United States
            Environmental Protection
            Agency
            Environmental Research
            Laboratory
            Corvallis OR 97330
EPA-600, 3-80-002
January 1980
            Research and Development
&EPA
Photochemical
Oxidant Air Pollution
Effects on a  Mixed
Conifer Forest
Ecosystem

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology.  Elimination  of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental Health Effects Research
      2.  Environmental Protection Technology
      3.  Ecological Research
      4.  Environmental Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific  and Technical Assessment Reports (STAR)
      7  Interagency Energy-Environment Research and Development
      8.  "Special"  Reports
      9.  Miscellaneous Reports

This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and materials.  Problems  are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                                        EPA-600/3-80-002
                                                        January 1980
          PHOTOCHEMICAL OXIDANT AIR POLLUTION EFFECTS ON A
                  MIXED CONIFER FOREST ECOSYSTEM
                            Final Report
                               Editor

                            0. C.  Taylor


                        Principal  Authors

              R. N. Kickert, J. R. McBride,  P.R.  Miller,
              C. P. Ohmart, R. J.  Arkley, D. L.  Dahlsten,
                 F. W. Cobb, Jr.,  J. R. Parmeter, Jr.,
                     R. F. Luck, and 0. C. Taylor

                       University of California
                    Riverside, California 92521  and
                      Berkeley, California 94720

                        Contract No. 68-03-2442

                            Project Officer

                             R. G. Wilhour

                         Terrestrial Division
             Corvallis Environmental Research Laboratory
                       Corvallis,  Oregon 97330

          This study conducted under the direction of
0. C. Taylor, Project Principal Investigator and Associate Director
              Statewide Air Pollution Research Center
                     University of California
                    Riverside, California 92521
             CORVALLIS ENVIRONMENTAL RESEARCH LABORATORY
                 OFFICE OF RESEARCH AND DEVELOPMENT
                U.S. ENVIRONMENTAL PROTECTION AGENCY
                     CORVALLIS,  OREGON 97330

-------
                                 DISCLAIMER
     This report has been reviewed by the Corvallis Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for pub-
lication.  Approval does not signify that the contents necessarily reflect
the views and policies of the U.S. Environmental Protection Agency, nor
does mention of trade names or commercial products constitute endorsement
or recommendation for use.
                                    ii

-------
                                  FOREWORD

     The San Bernardino National Forest  (SBNF) has been under stress from
photochemically produced oxidant air pollutants for more than three decades.
With the rapid industrial and urban growth in the South Coast Air Basin
during the past 20 to 30 years the impact on forest species has intensified.
Loss of ponderosa and Jeffrey pine trees has increased dramatically as
pollutant levels have risen and frequency and length of pollutant attacks
have expanded.  Pollutant effects on interrelated subsystems of the SBNF
ecosystem have been studied in 18 plots  established in selected regions of
the forest. The plots were selected to represent sites of varying pollutant
dosages while retaining as much uniformity of plant species and environment
as possible. Studies by scientists from  the Berkeley and-Riverside campuses
of the University of California collected data which will be used for a
group of linked models.  The models will aid in describing pollutant impact
on subsystems and such models should be  useful in anticipating or predicting
responses in other areas under similar conditions.  Data gathered during the
period of this contract will add significantly to information collected
under previous contracts and during the  subsequent two years of an EPA grant
for the purpose of refining the models of a western coniferous forest
ecosystem under stress from long—term exposure to photochemically produced
air pollutants.
                                    iii

-------
                                 ABSTRACT

     EPA Contract 68-03-2442 provided support for three years of the studies
to determine the chronic effects of photochemical oxidant air pollutants
on a western mixed conifer forest ecosystem.  A progress report for the
years 1974-'75 and 1975-'76 was published in the Ecological Research Series,
EPA-600/-3-77-104.  The report being submitted deals specifically with the
year 1976-'77 and is the final report for EPA Contract 68-03-2442 which has
funded a three year portion of the study initiated in 1972 and is scheduled
to terminate May 31, 1980.

     A computer data bank was partially developed in the early years of
the study at the Lawrence Livermore Laboratory and was subsequently revised
and moved to the computer at the University of California, San Francisco.
Verification and auditing of datasets is well underway and several sets are
now ready for cross-disciplinary analysis for modeling.  Computer simulation
programs have been written for some of the subsections.

     Subsystems which received greatest attention during this study period
were:  major tree species response to oxidant dose; tree population dynamics;
tree growth; moisture dynamics; soil chemical and physical properties;
tree mortality relative to disease, insects and other factors; epidemiology
of forest tree pathogens with emphasis on Fomes annosus; cone and seed
production; tree seedling establishment; litter production and litter
decomposition relative to microfloral decomposer populations.  Progress is
being made in preparation of models for the purpose of describing the
behavior of the interlinked subsystems.  Since much progress has been made
in verifying accuracy of data and of identifying information in the data
bank the study of subsystems interaction should be accelerated.

     This report in conjunction with the Ecological Research Series report
EPA-600/3-77-104 is submitted to fulfill the requirement for a final report
for EPA Contract 68-03-2442.
                                    iv

-------
                                  CONTENTS
                                                                   Page

Disclaimer Notice	•	       ii
Foreword	      iii
Abstract	•	       iv
Contents 	       v
Figures	       vi
Tables  	       x
Abbreviations and Symbols   	  .....      xiv
Acknowledgements 	  ..........       xv

     1.  Introduction  	 1

     2«  Ecosystem  Simulation Modeling  	       3

     3.  Tree Population Dynamics Subsystem  	       28

     4.  Canopy Response Subsystem  	       38

     5.  Tree Growth Subsystem	       65

     6.  Physical and  Chemical  Properties of  Soil, Including
         Moisture Dynamics  	       77

     7.  Stand Tree Mortality Subsystem:  Bark Beetle Popu-
         lation Dynamics	       86

     8.  Forest Tree Pathogen Epidemiology  Subsystem  	      115

     9.  Cause and  Extent of Tree Mortality	      132

    10.  Tree Seedling Establishment Subsystem 	      136

    11.  Cone & Seed Production For Dominant  Conifer Tree
         Reproduction  	      143

    12.  Litter Production  Subsystem 	      150

    13.  Foliage Litter Decomposition Subsystem:  Microbial
         Activity and Nutrient  Cycling  	      167

References	•	      179
Appendix	      185

-------
                                   FIGURES

Number                                                                  Page

    1   The system simulation modeling process  with  bold arrows
        showing information  flows  which  were  behind  schedule and
        consequently were  retarding modeling  progress  	      4

    2   Information structure for  the computerized SBNF  data base
        accesssible from remote  dial-up  terminal  via UCSF/CMS
        time-sharing mode  	      8

    3   Example of Part III  of the SBNF  data  dictionary:  Dataset
        descriptors	     11

    4   Data processing environments used for simulating modeling
        and data base manipulation on IBM 370/145 computer at the
        U.C. San Francisco,  via  30 character  per-second  telecommuni-
        cations 	     13

    5   Simulation sequence  between subsystems  showing associated
        datasets for submodel quantification  and validation  .....     21

    6   Facies map of Dogwood Plot	     31

    7   Phenogram illustrating cluster formation  	     35

    8   Territorial map of discriminant  score 1 (horizontal)  vs.
        discriminant score 2 (vertical)  . .'	     36

    9   Trend of seasonal oxidant  dose at a representative San
        Bernardino mountain  station from 1968-1977   	     50

   10   Seasonal pattern of potential transpiration,  ratio of
        actual over potential transpiration and cumulative oxi-
        dant dose at Deer Lick,  1976	     52

   11   Comparison of injury to  the 1975 needle whorl with poten-
        tial transpiration and the ratio of actual over potential
        transpiration at Camp Angelus in 1976	     53

   12   Relationship of daily potential transpiration and daily
        oxidant dose at Camp Angelus during May through September,
        1976	     54
                                    vi

-------
Figures Continued

Number                                                                  Page

   13   Changes of  1975 needle whorl  injury  in  1976  and  1977  at
        Camp Angelus	     56

   14   Changes in  the percent of  the total  needle length of
        selected ponderosa pines with chlorotic mottle in relation
        to two measures of cumulative oxidant dose	     58

   15   Average height growth of ponderosa and Jeffrey pine sap-
        lings at plots experiencing different levels of  chronic
        oxidant injury	     61

   16   Height growth of ponderosa pine  saplings  in  three injury
        categories  at Camp Paivika between 1967 and  1976	     62

   17   Growth trends for trees on the COO plot from 1946 to  1975  in
        the San Bernardino National Forest   	     74

   18   Growth trends for trees on the BP plot from  1946 to 1975
        in the San  Bernardino National Forest	     74

   19   Growth trends for trees on the GVC plot from 1946 to  1975
        in the San  Bernardino National Forest	     75

   20   Growth trends for trees on the DL plot from  1946 to 1975
        in the San  Bernardino National Forest	     75

   21   Growth trends for trees on the NEGV  plot  from 1946 to 1975
        in the San  Bernardino National Forest	     76

   22   Growth trends for trees in the HV plot from  1946 to 1975
        in the San  Bernardino National Forest	     76

   23   Moisture retention curves  for Dogwood plot,  Site 2 	     85

   24   Graphic summary of the population sampling procedures used
        for the western pine beetle showing  datasets and the  type
        of information included for the  San  Bernardino study   ....     88

   25   Cost of sampling for fixed variances as a function of bark
        sample size for total larvae  of  the  Jeffrey  pine beetle, 2
        bark sample units per height,  San Bernardino National Forest,
        1974	    109

   26   Sample variance as a function of bark sample size for fixed
        costs for total larvae of  the Jeffrey pine beetle, 2  bark
        sample units per height, San  Bernardino National Forest, 1974    HO
                                    vii

-------
Figures Continued

Number                                                                  Page
   27   Variance as a function  of  cost  for  different  number  of
        1000 cm^ samples per height  for gallery  length  of  the
        mountain pine beetle in pondrosa pine, San Bernardino
        National Forest, 1974	    Ill

   28   Relationship between oxidant air pollution injury  and
        proximal colonization  of inoculated pine roots  by
        Fomes annosus	    118

   29   Age class distribution  of  ponderosa and/or Jeffrey pine  trees
        on three plots   	  .....    145

   30   Crown class distribution of  ponderosa and/or  Jeffrey pine
        trees on three  plots	    146

   31   Diameter class  distribution  of  ponderosa and/or Jeffrey  pine
        trees on three  plots	    147

   32   Height  class distribution  of ponderosa and/or Jeffrey pine
        trees on three  plots	    148

   33   Total pine needle  litter collected  related to oxidant injury
        score 1975, 1976	    152

   34   Size of pine needles in litter  related to  oxidant  injury
        score—high and low rainfall plots	    154

   35   Nitrogen content of pine needle litter related  to  oxidant
        injury  score—high rainfall  plots 	    156

   36   Nitrogen content of pine needle litter related  to  oxidant
        injury  score—low  rainfall plots 	    157

   37   Potassium content  of pine  needle litter  related to oxidant
        injury  score—high rainfall  plots 	    158

   38   Potassium content  of pine  needle litter  related to oxidant
        injury  score—low  rainfall plots 	    159

   39   Phosphorous content of  pine  needle  litter  related  to oxidant
        injury  score—high rainfall  plots 	    160

   40   Phosphorous content of  pine  needle  litter  related  to oxidant
        injury  score—low  rainfall plots 	    161
                                    viii

-------
Figures Continued

Number                                                                  Page


   41   Calcium content of pine needle  litter  related  to  oxidant
        injury score—high rainfall plots  	    162

   42   Calcium content of pine needle  litter  related  to  oxidant
        injury score—low rainfall plots   	    163

   43   Soluble phosphorous content of  surface soils related  to
        oxidant injury score of pine  trees—high  rainfall plots  .  .  .    165

   44   Soluble phosphorous content of  surface soils related  to
        oxidant injury score of pine  trees—low rainfall  plots   .  .  .    166

   45   Source and destination of 960 decomposition study envelopes  .    169

   46   Source and destination of 160 decomposition study envelopes  .    170

   47   Evenness  (E) in a three species community   	    173

   48   Percent weight loss (30C) incurred by  needle litter on four
        plots, representing the range of air pollution impact on
        ponderosa and Jeffrey pines    	    175

   49   Percent weight loss (30C) incurred by  needle litter follow-
        ing transfer between and within species   	    176
                                     IX

-------
                                  TABLES

Number                                                                  Page

   1   Part 1 of the SBNF data dictionary:  The dataset  index
       CMS Time-sharing commands used for on-line  file management
       of SBNF data base	      9

   2   CMS Time-sharing commands used for on-line  file management
       of SBNF data base	     14

   3   CMS Time-sharing edit-environment commands  used for finding,
       correcting, and displaying the contents of  a dataset or
       program file in the  SBNF data base	     15

   4   Status of distributed versus centralized data processing
       of SBNF data base for various subprojects as of August, 1977  •     17

   5   Distribution of datasets by development stage for the SBNF
       data base at U. C. Berkeley	     20

   6   Number of facies identified on the 18 permanent plots used
       to monitor air pollution injury to forest trees   	     32

   7   Species composition, tree height, and cover of facies on
       Dogwood plot	     33

   8   Major inputs, internal operations and outputs of  the trans-
       piration simulator	     41

   9   Description of information collected to describe  the within
       season development of oxidant injury symptoms on  ponderosa
       pine (PP), Jeffrey pine (JP), white fir (WF), black oak
       (BO)	     43

  10   Subjective categories for description of oxidant  injury
       symptoms on ponderosa (PP), Jeffrey pine (JP), white fir
       (WF) , and black oak  (BO)	     45

  11   Description of data  types and frequency of  data collection
       for each type at major vegetation plots in  1976 and early
       1977	     47

  12   Frequency of different transitional combinations  of five
       classes of spring and summer days	     51

-------
Tables Continued

Number                                                                  Page

  13.  Trends in ponderosa and Jeffrey pine  chronic  injury  scores
       and tree death at eighteen permanent  plots  	     64

  14.  Annual precipitation  (cm) at  Squirrel Inn #2  weather station
       and an analysis of variance among  the ten year  intervals
       that correspond to the ten year growth period	     67

  15.  Rainfall data for the six study plots (mean annual pre-
       cipitation cm)	     68

  16.  Mensurational descriptions of  the  six sample  plots	     70

  17.  Age groups, soil types, mean  annual radial  growth incre-
       ment (mm) and standard deviation for  each growth period,  F
       values from ANOVA, and explained variance (w^)  for each strat—
       fied group of sample  trees on  each plot	     72

  18.  Soil water content at various  matric  suctions for three pro-
       file depths	     80

  19-  Soil water content at various  matric  suctions for profile
       depths of 200 cm or greater	     83

  20.  Available soil water  and clay  content as percent of  the
       whole soil for tree soil depth intervals	     84

  21.  Height, diameter, and length  of infestation for western
       pine beetle sample trees 	     89

  22.  Western pine beetle infested ponderosa pine that were sampled
       between 1973 and 1976 ranked by oxidant damage  class 	     95

  23.  Western pine beetle mean egg  dissection variable by  year
       and generation for whole ponderosa pines, San Bernardino
       National Forest, 1973-1976    	     96

  24.  Variables calculated  from western  pine beetle egg disc
       sample dissection by  generation from  1973 to  1976.   San
       Bernardino National Forest    	  .     97

  25.  Correlation of western pine beetle egg disc dissection
       variables with year,  generation and tree oxidant ratings.
       San Bernardino National Forest, 1973-1976   	     97

  26.  Means by generations  on the last sample date  for western
       pine beetle brood, parasites and predators  from X-ray analysis
       of sample bark discs.  San Bernardino National  Forest, 1973-
       1976	     98

                                      xi

-------
Tables Continued

Number                                                                  Page

  27.  Significance of multiple  regression  coefficients  for  the
       last sample date of X-ray, rearing,  and  sticky  cartons  for
       the western pine beetle and  its  natural  enemies.   San Ber-
       nardino National Forest,  1973-1976  	     99

  28.  Mean western pine beetle  and natural enemy  emergence  by
       generation for the last sample date  of laboratory reared
       discs. San Bernardino National Forest, 1973 to  the first
       generation of 1975	    100

  29.  Mean western pine beetle  and natural enemy  emergence  by
       generation for the last sample date  of sticky cartons (field
       research).  San Bernardino National  Forest,  1973  to the forest
       generation of 1976	    100

  30.  Results of using Tang's calculation  on Jeffrey  pine beetle
       data:  Tree effect	    101

  31.  Results of using Tang's calculations on  Jeffrey pine  beetle
       data:  Height effect	    101

  32.  Results of using various  tests on attack density,  sample
       size of data, to determine differences between  pairs  of
       trees, values in attacks/cm^ 	    102

  33.  Preliminary summary of final smog damage ratings  for  pines
       killed insects on established vegetation plots, 1973-1975   .  .    102

  34.  Infection of inoculated pine roots with  Fomes annosus in
       relation to the severity  of  air  pollution injury  	    116

  35.  Colonization of inoculated pine  roots by Fomes  annosus  in
       relation to the severity  of  air  pollution injury  	    117

  36.  Number of pine stumps inoculated with Fomes annosus by
       site and species	    120

  37.  Relationship between  the  surface colonization of  inoculated
       pine stumps by Fomes  annosus and the severity of  air  pollution
       injury	    121

  38.  Downward colonization and colonization rate of  Fomes  annosus
       in inoculated pine stump  related to  air  pollution injury  ...    122

  39.  Volume of inoculated  stumps  colonized by Fomes  annosus  in
       relation to air pollution injury 	    123
                                    xii

-------
Tables Continued

Number                                                                  Page

  40.  Influence of ozone on  the  linear  growth  rate  and  conidial
       production of Fomes annosus   	    125

  41.  Influence of ozone on  conidial  germination of Fomes  annosus
       (Isolate:  JL1)	    126

  42.  Influence of ozone on  conidial  germination of Fomes  annosus
       (Isolate:  PP1)	    127

  43.  Influence of ozone exposure on  conidial  germination  and
       germ tube extension of Fomes  annosus	    128

  44.  Effects of ozone  on colonization  of pine discs by Fomes
       annosus	    129

  45.  Influence of ozone on  Fomes annosus conidial  germination
       through successive generations  of  exposure 	    131

  46.  Tree mortality by cause and forest type  in the San Bernardino
       mountains in 1976	    134

  47.  Tree mortality by pest complexes  and  species  in the  San
       Bernardino mountains in 1976	    134

  48.  Percent of mortality centers  in the mixed conifer and yellow
       pine forests in relation to oxidant injury 	    135

  49-  Pathogenicity test summary	    137

  50.  1975-1976 SBNF seedling establishment  field study, germinated
       seeds, means per  screened  mini-plot in May, 1976	•    139

  51.  1975-1976 SBNF seedling establishment  field study, sur-
       viving seedlings, means per screened  mini-plot as of July
       1976	    139

  52.  Status in acquiring plot data on  the  19  study plots  located
       in the San Bernardino mountains of southern California ....    144

  53.  Needle size and content of N, P, K and Ca in  needle  fall  .  .  .    155

  54.  Estimated total radiation  (R) and  temperature (T) accumu-
       lated during one  clear day at measurement points  beneath
       the integrated field decomposition study trees 	    177
                                    xiii

-------
                     LIST OF ABBREVIATIONS AND SYMBOLS

ABBREVIATIONS

SCAB        — South Coast Air Basin
SBNF        — San Bernardino National Forest
CP          — Camp Paivika vegetaton plot
BP          — Breezy Point vegetation plot
TUN 2       — Tunnel 2 vegetation plot
DWA         — Dogwood A vegetation plot
DWB         — Dogwood B vegetation plot
DL          — Deer Lick
SF          — Sky Forest vegetaton plot
UCC         — University Conference Center vegetation plot
COO         — Camp 0-Ongo vegetation plot
GVC         — Green Valley Creek vegetation plot
NEGV        — Northeast Green Valley vegetation plot
SV          — Snow Valley vegetation plot
BL          — Bluff Lake vegetation plot
SC          — Sand Canyon vegetation plot
HV          — Holcomb Valley vegetation plot
CA          — Camp Angeles vegetation plot
SCR         — Schneider Creek vegetation plot
BF          — Barton Flats vegetation plot
CAO         — Camp Osceola vegetation plot
HB          — Heart Bar vegetation plot
PP          — ponderosa pine
JP          — Jeffrey pine
WF          — white fir
dbh         — diameter at breast height
ppm         — parts per million

SYMBOLS
„£          — microgram
Oo          — ozone
                                    xiv

-------
                              ACKNOWLEDGMENT S

     Administrative services for the project were coordinated by 0. Clifton
Taylor, Associate Director, Statewide Air Pollution Research Center, Univer-
sity of California, Riverside, CA  92521.

     Portions of this report were contributed by the following authors
and assistants:
Pages

iii, iv
  1
Author

0. Clifton Taylor
Contribution

Foreword, Abstract,
Introduction
                Ronald N. Kickert
                Assisted by
                Robert Thomson
                                Ecosystem Simulation
                                Modeling
 28
 38
 65
 77
 86
115
132
Joe McBride
Paul Miller
Cliff Ohmart
Reviewed by Rick Laven
and project to be con-
tinued by Rick Laven

Rodney J. Arkley,
P- L. Gersper and
R. Glauser

Donald L. Dahlsten
Fields W. Cobb and
Robert L. James

Joe R. McBride,
Donald L. Dahlsten,
and Fields W. Cobb
Tree Population
Dynamics Subsystem

Oxidant Dose — Canopy
Response Subsystem

Oxidant - Tree Growth
Physical and Chemical
Properties of Soils and
Moisture Dynamics

Stand Tree Mortality
Subsystem - Bark Beetle
Population Dynamics

Epidemiology of Forest
Tree Population

Cause and Extent of
Tree Mortality
136
Fields W. Cobb
Forest Tree Seedling
Establishment
                                    xv

-------
Acknowledgments continued

Pages           Author

                Robert F. Luck



                Rodney J. Arkley

                John R. Parmeter
                and J. N. Bruhn
Contribution

Cone and Seed Production
for Dominant Conifer Tree
Production

Litter Production

Foliage Litter Decom-
position
      Particular  thanks  is  extended  to  the District  Rangers  and  Staff  of
the Arrowhead, Big  Bear, and  San  Gorgonio Districts of  the  San  Bernardino
National  Forest  for their  interest  and cooperation  in regard  to field study
activities.

      Special  thanks is  also extended to the many  research assistants,
staff research associates  and others who have made  this  study possible.  A
few of  these  participants  include:

      Maureen  A.  Thomas
      Kathrynn E.  Banbury
      Nancy L. Bruhn
      David L. Rowney
      Isabel F. Alvarez
      G. Nick  McKibben
      William  A.  Copper
      Diana Doyal
      Edith Reisner
      Robert Van  Doren

      A special expression  of  thanks is extended to  Donna M. Shaw for  typing
of the manuscript,  to Paul Miller for  proofing and  invaluable assistance
in compiling  the report and to Maureen Thomas for drafting  and  preparation
of many of the figures.
                                     xvi

-------
                              INTRODUCTION

     This is a progress report for the final year of a study partially
supported by EPA Contract 68-03-2442.  This report and a previous  report
designated as EPA-600/3-77-104 constitute the final report for the three
year contract which has supported a multidisciplinary research effort
initiated in 1972.  The study is expected to terminate May 31, 1980.

     This continuing-long term study is an effort to identify and  quantify
chronic effects of oxidant air pollutants on individual units of a forest
ecosystem and to model the interactions between units which are initiated,
stimulated or driven by the pollutant impact.  The units of the ecosystem
selected for study were those considered to be most susceptible to pollutant
impact those units which are known to play a very important role in struc-
turing the total ecosystem character.  We anticipate that models developed
for this study will be useful:  to design forest and recreational  area
management programs; to predict long term changes in ecosystem structure
when air pollutants are present; and perhaps to aid in establishing reason-
able pollutant standards to protect against serious undesirable changes in
the ecosystem.

     The forest ecosystem is subjected to numerous stresses which  favor
the development of some organisms and suppress development of others.
Air pollutants represent an additional man made stress in the complex,
therefore, any study of long-term chronic air pollutant effect must be
accompanied by evaluations of impact from other stress factors.  Since the
intensity of stress produced by any of these factors varies widely over time
and since plant response is affected by interaction of the stress  factors
evaluation of the air pollutant involvement in ecosystem changes become a
very complex study.

     The San Bernardino National Forest (SBNF) has been exposed to an
increasing annual dosage of photochemical oxidants during the preceding 3 or
4 decades as industrial and urban development in the South Coast Air Basin
(SCAB) expanded at a phenomenal rate.  Abnormalities, later identified as
oxidant air pollutant injury, were causing concern among residents  and U.S.
forestry officials in the early 1950's.  The injury on ponderosa pine was
initially thought to be associated with hydrogen fluoride and perhaps other
air pollutants released by specific industries which were relatively new to
the SCAB.  Research during the 1950's largely dispelled this theory and
implicated the oxidant air pollutants which are the responsibility of a
broad sector of man's activities.

     The SBNF, located at the east end of the South Coast Air Basin, is
subjected to an ebb and flow of polluted air from the SCAB as the  alternate
diurnal "pumping" of high desert and marine air through the SCAB occurs.

-------
Typically, during the recognized smog season, air flow during the morning
is from the east toward the  coastal region; and as the day warms, the
flow is essentially reversed to deliver marine air laden with oxidant
pollutants to the SBNF.

