor
*
*
*
* This threshold is outside the numerical range of data collected
                                          4-8

-------
TABLE 4-9. POPULATION PROPORTIONS FOR CROWN GROUPS IN STANDARD FEDERAL REGION
    IV FROM CDF ANALYSIS. Numbers represent the percentages of the  population classified as poor,
    subnominal, nominal, and optimal.  Numbers in parenthesis are the lower and upper bounds of the CDF
    90% confidence interval.
Crown Groups
Hardwoods-Oaks
Miscellaneous
Hardwoods-
Open Canopy
Large Leaves
Hardwoods-
Small Leaves
Optimal
0.92
0.95
0.92
0.87
(0.83
(0.76
(0.77
(0.77
0.99)
0.99)
0.99)
0.96)
Nominal
1.00
1.00
0.98
0.99
(0.94
(0.91
(0.82
(0.87
1.00)
1.00)
0.99)
1.00)
Subnominal
0.00
0.00
0.02
0.01
(0.00
(0.00
(0.01
(0.00
0
0
0
0
06)
09)
18)
13)
Poor
*
*
*
*
* This threshold is outside the numerical range of data collected
characterized as  nominal  and optimal.  In  most
cases where portions of the population fell into a
subnominal category, it represented less than three
percent of the population. It is not possible to infer
causality for any crown condition status observed in
Detection Monitoring.
Discussion

    The  importance  of  setting  realistic  CDF
thresholds should be apparent from the approach
presented in this  section;  the  determination of
nominal and subnominal proportions is dependent
on the accuracy of the CDF thresholds.  We have
set thresholds on the best information available to
us  today.   However, if the nominal-subnominal
thresholds for defoliation are too low, then a larger
percentage  of  the  population  will  be   in  a
subnominal condition. If they are too high, a larger
percentage of the population will be in a nominal
condition.  Since we found most of the population
for  species and crown groups to be nominal, our
interest is in whether or not the thresholds are too
high (i.e., everything  looks good). To address this
possibility, we should then give greater weight to
the even smaller percentages of the population that
are in a  subnominal condition ( i.e.,  they are
subnominal  even  when  the  threshold is very
lenient).  If future  research information indicates
that the thresholds should be more robust, then
those proportions of the population that are  now
subnominal will  become poor.

    Monitoring  techniques research on-frame (on
the Detection  Monitoring  plots), off-frame,  and
computer research are needed to set consistent
and tailored threshold values for each species,
forest type, and/or crown group.  At a  minimum,
one to three representative species in each crown
group should be intensely evaluated to precisely
determine threshold values for both growth and/or
survivability (primarily) and aesthetics (secondarily).
Future analyses  should  have  the  flexibility to
evaluate tree crown condition both on the basis of
visual appearance and on growth.  We should have
threshold values  specific for both  growth  and
aesthetics,  however  growth is considered more
important   because  the   productivity   and
sustainability  of species and forest types  are
dependent on growth and not on human evaluation
of desired appearance.
                                             4-9

-------
    Standard  Federal  Regions (SFR)  Boundaries
                                        SFR l&li
                                                SFR III
Figure 4-2. Standard Federal Regions used for the analyses.
                              4-10

-------
5.  Characterization of Regional
Overstory   Tree    Species
Diversity on FHM Plots


S. Cline

Background

       Conservation  and  maintenance  of
animal and  plant  species  diversity  is an
important public  value.  Physical alteration  or
destruction of habitats is a preeminent threat to
the maintenance of  animal and  plant species
diversity (EPA 1990). Forest Health Monitoring
(FHM) plans to monitor the status and trend in
condition of vegetation and habitat structure.
Through this approach, vegetation diversity can
be assessed directly, and animal diversity can
be assessed indirectly through its relationship
with vegetation and habitat structure.

       Depending upon the specific objective
and  scale  of interest, many vegetation  or
habitat indicators could be measured (Figure
5-1).  A complete set of measurements would
permit  assessing   the   heterogeneity  and
complexity  of vegetation and habitat structure,
but the limited set of measurements that were
made in  1991 do  support some preliminary
analyses.   The objective of this chapter  is  to
present  quantitative  baseline  estimates  of
overstory tree species diversity as indicators of
forest condition  for specific  forest types  in
twelve states in  the northeast and  southeast
United States.
Data Set Characteristics

       Only data for trees >12.7 cm diameter
breast  height  (DBH) measured on forested
subplots in 1991 were used to calculate tree
species diversity (see Chapter 2).  As  such,
each plot data set represents a segment of the
whole tree community composed of similarly
(large)  sized, mature trees that interact in the
forest overstory, in contrast to tree sapling and
seedling  communities,  which are smaller  in
stature, and  interact more directly  in lower,
environmentally  distinct,  strata.   With this
analytical   approach,   diversity   measures
indicate the community structure of organisms
similar  in  habitat  or  microhabrtat,  size,  life
history traits,  and resource utilization (Hurlbert
1971).

        A total  of  925  sample  plots were
separated  into  two   regional  data  sets:
Northeast (NE) and Southeast (SE).  The NE
data  set  represents  forests  of  northern
glaciated  ecoregions (Omernik  1987).   It is
made up of the 202 forested (out of 245 total)
plots  in Maine,  New  Hampshire,  Vermont,
Rhode Island, Connecticut and Massachusetts.
These plots were dominated by the northern
hardwoods forest type and the spruce-fir forest
type groups.   The  most  frequently  recorded
tree species  on NE sample  plots were  red
maple, balsam fir, and red spruce.

        The SE data set represents forests of
southern,  humid  ecoregions (Omernik 1987).
It includes the 420 forested (out of 680 total)
plots from New  Jersey, Delaware, Maryland,
Virginia, Georgia, and Alabama. These plots
were dominated by the oak forest type and the
pine forest type groups. The most frequently
recorded tree species on SE sample plots were
loblolly pine, sweet gum, and red maple.
Index Selection and Calculation

       Hill's   (1973)  series   of   diversity
measures was selected. The series consists of
diversity  measures based on the number of
species present (S), the exponent of Shannon's
index (eHl, Shannon and Wiener 1949), and the
reciprocal of Simpson's index (1/D,  Simpson
1949),   and  represents   various   symbolic
measures of the number of species present
based on different levels of analysis of their
rarities (Hill  1973,  Table  5-1).  The highest
diversity  values  are  measured  using  S,
because  it includes species presence without
regard  to rarity.  The lowest diversity values
are derived from 1/D because it weighs the
abundant species  more  heavily  than rarer
                                         5-1

-------
T3"c3
o>
CO

u.
          I     /     tXl
O

'•5
.9

1
^
<= at
1 =
g 0

pi  si  mi  I?  fi-e

III  33  III!  II  111

         |
                                                I
CO


CO
CO
CD

g
ca

15
.9
o>
o

8
CO
 O



 .C


 15
 
-------
 
I

O
0.
o
c
_g
ea
eo
a
ea
ea
<






A
3
£







Definition
en
c
V
E
u
W

o
c
o
•§
1
«
V)

'o
Si
en
§
number of
quadrats
E t
s!


en
u 
1
M
U.
O
5 «
£ H
3 ca
c c
O CQ
P.£



M
(rt
1
o
(2

1
•s
X-*.
J
S
S M
sl
8-i
2 "
•"* fc«
ji
« §
* en
§u
'3
45 S
-C vi
trt
o
1
ea
(«M
0
8
i
.a
C9
O
.>
ea
"S g:
11?
c£.i:

1
§
VI
O
o
c
*«
*• i
o
S
ea
"3
en
5
•a
u
'eb
|f
11-
2 5













The distribution of abundance
among the species in a
community



en

C
C
S
u







>^
11
8*8
b. 3
en 0-5
s « 1*
^u J5 '-'•B.
w E 2 E
O 3 "»
eo — « H
z = 8.1
1 -=Mfl
". u § i
a, •• E «>
o a g g
in O 5 §
3 2 8 2


c
* 5^
§ 5
I i
x 53 x g-
t> _g-. o E
"2 S>^ "5 ^
— — . ov £ >-^
e Q* "~* f- P* —
§ w u § a
s ' g a w
i » ~ | i,
55 £ > oo O
.£ en
Synthetic measures Ihat are
sensitive to both richness and
evenness. Related to the
uncertainty of identity of an
individual randomly selected
an area. Uncertainty increase
with richness and evenness.



>>

e
y
>
a
                                5-3

-------
species.   Intermediate  diversity  values are
obtained using eH>  because  it weighs  rarer
species more heavily than 1/D.

       Overstory tree species diversity values
based upon S, eH>, and 1/D were calculated for
each plot. Species richness (S) was expressed
as species density, the number of tree species
>12.7 cm DBH per 672 m2 (the total area of a
plot).   The  Shannon (H') and Simpson (D)
indices  were  calculated  from  the relative
abundance (p1) of each tree species >12.7 cm
DBH on the plot. In this analysis, the pi values
were the weighted mean basal area in m2/ha.
First the basal area of each tree >12.7 cm DBH
was calculated in m2, which was then multiplied
by an expansion factor (ca 59.48) to estimate
basal area in m2/ha.  Next, basal area of each
tree was summed by species on each subplot
and multiplied  by a weighting factor based on
the percent subplot  area forested.  Then the
weighted basal areas of each tree species
were summed  across subplots, and divided by
the total  percent area forested. Finally,  each
species-specific mean weighted basal area was
divided by the mean weighted basal area of all
trees on the plot.  When trees  >12.7 cm DBH
were absent on the plot (i.e.,  basal area was
zero), S, eH>, and 1/D were all set to zero.

       Both Hl and D are sensitive to changes
in species richness (S) and to the evenness of
the distribution of p1 values among species.  H'
increases with  species  richness from  a
minimum value of 0 when only one species  is
present.  For a given number of species, the
maximum value of H>  is obtained  when all
species  are equally abundant;  H'max  equals
ln(S). The value eH>  is the number of equally
abundant species equivalent to the sample
diversity H>. In contrast, D increases as species
richness decreases.  The maximum value of  D
= 1.0 is obtained when one species is present.
Similarly,  the  value  1/D  is the  number of
equally abundant  species in a sample with  a
dominance of D, and is comparable to eH>.  In
comparison to H', D is less sensitive to changes
in the abundance  of  rare species  of the
community.
Analysis

       The baseline status of overstory tree
species diversity in NE and SE regions was
expressed   using   cumulative   distribution
functions (CDF; see Chapter 2 and Appendix
A). A CDF characterized the regional variation
in overstory tree species diversity in 1991.  In
each  region a  CDF was constructed  based
upon the diversity values using S, eH>, and 1/D.
The CDFs were analyzed for intra- and inter-
regional patterns.

       While a statistical summary such as
this cannot be used to infer cause and effect
relationships, associations among variables can
be explored.   For example,  what are the
predominant  characteristics  of  the  forest
population with relatively  high overstory tree
species diversity  values?  How do  these
features contrast with  conditions associated
with low diversity values?  Consequently, an
exploratory analysis of the association of eH>
diversity values and forest type group, stand
origin, stand  size  class, disturbance history,
and condition code was  conducted in each
region (see Chapter 2).  First the range of eH'
diversity values was divided into five categories
(range in  eH> values):  none (0), very low  (>
0-1), low (>  1-3), medium (> 3-5), high (> 5).
Then  for  each variable,  the  proportion  of
sample plots within each  diversity and forest
condition  category were  determined.    The
trends  based   upon   eH'  diversity   as
representative of trends based upon S and 1/D
diversity values.
Results and Discussion

Intra-Regional Patterns

        Southeast  Region-Comparison   of
Diversity Indices

        The SE CDFs for S, eHl, and 1/D had
similar shapes, but the cumulative probabilities
at   any  given   diversity   value  differed
systematically as S < eHl < 1/D (Figures 5-2
(b),  5-3 (b),  5-4  (b)).   For example,  when
                                              5-4

-------
           (a)
              1.0 T
           (b)
                1  T
             0.8 -
      F(x)   0.6-•
             0.4 -
             0.2  -
                  0
                                      4         6         8        10

                                    Regional Overstory Tree Diversity
12
—I	1	1	I—
   4         6         8         10

 Regional Overstory Tree Diversity
12
14
14
Figure 5-2.     Cumulative distribution function of (a)  regional overstory tree diversity in the NE using
               species richness, and (b) regional overstory tree diversity in the SE using species richness.
                                               5-5

-------
            (a)
                 0.0
            (b)
                 0.0
2.0           4.0          6.0          8.0

     Regional Overstory Tree Diversity
2.0           4.0           6.0          8.0


     Regional Overstory Tree Diversity
10.0
10.0
Figure 5-3.     Cumulative distribution function of (a) regional overstory tree diversity in the NE using
               Simpson's Index, and (b) regional overstory tree diversity in the SE using Simpson's Index.

                                               5-6

-------
         (a)
            1.0 T
               0.0        2.0        4.0        6.0        8.0       10.0

                                  Regional Overstory Tree Diversity
         (b)
            1.00 T
            0.80-
     F(x)    0-60 -
            0.40 -
            0.20 -
            0.00
                        -t-
                                                                   12.0
                                                                                 ^
Figure 5-4.
                0.00      2.00      4.00       6.00       8.00      10.00      12.00


                                  Regional Overstory Tree Diversity
Cumulative distribution function of (a) regional overstory tree diversity in the NE using the
Shannon-Weiner method, and (b) regional overstory tree diversity in the SE using the
Shannon-Weiner method.
                                              5-7

-------
overstory tree species diversity equalled 2, the
cumulative  probabilities  (± 90%  confidence
limit) were 0.26 (0.05), 0.33 (0.05), and 0.41
(0.05) based upon S, eH>, and 1/D, respectively.
Diversity values of eH>  equalling 1.6, 2.9, and
4.6 defined the respective 0.25 (0.04), 0.50
(0.05), and 0.75 (0.05)  cumulative probabilities
of the SE forest population.

        Differences   in   the   cumulative
probabilities  were  expected   because  of
systematic differences in the calculation of Hill's
(1973)  series of diversity indices, as  noted
earlier.   Although these  differences among
diversity indices were not statistically significant
and tended to decrease  as diversity values
increased, the tendency for diversity values to
be  index-dependent suggests  that  only the
same index should  be compared to assess
community changes or environmental  effects.
        Southeast Region-Forest Conditions
Associated with High and Low Overstory Tree
Species Diversity
       Qualitative  results of an  exploratory
analysis  of the association of eH> diversity
values and forest condition variables in the SE
region follow.

As overstory tree species diversity increased:

• The predominant forest type group changed
  from pine to oak.  Both pine and oak types
  had eHl values in all diversity categories, but
  the  proportion  of  the pine type decreased
  while that  of the oak increased with higher
  diversity values.

• The predominant  stand  origin condition
  changed from planted softwoods to naturally
  regenerated.    The   planted   softwoods
  condition dominated  the   "none" overstory
  diversity  category,  but  the  proportion of
  naturally  regenerated  and  planted  stand
  origin were approximately equal in the "very
  low" diversity category. However, plots with
  the planted softwoods conditbn were absent
  from the "high" diversity category.
  The predominant stand size class changed
  from  seedling/sapling  to  sawtimber.  One
  might  expect all seedling/sapling to exhibit
  low or no tree  diversity because fewer (or
  no)  individual  trees > 12.7 cm  DBH   are
  present compared  to  the sawtimber  size
  class; however, results showed that both size
  classes  had  eH'  values  in  all diversity
  categories.  Instead, this trend appeared to
  be  more closely associated with changing
  forest  conditions  within  each  size class.
  Specifically, within the seedling/sapling size
  class,  a planted softwood origin/pine forest
  type group condition predominated at no and
  low diversity values, but the proportion of this
  stand condition type decreased as stand size
  increased.

  The proportion of plots with a  disturbance
  history condition of "none" increased while
  the proportion with disturbance history coded
  as  "cutting"  decreased.  However only the
  proportion  of plots  in  different diversity
  categories   changed    with   disturbance
  conditions,    since   disturbance  condition
  "none" and "cutting" both had eH' values in
  all diversity categories.

  The proportion of plots with >  1 condition
  code  increased  while the proportion of  plots
  with 1 condition code decreased. As noted
  in  Chapter 2, condition code is a composite
  of  land  use, forest  type, size class, stand
  origin, and  disturbance  history; condition
  code  on a plot changes if any one of these
  variables changes.  Multiple condition codes
  on a plot  indicates multiple habitats and  thus
  increased  diversity  was expected as the
  number  of conditions on a plot increased.
  However, even though the proportion of  plots
  with  1   condition  decreased as overstory
  diversity increased, it  was the predominant
  condition code in all diversity categories.
       Northeast   Region-Comparison   of
Diversity Indices

       The NE CDFs for S, eH', and 1/D also
had  similar shapes, and as in the SE,  the
cumulative probabilities at any given diversity
value differed systematically as S < eH' < 1/D
                                               5-8

-------
(Figures 5-2 (a), 5-3 (a), 5-4 (a)). For example,
when overstory tree species diversity equalled
2,  the  cumulative   probabilities   (±  90%
confidence limit) were 0.17 (0.07), 0.20 (0.08),
and 0.32  (0.08) based upon S, eH>, and 1/D,
respectively.  Diversity values of eH> equalling
2.2, 3.6, and 4.6 defined the respective 0.25
(0.08), 0.50 (0.08), and 0.75 (0.08) cumulative
probabilities of the NE forest population.
        Northeast Region-Forest Conditions
Associated with High and Low Overstory Tree
Species Diversity
        Qualitative  results of an  exploratory
analysis of the  association of eH> diversity
values and forest condition variables in the NE
region follow.

As overstory tree species diversity increased:

• The proportion of plots in the spruce-fir forest
  type group increased while the proportion in
  the northern hardwoods and pine forest type
  group decreased. Only the pine type had
  eH' values in  all diversity categories, and
  although the  proportion of the spruce-fir type
  increased  with  higher diversity categories,
  the   northern   hardwoods    type  was
  predominant  in  all  diversity categories but
  "none".

• There was  no  trend  in the stand origin
  condition.  Results  indicated that nearly all
  stands   encountered    were    naturally
  regenerated,   with  the  exception of a few
  planted softwoods stands.

• The predominant stand size class changed
  from  seedling/sapling   to  sawtimber. This
  trend  was  not as  clear here as in the SE
  because  in  the  NE, few seedling/sapling
  stands were  sampled  in comparison to the
  sawtimber class.  Similar to the SE, though,
  both size classes were  present in a range of
  diversity   categories.  In contrast to the SE,
  this  trend in  the  NE could not be closely
  associated with  specific changing  forest
  conditions within each size class.
• There was no trend in the disturbance history
  condition.   The  eH> diversity   values  of
  disturbance conditions  "none"  and "cutting"
  were  both distributed in about the same
  proportion  across  all  diversity categories.
  This contrasted with  the SE result, where the
  proportion  of  plots in  different  diversity
  categories   changed   with  disturbance
  conditions.

• The proportion of plots with > 1  condition
  code increased while the proportion of plots
  with 1 condition code decreased, matching
  the trend found  in  the SE.  Similarly, even
  though  the proportion  of  plots  with  1
  condition decreased as overstory  diversity
  increased, it was the predominant condition
  code in  all diversity categories  in theNE.
  Plots with > 1  condition  code were absent
  from  the  none   and  very   low  diversity
  categories.
Inter-Regional Patterns

Regional Differences

        The CDFs of the regions were most
clearly distinguishable at low diversity values
(e.g. < 3).  However, even here the regional
CDFs were not statistically different. All three
diversity indices  showed that  the SE forest
population  had  more  low  overstory  tree
diversity values compared to the  NE.   This
difference  was associated more closely with
the predominance of a particular combination
of forest conditions (i.e., a higher proportion of
the pine forest type group by planted softwood
stand origin in the SE than in the NE) than with
regional differences in the relative proportions
of the stand size classes.

Index Sensitivity

       The diversity value S  was the most
sensitive indicator of differences in overstory
tree species diversity between regions because
the CDFs based on  S were  more distinct
between regions over a wider range of diversity
values than the CDFs based on either eH> or
1/D. While the CDFs for S, eHl, and 1/D all
                                           5-9

-------
showed similar differences between the NE and
SE at  low diversity values,  only S showed a
difference beyond diversity  values  of 6.  For
example, at a diversity of 2, the differences in
cumulative probabilities between regions were
0.08, 0.13, and 0.09 based on S, eH>, and 1/D,
respectively. In contrast,  at a diversity of 6, the
differences in cumulative  probabilities between
regions were 0.07, 0.01, and 0 based on S, eHr,
and 1/D, respectively.
                                               5-10

-------
6.  Selected Climatic Data
    Summaries


E. J. Cooler and L Truppi

Introduction

       This  chapter  summarizes  selected
climate  conditions  and events that, through
experience, experimentation or research, are
known to impact forest ecosystem status and
health.   Data  have  been assembled  and
processed from a variety of sources listed at
the end  of this chapter. Every effort has been
made  to   ensure   information  quality  and
consistency, but it must be recognized that this
information represents regional summaries and,
in many cases,  does not reflect conditions at
particular places within the region.

Synopsis

       October, 1990 through  September,
1991 was marked by severe tropical storms, a
record warm winter with far  below normal
snowfall, early springtime warmth, a severe
spring ice storm, a  long, hot summer,  and a
visit from Hurricane Bob.

       In  October, 1990,  the  remnants of
tropical storms Marco and Klaus were among
several storm systems that brought excessive
rainfall to the eastern  United States.  States
from Pennsylvania  to Georgia were affected.
The storms produced heavy  rain  and high
winds in Maryland, Virginia, and North Carolina.
November and December were generally warm
and quiet.

       January 1991 was the  second wettest
in 96 years of record in the Southeast. Florida,
Louisiana, Alabama and  North  Carolina were
particularly hard hit.

       Nationwide,  February was the second
driest and third warmest February since 1895.
Only 1947 was drier and only 1932 and 1954
were warmer. Particularly warm temperatures
were reported February 1-10.
       By March,  it was being said that the
Middle Atlantic  states  and Southern  New
England had experienced an almost winterless
1990-1991.  Boston reported the warmest six-
month winter (October-March) in the last 121
years.    In  sharp contrast,  an ice storm,
described as  the  worst in  30 years,  was
reported in New York.

       The fifth  warmest April  in the last 96
years was reported in the Southeast. It was
the  sixth  warmest April in the Northeast.
Record numbers  of  April  tornados were
reported  in New York (4) and Pennsylvania
(10).

       Connecticut,  New  Hampshire,  New
Jersey and  Rhode Island  experienced  their
warmest   Mays  ever,   while    Florida,
Massachusetts,  New  York,  Pennsylvania,
Vermont and Virginia experienced their second
warmest Mays.

       June and July were wet but uneventful.
August witnessed the passage of Hurricane
Bob.   Although  the path of Hurricane  Bob
impacted  Massachusetts, Rhode Island and
northeastern  Maine  most  directly,  severe
weather associated with the accompanying
trough of  low pressure  was responsible for
more wide spread damage.  A new record for
August tornados was  set in North Carolina
(10).

Precipitation

       Climate division precipitation  data for
October,   1990  through  September,  1991
indicate most of the eastern seaboard  and
southeast  received well above average rainfall
(Figure  6-1).    The greatest  excess  was
reported in Alabama, where the 1991  total was
60 cm more than  the 30-year division average.
The exception to these wet conditions was the
Mid-Atlantic states.   Here, precipitation totals
as much  as  24  cm  below  30-year division
averages  were recorded.  Figure 6-2 gives
these  precipitation  values  an   historical
perspective.  The 1991 conditions ranged from
the wettest 10% of each division's  30-year
record  (northwestern   Alabama,  northern
                                         6-1

-------
       Precipitation  Deviation
       (in  centimeters)
Figure 6-1.  Deviation of annual climate division precipitation (October 1990 - September 1991)
           from 1961-1990 average conditions. Positive values represent above-average rainfall
           and negative values represent below-average rainfall.

                                       6-2

-------
         Cumulative  Frequency  1961-1990
         fflH Greater  Than  90%
         f§§§§§§§§§| 90%  To  75%
         BBSS 75%  To  50%
         WMh 50%  To  25%
              25%  To  10%
              Less  Than  10%
Figure 6-2.  Probability of a climate division receiving less precipitation than the October 1990 -
          September 1991 total.  Probabilities are based on 1961-1990 climatological records.
                                    6-3

-------
Florida, central and  eastern Georgia, South
Carolina,  Maine,  and  portions  of  New
Hampshire and Massachusetts) to  the driest
10% (southern Maryland). The excessively wet
areas are largely the result of decaying tropical
storms in October, 1990 and Hurricane Bob in
August, 1991.  The  dry  area  lies  along the
eastern edge of a  severe drought region
extending along the upper Ohio Valley.

Hurricanes

       Figure 6-3 highlights counties that have
experienced hurricane passages since October,
1984.  The most numerous reports come from
Eastern Texas (Cass County)  and Northern
Louisiana (Caldwell County). Figure 6-3 was
constructed by counting the number  of times a
plotted  hurricane path  passed  through  or
touched a county.  This explains the  narrow,
well defined patterns of most impact zones and
generates a conservative estimate of stressed
area. There have been 14 hurricanes during
this  period whose paths have crossed the
assessment  area.   Only those  storms with
reported winds in excess of  119 km/h (class 1
hurricane) have been plotted.  Two extreme
hurricane  events  were  reported during  this
period.  The first, Hurricane Hugo,  passed
through North  and  South Carolina during
September, 1989. The second, Hurricane Bob,
occurred in August of 1991.  Both storms were
class 3 hurricanes, containing winds in excess
of 178 km/h.

       Figure 6-4 shows hurricane contact for
October, 1990 through September 1991.  Only
one storm, Hurricane Bob was reported to pass
through the assessment area. In August, 1991
Hurricane Bob and its accompanying trough of
low pressure brought destruction to  eastern
New England and excessive rains to  most of
the Northeast. The hurricane center passed 25
km east of Cape Hatteras with winds of 181
km/h.  At these wind speeds, foliage can be
torn  from trees and large trees blown down.
The  eye passed east of Delaware Bay and
moved northeast, missing  Long Island.   It
passed over Block Island and made landfall on
the Rhode Island shore,  near Narragansett
Bay. The storm moved east of Providence and
Boston and later crossed northeastern Maine.
Excessive rainfall accompanied the western
periphery of the rather  narrow northeastward
track. Up to 229 mm of rain were reported in
Maine.   Massachusetts  and  Rhode  Island
reported the greatest amount of wind damage.
Historically,  Bob  resembled the  September
"Gale"  of  1869  and the "Enigma" storm  of
August,  1788  which impacted  New Jersey,
central  Hudson Valley and   western  New
England.

High Wind Events

       Figure 6-5 contains the frequency  of
report by county of winds in excess of 70 km/h.
All severe weather analysis must be interpreted
with care. To be entered into the data base, all
events must be verified. This  means that the
number  of reports  is  highly  dependent  on
population density.  For instance, most of the
counties reporting the  greatest frequency  of
high  wind events  also coincide with  major
metropolitan areas, e.g., Dallas-Ft  Worth, TX,
Shreveport,   LA,   Birmingham,  AL,   and
Pittsburgh,   PA.    Still,  general   areas   of
widespread activity  can be determined from
these data.  Key areas during this five year
period are: 1) the Texas, Louisiana, Arkansas
boarder area; 2) most of Alabama; 3) most  of
South Carolina and eastern North Carolina; 4)
the southwestern edge of the Florida peninsula;
and   5)   Pennsylvania,   New  York   and
Massachusetts.

       Figure 6-6  shows  areas  of  activity
during the 1991  sampling year.  The Florida
area is  associated  with a decaying tropical
storm in April. Sustained winds of 128 km/h for
a full hour were reported, with  gusts of 169
km/h reported in Tampa. Many considered this
event to be worse than  Marco  (October 1990)
and  in some  places,  worse  than  Hurricane
Elena (September 1985). While much of the
Northeast sustained high winds associated with
the passage of Hurricane Bob, those that are
highlighted  are  largely  the result of  winds
associated with severe ice and snow storms
during March, 1991.
                                             6-4

-------
                                                                                                              o
                                                                                                              I—



                                                                                                              1


                                                                                                              CO

                                                                                                              Q.
                                                                                                              CD
                                                                                                              03




                                                                                                              3
                                                                                                              o
                                                                                                              1
                                                                                                              §
                                                                                                              CD
                                                                                                              o

                                                                                                              I*

                                                                                                              (D

                                                                                                              &
                                                                                                              s

                                                                                                              _>,


                                                                                                              I
                                                                                                              o


                                                                                                              9
                                                                                                              CO
                                                                                                              CD
 =    \  \m •  I
ic    I—I  Wa •  •
                                                    6-5

-------
                                                                                                       Ol
                                                                                                       O5
                                                                                                       0>

                                                                                                       Q.
                                                                                                       03

                                                                                                       CO
                                                                                                       (0
                                                                                                      .a
                                                                                                       o
                                                                                                       8

                                                                                                       §

                                                                                                       o
                                                                                                      co

                                                                                                      O)
I
                                          6-6

-------
OJ
en
               <3 <=>  1=9
               o  o  o
                                                                                       CO
                                                                                       (O
                                                                                       (0

                                                                                       I
                                                                                       OT
                                                                                       T3
                                                                                       
-------
                                             CO
                                             CO
                                             CD


                                             s
                                             CD
                                             CO

                                             •§
                                             CD
                                             a.
                                             CO


                                             I

                                             5


                                             £
                                             c

                                             'co
                                              CD  .


                                              CD 
-------
Ice Storms

        No geographically  detailed, digitized
record of ice and freezing rain events currently
exists.  A county-scale data base has been
created for this assessment.  Published NOAA
records describing the  events were reviewed.
Minor   events  (identified   by   reports   of
occasionally slick roads) were not included in
the digitized  set.  National  Weather Service
forecast zones were used when specific county
lists were not provided.  Rgure 6-7 highlights
active  areas  between October, 1984 and
September, 1991. The largest number of ice or
freezing rain events are reported  in New York,
Vermont,  New  Hampshire  and  Maine.   A
second active area is the western mountains of
South Carolina. As with winds, it appears that
these reports are influenced by population and
level of vehicular traffic, so  Figure 6-7 should
be taken as an indication of icing regions rather
than  an  exhaustive   list  of  point  events.
Particularly severe  icing events  are listed in
Table 6-1.   Figure 6-8 shows  those  areas
reporting  icing conditions  during  the  1991
season.  The most frequently impacted areas
lie in New York, New Hampshire, Vermont, and
Maine.   New York state, in particular, was
impacted by a severe ice storm in March, 1991
which was described as the worst in 30 years.
                 Nineteen counties in western and northern New
                 York were declared disaster areas.