     Evaluation of direct  effects  of oxidant pollutants in the SBNF has
for the most part been confined to the dominant tree species although  it
is known  that a wide variety of species of green plants are adversely
affected.  These primary producers are critical elements in an ecosystem to
provide food and shelter for all other organisms in the system.  Chronic
injury to the green vegetation may over time significantly change the  source
of energy, protection and  general  habitat of numerous consumer organisms.
An understanding of changes  in plant communities suffering from  air pollu-
tant  injury  is  essential if  one is to predict the fate of an ecosystem
impacted  by  a growing and  changing industrial and urban complex*

      This study, including the modeling effort, is based on the  assumption
that  the  effect  of  air pollutants  or any other  stress element will be
ultimately transferred to  numerous other units  of the ecosystem.  It would
then  be expected that  a  gradual or insidious but significant change  in the
ecosystem might be  expected.

-------
ECOSYSTEM SIMULATION MODELING OF MIXED CONIFER FOREST UNDER PHOTOCHEMICAL
AIR POLLUTION

Introduction

     Can experiments be conducted on forest  land, using different kinds  of
long-term air pollution trends to determine  ecosystem response, without
actual manipulation of the forest, and without decades of waiting for  the
results?  This question leads to the following objectives of  the SBNF
program: 1) to design forest ecology systems models  for forecasting  ecolog-
ical effects of photochemical air pollutants in  southern California  mixed
conifer forest ecosystems; 2) to evaluate  the adaptability of  systems  models
to other pollutant types and other forest  types; and 3) to evaluate  the
forecasted consequences of photochemical air pollutants in forest ecosystems
in terms of human welfare effects.  A discussion of  these objectives,
together with a review of other scientists'  published thoughts related to
them, can be found in a previous progress  report (Kickert, 1977).

Methods

     General systems philosophy and systems  analysis techniques involving
digital computer simulation modeling methods were applied in working toward
the objectives of the project (Figure 1).

System Model Development--

     The first step in the development process was to clearly  define the
various problem solving goals for which the  simulation models  would  be used.
This was done by extensive discussion with all project scientists and
consideration of the results reported by Kickert (1977).  The  discussions
were directed at drawing out of each investigator a  qualitative description
of the relations between changeable properties of the forest as a system, as
he conceived them to exist in the subsystem(s) pertinent to his major
role(s) in the project.  The relationships were  then diagrammed and  inte-
grated into a graphic flow model (Figure 28  in Kickert, 1977).  The  next
step is to convert the set of graphic models into sets of transfer functions
to describe annual rate of change expressed  as finite difference equations.
Conditional logic for threshold and time-delay conditions typical of biolog-
ical phenomena will be included.  The mathematics and associated qualitative
conditional logic is written in a high-level computer language.  We  will
then draw upon the project data base, assembled  from the data  collected  by
the investigators, and use quantitative relations discovered by them,  to
refine the mathematical form of transfer functions in the various submodels,
to set the values of species and site parameters, and to serve as external
data for driving certain functions.  Methods of  handling the  data base are
discussed under achievements below because these have been reorganized and

-------
                    Model Planning



CO
1
o
u_
Q.
t-
o








1 	
(GOAL DEFINITIONS J

1

TABLES
U
A
*
Model
n
Model
Development
A
r / \
^ FLOW PARADIGM _/ ™$™\
TIME-SPACE \^ Mnnns /"
RESOLUTION \ nuutLs /
V
J
1 	
1
! Ql
i
i
1 cot
1
1 , 	
* 1 	
Application |
~{ 1
1 i
— . , 	
SYSTEM MODEL ' 1

t !_
DERIVATION OF
PROJECTIONS ON
SUBSYSTEMS 6
WHOLE SYSTEM |
V

GpUBLICATION-\
lEMONSTRATION J
1 	
MODIFY
/
r
\
\ ORIGINAL
DEVELOPMENT
I
i 	 	 1
FLOW CHART
COMPONENTS
& CONTROLS
J
i '
JANTIFICATION
6 •<
;ONDITIONAL
LOGIC
! i
1PUTER CODING
DEBUGGING
*
VALIDATION
1
SENSITIVITY
ANALYSIS


Figure 1.  System simulation modeling  process with bold arrows showing
           information  flows which were  behind  schedule and consequently
           were retarding modeling progress.

-------
re-defined during the period which  this  report  covers.

Model Reliability Evaluation and Use—

     After conversion to mathematical computer  programs  and  the  essential
debugging, the  set of simulation programs  are run  and their  behavior  is
compared to observations of real forests change over a number  of years.
When this evaluation of reliability is acceptable, we then perform experi-
ments on the  simulated forest  to determine the  probability of  long-term
responses of  various forest properties under alternative future  trends in
air quality,  forest meteorology, and harvesting practices externally  imposed
on the  simulated ecosystem.

Transfer of Ecological Discoveries  to Social Scientists—

     Results  of these experiments are discussed by project investigators
and subsequently communicated  to scientists outside of the project who
are in  a position to interpret the  economic, social, and political signifi-
cance as indicated in Table  10 in Kickert  (1977).  A cost-benefit evaluation
of alternative  air pollution control strategies in the South Coast Air
Basin might possibly be made in agreement  with  Westman's (1977)  warnings
by the  combined use of this forest  simulation package with parts of the
MATHAIR (MATHTECH, 1976) computer model  and an  appropriate meso-scale
meteorological  transport model. The MATHAIR model  assumption for the  local
source-local  impact is not valid for the Los Angeles Basin - SBNF geography.

     Documentation is produced for  each  submodel consisting  of a
word model description of relationships, graphic flow chart, mathematics,
computer program source code,  results of reliability evaluation,  and  results
of experiments  performed using the  forest  simulation package.

Dependence of Simulation Modeling on Data  Base  Management—

     Conversion of conceptual  models of  various subsystems into  mathe-
matical algorithms and subsequently into computer  programs has been designed
to be dependent as much as possible on the many datasets being acquired
in the  various  subprojects of  the SBNF program. The data base is designed
to serve as documentary evidence for the reality and validity  of (1)  quan-
titative relations used to construct the simulation models;  (2)  the behavior
of the  set of models when run  on the computer;  and (3) the kinds of eco-
systemsuccession forecasting experiments conducted by using  the  set of
simulators.   The bold arrows in Figure 1 outline the broad framework  of
dependence between the data base and the information requirements for
simulator development.  Until  recently,  the lack of a documented and  opera-
tional  procedure for managing  the SBNF data base delayed progress in  the
development of  our forest modeling  activity.

Discussion of Developments and Achievements During 1976-1977

     Progress is described in  terms of the sequence of steps being followed
in the  simulation modeling process  presented above and management of  the
computerized  SBNF data base which supports the  modeling  process. These

-------
accomplishments all relate to the first general objective of validating
quantitative relations used to construct  the simulation models. It  is
necessary to completely meet this first objective before the second and
third can be accomplished.  For this reason, we do not yet present  results
of actual computer experiments using the  forest system simulation models  for
various possible future oxidant air pollution  trends.

Organization of Computerized Data Base Management Procedures for Air
Pollution Ecological Effects—

     Some recent historical perspective is helpful as an aid in presenting
the achievements in managing the SBNF data base.

     From January 1974, through December  1976, the Corvallis Environmental
Research Laboratory and the Lawrence Livermore Laboratory (LLL) had an
interagency agreement for the LLL to design and develop a data management
system using data collected in the SBNF program.

     In late 1975 and early 1976, it became apparent that such a task
would not be completed by December 1976,  and it was anticipated that the
agreement would be renewed for at least an additional year.  In August and
September 1976, it was discovered that this extension would not take place.
In January 1977, the Ecosystem Simulation Modeling subproject acquired
the added responsibility of retrieving all of  the data files which  project
investigators had submitted to the LLL, and of organizing a computerized
data management system useable on the IBM 370/145 at U.C. San Francisco,
The ecological modeling computer work was also being done on this computer.
Under the prior arrangement, data management was done on the computer at
Livermore, but the change to UCSF led to  a significant improvement  in model
design planning and data management.  Both activities could now be  done on
the same computer system at UCSF.  The transfer of responsibility in January
1977, was an abrupt one due to circumstances beyond our control.  Clear,
comprehensive documentation of the status and contents of datasets  associ-
ated with the various subprojects was unavailable.  These conditions,
together with the fact that development of a management procedure for
the data base inherited without a budget  between January and June 1977, was
behind schedule with respect to modeling  needs, required us to relax empha-
sis on model development in order to organize the data base problem.

     Dataset verification and auditing—  Although written for a corporate
business environment, we applied the philosophy of Wilkinson (1977)  in
beginning a data processing audit, both through-the-computer, and around-
the-computer, for the correction of data  sets which we had acquired.
We inherited virtually no documentation on the nature of any verification
which might have previously been done on  individual datasets or between
datasets having common cross-referencing  data  elements, such as tree
tag numbers, or species codes.

     One objective of the audit was to discover whether datasets which  the
field investigators assumed we had acquired were missing.  Another  objective
was to uncover discrepancies in data elements between different years for a
given dataset, and between datasets where the  same data elements were used

-------
in each.  The intention was to assist  field  investigators  in revealing any
errors which reside in the datasets.

     The SBNF data base structure:  the  data dictionary—  The general
organizational structure of the data base  is shoxm  in  Figure 2.   A signifi-
cant achievement has been the establishment  of  an on-line  data dictionary.
Schussel (1977) describes this as "a repository of  information about the
definition, structure, and usage of data.  It does  not contain the actual
data itself" (sic).

     We have structured this into a data set index  (Table  1),  dataset
definitions (Appendix 1), and dataset  descriptors (example in Figure 3).
A computer terminal user can simply log-on,  and proceed  through  these
levels of increasing detail of information in search of  specific kinds of
datasets on ecological effects under air pollution  in  the  forst.  This can
be done by using the procedure described under  the  discussion of Central-
ized Data Base Approach which appears  later  in  this report.

     The dataset progress status chart— There  is a sequence of  distinct
stages  through which datasets advance, from  the time that  the decision
is made by a field  investigator to collect a certain kind  of data, until
that time when a written report is produced  which contains the description
and results of analysis of the dataset.  At  any given  time during this
research program, various datasets progress  at  various rates through all of
these stages. Merely tracking down the descriptive  information on a dataset
by using the data dictionary does not  inform one as to whether or not  the
dataset has reached a stage where the  data are  presently analyzeable on the
computer. In order  to assess the status  of any  of the  datasets in the  SBNF
data base  (Figure 2) at a given instant, an  on-line Dataset  Progress Status
file has been established and is updated on  a weekly basis.   A listing of
this file, as of August 31,  1977, is presented  in Appendix 2. This feature
of the  SBNF data base enables us to  track  the status of  a  dataset through
the various stages  of preparation, from  left to right  in Appendix 2, so that
analysis can then be performed, using  that dataset. It  also allows us to
see where we are in terms of stages  of a dataset's  analysis  for  systems
modeling, for data-sharing among investigators  in the  project, and subse-
quently, for external requests for data.  Datsets whose  entries  extend
to the  right of  the vertical bold line in  Appendix  2 are ready for, or
are presently under analysis.  Those that  do not are still in a  stage  of
data preparation.   The  information categories "NEW  DATA" and "VERIFICATION"
pertain to datasets which have not yet been  verified by  the original
investigator.  "FORMAT  APPROVED" only  pertains  to new  types of data
collection efforts  as they may arise.  This  stage is intended to call
to the  attention of the investigator that  some  aspect  of his data form
format  will induce  subsequent delays in  data processing^  If its being
altered presents no problem  for the  logistics of field data collection,
he is advised as to what change to make  for  his benefit  later in the data
processing stage.

     The category "DESCRIPTOR ENTERED?"  shows a record of  whether that
document has been placed in  the data dictionary.  Appendix 2 shows that a
number  of datasets  have not been covered in  this way since we took on  this

-------
                                                                            .DATASET INDEX
                                        DATA  DICTIONARY
                                DATASET  DEFINITIONS
00
               SBNF
           DATA  BASE
DATASET PROGRESS
      STATUS
                                        DATASETS
                                                                            •DATASET DESCRIPTORS
                                                                             DATASET DESCRIPTORS
                                                                             DATA
         Figure 2.  Information structure for the computerized SBNF data base  accessible from remote
                   dial-up terminal  via UCSF/CMS time-sharing mode.

-------
TABLE 1.  PART I OF THE SBNF DATA DICTIONARY:  THE DATASET INDEX.
age class
air pollutant
air temperature
basal area, tree

cations, soil
cones
cover
crown data, tree

density, tree
diameter, breast height
disease, tree

elevation
fire effects
foliage

geographic  coordinates
height, tree
height growth, tree
index
insect risk
insects

litter, needle

location, tree
moisture, soil
mortality,  tree

needle leaf, condition
needle leaf, length
needle leaf, retention
net radiation
nutrients
organic matter content,
soil
oxidant, ambient con-
centration
ozone, ambient con-
centrtion
pH, soil
plot, vegetation-,
logistics
plot, super-

plot, sapling-
STAGE, FIRESTAG, PLOTREGN, CTREE, STNDSITE
OXIDINDX, OXIDANT, PLOTOXID
FSMTINDX, FSMET, HMET, PLOTMET
TREEVEG, SPRMORT1, SPRMORT2, SPRMORX,
STNDSITE
SXSCAT
CONE, GCONE
SHRUBVEG, PLOTREGN, PLITR, STNDSITE
CTREE, SPRMORT1, SPRMORT2, SPRMORX
TREEPEST
TREEVEG, FIRETREE, SPRMORT1, SPRMORT2
TREE, SPRMORT1, SPRMORT2, TREEPEST
DISU, FASP, SPRMORT1, SPRMORT2, STNDSITE,
TREEPEST
PLOTINDX, STNDSITE
FIRETREE, FIRESTAG, FIRESRUB
STOMRES, OZFLUX, TREE, CTREE, SAPTREE,
SAPSURF, FLDECOMP
PLOTINDX
CTREE, SPRMORT1, SPRMORT2, STNDSITE, TREEPEST
SAPGRO
PLOTINDX, FSMTINDX, OXIDINDX
TREE, SAPTREE
ISURV, BTREE, EGG, REAR, STIK, XRAY, SPRMORT1,
SPRMORT2, SPRMORX, STNDSITE, TREEPEST
PLTR, PNFALL, TREELIT, LITMAS, FLDECOMP,
LITRKEM, SLSS
TRID
MOIST, MATRIC, LSOIL
TREEMORT, TREE, SPRMORT1, SPRMORT2, SPRMORX,
STNDSITE, TREEPEST
TREE, SAPTREE
TREE, SAPTREE
TREE, SAPTREE
FSMET
SXSCAT, SFCSOLKM, LITRKEM, DRIP

STEXOM

OXIDINDX, OXIDANT, PLOTOXID

OXIDINDX, OXIDANT, PLOTOXID, OZFLUX
STEXOM

PLOTINDX
SPRMORT1, SPRMORT2, SPRMORX, STNDSITE,
TREEPEST
SAPTREE, SAPGRO, SAPSURF

-------
TABLE 1.  (CONTINUED)
precipitation;
publications
radial growth, tree stem

regeneration, tree
relative humidity, air
seedling
seeds; see regeneration
shrub
site
slope
smog injury  score
soil

species composition
species, tree
succession
texture, soil
tree tag number
wind direction
wind speed
HPREC, PLOTPREC, DRIP
EPA/SBNF contracts & grants;  SBNFPUBS
TREEGRO, TREEGR02, BOGRO, SPRMORX,
TREEPEST
PLOTREGN, STAGE, CONE, GCONE, PLOTSEED,
SAPS, LSOIL, SSAS, SLSS,  SPRXRGNS,
SPRXRGNC
FSMTINDX, FSMET, HMET, PLOTMET
PLOTREGN, SAPS, STAGE

SHRUB, SHRUBVEG, FIRESRUB, STNDSITE
PLOTINDX
TREESOIL, STNDSITE
TREE, SAPTREE, SPRMORT1,  SPRMORT2, SPRMORX
SXSCAT, STEXOM, MOIST, MATRIC, TREESOIL,
SFCSOLKM, LSOIL, STNDSITE
TREEVEG
TRID
STAGE, PLOTREGN, FIRESTAG, STNDSITE
STEXOM, STNDSITE
PLOTINDX, TRID, TREE, CTREE,  CONE, TREEGR02,
BOGRO, DISU, ISURV, TREEMORT, TREESOIL,
PNFALL, TREELIT, LITMAS,  FLDECOMP, DRIP,
LITRKEM, SFCSOLKM, SAPTREE, SAPGRO, SAPSURF
FSMTINDX, FSMET, HMET
FSMTINDX, FSMET
                                     10

-------
   DATASET NAME:  SAPGRO   (PINE SAPLING ANNUAL HEIGHT GROWTH)
   INVESTIGATOR; PAUL  R.  MILLER,  U.C. RIVERSIDE,  (7 14 ) - 787-366 1
   STATUS OF EXTERNAL  AVAILABILITY: CLOSED
   DATASET DESCRIPTOR  AUTHOR/DATE: R.N. KICKERT,  FORESTRY, U.C. BERKELEY;  8/5/77
   DATA SITES: BL (JEFFREY PINE), BP (PONDEROSA  PINE),  CA (PONDEROSA  PINE),
               CAO,  CP,  DWA,  HB,  HV,  SF, TUN2;
   DATA RECORD SEQUENCE:  ONE  RECORD PER TREE; RECORDS GROUPED BY PLOT;

   DATA ELEMENT SEQUENCE:  [X]  POSITIONAL,   [~~] FREE-FIELD,  OR  [""] KEY-IDENTIFIER?
                           Li                [  ]                   [J
   RECORD FORMAT DESCRIPTION:  (VARIABLE NAME7 COLUMN NUMBERS FOR VARIABLES'  FIELDS,
                               PHYSICAL UNITS/ IF  ANY/,  ESTIMATED OBSERVATIONAL
                               ERROR TOLERANCE)
    1. PLOT INDENTIFIER,  COL  1, A(4);
    2. TREE TAG NUMBER,  COL 5,  F(4);
    3. ANNUAL INTERNODE  LENGTH GROWTH, COL  9, 18(IX,F3,0),  MILLIMETERS,  +/- 30 MM;

   DATA COLLECTION  DATES:    1976;
   NOTES, QUALIFICATIONS,  LIMITATIONS (BY VARIABLE NAME):
    2. INFORMATION  ON  SPECIES,  EXACT LOCATIONS,   1975 TOTAL TREE HEIGHTS,  STEM
       DIAMETERS, AND  ANNUAL  CROWN CONDITION, FOR EACH TAGGED SAPLING CAN  BE
       FOUND IN THE DATASET NAMED  'SAPTREE';
    3. FROM LEFT TO RIGHT, ALONG THE RECORD,  A MINIMUM OF 10 INTERNODE  LENGTH
         VALUES CORRESPOND TO THE YEARS
        1976, 19-75,  1974,...  1967.  ADDITIONAL VALUES,  UP TO A TOTAL  OF 18 VALUES,
        MAY BE FOUND IN  A SINGLE LOGICAL RECORD,  WHERE THE 18TH VALUE CORRESPONDS
        TO THE YEAR 1959.  THE VARIABLE RECORD LENGTHS PRIOR TO 1967 ARE A  RESULT
        OF THE INVESTIGATORS  GOING BACK ONLY  AS  FAR AS THEY FELT THEY COULD
        ACCURATELY  DETERMINE  ANNUAL INTERNODE  INCREMENTS.
Figure 3.  Example of Part III of the SBNF Data Dictionary:  Dataset Descriptors.

-------
activity.  If the category "KEYPUNCHED?" contains an "N" for a dataset, then
we have not been given a card deck for that dataset by the respective
investigator.  The category "REDUCTION PLAN DEFINED?" refers to whether a
specific detailed plan for quantitative analysis of the dataset has been
defined for the purpose of identifying certain transfer functions in one of
the subsystem models, or for the purpose of evaluating the reliability of a
part of the modeling package.

     The datasets— The third portion of the SBNF Data Base structure
(Figure 2) is the collection of datasets.  Each one is a file stored on
magnetic disk, with a back-up copy on magnetic tape, and has a copy of
its descriptor located at the beginning of the file-

     Centralized data base approach— The orientation used in maintaining
the SBNF data base has been a centralized approach in that all datasets are
kept collectively on only one computer system, the IBM 370/145 at the
University of California, San Francisco (Medical Center).  Within the
Ecological Modeling and Data Management activity, the data manipulation
environments are diagrammed in Figure 4.  Datasets are read in from card
decks to magnetic disk under the OS  (Operating System) environment and
transferred to magnetic tape as backup.  The datasets are transferred from
OS disk to mini-disk in the Conversational Monitor System (CMS) environment
in preparation for work.  CMS is a general purpose time-sharing system
operating under VM/370.  In addition to the datasets which are under
immediate use, other units of the SBNF data base such as the Data Dictionary
and Progress Status are maintained on CMS mini-disk for immediate tele-
phone access.  These relationships are diagrammed in Figure 4.  In keeping
with the dependence discussed earlier (Figure 1) of ecological systems
modeling on analysis of datasets in  the SBNF Data Base, model development is
conducted in the same CMS environment, including storage of the computer
programs designed to simulate various subsystems.  To dial up and interact
with the datasets in CMS, we use a DataMedia model 1520A video screen, key
board terminal, a Diablo 1620 HyTerm printer terminal, and an Execuport
320 portable thermal printer terminal.

     Readily available CMS commands  can be used to manipulate dataset files
and the contents within files.  Rapid manipulation of data between datasets
can easily be done byusing the CMS commands shown in Table 2-  Just about
any kind of searches desired can be made on data within a given dataset with
commands as shown in Table 3 which are immediately available at the terminal
in the TECO and CANDE interactive time-sharing environments on the DEC
PDP-10 and Burroughs computers respectively.  This means that any eventual
use of the SBNF data base on other main frame computers elsewhere, perhaps
by other environmental scientists, should be just as useable as our capabil-
ity on the UCSF IBM computer.

     Aside from using the CMS commands for retrieved and displayed various
kinds of data, summarization and analysis of data are done by entering
either the SPEAKEASY mode or SPSS mode in the on-line environment, or by
submitting a batch job, via remote job entry, in OS to use the BMDP statis-
tical programs (Figure 4)•  SPEAKEASY is a simple interactive data manipula-
tion language containing an immense  number of built—in functions for

                                     12

-------
             CMS INTERACTIVE
                                            OS BATCH
                    St or age  for
                    immed i at e
                                                      LINE
                                                      PRINTER
                                                      Evans  HaJJ.
    St or age  for
    frequent
    use
                    Stor age  for
                    infrequent use
                    & back
       Corval1i s
       Environment al
       Research
       Labor at ory
••^•^^MfeUC-A^^^VBV       -^^*^H^^^^M«lmHI^M«HV^^^B
Ecological Modeling Subproject
    U.C. Berkeley
                     Statewide  Air Pollution
                      Research  Center
                     U.C. Riverside
Figure  4.   Data processing environments used  for simulation
            modeling  and data base manipulation on IBM  370/145
            computer  at University of California, San Francisco,
            via 30  character-per-second telecommunications.
                                13

-------
TABLE 2.  CMS TIME-SHARING COMMANDS USED FOR ON-LINE FILE MANAGEMENT
          OF SBNF DATA BASE.
  Command

EDIT



READCARD

COPYFILE



RENAME

TYPE


PRINT


COMPARE


LISTFILE


ERASE
                    Function
construct a new file by inputting through the
terminal, or change, or examine data within, an
existing dataset file (see TABLE 3);

construct a new file by reading a card deck;

combine several files into one file; rearrange
the contents of records in a file; add one file
to the end of another;

change the name of a file;

type the contents of a file on the printer at
Evans Hall on the UCB campus;

type the contents of a file on the printer at
Evans Hall on the UCB campus;

compare all or part of the records in two files
and type the records that are not identical;

list information about the files which are stored
on disk;

delete the specified file from disk.
                                    14

-------
TABLE 3.  CMS TIME-SHARING EDIT-ENVIRONMENT  COMMANDS  USED  FOR FINDING,
          CORRECTING, AND DISPLAYING THE  CONTENTS  OF  A DATASET OR PROGRAM
          FILE IN THE SBNF DATA BASE.
  Command
INPUT
                    Function
creates new lines typed into the file by the user
at the terminal
LOCATE
TOP
locates the next line in the file that contains a
specified character string (and types the line at
the terminal)+

moves the line pointer back a specified number of
lines in the file
UP
moves the line pointer back a specified numbr of
lines in the file
BOTTOM
moves the line pointer to the position following
the last line in the file
DOWN
moves the line pointer forward a specified number
of lines in the file
NEXT
CHANGE
REPLACE
DELETE
moves the line pointer forward one line (and types
the line at the terminal)+

changes a specified character string in the line
to a new character string; can be used to search
for a specified character string anywhere in the
file and then type out the line in which the
string is found

changs the current line content according to the
terminal user's request

beginning at current line, erases the specified
number of lines from the file
TYPE
beginning at current line, types out the contents
of the specified number of following lines in the
file
FILE
terminates the current editing session for the
file and stores the file on disk
+ assuming VERIFY command is ON
                                    15

-------
performing mathematical and statistical analyses on data arrays and vectors,
for manipulating arrays in various ways, and for graphing data.  SPSS is  the
Statistical Package for the Social Sciences which is documented in Nie et
al. (1975). BMDP is the set of statistical analysis programs documented in
Dixon (1975).

     While our base of operations is at U.C. Berkeley, we have easily
accessed, with no difficulties, the various units of the SBNF Data Base
just described, and some of the simulation programs, while at the Statewide
Air Pollution Research Center, U.C. Riverside, and the Corvallis Environmen-
tal Research Laboratory, Corvallis, Oregon.  The potential exists for any
investigator in the SBNF program to directly interact with the UCSF computer
with those datasets that have been verified.  The extent to which this
happens from now on depends upon the desires and motivations of the investi-
gators.

     Distributed processing approach— While a centralized approach to
maintaining the SBNF data base has been employed, the approach to processing
of datasets has evolved in a distributed manner.  Up to the present, several
investigators in the project have only operated on their own datasets
within  their own data processing environments.  This is evident in our
assessment, as shown in Table 4, of the amount of data processing which has
been done by investigators on computer systems other than the one used to
maintain the centralized SBNF data base since January 1977, and the distri-
bution  of data-related requested from the investigators to this subproject.
Project investigators haven't reached the stage of conducting integrated
data analysis of their own datasets with those assembled by their colleagues.
This is probably because not enough years of data had accumulated prior to
this time and also because the entire data base was not in a readily access-
ible computer environment.  In addition, it is natural that we will be doing
much of this for transfer function identification in system model develop-
ment. As the project approaches a stage of synthesis in the next few years,
a decision may be adviseable from the project investigators as to whether it
is in their best interest, from the viewpoint of trans-disciplinary data
analysis, for them to continue solely with a distributed data processing
approach.

     Specific data processing tasks achieved— In 1*973, the Soils subproject
placed  soil moisture sensors at various depths on 22 sites in the 18 vegeta-
tion plots.  These have been interrogated at weekly, or biweekly, intervals
since that time by personnel out of U.C. Riverside.  The Soils subproject
also took field soil samples to the laboratory to develope calibration data
so that the field data on electrical current passing through the moisture
sensors could be converted to log resistance values and then to percent soil
water values.  Data processing to accomplish these  steps was expected  to  be
finished under the previous EPA/LLL agreement during January 1974, through
December 1976, as discussed earlier in this report.  The fact that this did
not materialize precluded the avilability of percent soil water data and
essentially halted progress in further development  of our forest stand
moisture simulation model.  In the first half of 1977, we tackled this
deliquent data processing task and by August 1977,  the establishment of the
MOIST (reduced) dataset was 95 percent accomplished.

                                    16

-------
TABLE 4.  STATUS OF DISTRIBUTED VERSUS CENTRALIZED DATA PROCESSING OF
          SBNF DATA BASE FOR VARIOUS SUBPROJECTS AS OF AUGUST 1977.
Submodel
WATER
CANOPY
TREEGROW
ROOTS
BEELTE
mortality
LITTER
LITDECAY
CONE
SEEDLING

STNDCOMP
Relevant
investigator
Arkley
Miller
Laven
Cobb
Dahlsten
superplot
Arkley /Miller
Bruhn
Luck
Cobb
McBride
McBride
Independent
data processing
little
much
little
some
much
little(?)
some
some
some
much
much
much
Dependence on
data management
subproject
much
little
much
some
some
much ( ? )
much
much
some
little
little
little
                                    17

-------
     Another task achieved was the conversion of raw field data on plot
tree ring widths (TREEGRO dataset) obtained as far back as 1920, to  a  form
which is now useable for analysis in developing the tree STEM growth sub-
system simulator.

     Accomplishments in auditing datasets—In order to prepare the Data
Dictionary and Data Progress Status Chart, our audit of the datasets led to
results which in some cases revealed further data needs for modeling,  and  in
other cases led to improvements in the consistency of information within and
between datasets.

     By auditing the data elements in SPRMORT1, SPRMORT2, SPRMORX, STNDSITE,
and TREEPEST, we discovered that data were not being obtained that would
enable calculation of the stand tree mortality as a percentage of the
total stand stocking density, by species.  We would have the estimates of
numbers of recently killed trees, by apparent causal agents and by species,
but we would not be able to relate this to the population size of the  stand
at the superplot spatial scale.  This deficiency would hinder the evaluation
of reliability of the systems models.

     The audit of data elements also revealed that we were not going to
obtain the kind of dead tree data necessary to tell the computer how to kill
a tree in a biologically reasonable way during simulation.  Subsequently,
the data elements shown in the SPRMORX dataset (see Appendix 1) were defined
for field data collection.  These include, for each mortality center dead/
damaged tree, recent radial growth increments, and height to lowest  branch
bearing needles.

     Additional supplementary datasets became identified as needing  to
be established in the SBNF data base.  These include SBNFPUBS, PLOTINDX,
FSMTINDX, OXIDINDX, and PLOTSEED.  A description of each of these is found
in APPENDIX 1.

     Several cases resulted in improvement in consistency of dataset con-
tents. Since several different datasets contained species identification and
tree tag numbers, we ran a program to compare for taxonomic agreement across
all the datasets, tree-by-tree, for all vegetation plots, and had a  list
printed of any and all disagreements.  The need to verify agreement between
datasets on this data element was fundamental in order to proceed with
any other tree-related data analysis.  When the task was done, it became
evident that several datasets from various years and/or different invest-
igators contained discrepancies as to the taxonomy to be associated with
a given tree number.  Two major reasons for these discrepancies seemed to be
the degree of hybridization which occurs in some areas between ponderosa
pine, Jeffrey pine, and Coulter pine, and the problem of mis-reading the tag
number of a tree when making and recording observations.  Several project
investigators subsequently used these lists in the field to recheck  specific
trees and plots, which led to improved consistency of data in TRID,  CTREE,
TREE, DISU, ISURV, and TREEGRO datasets.

     For both years, 1974, and 1976, of the Disease Survey (DISU) dataset,
a comparative listing was made by the computer, tree-by-tree, for any

                                    18

-------
differences in the data entries for a given  tree between  the  two  surveys.
The results were returned to the plant pathology subproject for resolution
of inconsistencies.

     Other internal examinations of TRID and TREE  led  to  enhancements  in
their contents.

     Summary of data base status— We have used the Dataset Progress Check-
list (Appendix 2) to assess the present overall state  of  the  project
data base.  Table 5 contains a summary of the  findings.   The  proportion
of the data base in various stages indicates that  we are  ready to  concen-
trate more on cross—disciplinary data analysis for modeling in the next
year than has been possible since data collection  began four  years ago.
However, we urge the subproject leaders to make sure they prepare  and  submit
verified card decks as soon as possible following  data collection, so  that
joint analysis between their subprojects and the Ecosystem Modeling subpro-
ject can be done with a minimum of delay.

Developments in Ecological Subsystem Modeling—

     A previous report (Kickert, 1977) highlighted the interrelational
structure between various submodels being developed.   A population dynamics
accounting (STNDCMP) for trees in the forest stand is  driven  by submodels
dealing with tree regeneration (SEED, SEEDLING, LITTER, LITDECAY)  and
stand mortality (ROOTS, BEETLE).  Each of these model  subsets is driven  by a
stand moisture subystem (WATER), as well as by external inputs of  air
quality monitoring data (OXIDANT dataset and PLOTOXID  dataset).

     In Figure 1, the first important link for use of  the project  data
base in the modeling activities is evaluating  the  quantitative nature  of the
relations which have been hypothesized in the  flow chart  of the various
ecological subsystems (Kickert, 1977). A computer  subroutine  is being
written to simulate  each subsystem is in the process.  The sequence
in which each of the various subroutines will be activated on the  computer,
for passing information from one simulated subsystem to another, is shown
from left to right in Figure 5.  Prior to the  computer terminal user telling
the simulators to begin running, the user is first given  a series  of options
for running the simulation.  These options are for setting numerical values
for:  starting year and ending year; site and  tree species parameters
which will not change during the simulation; initial forest stand  condi-
tions; the nature of long-term trends in meteorological conditions which the
user wants to drive the simulation; and the  format of  the output display to
be used. All of these options have default choices built-in,  with  the
additional option of being able to display the default values at the term-
inal, so the user may avoid making decisions to override  these if  so desired.
Figure 5 also shows which of the datasets, listed  in Table 1  and Appendix  1,
are being, or will be, used to quantify relations  in each of  the subsystem
models, and which will be used to provide the  external physical environ-
mental data to drive the set of simulators.

     Computer simulation programs have been written for WATER, TREEGROW,
and partially for BEETLE.  Details of the CANOPY submodel structure have

                                    19

-------
TABLE 5.  DISTRIBUTION OF DATASETS BY DEVELOPMENT STAGE FOR  THE  SBNF  DATA
          BASE AT UC BERKELEY.


                                 Number of             Percentage  of
   Development stage+            datasets                data base
1A. Descriptor not yet
    written for Data                49                       79
    Dictionary-H-

1B. Dataset not yet key-
    punched (or converted)//         30                       48

2.  Dataset card deck entered
    but not yet verified            16                       26

3.  Dataset verified and
    corrected on disk##             10                       16

4.  Dataset presently in
    analysis for modeling            6                       10
    within the Ecological
    Modeling - Data Mgmt
    subproject
+  does not pertain  to updates for post-January  1977, for prior
   existing datasets, but does include new datasets  through August  1977;

-H- this stage  is not mutually exclusive as the subsequent stages
   are, so it  should not be  compared against  them;

#  raw data converted to a new form, or transferred  from the U.S. Forest
   Service to  the  SBNF data  base;

## some datasets counted in  this stage may be in the next stage at
   UC Riverside.
                                     20

-------
                 Simulation-User s Choice
                                                     Simulation Sequence
Response
N3
1 scare i
/Over A.
/ ride \
(defaults?
( Starting/Ending years)
— ta> r i 1 1 1 1

FI
1 L1
,cay|
.DECOMP, DRIP

LTKKEM, St'CSOLKM |
	 1 . _ Seedlinq 1
Litter 1 'ii •
1 0 I I
PLITR I
PNFALL
TREELtT
LITMAS
I
CTREE, CONE, GCONE
PLOTSEED
I-»J Water 1 	 _ Canopy 1 	 ». Treegrow 1 	 1
'••••U 3 1 4 j
/'parameters forA MQ
Vsite & speciesy Mi
STEXOM, SXSCAT
(initial Forest Standj
STAGE
(Driving data generators\
OXIDANT, PLOTOXID ^\
HPREC, PLOTPREC
HMET, FSMET, PLOTMET
(Output display format)
L — 	 	
1ST OZFLUX
TRIG STOMRES
INJBIWK
\
ha
^X^/ 0
TRID ' TREEGRO:
SAPS
LSOIL
SSAS
SLSS
PLOTREGN

»- r"


TREE BOGRO . , . |
SAPTREE SAPGRO mortality!
floot-lo I 	

BTREE, EGG,
REAR, STIK,
| XRAY
rvest 	 ^ Roots I 	 1
3 1
•i tit fi
DISU
ISURV
TREEMORT
SPRMORT1
SPRMORT2
SPRMORX _ ..
1 FASP TREEPEST /!
,er \ ^~
Venter rupt^/


Stand
moosition
12
STAGE
TREEVEG
FIRESTAG
FIRETREE
'rimary \
)utput 1
)ispjay I


            Figure 5.   Simulation sequence  between subsystems (numbered  rectangles), showing
                       associated datasets  (capitalized) for submodel quantification and
                       validation.

-------
been worked out, and the next step will be  to convert  it  into program  code.
ROOTS, LITDECAY, LITTER, SEED, and SEEDLING subsystems  should have  the first
versions of the simulation programs written in  the next six months.

     The stand moisture subsystem simulator— The WATER submodel  is  used
to take precipitation and air temperature data, and simulate biweekly
changes in soil water in the tree root soil depth.  This  is then  used  as
input to the CANOPY subsystem, along with ambient oxidant air pollution
data.  The WATER submodel was a modification of that reported by  Sollins  et
al.  (1976) from the Coniferous Forest Biome.  We ran our  version  on  the
computer and found it to behave in an intuitively reasonable fashion.
Further work in adapting this submodel to the southern  California forest
sites had to be postponed because of unavailability of moisture data and  the
entire data base problem encountered during the past year.  Because  of  the
data base achievements described earlier in this report, we are now  in the
position to resume development and application  of this  subsystem  model  to
our needs and goals. A new detailed flow chart was constructed for the  logic
of this submodel. The one supplied by the authors of the  original model
(Sollins et al., 1974) was too general to clearly portray the model's
relations in the midst of critical review sessions with the soils subproject
investigator.

     The tree stem growth and canopy subsystems simulators— A second
example of our effort to build on the work  of our colleagues elsewhere is
found in our subsystem simulators for tree  canopy changes and tree  stem
growth.  We analyzed the structure of a forest  succession simulator, called
SUCSIM, developed in the Coniferous Forest  Biome.  Reed (1976) has described
the  theoretical ecology foundation for the  model, and has documented the
mathematical details of the various functions (Reed and Clark, 1976).   This
model was developed as an adaptation of the basic tree  growth theory earlier
developed in the easterb United States for  deciduous forest systems  simula-
tion by Botkin et al.  (1972).  In contrast  to the many  mensurational forest
growth models based on site-index, Reed's simulator has trees grow  in
response to physiological definition of tree species niches, with regard  to
availability of needed environmental resources  for growth, light, heat,
moisture and nutrients. Because of this particular theoretical base, this
model appears more suitable for restructuring for the  study of growth
effects of environmental pollutants than do other tree  growth models.

     The structure of the Reed model can be viewed as  consisting  of  two
modules relating to individual tree growth, a crown module, and a stem
module.  These feed into three modules at the tree population or  stand
level, a population regeneration module, a  population  mortality module, and
a population update module.  The primary variables which  describe the  state
of the system in this model at any time are basal diameter of individual
trees, height of individual trees, total leaf biomass  of  individual trees,
age  and species of individual trees, diameter at breast height of individual
trees, and number of live trees in the population.  The only state-variable,
and  respective mathematical functions, in this  list for which we  presently
have no data in the SBNF data base are those related to leaf biomass.

     We have assembled the  computer code for Reed's stem  module as  our

                                    22

-------
TREEGROW subsystem, and are restructuring his  crown module  to  respond to air
pollution injury so we can use  the modified version as  part of our CANOPY
subsystem. We are revising his  simplified population  regeneration and
mortality modules since the SBNF program was designed to  investigate  and
model these processes at a higher resolution (SEED, SEEDLING,  LITTER,
LITDECAY, and ROOTS, BEETLE). What we have identified as  his population
update module is similar to our STNDCMP subroutine.   This is the  master
calling procedure that keeps account of how many  trees  of various ages and
species exist at any given time during the simulation.

     Quantification of site and species parameters and  transfer functions—
As indicated under the previous discussion on  the progress  status of  data-
sets, we have now organized the data base sufficiently  to move more into
quantitatively analysis of transfer functions  for the various  submodels.
Our audit of the datasets and simulataneous conceptual  development of
submodels has led to the discovery that certain kinds of  analyses are
possible and data collection techniques are available that  could  improve the
quantification process for various subsystem models.

     A design of foliar biomass analysis for trees under  varying  degrees of
oxidant air pollutant injury is suggested to improve  useability of the tree
stem growth submodel. Functions in the original stem  growth simulator  now
used were constructed for essentially pollution-free  forests.  This analysis
is necessary to couple the oxidant-foliar injury  submodel (CANOPY) with the
tree growth submodel (TREEGROW).

     Development of the oxidant-uptake and crown  foliar injury submodel
is designed on the hypothesis that changes in  transpiration affect the
amount of oxidant taken into leaf tissue and the  subsequent visible injury
found over long periods.  We suggest that the  appropriate investigator
strongly consider an experimental treatment on pine foliage to demonstrate
the degree of validity of this  hypothesis.  This  suggestion involves  repeat-
edly treating samples of new foliage, during Spring-time, at a known heavily
polluted site, to applications  of an antitransparent, and then following
the seasonal changes in the various foliar injury symptoms  typically  assoc-
iated with ambient oxidants.

     To apply the stand moisture simulator, WATER, to forest plots in  the
SBNF, so that the hypothesis on transpiration  control of  oxidant  uptake may
be simulted, we suggest that the soils subproject strongly  consider perform-
ing a hammar seismograph analysis to determine the maximum  soil depth  on
those vegetation plots where this information  is  not  already available as a
result of the past installation of soil moisture  sensor profiles.

     For a part of the organic LITTER dynamics submodel which  responds
to the oxidant uptake crown injury subsystem,  we  suggest  that  the appropri-
ate investigator consider getting data on:  1) the distribution of various
areal densities of coarse woody litter on the  ground  on vegetation plots;
and 2) the time rate of change  of the amount of woody matter which falls
from various tree species after individuals are killed  by a pest  complex.

     Design of approach for evaluation of reliability of  the simulation

                                    23

-------
package—With regard to the simulation models developed  in  this  project,
the project officer has advised that  the  stage at which  the Corvallis
Environmental Research Laboratory desires  to receive  information is  with
the documentation of the reliability  of the models  as  compared to actual
real forest response.  Reliability of the  models can  be  examined using  a
method of successive approximation. An easier, but  less  objective, method  is
to obtain "the reactions of experienced field observers  to  the predictions"
of the simulation package  (Botkin et  al.  1972).  A  more  objective, but  more
difficult, method is to run the simulation with historical  input data and  to
compare the computer model output behavior with historical  data  on the  same
output variables.

     In some cases, there  appear to be adequate types  of data for evaluating
the reliability of individual submodels,  such as WATER and  CANOPY, over
the relatively short-term  study period 1973 through present.  However,  we
are convinced that the reliability of the  entire linked  set  of subsystem
simulators, producing simulated annual output on numbers of  trees by species
by age, can only be evaluated over a  long  time span of data. If  we knew what
the tree species composition on the vegetation plots were in 1920 and
1950, we could initialize  the simulation package for  1920,  and using histor-
ical weather data from HPREC and HMET datasets (Appendix 1), run the simula-
tors to 1950, covering the recent pre-air  pollution era.  We could then
compare simulated forest tree species composition at "1950"  on the computer
with the actual 1950 known composition.   By using quantitative techniques
developed by Miller for reconstructing the increasing  oxidant air pollution
trend from 1950 to 1973, and using monitored data for  1973  through present,
we could run the simulators for the air pollution era  and compare simulated
forest species composition at "1973"  through "present" on the computer  with
the known field composition data for  the  same years.   For the latter, the
datasets STAGE and TREEVEG, collected in  1973, are  in  the SBNF data  base.
Assuming these reliability evaluations showed that  the simulators tracked
the field data closely enough for 1950 and the mid-1970's to preclude the
decision that the models'  behavior were unreliable, then we  could begin
using the simulation package to perform "what if..." experiments (Figure 1
herein, and 26 in Kickert  (1977)) on  the  computer for  the future period 1980
through 2025.  We are using the philosophy that one cannot  directly  prove
that a simulation model is relizable; one  can only  fail  to  show  that it is
unreliable, the initial a  priori assumption, after  repeated  attempts to
discover the unreliable behavior.

     All investigators in  the SBNF project must be  aware that the ultimate
usefulness and acceptability of all of the subsystem simulators  and  accept-
ability of all of the subsystem simulators to outside  users, including
the Corvallis Environmental Research  Laboratory, pivots  on  our ability
to evaluate the reliability of the simulators produced in this program.
Inability to evaluate reliability could lead to future potential users
regarding these ecological system simulators as simply academic  exercises.

     At present, the research design  does  not include  obtaining  the  right
kind of numerical data on  the SBNF which would allow  for evaluating  the
reliability of the simulation models  over  the time  span  1920 through 1950
(the clean air era), and into 1950 through 1970/80  (the  present  air

                                    24

-------
pollution era).

     In order  to evaluate  reliability  of  the simulation package for prospec-
tive users, we urge the vegetation  investigators  to consider collect-
ing data, for  the  18 permanent vegetation plots,  on tree stump and snag
locations, species, diameter, year  of  death  or  cutting,  and  age when death
or cutting occurred.  These  data  are needed  for synthesis with the stand age
data (STAGE in Appendix 1) collected in  1973, in  order  that  reliability
analysis can be performed  as objectively  as  possible.

CONCLUSIONS

     The project officer has advised that the stage at which the Corvallis
Environmental  Research Laboratory desire  to  receive information on simula-
tion models developed in this project  is  with the documentation of reliabil-
ity of the models  as compared to  actual real forest responses.   Reliability
of the models  can  be examined using a  method of successive approximation.
The easiest, but least objective, method  is  to  obtain "the reactions
of experienced field observers to the  predictions" of the simulation pack-
age. A more objective, but more difficult, method is to  run  the simulation
with historical input data and to compare the computer model output behavior
with historical data on the  same  output variables.  At present,  the research
design does not include obtaining the  kind of numerical  data on the SBNF
which would allow  for evaluating  reliability of the simulation models
over the time  span 1920 through 1950 (the clean air era),  and the 1950
through 1970/80 (the present air  pollution era).   In some cases,  there
appear to be adequate types  of data for evaluating reliability of individual
submodels.

     The proportion of the SBNF Data Base in various developmental stages
indicates that we  are ready  to concentrate more on cross-disciplinary
data analysis  for  model development in the next year than has been possible
since data collection began  in the  SBNF project 4 years  ago.

RECOMMENDATIONS

Planning Future Project Activities  In  Manuscript  Preparation

     For purposes  of planning data  processing tasks and  associated manu-
script preparation by the  project investigators,  we suggest  that  the princi-
pal investigator,  in consultation with subproject leaders, define all of the
research activities felt to  be necessary, and all of the  manuscripts con-
templated, within  this program during  June 1978 through May  1980.

Reliability Evaluation of-Ecosystem Simulation  Submodels

     In order  to evaluate  the reliability of the  simulation  package for
prospective users, we urge the vegetation investigators make  every effort
to collect data, from the  18 permanent vegetations plots  to  show:   tree
stump and snag locations,  species, diameter, year of death or cutting,  and
age when death or  cutting  occurred.  These data are needed for  synthesis
with the stand age data collected in 1973, in order that  reliability

                                    25

-------
analysis of the simulation models can be performed  as  objectively as  possi-
ble.

Development of Ecosystem Simulation Submodels

     For usability of the tree  stem growth  submodel, we  suggest  that
a foliar biomass analysis be designed for trees  under  varying  degrees of
of oxidant air pollutant injury.  The functions  in  the original  stem  growth
simulator being used were constructed for essentially  pollution-free  forests
and analysis is necessary to couple the oxidant-foliar injury  submodel
with the tree stem growth submodel.

     Development of the oxidant-uptake and  crown foliar  injury submodel
is designed on the hypothesis that changes  in  transpiration  affect  the
amount of oxidant taken into leaf tissue and the subsequent  visible injury
found over long periods.  We suggest that the  appropriate  investigator
strongly consider an experimental treatment on pine foliage  to demonstrate
the degree of validity of this  hypothesis.  This suggestion  involves  repeat-
edly treating samples of new foliage, during Spring-time,  at a known  heavily
polluted site, to application of an antitranspirant, and then following
the seasonal changes in the various foliar  injury symptoms typically  assoc-
iated with ambient oxidants.

     In order to apply the stand moisture simulator to forest plots in
the SBNF, so that the hypothesis on transpiration control  of oxidant  uptake
may be simulated, we suggest that the soils subproject strongly  consider
performing a hammarseismograph  analysis to  determine the maximum soil
depth on those vegetation plots where this  information is  not already
available as a result of the past installation of soil moisture  sensor
profiles.

     As a part of the organic litter dynamics  submodel which responds
to the oxidant-uptake crown injury subsystem,  we suggest that  the appro-
priate investigator consider getting data on 1) distribution of  various
areal densities of coarse woody litter on the  ground on vegetation  plots,
and (2) the time rate of change of the amount  of woody matter which falls
from various tree species after individuals are killed by  a  pest-stress
complex.

     For the datasets to be submitted to the Data Management subproject,
we urge subproject leaders to make sure that they prepare  and submit  veri-
fied card decks as soon as possible following  data  collection, so that
joint analysis between their subprojects and the Ecosystem Modeling sub-
project can be done with a minimum of delay.

Interested Cooperating Agencies

     The funding agency should  recognize and act upon  the  need of the
systems ecologist to have as clear a definition as possible  of the  ways in
which the funding agency could  use the information being sought  from  the
research project.  It is recommended that the  project  systems  ecologist,
agency project officer, and other informed  agency personnel  obtain  a

                                    26

-------
conceptual information flow model describing how the environmental informa-
tion (data base, computer models, and ecological insights) resulting from
this research can be made available.  Through a computerized information
transfer, delivery could be made to other environmental scientists, admin-
istrators, legislators, and interested general public, for the purpose of
evaluating secondary standards for photochemical air pollutants, determining
possible consequences of alternative ambient oxidant trends, and identify-
ing alternaive forest management practices.

     Research proposals which indicate that a variety of different kinds
of related data collection is planned by more than one investigator, should
be required to show evidence in the proposal that a usable data base
management system and data dictionary processor are already available
and will be used at the time that data begin to accumulate.  This should
expedite the rate at which collected data are analyzed.
                                    27

-------
TREE POPULATION DYNAMICS SUBSYSTEM

Introduction

     Environment and particularly stress strongly  influence  successional
change in plant communities and contribute significantly  to  structure
the composition of these communities.  Photochemical oxidant air pollutants
are stress factors which invade the San Bernardino National  Forest  (SBNF)
and may play an important part in directing successional  changes.   A gradi-
ent in air pollutant exposure is recognized in the SBNF and  differential
susceptibility of plant species to the pollutants has been demonstrated.
Therefore, it can logically be hypothesized that the oxidant air pollutants
may strongly influence successional change and through years or decades
of exposure may be an important factor in determining plant  community
structure.

     The vegetation subsystem project is focused on describing:  1) plant
communities within the mixed conifer forest type in the San Bernardino
Mountains, and 2) the impact of oxidant air pollutant on  successional
changes in these communities.  Initial characterizaton of major plant
communities has been reported by McBride (1977).  This report summarizes two
studies conducted during 1976-77 of community description and successional
change.  One study was designated to identify sub-units (facies) occurring
within major plant communities and the second was aimed at classifica-
tion of forest sites as a first step in the description of plant succession.

Research Objectives

1.   To identify and map facies within the plant communities dominated
     by yellow pines in the San Bernardino mountains.

2.   To classify sites within the Jeffrey pine dominated  forests of the
     San Bernardino Mountains on the basis of environmental  parameters.