                 Last Hard Spring Freeze (-2.2? C)

                        The   Middle  Atlantic  states   and
                 southern New England experienced an almost
                 winterless 1990-1991  season.  This record
                 warmth is reflected in the extremely early dates
                 of last hard freeze shown in Figure 6-9. Areas
                 reporting their last hard freeze for Winter,  1991
                 in January, included Mississippi, New York and
                 Pennsylvania. Although last spring frost dates
                 as much as 17 weeks early were reported (as
                 compared to  median  date),  most locations
                 recorded last freeze dates two to three weeks
                 ahead of median dates.  This, combined with
                 the severe March ice storm discussed earlier,
                 could  have resulted in significant damage to
                 first  flush foliage.  Florida is the only state in
                 the  study region to report spring frost dates
                 later than the median date. Temperatures at or
                 below -2.2°C  occurred  two to seven weeks
                 late.

                 Data and Information Sources

                 Synopsis:
                 Weatherwise, American Meteorological Society,
                 Boston, MA
                Table 6-1.  Particularly severe icing events in recent history.
               Date
                       Place
           December, 1986
           January, 1987
           February, 1987
           March, 1987
           January, 1988
           February, 1989
           March, 1991
North Carolina
Pennsylvania
North Carolina and South Carolina
New York
Alabama, Georgia, North Carolina and South Carolina
Georgia
New York
                                          6-9

-------
     I ce  Frequency ,  1984-1991
     CZJ  0


           6  To 10
          11  To 15
          Greoter  Than  16
Figure 6-7.  County frequency of ice storm events, October 1984 - September 1991.

                                  6-10

-------
    Ice  Frequency,  1990-1991
          0



         Greater  Than  16
Figure 6-8.  County frequency of ice storm events, October 1990 - September 1991.
                                   6-11

-------
 Dev i o t i on
 (Number  of  Days)
       Earlier  than  60
       -60  To  -45
       -45  To  -30
       -30  To  -15
       -15  To    0
          0  To   15
        15  To   30
       Later  than  30
Figure 6-9. Deviation (days) of 1991  date of last hard spring freeze  (24-hour  minimum
          temperature equal to or less than -2.2° C) from 30-year median conditions. A positive
          value indicates a late spring freeze and a negative value indicates an early spring
          freeze.
                                     6-12

-------
Precipitation:
Time biased corrected divisional temperature-
precipitation-drought Index, TD9640, National
Climatic Data Center, Asheville, NC.

Hurricanes:
Weatherwise (1991 hurricane tracks and storm
description).

Jarinen, B.R., C.J. Neuman and MAS. Davis,
1984.  "A tropical cyclone Data Tape for the
North Atlantic Basin, 1886 - 1983:  Contents,
Limitations  and Uses."   NOAA  Technical
Memorandum NWS NHC 22 and data tape
updated through 1990, National Climatic Data
Center, Asheville, NC.

High Wind Events:
Mr.  Preston   Leftwich,  National   Weather
Service,   National  Severe  Storm   Forecast
Center, Kansas City, MO.

Ice Storms:
Storm  Data, National Climatic Data Center,
Asheville, NC.

Last Hard Freeze:
National   Weather  Service,  Cooperative
Observer Network, Tape TD3200, summary of
day, National Climatic Data Center, Asheville,
NC.
                                         6-13

-------
 7.     Status  of  Major  Forest
 Insects  and Diseases  in   the
 Eastern  United States,  1991

 R.    L.    Anderson,    W.    G.
 Burkman, and  W. H. Hoffard


 Introduction

       This is a summary of significant pest
 activity observed in the eastern  United States
 in 1991. The information is organized by major
 forest type group and by pest within forest type
 group.  Detailed  information is provided for
 hemlock  woolly  adelgid,  hemlock  loopers,
 eastern  spruce  budworm,  fusiform   rust,
 southern  pine beetle,  littleleaf disease, gypsy
 moth, oak decline, beech bark disease, and
 dogwood anthracnose.  Brief summaries are
 given for other pests.  The information was
 compiled from state forestry and agriculture
 agency reports, from Forest Service research,
 inventory, and Forest  Pest Management data
 bases such  as  the  Southern Pine Beetle
 Information System and Forest  Inventory and
 Analysis, and from Forest Hearth  Protection
 information.

 New Problems of Special Interest

       A new disease of blackgum (cause
 unknown) has been found in the Appalachian
 mountains,  damaging  trees in  a  three-state
 area. Thousands of cabbage palms have been
 found dead (cause unknown) along the Florida
 Gulf coast. The black twig borer, an introduced
 ambrosia borer, is on the increase, affecting a
 large number of tree species. Hurricane Bob
struck New England and caused widespread
devastation.
Eastern    White-Red-Jack
(includes Eastern Hemlock)
Pine
       The redheaded jackpine sawfly caused
2,000  ha of defoliation  in  Wisconsin and
resulting mortality is expected to be heavy in
1992.

       The jack pine budworm  defoliated
approximately 20,000 ha in the upper peninsula
of Michigan and 190,000 ha  in  the  lower
peninsula.    Populations  in  Michigan are
expected to stabilize or decrease. There were
200  ha defoliated  in  Pine,  Beltrami,  and
Hubbard counties in Minnesota.

       The jack pine sawfly defoliated  2,800
ha in northern Minnesota.

       Red pine adelgid  populations  were
unchanged   in   1991    in   Connecticut,
Massachusetts,  New York, and Rhode Island.

       The red  pine  scale  has  shown
increased activity after several years of minimal
activity, and threatens to damage 400 ha of red
pine in southeastern Connecticut. Populations
are reported to be stable in southeastern New
York, due  to reduced numbers  of host trees.
Heavy defoliation and  mortality continued  in
southern  Rhode Island and is expected to
spread throughout the range of  red pine within
the state.

       Red pine beetles caused mortality  in
scattered pockets on 175 ha in Baraga County,
Michigan.

       The white pine weevil affects eastern
white pine, jack pine, red spruce,  and  white
spruce. Statewide damage continued to occur
in Maine, with 170,000 ha affected. Levels are
expected to remain high but stable.  No heavy
damage was  observed  in  Vermont, where
levels were similar to those in 1990.

       The elongated  hemlock scale  was
found  in association with the hemlock woolly
adelgid in  Connecticut.   There were severe
infestations  in   southeastern   New   York,
increasing  in some stands and dying out  in
areas  where   hemlock  woolly  adelgid
populations were high.

       Hemlock looper refers to two species,
Lambdina  athasaria and  L.  fiscellaria.   In
Connecticut, L athasaria caused about 600 ha
                                         7-1

-------
of defoliation, and populations appeared to be
increasing.  The infestation of L fiscellaria
expanded to 130,000 ha of moderate to heavy
defoliation. About 1,000 ha had greater than
50%  tree  mortality.    Lambdina  athasaria
affected 4,600 ha in Maine with light defoliation.
Defoliation from L. athasaria occurred on 300
ha in Massachusetts and 1,100 ha  in New
Hampshire.   Populations of L fiscellaria  in
Vermont increased dramatically in 1991. Moths
were commonly seen throughout the area, but
no defoliation  occurred.   Damage from  L
athasaria  was found on 600  ha in southern
Vermont compared with 4 ha in 1990.

       Scleroderris  canker was unchanged at
several isolated places in Maine. In the upper
peninsula, the canker was found on 100 ha and
there has  not been significant mortality for 14
years.  The disease  was not found in any new
locations  for the fifth consecutive year  in
Vermont, but about 400 ha are infected in the
northern part of the state.

       White  pine  blister  rust was reported
throughout Maine.   In  Vermont,  significant
mortality and top dieback occurred in a mature
stand and heavy  damage occurred in young
plantations in the southern  part of the state.
Small infestations were reported  in Indiana.
Active blister rust cankers were found at 10 of
the 52 grid points surveyed in West Virginia.

       The hemlock woolly  adelgid  is  an
introduced insect  which attacks both eastern
hemlock  and  Carolina  hemlock.     It   is
established in the Blue Ridge Mountains in the
vicinity  of  the Shenandoah  Valley  and  is
spreading south (Figure 7-1). In the northeast,
the  adelgid   is  reported  in  Connecticut,
Maryland,  Massachusetts,  New Jersey, New
York,   Pennsylvania,   Rhode   Island,   and
Vermont.    Hemlock stands  in   previously
infested areas have been re-infested. The first
occurrences in native hemlock stands were
reported  from  Maryland.    The   infestation
appears   to   be   spreading   in  western
Massachusetts, but  it is still contained in the
Springfield  area.    There  are unconfirmed
reports of the insect in the eastern part of the
state.     New  Jersey  reports   moderate
infestations of native hemlock.  In southeastern
New York, moderate to heavy mortality was
reported on about 6,000 ha. The infestation is
intensifying  and  spreading,  and  hemlock
mortality is increasing.  The  hemlock woolly
adelgid was reported from five new locations in
Pennsylvania.   The infestation increased to
three of the  five counties in Rhode  Island,
spreading mostly in suburban areas along the
coastline.     Some  areas  reported  heavy
defoliation.   The  insect  was  introduced to
Vermont in 1990. Eradication efforts continued
at the  infected site where  several additional
seedlings were found infected.  No evidence of
the pest was found in native hemlock stands in
the valley below the infected site.

Eastern Spruce-Fir

        Eastern spruce budworm (Table 7-1)
caused light  to  moderate   defoliation   on
approximately  44,000 ha  in Minnesota during
1991.  This was a decline from the 80,000 ha
recorded in 1990. No defoliation was reported
from any of the other Lake States (Michigan
and Wisconsin).  For the second year in a row,
no defoliation was reported from Maine,  New
Hampshire, New York, or Vermont. However,
trap catches in both New York and Vermont
were higher than last year.

       Spruce and fir insects of the Southern
Appalachian Mountains have been the subject
of much controversy over  the past decade.
Most questions raised  concern the cause of
catastrophic tree mortality occurring in  this
high-elevation ecosystem. An evaluation of the
standing  dead  tree  losses   was made  in
1984-85  through  a  combination of aerial
photography and ground  sampling.  A  100
percent photographic coverage was made over
the Great Smoky  Mountains  National  Park,
Roan  Mountain,  Grandfather Mountain,  the
Black  Mountains, Balsam Mountains,  Mount
Rogers, and the White Top Range. A total of
232 photo plots and 123 ground plots were
established over all of these  areas with the
exception  of  Mount Rogers  and White Top
Mountain  which were experiencing significant
tree mortality at the time of the survey.

       The survey showed that approximately
26,500  ha of host type was  present in these
                                             7-2

-------
O  
                                                                                                              (Q

                                                                                                              CO
                                                                                                              (O
                                                                                                              <0
                                                                                                              a.

                                                                                                              o5
                                                                                                              £

                                                                                                              £

                                                                                                              m
                                                                                                              'c
                                                                                                             .1
                                                                                                              (O

                                                                                                             T3
                                                                                                              0)
                                                                                                             _
                                                                                                             o

                                                                                                             §
                                                                                                             ^
                                                                                                             o
                                                                                                             _o

                                                                                                             
LL
                                                     7-3

-------
           Table 7-1.  Defoliation by eastern spruce bud worm by state for 1990 and 1991.
   State
Hectares defoliated
1990    1991
                             81,000   44,000
  Percentage of the
spruce/fir type defoliated
1990          1991
Maine
Michigan
Minnesota
New Hampshire
New York
Vermont
Wisconsin
0
1,000
80,000
0
0
0
0
0
0
44,000
0
0
0
0
0
0.1
6.7
0
0
0
0
0
0
3.7
0
0
0
0
                             1.2
               0.6
areas and that  74% of the total host type
occurred  in  the  Great  Smoky  Mountains
National  Park.   The  survey also showed  a
range of 3 to 14% mortality in red spruce and
44 to 91% mortality of Fraser fir. The balsam
woolly adelgid was indicated as a major cause
of fir mortality.

       Balsam  woolly  adelgid  populations
were high again this year.  Mortality continues
throughout the range of  Fraser fir in this area.
A large number of trees are exhibiting thinning
crowns   and   dieback   in   the  southern
Appalachian mountains  of  Tennessee, North
Carolina, and Virginia.  The insect was also
found throughout  most of central and southern
Maine, but populations have declined there and
in southern Vermont following several years of
continued low populations.

       Within New Hampshire, New York, and
Vermont,  the  conifer  swift  moth   is  still
considered to be a factor  in  high elevation
spruce-fir  stands.  High  elevation, permanent
plots continued  to be monitored  for spruce
mortality, but no  increase in occurrence was
reported.

       The spruce beetle  affects white, red,
and black spruces. Heavy infestation continued
on about  3,600  ha in northern  and  western
areas of Maine  and mortality  in  spruce was
                    over 25% in those areas.  Scattered damage
                    was identified on over 400,000 ha. The attack
                    appears to have peaked in 1990 and 1991 in
                    much of this area and is expected to be stable.
                    New York reported infestation on over 3,200 ha
                    in Hamilton and Fulton counties. Mortality was
                    moderate on 1,200 ha and heavy on 600 ha.

                           Stillwell's  Syndrome,  a  decline  of
                    balsam fir that is associated with Armillaria root
                    disease, is present throughout Maine.   The
                    general  trend  is  expected to  remain at  low
                    levels unless a new  stress factor, such as a
                    spruce budworm outbreak, occurs. Scattered,
                    low numbers of afflicted fir were noted on over
                    2,800,000 ha of the spruce-fir forest type.  No
                    significant concentrations of the syndrome were
                    detected and tree mortality was minimal.

                           Cytospora canker is present throughout
                    the range of red spruce.

                    Southern  Pines

                           Fusiform  rust continues as the most
                    damaging disease of the loblolly and slash pine
                    types  in the South. Almost 1 in 3 ha in these
                    forest types show 10% or more trees infected
                    with potentially lethal  cankers.  Georgia is the
                                              7-4

-------
most heavily  affected  state  (2  million  ha)
accounting  for  30% of  all  infected lands.
Alabama  and  Mississippi  also  have  over
800,000  ha each  infected.   Together, these
three states account for nearly  60% of  all
affected   areas.     Table  7-2   summarizes
infections of fusiform rust, and  Figures 7-2 and
7-3 show the fusiform rust hazard ratings for
slash pine and loblolly pine.

       Southern pine beetle  infested nearly
four million ha in 1991, a  150%  increase over
the  previous year  (Figure  7-4).   Heaviest
activity shifted from the western  Gulf states to
the east,  with Alabama accounting for  over
37% of all outbreak acreage.  The outbreak in
the Appalachian mountains has collapsed, but
populations have expanded in  the piedmont of
Georgia,  South Carolina, and  North  Carolina.
A total of  6 ha were newly infested in 1991 in
Maryland  and is the first report since 1987.

       Southern  pine   beetle   populations
continue    to  cause   significant   mortality,
particularly in national forest wilderness areas
where the beetle poses a threat  to colonies of
the red-cockaded woodpecker,  a federally listed
endangered species.  This bird nests in trees
which are highly susceptible to  attack by the
southern pine beetle.

       Table 7-3 compares outbreak acreage
across the South in  1990 and 1991  by state.
Arkansas,  Florida, Kentucky,  Oklahoma, and
Tennessee were not epidemic in either year.
        Littleleaf is a decline disease affecting
shortleaf pine and, to a lesser extent,  loblolly
pine.  It results from the interactions of feeder
root disease,  poor internal drainage of clay
soils eroded after abandonment of agricultural
lands, and older, physiologically mature trees.
Bark  beetles,  especially  the  southern  pine
beetle,  preferentially  attack  littleleaf-affected
stands.   The range of littleleaf disease at its
height in the 1940's and  '50's encompassed
165 counties with a total area of 19.6 million ha
in  the Piedmont Plateau  from Mississippi to
Virginia,  and it included areas in  Tennessee
and Kentucky (Figure 7-5,  Table 7-4).
        Littleleaf  disease  is  probably  less
prevalent  today  due  to  regional  trends in
forestry such as increases in  industrial forest
land ownership with associated lower rotation
ages,  preference  for the  less susceptible
loblolly   pine   over   shortleaf   pine   for
reforestation,  and  urbanization.   However,
specific  information   on   current   disease
distribution is lacking.  A regional analysis of
littleleaf vulnerability   is underway,  utilizing
existing  information  on  historical  disease
distribution, soil properties, and the distribution
of susceptible forest types.

        Pine engraver beetles populations were
active and widespread throughout the South.
In nearly all cases, there was an association
with drought.

Oak-Hickory

        Gypsy  moth populations declined in
1991 throughout  much of the eastern United
States,  where  approximately  1.6 million ha
were defoliated compared  to  more than 2.8
million  ha in 1990 (Table 7-5). The sharpest
decline  occurred  in  Pennsylvania,  where
defoliation was 0.5 million ha compared to 1.4
million   ha  in   1991.   Defoliation  increased
significantly  in the New  England  states of
Maine, New  Hampshire, and Massachusetts
and continued  to  expand into new areas in
Michigan,  Ohio, and West Virginia.   In 1990,
approximately 6% of the susceptible host type
was defoliated, but in  1991  it was only 3%.

        In 1991, approximately 332,000 ha of
cooperative  State  and Federal  lands were
treated  to  protect  forested   communities,
recreation areas, parks, and high-value forests
from the gypsy moth.  This represents about
one-third  less  than  was  treated  in 1990.
Private landowners may have  treated another
0.2 million ha.

        Although   it  is  early,   suppression
estimates  for   1992  suggest that  at least
342,000 ha will be treated with large state
suppression projects  in Maryland,  Michigan,
Pennsylvania,  and West  Virginia.   National
Forest and other Federal suppression is again
expected to be between 20 and 30 thousand
ha.
                                               7-5

-------
Table 7-2. Area and percent of host type infected with fusiform rust by state, 1991.
State (survey yr.)
Alabama (82)
Arkansas (88)
Florida (87)
Georgia (89)
Louisiana (84)
Mississippi (87)
North Carolina (90)
Oklahoma (86)
South Carolina (86)
Texas (86)
Virginia (86)
TOTALS
Area infected
(ha)
1,061,243
124,445
3,949
2,016,985
722,490
817,209
452,047
9,119
745,160
252,961
28,556
6,234,164
Percent of
susceptible type
34
8
23
53
30
32
29
6
40
12
4
30
Table 7-3. Southern Pine Beetle Outbreak Area 1990 and 1991.
State
Alabama
Arkansas
Georgia
Florida
Kentucky
Louisiana
Mississippi
N.Carolina
Oklahoma
S.Carolina
Tennessee
Texas
Virginia
1990
0
0
0
0
0
0
0
45,084
0
939,540
0
728,745
0
1991
1,659,919
0
130,180
0
0
568,097
458,119
40,067
0
977,179
0
164,972
14,188
TOTAL
1,713,369
3,988,876
                                              7-6

-------
                                                    op
                                                    cc
                                                     
-------
                                                        CD

                                                        cc


                                                         c"
                                                         
                                                        09
                                                        C

                                                        "cx
                                                        (O
                                                        CO
                                                        •5
                                                        en
                                                        (O

                                                        u_
                                                        eo
                                                        03
7-8

-------
                                                        
                                                        Q.

                                                        to

                                                        2

                                                        £
                                                        S
                                                        ffl
                                                        o
                                                        13

                                                        0)

                                                        '5.

                                                        c
                                                         o

                                                         M
                                                         (0
                                                         0)
                                                         CO

                                                         ffl
                                                        OCC
                                                        r

                                                         £

                                                        A
                                                        U-
7-9

-------
Figure 7-5.  Historic range of littleleaf disease (Forest Pest Management, R-8).
                                         7-10

-------
Table 7-4. Area and number of counties within the historic range of Irttleleaf disease.
State
Hectares
(million)
Counties
(number)
AL
GA
KY
MS
NC
SC
TN
VA
 5.3
 4.7
 0.3
 0.1
 3.0
 2.8
 1.3
 2.1
Total    19.6
 29
 59
  3
  1
 25
 18
 13
 17

 165
Table 7-5. Area of defoliation by gypsy moth by state, 1990 and 1991.
       State
                        Hectares defoliated

                       1990          1991
     Connecticut
      Delaware
      Massachusetts
      Maryland
      Maine
      Michigan
      New Hampshire
      New Jersey
      New York
      Ohio
      Pennsylvania
      Rhode Island
      Vermont
      Virginia
      Washington, D.C.
      West Virginia

      TOTAL
                      71,488
                       1,534
                      33,844
                      53,871
                      109,487
                      145,076
                      53,927
                      174,589
                      143,385
                         47
                     1,764,251
                          0
                       25,506
                      240,486
                          4
                      137,144

                     2,954,640
                       20,305
                        5,455
                      114,228
                       30,444
                      248,789
                      253,720
                       73,227
                       68,785
                       71,239
                         140
                      498,002
                          0
                        1,456
                      249,474
                         51
                       45,709

                     1,752,263
                                            7-11

-------
           The gypsy moth continues its westward
    and southward spread (Rgure 7-6). In Illinois,
    1,442 moths were trapped, the second largest
    catch ever recorded.  Moth catches continued
    to increase in  northern Indiana  counties and
    limited ground control efforts took  place  in
    Kosciusko County.   In Minnesota,  51  male
    moths were trapped from 31 sites, down from
    a high of 126 in 1990.  In Iowa, 60 male moths
    were trapped from 41 traps in seven counties
    and in Missouri, statewide trapping continues.
    Male moths were trapped  in several cities, 25
    of 36 in St.  Louis. Most  moths in Iowa and
    Missouri  were  associated  with   infested
    ornamental spruce shipped from Pennsylvania.

           Mortality caused by gypsy moth varies
    between stands, with some stands experiencing
    high  losses.   Severe summer  drought  and
    secondary pests contributed to this mortality.
    Still, the majority of stands remain adequately
    stocked.

           Oak  decline is a complex, slow-acting
    disease syndrome involving the interactions of
    predisposing factors  such  as  climate,  site
    quality, or tree age;  inciting stress  such as
    drought or insect defoliation; and contributing
                                      organisms  of  secondary  action  such  as
                                      Armillaria  root  disease  and  the two-lined
                                      chestnut borer.  Decline is characterized by a
                                      gradual but progressive dieback of the crown.
                                      Mortality commonly results among susceptible
                                      trees but usually  not until after dieback has
                                      been  present for several years.   Mature
                                      overstory trees are most often affected.  Oak
                                      decline and mortality is widely distributed over
                                      the eastern half of the United States and has a
                                      long history.  Episodes of damage have been
                                      reported for more than  130 years.  Since the
                                      turn of the century, at least 26 episodes have
                                      been estimated to have  occurred; some  in
                                      each decade except the 1940's.

                                             Forest Inventory  and Analysis data
                                      from 12 Southern  states were used to assess
                                      oak decline on  a  regional scale (Table 7-6).
                                      The dieback pest/damage code was used to
                                      key on plots with decline damage.  Average
                                      annual mortality volume per unit area (oaks) on
                                      affected  and  unaffected  plots  was  then
                                      compared as a measure of decline  severity
                                      (Figures 7-7 and 7-8).
               Table 7-6. Incidence and impact of oak decline in the Southeast, 1991.
State
Affected
Affected Area Incidence
Unaffected
Dead Volume
Dead Volume
(ha) (%) (%)
Alabama
Arkansas
Florida
Georgia
Louisiana
Mississippi
N. Carolina
Oklahoma
S. Carolina
Tennessee
Texas
Virginia

TOTAL
 107,566
 152,964
  67,091
 111,144
  11,385
  45,733
 288,853
   7,400
  34,824
 274,416
  44,753
 440,441

1,586,569
6.87
6.38
18.65
7.82
2.32
3.48
19.63
0.92
5.49
12.02
4.43
19.13

9.88
1.08
1.22
2.43
1.36
1.90
0.88
1.30
2.00
2.49
1.71
2.20
1.53

1.49
1.00
0.97
1.90
1.02
1.09
0.78
1.00
1.35
1.26
1.15
1.55
0.86

1.04
                                             7-12

-------
Figure 7-6.  Gypsy moth defoliation, 1984-1991 (Forest Pest Management, R-8 and NA).
                                       7-13

-------
                                                             to
                                                             0)
                                                            a.
                                                             CD

                                                             O
                                                             o
                                                            'o>
                                                             £

                                                             E
                                                             CD
                                                             (O

                                                             CD
                                                             CD




                                                            1
                                                            •
                                                            .
                                                            CO
                                                            o
                                                            :x
                                                            JD

                                                            •o
                                                            CD
                                                            t3
                                                            co

                                                            co
                                                            CO

                                                            2
                                                            a.

                                                            <  -^
                                                            iZ  °?
                                                            ^  CC
                                                            co    .
                                                             E

                                                             =
                                                             S
                                                             I
                                                             O)
                                                             u_
7-14

-------
                                               w
                                               £

                                              £
                                              .2
                                              O)

                                              2


                                              
-------
       In  the  northeast, oak decline  was
reported   in  Iowa,  Michigan,   Minnesota,
Missouri, and Wisconsin.  There were 300,000
ha of scattered mortality caused by attacks of
the two-lined chestnut borer and armillaria root
disease,  predisposed  by the droughts  of
1987-1989.

       Variable oak leaf caterpillar defoliated
over 400,000 ha in northeast Texas in 1991.
Damage is more spectacular than serious and
trees normally recover.

       Fall  webworm  populations   in
Massachusetts were heavy in 1991, especially
in areas with high proportions of black  cherry.
There were 1.2 million ha affected statewide in
Missouri.    Moderate  to heavy  defoliation
occurred in scattered areas in New Hampshire,
especially along the seacoast. There were
isolated occurrences of complete defoliation in
New York.   Defoliation  was  widespread  in
Vermont, especially along roadsides. Incidence
has  noticeably increased over  1990,  with
complete defoliation occurring in some  areas.

       Oak leaftier population levels were low
in Maine  with  no  reportable   defoliation.
Conditions are expected to remain the same in
1992.  In   Vermont, very few moths were
collected  in pheromone traps  except  in
Rochester    where    numbers   increased
substantially from  last year. No damage was
reported anywhere in the state.

       Oak skeleton izer populations caused
heavy  leaf  skeletonizing on  shingle  oak
throughout  Missouri.  The last outbreak was in
1983, with the next outbreak predicted in 1999.

       The orangestriped oakworm caused
heavy defoliation of 1,400 ha in Montcalm and
Ionia  counties  in  the lower peninsula  of
Michigan.  The insect was also very prevalent
in the Southern Appalachians.

Northern Hardwoods

       Beech bark disease (Nectria coccinea
var. faginata in  association with beech scale)
seemed to  be down in Maine, although heavy
mortality and dieback occurred from the West
Virginia disease complicated by a severe May
freeze.    Scattered  dieback  and  chlorosis
occurred in southern Vermont with an increase
in Nectria fruiting.  Northeastern Vermont had
a decrease of occurrence and the Champlain
Valley had a slight increase of the disease.
Despite a decrease in the area infested by the
scale, about 1,600 ha in New York still showed
heavy infection  (70-100%).   Infection levels
were static in Connecticut.  The disease also
continued to  cause dieback  and  mortality
throughout the other New England states.  The
disease   was  widespread   in   northern
Pennsylvania counties.   In West Virginia, the
presence  of the  disease  continued to  be
reported on the  Monogahela National Forest
though  expansion  in 1991  may have been
limited due to dry weather.

        Elm spanworm defoliated 12,065 ha in
Pennsylvania including areas of the Allegheny
National Forest.

        Sugar and red maple were affected by
maple   decline  in   Maine,  Michigan,
Pennsylvania,  and Vermont.    Maple stands
continued to  be monitored  throughout  the
region.  Symptoms remained static in western
Maine with less than 10% crown dieback. In
the upper peninsula of Michigan, dieback  and
mortality occurred on sites with poor drainage,
heavy or shallow soils, and subjected to forest
tent caterpillar defoliation during the droughts
of 1987-1989.  Some reports of damage in
Pennsylvania may possibly be associated with
pear thrip activity.  In northern Vermont, 1,347
ha of  decline and mortality were observed in
scattered locations.  In general, trees looked
healthier than in previous years due to vigorous
spring flush. Dry summer conditions may lead
to some decline in  1991.

        Pear  thrips  affected   sugar  maple
throughout Maine,  Iowa, Maryland, Michigan,
Minnesota,  Pennsylvania,   Rhode   Island,
Vermont, and Wisconsin.   Population levels
continued to be  low in central and southern
Maine with some light damage observed.  An
area of 1,170 ha was surveyed for the pest.
Populations were very low throughout Maryland
and Pennsylvania.  First report of pear thrips in
                                             7-16

-------
Minnesota was found in Hennepin and Carver
counties.   It was also found in Wisconsin,
Michigan, and  Iowa in  1991.    Some light
damage was reported on survey plots in Rhode
Island.  Damage in Vermont declined in 1991
and was reported light throughout the state and
no  defoliation   was  aerially  mapped.
Overwintering soil counts decreased about 90%
from the previous year and counts were also
reported  lower than  1990.   Spring  weather
conditions which promoted rapid leaf expansion
contributed to lower damage levels.  Dieback
continued in one southern Vermont  maple
stand that was severely defoliated in 1988 and
1989 and also stressed by drought conditions.

       Sapstreak affected some sugarbushes
in the northern part of New York.

       Maple trumpet  skeletonizer  caused
defoliation in several counties in northern New
York,  an increase from last year.  Damage in
Vermont was scattered and variable throughout
the state.

       The   maple   leafcutter   defoliated
approximately 4,000 ha in northern New York.
About 1,200 ha of defoliation were observed in
southern Vermont. Lighter damage could be
found  throughout the  region, especially on
regeneration. Damage  was also found on 824
ha in northern Vermont and populations are
increasing.