Literature Review

     Variation in forest composition in the San Bernardino Mountains has
been discussed by Horton (1960), Minnlch et^ a.1 (1969), Miller and McBride
(1973), McBride (1973), and McBride (1977).  These authors identified  a
variety of forest communities at the association level  (Braun-Blanquet,
1932).  The classification developed by McBride (1977) established  five
associations within the general yellow pine types:  ponderosa pine  forest,
ponderosa pine-white fir forest, ponderosa-Jeffrey pine forest, Jeffrey  pine
forest, Jeffrey pine-white fir forest, but these associations have  not
previously been subdivided into facies.
                                     28

-------
     Forest succession in the San Bernardino Mountains was  discussed
in general terms by Miller and McBride  (1973), McBride (1973),  and McBride
(1977). Wildfire was determined by these authors to have an important
control over forest regeneration and age structure.  Minnich (1974) eval-
uated the role of major fires in initiating secondary succession  over  large
areas of the San Bernardino Mountains where the yellow pine type  was adja-
cent to extensive chaparral areas.  As  suggested that both  the  rate and
pattern of recovery of forest tree species following five was variable.
Specific classification of sites on the basis of environmental  parameters
has not been used previously to study forest succession.

Materials and Methods

Field Sampling Techniques—                    •

     Eighteen permanent plots established in 1972 and 1973  (McBride, 1977)
were used in the identification and mapping of facies.  The facies were
recognized on the basis of species composition, plant height, and cover
in the tree, shrub, and herb layers of  the forest.  Boundaries  were es-
tablished when a change in species composition or cover (more than 25%)
occurred.  The basic procedure was to walk the entire plot  in order to
survey the variation in tree, shrub, and herb layers before mapping of the
facies.  The recognition variable (species composition, plant height, and
cover) were recorded for each facies.

     Data collected on 45 Jeffrey pine plots of the 83 temporary  plots
established in 1974 to in investigate forest condition as a function of time
since the most recent fire (McBride, 1977) were used in the classification
of sites.  No additional field data were collected from these plots in
1976-77.

Laboratory Analysis Procedures—

     No laboratory analysis procedures were appplied to the facies identi-
fied and mapped on the 18 permanent plots.

     The field data obtained from the Jeffrey pine plots along  with data
obtained from published sources (i.e., U.S.G.S. topographic maps) were used
to classify sites with a numerical taxonomic clustering technique described
by Sneath and Sokal (1973).  The following parameters were  used for the
classification:

     1.  elevation
     2.  slope
     3   radiation index
     4.  precipitation
     5.  soil depth
     6.  water surplus
     7.  water-holding index
     8.  percent clay
     9.  percent sand
    10.  soil fraction greater than 2 mm

                                    29

-------
    11.  A horizon pH, color and chroma
    12.  C horizon pH, color and chroma
    13.  slope aspect
    14.  percent rock cover
    15.  percent bare ground
    16.  position of plot on slope
    17.  length of slope
    18.  microrelief (i.e., concave, flat, convex)
    19-  macrorelief (i.e., level, undulating, rolling, hilly, steep)

Based on the values of each of the above parameters similarity coefficients
were derived that compared each site with every other site.   (A similarity
coefficient is a numerical representation of the overall similarity between
two sites).  Sites that had the highest average similarity values were
grouped together thereby defining clusters containing members (sites)
that possessed a high degree of resemblance.  These clusters  are represented
in a tree-like diagram (i.e., phenogram) that is a two-dimensional represen-
tation of the interrelation of the study sites (Fig. 6).

    Stepwise discriminant analysis was subsequently applied to determine
if any of the sites should be reassigned to other clusters and to evaluate
the relative importance of each parameter used to define these clusters.

Results and Discussion

    A total of 189 facies were identifed on the 18 permanent  plots.  Seven-
teen of these occurred on more than one plot (Table 6).  The  facies map
(Fig. 6) and descriptive data (Table 7) for the Dogwood plot  is presented
as an example of the maps and data prepared during the study-  Maps and data
for all 18 permanent plots are available from the Forest Ecology Laboratory,
Department of Forestry and Conservation, University of California.  The
facies maps will provide a basis for relating forest regeneration and plant
succession to local variations within the forest types.

    The phenogram produced by cluster analysis revealed five  fairly distinct
clusters (Fig. 7).  As one moves along the horizontal axis the first
visually apparent cluster includes sites 78 through 50.  The  membership
for the second, third, and fourth clusters are respectively:  sites 66
through 55; sites 62 through 59; and sites 71 through 46.  The last cluster,
sites 60 through 64, can be viewed as a loose assemblage of sites that are
least like any of the other sites.

    Results of discriminant analysis do not reveal any significant changes
in group membership.  However, analysis of the territorial map produced by
discriminant analysis (Fig= 8), depicting a cluster summary of discriminant
scores 1 and 2 for each site, reveals with which parameters the  clusters
are associated.  The horizontal axis represents a moisture complex increas-
ing to the right.  Precipitation and water surplus variables  provided  the
greatest influence in this complex.  The veritical axis represents an
exposure complex decreasing upwaves.  The greatest influence  in  this complex
was provided by aspect, radiation index, and macrorelief variables.  The
map illustrates how the clusters are evaluated along these environmental

                                    30

-------
             DOG WOOD-165m
Figure 6.  Facias map of Dogwood Plot (see Table  7 for description of
          facies).
                                 31

-------
TABLE 6.  NUMBER OF FACIES IDENTIFIED ON THE  18 PERMANENT  PLOTS USED TO
          MONITOR AIR POLLUTION INJURY TO FOREST TREES.
                    Plot
                          Total
      Number of Facies
        Distinct     Common  to
        to Plot     Other Plots
	 - — -* ' — 	
Ponderosa pine


Breezy Point
U.C. Conference Grounds
N.W. Camp Paivika
10
4
8
9
4
5
1
0
3
Ponderosa pine-
  White fir
Ponderosa-Jeffrey
   pine
Jeffrey pine-
  White fir
Jeffrey pine
Camp Angeles
Camp 0-ongo
Dogwood
Schneider Creek
Sky Forest
Tunnel Two

Barton Flat
Green Valley Ck
Camp Osceola

Bluff Lake
Heart Bar
N.E. Green Valley
Holcomb Valley
Deerlick

Sand Canyon
13
11
11
15
15
13

 8
23
 9

15
 9
11
13
10

16
 8
 9
11
14
13
11

 8
20
 6

 9
 9
 6
 6
 9

15
5
2
0
1
2
2

0
3
3

6
0
5
7
1
                                    32

-------
    TABLE  7.  SPECIES  COMPOSITION,  TREE HEIGHT,  AND COVER OF FACIES ON DOGWOOD PLOT,
FACIES ;
NUMBER ;
TREE LAYER
Upper Middle Lower
6-J!
rH H
« CD
•U >
0 O
H U
1 I 0
I
2
I
3
4
5
6
7
8
9
i
5-25
50-75
1-5
0
n
25-5
0
75-100
4->
•a
•rt
1

50+
50+
50+


50+

50+
>
?
en s^s : B^
cu ;
•H M r i-l H
on) ; cd a)
a) > : 4-j >
& o i o o
CO O ! H O
i
t

IC5-25
PP50-75
1C 1-5


PP25-50
1C 1-5

WF 5-25
1C 5-25
SP 5-25
i
0
25-50
,
50-75
0
0
0
50-75
75-100
50-75
i
4-1
•a
•H
3

15-25
25-50



10-25
10-25
25-50
Species
Cover %

1C 5-25
BO 5-25
IC50-75
PP 1-5



PP50-75
1C 5-25
WF 1-5
B075-100
PP 5-25
1C 1-5
PP50-75
PP 1-5
WF 1-5
1C 1-5
6^2
H M
tfl 0)
4-1 >
O O
H 0
0
25-50
0
1-5
0
0
5-25
75-100
i
CO &-S
4-1 
0) ft O
Cd ; en u

10-25 WF 1-5
PP 5-25
1C 5-25

6-10 1C 1-5
BO 1-5
i
|
\
i
•j
6-10 PP 5-25
1C 5-25
6-10 PP75-100
1C 5-25
i
25-50 | 10-25 SP 5-25
1
GROUND
SHRUB LAYER LAYER
B-S
rH ^
rt a)
4-> >
0 0
H 0
0
0
1-5
0
0
0
0
0
0
4-»
•a
•H
3


>3






Species
Cover %


B01-5






CO &^
0)
•rl M
O 
ft O
en o
BG 50-75
G 25-50
BF 1-5
L 75-100
BF 50-75
G 50-75
L 75-100
BF 1-5
BF75-100
F 1-5
BF 1-5
L 1-5
L 50-75
BG25-50
BF 5-25
G 1-5
BG75-100
L75-100
BF25-50
G 1-5
L75-100
BF 1-5
L75-100
BF 1-5
OJ
OJ

-------
TABLE 7.  CONTINUED
FACIES
NUMBER
10




11



TREE LAYER

6^
T-l M
ctf CU
4J >
O O
H U
85-100




5-25



Upper Middle Lower
Height
50+




50+



Species
Cover %
PP75-100




PP 5-25



^s
i-i M
ti 
O 0
H 0
0




50-75



4-1
&
M)
•H
CU
W





10-25



Species
Cover %





PP25-50
B025-50
SP 5-25
WF 5-25
&-S
.H h
ed OJ
•U >
0 0
H O
50-75




5-25



jj
,£5
bO
•H
a)
ffi
10-25




6-10



CO B-S
OJ
•H M
O 0)
0) >
a o
C/3 O
WF25-50
PP 5-25
BO 5-25
SP 5-25
1C 1-5
WF 5-25
PP 5-25
SP 5-25

SHRUB LAYER
6-5
i-l h
n) a)
J-> >
0 O
H 0
0




0



w
M
•H
a)
W









Species
Cover %









GROUND
LAYER
Species
Cover %
L75-100
BF50-75
G 1-5


L75-100
BF 5-25


*Height in feet
PP = ponderosa pine;  BO = black oak;  SP = sugar pine;  WF = white fir;   1C = incense cedar;
BG = bare ground;  G = grass;  BF = bracken fern;  L = litter.

-------
      .60 r-
      .65
      .70-
      .75
      .80
    O
    z
      .85

    ^
    at



    I .90
    1/1
      .95
      1.0
f?
                                            OINJ
n
                                  PLOT  NUMBER
Figure 7.  Phenogram illustrating cluster formation.
                                      35

-------
-5
5.00-
3.75-



2.50-

1.25-


0-


•1.25-




2.50-

3.75-
•5.00-
•°° -2.50 0 2.50 5
11 ii/ii
t
i
3
i
i
<
/ 4
"**• -« ' N
"~x S "^^
^\ 5 ^^,
---'
\ *-~
0 \ *-'*
z v— - ^
1
1
1
1

1
'
1 1 1 I 1 1 1
-3.75 -1.25 1.25 3.75
.00
•5.00
-3.75



-2.50

-1.25


-0


—1.25




--2.50

--3.75
•-5.00
Figure 8.  Territorial map of discriminant score 1 (hortizontal) vs.
           discriminant score 2 (vertical).  (Number indicates group
           centroid of respective clusters.)
                                    36

-------
complexes.  Each of the four clusters are located  in  a  different  quadrant
indicating that these two environmental complexes  successfully  distinguish
the clusters.  The fifth cluster, composed of anomalous  sites according to
phenogram structure, is located in the center of the  map  thereby  substanti-
ating its status.

    Cluster #1 (located in the lower right quandrant  of  the  territorial
map) is exposed to an evnironment that is associated  with  a  high  moisture
complex and a high exposure complex.  Cluster #2 is located  in  the  lower
right quadrant of the map.  This indicates an environment  that  is associated
with a low moisture complex and a high exposure complex.   Cluster #3,
located in the upper right quadrant, is distinguished by  its association
with the low moisture and low exposure complexes.  Cluster #4,  situated
in  the upper right quadrant, is characterized by a high moisture  complex
and a low exposure complex»

    Since objective techniques were used to group  these  sites,  intra-
cluster variation was minimized and each cluster can  therefore  be considered
relatively homogenous.  This allows for data collected from  these sites
to  be used to predict plant succession.  However,  since  these clusters  are
ordinated along the axes of the territorial map and hence  exposed to
different environments, variations in plant successions between clusters  is
likely.  This assumption will be tested in the next phase  of the  study.
                                    37

-------
OXIDANT DOSE - CANOPY RESPONSE SUBSYSTEM

Introduction

    A number of different data gathering activities have been undertaken in
order to provide the background needed to reach  the following objectives:

    1)   Determine the influences of the May through October oxidant
         dose and climate on biweekly increases  of injury symptoms  on
         the foliage of ponderosa and Jeffrey pines and black oak.

    2)   Investigate the consecutive year results of tree injury  in terms  of
         both the oxidant injury index or score  of overstory trees  and  the
         height growth of sapling trees.

    In objective 1, the data sets that provide hourly temperature and
relative humidity information near vegetation plots (FSMET) or on vegetation
plots (PLOTMET) were used to compute cumulative  daily transpiration for the
whole season using a transpiration simulation model (Reed and Waring,
1974). Other data required by the model are designated as STOMRESIST.
Operationally they consist of biweekly predawn xylem water potential
measurements and associated minimum daily stomatal resistance measurements.
A second type of data needed for objective 1 are the hourly ozone or total
oxidant averages at stations nearest the vegetation plot (OXIDANT)  and  on
the vegetation plots (PLOTOXID).  A third data set required by objective 1
is the biweekly change of injury to needles or leaves of selected trees at
each of five vegetation plots (INJBIWK).  Injury data is gathered monthly  at
the remaining vegetation plots.

    It will be necessary to couple the output of objective 1 with other
data sets described for objective 2 (below) after sufficient testing has
established the relationships of transpiration (an integrater of  climate),
oxidant dose and single season foliage injury.

    In objective 2, several effects of multiple  years of injury are being
investigated concurrently but independent of the output from objective  1.
The data sets employed here include the annual evaluation of injury to
all plot trees (TREE and TREEMORT) and to saplings nearby (SAPTREE).  A
retrospective measure of annual height growth of saplings in relation
to oxidant injury (SAPGRO) is also included.

    A supplemental data set (HMET) includes daily maximum and minimums
of temperature, relative humidity and a one time daily wind and precipita-
tion measurement from 9 Forest Service Stations  in the San Bernardino
Mountains starting in 1976.
                                     38

-------
Materials and Methods

Equipment and Calibration Methods—

     Meteorological—The remote  stations  for  measurement  of  temperature,
relative humidity, winds and net radiation,  (FSMET)  at  Camp  Paivika,  Sky
Forest and Barton Flats and their  locations with  respect  to  vegetation plots
have been described earlier (Miller,  et al.  1977).   Hygrothennographs
in standard weather instrument enclosures were  placed on  or  near vegetation
plots.  In May  1976, enough instruments were  available  to place  one  from
June through October at each location:  Camp  Angelus (CA), Deer  Lick  (DL) ,
and Heart Bar (HB).  These instruments were within  0.8  to 1.6  km of  the
vegetation plot of the same name.  In May of  1977,  instrument  shelters
with hygrothermographs were placed at the following  vegetation plots  for
the entire summer season:  CA,DL,  Tunnel  2  (TUN2) and Camp Osceola  (CAO).
The three remaining hygrothermographs were  rotated  among  the other vege-
tation plots in the spring of  1977 with measurements for  periods ranging
from 2-3 weeks.

     Air pollution—DASIBI Model 1003AH ozone photometers were maintained
at the three FSMET stations (CP, SF and BF) in  both  1976  and 1977.   These
instruments were calibrated at the California Air Resources  Board Labora-
tory in El Monte.  Mast Model  724-2 total oxidant analyzers  were calibrated
by one DASIBI which was reserved for  this purpose at the  Statewide Air
Pollution Research Center  (SAPRC). The altitude  correction  factor for
each mountain station was  calculated  by dividing  the larger  pressure  height
(millibars) at  the SAPRC by the  smaller pressure height at each  mountain
station.  Placement of Mast analysers was coordinated with the schedule
described for hygrothermographs  on the vegetation plots.   Mast analysers
and their strip chart recorders  had to be battery powered at the vegeta-
tion plots.  Four, 6 volt, 217 amp-hr batteries were required  to operate
the Mast analysers when used with  a DC to AC  inverter.  Each set of batter-
ies lasted about 4 days.   The  strip chart recorders  were  powered by  an
internal rechargeable battery  pack.   The  delivery of these Datamart Model
D/M #755-M-M recorders was delayed because  of a redesign  required to
lower power drain.  Consequently,  the on-plot oxidant measurements did
not begin until May 1977.  The original schedule  specified an  August
1976 beginning  date.

Measurement of  Biological Variables—

     Data inputs to the transpiration simulation model—The  Reed and  Waring
(1974) transpiration simulator requires both  meteorological  data, namely
hourly temperature and relative  humidity, and biweekly  measurements  of
predawn xylem water potential  and  the associated minimum  daily stomatal
resistance (rs).  Reed and Waring  define  xylem  water potential in its
absolute terms  (Ibs/in^) and refer to it  as predawn  plant moisture stress
(PPMS).  As soil moisture becomes  more depleted during  the growing season
both PPMS and minimum daily resistance are  expected  to  increase, thus,
the daily actual transpiration (Ta) decreases although  atmospheric demand
or potential transpiration (Tp) may remain  large.
                                     39

-------
     Stomatal resistance was related to the only available method  of  measur-
ing transpiration, namely, infiltraton pressure  (IP) in a pressure porometer
(Fry and Walker,  1967).  A branch of ponderosa pine was cut, mounted  on a
clamp and exposed to a turbulent air flow at 2000 ft-c and 25  C  in the
laboratory during a 220 minute period.  Initially, and at 10 minute,  and
later at 30 minute intervals, it was weighed to determine water  loss  due to
transpiration. At similar intervals a needle faside was removed to deter-
mine the corresponding IP.  Resistance (rs) was calculated at  each inter-
val:
            rs  =  vapor density gradient = sec cm~l
                        transpiration


The resistance values ranged from 27 to  115 sec cm"1.  The following
relationships from Reed and Waring  (1974) were used to calculate  rs as  a
function  of IP over  the range available:
             In  rs =  1.433 + 0.053 IP
The key equation necessary to simulate transpiration was the regression
between PPMS and minimum daily IP, the equation also from Reed and Waring
(1974) was:
             IP(lb)  =   (2.50 x PPMS  [bars]) - 0.641, (r =  .53)
Daily ppms values were interpolated between the biweekly data points.
Actual  transpiration  (Ta) was then calculated by solving for diffusion
conductance  (DC) in the equation:
            DC  =   164.56


Potential transpiration  (Tp) was similarly calculated by assuming a maximum
value or the largest observed for DC during the entire season at each plot
location.  Table 8  summarizes the steps described above.

     The regression equations above were used for a preliminary run of
the transpiration simulator for two plot locations, Camp Angelus and Deer
Lick during the May through September period, 1976.  In addition, the ratio
Ta/Tp was calculated daily as a measure of the relative drought stress  at
each location and at different times during the season.  This ratio is
proposed as one criterion for testing the quantitative relationship between
plant moisture status throughout the season and ozone dose response.

     Concurrently, an effort has been under way to improve the accuracy of

                                    40

-------
TABLE 8.  MAJOR INPUTS, INTERNAL OPERATIONS AND  OUTPUTS  OF THE TRANS-
          PIRATION SIMULATOR
         Variables
Source or Product Dataset
Inputs

Mean Hourly Temperature  ( C)

Mean Hourly Relative Humidity  (%)
         FSMET and

         PLOTMET
Predawn Plant Moisture Stress  (PPMS)
(interpolated between biweekly
determinations)

Minimum Daily Stomatal Resistance  (rs)
(as a function of PPMS)
         STOMRESIST
Internally Generated Parameters

Vapor Pressure Deficit  (VPD)
     (night and day)

Diffusion Conductance (DC) =  l/rc
Outputs

Daily and Cumulative:

  Actual (Predicted) Transpiraton  (Ta)

  Potential Transpiration  (Tp)

  Ratio:  Ta/Tp
         TRANSPR
                                    41

-------
the regression between minimum daily stomatal resistance rs and PPMS
without relying on the pressure porometer so that rs can be substituted
for IP directly.  A diffusion porometer fabricated and calibrated by the
Lawrence Livermore Laboratory (Bingham and Coyne, 1977) was used to obtain
some mimimum rs data.  Unfortunately the unusual amount of summer rain
during 1976 and 1977 prevented us from obtaining sufficient data points at
moderate and high levels of PPMS.  STOMRESIST is designated as the data file
for PPMS and rs.  Plot locations where PPMS and rs data have been
obtained include DWA, TUN2, DL, CA, CAO and HB. Trees were selected as close
as possible to  the soil moisture sensor column in each plot or immediately
outside the plot.  Two to  three trees were selected at each plot; one
represented an  overstory tree larger than 30 cm dbh and the second an
understory tree less than  30 cm dbh.  Trees with moderate to severe oxidant
injury were not selected at this time because the limited amount of time
available during the predawn period (3 to 6 am) made it impossible to
observe all injury classes especially because travel time between the 3
plots was a big limiting factor.  The selection of healthy or slightly
injured trees was done to  provide the initial data because these trees are
expected to have the greatest survivorship and thus a requirement for
predictive information.  The concurrent studies being done by Coyne and
Bingham (1978)  include three injury categories, e.g. very slight, slight to
moderate and moderate to severe at one ponderosa pine stand near Crestline.
Their data will be helpful to us when it  is reported.

     Estimates  of injury to foliage at intervals during the season—During
the 1976 and  1977 summer seasons  five plots were visited every two weeks
to observe the  amount of visible  injury  to  selected ponderosa or Jeffrey
pines, white fir and black oak.   A minimum  of  two  trees of  each  species in
each of six oxidant  injury score  categories were  selected  throughout for
continuing observation.  The ponderosa  and  Jeffrey pine  categories  included
0-8 (very severe),  9-14  (severe),  15-21  (moderate),  22-28  (slight),  29-35
(very slight) and 36 and higher  (no visible injury).   White fir  and  black
oak were also divided into six  catagories.   Usually about  three  categories
were present on each plot  as determined  from the oxidant  injury  scores  from
1975.  Preselected  trees were  first  inspected so that disease or insect
problems which  may  interfere with the development of  oxidant injury symptoms
could be avoided.   Three branches  on  each selected tree  species  were tagged
except for white fir which had  one tagged branch.   On the ponderosa and
Jeffrey pines,  eight needle  fasicles  were labeled in each successively
younger needle  whorl starting with the  1975 needle whorl.   Each of  the eight
needle fasicles were enclosed  in  a loop  made with one length of  colored,
vinyl covered copper telephone wire.   Table 9 summarizes the kinds  of  data
gathered either biweekly or monthly  in  1976 and 1977 from mid-June through
the end of September.

     On ponderosa and Jeffrey pines  a "3 M" device was used to measure
total needle length  affected by chlorotic mottle.   This  "metric mottle
measurer" was a transparent plastic  tube about 30 cm long and 0.5 cm,
i.d.  It was painted black on one  third  of  the external  surface for the
full length, and etched every 0.5  cm.   This device which was developed
by Tom Quick was proven to be useful.  When the needle faside was


                                     42

-------
TABLE 9.  DESCRIPTION OF INFORMATION  COLLECTED TO DESCRIBE THE WITHIN
          SEASON DEVELOPMENT OF  OXIDANT  INJURY SYMPTOMS ON PONDEROSA
          PINE  (PP), JEFFREY PINE  (JP),  WHITE FIR (WF), BLACK OAK (BO).
  Species
    Type of Data
Frequency
PP and PJ
WF
BO
For each labeled branch,
and whole needle whorls,
starting with 1975:
—Total number of              June and
  needles per whorl            September

For the same labeled
branch and for 8 labeled
or "wired" needle fasicles
in each whorl since 1975:
—Number of the 8 needles      Biweekly
  remaining.                       or
—Total length of each         Monthly
  of the 8 needles (cm).
—Portion of the total
  length with identifi-
  able symptoms (cm).
—Intensity of injury
  symptoms on each needle
  (Table 10).

For the single labeled
branch:
—Number of annual needle      Biweekly
  whorls retained (first          or
  observation in 1975).        Monthly
—Intensity of symptoms
  on needles or each
  whorl (Table 10).

For each labeled branch:
—Intensity of symptoms        Biweekly
  on representative               or
  leaves (Table 10).           Monthly
                                    43

-------
slipped into the tube for measurement  (down to  the  top of  the  fasicle  wrap)
the needle was isolated momentarily against a dark  background  which  aided
the measurement of chlorotic mottle.   In Table  10 the different  subjective
categories are described for characterizing symptom intensity.   These
categories are more refined than those used in  the  end-of-season oxidant
injury score.  In that case 0 = severe chlorotic mottle and advanced necro-
sis, 2 = any discernable chlorotic mottle and 4 - an uninjured green needle.
In the refined version, the greater injury is assigned a larger  number so
that injury resposes would have a positive slope.   Scores  of 7,  6, and 5  for
the pines, white fir and black oak respectively signaled needle  or leaf
abscision.

     Records of vertical growth on ponderosa and Jeffrey pine  saplings
1967 to 1976—In September of 1976, the internodal  growth  of saplings
was measured with the aid of a fruit picker's ladder (SAPGRO).   Internodes
preceding  1967 were also measured when they could be confidently  ident-
fied.  The oxidant injury scores of each of the 50  ponderosa or  Jeffrey
pines in the plots was last determined in 1975  (SAPTREE).  These  plots
were located within or nearby the following major vegetation plots:  SF,
CP, DWA, BL, BP, TUN2, CA, HV, CAO and HB.

     Annual measurement of oxidant injury to trees  larger  than 10 cm dbh
at major vegetaton plots—The procedure has been described in  earlier
reports.   Comparisons between years is done by  using the paired  t test
to compare each tree with itself from one year  to the next.