       Ash dieback (ash yellows) has been
reported in stands of green, blue, black, white,
and brown  ash in Indiana, Iowa, Maine, New
York, Pennsylvania, and Vermont. Ash decline
and ash yellows are epidemic in northeastern
Indiana and can  be  found  throughout  the
northern half of the state. Annual mortality was
estimated at 3% of the ash population.  The
disease increased in urban trees in  Iowa in
1991.   Maine reported  800 ha  affected  in
Aroostook county, particularly along  streams
and other  wet areas.   This was  a  slight
increase from 1990.   Approximately 300 ha
were  severely  affected  (70-100%  of  trees
symptomatic or dead),  and it appeared to be
something other than ash yellows. The trend is
expected to increase. Ash decline continued to
be associated with ash dieback in New York.
There were nine counties  in Pennsylvania that
reported  symptoms:      Lehigh,   Adams,
Cumberland, Mifflin, Clarion, Erie, Perry, Junita,
and  Clinton.   In Vermont, dieback remained
common  and  stable   especially   in   the
Champlain Valley area and other low elevation
sites with a high stocking of pole size  trees.

        Butternut  canker   is   widespread
throughout the range of butternut.  Up to 100%
mortality has  occurred in some stands and
large geographic areas such as Virginia and
North Carolina have lost 77% of their butternut
over the last 20 years. The canker was found
in Vermont, with new infections detected in
Orange, Lamoille, and Chittenden Counties.
Twenty-eight  counties  are now affected  in
Wisconsin.

        Eastern tent caterpillar nests were very
common along roadsides in Connecticut.   In
Massachusetts, damage has been increasing
over the last several years, especially in  the
eastern portion of the state.  In Michigan,  tow
to moderate  defoliation occurred on over
800,000 ha  in the southern lower peninsula.
Occurrence   was   widespread   throughout
northern Vermont,  similar to last year's level.
However, in southern Vermont, the tents were
less  common.

        Forest tent caterpillar was reported with
widespread light defoliation in  eastern North
Carolina and with  moderate defoliation along
the Roanoke River drainage.  Light scattered
defoliation was observed in Tennessee and
Virginia.   Severe defoliation  occurred   in
western Mississippi and throughout Louisiana,
but  overall  population  levels are declining.
Build-ups  occurred in Texas and Arkansas.
There was no reported defoliation in  Indiana
this year.  Low populations were  reported in
Maine   and  Vermont  with only  scattered
individuals and no defoliation observed.   In
Michigan, 79,202 ha of defoliation occurred in
the upper peninsula.  This was a decline from
265,600 ha in 1990, and appears to signal  the
end of an outbreak that began in 1988.  Only
the  extreme  eastern  upper peninsula may
sustain defoliation in 1992. In Minnesota, light
to moderate defoliation occurred on 49,800 ha
and heavy defoliation on 168,000 ha. This was
the third successive year of some defoliation
on over 1,740,000 ha.  However, the intensity
                                         7-17

-------
has declined and is  expected to continue to
decline in 1992. In New York, over 8,000 ha of
moderate to heavy defoliation was reported in
the Adirondack region.  On over half of  the
area  affected,  the  trees  did  not  refoliate.
Populations are expected to increase in 1992.
In Wisconsin, 19,560 ha were  defoliated in
Marinette and Oconto counties. This appears
to be the  end of an outbreak that began in
1986.   Defoliation,  together with  drought,
resulted in widespread oak mortality.

       Fall   cankerworm  caused  light
defoliation on Cape  Cod where populations
appear to be increasing.

       Cherry  scallop  shell   moth-caused
damage was heavy within stands composed of
a high percentage of black cherry in  central
and western  Massachusetts. The infestation
occurred on 2,544 ha and  the population is
expected to continue to cause damage in  the
next  year  or  two.   This  is  the  second
consecutive year of damage in some areas in
northern New York.   There was some heavy
defoliation in Vermont, where the  pest was
common in the southern portion of the state.

       Saddled prominent populations   are
stable at low levels  and are not  expected to
change in the near future.  Numbers of moths
caught  in pheromone traps  were  minimal.
Infestation was reported to be heavy on 800 ha
in Berkshire County in western Massachusetts.
Presence of a disease in the population was
indicated.   In Vermont,  populations declined
and   no  defoliation  was  aerially mapped.
Larvae were occasionally observed and small
numbers of  moths were caught  in light  and
pheromone traps.   A fungal  pathogen was
suspected to have caused the decline.

Aspen-Birch

       Birch dieback  continued  in  Maine
affecting 13,890 ha in the Rangeley region and
20,000 ha in Washington county.  Defoliation
was light throughout  the area.  In an  annual
plot  survey,  59-66%  of  the  paper birch
exhibited top dieback and 3-9%  died.  Tree
health as reflected by crown condition improved
from  1990.     In northern  Vermont,  trees
appeared  to  be recovering,   but scattered
mortality was still occurring.  Widespread birch
defoliation  and  dry   summer  weather  is
expected  to  cause  dieback  in  southern
Vermont in the future.

       Scattered defoliation caused by  the
birch casebearer was recorded on over 40,000
ha in central and eastern portions of Maine.
Populations  were  higher than expected in
coastal areas.  The insect was also present in
the northern counties.

       Two  species   of  birch  leafminers
affected gray, paper,  and yellow birches in
Vermont and Maine. Populations were high in
some locations  in central and eastern coastal
portions of Maine  affecting over  48,000  ha.
Damage from this pest  was widespread in
southern Vermont with 2,000 ha of defoliation.
Over  10,000 ha were  moderately to heavily
defoliated  in northern Vermont.

       The  birch skeletonizer,  which caused
moderate to severe late season defoliation on
4,000 ha in Maine, was most visible on higher
elevation  paper birch  sites in the  western
portion  of the  state.    Defoliation  was  up
noticeably from  last year.  Over 4,500 ha were
affected in the Adirondack region of New York
with about half  of the area heavily defoliated.
In Vermont, there were scattered occurrences
of moderate to heavy defoliation.

       The  bronze birch borer continued to
attack paper birch in the northern regions of
Michigan, Minnesota, and Wisconsin where the
effects of the 1987-89 drought linger.

       Hypoxylon  canker   is  a  common
occurrence throughout the range of aspen.

Special  Topics

Dogwood anthracnose

       Since first discovered in the South in
northern Georgia (1987), dogwood anthracnose
continues  to expand  rapidly throughout  the
range of  the tree.  To  date,  120  counties,
encompassing some 2.3 million ha in seven
southeastern states,  have  shown confirmed
infections  (Figure 7-9).  The disease is more
prevalent  in   moist,   cool  sites  such  as
                                             7-18

-------
Figure 7-9.  Dogwood  anthracnose occurence  in  the  eastern  United  States  (Forest  Pest
           Management, R-8 and NA).
                                       7-19

-------
            Table 7-7. Area affected by Dogwood Anthracnose in the Southeast, 1988-91
State
1988
1989
1990
1991
Alabama
Georgia
Kentucky
N.Carolina
S.Carolina
Tennessee
Virginia
.
62,298
0
50,003
64,086
102,499
5,438

149,561
0
215,275
119,706
265,572
140,258

337,215
0
615,291
298,452
531 ,656
487,876

429,013
0
758,593
325,732
827,489
1,461,651
Notes:
(1) Alabama data were not available for this report.
(2) Overall estimate in the text does not equal the sum of the states in the table because the values are
based on the total areas for all states combined.
north-facing   slopes   and   beneath  dense
overstories.

       The  disease  was present throughout
Connecticut and Maryland.  Infections occurred
in eastern  Massachusetts and southeastern
New  Hampshire. The disease was identified
infecting dogwood throughout Long Island and
all along the southern tier of New York. It was
also present in Monroe County which is along
Lake  Ontario.   The  disease was present in
most counties in Pennsylvania.  All 55 counties
in West  Virginia  had dogwood anthracnose
reported.
       Table   7-7   shows   the  estimated
cumulative acreage increase by state by year
since first quantified in 1988 in the South.
Cabbage Palm Mortality

       The Florida Department of Agriculture
reported thousands of cabbage palms dead or
dying along the Gulf coast (cause  unknown).
The area affected is approximately 20 km long
and 1 to 3 km wide between Crystal River and
Cedar Key. Palms of  all  ages, but primarily
older ones, are affected. Trees on islands as
well as along the coastline are afflicted.
                                 Black Gum Disease

                                        Black gum disease (cause unknown)
                                 intensified  in  the  Southern  Appalachian
                                 mountains, principally in the tri-state area of
                                 Tennessee,   Georgia,  and  North  Carolina
                                 (Figure 7-10).  This condition appears to have
                                 great potential significance. Its occurrence in
                                 frequent association with dogwood anthracnose
                                 suggests that the two might have  a similar
                                 etiology.

                                 Dutch Elm Disease

                                        Dutch elm  disease remains common
                                 throughout the range of American elm with the
                                 more  virulent  strain causing  mortality  in
                                 previously unaffected trees. It is considered to
                                 be the most  important forest and  shade tree
                                 problem in many states.

                                 Pitch Pine Looper

                                        Pitch   pine  looper (Lambdina spp.)
                                 caused the first year of noticeable defoliation
                                 from L pellucidaria in Massachusetts since the
                                 mid 1970's. About 200 ha were defoliated and
                                 the  infestation  is   expected   to  increase.
                                 Approximately 81,000 ha were defoliated by L.
                                             7-20

-------
                                                        op

                                                        CC



                                                        
-------
athasaria in the Pine Barrens in New Jersey.
Some tree mortality may occur. L pellucidaria
caused moderate to heavy damage on about
250 ha on Long Island in New York.

        Eastern Larch Mortality

        Eastern   larch  mortality   was  still
common throughout  most of Maine.   Up  to
100%  mortality  occurred  in  some  stands
infested with the  eastern larch beetle.  New
attacks were most  common in  previously
infected  stands.   However, new localized
outbreaks occurred in eastern counties.  Only
8 ha of mortality were detected in Vermont.

        European larch canker reported  no
increases in  1991.   The area affected was
approximately 2,600 ha and remains relatively
unchanged.  Infection was heavy at two coastal
epicenters and  light to moderate elsewhere.
No major increases are expected in the future.
The quarantine was reviewed and  remains in
effect.

General Stressors

       Armillaria  root rot-caused mortality  in
northern red oak, northern pin oak, scarlet oak,
and red pine was scattered over 200,000 ha in
Wisconsin,  Minnesota, Iowa, and  Michigan
related to regional drought in previous years.

       Anthracnose was especially prevalent
this year in the Appalachian mountains, due no
doubt to the unusually wet  spring.   Maples
were perhaps most conspicuously affected, but
various other hardwoods were also damaged.
Light damage was observed in Vermont and
Maine on American sycamore, maple and other
hardwoods, down  dramatically from 1990.

       Annosus root disease continues as the
most  serious root  disease  of  the  South.
Southern pine  beetle,  Ips spp.,  and  black
turpentine  infestations frequently  occur  in
infected stands. Trace to light infections were
present in red pine plantations in Maine.  The
trend was expected to gradually increase.

       Decay continues as one of the most
important   problems   in  hardwoods.   The
condition is widespread and is often associated
with fire, logging injury, or storm damage.

Cyclic Hardwood Insects

        Incidence of the black  twig borer, a
recently introduced ambrosia borer, is on the
increase.  It is sometimes found  in association
with fusarium canker, especially in shade trees.

        Buck moth defoliation was prominent in
Tennessee on  ridge tops and  on the upper
slopes  of forest  stands  of   the  Western
Highlands Rim.  Scattered defoliation was also
reported throughout  Virginia.   Moderate  to
severe defoliation occurred in the New Orleans
metropolitan area.   Populations have been
building in Delaware for the past few years.

       About 65,000 ha were defoliated by the
fruit tree leafroller in the Atchafalaya basin of
south central Louisiana.

       Brood  XIV  of the  periodical  cicada
caused considerable twig damage to various
hardwood species in western North Carolina.
Eastern  Tennessee  and  the   Cumberland
Plateau  also  incurred  moderate  to  heavy
damage.

Ozone

       Ozone  damage  symptoms appeared
throughout the  South in 1991.  Tipburn was
observed in some eastern white pine families.
Indicator plants were  used to assess ozone
levels in wilderness areas in the South, with
light to moderate damage on sensitive native
species.  Ozone symptom surveys continued to
be  conducted  in Wilderness Areas  on  the
White and Green Mountain National Forests.
Evaluations showed light damage occurred on
sensitive native plants.

Windstorms

        In Wisconsin,  8,715  ha  in Chippewa,
Lincoln, Langlade, and Oneida counties were
damaged by a storm on May 29, 1991. There
were  28,700  ha in  Rusk, Barren, Chippewa,
and Price counties affected by another storm
on June 7.
                                         7-22

-------
    Drought
       Residential  and  rural  oaks  were
affected  in  Delaware.   Light mortality  was
reported in Indiana. The species most affected
were yellow poplar, sugar maple, white pine,
and black cherry.  Drought affected jack, red,
and eastern white pine in Iowa. There were 89
ha  of  tree  mortality  in  small  pockets  in
plantations.   This mortality was related  to
lingering  effects of 1987,  1988, and 1989
droughts.    Scattered  mortality  occurred  to
balsam   fir   and  paper  birch  in   Itasca,
Koochiching,  and  St.   Louis   counties  in
Minnesota.   Lake  and Carlton counties  also
had scattered birch mortality. Drought has also
affected forests  in central and western  New
York.  Drought conditions in June and July led
to leaf  scorch,  chlorosis,  leaf  curling  and
dropped  leaves  on hardwoods  in Vermont.
Symptoms are expected to show for the  next
couple of years.  In West Virginia, drought
conditions were worst in northern counties.

Hurricane Bob

       Coastal areas were hard hit with salt
spray damage, conifers being most susceptible.
Forest trees and ornamentals were broken or
blown down in  areas  throughout  the coastal
New   England   region.     Cape   Cod,
Massachusetts was the hardest hit with pockets
of severe damage occurring  throughout  the
area.   Most susceptible were  black  locust,
pines, older oaks, and  other stressed trees
being uprooted and broken. Remnants of the
damage  are  expected to remain for  years.
Many trees were permanently tilted in  certain
larch plantations in Maine.

Ice Storms

       In Indiana, 40,000 ha of jack pine in
forest, fence row, and yard trees in 13 counties
were damaged.  Damage included uprooted
and broken limbs and tops. Up to an  inch of
ice damaged trees in western and central New
York.   Monroe  County  reported the most
damage with 30% of the trees severely  broken
and another 30-50%  moderately damaged.
Most species affected were ash, red and silver
maples, pines, larch, willow, and other stressed
or weakened trees.  The storm also caused
limbs to droop in the summer and internal
injuries  which may take more time to detect.
Salvage of timber and pruning,  removal, and
planting of ornamental trees is being  done.
Over three million cubic yards of  debris was
cleared  from Monroe County. A severe storm
in March,  1991  snapped trees on over  1,740
ha acres in Wisconsin.

Spring  Visits

        As a   part  of the  Forest  Health
Monitoring initiative,   crews  in   Alabama,
Georgia, and Virginia are making spring visits
to the FHM plots.  Activity in all  states was
minimal  this   year.    Only  some   minor
frost/freeze damage was reported, principally in
Alabama.

Aerial Photography of Georgia FHM Plots

        Photointerpretation of Georgia Forest
Health Monitoring plots showed that in the vast
majority of cases (about 98.5%), tree mortality
averaged  1% or less  per stand in which the
FHM plot  was located.  The most significant
single cause of mortality in any one stand (6%)
was  attributed to fire.

Acknowledgements

        The information from the Northeastern
Area was  compiled by personnel from USDA
Forest  Service,  Forest  Health  Protection:
Margaret  Miller-Weeks,  Susan  Cox,  and
Florence Peterson of  the Durham, NH Field
Office; Dan Twardus of the Morgantown, WV
Field Office; Manfred  Mielke of  the  St.  Paul,
MN Reid Office; and Charles Hatch of the Area
Office in Radnor, PA.

        Information from the  Southeast was
compiled by personnel of the USDA Forest
Service, Forest Pest  Management offices  in
Asheville,  NC,  Pineville, LA, and Atlanta, GA:
James D. Ward (Gypsy moth); Jose F. Negron
(hemlock  woolly  adelgid);  Steven  W.  Oak
(littleleaf disease); Dale  Starkey  (oak wilt);
Nolan Hess (annosus root rot);  John Taylor,
Walter Salazar,  Carol Scott (Georgia FHM Plot
Photointerpretation); Wes Nettleton (southern
                                            7-23

-------
pine  beetle);  and  Steve  Holtzman   (GIS
support).

       The hemlock woolly adelgid distribution
map   is  a   collaborative  effort  by   the
Southeastern Forest Experiment Station (Forest
Inventory and Analysis), the Department of the
Interior (Shenandoah National Park and  Blue
Ridge  Parkway),  and  the  Department  of
Entomology, Virginia Polytechnic Institute and
State University.

       Additional information was contributed
by   personnel   from  the  following  state
organizations:
Alabama Forestry Commission  (Jim  Hyland,
FHM Coordinator)
Arkansas Forestry Commission
Connecticut   Department  of  Environmental
Protection, Bureau of Forestry
Delaware Department of Agriculture
Florida Division of Forestry
Georgia Forestry Commission (Terry Price and
Richard Jernigan, FHM Coordinators)
Illinois Department of Conservation, Division of
Forest  Resources
Illinois  Natural History Survey
Louisiana Forestry Commission
Maine Forest Service
Massachusetts Division of Forests and Parks
Maryland Department of Agriculture
Michigan Department of Agriculture
Michigan Department of Natural Resources,
Division of Forestry
Minnesota Department  of Natural Resources,
Division of Forestry
Mississippi Forestry Commission
Missouri Department of Conservation, Forestry
Division
New Hampshire Division of Forests and Lands,
Department   of  Resources and  Economic
Development
New Jersey Department of Agriculture
New York State Department of Environmental
Conservation, Division of Lands  and Forests
North Carolina Forest Service
Ohio   Department  of  Natural Resources,
Division of Forestry
Oklahoma Forestry Division
Pennsylvania Bureau of Forestry, Division of
Pest Management
Rhode  Island  Division of  Agriculture  and
Marketing
South Carolina Forestry Commission
Tennessee Division  of Forestry
Texas Forest Service
Vermont  Division  of  Forests,  Parks,  and
Recreation, Forest  Resource Protection
Virginia  Division of Forestry (Tim Tigner, FHM
Coordinator)
West Virginia Department of Agriculture, Pest
Industry Division
Wisconsin Department of Agriculture
Wisconsin Department  of Natural Resources,
Forestry Division
                                         7-24

-------
8.   Selected Air Quality and
    Deposition  Data Summaries


    D. S.  Shadwick and L. A.
    Smith

       This   chapter   presents   maps  of
interpolated  ozone  data  for  low  elevation
regions and  maps  and  regional  summary
statistics for ions found  in precipitation.  The
selection of air quality and deposition data for
inclusion in this report has been motivated by
several concerns.  First, the  air constituents
that have been summarized are suspected to
be in high enough concentrations to adversely
affect vegetation immediately or in the future.
For example, this is  the reason that gaseous
NO2 is not included.  Second, only data  from
monitors that cover a broad geographic  area
and  that come from a  sufficiently  dense
network are  included. This criterion excluded,
for   example,   maps   of  dry  deposition
constituents.    Finally,  pollutants  that  are
typically  point source oriented (for  example,
SO2 and fluoride) have not been included.  The
review by Garner et al.  (1989)  is a good
overview of possible pollutant effects on forests
in the eastern  U.S.

Ozone

       The information consists of interpolated
(i.e., kriged) maps of the seasonal ozone W126
index (ppm-h):
                        —
              l+4403exp(-126 x
where  O, is the  Ith hourly average  ozone
concentration (in ppm) and the index includes
all  hourly  average  concentrations for the
months April-October, inclusively. The rationale
for  the  emphasis  on  peak  ozone  hourly
average  concentrations  and  examples  of
applications  of the index are in Lefohn et al.
(1992).

       Maps of the interpolated W126  values
and the relative errors (interpolation standard
deviation divided by the interpolated value)
were  prepared for the years  1985-1989.
The area covered by the maps includes all
states east  of,  and adjacent  to,  the
Mississippi River (approximately 96° west
longitude).  The maps are of kriged values
on a  1/2°x 1/z° (latitude by longitude) grid.
Figures 8-1, 8-2, 8-3, 8-4, and 8-5 show for
each year, smooth contours of the gridded
values at  10  ppm-h intervals.    The
seasonal W126 values shown on the maps
are generally lower than values observed
at sites in southern California where ozone
has been linked to forest  damage.   For
example, seasonal W126  values for  the
Lake Gregory, CA ozone monitor (AIRS ID
060710005) are generally in excess of 200
ppm-h.

       The ozone data sources were the
U.S.   EPA's   Aerometric   Information
Retrieval  System (AIRS), National  Dry
Deposition Network (NDDN), and Mountain
Cloud  Chemistry  Program (MCCP) data
bases. All data used in the analysis have
passed  through the   appropriate quality
assurance programs  (AIRS, Office of  the
Federal  Register  1988;  NDDN   1989;
MCCP 1987).  Only those  sites with land
use designations 'suburban' or 'rural'  in the
AIRS  data base were used. Furthermore,
only data for which the actual data capture
was at least 75% of all possible hourly
average  values  for  the  April-October
'ozone season' (at least 3852 actual out of
5136 possible hourly average values) were
included.  Details on corrections for  the
seasonal W126 index to account for less
than 100% data capture and  the kriging
procedures used  in  the analysis of  the
ozone data can be found in Lefohn  et al.
(1992).

       To  suggest ways that the kriged
ozone information across  years can  be
aggregated, an example of  a  'gradient'
analysis  is shown in Figure  8-6.   The
method  of  aggregating  the  kriged
information  to  produce  the  figure is
explained in Lefohn et al.  (1992) for  the
'Student  An a I y  s i  s'.  This  illustrative
                                         8-1

-------
                                                                    70.00
Figure 8-1.  Map of kriged seasonal ozone W126 index at 10 ppm-h intervals for 1985.
                                       8-2

-------
                                                     E in  p pm  h
                                                    0.00 -
                                                    0.00 -
                                                   20.00 -
                                                   30.00 -
                                                   40.00 -
                                                   50.00 -
                                                   60.00 -
10.00
20.00
30.00
40.00
50.00
60.00
70.00
                                                   Over 70.00
Figure 8-2.  Map of kriged seasonal ozone W126 index at 10 ppm-h intervals for 1986.
                              8-3

-------
                                                              E   in  p pm  h

1
2
3
4
5
6
0
0
0
0
0
0
0
0
V
.0
.0
.0
.0
.0
.0
.0
e r
0 -
0 -
0 -
0 -
0 -
0 -
0 -
70
1
2
3
4
5
6
7
.0
0
0
0
0
0
0
0
0
.0
.0
.0
.0
.0
.0
.0

0
0
0
0
0
0
0

Figure 8-3.  Map of kriged seasonal ozone W126 index at 10 ppm-h intervals for 1987.
                                  8-4

-------
                                                          ONE  in  p pm  h

1
2
3
4
5
6
0
0.
0.
0.
0.
0.
0.
0.
v e
0
0
0
0
0
0
0
r
0 -
0 -
0 -
0 -
0 -
0 -
0 -
70.
1
2
3
4
5
6
7
0
0.
0.
0.
0.
0.
0.
0.
0
00
00
00
00
00
00
00

Figure 8-4.  Map of kriged seasonal ozone W126 index at 10 ppm-h intervals for 1988.
                                 8-5

-------
                                                        Over  70.00
Figure 8-5. Map of kriged seasonal ozone W126 index at 10 ppm-h intervals for 1989.
                                8-6

-------
Figure 8-6.  An example of a 'gradient' analysis showing one way that kriged ozone information
            across years can be aggregated.
                                         8-7

-------
example  may  or  may  not  be  the  'best'
method of analysis for other questions that
could be asked of the data. The analysis was
designed  to identify  regions  where seasonal
ozone W126 values were high (or low),  not
only  within  a  year, but  also  across  years
relative to other regions.  Noteworthy features
of the map  are the  regions  where ozone is
relatively  high each year.  One such region
extends   continuously   along  the  Atlantic
seacoast  and  extends  roughly  to   the
Appalachian Mountains.  A second region of
smaller area is centered  over the Ohio River
Valley near the Ohio-Indiana border.

Wet Deposition

       This section summarizes the analysis
of wet deposition of hydrogen, sulfate, nitrate,
and  ammonium ions for the calendar year
1990.   Data sources  included the National
Atmospheric  Deposition  Program  (NADP)
(National   Atmospheric   Deposition
Program/National  Trends   Network  1991).
Aubertin  et  al.  (1990) describe  the  NADP
quality assurance program.   To ensure that
representative  values were  used,  the same
data completeness criteria adopted by NADP
were applied here before an annual value was
accepted as valid (NADP  1991).

       The value of pH=5.0 is chosen here as
the benchmark for  natural  precipitation  in
accord with  the suggestion by Charison and
Rodhe (1982) that  pH=5.0  is a  reasonable
midpoint for their hypothetical range of natural
precipitation  pH values from pH=4.5 to pH=5.6.
The value of pH=5.0 is also in good agreement
with  values actually  monitored  at remote
locations (Galloway et. al. 1982; Galloway and
Gaudry 1984).

       Figures 8-7, 8-8, 8-9, and 8-10 display
the 1990 annual levels of precipitation-weighted
mean pH,  sulfate deposition, nitrate deposition,
and ammonium deposition, respectively, for the
eastern United States. Figures 8-7 through 8-9
indicate that in 1990, precipitation was most
acidic and sulfate and nitrate deposition levels
highest in an   area  extending  roughly from
eastern Michigan, down to southern Indiana,
across Ohio, West Virginia, and Pennsylvania,
and into New York and southern New England.
Levels generally  decrease  in all directions
away from this area.  This general pattern is
consistent from year to year for each of these
ions.   Figure  8-7 indicates that  almost the
entire  eastern United  States  experiences
precipitation  more  acidic  than  would be
expected from natural processes; that is, nearly
all the eastern United States has precipitation
with an average pH  less than 5.

       Ammonium deposition does not have a
clear  spatial pattern (Figure  8-10).  In 1990,
ammonium levels tended to  increase  with
distance from the  coast. However, this same
picture is not  necessarily obtained for other
years (results not  shown).

       Wet deposition data from certain areas
of special interest  to Forest Health Monitoring
(FHM) were examined. These areas were:
Alabama-Georgia,   mid-Atlantic   (Virginia,
Maryland, Delaware, and New  Jersey), and
New   England   (Maine,   Vermont,    New
Hampshire, Massachusetts, Connecticut, and
Rhode Island).  A few sites that are near, but
outside, these  areas were also included.  For
example,  a few  sites were  used  that  are
located just north  of the Alabama-Tennessee
border and in  eastern New York.  Table 8-1
shows median annual values over the entire
period of  monitoring  for  which  data  are
available  (1979-1990).   The  physiographic
regions  listed  in Table 8-1 are based upon
those outlined  in  The  National Atlas  of the
United States  of America  (US  Geological
Survey 1970) and a geography text by Wheeler
etal. (1969).

       Some   physiographic  regions  had
relatively little data. The statistics in Table 8-1
roughly agree with Figures 8-7 through 8-10, in
that more acidic precipitation  and  higher
deposition levels occur in the mid-Atlantic area
than in the other two areas. Within each of the
three  areas,  acidity  and deposition  levels
appear to be lower in the coastal areas  than in
the other physiographic regions.
                                              8-8

-------
Figure 8-7.  The 1990 annual levels of precipitation-weighted mean pH for the eastern United
            States.
                                          8-9

-------
Figure 8-8.  The  1990 annual levels of precipitation-weighted  mean sulfate deposition for the
            eastern United States.
                                          8-10

-------
Figure 8-9.  The  1990 annual levels of precipitation-weighted mean nitrate deposition for the
            eastern United States.
                                          8-11

-------
Figure 8-10.   Thel 990 annual levels of precipitation-weighted mean ammonium deposition for
              the eastern United States.
                                        8-12

-------
                 Table 8-1. Median values of annual summary statistics, 1979-1990.




AL-GA:
coastal plain
Piedmont
mountains
uplands
mid-Atlantic:
coastal plain
Piedmont
mountains
New England:
lower
highlands
Number
of
site years"


34
10
12
4

27
10
26

19
74


-fid.


4.69
4.54
4.56
4.51

4.40
4.26
4.36

4.49
4.38

sulfate
dep."


14.20
18.82
22.42
20.29

21.23
25.63
30.14

20.64
21.19

nitrate
dep.
kg/ha

7.65
8.87
10.89
10.36

12.31
16.23
16.49

11.74
15.22

ammonium
dep.


1.43
1.79
2.08
1.97

2.12
2.34
2.69

1.48
1.99
a A site-year is one year of monitoring (subject to meeting data completeness criteria) at one site.
b Deposition
                                             8-13

-------
9.  References
Anderson, R.L and R.P. Belanger.  1987. A
crown rating method for assessing tree vigor of
loblolly and shortleaf pines.  In: Phillips  DR,
comp.  Proceedings  of  the fourth  biennial
southern  silvicuttural  research conference.
November 4-6,  1986; Atlanta,  GA.   General
Technical Report SE-42, USOA Southeastern
Forest Experiment Station, Asheville, NC. 538-
543.

Anderson, R.L, W.G. Burkman, (.Millers,  and
W.H.  Hoffard.  1992.   Visual crown  rating
model for upper canopy trees in the eastern
United  States.    USDA  Forest   Service,
Southeastern   Region,   Forest  Pest
Management. 15 pps.

Anderson, R. L and I. Millers. 1992.  Tree
groups   based   on   foliage  and   crown
characteristics.  USDA Forest  Service white
paper. Asheville, North Carolina. 7pp.

Aubertin, G.M.,  D.S.  Bigelow, and B.A. Malo
[eds.J.   1990.  Quality  Assurance  Plan:
NADP/NTN Deposition  Monitoring.  National
Atmospheric  Deposition  Program.  Natural
Resource Ecology Laboratory, Colorado State
University, Fort Collins, CO.