     Summary of data gathering activities—Table 11 shows when and where
eight data sets are obtained.  Daily and cumulative daily  transpiration
(TRANSPR)  is calculated from FSMET, PLOTMET and STOMRESIST but it is
included in the table thus bringing the total number of data sets to nine.
The oxidant injury scores for saplings (SAPTREE) is not included  in  the
table because this information has not been obtained on a  regular annual
schedule as originally planned.  SAPTREE and TREE must be  obtained during
the September to November period each year; TREE has been  given  first
priority.  Good weather and qualified manpower  were available  in  1973
and 1975 to allow SAPTREE to be obtained.  It must be done in  1978 without
fail.

Results and Discussion

Meteorological Effects on Seasonal Oxidant Dose—

     One method of summarizing the influence of seasonal climate  on  oxidant
dose was introduced in the last progress report (Miller, et al.  1977),
namely, the frequency of five classes of meteorological patterns  in  southern
California (McCutchan and Schroedor, 1973) and  the  oxidant dose  related to
single and consecutive days of each type.  The  following are brief definition
of these types:

     1)  Hot dry continental air all day (Santa Ana)
                                    44

-------
TABLE 10.  SUBJECTIVE CATEGORIES FOR DESCRIPTION OF OXIDANT INJURY SYMP-
           TOMS ON PONDEROSA  (PP), JEFFREY  PINE  (JP),  WHITE FIR (WF),  AND
           BLACK OAK (BO).
  Species
Numeric Category
Description of Leaf Symptoms
PP and JP
                       -Completely grass green  (PP)
                        Completely gray green (JP)

                       -Slight chlorosis or very
                        slight chlorotic mottle

                       -Distinct, bright yellow
                        chlorotic mottle

                       -More intense chlorotic mottle
                        and some uniform chlorosis

                       -Intense mottle with necrosis
                        appearing at needle tips, not
                        exceeding the distal 1/3 of
                        the needle

                       -Intense mottle with necrosis
                        occupying the distal 2/3 of the
                        needle

                       -Entire needle necrotic, appear-
                        ing dry and brown

                       -Needle abscission
WF
                       -Completely green or gray-
                        green

                       -Light green and/or chlorotic
                        mottle barely distinguishable
                        on the sides (thinnest part of
                        the elliptical cross section) or
                        at the needle tip
                    (continued on next page)
                                    45

-------
TABLE 10. CONTINUED
  Species
Numeric Category
Description of Leaf Symptoms
WF continued
                       5

                       6
                       -Mottle more definite  (bright
                        yellow) sometimes uniformly
                        chlorotic

                       -Intense mottle and the tip 1/3
                        of the needle is necrotic

                       -Uniform yellow and at least 2/3
                        of the needle is necrotic

                       -Needle is entirely necrotic

                       -Needle abscission
BO
      0

      1
-Leaf completely green

-First evidence of interveinal
 chlorosis, chlorotic mottle
 or necrotic lesions mainly
 on upper surfaces

-Moderate levels of interveinal
 chlorosis, chlorotic mottle
 and/or necrosis mainly on
 upper surface

-More severe than 2 with
 necrosis extending to the
 lower leaf surface

-Whole leaf is necrotic, both
 surfaces
                                     46

-------
TABLE 11.  DESCRIPTION OF DATA TYPES AND FREQUENCY OF DATA  COLLECTION FOR
           EACH TYPE AT MAJOR VEGETATION PLOTS  IN 1976 AND  EARLY  1977.
Plot Spec- FSMET/
Name ies OXIDANT
COO p,f,o 	
TCP r> f n _____
or p , t 9 u _— ___
CP p,o hourly
SF p,f»o hourly
rvTJA r\ f rt _____
UCC p , o 	
TTTKT7 n f n _____
TIT -i f n _____
rw r> T f n —————
BL j,f 	
VWfW -i f _____
Of ,, f n _____
TTU •? -F r> _____
5P.R n f n 	
Acronyms for data types
STOM- INJ-
PLOTMET PLOTOX RESIST BIWK TRANSPR SAPGRO
	 	 	

	 	 	
_____ _____ on/near
plot/
biwk
hourly hourly on/near
plot/
biwk
hourly hourly on/near
plot/
biwk
	 	 	



plot/mo 	 	

pxot/iiio _____ ann.
plot /mo 	 ann.
plot- daily/ ann.
sap/ cum.
biwk
plot /mo 	 	
plot- daily/ ann.
sap / cum .
biwk
plot/ daily/ — — — —
biwk cum.
p j,o t./ mo ______ ___ .___
plot /mo -— — — — — —


pxo t. / mo —_-«.___. ann. •
rvl nt* lmr\ ..-..._.- -___.-__
TREE
ann.
ann.
ann.
ann.
ann.
ann.
ann.
ann.
ann.
ann.
ann.
ann.
ann.
O1"»f^
                           (continued on next page)
                                    47

-------
TABLE 11. CONTINUED
                                   Acronyms for data  types
Plot  Spec-
Name  ies
       FSMET/                    STOM-   INJ-
       OXIDANT  PLOTMET  PLOTOX  RESIST    BIWK
                                   TRANSPR  SAPGRO  TREE
CA



BF

CAO



HB
p,f,o  hourly



p,j,o  hourly

j,f,o  	
hourly
                 near    plot-     daily/   ann.
                 plot/   sap/      cum
                 biwk    biwk

                 	   plot/mo   	    	
hourly  on/near plot-     daily/   ann.
        plot/   sap       cum
        biwk    biwk
                                       near
                                       plot
                                       biwk
                                         plot/mo   daily/   ann.
ann.



ann.

ann.



ann.
   Species present:   p - ponderosa pine
                      j - Jeffrey pine
                      f - white fir
                      o - black oak

   Comprised of biweekly predawn plant moisture stress  (PPMS) and associated
   stomatal resistance (rs) measurements made on 2 ponderosa or Jeffrey
   pines
     10 cm dbh; located  in plots or immediately outside plots within  30 m  of  the
   soil moisture sensor  column.
   Needle injury data taken from selected plot trees of all species  >10  cm
   dbh and from sapling ponderosa and Jeffrey pines < 10 cm dbh at 4 plots
   at monthly  (mo) or biweekly  (biwk) intervals.
   Sapling height growth  (SAPGRO) and the oxidant injury score  (TREE)
   are determined annually  (ann.).
                                    48

-------
     2)  Relatively dry forenoon, modified marine  air  in  the afternoon;
         very hot (heat wave)

     3)  Moist, modified marine air, hot  in  the  afternoon

     4)  Moist, modified marine air, warm in the afternoon

     5)  Cool moist, deep marine air throughout  the  day.

     It was concluded that consecutive  occurrences of  Class  3 days  were  the
most numerous in  1974 and resulted  in the highest  daily dose of  oxidant.
Consecutive Class 4 days followed very  closely behind  in  frequency  and  the
size of the resultant oxidant dose.  The  primary reason for  the  higher dose
in 1974 (Fig* 9) was the larger number  of Class  3  followed by 3  transitional
combinations  (Table 12).  The lower seasonal oxidant dose in 1976  (Fig.
9) was evidently  associated with a higher frequency  of transitional com-
binations which induce lower daily oxidant doses,  namely, 5-5, 4-5, 5-4
and 1-1 (Table 12).  In general, the 1976 season was cooler  and  marked by
greater than  usual rainfall particularly  on  September  11, when a tropical
storm persisted over southern California.  The cumulative oxidant dose
in June 1976  was  particularly lower than  the June  of 1974 and 1975  (Fig = 9).
There were 13 Class 5 (cool, moist) days  in  June 1976  compared to 7 in  1975
and 4 in 1974.  These results give a general view  of the  trends  in  seasonal
oxidant dose  in relation to climate.  The results  in the  following  section
will attempt  to show the coupling between seasonal climate,  oxidant dose and
tree injury response.

Daily and Cumulative Transpiration—

     Daily and cumulative transpiration at Camp  Angeles and  Deer Lick,
1976—The transpiration simulator provides both  actual and potential
transpiration on  a daily basis.  Potential transpiration  is  a measure of
atmospheric demand.  The ratio of actual  to  potential  transpiration (Ta/Tp)
is a useful index of the relative drought stress at different locations as
well as throughout the season at the same location.  Several observations
can be listed:  First, the time series  in Figures  10 and  11  suggest that
higher cumulative weekly oxidant doses  coincided with  periods of higher
potential transpiration throughout  the  season at both Deer Lick  and Camp
Angeles.  The scatter diagram (Figure 12) relating daily  oxidant dose to
daily potential transpiration at Camp Angeles in 1976  further suggests a
relationship  between transpiration demand (Tp) and concurrent oxidant dose.
Second, the ratio Ta/Tp at both DL and  CA responded by becoming  lower
following continuous periods of moderate  to  high potential transpiration and
higher following significant rain on Julian  dates  211  and 254.   Third, it  is
possible that a high oxidant dose occurring  during a period  of high Ta/Tp
(higher actual transpiration) would be  more  injurious  than in the opposite
circumstance, namely, a low Ta/Tp.  The reason may be  that ozone flux to
needle tissue may be larger during periods of higher actual  transpiration
(high values  of Ta/Tp).  For example, in  Figure  10 ozone  flux could be
expected to be larger in early July than  in  late August.  In Figure 11, the
first small peak of Ta/Tp coincides with  a short period of high  oxidant in
mid-June and  the lowest value of Ta/Tp  in late august  also coincides with

                                    49

-------
U1
o
     ID
      O
      fO
         5.0-
         4.5-
         4.0-
 o»

 £ 3.5
A|
 k.

 f 3.0-
KJ_
       o>
       a.
         2.5-
 UJ 2.0
 CO
 O
 O
 ,_  1.5
§'-°
O

< 0.5

O
H  o-
PERCENT OF  TOTAL POSSIBLE  HOURS OF DATA-y
JUNE —SEPTEMBER, RIM FOREST/SKY FOREST   /
 89
 84

 SE
 90

 AU
                90


                JL
                90

                JU
90
                       82
                             94
86
85
                      THREE-YEAR MOVING AVERAGE
100
                              100

                              SE
                                                     too


                                                     AU
                                      100


                                       JL
                                      too

                                       JU
IOO    96
                                            IOO
                                             SE
                                     100


                                     AU
                                     100


                                     JL
                                     100


                                     JU
                                                                   90


                                                                   JU
               1968   1969   1970  1971    1972   1973   1974   1975   1976  1977
     Figure 9.  Trend of seasonal oxidant dose at a representative San Bernardino mountain
               station from  1968-1977.

-------
TABLE 12.  FREQUENCY OF DIFFERENT TRANSITIONAL  COMBINATIONS  OF FIVE
           CLASSES OF SPRING AND SUMMER DAYS-
Transitional
Combinations
4
3
5
2
3
4
4
5
1
2
3
2
4
fi
**/
5
2
4
5
3
4
1
3
2
4
2
n
32
53
22
8
11
9
8
9
6
6
4
1
2
1974
Percentage
18.7
31.0
12.9
4.7
6.4
5.3
4.7
5.3
3.5
3.5
2.3
0.6
1.2
n
54
17
22
16
8
9
8
7
9
6
6
4
2
1975
Percentage
32.0
10.1
13.1
9.5
4.8
5.4
4.8
4.2
5.4
3.6
3.6
2.4
1.2
n
37
23
34
8
8
16
7
13
12
4
3
3
0
1976
Percentage
22.0
13.7
20.2
4.8
4.8
9.5
4.2
7.7
7.1
2.4
1.8
1.8
0
+Day classifications were obtained from Morris H. McCutchan,
 Project Leader, Fire Meteorology Project, Pacific Southwest
 Forest and Range Experiment Station, Forest Service, U.S.D.A.,
 Riverside, CA.

*The 12 remaining possible combinations occurred only 3 percent
 of the time or less and are omitted.

**The dose (ppm-hrs) on the second day of the most common transi-
  tional combinations is indicative of the pollution potential,
  for example, the combined data from 1974 and 1975 (Miller, et
  al., 1977) showed the following doses associated with these
  combinations:  3-3 = 2.09, 3-4 = 1.90, 2-2 = 1.88, 4-4 = 1.76,
  5-5 =0.78 and 1-1 =0.38 ppm-hr.
                                    51

-------
         V)
         cc
         <
         CO
         UJ
         cc
         I


         2
         QL
         Q.

         UJ
         to
         O
         o
         z
         <
         o

         X
         o
         o


         
-------
Cn
                                         CAMP   ANGELUS  —  1975  NEEDLES

                                                                      (INJURY SCORES 0-18)      Y-JSIN(I6.4-I-0.2IXJ]
                                                              <£
                                                              £
                                                                80-
                                                              Ul
                                                              Ul
                                                              z 40-
         t- 20-
         o
         z
         -I
         _  0-
                                                              Ul
                                                              ^60-
                                                              >
                                                              - 40-
                                                              x

                                                              i
                                                              u 20-
                                                              o
                                                              z
                                                              u>
                       T—T—T^—n—r
                       ISO   180   260   220  240   260
                                                                                   $   *
                                                                     I      p    |   I       p       T-p
                                                                    JUNE     JULY     AUG     SEPT    OCT
                 (INJURY SCORES 19-21)      Y-|siW39.3+0.30x]2
                                                                    JUNE '     'JULY  '   AUG     SEPT  '  OCT
JULIAN DATES

  1976
~n—n—i   i  i"  "ir—i—n
  160   180   200  220   24"  "~   -J
                                 r"-2-^
                                                                                           0  260  280
                                                                                                              -7?
                                                                                                              -2ui
                                                                                                               ui
                                                                                                              02
                                                                                                               Ul
                                                                                                              Oz
           Figure 11.   Comparison  of injury to the  1975 needle  whorl  with  potential  transpiration
                         and  the ratio of  actual over potential transpiration at  Camp  Angelus in
                         1976.

-------
         2.4
         2.2
         2.0
          1.8
          1.6
          1.4
        •B
        Q- 1.2
        a.
        u
        
-------
a high dose period.  A scatter diagram  (not  shown)  relating the weekly
oxidant dose (ppm-hr) to the average  of  the  daily values  of Ta/Tp for the
same week for 18 weeks at Camp Angelus  in  1976  did  not  suggest a relation-
ship between two variables over a  range  of 0.14 to  0.35 for Ta/Tp.   An
adjusted ozone dose for the whole  season might  be calculated by multiplying
daily Ta/Tp times the daily dose but  it  appears that  a  wider range  for the
Ta/Tp variable will be needed to more carefully examine the modifying effect
that it may have on the development of needle symptoms.  MuKammal (1965)
found that by multiplying the daylight  oxidant  dosage by  the coefficient  of
evaporation he could reduce the scatter  of data points  describing dose-
injury response of tobacco.

     At this time, we have calculated seasonal  transpiration at only two
of the six plots where data is available (CA, CAO,  HB,  TUN2,  DWA and
DL).  It will be helpful to find the  seasonal ranges  of Ta/Tp at other
plots which are expected to be both drier  and more  moist  than CA and DL.
Since three of the plots are dominated by  ponderosa pine  (CA,  TUN2  and
DWA) and the remaining three have  Jeffrey  pine  it may be  possible to
evaluate differences in transpirational  behavior that may be  related to
species.  When the data from 1977  is  considered in  addition,  we have pro-
cessed only one sixth of the available data.  An improvement  in the preci-
sion of the regression formula relating  predawn xylem water potential and
minimum daily stomatal resistance  will be  incorporated  as soon as possible.

     The relationship between soil moisture  content and transpiration
(Ta) can be determined, as proposed in  the systems  modeling plan by Kickert
in Miller, et al (1977), in cooperation  with Rod Arkley using the relation-
ship in the model described by Thompson  and Hinckly (1977).

     Development of ozone injury to foliage of  ponderosa  and  Jeffrey
pines, 1976 and 1977—The preliminary results showing the relationships
between Tp, Ta/Tp, and cumulative  oxidant  dose  (weekly  and  seasonally)  (and
development of oxidant injury symptoms  to  ponderosa pine  needles) is shown
in both Figure 11 and 13 for Camp  Angelus  in 1976.  The increase in the
percentage of needle length exhibiting injury of any  intensity (Table
11) and the increase of injury severity  on the  same needles versus  Julian
date show differing responses depending  on the  oxidant  injury scores of
ponderosa pines at Camp Angeles.   Needle injury was barely  detectable on
Jeffrey pines at Deer Lick in 1976; no data are presented in  graphic form.
A regression line was calculated for  "injured length" at  CA because it  is
derived from a continuous metric scale,  but not for the needle injury index
because it is the product of a subjective  judgement for which scale units
cannot be assumed to be of uniform size  at all  points of  the  scale.   It
is evident however, that injury intensity  increased moderately for  the  more
sensitive group of trees (0-18) compared to the less  sensitive group (19-21)
The 1975 needles began the 1976 season with some injury accumulated in
1975. The new needles produced in  1976  (not shown in  Figure 11)  did not show
injury until the September observation and then about 12  percent of the
needle length exhibited injury.  The  increase of injury to  the 1975 needles
in 1976 (1 year old) and 1977 (2 years old) is  shown  in Figure 13.   From
1976 to 1977, the percent of the needle  length  with injury  (chlorotic
mottle) of 1975 needles increased  with very slight  recovery over the 1976-77

                                    55

-------
Oi


100-

80-
H
§60-
i
J40-
Q
U
£
o20-
z
I-
(9
Sn


gao-

K
360-
Z
3
g40-
£
U.
°20-
o
z
3 0-



CAMP ANGELUS — 1975 NEEDLES
(INJURY SCORES 0-18) Y- |?IN( 16.4 4- 0.21 X TJ *
. . • • 100-
. :__^_— -
: !^--*^* — ^"""^
1 ' ' ' » t
• t T
4

• *
t +
;*»>**



_ 80-
.? £
-*- D:
-6-3 60-
% -
Z UJ
-4> o 40-
o: uj
,3 UJ
-3-j z
-2 g *«0-
"! S 0
111 "^™
/N -5 UJ XV
JUNE JULY AUG SEPT OCT
(INJURY SCORES 19-21 ) Y -[siN (-39.3+ 030 xT|2
80-
p
~
5 60-
i
. • ^s' w
' .s"^ D 40-
.s^ UJ
•^^•' I i z°-
r^'^ . '
• •


*,**** -i —
«• "J r:
•s§ |
-4>- " 0-
ir J
ll
"' Ul
Ul
_rt Z
(INJURY SCORES 0-18) Y« |JN(48.7 + -I3XT| 2
t ftit
v 	 -; 	 1 » r
! :
* * * 1 *
*
4 4
* i t * +
4 4
4 4 * 4 T
t t




JUNE JULY ' AUG' 'S'EPT ' OCT
(INJURY SCORES 19-21) • • Y- QlN(-2l +^8xTj2
•
.

.
^^
^s^ .
* — . '.
' ^^^
r ; ' :
; 4 4 4 4

. « t t {
* $ i * t






7
±
6x
•S
-4>-
or
31
_i
iS
UJ
oz









X

tr
_i
~ UJ
UJ
*^^
— i — i n i ' i M^ n " < i i« i ' i 1 1 i i i
160 180 200 220 240 260 280 160 180 200 220 240 260 280
JULIAN DATES 1976 JULIAN DATES 1977
          Figure  13.   Changes  of  1975  needle  whorl injury in 1976 and 1977 at Camp Angelus.

-------
winter for trees in the oxidant  sensitive  (0-18)  score range whereas the
less sensitive group of trees  (19-21) may  have recovered slightly over
winter.  The regression lines  for needle length with injury show increases
of injury in 1977; symptom intensity  (needle  injury index)  increased similar-
ly.  Needle abscision was not  observed  until  more than 90 percent of the
needle length showed injury on trees  in the sensitive group (0-18).

     As illustrated in Figure  13, the fate of 1975 needles  can be followed
year after year.  The same is  true  for  1976 and 1977 needles.   Other
information that can be derived  from  the INJBIWK  data set includes  the
rate of needle elongation of the new needles  each year;  total  length
of needles; the longevity of needles; and  the annual change in green or
mottled needle surface area relative  to tree  sensitivity.

     The biweekly changes in injury to  white  fir  are not reported because
the lower injury response compared  to ponderosa and Jeffrey pine has
resulted in little change of needle condition or  needle  retention over
the single 1976 observation year.   Black oak  injury data and the monthly
observations of injury to all  four  species has been keypunched and  given
preliminary processing.  These results  will be reported  in  conjunction
with the 1977 injury data.

     Progress in development of  a single season ozone dose, foliage  injury
model—Transpiration, oxidant  dose  and  foliage injury data  sets for  CA and
DL have been completed for 1976  and 1977.  A  preliminary analysis of the
relationships between injury to  1975 needles  from the moderate injury (19-21
injury score) group of ponderosa pines  at  Camp Angeles in 1976 (Figure 13,
lower left) and the cumulative oxidant  dose alone or the cumulative  dose
multiplied by the ratio of actual to potential transpiration (Ta/Tp) is
shown in Figure 14.  The biweekly mean  of  daily Ta/Tp was multiplied times
the dose increment for the same  period. The  effects of  the multiplication
were to "weight" each biweekly dose increment and to change the total of the
cumulative seasonal dose from  160 to  38 ppm-hr.  It is evident from  the
small range of values for Ta/Tp  at  Camp Angeles in 1976  (Figure 11,  lower
left) that the ratio was nearly  a constant until  the very end  of the season
(Julian Date 250 to 265); it did not  "weight" the dose in any  significant
way.  In this example, either  dose  injury  relationship in Figure 14  would be
acceptable.  The analyses that will be  done with  data from  other plots may
test the utility of the transpiration ratio in describing dose if the range
of the ratio is larger.  The cumulative dose  documented  on  a daily or bi-
weekly basis seems to be an appropriate unit  for  expressing dose. The
geometric mean for biweekly increments  of  the whole season  does not  seem to
provide a scale with fine enough resolution but it should be tested.

     Existing dose response models  (Larsen and Heck, 1976)  deal with single
concentrations for periods not exceeding 8 hr. In chronic  exposures, the
dose pattern is usually characterized by a series of high concentration
episodes occurring at random intervals  and linked by consecutive days with
lower or moderate concentrations of oxidant;  this generates new questions.
What sequences or seasonal patterns of  high concentration episodes  are most
injurious?  It is assumed that periods  of  high actual transpiration  would
contribute to more rapid injury  development because of a larger ozone flux

                                    57

-------
  to
    40
  OL

  I30

  (T

  | 2O

  UJ

  o  I0

  o
  X

  £  o
O UJ
£ Z 40
UJ <
o5

ct gj 3O

  z
  o
  o.
  a: 20
      10
  0>
CAMP ANGELUS -1976
  FIVE TREES,  MODERATE  INJURY  (19-21)
                                                             267
         141
                    JULIAN DATE
                       197
                                            Y=CSIN( 1.5+ i.
             6     10     14     18    22    26     3O    34
          BIWEEKLY CUMULATIVE OXIDANT DOSE (PPM-HR) X
                                      BIWEEKLY AVE. TQ/T,
                                                      38
         141
                    JULIAN DATE
                        197
                                                             267
                                                  (2.8 + 0.22 X)]]2
    20    40     60    80    100    120    140    I6O
 CUMULATIVE OXIDANT DOSE,   PPM-HR  (BIWEEKLY)
Figure  14.  Changes in the percent of the total needle length of
           selected ponderosa pines with chlorotic mottle in rela-
           tion to two measures of cumulative oxidant dose.
                                 58

-------
to foliage, but is there a point  on  the  lower end  of  the available soil
moisture curve where the lowered  transpiration rate (that may limit ozone
flux to needles) increases to  drought  stress  proportions?  Drought stress
would mainly accelerate abscision of the needles already injured by ozone.

     The worst case conditions  that may  cause the  greatest injury in a
single season are hypothesized  as follows:  Abundant  soil moisture at
the beginning of the season followed by  hot weather in June and July which
would result in numerous closely  spaced  episodes of high ozone concentra-
tions, and a continuing high potential transpiration  throughout August
and September.  The Ta/Tp ratio would  decrease gradually until late July
and oscillate only slightly at  a  low level for the remainder of the summer
season.  The drought stress induced  in August and  September may decrease
carbon fixation in the remaining  needles in addition  to accelerating needle
abscision.

     Some very important results  have  been obtained by Coyne and Bingham
(1978) that focus on changes in stomatal behavior  of  pole-size ponderosa
pines in the San Bernardino mountain near Camp Paivika and of container-
grown trees under greenhouse conditions. The most important variables
measured in the study included:  predawn and  daylight xylem water potential
and stomatal conductance in relation to  amount of  chronic injury to pon-
derosa pines under field conditions.   Some of these results run counter
to the hypothesis that has been developed in  our report,  namely, that
injury during a given season is mediated mainly by stomatal resistance,
i.e., high predawn xylem water  potential would result in a lower stomatal
resistance (increased conductance), hence increased pollutant flux to needle
tissue.  Their data shows that  as needles on  individual trees showing
increasing amounts of chronic  injury (lower oxidant injury scores) become
more injured as the summer season progresses, their stomatal resistance also
increases.  This event is accompanied  by drastic declines in apparent
photosynthesis.  In summary:   1)  Increased needle  injury is associated with
increased stomatal conductance  and decreased  stomatal resistance and de-
creased water loss; 2) The relationship  between xylem water potential
(container grown trees) and stomatal conductance is basically sigmoid for
both uninjured and ozone-injured  trees but as xylem water potential in-
creases over a arange from 0 to 25 bars  the steep  part of the curve is
between -10 and -14 for uninjured trees  and -14 to -16 for ozone injured
trees; 3) Increased stomatal resistance  would decrease pollutant flux to
needles. 4) Because other internal variables  may also influence stomatal
behavior, e.g. higher levels of abscissic acid, "the  influence of water
stress can more readily be described as  an operational limit beyond which
stomatal aperture can not increase."