Belanger,  R.P. and R.L Anderson.  1991. A
guide for visually assessing crown densities of
loblolly and shortleaf pines.   USDA Forest
Service, Research Note SE-352. 1 p.

Byers,  G.E., and  R.D. Van Remortel (eds.).
1991.     Environmental   Monitoring   and
Assessment Program Forests:   Laboratory
Methods   Manual  for   the   Forest  Health
Monitoring Demonstration and  Pilot Projects.
Internal Report.  U.S. Environmental Protection
Agency, Las Vegas, NV.

Charlson,  R.J. and H. Rhode.  1982. Factors
controlling the  acidity of  natural  rainwater.
Nature 295:683-685.

Chojnacky, D. 1991. Eastern  Forest Health
Monitoring:Field Measurements Guide. USDA
Forest Service, Forest Survey, Intermountain
Research Station, Ogden, UT.

Cochran, W.G. 1977.  Sampling Techniques,
Third Edition. New York: John Wiley and Sons,
Inc.

Conkling, B.L, and G.E. Byers (eds.).  1991.
Environmental   Monitoring and  Assessment
Program - Forests: Field Methods Manual for
the Forest Health Monitoring Demonstration
and Pilot  Projects.   Internal Report.   U.S.
Environmental  Protection Agency, Las Vegas,
NV.

Conkling, B.L  and G.E. Byers (eds.).  1992.
Forest Health Monitoring Field Methods Guide,
Internal Report. U.S. Environmental Protection
Agency, Las Vegas, NV.

Dolph, K.E. 1988. Predicting height increment
of young-growth mixed  conifers in the  Sierra
Nevada.   USDA Forest Service Research
Paper, PSW-191. 7 pps.

Einhaus, R.L, D.M. McMullen, R.L  Graves,
and   P.H.  Friedman.     In  Preparation.
Environmental   Monitoring and  Assessment
Program Quality  Assurance  Program  Plan.
Environmental Monitoring Systems Laboratory,
Office  of  Research and  Development, U.S.
Environmental  Protection  Agency, Cincinnati,
Ohio.

EPA. 1990. Threats to biological diversity in the
United States.  Office of Policy, Planning, and
Evaluation. PM-223X.57p.

Francis, J.K.  1986. The relationship of bole
diameters   and   crown  widths  of   seven
bottomland hardwood species.  USDA Forest
Service, Research  Note, SO-328,  October,
1986.  3 pps.

Galloway,  J.N. and A.  Gaudry.  1984. The
composition of precipitation  on  Amsterdam
Island,   Indian  Ocean.   Atmospheric
Environment. 18:2649-2656.

Galloway, J.N., G.E. Likens, W.C. Keene, and
J.M.   Miller.   1982.  The  composition   of

-------
presipitation in  remote areas  of the  world.
Journal of Geophysical Research: Oceans and
Atmospheres. 87:8771-8786.

Garner, J.H.B., T. Pagano, and E.B. Cowling.
1989. An evaluation of the role of ozone acid
deposition, and other airborne pollutants in the
forests of eastern North America. Gen. Tech.
Rep. SE-59, U.S. Dept. of Agriculture, Forest
Service,  S.E.  Forest  Experiment  Station,
Ashevilte, NC. 172 pp.

Grano, C.X.  1957.  Growth of loblolly pine
seed trees in relation to crown density. Journal
of Forestry, 55(11):852.

Hill, M.O. 1973.  Diversity and  evenness:  a
unifying   notation  and   its   consequences.
Ecology 54:427-432.

Horvitz,  D.G. and D.J. Thompson. 1952.  A
generalization of sampling without replacement
from a finite universe. J. American Stat. Assoc.
47:663-685.

Hurlbert, S.H. 1971. The nonconcept of species
diversity:  a critique and alternative parameters.
Ecology 52: 577-586.

Iman, R.L and W.J. Conover. 1983. A modem
approach to statitics. John Wiley & Sons, New
York.

Kutman,  H.M.    1971.    Effects  of  insect
defoliation on growth and mortality of trees.
Annual Review of Entomology, 16:289-324.

Lefohn, A.S., C.M. Benkovitz, R.L. Tanner, D.S.
Shadwick, and LA. Smith. 1990. 'Air quality
measurements and characterization for effects
research.  Report  7." in  NAPAP  State   of
Science  and  Technology.   National  Acid
Precipitation   Assessment  Program,
Washington, DC, 1990.

Lefohn, A.S., H.P. Knudsen, D.S. Shadwick,
and K.A.  Hermann. 1992.  Surface-level ozone
exposures in the eastern United States and the
potential for vegetation effects. ]n_: Transactions
of  the   Tropospheric    Ozone  and  the
Environment Specialty Conference I
Pittsburgh, PA.
A&WMA
Linthurst, R.S., D.H. Landers, J.M. Eilers, D.F.
Brakke,  W.S. Overton,  E.P.  Meier, and R.E.
Crowe.  1986. Characteristics of lakes in the
Eastern  United States.  Volume I: Population
descriptions   and   physico-chemical
relationships. EPA-600/4-86/007A, U.S. EPA,
Corvallis, OR.

MCCP. 1987. MGCP Ozone Monitoring Quality
Assurance  Project  Plan (Revision  No.  2,
February, 1987).

Millers, I., R.L. Anderson, W.G.  Burkman, and
W.H. Hoffard. 1992.  Definitions, Acquisition,
and  Use of Crown Measurements.   USDA
Forest  Service, State and  Private Forestry,
Northeastern Area, Forest Health  Protection.
19 pps.

National  Atmospheric  Deposition  Program.
1991. NADP/NTN Annual   Data  Summary.
Precipitation Chemistry in the  United  States
1990. Natural Resource Ecology Laboratory.
Colorado State University. Fort Collins.

National Atmospheric Deposition Program (IR-
7)/National  Trends Network. 1991. Tape of
Weekly  Data:  July,  1978  -   April,  1991.
NADP/NTN   Coordination   Office.   Natural
Resource Ecology Laboratory. Colorado State
University. Fort Collins.

NDDN. 1989. National Dry Deposition Network,
Project Quality Assurance Plan.  Environmental
Science and Engineering, Inc. Gainesville, FL
(ESE No. 86612-0212-3140).

Office  of  the  Federal  Register,  National
Archives and Records  Administration. 1988.
Code of Federal Regulations (CFR), Title 40.
U.S. Government Printing Office, Washington,
DC.

Omernik,  J.M.  1987.   Ecoregions  of  the
coterminous United States. Ann. Amer. Geog.
77: 118-125.
                                             9-2

-------
Overton,  W.S.,  D. White, and  D.L Stevens.
1990. Design report for EMAP (Environmental
Monitoring   and   Assessment   Program).
EPA/600/3-91/053,   U.S.   Environmental
Protection Agency, Office of Research  and
Development, Washington, DC.

Ricklefs,  R.E.   1987. Community  diversity:
relative roles of local  and regional processes.
Science 235:167-171.

Riitters, K.H., M.L Papp, D.L. Cassell, and J.
Hazard   (editors).  1991.     Forest  Health
Monitoring Plot Design and  Logistics Study.
EPA/600/S3-91/051.    U.S.   Environmental
Protection Agency, Office of Research  and
Development, Washington, DC.

Schmitt, D.M., D.G. Grimble,  and J.L Searcy.
1984.   Managing  the spruce  budworm in
eastern North America. USDA Forest Service,
Agriculture Handbook No. 620. October, 1984.
192 pps.

Scott, C.T. 1991.  "Optimal Design  of a  Plot
Cluster for Monitoring", in: Proceedings from
The Optimal  Design of Forest  Experiments and
Surveys,  1991.

Shannon, C.E.  and  W.  Wiener. 1949.  The
mathematical theory of communication. Univ.
of Illinois Press, Urbana.  117  pp.

Simpson, E.H. 1949. Measurement of diversity.
Nature 163: 688.

Sprinz,  P.T. and  H.E.  Burkhart.    1987.
Relationships between tree crown, stem,  and
stand characteristics in unthinned loblolly  pine
plantations.  Canadian  Journal of  Forestry
Research, 17(6):534-538.

Stanley, T.W.,  and S.S. Vemer.  1985.   The
U.S. Environmental Protection Agency's Quality
Assurance Program. IN: Quality Assurance for
Environmental  Measurements.   ASTM  STP
867. pp. 12-19. American Society for Testing
and Materials, Philadelphia, Pennsylvania.

Stevens,  D.L, A.R. Olsen, and D. White. In
Review.     Environmental   Monitoring   and
Assessment Program  -  Integrated  Sampling
Design. U.S. Environmental Protection Agency,
Corvallis, OR.

Stolte,  K.W., D.M.  Duriscoe, E.R. Cook,  and
S.P. Cline.  1992.   Chapter 7 - Methods of
assessing responses  of trees,  stands,  and
ecosystems to air  pollution.  In: Binkley  DB,
Olson RL, and Bohm M, Pollution impacts on
forest ecosystems  in western United States.
Sringer-Verlag, New York, In Press.  65 pps.

U. S. Geological Survey. 1970. The National
Atlas of the United States of America. U. S.
Geological Survey.  Washington.

Wheeler,  J.H.,  J.T.  Kostbade, and  R.S.
Thoman. 1969. Regional Geography of  the
World.  Holt, Rinehart, and Winston. New York.
                                         9-3

-------
Appendix A.  Statistical Design
D. L.  Cassell

Design Objectives

        The design objectives can be stated as
follows. First, we wish to be able to statistically
estimate the current status and extent, as well
as changes and trends  in,  indicators  of the
condition of the  nation's forest ecosystems.
Furthermore, this is to be done on a regional
and  a national  basis, with  known  statistical
confidence in our estimates.  Second, we want
to monitor indicators of pollutant exposures and
habitat  condition,  and   seek  associations
between human-induced stresses  and  the
ecological condition of the forests. Third, we
want  to be able to  provide yearly  statistical
summaries and periodic interpretive reports on
ecological  status  and  trends  to  resource
managers and the public.

        Specific design criteria were essential
to achieve these objectives. There had to be a
consistent  and realistic representation of all
ecological resources and environmental entities
through the use of probability samples.  The
design had to be flexible to accommodate post-
stratification  and  aggregation  for  many
alternative  subpopulations such  as  states or
other areas of special interest.  There had to
be a  provision  for responding quickly to  new
questions or issues.  The spatial distribution of
the sample of any  resource needed  to be
arranged   according  to   the  population
distribution of the resource. Periodic revisiting
of all sampling sites was also required.

        These requirements led directly to the
EMAP/FHM design (Overton et al. 1990). This
design has been peer reviewed by a panel of
the American   Statistical  Association.    The
design has also been extensively reviewed by
Forest Service  and EPA statisticians, among
others.

The  Monitoring Grid

        The design uses a triangular grid so
that  the  nation   is  tesselated  (tiled)   with
hexagons,  each covering approximately 635
km2.  Within each 635 km2 hexagon, another
hexagon of approximately 40 km2 represents a
1-in-16  sample of the resource.  Each such
'40-hex' represents  the EMAP Tier I sample
that will  be characterized using remote-sensing
techniques.  Each 40-hex is uniformly offset
from the center of the larger hexagon. Within
a 40-hex, a point can be located on the ground
where one plot, or a constellation of plots, can
be  physically  established.   On  these plots,
measurements are made to gauge the selected
forest condition and  stress indicators. To give
some idea of the scale of this monitoring, the
distance between ground plot locations is about
27  km.   For the U.S., the grid  yields about
12,600  potential ground plot  locations.   Of
these, about 4,000 are expected to be forested.

        The national  design  provides  for
measurements at four-year intervals.  The plots
measured in any one year are selected so that
the spatial structure of the plots is preserved; in
other words, so that the  samples are spread
uniformly  over time  and   space.    This
arrangement of sites across space and time is
often referred to as an 'interpenetrating' design,
or a rotating panel design.  The design permits
analysis using traditional Horvttz-Thompson
procedures  (Horvitz  and Thompson  1952;
Overton et al.  1990).

Benefits  of the Triangular Grid

        The triangular grid is systematic with a
random  start.    The  centers  of the larger
hexagons have been randomly  located  by
selecting a single random point in space and
moving  the entire pattern so that an arbitrary
hexagon center is  on that  point.  Then, a
random  point  is located within that hexagon,
and the entire systematic grid of smaller 40-
hexes is fixed by centering one of them on that
point. The triangular grid is spatially compact,
provides uniform spatial coverage, and is very
flexible.  The grid density can be altered easily,
without  destroying  the overall temporal and
spatial structure of the  measurements.   In
addition, a triangular grid is less  likely than a
square  grid to coincide with artificially  linear
features such as state boundaries.
                                          A-1

-------
        In forestry,  regular spacing of sample
points generally leads to population estimates
with smaller sampling  variances than  simple
random  sampling.    The  uniform   spatial
coverage  means  that  for a given density,
resources in the landscape  are sampled in
proportion to their abundance, area, and spatial
pattern. Use of an interpenetrating sample and
a fixed remeasurement cycle also means that
this is achieved every year. The triangular grid
allows a wide variety of grid magnifications and
reductions,  and it  can  handle special cases
such as subpopulations of special interest.

Evaluating Status and Change

       There  are  several  key  issues  in
monitoring   across  both  time  and   space.
Current status is best estimated by including as
many population units as  possible, because
greater coverage enhances the identification of
subtle subpopulation differences.  But detection
of  trends   is   best  done   with  repeated
observations of the same units over time, with
the interval  between observations chosen on
the basis of the signal-to-noise ratio  of  the
measurements.    Observer  effects  can  be
important; too long  a time between visits may
mean  that  field measurements  cannot  be
properly calibrated,  while too short a time may
lead to unwanted impacts such as trampling of
the understory.

       To  answer  questions  about   the
condition of  forest ecosystems, it is necessary
to specify explicitly what set of forests will be
analyzed.   This  set   becomes  the  target
subpopulation,  the areal extent of  forested
ecosystem about which estimates of conditions
will be made.  Target subpopulations can be
defined for any  region or attribute. They serve
two main purposes: 1) to increase the precision
of condition  and trend estimates  by controlling
extraneous variation, and 2) targeting  specific
sets of resources for reporting and assessment.

Stratification

       The sample design currently used does
not employ stratification. This topic continues
to be debated because stratification is  usually
very useful for reducing sample variances, but
no  clear choice of stratification method has
emerged from  the debate.   One argument
against stratification is that if it is based  on
current  resources,  then distinctly non-optimal
sample designs would be forced  upon any
future  study of the  same forest  resource
because forests normally change over time.  In
many cases, insufficient data are available  to
perform the proposed stratifications  at the
desired scale, and classification errors as small
as 20 percent would erase any potential gains
in precision.   It is still  an  open question
whether stratification can be applied effectively
at larger scales in the immediate vicinity  of
ground plots.

Selection of Ground Plots

       The   field   sampling   involves  the
collection of EMAP Tier II  measurements that
will be  used to calculate values  for forest
condition indicators.  One key feature is the
statistical selection  of ground plots so that the
data  represent  a   probability  sample.    in
comparison with current and well-known Forest
Service Forest  Inventory and Analysis (FIA)
procedures,  the FHM method of Tier II site
selection is equivalent to selection of a single
FIA  photo point (i.e., stage one of  the FIA
multi-stage sample). This permits a linkage to
the FIA statistical design.  The FIA photo point
grid for a region is overlaid on a 40-hex.  Then
the closest  FIA  photo point is  selected.   In
cases  where  the  selected  photo point  is
already an FIA plot, the FHM plot will be offset
from the photo point. This procedure was used
in all but one of the states implemented  in
1991.

        Because the selected photo point is the
one closest  to the  center  of the 40-hex, and
because the photo point  grid  is statistically
independent of the EMAP/FHM  hexagon grid,
the probability of selecting that particular photo
point is proportional to the area that photo point
represents in  the  FIA sample design.   To
estimate this probability, the coordinates of the
photo  points around each hex  center are
entered into a data base and  plotted on an
equal-area projection map.  Then,  for the
                                              A-2

-------
selected photo  point,  a geographic  analysis
system computes the area of ail land closer to
this  point than  any other photo point.  The
larger this area, the more likely it is that the
hex  center  would fall  in  this area.   The
inclusion  probabilities are  calculated  as  a
constant multiple of this area.

        Thus,   inclusion  probabilities   are
proportional to the area that the FIA photo point
represents in the photo point grid. This means
that  the inclusion probabilities can be used to
provide population estimates for all  of  our
variables  of interest.   In 1991,  it  was  not
feasible to complete the necessary analysis,
and  some  inclusion  probabilities are  only
approximately correct.  The additional variance
component  due to the approximation is  not
included in the statistical estimation procedures
used in this report.  Strictly  correct inclusion
probabilities will be available for the next report.

        Another approach is  needed for  the
regions  where  FIA  photo  points  are  not
available. In these regions, the ground plot can
be located at a random offset and azimuth from
the center of the  40-hex. These points  are
then added to the list of photo points that are
interpreted by the FIA process. This 'reversed'
procedure was used in the one  state that did
not  start  with  the  FIA  photo  points  as  a
sampling basis.  Since the same probabilistic
procedure was performed within each 40-hex,
all ground plots in that state  have the same
inclusion probability.

Ground Plot Structure

        Another key  feature  of the Tier  II
measurements is the flexible design of ground
plots for multiple categories of measurements.
The  plot  designs  used  in EMAP/FHM have
been developed jointly by the Forest Service
and  the EPA, based  on an optimally cost-
effective  plot design  for FIA  mensuration
measurements (Scott 1991).  A  circular, one-
hectare plot contains four fixed-radius (7.32
meters) subplots for field measurements. The
subplots are arranged in an equilateral triangle
with  an additional subplot at the center of the
triangle. The centers of the outer subplots are
36.6  meters from the center of the central
subplot.

       This basic  plot  design  has  been
evaluated for many  indicators (Riitters  et al.
1991) using sampling theory (Cochran 1977) to
estimate optimal  numbers  of plots, subplots,
and observations within subplots. The results
suggested  that the plot design was adequate
for current  FHM indicators,  in the areas where
FHM has been implemented. Furthermore, the
indicators evaluated on the  FHM  plots are thus
implicitly defined as being sampled over a
specific area of forested land.

       Fixed-radius   plots  have   several
advantages over the variable-radius plots that
are used often in forest inventory.  Variable-
radius plots are more efficient for measuring
the current status of characteristics that  are
correlated with tree size. But fixed-radius plots
are generally  more  efficient  for  measuring
quantities that are not correlated with tree size,
which is the case for most FHM indicators.
Another advantage is that change over time is
more easily estimated by using fixed-area
plots.

       A subset of the  total  ground plot is
sometimes used for a given measurement.  In
those cases, a factor was calculated to  show
what  proportion of the plot was used in  the
analysis. This provided an expansion factor so
that all variables could be summarized on an
equal-area  basis.  Since only areal estimates
were  computed, this was combined with  the
inclusion probability  for the plot to give an
inclusion probability  for the subset.  Another
alternative  is to use multiple-stage sampling
procedures.

       For all  estimates computed  over  the
forested  portion of  the plot,  the expansion
factor was the inverse of the proportion of plot
in forested conditions. For estimates computed
over specific forest type groupings or crown
groups, the expansion factor was  the inverse of
the proportion of the plot in that  classification.
For estimates based on individual tree species,
the expansion  factor was  the inverse of  the
proportion   of  plot  in  forested  conditions,
multiplied by the  inverse of the proportion of
                                           A-3

-------
 the plot's basal area covered by the specified
 species.

 Statistical Procedures for this Report

        In  this report,  cumulative distribution
 functions (CDFs) were generated for the data
 collected at  the ground plot locations.  The
 estimated CDF at a value of, for example x,
 equals the estimated proportion of  the areal
 extent of the real resource being examined that
 has  values  less than or equal to x.  The
 Horvitz-Thompson estimate of the CDF at x is
 given by:
                     gU
 (Cochran 1977) and approximate joint inclusion
 probabilities (Stevens et al. in review).  The
 approximation to ^ is given by:

The approximate  90%  confidence  bounds
about each CDF were calculated by assuming
that the CDF estimates approximated normal
distributions.   The  confidence  regions were
obtained as the estimated CDF value, plus or
minus 1.645 times the estimated standard error
of the  CDF value.  The standard error was
computed as the square root of the estimated
variance  of  the  CDF  value,   where  the
estimated variance was calculated by:
                                                                         (-
                                                                             Tlj
where
Vi = 1 if the variable is not greater than x, and
    0 otherwise
it,  = inclusion probability for plot i
n  = number of plots in this subset of the data.

For example, in the CDF of basal area for the
spruce-fir forest type group in the New England
area, the height of the CDF at x = 20 m2ha"1 is
the  estimated   percent  of  the  underlying
population of spruce-fir forests in New England
with basal area less than 20 m2ha"1. The CDF
rises  monotonically to 100% at the maximum
observed value of x.

       The  estimated  variance  of  this
percentage is based  on both the first-order
inclusion probability (it,) and the joint inclusion
probability  (jt,j).   The  first-order  inclusion
probability  was described earlier.  The joint
inclusion probability is the probability that both
sites  / and j  are selected for inclusion in the
sample.  The joint inclusion probabilities must
be  non-zero to  obtain  unbiased  variance
estimates.  In the systematic sampling design
of   EMAP/FHM,   some  joint   inclusion
probabilities are zero  or even unknown.  An
approximate variance estimate was obtained by
using the  Yates-Grundy  variance  estimate
where
               A  =  V —
In application, the variance estimates assumed
a fixed sample size within  each analyzed
subset. The analyzed subset may have been
the region  as a whole,  the  area of a major
forest type grouping within a region, or the area
within  a region covered by  a particular tree
species.  For such subsets, the sample size is
a random variable depending on the position of
the sampling  grid  and  the  selection  of  the
major  forest type  grouping  or tree species.
This variance component  has  not  been
incorporated into the estimation procedures.
                                             A-4

-------
Appendix    B.        Quality
Assurance Program  for Forest
Health Monitoring for 1991

G. E. Byers and C. J. Palmer
Quality Assurance Within EMAP

       The potential temporal and geographic
scope of EMAP is immense, compared to most
environmental   monitoring   programs.
Information from EMAP will be used  in risk
assessments  that  are  integrated   across
ecological  resource  types.    To  provide
information that is of sufficient quality (and thus
usefulness)   to   address  the   program's
objectives, a  comprehensive and  integrated
quality assurance (QA) program is required for
both EMAP as  a  whole and for the Forests
component of EMAP.

       EMAP participates in the Environmental
Protection  Agency's  mandatory   quality
assurance program (Stanley and Verner 1985).
The overall policies, organization objectives,
and functional responsibilities associated with
the EMAP QA program are documented in the
quality  assurance  program  plan (QAPP)
(Einhaus et al. in preparation). The part of the
QA program that is implemented by the Forest
Health   Monitoring   (FHM)   program   is
documented by  an annual quality assurance
project plan (QAPjP).  It describes the plan for
achieving data quality objectives established for
all data acquisition activities. It also serves as
a reference for all QA activities within FHM.

The  1991   FHM  Quality Assurance
Project Plan

       The  purpose  of  the  QAPjP is   to
describe the organization and objectives of the
FHM program and the QA  activities needed to
achieve certain data quality requirements. The
data  quality  requirements  for many  FHM
measurements  cannot be  specified with
experience using  those  measurements  for
assessments.  During the  formative stage  of
FHM, the QAPjP is being used to assess and
control measurement errors that may enter the
system during the field measurement phase,
sample preparation, and sample analysis.  The
QAPjP attempted to identify all environmental
measurements that are being made and to
identify  specific  processes  within   each
measurement that could  introduce possible
sources of error or uncertainty in the resulting
data.  Methods, materials, and schedules for
assessing  the  error contributed  by  each
process were  also be  addressed in some
cases.  The  1991  QAPjP  also described the
QA activities and assessment criteria that were
implemented to ensure that the data bases met
all data quality objectives (DQOs) that  had
been established,  using defined  criteria  and
procedures for assessing statistical control for
each measurement parameter in many cases.
Because of the complexity and variety of data
acquisition activities, the  organization of the
QAPjP was modified from  that recommended
by  EPA's Quality  Assurance Management
Staff.

       Table   B-1  summarizes  the  14
measurements,  including  mensuration  and
visual damage, for which data were collected
during   the   summer  of   1991.     The
implementation  of  mensuration  and  visual
damage measurements have been designated
the responsibility of specific indicator leaders.
There is close  cooperation between indicator
leaders, the indicator development coordinator,
the QA coordinator for EMAP-Forests, and the
logistics coordinators.   Following a format
provided by the coordinators (Table B-2), each
indicator  leader specified  appropriate  QA
procedures for their measurements.

       The  standard operating  procedures
(SOPs) for each indicator are not  included in
the 1991  QAPjP, but are printed in separate
methods manuals.  The QAPjP will be revised
as necessary to reflect changes in procedures.
To  ensure proper interactions  among  the
various data acquisition  and management
components of  FHM, all project personnel are
familiar with the policies and objectives outlined
in the QAPjP and other FHM documents.
                                        B-1

-------
Table B-1.  FHM measurements subject to 1991 QA requirements.
Mensuration Data
Visual Damage Indicator
Soil Productivity Indicator
Foliar Chemistry Indicator
Elemental Analysis of Tree Cores Indicator
Evaluation of Tree Branch Damage
Evaluation of Root Disease
Photosynthetically Active Radiation Indicator
Vegetation Structure Indicator
Root Sampling for Evaluation of Root Disease and Mycorrhizae
Wildlife Habitat and Population Measurements
Tree Height
High Resolution Areal Photography
Global Positioning System
Methods   Manuals   for   Standard
Operating Procedures

       Forest Health Monitoring utilizes SOPs
which are also called "methods" or "protocols"
(ConMing  and Byers 1991;  Byers  and Van
Remortel   1991;   and   Chojnacky  1991).
Environmental monitoring SOPs are devised for
sampling and analysis, data management, QA,
reporting activities, accounting, project finance
and contracts, and  in analysis and integration
phases of the project. The use of written SOPs
helps to  ensure  consistency in  planning,
implementation, and analysis activities over
time and among personnel for routine activities
within an  organizational unit.   To  ensure
consistency of data among the FHM program
measurements,  SOPs must be cooperatively
developed.

       The FHM managers are responsible for
determining which activities require SOPs and
for ensuring that they are developed,  reviewed,
and  implemented.   The  research personnel
closest  to  the  actual implementation  of an
activity  (indicator leaders)  develop specific
SOPs. The QA coordinator has responsibilities
in SOP identification, consistency, evaluation of
method or protocol status, and training.
       Due  to the many and varied data
gathering activities and the existence of current
Forest Service manuals for  mensurational
methods for FHM 1991, the program managers
deemed  it appropriate  to produce separate
documents  for methods manuals.    These
documents  are an extension  of the QAPjP.
This  separate   document    approach  is
appropriate because of the needs for  stand-
alone methods manuals for field measurements
and laboratory analyses, and because of the
sheer amount of all required documents.

Quality Assurance Responsibilities
in 1991 FHM Program

       The  success of  the  FHM  program
depends  on the willingness of all participating
agencies to cooperate as full partners  in QA.
The   roles  and  responsibilities  have  been
identified  to   encourage  cooperation and
successful implementation.   In general, the
EPA  is  responsible  for  preparing  planning
documents. Field activities are coordinated by
the Forest Service. Evaluation of the results is
a shared activity.  The individuals  who are
most responsible for the success of QA in FHM
are the  indicator  leaders, regardless  of the
agency from which they come. Table QA-3
lists persons  responsible for QA activities in
1991.
                                             B-2

-------
Table B-2.  Format for quality assurance for implementation measurement documentation for QAPjP.


Field Operations Plan
       Plot locations
       Personnel (field crew requirements, etc.)
       Measurement activities
              Description of measurement site
              Description of measurement procedure
              Table of parameters and measurement techniques
              Data entry and reporting
                      Documents available
                      Data qualifiers
                      Reporting units
                      Evaluation of measurement quality data
                      Communications
              Data custody
       Documentation
QA Operations Plan
       Training
       Batch analysis
       Independent measurement quality samples (test trees, etc.)
       Measurement error samples (control duplicates, reference samples, calibration check samples,
              reagent blanks, detection check samples)
       Sources of measurement error (field spatial variability, temporal variability, between-crew, within-
              crew, wrthin-site)
       Measurement quality objectives (MQOs) for DPARCC (detectability, precision, accuracy,
              representativeness, completeness, comparability)
       Audits (on-site crew or tab audits)
QA Implementation Plan
       Training
       Control of data quality
              Field sampling/measurement and characterization (PARCC)
              On-site systems and crew audits
       Data verification - describing, defining, implementing
              Field measurement
Data Quality Assessment and Reporting Plan
       DPARCC
       Validation procedures
       QA reports to management
              Status reporting
              Formal reports
References
                                             B-3

-------
Table B-3.  Personnel Responsibilities in 1991 FHM Monitoring QA Activities.
Strategy Document
Study Plan
QA Program Plan
QA Project Plan
Methods Manuals:
 Field Research
 Sample Preparation and Analysis
 Eastern FHM Monitoring
Plot Reconnaissance:
       New England, Mid-Atlantic
       South and Southeast
Pretraining Workshops:
       New England
       Mid-Atlantic
       South and Southeast
Training Workshops:
       New England
       Mid Atlantic
       South and Southeast
Audit Crews:
       New England, Mid-Atlantic
       Georgia
       Virginia
       Alabama
Indicator Leaders:
       Mensuration
       Visual Damage
Design
Information Management
Other Forest Service Coordinators
C. Palmer
C. Palmer, J. Barnard
R. Graves
G. Byers

B. Conkling, G. Byers
G. Byers, R. Van Remortel
D. Chojnacky

W. Burkman
W. Bechtold

I. Millers. M. Miller-Weeks
W. Jackson, D. Twardus
W. Anderson, W. Hoffard

W. Burkman
W. Burkman
S. Alexander, R. Kucera, M. Baldwin

W. Burkman
W. Bechtoid, N. Snyder
J. Pemberton
D. May

W. Burkman, I. Millers
S. Alexander, W. Hoffard, R. Anderson
J. Hazard, D.Cassell
C. Liff
C. Eagar, N. Cost, J. Hyland, R. Brooks,
R. Beltz, V. Few, C. Scott
Data Quality Objectives (DQOs)

       The   EPA's   Quality  Assurance
Management Staff has developed a process for
establishing DQOs that can be applied to the
FHM QA program. Modifications to the process
are expected to be developed over time. At
present, DQOs are considered to be specific
statements of  the level of uncertainty a data
user  (presently  defined  as  the respective
indicator leaders) is willing to accept in a body
of environmental data, with respect to the kind
of scientific or policy question that motivated
the  data   collection  activity.   Data  Quality
             Objectives  are  definitive,  quantitative   or
             qualitative statements developed jointly by data
             users (e.g., scientists,  policy makers, interest
             groups) in conjunction with the QA staff.

                    The  DQO process  is an  iterative
             approach   that   balances   costs   versus
             uncertainty to achieve a desired or acceptable
             level of quality. This information can also be
             used   to  allocate  resources to   specific
             monitoring phases in order to generate data of
             sufficient quality  to  support  management
             decisions   or  answer  specific   scientific
             questions.
                                             B-4

-------
        Data quality, and therefore DQOs, may
be  defined  for several  levels of  FHM  data
collection. The first level is measurement-level
DQOs  (MQOs)  for  specific  measurement
parameters, estimated using existing or initial
baseline data.  These  DQOs may  define
acceptance  criteria for detectivity, precision,
accuracy, representativeness,  completeness,
and comparability  in  field and   laboratory
measurement data (Byers et al. 1990). Beyond
this, another  criterion  may be to  optimize
measurement uncertainty with respect to  non-
measurement sources of uncertainty (e.g., due
to sampling design constraints or naturally-
occurring spatial and temporal variability that
often is confounded within environmental data).

        Other levels are recognized at a higher
ecosystem  level  but they have not yet been
addressed seriously.  They are Indicator-level
DQOs  (IQOs)  derived  from  aggregated
parameter   data  for  ecological  indicators:
Resource-level DQOs (RQOs) derived from
aggregated  indicator  data  for the  EMAP-
Forests Resource Group, and; Ecosystem-level
DQOs (EQOs) from aggregated resource  data
for overall ecosystem assessments.

        Throughout the DQO-setting process,
there   should  be  communication  among
program management, policy-makers, program
coordinators, resource scientists, data analysts,
and scientists involved  in  the actual data
collection  activities.     Acceptance  criteria
established  during  the  DQO  development
process serve as  benchmarks for satisfying
data user requirements.   In FHM,  DQOs are
being established for  several levels of data
collection, e.g., sample measurement system,
measurement parameter, or indicator level for
various  indicators.   Included  in  the DQO
assessment  are  four  quantitative attributes:
detectability,  precision,   accuracy,   and
completeness.    Also  included   are  two
qualitative attributes; representativeness,  and
comparability.
Field   QA    for   Measurements
Implemented in 1991

       Several QA issues were of concern for
the 1991 field measurements of  mensuration
and visual damage. Their resolution for 1991
was based partly on logistical and budgetary
constraints.  The issues were:

• How  to  assess field   crew   precision,
  accuracy,   and   comparability  within a
  region.

• How  to   assess field  crew  precision,
  accuracy,   and   comparability  between
  regions.

• How to assess  trainer  crew  precision,
  accuracy,   and   comparability  between
  regions.

• What are  the  logistical  constraints using
  different crews  on  the  application ofQA
  techniques in  the  relatively  large  Forest
  Service regions, such as the Northeastern
  region and  Southeastern/Southern  regions
  of  the U.S.
• How beneficial and cost-effective will be
  the  different possible QA techniques.

Accuracy

1.      Two-person field crews were provided
       by  the  Forest  Service or  by  state
       forestry authorities.    An  adequate
       number of field  crews were hired to
       ensure   that   the   planned   1991
       measurement  program   was
       accomplished in the available time.

2.      Reference plots (also called 'standard'
       or 'accuracy' plots) were established at
       training sites  in  Durham,  NH,  and
       Asheville,  NC.  All trainers  measured
       the reference plots to establish 'true'
       values. The field crews also measured
       the reference plots  as part of the
       training program, and were judged
       under the same standard.
                                         B-5

-------
3.      Specific test and training trees  were
        used for training and testing the field
        crews'  abilities  to  make the  visual
        crown measurements. An expert team
        selected  approximately  40  trees  at
        each training site that reflected a broad
        range of visual damage conditions
        found in the forests.  The field crews
        were trained  on  these  trees  after
        classroom instruction.  Then the field
        crews were tested on another set of 20
        test trees, both individually as  crew
        members and  as two-man crews.

4.      Expert  trainers  were  used  to  form
        trainer crews for each of the regions
        for 1991.     All measurement  data
        used as standards on  the  reference
        plots were  established  by   these
        Forest   Service personnel  who are
        trained     and  experienced   in the
        measurement  techniques.   The four
        trainers  for   the   regions   crossed
        over  to each other's regions during
        pre-training  sessions to establish the
        national      standards   for  the
        measurements.

5.      Two-person audit crews visited field
        plots randomly during  measurements
        to observe protocol  application and to
        provide  real-time input  to  the field
        crews  to reduce  inaccuracy in the
        measurements.    The  audit  crews
        worked with the field crew,  assessed
        its  performance,   and  discussed
        deviations from protocol  when they
        occurred. The audit crews performed
        these audits about 1 to 4 weeks after
        the start of data  collection.

Precision

        Each field crew  remeasured  two plots
that it had previously measured.  These plots
were selected about 1/3 and 2/3 through the
data  collection  period.    This   permitted
estimation of within-crew variability.  Between-
crew  variability  was  assessed  from  the
reference plot measurements.
Comparability

        Each field crew collected data from the
reference plots at training in June and also at
a debriefing session in October to establish a
basis for comparability among crews within a
region.  Inter-regional comparability  is also of
great concern, and a technique was developed
in   1991,  but  it  was  not deemed proper
(financially, logistically, and legally) for all field
crews to collect data  inter-regionally, nor to
travel  to  inter-regional  training   sessions.
However, two field crews  from the  northeast
region were able attend the debriefing session
in  Asheville, NC, in October.  The southeast
and the northeast trainers  both measured the
reference   plots   (mensuration  and  visual
measurements) to reestablish accuracy data for
the plots that might  have  changed from the
pretraining   sessions    and  to  compare
measurements among trainers.    Detailed
results are  not yet available due to limited QA
resources,  but preliminary  analyses suggest
acceptable  within-crew  and   between-crew
precision.

Representativeness,  Completeness,
and Detectibility

        Representativeness   criteria   are
established by the plot design and plot sample
selection established for 1991. Completeness
and detectivity criteria have been established
for  some   of  the   components   of  the
implementation  indicators.     Measurement
parameters  for  some indicators were very
qualitative,  e.g., merely presence or absence
with no  detectivity criteria.

Data Verification

        Verification is the  act of determining
and controlling the quality of data. Verification
can be accomplished  manually, electronically,
or through remeasurements.   A systematic
approach to data verification ensures that all
data  are  subjected   to basic standards of
accuracy that verify the authenticity,  but not
necessarily   the  validity,   of  the  data.
Verification is accomplished by comparing data
at each  level of processing to established data
quality criteria. Most of the data collected were
                                          B-6

-------
entered into portable data recorders in the field.
Computer programs performed real-time logic
checks   of   most   entries,   automatically
determining if entries  were valid and logically
correct.

The  1991  FHM  Quality  Assurance
Report

       A  summary  of  quality  assurance
information collected  during the  1991 field
season is currently being prepared.  It will be
available for distribution in the late fall, 1992.
The  report will provide  information  on data
quality attributes for the indicators included in
this annual statistical summary as well  as
indicators  undergoing testing  in  pilot  and
demonstration studies.
                                              B-7

-------
Appendix C.   Nomenclature of Species in  Text and Forest Type and
Species Codes
Tree Species Nomenclature

American basswood
American beech
American chestnut
American elm
American holly
American hornbeam
American mountain-ash
American sycamore
apple
ash
Atlantic white-cedar
Austrian pine
baldcypress
balsam fir
balsam poplar
bear oak
bigleaf magnolia
bigtooth aspen
birch
bitternut hickory
black ash
black cherry
black locust
black oak
black spruce
black walnut
black willow
blackgum
blackjack oak
blue ash
blue spruce
bluebeech
bluejack oak
boxelder
brown ash
buckeye
butternut
cabbage palm
Carolina hemlock
catalpa
cherry
cherrybark oak
chestnut oak
Chinaberry
chinkapin oak
Tilia americana L.
Fagus grandifolia Ehrh.
Castanea dentata (Marsh.) Borkh.
Ulmus americana L.
Ilex opaca Ait.
Carpinus caroliniana Walt.
Sorbus americana Marsh.
Platanus occidentalis L.
Malus spp.
Fraxinus spp.
Chamaecyparis thyoides (L.) B.S.P.
Pinus nigra Arnold
Taxodium distichum (L.) Rich.
Abies balsamea (L.) Mill.
Populus balsamifera L.
Quercus ilicifolia Wangenh.
Magnolia macrophylla Michx.
Populus grandidentata Michx.
Betula spp.
Carya cordiformis (Wangenh.) K. Koch
Fraxinus nigra Marsh.
Prunus serotina Ehrh.
Robinia pseudoacacia L.
Quercus velutina Lam.
Picea mariana (Mill.) B.S.P.
Juglans  nigra L.
Salix nigra Marsh.
Nyssa sylvatica Marsh.
Quercus marilandica Muenchh.
Fraxinus quadrangulata Michx.
Picea pungens Engelm.
Carpinus caroliniana Walt.
Quercus incana Bartr.
Acer negundo L.
Fraxinus nigra Marsh.
Aesculus spp.
Juglans  cinerea L.
Sabal palmetto (Walt.) Lodd.
Tsuga caroliniana Engelm.
Catalpa spp.
Prunus spp.
Quercus falcata var. pagodaefolia Ell.
Quercus prinus L.
Melia  azedarach L.
Quercus muehlenbergii Engelm.
                                            C-1

-------
choke cherry (common)
cucumbertree
dwarf live oak
dwarf post oak
eastern cottonwood
eastern hemlock
eastern hophombeam
eastern larch
eastern redbud
eastern redcedar
eastern white pine
elm
European larch
Florida maple
flowering dogwood
Fraser fir
giant sequoia
gray birch
green ash
hackberry
hackberry (common)
hawthorn
hemlock
hickory
honey locust
jack pine
juniper
larch (introduced)
laurel oak
live oak
loblolly pine
longleaf pine
lowland blackgum
magnolia
maple
mockernut hickory
mountain maple
mulberry
northern pin oak
northern red oak
northern white-cedar
Norway spruce
Ohio buckeye
osage-orange
overcup oak
paper birch
Paulownia
pawpaw
pecan
persimmon
pignut hickory
pin cherry
Prunus virginiana L.
Magnolia acuminata L.
Ouercus virginiana Mill.
Ouercus stellata var. margaretta
Populus deltoides Bartr.
Tsuga canadensis (L.) Carr.
Ostrya virginiana (Mill.) K. Koch
Lam laricina (Du Roi) K.  Koch
Cercis canadensis L.
Juniperus virginiana L.
Pinus strobus L.
Ulmus spp.
Larix decidua Mill.
Acer barbatum Michx.
Cornus florida L.
Abies fraseri
Sequoia gigantea (Lindl.)  Decne.
Betula populifolia Marsh.
Fraxinus pennsylvanica Marsh.
Celtis spp.
Celtis occidental L.
Crataegus spp.
Tsuga spp.
Carya spp.
Gledftsia triacanthos L.
Pinus banksiana Lamb.
Juniperus  spp.
Larix decidua Mill.
Quercus laurifolia Michx.
Quercus virginiana Mill.
Pinus taeda L.
Pinus palustris Mill.
Nyssa sylvatica var. biflora (Walt.) Sarg.
Magnolia spp.
Acer spp.
Carya tomentosa Nutt.
Acer spicatum Lam.
Morus spp.
Quercus ellipsoidalis E. J. Hill
Quercus rubra L.
Thuja occidentalis L.
Picea abies (L.) Karst.
Aesculus glabra Willd.
Madura pomifera (Raf.) Schneid.
Quercus lyrata Walt.
Betula papyrifera Marsh.
Paulownia tomentosa (Thunb.) Steud.
Asimina tritoba (L.) Dunal
Carya illinoensis (Wangenh.) K. Koch
Diospyros virginiana L.
Carya glabra (Mill.) Sweet
Prunus pensylvanica L.
                                              C-2

-------
pin oak
pinyon
pitch pine
pondcy press
pond pine
post oak
prunus
quaking aspen
red maple
red mulberry
red pine
red spruce
redbay
redwood
river birch
rock elm
sassafras
scarlet oak
Scots pine
scrub oak (several)
serviceberry
shagbark hickory
shingle oak
shortleaf pine
Shumard oak
Sitka spruce
silver maple
silverbell
slash  pine
slippery elm
sourwood
southern magnolia
southern red oak
sparkleberry
spruce pine
striped maple
sugar maple
swamp chestnut oak
swamp tupelo
swamp white oak
sweet birch
sweetbay
sweetgum
table-mountain pine
tamarack
tulip-poplar
turkey oak
upland blackgum
Virginia pine
Quercus palustris Muenchh.
Pinus cembroides Zucc.
P. edulis Engelm.
P. monophylla Torr. and Frem.
P. quadrifolia Part.
Pinus rigida Mill.
Taxodium distichum var. nutans (Ait.) Sweet
Pinus serotina Michx.
Quercus stellata Wangenh.
Prunus spp.
Populus tremuloides Michx.
Acer rubrum L.
Moms rubra L.
Pinus resinosa Ait.
Picea rubens Sarg.
Persea borbonia (L.) Spreng.
Sequoia sempervirens (D. Don)  Endl.
Betula nigra L.
Ulmus thomasii Sarg.
Sassafras albidum (Nutt.) Nees
Quercus coccinea Muenchh.
Pinus sylvestris L.
Quercus spp.
Amelanchier spp.
Carya ovata (Mill.) K. Koch
Quercus imbricaria Michx.
Pinus echinata Mill.
Quercus shumardii Buckl.
Picea sitchensis (Bong.) Carr.
Acer saccharinum L.
Halesia spp.
Pinus elliottii Engelm.
Ulmus rubra Muhl.
Oxydendrum arboreum (L.) DC.
Magnolia grandiflora L.
Quercus falcata Michx.
Vaccinium arboreum Marsh.
Pinus glabra Walt.
Acer pennsylvanicum L.
Acer saccharum Marsh.
Quercus michauxii Nutt.
Nyssa sylvatica var. biflora (Walt.) Sarg.
Quercus bicolor (Willd.)
Betula lenta L.
Magnolia virginiana L
Liquidambar styraciflua L
Pinus pungens Lamb
Larix laricina (Du Roi) K. Koch
Liriodendron tulipifera L.
Quercus laevis Walt.
Nyssa sylvatica Marsh.
Pinus virginiana Mill.
                                             C-3

-------
water hickory                  Carya aquatica (Michx. f.) Nutt.
water oak                     Quercus nigra L.
water tupeto                   Nyssa aquatica L
white ash                     Fraxinus americana L
white basswood               Tilia heterophylla Vent.
white oak                     Quercus alba L.
white pine                     Pinus strobus L
white spruce                   Picea glauca (Moench) Voss
willow                        Salix spp.
willow oak                     Quercus phellos I.
winged elm                   Ulmus alata Michx.
yellow birch                   Betula alleghaniensis Britton
yellow buckeye                Aesculus octandra Marsh.
yellow-poplar                  liriodendron tulipifera L.
yellowwood                   Cladrastis lutea (Michx. f.) K. Koch
                                              C-4

-------
Insects Nomenclature
balsam woolly adelgid
beech scale
black twig borer
birch casebearer
birch leafminers
birch skeletonizer
bronze birch borer
buck moth
cherry scallop shell moth
conifer swift moth
eastern larch beetle
eastern spruce budworm
eastern tent caterpillar
elm spanworm
elongated hemlock scale
fall cankerworm
fall webworm
forest tent caterpillar
fruittree leafroller
gypsy moth
hemlock looper
jack  pine budworm
jack  pine sawfly
maple leafcutter
maple trumpet skeletonizer
oak leaftier
oak skeletonizer
orangestriped oakworm
pear thrips
periodical  cicada
pine engraver beetles
pitch pine  looper
redheaded jackpine sawfly
red pine adelgid
red pine beetle
red pine scale
saddled prominent
southern pine beetle
spruce beetle
variable oak leaf caterpillar
white pine weevil
Adelges piceae
Cryptococcus fagisuga
Xylosandais compactus
Coleophora serratella
Fenusa pusilla, and Messa nana
Bucculatrix canadensisella
Agrilus anxius
Hemileucia maia
Hydria prunivorata
Korscheltellus gracilis
Dendroctonus simplex
Choristoneura fumiferana
Malacosoma americanum
Ennomos subsignarius
Fiorinia extema
Alsophila po met aria
Hyphantria cunea
Malacosoma disstria
Archips argyrospila
Lymantria dispar
Lambdina athasaria and L. fiscellaria
Choristoneura pinus
Neodiprion pratti banksianae
Paraclemensia acerifoliella
Epinotia aceriella
Croesia semipurpurana
Bucculatrix spp.
Anisota senatoria
Taeniothrips inconsequens
Magicicada septendecim
Ips spp.
Lambdina pellucidaria, and L. athasaria
Neodiprion rugifrons
Pineus borneri
Ips pini
Matsucoccus resinosae
Heterocampa guttivitta
Dendroctonus frontalis
Dendroctonus rufipennis
Heterocampo manteo
Pissodes strobi
                                              C-5

-------
Diseases Nomenclature

annosus root disease          Heterobasidium anno sum
anthracnose                  Gloeosporium spp., Discula  spp., Apiognomonia venta, and Gnomonia
                             spp.
armillaria root rot              Armillaria spp.
butternut canker              Sirrococcus clavigignenta-juglandacearum
cytospora canker              Valsa kunzei (Cytospora kunzei)
dogwood anthracnose          Discula destrucliva
Dutch elm disease            Ceratocystis ulmi
European larch canker         Lachnellula wilkommii
fusiform rust                  Cronartium quercuum f. sp. fusiforme
hypoxyton canker             Hypoxyton mammatum
sapstreak                    Ceratocystis coerulescens
scteroderris canker            Ascocalyx  abietina (Gremmeniella abietina)
white pine blister rust          Cronartium ribicola
                                             C-6

-------
 Eastern Forest Type Codes


 Code         Definition

 0000          WHITE/RED/JACK PINE GROUP
 0010         Jack pine
 0020         Red pine
 0030         White pine
 0040         White pine/hemlock
 0050         Hemlock
 0060         Scotch pine
 0070         Ponderosa pine

 0100          EASTERN SPRUCE/FIR GROUP
 0110         Balsam fir
 0120         Black spruce
 0130         Red spruce/balsam fir
 0140         Northern white-cedar
 0150         Tamarack (eastern larch)
 0160         White spruce
 0170         Norway spruce
 0180         Larch (introduced)
 0190         Red spruce

 0200          LONGLEAF/SLASH PINE GROUP
 0210         Longleaf pine
 0220         Slash pine

 0300          LOBLOLLY AND SHORTLEAF PINE GROUP
 0310         Loblolly pine
 0320         Shortleaf pine
 0330         Virginia pine
 0340         Sand pine
 0350         Eastern redcedar
 0360         Pond pine
 0370         Spruce pine
 0380         Pitch pine
 0390         Table-mountain pine

0400         OAK/PINE GROUP
 0410         White pine/northern red oak/white ash
 0420         Eastern redcedar/hardwood
 0430         Longleaf pine/scrub oak
 0440         Shortleaf pine/oak
 0450         Virginia pine/southern red oak
 0460         Loblolly pine/hardwood
 0470         Slash pine/hardwood
 0480         Scarlet oak
 0490         Other oak/pine

0500         OAK/HICKORY GROUP

                                          C-7

-------
 0510         Post, black, or bear oak
 0520         Chestnut oak
 0530         White oak/red oak/hickory
 0540         White oak
 0550         Northern red oak
 0560         Yellow-poplar/white oak/northern red oak
 0562         Sweetgum/yellow-popular
 0564         Yellow-popular
 0570         Southern scrub oak
 0580         Black locust
 0590         Mixed central hardwood
 0592         Sassafras/persimmon
 0594         Central hardwoods reverting from field

0600          OAK/GUM/CYPRESS GROUP
 0610         Swamp chestnut oak/cherrybark oak
 0620         Sweetgum/Nuttall oak/willow oak
 0630         Sugarberry/American elm/green ash
 0650         Overcup oak/water hickory
 0660         Atlantic white-cedar
 0670         Bakfcypress/water tupelo
 0680         Sweetbay/swamp tupelo/red maple
 0690         Palm/mangrove/other tropical
 0692         Mangrove
 0694         Palm
 0696         Other tropical

0700          ELM/ASH/RED MAPLE GROUP
 0710         Black ash/American elm/red maple
 0720         River birch/sycamore
 0730         Cottonwood
 0740         Willow
 0750         Sycamore/pecan/American elm

0800          MAPLE/BEECH/BIRCH GROUP
 0810         Sugar maple/beech/yellow birch
 0820         Black cherry
 0830         Black walnut
 0840         Red maple/northern hardwoods
 0850         Red maple/central hardwoods
 0880         Northern hardwood reverting from field
 0890         Mixed northern  hardwoods

0900          ASPEN/BIRCH GROUP
 0910         Aspen
 0920         Paper birch
 0930         Gray birch

0990          INDETERMINANT/NONSTOCKED
 0998         Indeterminant
 0999         Nonstocked
                                          C-8

-------
Tree Species Codes
Code  Common Name
Genus
Species

010
011
012
014
015
016
017
018
019
020
021
022
041
042
043
051
052
060
058
059
061
062
063
064
065
066
067
068
069
070
071
072
073
081
090
091
092
093
094
095
	 OUIIWUUUS 	
fir sp.
Pacific silver fir
balsam fir
bristlecone fir
white fir
Fraser fir
grand fir
corkbark fir
subalpine fir
California red fir
Shasta red fir
noble fir
Port-Orford -cedar
Alaska-cedar
Atlantic white-cedar
Arizona cypress
Baker cypress
Redcedar/Juniper
Pinchot juniper
Red berry juniper
Ashe juniper
California juniper
Alligator juniper
Western juniper
Utah juniper
Rocky Mountain juniper
southern redcedar
eastern redcedar
one seed juniper
larch (introduced)
tamarack (native)
subalpine larch
western larch
incense-cedar
spruce
Norway spruce
Brewer spruce
Engelmann spruce
white spruce
black spruce

Abies
Abies
Abies
Abies
Abies
Abies
Abies
Abies
Abies
Abies
Abies
Abies
Chamaecyparis
Chamaecyparis
Chamaecyparis
Cupressus
Cupressus
Juniperis
Juniperis
Juniperis
Juniperis
Juniperis
Juniperis
Juniperis
Juniperis
Juniperus
Juniperus
Juniperus
Juniperis
Larix
Larix
Larix
Larix
Libocedrus
Picea
Picea
Picea
Picea
Picea
Picea

sp.
amabilis
balsamea
bracteata
concolor
fraseri
grandis
lasiocarpa var.
lasiocarpa
magnifica var.
magnifica var.
procera
lawsoniana
nootkatensis
thyoides
arizonica
bakeri
sp.
pinchotii
erythrocarpa
ashei
California
deppeana
occidentalis
osteosperma
scopulorum .
sillcicola
virginiana
monosperma
sp.
laricina
lyallii
occidentallis
decurrens
sp.
abies
breweriana
engelmannii
glauca
mariana








arizonica

magnflfca
shastensis





























                                      C-9

-------
Code  Common Name
Genus
Species
096
097
098
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
201
202
	 ouuwuuuo 	
blue spruce
red spruce
Sitka spruce
whitebark pine
bristlecone pine
knobcone pine
foxtail pine
jack pine
common pinyon
sand pine
lodgepole pine
coulter pine
shortleat pine
slash pine
Apache pine
limber pine
Mexican white pine
spruce pine
Jeffrey pine
Sugar pine
Chihuahua pine
Western white pine
bishop pine
longleaf pine
ponderosa pine
table mountain pine
monterey pine
red pine
pitch pine
Digger pine
pond pine
eastern white pine
Scotch pine
loblolly pine
Virginia pine
Singleleaf pinyon
Border pinyon
Arizona pine
southwestern white pine
washoe pine
four-leaf pine
Austrian pine
Mexican pinyon pine
bigcone Douglas-fir
Douglas-fir
Picea
Picea
Picea
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pinus
Pseudotsuga
Pseudotsuga
pungens
rubens
sitchensis
albicaulis
aristata
attenuate
batfouriana
banksiana
edulis
clausa
contorta
coulter!
echinata
elliottii
engelmannii
flexilis var. reflexa
flexilis var. reflexa
glabra
Jeffrey!
lambertiana
leiophylla
monticola
muricata
palustris
ponderosa
pungens
radiata
resinosa
rigida
sabiniana
serotina
strobus
sylvestris
taeda
virginiana
monophylla
discolor
ponderosa var. arizonica
strobiformis
washoensis
quadrifolia
nigra
cembroides
macrocarpa
menziesii
211   redwood
Sequioa
sempervirens
                                       C-10

-------
Code   Common Name
                            Genus
                     Species
                 -Softwoods	
212    giant sequoia
221
222
baldcypress
pondcypress
231    Pacific yew

241    northern white-cedar
242    western  redcedar

250    California torreya

260    hemlock
261    eastern hemlock
262    Carolina hemlock
263    western  hemlock
264    mountain hemlock
Sequoiadendron

Taxodium
Taxodium

Taxus

Thuja
Thuja

Torreya

Tsuga
Tsuga
Tsuga
Tsuga
Tsuga
giganteum

distichum
distfchum var. nutans

brevifolia

occidentalis
plicata

Califomica

sp.
canadensis
caroliniana
heterophylla
mertensiana
                                           C-11

-------
Code  Common Name
Genus
Species
300
310
311
312
313
314
315
316
317
318
319
321
322
330
331
332
333
341
350
351
352
353
354
355
356
361
367
370
371
372
373
374
375
376
377
378
379
381
	 nciiuwuuus 	
acacia
maple
Florida maple
bigleaf maple
boxelder
black maple
striped maple
red maple
silver maple
sugar maple
mountain maple
Rocky Mountain maple
Bigtooth Maple
buckeye, horsechestnut
Ohio buckeye
yellow buckeye
buckeye(except 331 ,332)
ailanthus
Alder
red alder
white alder
Sitka alder
thinleaf alder
European alder
serviceberry
Pacific madrone
pawpaw
birch sp.
yellow birch
sweet birch
river birch
water birch
paper birch
western paper birch
Alaska paper birch
northwestern paper birch
gray birch
chittamwood.gum bumelia
Acacia
Acer
Acer
Acer
Acer
Acer
Acer
Acer
Acer
Acer
Acer
Acer
Acer
Aesculus
Aesculus
Aesculus
Aesculus
Ailanthus
Alnus
Alnus
Alnus
Alnus
Alnus
Alnus
Amelanchier
Arbutus
Asimina
Betula
Betula
Betula
Betula
Betula
Betula
Betula
Betula
Betula
Betula
Bumelia
sp.
sp.
barbatum
macrophyllum
negundo
nigrum
pennsylvanicum
rubrum
saccharinum
saccharum
spicatum
glabrum
grandidentatum
sp.
glabra
octandra
sp.
altissima
sp.
rubra
rhombifolia
sinuata
tenuifolia
glutinosa
sp.
menziesii
triloba
sp.
alleghaniensis
lenta
nigra
occidentalis
papyrifera
papyrifera var. commutata
papyrifera var. neoalaskana
papyrifera var. subcordata
populifolia
lanuginosa
                                     C-12

-------
Code  Common Name
Genus
Species
391

400
401
402
403
404
405
407
408
409
421
422
423
430
450
451
452
460
461
462
463
471
475
476
477
478
479
481
491
500
521
531
540
541
543
	 narowooas 	
American hornbeam,
musclewood
hickory sp.
water hickory
bittemut hickory
pignut hickory
pecan
shellbark hickory
shagbark hickory
black hickory
mockernut hickory
American chestnut
Allegheny chinkapin
Ozark chinkapin
chinkapin
catalpa
southern catalpa
northern catalpa
hackberry sp.
sugarberry
hackberry
netleaf hackberry
eastern redbud
curfleaf mountain-mahogany
alder-leaf mountain-mahogany
hairy mountain-mahogany
birchleaf mountain-mahogany
dwarf mountain-mahogany
yellowwood cladrastis
flowering dogwood
hawthorn
common persimmon
American beech
ash
white ash
black ash
Carpinus

Carya
Carya
Carya
Carya
Carya
Carya
Carya
Carya
Carya
Castanea
Castanea
Castanea
Castanopsis
Catalpa
Catalpa
Catalpa
Celtis
Celtis
Celtis
Celtis
Cercis
Cercocarpus
Cercocarpus
Cercocarpus
Cercocarpus
Cercocarpus
Cladrastis
Cornus
Crataegus
Diospyros
Fagus
Fraxinus
Fraxinus
Fraxinus
caroliniana

sp.
aquatica
cordiformis
glabra
illinoensis
laciniosa
ovata
texana
tomentosa
dentata
pumila
ozarkensis
sp.
sp.
bignonioides
speciosa
sp.
laevigata
occidentalis
reticulata
canadensis
ledifolius
montanus
brevrflorus
betuloides
intricatus
kentukea
florida
sp.
virginiana
grandifolia
sp.
americana
nigra
                                      C-13

-------
Code  Common Name
Genus
Species
544
545
546
551
552
555
571
580
581
591
601
602
611
621
641
650
651
652
653
654
660
661
680
681
682
691
692
693
694
701
711
712

	 naiuwwua 	
green ash
pumpkin ash
blue ash
waterlocust
honey locust
loblolly-bay
Kentucky coffeetree
Mountain silvertell
Carolina silverbell
American holly
butternut
black walnut
sweetgum
yellow-poplar
Osage-orange
magnolia sp.
cucumbertree
southern magnolia
sweetbay
bigleaf magnolia
apple sp.
Oregon crab apple
mulberry sp.
white mulberry
red mulberry
water tupelo
ogeechee tupelo
blackgum
swamp tupelo
eastern hophornbeam,
ironwood
sourwood
paulownia, empress tree