     The small amount of data  that we  have described  in this report does
not dispute the results of studies done  by Coyne and  Bingham (1978) and we
will hasten to complete the analysis of  the remaining data so that our
research can be directed towards  the most crucial  questions pertaining to
the description of the chronic  ozone dose-injury relationship for ponderosa
and Jeffrey pine.

     Another variable that may  have an important influence on injury develop-

                                    59

-------
ment is needle phenology.  Earlier studies with container grown ponderosa
pines fumigated with ozone suggested that the injury from the  same  concen-
tration, 0.45 ppm, increased as the summer progressed.  A large number  of
trees were held in a filtered air greenhouse and at 4 week  intervals  (during
3 summers) a new group of 40 trees was fumigated.  Smaller  doses were
required to cause equivalent injury to current and one-year-old needles as
the summer season advanced (mid-June to mid-September).  These trees were
not water stressed (Miller, 1973).  If this observation holds  true  under
field conditions, it would suggest that higher dose episodes in late August
and September may result in increased injury under the regulation of unknown
and unspecified controls at the physiological level; these  controls may be
independent of stomatal behavior.

Annual  Shoot Growth of Ponderosa and Jeffrey Pine Saplings  From 1967-1976—

     Internode lengths of terminal shoots and oxidant injury score—The
link that may be the most practical in coupling oxidant injury to growth
of smaller trees is foliage surface area retained.  For example, Kozlowski
and Winget (1964) showed by removal of various proportions  of  needles from
Pinus resinosa that the old (all except the current year whorl) needles
provided the food reserve that accounted for four-fifths or more of all shoot
growth.  The combined reserves in the branches, main stem and  roots accounted
for less than 15 percent of shoot growth.

     The oxidant injury score used for both sapling and sawtimber sized
trees is comprised of the number of needle whorls retained  on  the main  stem
and a mid-crown branch for saplings and for the upper and lower crown in
larger  trees.  It is presently a crude index of foliage retained but may be
expanded to foliage surface area estimates in combination with other parame-
ters.   In Figure 15, the preliminary linear regression lines show the average
of the  annual shoot growth (internode lengths) between 1967 and 1976.   Most
of this period was characterized by high seasonal oxidant doses (Fig = 9).
The best correlation between injury score determined in 1975 and growth was
at those plot locations experiencing the highest doses namely  CP, SF and BP.
The maximum value for each line may be an approximate measure  of the site
quality where the 25 to 40 trees were growing.  The ponderosa  pine  plots had
shown greater height growth than the Jeffrey pine plots (HV, CAO, HB and BL).
This difference is believed to be associated with lower rainfall and lower
mean temperatures at the Jeffrey pine plots compared to the ponderosa pine
plots.  The regression lines in Figure 15 should be regarded as an  approxima-
tion because the independent variable (oxidant injury score) is not based on
continuous variable data but is a composite of discontinuous variables,
ranked variables and attributes.  One of the most important problems  that we
face is to find acceptable ways to use selected variables in the index  for
purposes of statistical analyses.

     At the ponderosa pine plots experiencing the highest doses of  oxidant
(Miller, et al. 1977) the sample populations could be divided  into  several
injury score groups which also exhibited growth differences commensurate
with their injury category (Fig. 16).  The year to year variability in  growth
of all injury score groups is definitely associated with rainfall amount but
the differences between groups is mostly attributable to seasonal oxidant

                                    60

-------
      CO
       i
      C0200
      C/5
      O
      o
      X


      §100
      o
      UJ
      X
         50
      <

      <
                         • COEFFICIENT OF CORRELATION
                                         .66
CA(PP)r-.l4

  BP(PP) r-.6l

  T-2 (PP) r«.29

  SF (PP)  r=.56
  HV (PJ) r«.23

  CAO (PJ) r-.3ft



  HB (PJ) 1T-.2B
                                                    BL (PJ)  r-.28
                       10          20          30
                        OXIDANT  INJURY SCORE
           40
Figure  15.  Average height growth of ponderosa and Jeffrey pine sap-
           lings  at plots experiencing different levels  of chronic
           oxidant injury.
                                 61

-------
         330-
         320-
         310-
         300-
         290-
         28O-
         270-
         26O-
         250-
        2* 240-
        Z 230-
        J 220-
        o
        z 2IO-
        w
        J 2OO-
z 170-
  160
  ISO
  I40H
  130
  120
  110 H
                 ANNUAL  TERMINAL GROWTH
                 P. PONDEROSA - CAMP PAIVIKA
                              INJURY SCORE
                                 (15-21)—8 TREES
                                 (9-14) —IOTREES
                                         (0—8)	8 TREES
          100   76 75 74 73 72  71 70 69  68 67
                          YEAR
Figure 16.  Height growth of ponderosa pine saplings in three injury
          categories  at Camp Paivika between 1967 and 1976.
                               62

-------
doses which increased towards  the  end  of  this  period  (see the three year
moving average in Figure 9 from  the  nearby  Sky Forest station).   In a pre-
vious study, saplings were placed  in carbon filtered  air  and  even the most
severely injured trees improved  but  there was  a 2-3 year  lag  before the
improvement in terminal shoot  growth was  evident  (Miller,  et  al.  1977).  Lags
may make it difficult to recognize the effects of  a single year with a low
oxidant dose and non-limiting  soil moisture.   Additional  analysis of the
SAPGO data will be done in order to  estimate the  effects  of the amount and
the timing of oxidant dose and precipitation relative to  annual shoot growth.

Annual Changes in Oxidant Injury Scores of  Ponderosa  and  Jeffrey  Pines
and Tree Mortality at Major Vegetation Plots—

     Oxidant injury scores—The  changes in  mean scores for each plot in
1976 included higher scores (improved  tree  condition) at  most of  the 18
plots. In Table 13, the changes  at individual  plots are shown between 1973
and 1976. In 1976, 10 plot scores  were significantly  higher (p =  .05), 2 were
significantly lower and the remaining  6 plots  did  not change  significantly.
The trend to improve tree condition  matches the decreasing seasonal oxidant
doses in 1975 and 1976 (Fig. 9).

     Tree mortality—Tree deaths related  to chronic oxidant injury decreased
also from 26 in 1974 to 19 and 11  in 1975 and  1976 respectively.   The
4-year accumulated mortality in  Table  13  is expressed as  percentage of
the original number of ponderosa and Jeffrey pines on the 30  m wide plots.
There are several exceptions where tagged trees fall  outside  the  30 m specifi-
cation because in some cases the stands contained  fewer trees >30 cm dbh;
in such cases trees were selected  at measured  distances outside the plot (to
avoid making the plot excessively  long) to  achieve the purpose of having 50
ponderosa or Jeffrey pines ^30 cm  dbh  in  each  plot-
                                     63

-------
TABLE 13.  TRENDS IN PONDEROSA AND JEFFREY PINE CHRONIC INJURY SCORES
           AND TREE DEATH AT EIGHTEEN PERMANENT PLOTS.
Plot
Name
COO
BP
CP
SF
DWA
UCC
TUN 2
DL
GVC
BL
NEGV
SC
HV
SCR
CA
BF(PP)
BF(JP)
CAO
HB
Score
1-8
9-14
15-21
1973
Injury Number
Score Dead
15.1
16.8
17.0
13.4
20.2
15.5
19.2
	
21.7
29.4
33.1
41.3
46.3
12.4
25.6
22.5
21.3
21.7
44.0
0
2
0
1
0
0
0
-
0
0
0
0
0
0
1
5
1
1
0
YEARS Accumulated
1974 1975 1976 Mortality
Injury Number Injury Number Injury Number 1973-1976
Score Dead Score Dead Score Dead Percent
12
16
16
13
16
15
17
18
20
31
32
47
47
11
17
19
20
24
39
.9
.5
.7
.8
.5
.9
.0
.6
.1
.8
.1
.3
.7
.7
.4
.3
.4
.6
.6
0
2
2
1
0
1
1
0
0
0
0
0
0
0
2
5
2
9
1
Interpretation:
= Very severe
-
Severe



= Moderate
12.5
14.5
14.7
15.3
16.0
16.3
17.0
19.2
23.1
30.5
40.6
44.5
44.6
12.5
17.9
20.5
21 = 5
25.7
33.9
22-28
29-35
36 &
4
1
0
2
2
0
1
1
0
1
1
0
0
1
1
2
' 1
1
0

above
13
15
17
15
18
16
18
20
21
34
36
46
50
16
17
23
23
33
38
=
=
.8 0
.5 0
.9++ 1
.0 0
.2++ 0
.0 0
.64"1" 0
.7 0
.0 0
.5* 2
.0 0
.0 0
.5^ 0
.7++ 1
.5 1
.0*"*" 6
• S4* 0
.2 0
.2^ 0
Slight
Very slight
No visible
6.
6.
3.
3.
2.
1.
2.
1.
0
2.
1.
0
0
4.
7.
10-
6.
8.
0.

injury
6
9
1
3
4
5
7
5

1
5


0
3
7
9
9
8


  Significant Difference (p = 0.05) between 1975 and 1976
                                    64

-------
EFFECT OF PHOTOCHEMICAL OXIDANTS ON TREE GROWTH  IN  THE  SAN  BERNARDINO
NATIONAL FOREST

Introduction

History of Oxidant Impact on Tree Growth—

     Symptoms of air pollution injury were  first  observed in  the  San
Bernardino mountains on ponderosa pine  (Pinus ponderosa Laws.)  in 1953
(Asher 1956), although the direct link  between cause  and effect was not
discovered until early 1960.  Symptoms  included  loss  of all but the current
season's needles, reduction in number and size of needles,  and  yellow
mottling of the needles.  Miller et al. (1963) confirmed that these injury
symptoms were produced by ozone.  McBride et_ al.  (1975) and Parmeter et  al.
(1962) demonstrated significant reductions  in both  radial and height growth
increment of ponderosa pine as a result of  oxidant  exposure.  Additional  data
was necessary to more completely document the impact  of oxidants  on tree
growth and to properly calibrate a stand growth model for the SBNF.
Research Objective

     1)  Document the impact of photochemical  oxidant  injury on  radial
growth increment of the major tree species occurring in  the SBNF.

     2)  To examine the mortality rates and  radial  growth patterns  of
forest trees in the SBNF with the aid of a stand development model.
Materials and Methods

Selection of Study Plots—

     Six vegetation plots were selected for examination.  A smog  injury
rating system, developed by Miller  (1973), was applied to the  trees on
each plot.  This system involves examination of a  tree with binoculars
and scoring it as to needle retention, needle condition, needle length
and branch mortality.  The six plots were  then classified according to
the most common injury scores.  Two plots  contained a majority of  trees
rated as severely injured, COO and BP.  Two plots  were rated as moderately
damaged, GVC and DL; and two plots had very slight or no visible  smog
damage, HV and NEGV.  The plots varied in  length from 100 to 300 m.

Sampling Procedure—

     During the summer of 1976 each conifer greater than 10 cm. dbh

                                    65

-------
(diameter at breast height) on the six study plots was cored with  an incre-
ment borer.  The core was taken at a height of  1.5 m on a  randomly chosen
side of the tree trunk.  The cores were returned to the laboratory,  dried at
70°C for 24 hours, mounted on boards and sanded, aged and  the growth rings
measured to the nearest 0.01 mm back to and including the  year  1920.  Each
tree was classified as to the soil type on which it occurred.

Statistical Analysis—

     Growth data from trees within and between  stands was  quite variable.
To reduce this variation as much as possible it was felt that the  trees
should be stratified before analysis.  Trees were divided  into groups
delineated by species, plot, age and soil type.  Within each group  the
growth data for the past 30 years (excluding 1976) was considered  for
analysis (i.e., 1946-1975).  This period was divided into  ten year  growth
periods; 1946-1955, 1956-1965, 1966-1975.  A period of this length  (30
years) should be adequate to detect growth trends as influenced by  photo-
chemical oxidants which have steadily been increasing in recent years
(Corn et al. 1975).

     Examining the growth rings on a tree core  is, in actuality, sampling
the growth of a given tree through time.  This  situation lends itself
to examination using a repeated measure analysis of variance (ANOVA) design
(Sokal and Rohlf, 1969).

     No statistical tests were performed to compare growth between plots.
Since so many variables among them are different it was felt that  they
were not quantitatively comparable.  However, it was felt  that, quali-
tatively, the growth trends for two similarly stratified tree groups on
different plots could be compared.

Results and Discussion

Impact of Oxidant on Radial Increment—

     Radial growth of forest trees can be affected by many different phe-
nomena.  In order to examine the effects of oxidants on tree growth  it
is necessary to eliminate other variables that  affect tree growth  to
ensure that there are no confounding interactions among these variables.
One important variable influencing tree growth  is precipitation.   One
weather station in the vicinity of the SBNF was selected as being  most
representative of the precipitation in the SBNF (Squirrel  Inn #2).   The
precipitation records for the years in question are presented in Table
14.  A one way analysis of variance was performed on these data to  detnnine
if any significant differences in precipitation existed among the  three
growth periods and none were found.  Therefore, significant differences  in
radial growth among the three growth periods cannot be directly attributed
to rainfall.

     It was necessary to insure that tree growth was not correlated with
precipitation during the years encompassed by this study.  Mean annual
precipitation, recorded at 3 weather stations in close proximity of  the

                                    66

-------
TABLE 14.  ANNUAL PRECIPITATION  (cm) AT  SQUIRREL  INN  #2 WEATHER STATION
           AND AN ANALYSIS OF VARIANCE AMONG THE  TEN  YEAR  INTERVALS  THAT
           CORRESPOND TO THE TEN YEAR GROWTH PERIODS.
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
n
X
Sx
Source
Between
Within
Total
35.89
83.29
109.32
55.85
99.26
147.32
36.68
120.19
78.97
63.35
10
83.01
34.61
d.f
Periods 2
Periods 24
26
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965



SS
1599.915
43009.607
44609.522
125-78
113.46
62.26
73.74
45.16
87.05
86.59
76.07
169.88
90.27
10
93.03
33.73
MS
799.958
1792.067

1966 151.33
1967 49.05
1968 200.25
1969 91.59
1970 88.54
1971 35.13
1972 102.13



7
102,57
52.98
F
.446 ns


Reject H0 if F > F2;24/.05 = 3'40
                                    67

-------
specific  study plots, is presented  in Table  15. Mean  annual  radial growth
from  several  trees on each plot was  used  in  a  regression analysis with mean
annual precipitation from the  appropriate weather  station.   No significant
correlations  were found between tree growth  and precipitation.

      Plot size, number of individual trees per plot,  species composition,
density,  and  basal area for  each plot are described in  Table 16.   Data for
the Deer  Lick plot is incomplete.

      COO  and  BP, the two plots that  have  a severe  smog  injury rating,
were  the  only two plots that showed  significant changes in radial growth
trends (Table 17, Figs. 17 to  22).   Growth in  all  of  the tree groups
in COO shows  a significant reduction (Fig; 17) but this trend was not  as
distinct  in the BP plot (Fig.  18).

      In the first growth period (1945-55), the yearly radial growth incre-
ment  for  trees in COO was much greater  compared to growth in the  other
plots, and was still slightly  greater in  most  groups  even after drastic
reductions by 1975 (Table 17,  Figs.  17-22).  This  could be explained,
in part,  by the predominately  younger age groups in the COO  plot, since
younger trees tend to grow faster than  older trees.   The site class of COO
may also  be higher.  Trees in  the BP plot were older  than most of the
trees in  COO  and showed a more variable growth response.  Older trees  may
not react as  strongly to oxidant exposure as young trees.  The youngest
group of  trees in BP were the  fastest growing  in 1945-56 and showed the
strongest decrease in growth through time.   Faster growing trees  may also
react more to oxidant exposure than  slower growing trees.  Since  the young-
est trees were usually the faster growing individuals it is  impossible to
determine which factor may be  responsible for  this strong reaction to
oxidant exposure.
TABLE  15.  RAINFALL DATA FOR THE  SIX  STUDY  PLOTS  (MEAN ANNUAL  PRECIPI-
           TATION cm)
                                         1946-55      1956-65      1966-75
Green Valley Weather Station
 for COO, GVC, DL, NEGV

Panorama Point Weather
  Station for BP

Big Bear Dam Weather
  Station for HV
83.84
           84.28
           61.74
86 = 59
            86.21
            80.92
97,94
+No available data.
                                     68

-------
     White fir trees on COO are affected just as  severely  as  the  pine
(Fig. 17) while incense cedar on BP did not seem  to be  responding to
oxidant exposure (Fig. 18).

     The two plots rated as moderately damaged  (GVC and DL) showed no  growth
response to oxidant exposure (Table 17, Figs. 19  and  20).  The presence of
visible foliar damage on these plots may indicate a threshold phenomenon.
Visible injury, or oxidant exposure, may need to  reach  a certain  level
before radial growth increment is affected.  Another  explanation  may be that
oxidant impact on radial growth may not be expressed  in the bole  of the tree
at 1.5m but may cause measurable increment loss  within the crown.  Williams
(1967) demonstrated that spruce budworm defoliation reduced radial increment
of grand fir, Douglas-fir, and Englemann spruce to a  greater  extent within
the crowns than near the ground. The same simulation  may exist here.

     The two plots with slight or no visible smog injury (HV  and  NEGV)
showed some significant growth trends (Table 17,  Figs.  21  and 22).  The
Jeffrey pine over 200 years old in NEGV show a  significant but small in-
crease in growth (Fig = 21).  This defines no signicant  trend  in growth.
Jeffrey pine on the HV plot from 41-50 and 51-60  years  old showed a signifi-
cant but small decline in growth.  This reduction was probably due to  the
natural, gradual decline in radial growth increment at breast height that
occurs as a tree ages (Duff and Nolan, 1953)•

     The younger trees of each species are growing faster  than the older
trees of the same species on any given plot.  White fir is the fastest
growing tree on each plot when compared to other  species of the same age.
It appears to be better adapted for faster growth than  the other  tree
species in the San Bernardino mountains but is just as  susceptible to
oxidant injury as ponderosa pine.

     Not all species occurred on each plot so it  is not possible  to make
statements on their susceptibility to specific  levels of oxidant  injury.

Stand Development Model—

     The data collected from the six study plots  has  been  sent to  Dr.
A. R. Stage, U. S. Forest Service, Intermountain  Forest and Range  Experiment
Station, Moscow, Idaho to be run through his Stand Prognosis Model.  No
results have been obtained from this effort to date.  Work on  this  aspect of
the project will continue.
                                    69

-------
TABLE 16.  MENSURATIONAL DESCRIPTIONS OF THE SIX SAMPLE PLOTS.
                          PP    SP
1C
WF
JP
                    BO
QW
                                                                        CL
COO No. of Trees/
Plot
170m Spp.
Composi-
tion(%)
Severe Density
Damage (#/ha)
Basal Area
(m2/ha)
BP No. of Trees/
Plot
100m Spp.
Composi-
tion(%)
Severe Density
damage (#/ha)
Basal Area
(m2/ha)
GVC No. of Trees/
Plot
300m Spp.
Compos i-
tion(%)
Moderate Density
damage (#/ha)
Basal Area
(m2/ha)
NEGV No. of Trees/
Plot
180m Spp.
Composi-
tion(%)
No Density
damage (#/ha)
Basal Area
(m2/ha)
46
37
90.2
14.52.
71
43
236.7
28.10








1
1
2.0
.72




11
4
12.2
2.93




3
2
5.9
1.53
30
22
100.0
11.86
10
4
11.1
1.63




1
20 |
1
1
16 1
1
39.2 |
6.00 |
1
1
1
1
1
1
1
1
I
1
62 |39
1
1
26 |16
1
68.9 143.3
4.94 |12.20
1
1
8 |65
1
1
10 |89
1
14.8 |120.4
1.05 |35.00
1
1
39
31
76.5
4.80
31
23
103.3
4.08
82
35
91.1
4.70








1
1
3.3
0.04
1
1
1.1
0.02



1
1
1















                                    70

-------
TABLE 16.  CONTINUED.

HV No. of Trees/
Plot
290m Spp.
Compos i-
tion(%)
No Density
damage (#/ha)
Basal Area
(m2/ha)
DL No. of Trees/
Plot
120m Spp.
Compos i-
tion(%)
Moderate Density
damage (#/ha)
Basal Area
(m2/ha)
Table Legend
PP Ponderosa pin«
SP Sugar pine
1C Incense Cedar
WF White fir

PP








A




SP




22
15







1C








JI
B(
QV
CI

WF
24
11
27.6
3.4
32
21
—

3
)
1
J

JP
168
83
193.1
22.07
55
37
^••B
•M^B

Jeffre
Black
Quercv
Cercoc

BO
4
1
4.6
.21
40
27
^•^A

;y pine
Oak
is wisl:
:arpus ]

QW










izenii
LedifoJ

CL
1
5 \
1
1
2 I
1
5.7 |
0.13|
1
1
1
1
1
1
1
1
1
1
+



.ius

Note:   From McBride,  J.  R.   1974.   Annual  report  of  the  vegetation  sub-
committee  for  the  fiscal year  ending.   In:   Taylor,  0. C.   Oxidant  air
pollution  effects  on  a western coniferous  forest  ecosystem.   Task D.
Statewide  Air  Pollution  Research Center, Riverside,  California.
                                     71

-------
TABLE 17.  AGE GROUPS, SOIL TYPES, MEAN ANNUAL RADIAL  GROWTH INCREMENT
           (mm) AND STANDARD DEVIATION FOR EACH GROWTH PERIOD,  F VALUES FROM
           ANOVA, AND EXPLAINED VARIANCE  (co2) FOR  EACH STRATIFIED GROUP OF
           SAMPLE TREES ON EACH PLOT.
COO Age Soil Type
PP
PP
PP
PP
WF
BP
PP
PP
PP
1C
51-60
51-60
41-50
41-50
41-118

61-100
41-60
61-100
61-100
PxClCm
PxClDm
PxClEm
PxClDm
PxClEm

PHclCm
PHclCm
PHclDm
Phc IDm
n
9
9
9
9
7

15
8
38
15
1946-55
3.26 +
2.89 +
3.39 +
3.76 +
3.87 +

1.25 +
1.96 +
1.51 +
1.78 +
.92
.54
1.55
1.29
.75

.65
.58
.74
.85
1956-65
2.13 +
1.96 +
2.36 +
2.25 +
3.24 +

1.52 +
1.21 +
1.78 +
1.58 +
.65
.38
1.14
.97
.98

.65
.68
.67
.66
1966-75
1.33 +
1.64 +
1.48 +
.78 +
2.47 +

1.75 +
.56 +
1.17 +
1.87 +
.76
,90
1.00
.33
.98

.62
.21
.72
.81
F
23.24
23.77
16.22
55.87
6.92

n.s.
27.00
6.65
n.s.
CO2
.510
.401
.280
.620
.286

	
.542
.047
	
GVC
JP
WF
WF
SP
1C
DL
JP
SP
61-200
61-200
51-60
TsEm
ExsaDm
TsEm and
ExsaDm
41 or older
over 31

101-200
data
61-200
data


no
available
no
available
14
14
9
10
7

33
17
.87 +
1.20 +
1.80 +
1.28 +
1.85 +

-55 +
.88 +
.41
-53
.85
.58
.81

.33
.37
.76 +
1.07 +
1.90 +
1.29 +
1.23 +

.52 +
.89 +
.34
.53
.76
.73
.58

.34
.34
.69 +
1.33 +
1.92 +
1.55 +
1.38 +

.44 +
1.00 +
.50
.89
.81
.81
.82

.31
.39
n.s.
n.s.
n.s.
n.s.
5.81

10.94
3.39
	
	
	
	
.112

.022
-022
WF 61-200    no       19
       data available
 NEGV
JP over 200 TsEf      17
JP 101-200  TsEf      20
1.09 +  .40   1.17 +  .41  1.17 +  .38   n.s.
 .45+  .16
 .77 +  .49
.45+  .22
.79 +  .42
.56 +  .78
.77 +  .34
4.37  .046
n.s.   	
                                     72

-------
TABLE 17.  (continued)
HV Age Soil Type n
JP
JP
JP
JP
WF
over 200
41-50
51-60
61-100
61-100
TAflDm
TAflCm
TAflCm
TAflBm
TAflBm
9
34
22
17
12
1946-55
.60 +
1.74 +
1.53 +
.75 +
1.00 +
.22
.68
.63
.39
.56
1956-65
.43 +
1.24 +
1.04 +
.64 +
.92 +
.14
.55
.53
.38
.46
1966-75
.43 +
1.21 +
1.11 +
.65 +
1.04 +
.18
.68
.64
.42
.64
F
7.48
43.43
26.77
n.s.
n.s.
C,2
.166
.126
.114
	
	
Table Legend:




        Code




        PxClCm




        PxClDm




        PxClEm




        PHclCm




        PHclDm




        TsEm




        ExsaDm




        TsEf




        TAflDm




        FAfICm




        TAflBm
            Soil Type




Pachic Xerumbrept on 5-10% slopes




Pachic Xerumbrept on 19-23% slopes




Typic Xerorthent




Pachic Ultic Haploxerolls on 10-15% slopes




Pachic Ultic Haploxerolls on 15-30% slopes




Typic Xeropsamments




Entic Xerorthents




Typic Xeropsamments




Typic Argixerolls on 15-20% slopes




Typic Argixerolls on 9-15% slopes




Typic Argixerolls on 3-9% slopes
                                     73

-------
o

_j
<
a
<
a:
z
z
UJ
   3.0-
   2.0-
   1.0-
       CAMP 0-ONGO
                                                         TxClEm

                                                         PxCIDm

                                                         PxCICm
                                                                 WF
                                                                 4ltoll8yrt
            T
         1946-55                    1956-65


                      TEN YEAR GROWTH PERIODS
                                                             1966-75
      Figure 17.  Growth trends  for trees on the  COO plot from 1946

                  to 1975  in  the San Bernardino National Forest.
 ^  3.0 H

 o
<

a

K  2.0-

_j

3
Z

<
    1.0
 UJ
          BREEZY POINT
                                                         • PHclCtn

                                                         •PHclDm
                                                                   pp
                                                                   61 to 100 yn
                                                                   1C 61 to 100
                                                                   PP 61 to 100
                                                                   PP 41 to 60
          1946-55                    1956-65                    1966-75

                           TEN YEAR GROWTH PERIODS

      Figure 18.   Growth trends  for  trees on the BP plot from 1946

                   to 1975 in the San Bernardino National Forest.
                                   74

-------
  s
  5
  I  3.0'
  (T
  O
  5  2.0-

  _J

  z>
  •z.