Fraxinus
Fraxinus
Fraxinus
Gleditsia
Gleditsia
Gordonia
Gymnocladus
Halesia
Halesia
Ilex
Juglans
Juglans
Liquidambar
Liriodendron
Madura
Magnolia
Magnolia
Magnolia
Magnolia
Magnolia
Malus
Malus
Morus
Morus
Morus
Nyssa
Nyssa
Nyssa
Nyssa
Ostrya
Oxydendrum
Paulownia
C-14
pennsylvanica
profunda
quadrangulata
aquatica
triacanthos
lasianthus
dioicus
sp.
Carolina
opaca
cinerea
nigra
styraciflua
tulip ifera
pomifera
sp.
acuminata
grandiflora
virgirriana
macrophylla
sp.
fusca
sp.
alba
rubra
aquatica
ogeche
sylvatica
sylvatica var. biflora
virginiana
arboreum
tomentosa


-------
Code  Common Name
Genus
Species
721
722
731
740
741
742
743
744
745
746
747
748
749
752
755
760
761
762
763
764

765
766
800
801
802
803
804
805
806
807
808
809
810
811
812
813

814
815
816
817
818
	 narowooas 	
redbay
water elm, planer tree
sycamore
cottonwood
balsam poplar
eastern cottonwood
bigtooth aspen
swamp cottonwood
plains cottonwood
quaking aspen
black cottonwood
fremont poplar
narrow/leaf cottonwood
silver poplar
mesquite
cherry, plum spp.
pin cherry
black cherry
chokecherry
plums, cherries,
except 762
Canada plum
wild plum
Oak
California live oak
white oak
Arizona white oak, Gray oak
swamp white oak
canyon live oak
scarlet oak
blue oak
durand oak
northern pin oak
emery oak
engelmann oak
southern red oak
cherrybark oak,
swamp red oak
gambel oak
Oregon white oak
bear oak, scrub oak
shingle oak
California black oak
Persea
Planera
Platanus
Populus
Populus
Populus
Populus
Populus
Populus
Populus
Popufus
Populus
Populus
Populus
Prosopis
Prunus
Prunus
Prunus
Prunus
Prunus

Prunus
Prunus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus

Quercus
Quercus
Quercus
Quercus
Quercus
borbonia
aquatica
occidentalis
spp.
balsamifera
deltoides
grandidentata
heterophylla
sargentii
tremuloides
trichocarpa
fremontii
angustifolia
alba
sp.
sp.
pensylvanica
serotina
virginiana
sp.

nigra
americana
spp.
agrifolia
alba
arizonica, grisea
bicolor
chrysolepsis
coccinea
douglassi
durandii
ellipsoidalis
emoryi
engelmannii
falcata var. falcata
falcata var. pagodaefolia

gambelii
garryana
ilicifolia
imbricaria
kelloggii
                                     C-15

-------
Code  Common Name
Genus
Species

819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
848
849
899
901
902
920
921
922
928
931
935
936
950
951
952
970
971
972
	 naruwuuus 	
turkey oak
laurel oak
California white oak
overcup oak
bur oak
blackjack oak
swamp chestnut oak
chinkapin oak
water oak
nuttall oak
Mexican blue oak
pin oak
willow oak
chestnut oak
northern red oak
shumard oak
post oak
delta post oak
black oak
live oak
interior live oak
dwarf post oak
dwarf live oak
bluejack oak
silverleaf oak
western Oak (Deciduous)
western Oak (Evergreen)
scrub oak
black locust
New Mexico locust
willow
peachleaf willow
black willow
diamond willow
sassafras
American mountain-ash
European mountain-ash
basswood
American basswood
white basswood
elm
winged elm
American elm

Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercus
Quercua
Quercus
Quercus
Quercus
Quercus
Robinia
Robinia
Salix
Salix
Salix
Salix
Sassafras
Sorbus
Sorbus
Tilia
Tilia
Tilia
Ulmus
Ulmus
Ulmus

laevis
laurifolia
lobata
lyrata
macrocarpa
marilandica
michauxii
muehlenbergii
nigra
nuttalii
oblongifolia
palustris
phellos
prinus
rubra
shumardii
stellata
stellata var. Mississippiensis
velutina
virginiana
wislizenii
stellata var stellata
sp.
incana
hypoleucoides
spp.
spp.
sp.
pseudoacacia
neomexicana
sp.
amygdaloides
nigra
eriocephala
albidum
americana
aucuparia
sp.
americana
heterophylla
sp.
alata
americana
                                     C-16

-------
Code  Common Name
Genus
Species

973
974
975
976
977
981
990
991
992
993
994
995
996
998
999

	 	 naruwoous 	
cedar elm
Siberian elm
slippery elm
September elm
rock elm
California laurel
Arizona ironwood
salt cedar
sparkleberry
chinaberry
Chinese tallowtree
tung-oil tree
smoketree
Other tree species
unknown or not listed

Ulmus
Ulmus
Ulmus
Ulmus
Ulmus
Umellularia
Olneya
Tamarisk
Vaccinium
Melia
Sapium
Aleu rites
Cotinus



crassifolia
pumila
rubra
serotina
thomasii
califomica
tesota
sp.
arboreum
azedarach
sebiferum
fordii
obovatus


                                     C-17

-------
Appendix D.  Conversion Factors for Common Measurement Units
             Metric unit
English equivalent
             1 centimeter (cm)
             1 meter (m)
             1 kilometer (km)
                                       Length
0.39370 inches (in)
3.28083 feet (ft)
0.62137 miles (mi)
                                       Area
             1 square centimeter (cm2)

             1 square meter (m2)
             1 hectare (ha)

             1 square kilometer (km2)

             1 m2/ha
0.15500 square inches (in2)
0.001076 square feet (ft2)
10.76387 square feet (ft2)
2.471044 acres (ac)
0.0038610 square miles (mi2)
247.104 acres (ac)
0.38610 square miles (mi2)
4.3560 ft2/ac
                                        D-1

-------
Appendix E.  Supplemental Tabular Summaries

       By the end of 1991,925 potential plot locations had been visited at least once in 12 eastern states,
and 628 plots had been established. Considering trees larger than 12.7 cm DBH, 14,307 trees of 109
identified species were measured in 1991.  Among the smaller trees, 4,438 saplings larger than 2.5 cm
DBH, and 26,688 seedlings larger than 30.5 cm tall were tallied in 1991, and eleven additional species were
identified.

       The objective of this Appendix is to describe the sample of plots and trees that were measured in
1991. This information can be used to help interpret the forest health indicators that are summarized in the
main body of the report. This information does not constitute population estimates for the forests, states,
or other  categories that were sampled. Population  estimates must be based on  specific statistical
procedures (Appendix A), and these tables are presented as background information only.

       A  simplification was made to prepare  the summaries of plot characteristics.  Each subplot was
placed into separate categories so that the sampling effort could be expressed in terms of the numbers of
subplots. In the field, however, there is  not a single category for any given plot or subplot. For example,
each plot may be made up of subplots of different land use categories, and a forested subplot may border
a plantation-natural forest boundary.  Field crews used a set of rules to identify and map distinctly  different
land  use categories or conditions within  a subplot.  The rules recognize different conditions if any one of
the following situations was different with a subplot: land use, forest type, stand origin, predominant size
class, or site disturbance.  The summary tables  (in this Appendix only) assigned each subplot to the
category of the most prevalent condition encountered on each subplot. The alternative of summarizing
according to the area of each condition sampled was not used because there is not a simple procedure for
re-expressing the resulting area statistics as numbers of subplots.

       Table E-5 through E-7 report the evidence of disturbance in the last five  years.  In Table E-5,
utilization disturbance includes evidence of harvest, commercial thinning, selective cutting, and  grazing.
In Table E-6, care disturbance includes site preparation, artificial regeneration, prescribed burning, and
miscellaneous silvicultural treatments such  as herbicide or fertilizer application.  In Table E-7, natural
disturbance includes evidence of damage by insects, disease, weather,  and wildfire.
                                              E-1

-------
                        Table E-1.  Number of subplots by state and land use.
  Land Use
                   State
AL  GA VA  MD DE  NJ  CT Rl  MA VT  NH  ME Total
               (no. of subplots)
Forest land
Crop land
Improved pasture
Natural rangeland
Idle farmland
Other farmland
Urban or developed
Marsh (non-forest)
Water
Access denied
Unsafe conditions
Data lost
512
74
70
0
37
12
100
0
13
0
14
0
553 374
173 43
29 77
0 0
12 17
2 4
61 85
21 3
12 13
33 12
4 0
36 12
58
48
4
0
0
0
30
4
4
0
0
0
5
11
0
0
0
0
12
4
0
0
0
0
57
12
4
0
1
0
35
8
0
0
3
0
41
0
4
0
0
0
19
0
4
0
0
0
8
0
4
0
0
0
4
0
0
0
0
0
59
0
7
0
4
0
27
8
3
0
0
0
83
20
8
0
1
4
8
0
0
0
0
0
120
2
0
1
0
0
10
3
0
0
0
0
484
15
4
0
15
0
22
0
0
0
0
0
2354
398
211
1
87
22
413
51
49
45
21
48
Total
832 936 640  148  32 120  68  16  108  124 136 540 3700
                                                 E-2

-------
                      Table E-2.  Number of subplots by state and major forest type.
Major Forest type
                       State
AL    GA   VA    MD   DE   NJ    CT    Rl    MA   VT
                   (no. of subplots)
NH   ME   Total
White/red/jack pine
Eastern spruce-fir
Longleaf/slash pine
Loblolly/shortleaf pine
Oak-pine
Oak-hickory
Oak-gum-cypress
Elm-ash-red maple
Maple-beech-birch
Aspen-birch
Other
Non-forest
Total
0
0
23
140
171
111
53
7
0
0
7
320
832
1
0
112
167
83
104
79
4
0
0
3
383
936
5
0
0
67
44
237
1
8
8
0
4
266
640
0
0
0
3
6
33
7
0
5
0
4
90
148
0
0
0
0
0
1
4
0
0
0
0
27
32
7
0
0
13
6
11
5
0
11
0
4
63
120
4
0
0
0
0
23
0
6
7
1
0
27
68
0
0
0
0
0
0
0
4
0
4
0
8
16
1
0
0
0
14
4
0
0
37
3
0
49
108
16
11
0
0
0
0
0
0
48
4
4
41
124
27
2
0
0
5
0
0
4
81
1
0
16
136
35
250
0
0
20
0
0
0
161
17
1
56
540
96
263
135
390
349
524
149
33
358
30
27
1346
3700
                                               E-3

-------
          Table E-3.  Number of forested subplots by major forest type and stand size class.
Major forest type
                 Stand size class
                      Seedling/   Non-
Sawtimber  Poletimber sapling      stocked
              (no. of forested subplots)
White/red/jack pine
Eastern spruce-fir
Longleaf/slash pine
Loblolly/shortleaf pine
Oak-pine
Oak-hickory
Oak-gum-cypress
Elm-ash-red maple
Maple-beech -birch
Aspen-birch
Other
63
125
63
123
193
333
108
21
221
12
3
32
106
43
105
65
88
20
7
100
10
8
1
32
26
158
91
100
21
5
37
8
5
0
0
0
0
0
3
0
0
0
0
11
Total
   1265
584
484
14
                                              E-4

-------
            Table E-4. Number of forested subplots by major forest type and stand origin.
                       Stand Origin
Major forest type     Natural   Planted
                   (no. of forested subplots)
White/red/jack pine
Eastern spruce-fir
Longleaf/slash pine
Loblolly/shortleaf pine
Oak-pine
Oak-hickory
Oak-gum-cypress
Elm-ash-red maple
Maple-beech-birch
Aspen-birch
Other
82
259
83
203
325
521
149
33
358
30
27
14
4
52
187
24
3
0
0
0
0
0
Total                  2070  284
                                               E-5

-------
       Table E-5.  Number of forested subplots by major forest type and utilization disturbance.
                          Evidence of
                       Utilization disturbance
Major forest type       No      Yes
                       (no. of forested subplots)
White/red/jack pine
Eastern spruce-fir
Longleaf/slash pine
Loblolly/shortleaf pine
Oak-pine
Oak-hickory
Oak-gum-cypress
Elm-ash-red maple
Maple-beech-birch
Aspen-birch
Other
88
235
101
332
292
446
125
31
294
26
13
8
28
34
58
57
78
24
2
64
4
14
Total                   1983    371
                                               E-6

-------
          Table E-6.  Number of forested subplots by major forest type and care disturbance.
Major forest type
  Evidence of
  care disturbance
 No      Yes
(no. of forested subplots)
White/red/jack pine
Eastern spruce-fir
Longleaf/slash pine
Loblolly/shortleaf pine
Oak-pine
Oak-hickory
Oak-gum-cypress
Elm-ash-red maple
Maple-beech-birch
Aspen-birch
Other
96
260
77
296
330
503
146
33
349
30
24
0
3
58
94
19
21
3
0
9
0
3
Total
2144
210
                                              E-7

-------
        Table E-7. Number of forested subplots by major forest type and natural disturbance.
Major forest type
  Evidence of
  natural disturbance
 No      Yes
(no. of forested subplots)
White/red/jack pine
Eastern spruce-fir
Longleaf/slash pine
Loblolly/shortleaf pine
Oak-pine
Oak-hickory
Oak-gum-cypress
Elm-ash-red maple
Maple-beech-birch
Aspen-birch
Other
96
263
132
377
341
486
138
32
358
30
27
0
0
3
13
8
38
11
1
0
0
0
Total
2280
74
                                               E-8

-------
               Table E-8.  Number of subplots by major forest type and terrain position.
Major forest type
         Terrain position



Slope       Flatland    Bottomland



         (no. of subplots)
White/red/jack pine
Eastern spruce-fir
Longleaf/slash pine
Loblolly/shortleaf pine
Oak-pine
Oak-hickory
Oak-gum-cypress
Elm-ash-red maple
Maple-beech-birch
Aspen-birch
Other
Non-forest
Total
82
247
8
167
168
362
15
14
323
26
11
1229
2652
14
16
115
194
160
144
55
11
35
4
12
92
852
0
0
12
29
21
18
79
8
0
0
4
25
196
                                               E-9

-------
              Table E-9.  Number of subplots by major forest type and elevation class.
                               Elevation class (ft)
Major forest type     <500    500-1500     1500-2500   >2500   Missing
White/red/jack pine
Eastern spruce-fir
Longleaf/slash pine
Loblolly/shortleaf pine
Oak-pine
Oak-hickory
Oak-gum-cypress
Elm-ash-red maple
Maple-beech-birch
Aspen-birch
Other
85
251
123
243
225
238
132
26
337
27
18
4
12
12
139
91
175
16
0
12
3
3
2
0
0
4
9
65
0
0
0
0
5
4
0
0
4
20
42
0
4
8
0
0
1
0
0
0
4
4
1
3
1
0
1
Non-forest             1035   217         71          14       9

Total                  2740   684        156          96      24
                                             E-10

-------
Table E-10.  Number of live trees greater than 12.7 cm DBH by species and crown position on
            subplots.
Species
balsam fir
eastern redcedar
larch (introduced)
tamarack
white spruce
black spruce
blue spruce
red spruce
shortleaf pine
slash pine
spruce pine
longleaf pine
table-mountain pine
red pine
pitch pine
pond pine
eastern white pine
Scots pine
loblolly pine
Virginia pine
Austrian pine
baldcypress
pondcypress
northern white
-------
Species
hackberry (common)
eastern redbud
yellowwood
flowering dogwood
persimmon
American beech
ash species
white ash
black ash
green ash
American holly
butternut
black walnut
sweetgum
yellow-poplar
osage-o range
magnolia species
cucumbertree
southern magnolia
sweetbay
apple species
Chinaberry
mulberry species
red mulberry
water tupelo
blackgum
swamp tupelo
eastern hophombeam
sourwood
Paulownia
redbay
American sycamore
balsam poplar
eastern cottonwood
bigtooth aspen
quaking aspen
pin cherry
black cherry
choke cherry (common)
white oak
swamp white oak
scarlet oak
black oak
northern pin oak
southern red oak
cherrybark oak
turkey oak
laurel oak
overcup oak
blackjack oak
swamp chestnut oak
Open -grown
0
0
0
0
0
0
0
3
0
1
0
0
0
3
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
0
2
0
3
0
2
0
0
0
0
0
0
0
1
0
Dominant
2
0
0
4
6
15
2
23
1
4
3
1
2
120
112
0
2
0
0
14
0
1
0
0
5
24
19
0
1
1
0
4
3
3
11
33
0
10
0
81
0
58
0
0
28
2
3
3
2
1
0
Crown position
Co-dominant
13
0
1
7
6
171
5
118
16
11
16
4
8
250
185
1
2
10
1
38
3
1
0
4
27
66
70
5
14
0
5
4
6
1
27
111
2
66
5
203
0
86
0
6
99
4
3
17
5
14
4
Intermediate
8
3
0
25
1
76
4
21
5
11
24
0
3
179
73
1
1
4
10
32
2
1
0
4
20
65
40
19
42
0
4
3
0
0
7
11
2
30
2
112
2
14
2
1
43
1
1
11
0
7
0
Suppressed
0
2
0
33
0
32
2
7
1
2
13
0
2
53
19
0
2
1
0
9
2
0
1
1
5
35
13
3
35
0
0
2
0
0
0
2
0
6
0
41
2
3
0
0
11
0
1
3
1
3
0
Total
23
5
1
69
13
294
13
172
23
29
56
5
15
605
390
2
7
15
12
93
7
3
1
9
57
190
142
27
92
1
9
13
9
4
45
160
4
114
7
440
4
163
2
7
181
7
8
34
8
26
4
E-12

-------
Species
chinkapin oak
water oak
willow oak
chestnut oak
northern red oak
Shumard oak
post oak
black oak
live oak
dwarf post oak
black locust
black willow
sassafras
American basswood
white basswood
elm species
winged elm
American elm
slippery elm
unknown / other
Open-grown
0
2
0
0
1
0
0
0
1
0
0
0
0
1
0
0
0
0
0
0
Crown position
Dominant
0
50
9
73
64
0
12
23
3
0
12
0
2
1
0
2
0
1
0
4
Co-dominant
1
144
21
220
209
1
30
65
16
0
10
3
15
8
0
6
11
9
3
6
Intermediate
3
57
6
96
46
0
23
37
1
1
9
3
13
5
0
1
6
3
5
0
Suppressed
0
18
4
26
14
0
11
8
0
2
5
0
5
2
1
0
2
0
3
2
Total
4
271
40
415
334
1
76
133
21
3
36
6
35
17
1
9
19
13
11
12
E-13

-------
Table E-11.  Numbers of trees tallied by species and size class (seedlings, saplings, and large trees >12.7
            cm DBH).
Species
balsam fir
Atlantic white-cedar
eastern redcedar
larch (introduced)
tamarack
white spruce
black spruce
blue spruce
red spruce
shortleaf pine
slash pine
spruce pine
longleaf pine
table-mountain pine
red pine
pitch pine
pond pine
eastern white pine
Scots pine
loblolly pine
Virginia pine
Austrian pine
baldcypress
po ndcy press
northern white-cedar
hemlock species
eastern hemlock
maple species
Florida maple
boxelder
striped maple
red maple
sugar maple
mountain maple
horsechestnut species
Ohio buckeye
yellow buckeye
serviceberry
pawpaw
birch species
yellow birch
sweet birch
river birch
paper birch
gray birch
Seedlings
3017
7
60
0
3
29
91
0
529
15
44
2
5
0
1
6
0
220
0
636
105
0
8
1
244
0
254
2
76
61
940
3292
1696
236
17
1
20
42
40
5
367
215
0
640
194
Saplings
400
0
23
0
17
18
76
0
145
24
33
0
9
0
1
13
0
57
0
325
28
0
1
0
36
1
58
0
0
2
85
506
118
6
0
0
0
8
3
11
59
30
2
80
49
Trees
585
0
49
1
16
78
21
2
654
278
455
13
119
11
12
103
14
613
4
1814
471
1
47
3
350
1
338
0
2
18
23
1463
464
0
0
0
6
1
0
22
257
92
16
309
64
Total
4002
7
132
1
36
125
188
2
1328
317
532
15
133
11
14
122
14
890
4
2775
604
1
56
4
630
2
650
2
78
81
1048
5261
2278
242
17
1
26
51
43
38
683
337
18
1029
307
                                             E-14

-------
Species
American hornbeam
hickory species
water hickory
brtternut hickory
pignut hickory
pecan
shagbark hickory
mockernut hickory
American chestnut
catalpa species
hackberry species
hackberry (common)
eastern redbud
yellowwood
flowering dogwood
hawthorn species
persimmon
American beech
ash species
white ash
black ash
green ash
honeylocust
silverbell species
American holly
butternut
black walnut
sweetgum
yellow-poplar
osage-orange
magnolia species
cucumbertree
southern magnolia
sweetbay
bigleaf magnolia
apple species
Chinaberry
mulberry species
red mulberry
water tupelo
blackgum
swamp tupelo
eastern hophornbeam
sourwood
Paulownia
redbay
American sycamore
Seedlings
341
477
0
6
18
4
2
148
71
1
1
7
85
0
1248
31
196
523
19
467
32
213
2
6
229
0
3
1307
516
0
34
21
11
185
2
8
2
5
11
51
787
79
156
300
0
29
1
Saplings
70
88
0
0
2
0
2
31
2
0
2
0
19
1
231
1
12
117
6
47
4
18
0
0
38
0
0
286
78
0
10
5
3
61
0
6
0
0
0
8
153
33
42
47
0
4
3
Trees
57
246
2
1
31
0
19
82
1
0
1
23
5
1
69
0
13
294
13
172
23
29
0
0
56
5
15
605
390
2
7
15
12
93
0
7
3
1
9
57
190
142
27
92
1
9
13
Total
468
811
2
7
51
4
23
261
74
1
4
30
109
2
1548
32
221
934
38
686
59
260
2
6
323
5
18
2198
984
2
51
41
26
339
2
21
5
6
20
116
1130
254
225
439
1
42
17
E-15

-------
Species
balsam poplar
eastern cottonwood
bigtooth aspen
quaking aspen
cherry, plum species
pin cherry
black cherry
choke cherry (common)
white oak
swamp white oak
scarlet oak
bluejack oak
northern pin oak
southern red oak
cherrybark oak
bear oak
turkey oak
laurel oak
overcup oak
blackjack oak
swamp chestnut oak
chinkapin oak
water oak
pin oak
willow oak
chestnut oak
northern red oak
Shumard oak
post oak
black oak
live oak
dwarf post oak
dwarf live oak
scrub oak
black locust
willow species
black willow
sassafras
American mountain-ash
American basswood
white basswood
elm species
winged elm
American elm
slippery elm
rock elm
Seedlings
27
0
16
202
7
171
695
17
519
15
96
50
2
381
8
73
19
234
0
41
7
17
853
4
212
244
396
0
137
191
29
0
48
81
36
1
17
612
31
14
1
75
110
54
31
2
Saplings
0
0
3
28
2
39
84
0
73
1
10
7
0
51
0
1
5
61
0
4
0
0
147
0
23
22
53
0
20
21
3
3
1
3
2
2
1
18
0
1
0
3
25
6
5
0
Trees
9
4
45
160
0
4
114
7
440
4
163
2
7
181
7
0
8
34
8
26
4
4
271
0
40
415
334
1
76
133
21
3
0
0
36
0
6
35
0
17
1
9
19
13
11
0
Total
36
4
64
390
9
214
893
24
1032
20
269
59
9
613
15
74
32
329
8
71
11
21
1271
4
275
681
783
1
233
345
53
6
49
84
74
3
24
665
31
32
2
87
154
73
47
2
E-16

-------
Species                 Seedlings   Saplings   Trees    Total
sparkleberry              18         6          0        24
unknown / other          138       50          12      200

total                   26,688     4438      14,407   62,681
                                          E-17

-------
Table E-12.  Number of live trees > 12.7 cm DBH tallied on subplots, by species and crown ratio class.
Species
balsam fir
eastern redcedar
larch (introduced)
tamarack
white spruce
black spruce
blue spruce
red spruce
shortteaf pine
slash pine
spruce pine
longteaf pine
table-mountain pine
red pine
pitch pine
pond pine
eastern white pine
Scots pine
loblolly pine
Virginia pine
Austrian pine
baldcypress
pondcypress
northern white-cedar
hemlock species
eastern hemlock
Florida maple
boxelder
striped maple
red maple
sugar maple
yellow buckeye
service berry
birch species
yellow birch
sweet birch
river birch
paper birch
gray birch
American hornbeam
hickory species
water hickory
bittemut hickory
pignut hickory
shagbark hickory
mockemut hickory
American chestnut
hackberry species
hackberry (common)
eastern redbud
yedowwood
flowering dogwood
persimmon
American beech
ash species
white ash
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
2
4
1
0
0
1
0
0
0
8
12
0
1
0
0
2
0
14
0
17
27
0
1
0
7
0
1
0
0
0
8
0
1
0
0
2
0
0
1
1
0
2
0
0
0
0
0
0
0
0
0
0
1
0
1
0
3
3
29
1
0
2
2
2
0
24
52
64
2
4
0
0
2
0
86
0
120
101
0
6
0
11
0
8
0
0
2
75
12
1
0
5
10
2
0
34
4
1
5
0
0
0
0
1
0
0
1
0
0
1
0
8
1
31
4
56
2
0
0
6
3
0
104
72
148
4
16
1
2
20
3
104
0
354
154
0
10
1
18
0
27
0
0
1
225
54
1
0
4
22
13
1
85
15
6
27
0
0
4
0
6
0
0
8
2
1
12
4
24
4
49
Crown
5
74
7
0
3
12
1
1
135
75
158
3
45
6
2
15
3
116
1
495
116
1
9
1
35
0
33
1
3
6
436
97
1
0
4
56
19
6
96
18
10
51
1
0
9
3
23
0
0
2
2
0
7
4
46
8
41
Ratio Class*
6 7
115
7
1
2
24
3
0
160
43
68
3
35
2
5
33
0
99
0
402
36
0
8
1
51
0
49
1
4
8
338
115
0
0
4
69
30
3
46
18
19
59
1
1
9
6
24
0
0
3
0
0
10
2
57
0
18
78
8
0
4
12
7
1
98
14
5
0
14
1
0
18
5
60
1
215
18
0
6
0
77
0
55
0
4
2
199
105
0
0
1
43
19
4
27
6
11
45
0
0
5
3
13
0
1
2
0
0
10
2
49
0
18
8
79
5
0
2
7
1
0
63
5
0
1
4
1
2
9
3
40
0
123
11
0
1
0
45
0
51
0
2
3
105
52
1
0
3
30
3
1
16
1
7
31
0
0
4
5
8
0
0
2
0
0
14
1
31
0
6
9
70
9
0
1
11
3
0
48
5
0
0
0
0
1
3
0
42
2
66
3
0
4
0
63
1
50
0
2
0
51
20
1
1
1
19
4
0
4
1
3
17
0
0
0
1
6
1
0
1
1
0
10
0
41
0
5
10
52
4
0
1
1
1
0
15
2
0
0
0
0
0
1
0
28
0
17
4
0
2
0
36
0
40
0
3
1
22
9
0
0
0
6
2
0
0
0
0
6
0
0
0
1
1
0
0
4
0
0
3
0
24
0
1
11
28
5
0
1
2
0
0
7
2
0
0
0
0
0
0
0
24
0
5
1
0
0
0
7
0
24
0
0
0
2
0
0
0
0
0
0
1
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
13
0
0
No.
trees
585
49
1
16
78
21
2
654
278
455
13
119
11
12
103
14
613
4
1814
471
1
47
3
350
1
338
2
18
23
1463
464
6
1
22
257
92
16
309
64
57
246
2
1
31
19
82
1
1
23
5
1
69
13
294
13
172
                                             E-18