  <
     1.0-
  UJ
6REEN VALLEY CREEK
-ExsaDm
-TsEm
-Both Soils

-Entire Plot
                                                                     JP 61 to 100
           1946-55
                            1956-65


                  TEN YEAR GROWTH PERIODS
      1966-75
Figure 19.   Growth trends  for  trees on the GVC  plot from 1946  to  1975

             in the San Bernardino National Forest.
   o  3.0-
   cr
   (S
     2.0
   <
   UJ
      1.0-
           DEER LICK
                                      NO  SOIL CLASSIFICATION
                                                                     WF
                                                                     6l»o2OO»
                                                           biro zooyn
                                                          -•SP6H0200
                                                                    •JP 101 to 200
            1946-55
                             1956-65


                   TEN YEAR  GROWTH PERIODS
      1966-75
Figure 20.   Growth trends  for trees on the  DL plot from 1946  to 1975

             in the San Bernardino National  Forest.
                                      75

-------
I
t-


o
cr
    3.0-
    2.0-
  <

  UJ
     1.0-
          NORTH EAST GREEN VALLEY
                                                        -TsEf
                                                                    JP
                                                                   •lOltoZOOyrs
           1946-55
                                    1956-65



                          TEN YEAR GROWTH PERIODS
1966-75
Figure 21.   Growth trends for trees  on the NEGV plot  from 1946 to 1975

             in the San Bernardino National Forest.
  o  3.0
  K
  O
  o
  <
  or
  z
  UI
     2.0-
     1.0 H
          HOLCOMB VALLEY
                                                        -TAflDm

                                                        -TAflCm

                                                        -TAflBm
                                                                   •JP6MOIOO


                                                                    'JPov»r200
           1946-55
                                        T
                                    1956-65



                          TEN YEAR  GROWTH  PERIODS
                                                                1966-75
Figure 22.   Growth trends for  trees on the HV plot  from 1946 to 1975

             in the San Bernardino  National Forest.
                                        76

-------
PHYSICAL AND CHEMICAL PROPERTIES OF SOILS, INCLUDING MOISTURE DYNAMICS

Introduction

     During 1976-'77 the questions addressed were   (1) What are the soil
moisture and temperature regimes and how do they relate to the suscepti-
bility of the vegetation to damage during periods of high oxidant air
pollutant concentrations and do they affect the impact of oxidants on other
organisms such as pathogenic fungi, arthropods and  litter decomposing
organisms?  (2) How do physical properties of the soils affect these mois-
ture and temperature relationships?  (3) Do the plant nutrients in and
chemistry of the soil affect the relationships described above?  In this
context, three kinds of soil characterization measurements carried out were:
soil moisture retention properties; soluble soil phosphorous; and identifi-
cation of the soil and slope with each tree on the major vegetation plots.
The first two characterization measurements were descriptive of the system.

     Soil moisture and temperature data collected continuously since summer
1973 can be used to document the complete water balance including periods
of soil moisutre deficit.  This data will then be related to the physiolog-
ical conditions of the vegetation measured as xylem water potential and
transpiration rate.

Material and Methods

Soil Moisture and Temperature Measurement—

     Soil moisture-temperature sensors (fiberglass moisture blocks) install-
ed at depths of 15, 31, 61, 92, 152, 214 and 274 cm were monitored every one
to three weeks depending upon the rate of change.   Field readings from the
moisture sensor were converted to percent water content (Pw) based upon
calibration curves for soil at each sensor site.  Soil samples were col-
lected at the sensor sites and water content was determined in the labora-
tory to construct the calibration curves.  The values obtained were used
mainly to describe the shape of the curves of soil moisture change with
time.  The values were then adjusted based upon soil moisture content
determined by mechanically sampling of the soil at  the same depths as the
moisture blocks with an auger boring within 1.5 meters of the site of
the sensors.  Moisture retention data also determined on the same sensor-
site samples were used as a further check upon the  accuracy of the final
corrected values.  This procedure was adopted because of the very large
number of moisture sensors emplaced in the field.   Soil temperature was read
directly from the sensors by means of calibration data stored in the compu-
ter program.  Both soil moisture and temperature can be monitored for
accuracy by comparing the results from a given sensor with those above and
                                    77

-------
below it in the soil column.  Any erratic behavior can be readily detected
and corrected.  A few sensors failed and were replaced due mainly to  the
wire leads being bitten off by pocket gophers.  The sensors were remarkably
stable and consistent in their behavior.

Nature and Pattern of Soils—

     Detailed slope and soil classification and mapping completed previously
on 18 major vegetaton plots were used to characterize the slope and soil
under each pine tree on the plots in order to develop relationships of
soil and stand dynamics.

     Soil properties measured—Particle size distribution, bulk density,
exchangeable and soluble cations, soil pH, organic matter and nitrogen were
previously measured in the soils on major vegetation plots (Miller et al.,
1977) During the 1977-'78 period soluble phosphate was measured on 138
surface soil samples taken under 40 trees selected for their relationships
to other subprojects dealing with arthropods, litter decompositon, and
pathogenic fungi.  The latter completes the analyses of the surface soil
samples, reported previously, which included organic carbon, nitrogen,
exchangeable sodium, potassium, calcium and magnesium.

     Specific analytical procedures—Moisture retention was determined
on soils from all major vegetation plots and on two aspect plots at the
same sites used for soil moisture monitoring by fiberglass moisture block
sensors.  Soil moisture tension was measured at 0.1, 0.33, 1, 5 and 15
atmospheres by the methods described by Richards in the book edited by
Black (1965).  However, duplicate 1.0 cm rings were used rather than 5
cm diameter rings described in the method for holding the soil on the
tension and pressure plates. Soil moisture retention was determined on each
site at depths of 15, 61, 91 cm and over 200 cm for deep soils.  Soluble
phosphorous was determined by a colorimetric method (Jackson, 1960) for the
surface soils collected to a depth of 7.5 cm of soil.

Results and Discussion

Soil Moisture Retention Characteristics—

     The computer programs for processing the large amount of data (over
8,000 field readings on the 153 sensors) has been completed and complete
curves were plotted from summer of 1973 to date (summer 1977) .  These
curves were not adjusted for the gravimetric control moisture sampling and
so are not included in this report -  Adjusted curves are being prepared
and will be made available for all participants in the project soon.
However, inspection of the nonadjusted curves show that the moisture-
temperature sensors have been remarkably stable throughout the 5 year
period, in that the maxima of winter moisture and summer minima are nearly
the same each year.  Also the shape of the moisture depletion curves vary
from year to year, and with increased depth of soil or variation in soil
texture.  Further, errors in field readings are readily detected as indi-
cated by individual variations from the smooth curves not accounted for by
sudden rainfall or abrupt changes in air temperature which may affect the

                                    78

-------
upper sensors.

     It appears that the soil moisture-temperature monitoring  during
the 5 year period has been eminently successful and can be used  effectively
in evaluating soil relationships to the impact of oxidant air  pollution
on the ecosystem.  Rates of soil moisture depletion will be  related to
transpiration by the vegetation and to moisture stress on the  plants.

     The results of the analyses are shown  in Table 18 and 19  and  a typical
set of moisture retention curves are shown  in Figure  23.  Water  held
in the soil at low tension (0.1 atm.) is an approximation of field capacity
of the soil to hold water when freely drained.  The minimum water  content
which is available for plant use is considered to be  about 15  atm; this
minimum is called the permanent wilting point.  Difference is  an approxima-
tion of the amount of water that can be stored in the soil which is avail-
able for plant growth.  Values of the differences in  percent water for  the
major vegetation plots are shown in Table 20 and are  called  "available  soil
water".  However, these values are only approximations and do  not  always
agree well with measurement of samples taken directly from the field, and
should be used with caution.  They will be  used to help in calibrating  the
soil moisture monitoring program, and for direct measurement of  soil mois-
ture sampled when wet in the spring and when dry in the late summer or  fall.
"Available soil water" indicated in Table 20 ranges from 7.17  and  7-35
percent by weight (Pw) in the two plots with the sandiest textures to nearly
24 percent in the surface of a number of plots where  the soil  is rich in
organic matter.  Available soil water tends to increase with increased  clay
content in the upper layers of these sandy  soils, but there appears to  be
little relationship between the two variables below 61 cm.
                                    79

-------
TABLE 18.  SOIL WATER CONTENT AT VARIOUS MATRIC SUCTIONS FOR THREE PROFILE
           DEPTHS.
Site
BF
BL
CAO
DW2
GVC
HV
NE12
NEGV
SCI
SC2
SV2
UCC
BP
COO
DW1
DW3
HB
CP
S22M
SCR
SF
TUN2

0-30
31.73
18.31
20.79
22.48
13.87
16.72
25.63
13.22
11.73
14.22
16,56
20.44
26.87
22.30
26.29
24.04
15.79
27.73
26.15
10.54
29.40
19.11

pw
1/10 BAR
30-61
26.88
15.78
20.45
22.22
14.88
12.89
18.59
12.85
12,20
13.45
n.d.
21.04
25.00
18.40
18.80
22.20
11.90
24.50
20.30
10.90
25.90
17-20

61-91 cm
23.02
12.23
19.44
17.86
12.59
13.05
17.29
11.10
13.85
11.17
n.d.
20.07
21.66
17.00
19.02
17.66
12.36
n.d.
18.63
10.50
21.21
n.d.
80
0-30
20.97
11.55
14.98
17.48
10.64
11.50
18.03
9.21
8.74
9.47
13.05
15.90
18.18
17.03
17.40
16.77
11.19
19.83
20.97
6.38
21.36
13.91

pw
1/3 BAR
30-61
15.24
8.37
12.91
13.82
10.25
8.99
12.82
7.71
7.87
6.77
6.59
14.65
16.05
12.85
13.45
13.65
7.47
15.74
14.31
5.93
17.17
11.00

61-91 cm
13.84
7.42
13.77
12.83
8.82
8.94
11.98
6.65
8.28
5.83
n.d.
15.26
17.88
13.55
15.62
13.32
9.42
n.d.
14.56
7.07
17.71
n.d.


-------
TABLE 18.  CONTINUED
Site
BF
BL
CAO
DW2
GVC
HV
NE13
NEGV
SCI
SC2
SV2
UCC
BP
COO
DW1
DW3
HB
CP
S22M
SCR
SF
TUN2
0-30
14.61
8.29
11.25
12,55
7.89
8.59
12.60
6.52
6.69
6.99
9.32
11.09
12.55
12.53
11.91
10.76
6.26
13.39
13.38
5.15
14.74
8.80
pw
1 BAR
30-61
10.15
5.31
9.04
10.21
7.52
7.25
8.83
5.36
6.09
4.66
4.58
11.60
12.10
10.34
9.24
10.43
5.72
12.45
11.65
4.56
13.34
7.71
61-91 cm
9.28
5.25
9.72
9.22
6.49
7.07
8.45
5.25
6.25
4.29
n.d.
12.27
13.29
9.86
10.30
8.54
6.39
n.d.
9.92
4.14
10.96
n.d.
0-30
8.81
5.39
6.33
7.46
3.97
6.23
7.59
4.30
5.00
5.66
5.21
5.60
7.96
8.50
7.08
7.06
4.09
8.86
7.48
4.11
9.00
4.74
pw
5 BAR
30-61
6.30
3.54
5.78
6.40
4.15
5.46
5.27
3.38
4.77
3.14
2.68
8.40
7.04
7.08
5.82
6.34
3.47
8.33
6.58
2.92
8.00
4.24
61-91 cm
5.79
3.54
6.38
6.29
4.00
5.36
5.44
3.15
4.74
2.93
n.d.
9.52
7.57
7.25
7.04
5.58
4.42
n.d.
6.29
2.66
6.94
n.d.
                                    81

-------
TABLE 18.  CONTINUED
Site
BF
BL
CAO
DW2
GVC
HV
NE13
NEGV
SCI
SC2
SV2
UCC
BP
COO
DW1
DW3
HB
CP
S22M
SCR
SF
TUN2
0-30
7.86
5.08
5.27
6.58
3.31
5.66
6.58
3.94
4.56
5.13
4.36
4.78
6.95
6.97
5.90
6.20
3.45
7.42
6.00
3.19
7.46
3.81
pw
15 BAR
30-61
5.84
3.32
5.25
6.16
3.71
5.15
4.87
3.25
4.64
3.14
2.37
7,91
6.47
6.44
5.23
5.80
2.95
7.84
5.99
2.44
7.18
3.93
61-91 cm
5.43
3.25
5.95
6.15
3.62
5.18
5.15
3.06
4.51
2.71
n.d.
9.09
8.16
8.64
6.92
5.06
4.07
n.d.
5.54
2.33
6.20
n.d.
                                     82

-------
TABLE 19.  SOIL WATER CONTENT AT VARIOUS MATRIC SUCTIONS FOR PROFILE DEPTHS
           OF 200 CM OR GREATER.
Site
BP
COO
SCR
SF
BL
DW2
GVC
NE13
UCC
1/10 BAR
200 cm
12.33
10.73
10.09
22.94
10.50
17.69
11.45
15.56
17.06
1/3 BAR
200 cm
7.62
4.14
6.22
18.62
6.45
12.92
7,55
9.37
11.40
1 BAR
200 cm
6.42
2.97
4.36
12.43
3.86
8.85
4.89
6.50
9.23
5 BAR
200 cm
4.70
2.19
2.75
7.21
1.80
4.98
3.15
3.73
6.30
15 BAR
200 cm
4.23
1.87
2.49
6.65
1.63
4.71
2.84
3.50
5.89
                                    83

-------
TABLE 20.  AVAILABLE SOIL WATER AND CLAY CONTENT AS PERCENT OF THE  WHOLE
           SOIL FOR THREE SOIL DEPTHS.
Site
BF
BL
CAO
DW2
GVC
HV
NE13
NEGV
SCI
SC2
SV2
UCC
BP
COO
DW1
DW3
HB
CP
S22M
SCR
SF
TUN2
0-30
Pw
23.87
13.23
15.52
15.90
10.56
11.06
18.05
9.28
7.17
9.09
12.20
15.66
19.92
15.33
20.39
17.84
12.34
20.31
20.15
7.35
21.94
15.30
cm
% Clay
9.0
4.9
9.9
5.4
4.9
9.1
7.4
4.4
5.4
5.6
6.8
4.9
7.9
6.3
6.3
8.9
5.0
7.9
7.9
3.0
10.3
5.2
30-61
Pw
21.04
12.46
15.20
16.06
11.17
7.74
13.72
9.60
7.56
10.31
6. ,63
13.13
18.53
11.96
13.57
16.40
8.95
16.66
14.31
8.46
18.72
13.27
cm
% Clay
7.8
3.6
10.6
4.1
2.3
9.6
6.9
4.3
3.9
4.4
4.0
12.7
6.4
5.0
6.7
8.9
4.7
7.9
7.1
3.2
9.5
5.6
61-91 cm
Pw % Clay
17.59
8.98
13.49
11.71
8.97
7.87
12.14
8.04
9.34
8.46
n.d.
10.98
13.50
8.36
12.10
12.60
8.29
n.d.
13.09
8.17
15.01
n.d.
7.6
4.6
13.0
6.3
2.7
7.9
6.3
2.1
2.8
2.9
n.d.
14.3
6.2
8.6
11.2
8.4
9.4
n.d.
7.2
3.3
7.0
n.d.
                                    84

-------
          20


           15.0 [-


           10.0
            5.0
        10
        DC
        <

        £  2.0
        O  1-5
         en
 1.0

0.8


0.6
         6  0.4
         CO
            0.2


           0,15


            O.I
                                 15 CM
              I
                          5        10         15        20        25

                         SOIL WATER CONTENT (Pw) % BY WEIGHT
Figure 23.  Moisture retention curves for Dogwood Plot  Site  2.
                                   85

-------
STAND TREE MORTALITY SUBSYSTEM—BARK BEETLE POPULATION DYNAMICS

Introduction

     Bark beetles in the genus Dendroctonus attack and kill conifers.
Except in epidemics (the last large outbreak in California was in  1962-
63) these beetles attack and kill trees that have been weakened or stressed
physiologically.  Some factors that predispose trees to attack are drought,
flooding, lightning strikes, mechanical injury due to logging, building  or
road construction, and root disease.  Photochemical oxidants likewise
predispose trees to attack by bark beetles and this has been reviewed by
Cobb ^t_ al.  (1968).

     Three species of bark beetle, Dendroctonus brevicomis Le Conte, D_.
ponderosae Hopkins, and _D. .leffreyi Hopkins, attack and kill pines in the
San Bernardino National Forest (SBNF).  White fir engraver, Scolytus ven-
tralis Le Conte, Ips emarginatus and Melanophila California Van Dyke are
also important species in the pest complex in SBNF.  Pines were chosen for
this study because ponderosa and Jeffrey pines are more susceptible to
oxidant air pollutants than other major tree species in the forest.  A major
part of the study was focused on the Dendroctnous genera of insects because
they appeared to be more important in the SBNF than other genera.  The
California flathead borer, M. California Van Dyke, on Jeffrey pine appears
to be of little importance but it may be recognized as a significant part of
the pest complex on ponderosa pine when the data are analyzed.  Ips emar-
ginatus is found in the pest complex on Jeffrey pine 22% of the time (Fig. 1
in McBride, Dahlsten and Cobb, in Miller 1977) and could be considered
important.  The three Dendroctonus species alluded to above were the only
bark beetles studied and by far the greatest effort was on the western pine
beetle, D_. brevicomis.  An earlier study (Stark et_ al., 1968) on the SBNF
showed that as the severity of oxidant damage to ponderosa pine increased,
the incidence of western pine beetle and mountain pine beetle, D_. ponderosae
infestations increased.  Further substantiation of the interaction between
oxidant damage to pines and the incidence of bark beetles was gathered by
Wood (1971).  A historical analysis on the Lake Arrowhead District of the
SBNF showed substantial increases in bark beetle caused tree mortality since
1951.  These data were obtained by examining the beetle control records.

Research Objectives

     The general objectives of the study were to determine susceptibility of
predisposed trees to bark beetles and the nature of the interrelationship
between oxidant damaged trees and beetle populations.  Specifically the
objectives were as follows:

     1)  To determine the degree of susceptibility of oxidant-injured

                                    86

-------
ponderosa pine to the western pine beetle and  the mountain  pine  beetle
and of Jeffrey pine to the Jeffrey pine beetle and  the California  flat-
headed borer.

     2)  To investigate the influence of oxidant-injured pine  trees  on
the success and productivity of broods of the  four  beetle species  listed
in objective #1.

     3)  To study the direct and indirect influence of photochemical
oxidants on the biology of bark beetles, with  particular emphasis  on the
insect associates, parasitoids, and predators.

     4)  To develop life tables for bark beetles by oxidant injury categor-
ies and, based on these tables, to develop predictive models of  beetle
activity with reference to stand type and pine oxidant-injury  level.

     5)  To determine the biological impact and relative importance  of
each of the beetle species in forest communities and what influence  they
have on stand change and forest succession.

Methods and Materials

Western Pine Beetle—

     Field sampling procedures and laboratory  analyses have been described
in detail for the western pine beetle (Dahlsten, 1974 and Dahlsten,  1977)
and are summarized in, Figure 24.  Basically each beetle generation  was
sampled twice or three times if it was an overwintering generation.   Four
different procedures were used so that each developmental stage  of the
western pine beetle and the insect associates, parasitoids, and  predators
could be accounted for. The type of information taken is shown in  Figure 24
as well as in Tables 21 through 33.

Mountain Pine Beetle and Jeffrey Pine Beetle—

     Procedures for the development of an optimum sampling  design  for
the mountain pine beetle on ponderosa pine and the  Jeffrey  pine  beetle
on Jeffrey pine have been described previously (Dahlsten, 1974 and Dahlsten,
1977).  A nested sampling design was used for  paired samples taken at
the lower, mid and upper portions of six beetle infested Jeffrey pines
and five infested ponderosa pines.  The optimum design is one  which  for
a fixed variance yields the lowest cost or for a fixed cost yields the
lowest variance.  Four sample sizes were used  and several population
attributes were measured, three are indicated  on the figures and tables.

     Variances were estimated for each sample  size  for each variable
within heights, within trees, and between trees using the following  formu-
las:
                                    87

-------
00
00
                                                  FIELD  SAMPLE

                                                    2 egg discs/odd ht [—


                                                     2 lorvol discs/ht  [•
       LAB  ANALYSIS
TREE
                                                                       TREE
                                                                       TBUG
                                                         SAMPLE
                                                                       TREE
                                                     2 larval discs/ht.
                                                   2 emergence cartons | — '
                                                                       TBUG
                                                                      STREE
      Egg Disc Dissection
^attacks
#eggs
gallery length
early larvae
                                                                             X-ray
                                                                            Rearing
                                                                             X-ray
                                                                            Rearing
            parent adults
            early larvae
            predators
            associates
            brood adults
            parasites
            predators
            associates
            teneral adults
            pupae
            late larvae
            parasites
            brood adults
            parasites
            predators
            associates
        Sticky Cartons
        brood adults
        parasites
        predators
        associates
               Figure 24.   Graphic  summary of  the population sampling procedures  used for  the western
                             pine beetle  showing datasets  and the type of information  included for the

                             San Bernardino study.

-------
    TABLE 21.  HEIGHT, DIAMETER, AND LENGTH OF INFESTATION FOR WESTERN PINE BEETLE SAMPLE TREES SAN

               BERNARDINO NATIONAL FOREST, 1973-1976.
oo
VO
Tree
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Yr.-Gen.
73-1
73-1
73-1
73-1
73-1
73-1
73-1
73-1
73-1
73-1
73-1
73-1
73-2
73-2
Tree
Ht.
(m)
18.3
22.9
19.2
28.0
29.0
25.9
43.9
29.9
13.7
19.5
20.7
17.1
18.0
16.5
DBH
(cm)
34.4
44.9
41.7
50.9
—
34.4
87.9
52.5
48.4
36.0
31.5
28.0
29.6
23.2
Top infest.
Ht. Diam.
(m) (cm)
16.5
13.5
13.5
15.5
19.5
13.5
34.5
22.0
12.0
12.0
15.0
12.0
10.8
11.5
15.9
29.6
23.2
32.8
31.8
23.9
23.9
15.9
1.0
22.6
18.1
16.9
17.2
15.0
Bottom infest.
Ht. Diam.
(m) (cm)
1.5
1.5
.5
.5
3.0
.5
.6
.1
.4
.5
.2
.6
.4
.3
34.4
44.9
45.8
57.3
-^
33,7
90.7
62.4
53.8
39,5
34.4
29.9
32.2
27.7
Lowest
sample
On)
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1,
5
5
5
5
5
5
5
5
5
5
5
5
5
5
Highest
sample
(m)
15.
13.
13.
15.
19.
12.
34.
21.
10.
12.
15.
12.
10.
10.
0
5
5
0
5
0
5
0
5
0
0
0
5
5
(continued)

-------
TABLE 21.  (continued)
Tree
No.
15
16
17
18
19
20
21
22
23
24
525
526
527
528
Yr.-Gen.
73-2
73-2
73-2
73-2
73-2
73-2
73-2
73-2
73-2
73-2
74-1
74-1
74-1
74-1
Tree
Ht.
(m)
24.0
17.7
27.5
18.7
34.0
23.0.
22.0
25.0
2Q.Q
25.5
22.8
16.0
—
30.0
DBH
(cm)
53.8
26.4
56.0
28.0
76.4
30.6
33.4
46.2
26.4
36.3
69. .7
39 ..2
73.2
___
Top
Ht.
Cm)
16.8
12.0
22.7
10.2
28. Q
13.1
15.0
18.2
10.0
14.2
15.0
10.9
27.0
24.0
infest.
Diam,
(cm)
23.9
12.4
25.8
21.0
27,1
19,7
22.3
19.7
19.. 1
21.7
51.6
24.2
27.1
45.8
Bottom
Ht.
(m)
.5
.1
.2
.1
.2
.4
,3
3.0
.4
.5
.5
0.0
0.0
—
infest .
Diam,
(cm)
60 .,8
38.2
60.5
34,7
76.4
32,2
43.3
58.9
32.5
39.5
75.4
43.6
87.5
— .,
Lowes t
sample
Cm)
1,5
1.5
1.5
1,5
1.5
1,5
1.5
1.5
1.5
1,5
1.5
1,5
1.5
1,5
Highest
sample
(m)
15.0
12.0
22.5
9.0
19.5
12.0
9.0
16,5
9,0
7.5
10.5
10.5
27.0
24,0
                                                                                 (continued)

-------
TABLE 21. (.continued)
Tree
No.
529
530
531
532
533
534
vo
In***
535
536
537
538
539
540
541
542
Yr.-Gen.
74-1
74-1
74-1
74-1
74-1
74-1

74-1
74-1
74-1
74-2
74-2
74-2
74-2
74-2
Tree
Ht.
(m)
16.2
25,0
22.0
30.2
14.4
19.2

28.4
25.0
26.0
13.7
12.8
13.7
28.9
16.3
DBH
(cm)
30,6
60.5
94.5
67.8
71.6
51.6

96.1
48.4
50. Q
30.2
25.2
53.5
69.1
79.3
Top
Ht.
Cm)
13.0
25.0
21.0
23.6
14.4
16.8