-------
Species
black ash
green ash
American holly
butternut
black walnut
sweetgum
yetow-poplar
osage -orange
magnola species
cucumbertree
southern magnolia
sweetbay
apple species
Chinaberry
mulberry species
red mulberry
water tupelo
blackgum
swamp tupelo
eastern hophombeam
sourwood
Pautownia
redbay
American sycamore
balsam poplar
eastern cottonwood
bigtooth aspen
quaking aspen
pin cherry
black cherry
choke cherry (common)
white oak
swamp white oak
scarlet oak
bluejack oak
northern pin oak
southern red oak
cherrybark oak
turkey oak
laurel oak
overcup oak
blackjack oak
swamp chestnut oak
chinkapin oak
water oak
willow oak
chestnut oak
northern red oak
Shumard oak
post oak
black oak
live oak
dwarf post oak
black locust
black willow
sassafras
1
0
0
0
0
0
2
2
0
0
0
0
1
0
0
0
0
0
0
1
0
2
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
1
0
0
0
0
2
0
0
1
0
0
0
1
0
0
2
1
0
0
0
1
10
3
0
0
0
0
2
0
0
0
0
0
3
3
0
0
0
1
0
1
0
2
2
1
3
0
2
0
2
0
0
2
0
0
0
0
0
0
0
3
1
2
2
0
1
0
0
0
1
0
0
3
5
2
0
0
1
35
5
0
0
0
0
3
0
0
0
0
1
14
15
0
4
0
0
1
1
2
6
30
0
14
1
9
1
10
0
2
7
0
0
1
0
1
0
0
5
1
14
11
0
4
3
0
0
14
0
4
4
10
4
0
2
1
101
63
0
0
0
0
19
0
0
0
5
20
23
36
4
22
0
0
2
1
2
14
70
2
33
2
33
0
18
0
1
27
2
1
4
0
5
0
2
27
4
42
70
0
7
11
3
0
8
1
8
Crown
5
3
7
2
2
6
150
130
0
2
2
2
28
0
0
0
1
22
54
41
6
31
0
1
3
1
0
16
39
1
35
2
90
2
50
1
1
54
2
2
8
2
11
0
0
59
10
114
107
1
24
34
8
1
7
0
13
Ratio Class'
6 7
2
6
9
1
2
136
97
0
5
2
2
23
4
1
0
1
10
45
30
7
21
0
4
1
2
0
5
15
0
18
2
129
0
54
1
1
51
3
2
7
4
4
0
2
74
9
110
79
0
18
42
5
0
2
4
6
0
6
9
0
2
86
55
1
0
4
5
14
1
1
0
1
3
28
8
6
10
0
2
4
2
0
2
2
0
7
0
98
0
11
0
2
27
0
2
2
2
2
1
0
39
5
62
38
0
12
26
1
0
1
1
3
8
1
2
9
0
1
43
22
0
0
4
2
0
1
1
0
1
0
11
7
3
1
0
1
2
1
0
0
0
0
3
0
34
1
11
0
0
5
0
1
5
0
1
2
0
27
3
40
19
0
7
9
2
0
1
0
0
9
1
2
15
0
1
27
7
1
0
3
1
3
0
0
0
0
0
6
0
0
0
0
0
0
0
0
0
2
0
0
0
33
0
2
0
0
• 6
0
0
5
0
0
1
0
19
3
22
7
0
2
6
2
1
1
0
1
10
0
0
8
0
0
11
3
0
0
0
0
0
0
0
1
0
1
2
1
1
1
1
0
0
0
0
0
0
0
0
0
8
0
5
0
0
2
0
0
1
0
0
0
0
15
2
7
1
0
0
2
0
1
0
0
0
11
0
0
4
0
0
4
3
0
0
0
0
0
1
0
0
0
0
4
0
0
0
0
0
0
0
0
0
0
0
1
0
2
0
0
0
0
0
0
0
1
0
1
0
0
3
2
0
0
0
0
0
21
0
0
0
0
No.
trees
23
29
56
5
15
605
390
2
7
15
12
93
7
3
1
9
57
190
142
27
92
1
9
13
9
4
45
160
4
114
7
440
4
163
2
7
181
7
8
34
8
26
4
4
271
' 40
415
334
1
76
133

3
36
6
35
E-19

-------
Species
American basswood
white basswood
elm species
winged etm
American elm
slippery elm
unknown / other
1
0
0
0
0
0
0
0
2
0
0
0
0
1
0
0
3
1
0
0
0
1
2
2
4
3
1
1
1
1
1
2
Crown Ratio Class*
567
4
0
4
2
2
1
4
6
0
2
6
6
5
2
1
0
0
3
0
1
2
8
2
0
1
3
0
1
0
9
0
0
1
1
2
0
0
10
0
0
0
3
0
0
0
11
0
0
0
0
0
0
0
No.
trees
17
1
9
19
13
T1
12
Total              19     166    943   2393   3502   3077   1867   1070    756    364    150  14307
•Crown ratio classes, in percent:
        1=0              7 = 51 - 60
        2 = 1 -10          8 = 61 - 70
        3 = 2-20          9 = 71-80
        4 = 21-30         10 = 81-90
        5 = 31-40         11 =91 -99
        6 = 41 - 50
                                                 E-20

-------
Table E-13.  Number of live trees > 12.7 cm DBH tallied on subplots, by species and crown density class.
Species
balsam fir
eastern redcedar
larch (introduced)
tamarack
white spruce
black spruce
blue spruce
red spruce
shortteaf pine
slash pine
spruce pine
longteaf pine
table-mountain pine
red pine
pitch pine
pond pine
eastern white pine
Scots pine
loblolly pine
Virginia pine
Austrian pine
baldcypress
pondcypress
northern white-cedar
hemlock species
eastern hemlock
Florida maple
boxelder
striped maple
red maple
sugar maple
yellow buckeye
serviceberry
birch species
yellow birch
sweet birch
river birch
paper birch
gray birch
American hornbeam
hickory species
water hickory
bittemut hickory
pignut hickory
shagbark hickory
mockemut hickory
American chestnut
hackberry species
hackberry (common)
eastern redbud
yellowwood
flowering dogwood
persimmon
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
2
4
1
0
0
0
0
0
1
1
0
0
0
0
0
0
0
5
0
3
3
0
0
0
9
0
0
0
0
0
5
1
1
0
0
1
0
0
4
0
0
0
0
0
0
0
0
0
0
0
0
0
3
0
3
9
0
0
2
1
1
0
5
9
21
0
0
0
0
0
0
19
0
48
29
0
2
1
14
0
8
0
0
0
32
5
0
0
0
5
0
0
2
3
1
0
0
0
0
0
1
0
0
0
0
0
1
0
4
26
3
0
2
3
3
2
26
33
133
3
14
2
0
2
0
52
0
169
87
0
4
1
40
0
19
1
1
0
97
9
2
0
3
6
1
1
11
8
2
7
0
1
1
0
3
0
0
0
1
0
4
3
Crown Density Class*
567
89
8
0
4
9
0
0
78
119
198
6
41
5
2
8
1
93
0
551
163
0
12
1
70
0
74
1
3
4
248
54
1
0
2
34
15
4
34
16
11
40
0
0
4
1
17
0
0
13
2
0
11
7
94
18
1
1
11
6
0
116
79
71
3
43
4
1
19
10
144
0
600
119
1
15
0
61
0
90
0
9
2
384
110
1
1
5
48
22
7
56
15
20
84
2
0
16
6
21
0
0
6
0
1
20
2
130
6
0
2
22
7
0
171
19
30
1
19
0
6
23
3
132
2
324
46
0
9
0
77
0
70
0
3
7
352
127
1
0
3
58
31
2
79
13
13
67
0
0
7
7
28
0
1
4
2
0
21
1
8
121
8
0
4
18
1
0
164
13
2
0
2
0
2
24
0
104
0
92
15
0
3
0
57
0
46
0
2
5
228
89
0
0
0
50
13
1
77
6
8
34
0
0
2
5
8
1
0
0
0
0
5
0
9
80
3
0
1
12
3
0
76
3
0
0
0
0
1
24
0
55
2
26
9
0
1
0
20
1
20
0
0
4
98
56
0
0
7
46
9
1
41
3
2
12
0
0
1
0
2
0
0
0
0
0
2
0
10
28
1
0
0
1
0
0
17
2
0
0
0
0
0
3
0
8
0
1
0
0
1
0
2
0
8
0
0
1
16
12
0
0
2
9
1
0
5
0
0
1
0
0
0
0
2
0
0
0
0
0
1
0
11
4
1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
3
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
No.
trees
585
49
1
16
78
21
2
654
278
455
13
119
11
12
103
14
613
4
1814
471
1
47
3
350
1
338
2
18
23
1463
464
6
1
22
257
92
16
309
64
57
246
2
1
31
19
82
1
1
23
5
1
69
13
                                            E-21

-------
Species 1
American beech
ash species
white ash
black ash
green ash
American holy
butternut
black walnut
sweetgum
yellow-poplar
osage-orange
magnolia species
cucumbertree
southern magnolia
sweetbay
apple species
Chinaberry
mulberry species
red mulberry
water tupeto
blackgum
swamp tupeto
eastern hophombeam
sourwood
Pautownia
redbay
American sycamore
balsam poplar
eastern cottonwood
bigtooth aspen
quaking aspen
pin cherry
black cherry
choke cherry (common)
white oak
swamp white oak
scarlet oak
bluejack oak
northern pin oak
southern red oak
cherrybark oak
turkey oak
laurel oak
overcup oak
blackjack oak
swamp chestnut oak
chinkapin oak
water oak
willow oak
chestnut oak
northern red oak
Shumard oak
post oak
black oak
live oak
dwarf post oak
black locust
black willow

0
0
0
0
0
0
0
0
2
2
0
0
0
0
0
0
0
0
0
0
0
1
0
2
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
1
0
0
0
0
2
0
0
1
0
0
0
0
0
2
9
0
1
2
1
0
0
0
3
1
0
0
0
0
2
0
0
0
0
0
0
1
0
0
0
0
0
0
1
8
5
0
2
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
3
11
0
8
3
2
0
0
0
12
1
0
0
0
0
1
0
0
0
1
0
5
4
2
2
0
1
0
0
0
3
23
0
4
0
12
0
2
0
0
6
0
0
0
0
2
0
0
3
0
2
6
0
2
3
0
0
3
0
Crown Density Class*
4567
19
2
5
4
3
0
2
1
54
6
1
0
0
0
9
0
0
0
2
1
8
25
1
9
0
2
4
1
1
0
16
1
14
0
22
1
10
0
1
15
0
3
1
3
6
0
0
19
3
35
32
0
14
10
1
2
14
3
45
4
31
5
7
2
2
5
135
64
0
1
0
2
26
2
3
0
4
12
42
54
6
31
0
2
2
0
2
6
31
1
16
0
82
2
47
1
2
33
1
2
3
2
4
0
2
70
3
120
90
0
28
26
5
0
6
1
62
2
38
5
7
14
1
4
194
138
1
1
5
3
35
3
0
1
1
34
54
49
9
28
0
2
5
3
0
7
22
1
27
2
128
0
55
0
3
66
3
1
10
3
5
1
1
90
18
151
71
1
19
52
7
1
5
2
61
2
36
3
5
23
0
5
138
113
0
1
3
6
13
0
0
0
1
9
43
7
6
12
0
1
2
3
0
10
28
1
23
2
124
1
29
1
0
37
1
1
14
0
6
1
1
57
11
74
73
0
10
27
6
0
5
0
8
62
3
29
0
4
13
0
0
43
48
0
4
5
1
7
1
0
0
0
1
23
1
1
5
1
0
0
2
0
3
21
0
15
1
50
0
15
0
1
22
2
1
6
0
1
1
0
26
5
23
44
0
2
9
2
0
2
0
9
17
0
20
1
0
4
0
0
19
17
0
0
2
0
0
1
0
0
0
0
11
0
1
3
0
1
0
0
0
3
13
0
10
1
19
0
2
0
0
1
0
0
0
0
1
1
0
6
0
7
17
0
0
5
0
0
1
0
10
7
0
4
0
0
0
0
0
4
0
0
0
0
0
0
0
0
0
0
0
3
0
1
0
0
0
0
0
0
5
1
0
3
1
0
0
1
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
11
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
No.
trees
294
13
172
23
29
56
5
15
605
390
2
7
15
12
93
7
3
1
9
57
190
142
27
92
1
9
13
9
4
45
160
4
114
7
440
4
163
2
7
181
7
8
34
8
26
4
4
271
40
415
334
1
76
133
21
3
36
6
E-22

-------
Species
sassafras
American basswood
white basswood
elm species
winged elm
American elm
slpperyelm
unknown / other
Total
1
0
0
0
0
0
0
0
0
17
2
0
0
0
0
0
1
0
0
82
3
0
0
0
1
0
0
1
0
345
4
3
1
0
1
1
0
1
0
1179
Crown Density Class*
567
9
10
0
3
8
3
4
4
3168
12
1
1
1
5
6
3
3
3828
5
2
0
2
2
2
2
0
2977
8
3
3
0
1
1
1
0
3
1728
9
3
0
0
0
2
0
0
2
812
10
0
0
0
0
0
0
0
0
155
11
0
0
0
0
0
0
0
0
16
No.
trees
35
17
1
9
19
13
11
12
14307
•Crown density classes, in percent:
        1=0              7 = 51 - 60
        2 =  1 -10          8 = 61 - 70
        3-11-20         9 = 71-80
        4 = 21 - 30        10 = 81 - 90
        5 = 31-40        11=91-99
        6 = 41 - 50
                                                 E-23

-------
Table E-14. Number of live trees > 12.7 cm DBH tallied on subplots, by species and crown dieback class.
Species
balsam fir
eastern redcedar
larch (introduced)
tamarack
white spruce
black spruce
blue spruce
red spruce
shortteaf pine
slash pine
spruce pine
longleaf pine
table-mountain pine
red pine
pitch pine
pond pine
eastern white pine
Scots pine
loblolly pine
Virginia pine
Austrian pine
baldcypress
pondcypress
northern white-cedar
hemlock species
eastern hemlock
Florida maple
boxelder
striped maple
red maple
sugar maple
yellow buckeye
service berry
birch species
yellow birch
sweet birch
river birch
paper birch
gray birch
American hornbeam
hickory species
water hickory
bittemut hickory
pignut hickory
shagbark hickory
mockemut hickory
American chestnut
hackberry species
hackberry (common)
eastern redbud
yellowwood
flowering dogwood
persimmon
American beech
ash species
1
373
28
1
10
63
14
2
465
215
398
6
96
9
5
15
13
432
1
1446
314
1
25
2
173
1
228
1
5
7
477
189
1
1
9
91
53
4
60
15
22
93
2
0
22
13
36
1
0
0
2
1
38
8
83
12
2
194
18
0
4
15
6
0
185
53
51
6
23
2
7
76
1
161
3
348
140
0
20
1
153
0
97
0
12
13
879
256
3
0
11
157
39
11
222
30
31
145
0
1
7
6
42
0
1
22
1
0
22
5
173
1
3
8
1
0
0
0
0
0
2
7
4
1
0
0
0
12
0
13
0
14
12
0
2
0
10
0
7
1
1
2
66
11
0
0
2
5
0
1
9
15
3
6
0
0
2
0
3
0
0
1
1
0
4
0
12
0
4
4
0
0
1
0
0
0
0
2
1
0
0
0
0
0
0
2
0
1
1
0
0
0
8
0
4
0
0
1
15
3
0
0
0
2
0
0
10
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
8
0
Crown Dieback Class'
5 6
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
0
1
0
0
0
0
2
0
1
0
0
0
10
2
0
0
0
0
0
0
0
3
0
1
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
3
2
0
0
0
0
0
0
0
0
0
5
0
0
0
0
1
0
0
3
1
0
1
0
0
0
0
1
0
0
0
0
0
1
0
3
0
7
2
1
0
1
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
2
0
0
0
0
0
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
9
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
0
0
0
0
0
0
0
0
0
2
0
1
0
0
1
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
• o
6
0
10
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
3
0
1
0
0
0
0
1
0
0
0
0
0
4
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
2
0
11
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
3
0
No.
trees
585
49
1
16
78
21
2
654
278
455
13
119
11
12
103
14
613
4
1814
471
1
47
3
350
1
338
2
18
23
1463
464
6
1
'22
257
92
16
309
64
57
246
2
1
31
19
82
1
1
23
5
1
69
13
294
13
                                            E-24

-------
Species 1
white ash 50
black ash 6
green ash 7
American holly 21
butternut 1
black walnut 5
sweetgum 343
yellow-poplar 276
osage-orange 0
magnolia species 3
cucumbertree 12
southern magnolia 3
sweetbay 50
apple species 5
Chinaberry 3
mutoerry species 0
red mulberry 2
water tupeto 49
blackgum 107
swamp tupeto 73
eastern hophombeam 14
sourwood 34
Paulownia 1
redbay 4
American sycamore 8
balsam poplar 0
eastern cottonwood 1
bigtooth aspen 8
quaking aspen 21
pin cherry 1
black cherry 36
choke cherry (common) 5
white oak 177
swamp white oak 0
scarlet oak 23
bluejack oak 0
northern pin oak 3
southern red oak 66
cherrybark oak 0
turkey oak 2
laurel oak 22
overcup oak 5
blackjack oak 2
swamp chestnut oak 3
chinkapin oak 4
water oak 105
willow oak 15
chestnut oak 87
northern red oak 57
Shumard oak 1
post oak 29
black oak 29
live oak 7
dwarf post oak 1
black locust 9
black willow 2
sassafras 8
2
101
5
16
35
4
8
228
106
1
4
3
8
40
2
0
0
5
8
75
54
10
49
0
2
5
7
3
32
115
2
64
2
235
4
105
2
3
105
7
5
11
3
15
1
0
141
22
298
243
0
40
90
12
1
19
3
24
3
11
3
3
0
0
0
16
5
0
0
0
1
0
0
0
0
2
0
8
8
1
5
0
1
0
1
0
2
12
1
9
0
19
0
23
0
1
5
0
0
0
0
5
0
0
17
2
21
29
0
3
9
2
0
3
1
3
Crown Dieback Class*
4567
2
4
0
0
0
0
8
0
0
0
0
0
0
0
0
1
0
0
0
2
0
1
0
0
0
0
0
1
3
0
4
0
6
0
4
0
0
1
0
0
1
0
1
0
0
3
0
4
4
0
1
4
0
0
0
0
0
5
2
2
0
0
1
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
2
0
0
0
1
0
4
0
0
1
0
0
0
0
2
0
0
1
0
0
1
0
0
1
0
0
1
0
0
2
0
0
0
0
0
4
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
1
0
0
0
0
3
0
0
0
0
0
2
0
0
0
0
1
0
0
0
0
0
1
0
2
0
0
1
0
0
1
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
1
2
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
1
0
0
8
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
3
0
0
0
0
0
0
0
1
0
2
0
0
0
0
0
0
0
0
0
9
0
0
1
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
10
1
2
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
1
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
1
0
0
0
0
0
0
11
0
0
0
0
0
1
2
2
0
0
0
0
2
0
0
0
0
0
0
1
0
1
0
1
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
2
0
0
No.
trees
172
23
29
56
5
15
605
390
2
7
15
12
93
7
3
1
9
57
190
142
27
92
1
9
13
9
4
45
160
4
114
7
440
4
163
2
7
181
7
8
34
8
26
4
4
271
40
415
334
1
76
133
21
3
36
6
35
E-25

-------
Species
American basswood
white basswood
elm species
winged elm
American elm
slippery elm
unknown / other
1
9
1
6
7
5
7
6
2
8
0
2
10
7
4
6
3
0
0
0
1
0
0
0
Crown Dieback Class'
4567
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
8
0
0
0
0
0
0
0
9
0
0
0
0
0
0
0
10
0
0
0
0
0
0
0
11
0
0
0
0
0
0
0
No.
trees
17
1
9
19
13
11
12
Total              7429   6074    471     120     55     44     24      17     23     27     23   14307


'Crown dieback classes, in percent:
        1=0               7 = 51 - 60
        2 = 1 -10           8 = 61-70
        3 = 11 - 20          9 = 71 - 80
        4 = 21-30         10 = 81-90
        5 = 31 - 40         11 = 91 - 99
        6 = 41 - 50
                                                  E-26

-------
Table E-15.  Number of live trees > 12.7 cm DBH tallied on subplots, by species and crown transparency
class.
Species
balsam fir
eastern redcedar
larch (introduced)
tamarack
white spruce
black spruce
blue spruce
red spruce
shortteaf pine
slash pine
spruce pine
longleaf pine
table-mountain pine
red pine
pitch pine
pond pine
eastern white pine
Scots pine
loblolly pine
Virginia pine
Austrian pine
babcypress
pondcypress
northern white-cedar
hemlock species
eastern hemlock
Florida maple
boxelder
striped maple
red maple
sugar maple
yetow buckeye
service berry
birch species
yellow birch
sweet birch
river birch
paper birch
gray birch
American hornbeam
hickory species
water hickory
bittemut hickory
pignut hickory
shagbark hickory
mockemut hickory
American chestnut
hackberry species
hackberry (common)
eastern redbud
yelowwood
flowering dogwood
persimmon
American beech
1
0
1
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
0
6
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
3
2
264
23
0
3
38
14
0
325
136
328
3
54
3
8
34
8
157
3
863
136
1
14
0
59
1
96
0
15
12
604
269
1
1
12
131
39
3
71
4
27
180
2
0
24
15
69
1
1
8
2
1
60
9
119
3
258
18
1
7
32
7
1
292
119
93
8
62
5
3
46
6
292
1
870
240
0
33
2
151
0
131
2
2
8
602
177
2
0
7
107
42
13
171
22
20
55
0
1
7
3
11
0
0
14
0
0
6
3
118
Crown Transparency Class*
4567
45
2
0
3
8
0
0
28
18
30
2
1
2
0
8
0
90
0
54
49
0
0
1
88
0
71
0
1
1
160
14
1
0
3
14
8
0
51
27
6
3
0
0
0
1
2
0
0
1
2
0
0
1
27
10
4
0
2
0
0
0
5
3
3
0
1
1
1
4
0
30
0
18
16
0
0
0
39
0
26
0
0
0
56
2
0
0
0
1
2
0
11
8
3
6
0
0
0
0
0
0
0
0
0
0
1
0
13
4
0
0
0
0
0
0
2
1
0
0
0
0
0
5
0
19
0
1
10
0
0
0
8
0
7
0
0
1
19
0
0
0
0
2
0
0
4
1
0
0
0
0
0
0
0
0
0
0
0
0
1
0
2
2
1
0
0
0
0
1
1
1
1
0
0
0
0
2
0
9
0
1
8
0
0
0
2
0
3
0
0
1
11
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
2
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
6
0
0
6
0
0
0
2
0
1
0
0
0
4
1
1
0
0
1
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
9
1
0
0
1
0
0
0
0
0
0
0
0
0
0
2
0
5
0
0
3
0
0
0
1
0
3
0
0
0
2
1
0
0
0
• o
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4
10
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
2
0
1
2
0
0
0
0
0
0
0
0
0
2
0
0
0
0
1
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
11
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
0
0
1
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
1
No.
trees
585
49
1
16
78
21
2
654
278
455
13
119
11
12
103
14
613
4
1814
471
1
47
3
350
1
338
2
18
23
1463
464
6
1
22
257
92
16
309
64
57
246
2
1
' 31
19
82
1
1
23
5
1
69
13
294
                                            E-27

-------
Species
ash species
white ash
black ash
green ash
American holly
butternut
black walnut
sweetgum
yeHow-poplar
osage-orange
magnolia species
cucumbertree
southern magnolia
sweetbay
apple species
Chinaberry
mulberry species
red mulberry
water tupeto
blackgum
swamp tupeb
eastern hophombeam
sourwood
Paulo wnia
redbay
American sycamore
balsam poplar
eastern cottonwood
bigtooth aspen
quaking aspen
pin cherry
black cherry
choke cherry (common)
white oak
swamp white oak
scarlet oak
bluejack oak
northern pin oak
southern red oak
cherrybark oak
turkey oak
laurel oak
overcup oak
blackjack oak
swamp chestnut oak
chinkapin oak
water oak
willow oak
chestnut oak
northern red oak
Shumard oak
post oak
black oak
live oak
dwarf post oak
black locust
black willow
1
0
0
0
0
0
0
0
3
1
0
0
0
0
0
0
0
0
0
0
1
1

0
0
0

0

1
0
0
0

0
0
0
0
0
0
0
0
1
0
1

0
0
0
1
2
0
0
1
0
0
0
0
2
7
55
2
10
37
2
7
490
312
1
6
14
11
72
1
1
0
6
45
128
63
0 13
60
1
8
0 6
2
0 0
11
17
1
43
0 6
260
2
65
1
6
121
7
1
25
6
15
0 4
4
164
29
202
122
1
49
59
6
0
8
3
3
6
86
17
6
19
3
6
99
63
1
0
1
1
19
4
0
0
3
12
48
58
6
23
0
0
5
7
2
19
73
2
46
1
131
2
72
1
1
41
0
3
8
2
7
0
0
97
8
113
136
0
20
55
12
1
11
3
Crown Transparency Class3
4567
0
21
1
6
0
0
1
9
7
0
1
0
0
1
2
1
0
0
0
10
18
5
7
0
1
0
0
1
5
36
1
17
0
29
0
17
0
0
10
0
3
0
0
2
0
0
9
0
62
45
0
4
12
3
1
11
0
0
6
2
4
0
0
1
2
2
0
0
0
0
0
0
1
1
0
0
2
1
1
1
0
0
1
0
0
0
8
0
5
0
5
0
3
0
0
8
0
1
0
0
0
0
0
1
1
16
15
0
2
2
0
0
2
0
0
1
0
1
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
3
0
0
0
7
0
4
0
0
0
0
0
0
0
1
0
0
0
0
7
5
0
0
2
0
1
4
0
0
2
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
2
0
1
0
1
0
2
0
0
0
0
0
0
0
0
0
0
0
0
6
5
0
0
2
0
0
0
0
8
0
1
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
3
0
0
0
0
0
0
0
0
9
0
0
0
1
0
0
0
1
1
0
0
0
0
1
0
0
0
0
0
0
0
2
0
0
0
0
0
0
1
15
0
1
0
3
0
0
0
0
1
0
0
0
0
0
0
0
0
0
3
1
0
0
0
0
0
0
0
10
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
4
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
2
0
0
0
0
0
0
0
11
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
6
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
1
0
0
0
0
0
No.
trees
13
172
23
29
56
5
15
605
390
2
7
15
12
93
7
3
1
9
57
190
142
27
92
1
9
13
9
4
45
160
4
114
7
440
4
163
2
7
181
7
8
34
8
26
4
4
271
40
415
334
1
76
133
21
3
36
6
E-28

-------
Species
sassafras
American basswood
white basswood
elm species
winged elm
American elm
slippery elm
unknown / other
1
0

0
0
0
0
0
0
2
20
0 6
1
4
9
7
7
6
3
15
8
0
2
7
3
3
6
Crown Transparency Class*
4567
0
0
0
2
2
2
0
0
0
1
0
1
1
1
0
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
8
0
0
0
0
0
0
0
0
9
0
0
0
0
0
0
0
0
10
0
1
0
0
0
0
0
0
11
0
0
0
0
0
0
0
0
No.
trees
35
17
1
9
19
13
11
12
Total
29   6909   5475   1188    364
130
72
39
54
27
20  14307
"Crown transparency classes, in percent:
        1=0               7 = 51 - 60
        2 = 1-10           8 = 61-70
        3 = 11 - 20          9 = 71 - 80
        4 = 21-30         10 = 81-90
        5 = 31-40         11 =91 -99
        6 = 41 - 50
                                                  E-29

-------
Table E-16.  Number of live trees > 2.5 cm DBH tallied on microplots, by species and crown ratio class.
Species
balsam fir
eastern redcedar
tamarack
white spruce
black spruce
red spruce
shortleaf pine
slash pine
longteaf pine
red pine
pitch pine
eastern white pine
loblolly pine
Virginia pine
baldcypress
northern white-cedar
hemlock species
eastern hemlock
boxelder
striped maple
red mapte
sugar maple
mountain maple
serviceberry
pawpaw
birch species
yellow birch
sweat birch
river birch
paper bjrch
gray birch
American hornbeam
hickory species
pignut hickory
shagbark hickory
mockemut hickory
American chestnut
haofcberry species
eastern redbud
yeflowwood
flowering dogwood
hawthorn species
persimmon
American beech
ash species
white ash
black ash
green ash
American holy
sweetgum
yellow-poplar
magnolia species
cucumbertree
southern magnolia
sweetbay
1
45
0
4
3
10
22
0
0
0
1
0
5
1
0
0
3
0
4
0
8
33
13
2
0
0
0
8
0
0
10
9
0
1
0
0
0
0
0
0
0
0
0
0
6
0
5
0
0
0
0
1
0
0
0
1
2
3
0
1
0
2
2
0
4
0
0
0
1
22
1
1
1
0
1
0
0
19
2
0
0
0
0
0
2
0
0
0
1
5
1
0
0
1
0
1
0
7
0
0
2
0
1
0
1
0
6
1
0
0
0
5
3
23
1
2
0
3
8
4
4
3
0
0
3
25
7
0
2
0
1
0
1
50
6
1
2
0
1
4
2
1
9
9
4
4
0
1
4
0
0
5
1
18
0
1
5
0
4
0
4
0
20
3
0
0
0
10
4
29
2
5
0
7
5
2
8
0
0
1
7
42
2
0
5
0
7
0
24
85
16
2
1
0
3
4
3
1
14
13
15
8
0
0
4
0
0
2
0
44
0
3
6
0
11
0
1
3
31
8
1
3
1
15
Crown Ratio Class*
567
41
0
1
1
15
13
8
2
3
0
0
4
35
5
0
2
0
3
1
7
109
21
1
2
0
0
7
9
0
14
8
14
16
0
0
7
0
1
3
0
58
0
2
8
2
6
2
5
3
51
13
1
0
0
17
60
2
2
2
12
21
4
0
3
0
6
9
43
2
0
9
0
5
0
19
85
22
0
1
0
3
12
7
0
12
7
11
17
1
1
8
0
0
6
0
42
1
4
18
2
10
1
5
4
49
20
4
1
1
12
34
6
1
1
10
15
4
3
0
0
2
6
36
2
0
2
0
4
1
7
63
10
0
0
0
2
7
4
0
13
3
16
12
0
0
4
0
1
2
0
25
0
1
21
1
3
1
1
6
45
10
2
0
0
0
8
23
4
0
0
6
14
1
3
0
0
1
5
48
2
0
3
0
5
0
12
33
11
0
0
0
1
12
2
0
4
0
4
7
0
0
4
0
0
0
0
18
0
0
11
1
3
0
1
5
41
11
0
1
1
0
9
51
4
0
0
5
18
0
3
0
0
2
4
40
5
0
3
0
9
0
5
24
13
0
2
1
0
2
0
0
4
0
3
12
0
0
0
1
0
0
0
12
0
1
19
0
4
0
0
7
26
7
1
0
0
0
10
48
2
0
4
2
14
0
6
0
0
1
7
16
1
0
1
0
9
0
2
4
3
0
0
1
1
3
1
0
0
0
2
6
0
0
0
0
0
0
0
4
0
0
16
0
0
0
0
7
12
2
0
0
0
1
11
43
2
1
7
4
13
1
0
0
0
0
6
17
1
0
5
1
10
0
0
1
1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
0
0
5
0
0
0
0
3
5
2
1
0
0
0
No.
trees
400
23
17
18
76
145
24
33
9
1
13
57
325
28
1
36
1
58
2
85
506
118
6
8
3
11
59
30
2
80
49
70
86
2
2
31
2
2
19
1
231
1
12
117
6
47
4
18
38
286
78
10
5
3
61
                                             E-30

-------
Crown Ratio
Species
apple species
water tupelo
blackgum
swamp tupelo
eastern hophombeam
sourwood
redbay
American sycamore
bigtooth aspen
quaking aspen
cherry, plum species
pin cherry
black cherry
white oak
swamp white oak
scarlet oak
bluejack oak
southern red oak
bear oak
turkey oak
laurel oak
blackjack oak
water oak
willow oak
chestnut oak
northern red oak
post oak
black oak
live oak
dwarf post oak
dwarf live oak
scrub oak
black locust
willow species
black willow
sassafras
American basswood
elm species
winged elm
American elm
slippery elm
sparkle berry
unknown / other
Total
1
0
0
0
0
3
0
0
0
0
10
0
9
4
5
0
0
0
0
0
0
2
0
1
0
0
5
1
1
0
0
0
0
0
1
0
1
0
0
1
1
0
1
0
241
2
0
0
4
2
0
3
0
0
0
1
0
0
3
1
0
0
2
2
0
1
1
0
0
1
1
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
121
3
0
0
13
5
2
6
1
0
1
2
0
4
8
3
0
1
0
1
0
1
3
0
9
3
2
7
1
2
0
0
0
1
1
0
1
6
0
0
3
0
1
0
17
361
4
1
3
28
10
5
7
1
2
0
3
0
10
15
11
0
3
0
9
0
1
15
0
28
2
4
10
3
2
1
0
0
0
0
0
0
1
0
1
4
0
0
1
13
638
5
1
2
33
8
10
14
1
0
0
1
2
8
19
10
0
2
0
7
0
0
13
1
22
2
3
12
3
4
0
2
0
0
1
1
0
6
0
1
3
1
2
1
7
724
6
3
1
32
3
7
7
0
0
1
4
0
3
13
15
0
1
1
11
0
0
12
0
29
4
2
11
3
2
2
0
0
1
0
0
0
2
1
0
5
1
0
3
5
751
Class"
7
1
1
16
3
2
4
1
0
0
2
0
3
10
13
1
2
3
7
1
0
7
0
13
4
5
1
4
3
0
1
0
0
0
0
0
2
0
0
3
3
1
0
3
507

8
0
1
12
1
9
2
0
0
1
0
0
0
9
5
0
0
0
6
0
2
1
1
19
2
3
3
2
3
0
0
0
0
0
0
0
0
0
0
5
0
0
0
2
387

9
0
0
11
1
2
2
0
0
0
3
0
1
2
5
0
1
1
7
0
0
4
2
16
2
0
3
2
2
0
0
1
0
0
0
0
0
0
• 0
0
0
1
0
0
357

10
0
0
1
0
2
2
0
0
0
2
0
0
1
4
0
0
0
1
0
0
3
0
10
2
2
0
1
1
0
0
0
1
0
0
0
0
0
0
1
0
0
0
0
210

11
0
0
3
0
0
0
0
1
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
141
No.
trees
6
8
153
33
42
47
4
3
3
28
2
39
84
73
1
10
7
51
1
5
61
4
147
23
22
53
20
21
3
3
1
3
2
2
1
18
1
3
25
6
5
6
50
4438
'Crown ratio classes, in percent:
1 =0
2 = 1-10
3= 11 -20
4 = 21 - 30
5 = 31 - 40





7 = 51
8 = 61
9 = 71
10 = 81
11 =91
-60
-70
-80
-90
-99













































6 = 41 - 50
                                         E-31

-------
Table E-17.  Frequency of visible tree damage types on all live trees greater than 12.7 cm DBH, by
            species.
Total no.
Species trees damages
balsam fir
eastern redcedar
larch (introduced)
tamarack
white spruce
black spruce
blue spruce
red spruce
shortteaf pine
slash pine
spruce pine
longteaf pine
table-mountain pine
red pine
pitch pine
pond pine
eastern white pine
Scots pine
loblolly pine
Virginia pine
Austrian pine
baldcypress
pondcypress
northern white-cedar
hemlock species
eastern hemlock
Florida maple
boxelder
striped maple
red maple
sugar maple
yellow buckeye
service berry
birch species
yeUow birch
sweet birch
river birch
paper birch
gray birch
American hornbeam
hickory species
water hickory
bittemut hickory
pignut hickory
shagbark hickory
mockemut hickory
American chestnut
hackberry species
hackberry (common)
eastern redbud
yellowwood
flowering dogwood
persimmon
585
49
1
16
78
21
2
654
278
455
13
119
11
12
103
14
613
4
1814
471
1
47
3
350
1
338
2
18
23
1463
464
6
1
22
257
92
16
309
64

246
2
1
31
19
82
1
1
23
5
1
69
13
76
22
0
2
3
0
0
46
75
94
9
28
4
7
22
5
90
0
438
197
0
8
0
59
0
69
1
7
11
588
140
4
0
7
64
21
7
87
26
5730
140
2
1
13
3
90
1
0
21
6
1
67
7
0
513
31
1
14
75
21
2
610
205
368
7
96
7
5
85
9
532
4
1416
297
1
41
3
298
1
276
1
11
13
975
351
2
1
15
206
74
11
233
42
33
134
0
0
20
16
20
0
1
7
1
0
22
8
1
3
1
0
1
0
0
0
2
0
1
0
0
0
0
1
0
6
0
4
0
0
0
0
9
0
3
0
0
0
31
3
1
0
0
4
0
1
6
0
1
3
0
0
1
0
4
0
0
0
2
0
6
0
Damage type'
2 3
number
17
6
0
0
2
0
0
16
9
12
1
3
1
0
2
0
12
0
39
16
0
0
0
28
0
9
0
0
7
188
63
0
0
1
22
8
3
39
3
7
26
0
1
1
0
5
0
0
10
2
0
8
1
3
3
0
0
0
0
0
1
5
40
0
4
0
0
9
2
1
0
179
59
0
2
0
0
0
0
0
0
1
40
15
0
0
2
8
1
1
7
0
2
12
0
0
0
0
6
1
0
0
2
1
2
1
4
2
1
0
0
1
0
0
2
2
2
0
6
0
6
0
0
1
0
22
11
0
1
0
1
0
7
0
1
0
21
2
0
0
0
0
0
0
1
0
0
18
0
0
2
0
35
0
0
2
0
0
1
0
5
21
4
0
0
0
0
0
8
1
1
3
3
0
0
0
0
2
0
10
5
0
0
0
5
0
4
0
0
2
48
12
0
0
2
8
2
0
15
4
1
8
0
0
0
0
5
0
0
0
0
0
1
0
6
12
1
0
0
0
0
0
7
1
2
0
2
0
0
6
0
15
0
13
3
0
2
0
4
0
6
0
0
1
42
5
0
0
1
6
1
0
2
2
4
9
1
0
2
0
4
0
0
0
0
0
3
0
7
0
0
0
1
0
0
0
0
1
0
0
0
0
0
0
1
1
0
16
4
0
0
0
0
0
0
0
0
0
4
2
0
0
0
2
2
0
3
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
                                             E-32

-------
Total no.
Species trees damages
American beech
ash species
white ash
black ash
green ash
American holly
butternut
black walnut
sweetgum
yellow-poplar
o sage-orange
magnolia species
cucumbertree
southern magnolia
sweetbay
apple species
Chinaberry
mulberry species
red mulberry
water tupeto
blackgum
swamp tupelo
eastern hophornbeam
sourwood
Paubwnia
redbay
American sycamore
balsam poplar
eastern cottonwood
bigtooth aspen
quaking aspen
pin cherry
black cherry
294
13
172
23
29
56
5
15
605
390
2
7
15
12
93
7
3
1
9
57
190
142
27
92
1
9
13
9
4
45
160
4
114
choke cherry (common) 7
white oak
swamp white oak
scarlet oak
bluejack oak
northern pin oak
southern red oak
cherrybark oak
turkey oak
laurel oak
overcup oak
blackjack oak
swamp chestnut oak
chinkapin oak
water oak
willow oak
chestnut oak
northern red oak
Shumard oak
post oak
black oak
live oak
dwarf post oak
440
4
163
2
7
181
7
8
34
8
26
4
4
271
40
415
334
1
76
133
21
3
139
9
44
2
30
9
5
12
317
216
2
8
4
1
75
1
0
1
15
22
128
70
10
88
0
4
8
3
1
16
39
1
58
5
285
1
149
0
1
111
5
2
10
7
22
1
0
175
24
431
125
1
54
92
15
1
0
178
5
135
21
7
48
1
8
347
207
1
0
11
11
36
6
3
0
1
35
102
85
21
30
1
5
7
6
3
31
124
3
71
3
224
3
60
2
6
101
2
6
27
3
11
3
4
139
21
135
230
0
36
64
8
2
1
3
0
4
1
4
0
0
3
15
2
0
0
0
0
2
0
0
1
0
0
3
1
0
6
0
1
0
0
0
2
2
0
6
0
12
0
14
0
0
5
0
0
1
1
1
0
0
2
0
19
5
0
4
6
1
0
Damage type'
2 3
number
59
2
15
0
5
5
2
1
62
31
0
2
2
0
10
0
0
0
2
2
21
13
1
13
0
0
1
2
0
4
6
1
5
3
27
0
9
0
0
6
0
0
1
1
1
0
0
18
2
27
11
0
6
20
2
0
7
0
1
0
3
1
1
0
40
14
0
1
2
0
8
0
0
0
0
4
9
12
1
3
0
0
0
0
0
0
3
0
3
1
31
0
21
0
0
10
0
0
1
1
2
1
0
25
2
12
9
0
6
8
3
0
4
8
0
0
0
1
0
0
0
10
14
0
0
0
0
3
0
0
0
0
1
2
1
2
2
0
0
4
0
0
0
0
0
2
0
21
0
29
0
0
13
0
0
0
3
2
0
0
7
3
81
13
0
3
9
0
0
5
16
0
5
0
1
0
0
1
18
9
0
1
0
0
3
0
0
0
0
0
5
0
0
2
0
0
0
0
0
1
2
0
4
0
34
0
7
0
0
16
2
0
0
0
3
0
0
10
4
31
15
1
2
6
1
0
6
4
0
2
0
1
0
1
0
22
7
0
0
0
0
4
0
0
0
1
1
12
4
1
1
0
0
1
1
0
0
1
0
7
0
9
0
2
0
0
1
0
0
0
0
2
0
0
5
2
14
4
0
4
1
0
0
7
3
0
0
1
0
0
0
1
3
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
3
0
3
0
2
0
2
0
0
0
0
0
0
0
0
0
0
4
0
1
1
0
0
0
0
0
E-33

-------
Total no.
Species trees damages
0
Damage type*
1 2 3
4
5
6
7
number
black locust
black willow
sassafras
American basswood
white basswood
ekn species
winged elm
American elm
slippery elm
unknown / other
36
6
36
17
1
9
19
13
11
12
45
4
18
11
1
10
10
4
8
6
8
3
22
7
0
2
11
9
5
7
4
0
4
0
0
0
0
1
0
0
7
2
4
2
0
0
2
0
0
1
2
0
1
0
1
2
0
1
0
1
5
0
0
1
0
1
2
0
0
0
2
0
0
1
0
0
0
0
1
2
1
0
2
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
0
Total             14307   5571    9829   230    987    664    391    381    276     66
                                                  E-34

-------
Table E-17, continued
Species
           Damage type
10    11     12     13
              number
14
15
16
                   17
balsam fir
eastern redcedar
larch (introduced)
tamarack
white spruce
black spruce
blue spruce
red spruce
shortteaf pine
slash pine
spruce pine
longteaf pine
table-mountain pine
red pine
pitch pine
pond pine
eastern white pine
Scots pine
lobloly pine
Virginia pine
Austrian pine
baldcypress
pondcypress
northern white-cedar
hemlock species
eastern hemlock
Florida maple
boxelder
striped maple
red maple
sugar maple
yellow buckeye
servfceberry
birch species
yellow birch
sweet birch
river birch
paper birch
gray birch
American hornbeam
hickory species
water hickory
bittemut hickory
pignut hickory
shagbark hickory
mockemut hickory
American chestnut
hackberry species
hackberry (common)
eastern redbud
yeHowwood
flowering dogwood
persimmon
American beech
2
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
3
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
1
0
0
0
0
0
0
0
6
0
0
0
0
0
0
0
3
0
3
6
0
0
0
0
0
31
0
1
0
9
0
1
0
0
0
0
0
1
1
0
5
0
0
0
0
6
0
0
0
0
0
3
1
6
0
0
0
0
0
0
0
4
1
17
0
4
0
0
0
0
10
0
9
5
0
0
0
0
0
0
1
1
0
3
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
1
0
2
0
0
0
0
0
0
27
10
2
3
2
0
1
2
27
0
76
66
0
2
0
2
0
3
0
2
0
74
9
0
0
0
1
1
0
2
1
4
14
1
0
0
1
16
0
0
6
0
0
6
2
1
0
2
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
5
1
0
0
0
0
0
0
0
0
0
11
4
0
0
0
1
0
1
0
0
0
6
0
0
0
0
0
0
0
1
0
0
3
0
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
6
1
0
1
0
0
0
0
0
0
0
5
0
0
0
0
1
0
0
0
0
1
3
0
0
1
0
0
0
0
1
0
0
2
0
1
0
0
0
0
0
0
0
0
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
16
5
0
0
0
1
1
0
2
0
0
14
0
0
0
1
3
0
0
0
0
0
8
0
2
2
0
0
0
0
0
0
0
5
0
1
0
0
0
0
0
0
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
0
0
0
0
0
0
0
2
1
1
0
0
0
0
0
0
0
5
0
0
0
0
0
0
3
0
0
0
39
13
0
0
0
6
2
1
8
1
4
7
0
0
3
0
1
0
0
0
0
0
9
0
5
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
                                              E-35

-------
Species
ash species
white ash
black ash
green ash
American holy
butternut
black walnut
sweetgum
yellow-poplar
osage -orange
magnolia species
cucumbertree
southern magnolia
sweetbay
apple species
Chinaberry
mulberry species
red mulberry
water tupeb
blackgum
swamp tupeb
eastern hophombeam
sourwood
Paukjwnia
redbay
American sycamore
balsam poplar
eastern cottonwood
bigtooth aspen
quaking aspen
pin cherry
black cherry
choke cherry (common)
white oak
swamp white oak
scarlet oak
bluejackoak
northern pin oak
southern red oak
cherry bark oak
turkey oak
laurel oak
overcup oak
blackjack oak
swamp chestnut oak
chinkapin oak
water oak
willow oak
chestnut oak
northern red oak
Shumard oak
post oak
black oak
live oak
dwarf post oak
black locust
8
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
1
0
0
0
9
0
0
0
2
0
0
0
1
4
0
0
0
0
1
0
0
0
0
0
5
4
2
1
0
0
0
0
1
8
19
0
1
0
9
0
9
0
0
0
0
0
0
0
1
0
0
1
1
77
12
0
2
11
0
0
4
10
0
0
0
0
0
0
0
4
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
2
0
0
1
1
0
0
0
0
0
0
15
0
1
4
0
0
1
0
0
0
11
5
3
0
5
0
0
4
55
79
0
2
0
0
15
0
0
0
1
12
38
11
0
40
0
2
1
0
0
0
0
0
16
1
26
0
9
0
1
11
0
0
2
0
1
0
0
25
2
89
15
0
6
6
1
0
7
Damage type
12 13
number
0
0
0
2
0
0
0
8
4
0
1
0
0
7
0
0
0
3
0
2
10
0
2
0
1
0
0
0
0
0
0
1
0
11
0
11
0
0
3
0
1
1
0
1
0
0
4
0
8
3
0
1
2
0
0
0
0
0
0
0
0
0
0
2
2
0
0
0
0
4
0
0
0
2
0
2
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
0
0
0
0
1
0
0
2
0
2
0
0
1
0
0
0
0
14
1
2
0
1
0
0
0
48
24
0
0
0
0
5
0
0
0
0
0
17
6
0
2
0
0
0
0
0
0
0
0
3
0
56
0
9
0
0
26
0
0
3
1
1
0
0-
33
4
26
16
0
9
4
2
0
0
15
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
16
0
3
0
1
1
0
0
13
3
1
1
0
0
6
0
0
0
5
1
5
4
1
3
0
0
0
0
0
0
1
0
2
0
20
0
15
0
0
13
2
0
1
0
6
0
0
15
1
17
6
0
7
9
0
0
0
17
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
E-36

-------
Species
black willow
sassafras
American basswood
white basswood
elm species
winged elm
American elm
slippery elm
unknown / other
8
0
0
0
0
0
0
0
0
0
9
0
0
4
0
1
0
0
0
0
10
0
0
0
0
0
0
0
0
0
11
1
6
1
0
0
1
1
2
2
Damage type
12 13
number
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
2
0
14
0
0
2
0
3
0
0
2
0
15
0
0
0
0
0
0
0
0
0
16
0
0
0
0
0
0
0
0
0
17
0
0
0
0
0
0
0
0
0
Total                14    265     96    871    125     49    362     13    277
                                                 E-37

-------
Table E-17, continued.
Species
balsam fir
eastern redcedar
larch (introduced)
tamarack
white spruce
black spruce
blue spruce
red spruce
snortleaf pine
slash pine
spruce pine
tongteaf pine
table-mountain pine
red pine
pitch pine
pond pine
eastern white pine
Scots pine
loblolly pine
Virginia pine
Austrian pine
baldcypress
pondcypress
northern white-cedar
hemlock species
eastern hemlock
Florida maple
boxelder
striped maple
red maple
sugar maple
yellow buckeye
serviceberry
birch species
yelow birch
sweet birch
river birch
paper birch
gray birch
American hornbeam
hickory species
water hickory
bittemut hickory
pignut hickory
shagbark hickory
mockemut hickory
American chestnut
hackberry species
hackberry (common)
eastern redbud
yeUowwood
flowering dogwood
persimmon
American beech
18
0
0
0
0
0
0
0
0
1
5
0
2
1
0
1
0
2
0
5
5
0
0
0
0
0
0
0
0
0
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
1
19
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
3
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
3
0
0
2
1
2
0
0
0
0
0
0
0
0
20
7
0
0
0
0
0
0
4
5
2
0
0
0
1
0
0
0
0
12
9
0
0
0
9
0
3
0
2
0
26
2
0
0
1
2
2
0
1
14
5
4
0
0
1
0
0
0
0
1
0
0
11
0
2
21
2
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
5
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
3
0
0
0
0
1
0
0
0
0
0
0
0
0
Damage type
22 23
number
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
3
0
0
0
0
0
0
0
2
0
8
0
0
0
0
0
0
0
0
0
0
2
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
24
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
6
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
25
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
2
26
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
27
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
99
1
1
0
0
0
0
0
0
1
0
0
0
0
0
1
0
5
0
16
1
0
0
0
0
0
0
0
0
0
11
2
0
0
0
2
1
0
0
0
0
1
0
0
0
0
1
0
0
0
0
0
1
0
16
                                             E-38

-------
Species
ash species
white ash
black ash
green ash
American holly
butternut
black walnut
sweetgum
yelow-poplar
o sage-orange
magnolia species
cucumbertree
southern magnolia
sweetbay
apple species
Chinaberry
mulberry species
red mulberry
water tupeto
blackgum
swamp tupelo
eastern hophombeam
sourwood
Paulo wnia
redbay
American sycamore
balsam poplar
eastern cottonwood
bigtootn aspen
quaking aspen
pin cherry
black cherry
choke cherry (common)
white oak
swamp white oak
scarlet oak
bluejack oak
northern pin oak
southern red oak
cherrybark oak
turkey oak
laurel oak
overcup oak
blackjack oak
swamp chestnut oak
chinkapin oak
water oak
willow oak
chestnut oak
northern red oak
Shumard oak
post oak
black oak
live oak
dwarf post oak
black locust
18
1
0
0
0
0
0
1
1
2
0
0
0
0
0
0
0
0
0
0
1
0
0
2
0
0
0
0
0
0
0
0
0
0
4
0
2
0
0
0
0
0
0
0
0
0
0
0
2
12
2
0
0
3
0
0
4
19
0
2
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Q
0
0
0
0
7
0
0
0
0
2
0
0
0
0
0
0
0
1
0
1
0
0
1
0
0
0
0
20
0
2
0
2
0
1
1
5
2
0
0
0
0
6
1
0
0
1
1
3
0
0
8
0
0
1
0
0
0
0
0
4
0
3
0
1
0
0
2
0
0
0
0
0
0
0
4
0
8
5
0
2
3
4
0
1
21
0
1
0
1
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
1
0
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Damage type
22 23
number
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
24
0
0
0
0
0
0
0
4
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
25
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
2
0
0
0
2
0
3
0
0
0
0
0
0
0
0
0
0
0
1
3
1
0
0
0
0
0
0
26
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
27
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
99
0
3
0
0
2
0
0
3
3
0
0
0
1
1
0
0
0
0
0
1
2
0
1
0
0
0
0
0
0
0
0
0
0
7
1
0
0
0
1
0
1
0
0
0
0
0
4
0
0
3
0
0
2
1
1
6
E-39

-------
Species
black willow
sassafras
American basswood
white basswood
elm species
winged elm
American elm
slippery elm
unknown / other
18
0
0
0
0
0
0
0
1
0
19
0
0
0
0
0
0
0
0
0
20
1
1
0
0
1
1
0
0
0
21
0
0
0
0
0
0
0
0
0
Damage type
22 23
number
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
24
0
0
0
0
0
0
0
0
0
25
0
0
0
0
2
0
0
0
0
26
0
0
0
0
0
0
0
0
0
27
0
0
0
0
0
0
0
0
0
99
0
0
0
0
0
0
1
0
0
Total
72
30
201
23
28
13
22
106
'Damage types (Chojnacky 1991):

 0       none
 1       dead part of a live tree
 2       open wound
 3       closed wound
 4       small holes or pinhotes
 5       crack or seam
 6       broken part
 7       conk or fruiting bodies or fuzz on surface or mold
 8       removed or missing part
 9       defoliation
10       resinosis or bleeding
11       crook or sweep
12       swelling
13       excessive branching
14       abundance of epteormic branches or water sprouts on trunk or base
15       abundance of seeds or cones
16       rotten branch stubs or excessive swelling at base of dead branches
17       stunted or dwarfed woody stem
18       deformed or twisted or curled woody stem
19       imbedded foreign object
20       leaning
21       general discoloration
22       pale green foliage
23       yellow-green foliage
24       leaves spotted
25       damaged leaves
26       distorted foliage
27       stunted foliage
99       other damage not listed above
                                                      E-40

-------
Table E-18. Numbers of seedlings and saplings tallied by species and vigor class on microplots.
Species
balsam fir
Atlantic white-cedar
eastern redcedar
tamarack
white spruce
black spruce
red spruce
shortteaf pine
slash pine
spruce pine
longteaf pine
red pine
pitch pine
eastern white pine
loblolly pine
Virginia pine
baldcy press
pondcypress
northern white-cedar
hemlock species
eastern hemlock
maple species
Florida maple
boxelder
striped maple
red maple
sugar maple
mountain maple
total
3017
7
60
3
29
91
529
15
44
2
5
1
6
220
636
105
8
1
244
0
254
2
76
61
940
3292
1696
236
horsechestnut species 1 7
Ohio buckeye
yellow buckeye
serviceberry
pawpaw
birch species
yellow birch
sweet birch
river birch
paper birch
gray birch
American hornbeam
hickory species
bittemut hickory
pignut hickory
pecan
shagbark hickory
mockemut hickory
American chestnut
catalpa species
hackberry species
hackberry (common)
eastern redbud
yellowwood
flowering dogwood
hawthorn species
persimmon
1
20
42
40
5
367
215
0
640
194

477
6
18
4
2
148
71
1
1
7
85
0
1248
31
196
Seedlings
vigor class*
1 2
2947
6
31
1
27
91
510
8
42
2
4
1
6
196
512
57
2
1
238
-
218
0
65
9
841
2283
1536
216
7
1
12
23
15
1
351
150
-
622
110
34175
190
1
7
0
2
54
19
0
0
3
23
-
549
14
82
45
1
25
0
0
0
10
7
2
0
1
0
0
17
105
41
6
0
6
-
34
2
9
37
85
847
135
16
1
0
1
14
25
4
11
55
-
16
78
152
237
4
11
4
0
68
48
0
0
2
54
-
603
17
99
3
25
0
4
2
2
0
9
0
0
0
0
0
0
7
19
7
0
0
0
-
2
0
2
15
14
162
25
4
9
0
7
5
0
0
5
10
-
2
6
14
50
1
0
0
0
26
4
1
1
2
8
-
96
0
15
Saplings
total vioor class
1 2
400
0
23
17
18
76
145
24
33
0
9
1
13
57
325
28
1
0
36
1
58
0
0
2
85
506
118
6
0
0
0
8
3
11
59
30
2
80
49
70
88
0
2
0
2
31
2
0
2
0
19
1
231
1
12
321
-
12
11
15
62
117
13
17
-
5
0
12
39
184
9
0
-
23
1
43
-
-
0
66
262
85
4
-
-
-
5
0
9
46
12
0
61
31
29
20
-
1
-
1
8
1
-
1
-
7
0
102
1
3
33
-
8
1
0
3
3
11
14
-
4
0
1
10
106
13
0
-
9
0
9
-
-
2
11
169
17
0
-
-
-
3
3
2
4
15
1
7
7
38
57
-
0
-
1
21
0
-
1
-
10
0
106
0
9
3
1
-
3
1
0
1
3
0
2
-
0
0
0
3
34
6
1
-
1
0
2
-
-
0
1
42
3
2
-
-
-
0
0
0
1
3
1
2
2
3
11
-
1
-
0
2
1
-
0
-
2
1
23
0
0
                                              E-41

-------
Species
American beech
ash species
white ash
black ash
green ash
honeybcust
sitverbeU species
American holy
black walnut
total
523
19
467
32
213
2
6
229
3
sweetgum 1307
yellow-poplar
magnolia species
cucumbertree
southern magnolia
sweetbay
bigteaf magnolia
apple species
Chinaberry
mulberry species
red mulberry
water tupeto
blackgum
swamp tupeto
eastern hophombeam
sourwood
redbay
American sycamore
balsam poplar
bigtooth aspen
quaking aspen
cherry, plum species
pin cherry
black cherry
516
34
21
11
185
2
8
2
5
11
51
787
79
156
300
29
1
27
16
202
7
171
695
choke cherry (common) 17
white oak
swamp white oak
scarlet oak
bluejack oak
northern pin oak
southern red oak
cherrybark oak
bear oak
turkey oak
laurel oak
blackjack oak
swamp chestnut oak
chinkapin oak
water oak
pin oak
willow oak
chestnut oak
northern red oak
post oak
black oak
live oak
dwarf post oak
519
15
96
50
2
381
8
73
19
234
41
7
17
853
4
212
244
396
137
191
29
0
Seedlings
vigor class*
1 2
384
16
394
31
92
2
6
88
0
835
254
27
14
6
107
2
8
0
3
9
41
382
18
67
171
13
1
15
9
176
2
126
427
15
232
0
33
43
1
216
2
72
6
161
29
1
4
581
1
103
62
212
76
88
18
-
115
3
68
1
110
0
0
116
1
422
225
7
7
5
72
0
0
0
1
2
10
351
58
84
116
14
0
7
6
11
4
45
233
2
261
5
59
7
1
147
6
1
10
63
9
5
9
235
3
92
158
165
55
83
9
-
3
24
0
5
0
11
0
0
25
2
50
37
0
0
0
6
0
0
2
1
0
0
54
3
5
13
2
0
5
1
15
1
0
35
0
26
10
4
0
0
18
0
0
3
10
3
1
4
37
0
17
24
19
6
20
2
-
Saplings
total vioor class
1 2
117
6
47
4
18
0
0
38
0
286
78
10
5
3
61
0
6
0
0
0
8
153
33
42
47
4
3
0
3
28
2
39
84
0
73
1
10
7
0
51
0
1
5
61
4
0
0
147
0
23
22
53
20
21
3
3
77
3
39
4
2
-
-
14
-
144
33
4
2
0
23
-
4
-
-
-
2
48
7
19
11
1
0
-
0
16
2
16
43
-
36
0
2
2
-
15
-
1
0
49
2
-
-
76
-
11
10
28
6
5
0
1
28
3
2
0
11
-
-
22
-
118
39
5
2
3
30
-
0
-
-
-
5
90
23
20
33
2
2
-
3
2
0
13
30
-
26
1
7
3
-
33
-
0
4
7
2
-
-
66
-
8
7
20
12
12
3
2
3
6
0
1
0
5
-
-
2
-
24
5
1
1
0
8
-
2
-
-
-
1
15
3
0
3
1
1
-
0
0
0
1
7

6
0
1
2
-
3
-
0
1
3
0
-
-
5
-
4
5
0
2
4
0
0
E-42

-------
Species
dwarf live oak
scrub oak
black locust
willow species
black willow
sassafras
total
48
81
36
1
17
612
American mountain-ash 31
American basswood 14
white basswood
elm species
winged elm
American elm
slippery elm
rock elm
sparkteberry
unknown / other
Total
1
75
110
54
31
2
18
138
26688
Seedlings
vigor class'
1 2
37
52
21
1
16
289
18
7
0
62
31
33
10
1
1
110
18632
11
25
13
0
1
286
13
6
1
10
71
20
19
0
9
15
6914
3
0
4
2
0
0
37
0
1
0
3
8
1
2
1
8
13
1142
Saplings
total vigor class
1 2
1
3
2
2
1
18
0
1
0
3
25
6
5
0
6
50
4438
1
1
0
1
0
3
-
1
-
1
10
2
1
-
2
31
2451
0
2
1
0
1
10
-
0
-
2
14
3
3
-
4
15
1463
3
0
0
1
0
0
5
-
0
-
0
1
0
1
-
0
4
294
•Vigor classification (Chojnacky 1991):
Class 1 plants have:
Class 2 plants have:
Class 3 plants have:
                          more than 1/3 of the crown in foliage
                          no dieback in the upper 1/2 of the crown
                         • more than 4/5 of the foliage normal
                         • any amount of crown in foliage
                         • any amount of dieback, anywhere
                         • between 1/5 and 4/5 of the foliage normal
                         • any amount of crown in foliage
                         • any amount of dieback, anywhere
                         • less than 1/5 of the foliage normal
                                                        E-43
•ft U.S. GOVERNMENT PRINTING OFFICE:  199« - 650-006/00219

-------