16.5
19.8
22.0
10.0
6.0
11.0
22.5
13.8
infest.
Diam.
(cm)
15.9
—
—
29.3
55.7
21.0

71.0
19.1
16.6
21.7
19.4
39.2
36.9
43.9
Bottom
Ht.
(m)
.4
0.0
—
.1
.7
0.0

.3
.9
.8
0.0
.2
0.0
.4
0.0
infest.
Diatn.
(cm)
35.0
65.3
93.0
82.4
74.8
64.0

110.8
50.0
49.3
33.7
29.9
57.9
78.3
88.5
Lowes t
sample
(m)
1.5
1.5
1.5
1.5
1.5
1.5

1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
Highest
sample
(m)
12.0
18.0
21.0
22.5
13.5
16.5

16.5
19.5
21.0
9.0
6.0
10.5
19.5
13.5
                                                                                   (continued)

-------
     TABLE 21.(continued)
l-o
Tree
No.
543
544
546
547
548
549
550
551
552
553
554
555
556
557
Yr . -Gen .
74-2
74-2
74-2
74-2
74-2
74-2
74-2
74-2
75-1
75-1
75-1
75-1
75-1
75-1
Tree
Ht.
(m)
26.8
22.9
—
20.0
25.0
22.6
19.0
27.0
18.0
21.3
15.6
20.1
21.9
12.5
DBH
(cm)
90.7
47.1
40.4
38.2
43.6
36.3
54.8 ,
63.0
42.0
45.8
39.5
65.3
52.8
32.8
Top infest.
Ht. Diam.
(m) (cm)
18.0
18.5
16.0
19.0
18.3
17.0
12.3
15.0
10.8
15.2
11.2
14.5
12.0
9.0
63.7
23.9
18.1
18.5
25.2
17.2
30.2
41.7
25.8
21.0
21.7
34.7
24.2
17.8
Bottom infest.
Ht. Diam.
(m) (cm)
.3
0.0
1.5
0.0
.5
0.0
2.0
.3
.1
1.5
.3
3.0
.1
•5
95.8
57.6
38.5
46.2
50.0
43.6
53.8
68.1
44.6
45.8
47.8
61.1
65.9
38.2
Lowest
sample
(m)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
3.0
1.5
1.5
Highest
sample
(m)
18.0
18.0
15.0
18.0
18.0
16.5
12.0
13.5
10.5
15.0
10.5
13.5
12.0
9.0
                                                                                        (continued)

-------
TABLE 21.  (continued)
Tree
No.
558
559
560
561
562
563
564
565
566
567
568
569
570
571
Yr.-Gen.
75-2
75-2
75-2
75-2
75-2
75-2
76-1
76-1
76-1
76-1
76-1
76-1
76-2
76-2
Tree
Ht.
On)
10.1
20.4
25.6
29.3
33.8
—
31.7
28.0
32.6
16.2
29.9
38.4
25.6
20.7
DBH
(cm)
48.4
36.9
67.5
62.1
75.4
—
74,5
37.6
111.7
38.8
52.5
95.2
71.0
44.9
Top infest.
Ht. Diam.
(m) (cm)
8.8
15.6
15.0
20.5
18.5
—
27.5
19.0
16.5
11.6
16.9
16.2
16.5
17.2
24.5
17.2
33.1
25.8
53.2
—
36.0
25.2
79.9
18.1
30.6
66.8
48.1
22.6
Bottom infest.
Ht. Diam.
(m) (cm)
0.0
Q.O
0.0
0.0
Q.Q
—
1.0
.4
.3
.1
0.0
0,0
.2
.1
54.1
44.2
77.7
63.7
84.7
—
—
42.0
121.0
40.4
65.9
111.4
78.3
54.8
Lowes t
sample
(m)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
7.5
1.5
1.5
Highest
sample
(m)
7.5
15.0
15.0
19.5
18.0
19.5
27.0
18.0
16.5
10.5
16.5
15.0
16.5
16.5
                                                                                 (continued)

-------
     TABLE 21.   (continued)
Tree
No.
572
573
574
575
Yr . -Gen .
76-2
76-2
76-2
76-2
Tree
Ht.
(m)
18
32
18
27
.9
.3
.6
.7
DBH
(cm)
41.4
83.1
92.0
77.4
Top infest.
Ht. Diam.
(m) (cm)
11.2
23.1
13.5
24.8
22.9
49.0
71.9
25.8
Bottom infest. Lowest
Ht. Diam. sample
(m) (cm) (m)
1.2
3.0
.5
1.3
43
80
96
78
.6 1.5
.5 6.0
.8 1.5
.6 1.5
Highest
sample
(m)
10
22
13
24
.5
.5
.5
.0
\o

-------
TABLE 22.  WESTERN PINE BEETLE INFESTED PONDEROSA PINES THAT WERE SAMPLED
           BETWEEN 1973 AND 1976 RANKED BY OXIDANT DAMAGE.
D . brevicomis
generation
1973-1
1973-2
1974-1
1974-2
1975-1
1975-2
1976-1
1976-2
Years
combined
Generations
combined
Damage class
1-5 6-10

3
1 3
3 4
2
2
2
3
1 7
3 12
4 19
11-15

4
5
3
2



7
7
14
16-20
2
3
1


2
1
1
4
6
10
21-25
2

3

2

1

8
0
8
26-30
5


2


1
1
6
3
9
31-35 36+
2 1
2



1 1
1
1
2 2
2 3
4 5
                                    95

-------
TABLE 23.  WESTERN PINE BEETLE MEAN EGG DISSECTION VARIABLES BY YEAR AND
           GENERATION FOR WHOLE PONDEROSA PINES, SAN BERNARDINO NATIONAL
           FOREST, 1973-1976.
	 	 ••"
Attacks
per dm^

Gallery
length
cm/dm^
Total eggs
per dm^

1st ins tar
larvae
per dm^
Eggs per
cm. of
gallery
Average
tree smog
rating
Year
Gen.
1

2
1

2
1

2
1

2
1

2
1

2
	
1973
2.

0.
79.

41.
71.

41.
59.

30.
0.

0.
28.

17.
88

93
3

7
1

2
4

4
90

97
5

3
1974
2,27

0.99
71.8

36.5
54.7

47.4
45.7

34.2
0.76

1.31
14.6

11.2
1975
2.89

1.48
58.7

46.5
70.9

63.8
53.5

51.8
1.24

1.35
13.8

19.3
1976
1.32

1.04
60.2

39.3
86.8

64.8
68.9

49.5
1.41

1.56
19.8

17.7
4 yr Sierra Nevada Miller
mean Blodgett Forest & Keen
1967-1970 mean 1960
2.

1.
67.

41.
70.

54.
56.

41.
1.

1.
19.

16.
34

20
5

0
9

3
9

5
08

30
2

4
1

1
52

45
74

78
47

56
1

1



.61

.56
.8

.8
.5

.8
.3

.3
.42

.69




1.25





69.8





1.65




                                   96

-------
TABLE 24.  VARIABLES CALCULATED FROM WESTERN PINE BEETLE EGG DISC SAMPLE
           DISSECTION BY GENERATION FROM 1973 TO 1976.  SAN BERNARDINO
           NATIONAL FOREST.
•
Eggs per attack

Gallery length
per attach
1st instar lar-
vae per attack
Average tree
smog rating
Gen.
1
2
1
2
1
2
1
2
1973
25.0
45.4
28.2
48.3
20.9
32,7
28.5
17.3
1974
28.3
66.0
36.1
51.6
23.4
47.6
14.6
11.2
1975
47.8
56.7
38.2
41.0
33.5
44.8
13.8
19.3
1976
70.8
75.8
49.2
47.0
57.0
57.8
19.1
17.7
4 Year
Mean
43.0
61.0
37.9
47.0
33.7
45.7
19.2
16.4
TABLE 25.  CORRELATION OF WESTERN PINE EGG DISC DISSECTION VARIABLES
           WITH YEAR, GENERATION AND TREE OXIDANT RATINGS SAN BERNARDINO
           NATIONAL FOREST, 1973-1976.

Attacks/dm2
Gallery length cm/dm^
o
Eggs, dm''
1st instar larvae/dm2
Eggs/cm, gallery L.
Gallery length/attack
Eggs/attack
1st larvae/attack
Generation
(-) -05
(-) .01
(-) NS
(-) -01
.01
.05
.05
.05
Year
NS
NS
.05
NS
.05
NS
.05
.05
Tree Smog Rating
.05
.05
.01
.05
NS
(-) .05
NS
NS
Notes:  NS   =  Not significant at 5% level or better.
       (-)   =  Negative correlation.
                                    97

-------
TABLE 26.  MEANS BY GENERATIONS ON THE LAST SAMPLE DATE FOR WESTERN PINE
           BEETLE BROOD, PARASITES AND PREDATORS FROM X-RAY ANALYSIS OF
           SAMPLE BARK DISCS.  SAN BERNARDINO NATIONAL FOREST, 1973-1976.
Sierra Nevada
San Bermardino National Forest Blodgett Forest

Total D.b.
brood/dm^
Total para-
sites
Total preda-
tor s/dm^
Mean tree smog
rating
Year 1973
Gen.
1 15.1
2 2.18
1 0.82
2 0.30
1 3.36
2 0.74
1 28.5
2 15.4
1974 1975
16.8 26.5
4.52 3.57
0.59 0.51
0.59 0.34
3.07 1.13
1.25 0.57
14.0 13.8
11.2 19.3
1976
12.4
7.42
0.18
0.16
1.17
0.37
19.8
17.7
Mean 1966-1970 Mean
17.7
4.42
0.52
0.35
2.18
0.73
19.0
16.9
13.6
6.95
0.452
0.613
1.59
0.434
—

                                   98

-------
TABLE 27.  SIGNIFICANCE OF MULTIPLE REGRESSION COEFFICIENTS  FOR  THE LAST
           SAMPLE DATE OF X-RAY, REARING, AND STICKY CARTONS FOR THE WESTERN
           PINE BEETLE AND ITS NATURAL ENEMIES,  SAN BERNARDINO NATIONAL
           FOREST, 1973-19761.
Gen.
Smog
Year rating D.b. Pred. Para.
Live D. brevicomis



Predators



Parasites



iFor D.b.,
X-ray
Rear
Stik
X-ray
Rear
Stik
X-ray
Rear
Stik
the variables
.001
.001
.01
.001
.001
NS
.05
NS
NS
of
.05
NS
NS
.001
.05
.01
NS
NS
NS
generation,
.05 	 NS NS
.05 	 NS NS
.05 	 .001 .05
NS NS 	 .001
NS .01 	 NS
NS .001 	 NS
NS .05 .001 	
NS .01 NS
NS .01 NS 	
year, and smog rating were analyzi
 together, without the effect of predators and parasites. The predators and
 parasites variables were then added to the equation to test their signifi-
 cance.
                                    99

-------
TABLE 28.  MEAN WESTERN PINE BEETLE AND NATURAL ENEMY EMERGENCE BY GENERA-
           TION FOR THE LAST SAMPLE DATE OF LABORATORY REARED DISCS, SAN
           BERNARDINO NATONAL FOREST, 1973 TO THE FIRST GENERATION OF
           1975.

Western pine
beetle per dm^
Predators
per dm
Parasites
per dm
Mean tree
smog rating
TABLE 29. MEAN
TION
SAN
1976


Western pine
beetle per dm^
Predators
per dm
Parasites
per dm
Mean tree
smog rating
Year
Gen.
1
2
1
2
1
2
1
2
	 - 	 _ , ,. 	 — — ________ 	 	 	 — 	 -•
1973 1974 1975
S
8.00 4.82 6.47
1.19 1.00
1.85 1.67 0.523
0.315 0.237
0.697 0.301 0.400
0.318 0.235
28.5 14.9 13.8
15.2 11.2
.B.N.F.
Mean
6.43
1.10
1.35
0.276
0.466
0.277
18.7
13.2
Blodgett
Forest
1966-1970
7.12
2.99
2.88
0.177
0.298
0.630
—
—
WESTERN PINE BEETLE AND NATURAL ENEMY EMERGENCE BY GENERA-
FOR THE LAST SAMPLE DATE OF STICKY CARTONS (FIELD RESEARCH),
BERNARDINO NATIONAL FOREST, 1973 TO THE FOREST GENERATION OF
*
Year
Gen.
1
2
1
2
1
2
1
2
1973 1974 1975 1976

3.86 3.30 1.77 1.60
0.810 0.297 0.763
2.11 1.50 0.746 0.759
0.91 0.459 0.583
2,20 1.33 1.75 0.146
0.359 0.343 0.358
30.4 14.0 13.8 19.8
15.4 11.2 19.3
S. B.N.F.
Mean
2.63
0.623
1.28
0.65
1.36
0.353
19.5
15.3
Blodgett
1966-1970
4.95
1.45
0.947
0.733
1.56
0.639
—
^ ,_
                                    100

-------
TABLE 30   RESULTS OF USING TANG'S CALCULATION OF JEFFREY PINE BEETLE
           DATA:  TREE EFFECT*
Quantity
Measured
Gallery
Length
Density


Attack
Density



TABLE 31.

Quantity
Measured
Gallery
Length
Density


Attack
Density



Sample Calcu-  50
Size lated 6 trees
100cm2 1.02 1.7
100cm2 1.14 1.7
250cm2 1.52 1.7
500cm2 1.91 1.7
1000cm2 1.82 1.7
100cm2 .87 1.7
100cm2 1.28 1.7
250cm2 1.78 1.7
500cm2 1.78 1.7
1000cm2 2.47 1.7
 50
18 trees
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
RESULTS OF USING TANG'S CALCULATIONS ON
DATA: HEIGHT EFFECT.
Sample
Size  Calculated
100cm2 .484
100cm2 .609
250cm2 .611
500cm2 .442
1000cm2 .257
100cm2 .272
100cm2 .272
250cm2 .365
500cm2 .424
1000cm2 .474

* 50
3 heights
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
§ 50 Avg= Conf.
3 trees effect found
2.0 .177
2.0 .166
2.0 .181
2.0 .170
2.0 .128
2.0 .0018
2.0 .0026
2.0 .0018
2.0 .0016
2.0 .0016
JEFFREY PINE BEETLE

 50 Conf.
6 heights found
1.15 .290
1.15 .42
1.15 .42
1.15 .25
1.15 .093
1.15 .10
1.15 .87
1.15 .18
1.15 .24
1.15 .28
67
41
95
99
99
50
87
98
99
99













                                   101

-------
TABLE  32.   RESULTS  OF USING  VARIOUS TESTS ON ATTACK DENSITY,  SAMPLE SIZE OF
            DATA, TO DETERMINE DIFFERENCES BETWEEN  PAIRS  OF  TREES.   VALUES IN
            ATTACKS/CM2.

Trees
compared
Mean differ
with
confidence
For
mean
dif.
Actual
mean
dif.
Rel.
that
fer
likelihood
mean dif-
by 0.25
Means
differ by
at least
compared by

1.2
1-3
1.4
1.5
1.6

.17
.98
.83
.99
.95

.0033
.0033
.0035
.0029
.00339

.0003
.0040
.0023
.0043
.00333
D
0.69
13.0
2.3
89.5
0.81
2D
.47
.71
,55
.94
.21


.0068

.0014

D  is  the measured difference in means.
TABLE  33.   PRELIMINARY SUMMARY OF FINAL SMOG DAMAGE RATINGS FOR PINES
            KILLED BY  INSECTS ON ESTABLISHED VEGETATION PLOTS,  1973-1975.
Tree
species"*"
PP
PP
PP
.
PP

JP
JP

Insect
species^
D.b.
D.p.
Mixed (D.b.
broods + D.p . )
Ips & M.c.
(combined)
D.j.
Mixed (D.j.
broods + Ips)
Number
of
trees
17
7

8

5
7

4
*
Oxidant injury score
Mean
9.9
10.4

11.8

15.2
11.6

13.0
SD SE
6.3 1,5
6.3 2.4

8.6 3.0

10.0 4.5
6.7 2.5

8.2 4.1
Range
1-21
6-25

1-30

2-32
3-19

4-23
*A11 scores given by P. Miller except those from  1973.
+pp = ponderosa pine;  JP = Jeffrey pine.


~T).b. = Dendroctonus brevicomis;  D.p. = D. ponderosa;  D.j
 jeffreyi; Ips = Ips sp; M.c. = Melanophila californica.
= D.
                                    102

-------
     Within levels
                  632
                                  - 7i>
                     .1-1 i=l
                      6-3-2-1


     Within trees:
                  6    3   -     -2
        0-2  .     *    2<7jk - 7k>
          WT     k=l  .1 = 1	
                      6-3-1

     Between trees:

                    6    -    3
         0-2   =     j (7k - y>
           BT      k=1
in which

            value °f variable  for  itn  bark sample,  jtlri level,  kth tree
         = mean of 2 bark  samples  at  jtn level,  k*-" tree
          mean of 6 bark  samples  for  k   tree
     y = mean overall 6  trees


     We assume, for a new population to be sampled,  if the variances were
similar to those of this population,  that the variance of an estimate
would be


      2      2        2         2
             BT      __WT       WL
             n       3'n      3'm-n


                                     103

-------
in which

          n = number  of  trees

          m = number  of  samples  per  each  of  the 3 levels

      Costs of sampling were  estimated by  the following formula

          C  =   $120  + $123  x  n  +  Cinm

in which  C^ is  the  cost  of cutting and measuring a bark sample of size
i, $123 is the  additional cost of  measuring  each tree  and $120 is the one
time  set-up cost  of measuring  each tree and  $120 is the one time set-up
cost  of sampling ;   These costs were  derived  from tallies of the time spent
doing the sampling  tasks, CIQQ = $2^17, ^25Q =  $2>77>  C500 = $4.07
and CIQOO = $6.10.
     Using  the  above  formulas  sampling was  simulated  for  various combina-
tions of numbers  of trees,  samples per tree and  sample  sizes.   For each
combination the corresponding  cost was also calculated.

     Application  of the Tang procedure — During the  initial  phase of the
analysis of the mountain pine  beetle and Jeffrey pine data  several types
of analyses were  used.  One procedure was found  to  be particularly valuable
and some calculations using data  on the Jeffrey  pine  beetle were used
as an illustration of the application of the Tang procedure.

     The Tang procedure supplies  the probability that an  analysis of variance
test will find  significance for a certain test given  an estimate of the size
of the effect of  the  treatment, an estimate of the  variance and the confi-
dence desired.  By use of this procedure the probability  that  an analysis
of variance calculation would  show significance  given that  the effect is
of a certain size can be determined.  Thus,  use  of  this test allows one
to conclude that  if an analysis of variance computation does not show sig-
nificance,  the  actual effect was  probably less than a given size.

     The Tang procedure is  as  follows: the  quantity (j) is  calculated,
where (j)2 is given by:


              ,2     m2d2
              $         i
                  kd2

     d  is the size of the ith effect
     a2 is the variance
     k  is the number of treatments
     m  is the number of replications of  each  treatment.

Separate computations are made for each effect in an experiment in which
more than one effect is tested.  m,k,d,
-------
effects.

     For a given combination of degrees  of  freedom,  confidence level,  and
value of Q-2, the probabilities of  finding significance can be found in
tables in statistical textbooks.   The  probabilities  are tabulated for
 = 2,2.5,2,2,5,3,4,5,6,7,8.

Vegetation Plot Surveys—

     Each numbered tree on each of  the 19 vegetation plots was examined
twice a year, in July and November.  The cause  of mortality,  if due to
an insect, was recorded if a tree  had  died  since the previous survey date.

Results and Discussion

Western Pine Beetle—

     The size, extent of infestaton, and the  location of  the  top and bottom
sample is summarized for each Ponderosa  pine  sampled from 1973 to 1976 in
Table 21.  Most of the sample trees fall in the middle size categories of
those attacked since the large trees are too  difficult to sample.   Also,
small trees were not sampled as the western pine broods were  often mixed  with
Ips spp.  All of the sample trees were smog rated using Miller's (1973)
scoring system.  The distribution  of sample trees by smog score is shown  in
Table 22.

     Evidence for the direct or indirect influence of photochemical oxidants
on the dynamics of western pine beetle populations is circumstantial.
However, it is interesting to compare  the San Bernardino  populations
with other western pine beetle populations.  Where possible the various
population parameters are compared with  data  from Blodgett Experimental
Forest (Dahlsten et_ _§!_•, 1974) where trees  are  not damaged by photochemical
oxidants but are weakened by a root disease,  Verticicladiella wagneri
(Dahlsten and Rowney, 1974)•  Some  comparisons  can be made with the data
summarized by Miller and Keen (1960) but their  information does not separate
out generation effects, which are  considerable.  The Miller and Keen
data was collected before smog damage  was prevalent  in southern California
and the data was collected at sites throughout  the state.

     Egg disk dissection variables—The  first samples taken at the beginning
of each generation were the egg disks.   The mean values by generation
for each of the variables sampled  is given  in Table  23 along  with mean
values from Blodgett Forest (Dahlsten  et^ al., 1974)  and Miller and Keen
(1960). Evaluation on a tree by tree basis  or by height was not done for
this analysis. The main objective was  to find possible photochemical oxidant
effects on the different life stages of  the beetle.

     Generally the first generation of beetles  attacking  in a given year
emerge from the overwintering brood trees.  These beetles set up what  is
referred to in the text as the first generation.  This generation is more
discrete in terms of development,  it is  the most successful and the most
abundant*  It is conceivable, too,  that  the condition of  the  host trees

                                    105

-------
in mid-June to July when the first generation begins may have  considerable
influence on the success of the generation.  The second generation  begins
from mid-August to September; at this time of year it is dry and warm,
and the incidence of photochemical oxidants is at its peak.  Stress on
the trees would be greatest at this time of year and the beetles may there-
fore be affected.  Note that the mean tree smog rating was usually  (3 of  4
years) lower (a lower score means more oxidant damage) in the  second genera-
tion (Table 23).

     Attack densities per square decimeter (dm2) were much higher in SBNF
in the first generation than those previously recorded at Blodgett  or by
Miller and Keen, Table 23.  The second generation was much lower and this
could well be the influence of photochemical oxidants on the trees,  i.e.,  it
takes fewer beetles to kill the trees because trees are weakened at this
time of year.  The low attack densities in 1976, generation one at  SBNF
could have been due to drought conditions or to a declining beetle  popula-
tion.  Note the higher densities in 1975, generation two, Table 22.   This,
too, could be related to moisture conditions.  There was a close relation-
ship between the cm of gallery length per dm2 and attacks and  the same
relationships hold, Table 23.

     The relationship between attacks and total eggs per dm^ was con-
sistent except at Blodgett.  However, there were small outbreaks at Blodgett
in 1967 and 1969 which could explain the jump in the number of eggs in
generation two.  It appears that whenever an outbreak occurs it is  preceded
by very high densities of generation two in the previous year.

     Density of first instar larvae was as to be expected based on  the
preceding discussion.  Egg mortality can be calculated by the  difference
between total eggs and first instar larvae.  The percent egg mortality at
SBNF was 19.7 in generation one and 23.6 in generation two. At Blodgett it
was 36.5% in generation one and 28.5% for generation two. There is  no
explanation for these differences except perhaps the influence of natural
enemies.  This illustrates the importance of developing life tables which
are not yet available.  Of particular interest is the large difference in
egg mortality between the first generation at SBNF and that at Blodgett.

     The eggs per cm of gallery length is usually higher in generation
two.  This is undoubtedly due to competition, and is an inverse relation-
ship between attacks and eggs per cm of gallery length.  The calculated
variables in Table 24 further demonstrate intraspecific competition.

     Since all trees were ranked by oxidant rating it was possible  through
covariance analysis to analyze for smog as well as generation  and year
effects, Table 25.  From the data in Table 24 it appears that  a direct or
perhaps indirect effect could be attributed to photochemical oxidants in
generation two.  Looking at the tree by tree smog rating shows significant
differences between high scoring (not damaged) and low scoring (oxidant
damaged) trees for most variables.  There is a negative correlation for
gallery length/attack and this is to be expected and is due to competition
as described above, i.e., the lower the smog score, the greater the gallery
length as the attack density is lower, Table 24.  There are some  significant

                                    106

-------
relationships by year and almost all correlations are  significant by  genera-
tion as would be expected by looking at Tables  23 and  24.

     X-ray analysis—The second sampling procedure is  designed  to analyze
the density of western pine beetle brood, its parasites and predators,
Figure 24.  As with the egg variables, generation one  is always higher
than generation two, Table 26.  Generation one  at San  Bernardino was  higher
than Blodgett but generation two was lower.  The very  low densities for
the second generation could well be an effect of smog  directly or indirectly
as described above.

     The parasite and predator densities are very low  but tend  to be  higher
in SBNF than Blodgett except for the second generation parasites, Table
26.  It is interesting and perhaps important that the  highest brood density
in the second generation at SBNF occurred when  both the predators and
parasites were the lowest.

     Generation, year, and oxidant ratings are  significantly different
for western pine beetle brood, Table 27.  Generation was significant  for
the predators and parasites but there is no influence  of smog at least
indirectly as determined by tree oxidant ratings.

     Rearing analysis—Each of the x-ray discs  was placed into  a carton
for rearing and the individuals emerging were identified and recorded.
Only the data from the last sample date was considered, as it corres-
ponded more closely to actual emergence in the  field and could be compared
with the sticky carton samples.  The rear data  has only been analyzed
through the first generation of 1975, Table 28.  The same trends are  shown
with this sampling procedure, i.e., generation  one emergence was higher for
western pine beetle brood, parasites and predators than generation two.
Significant effects of tree smog rating are shown only for western pine
beetle brood, Table 27.  There were significant effects of generation
and year on predators but not on parasites.

     A possible link between western pine beetle success and smog is  the
low mean emergence of beetles in the first generation, Table 28.  The
densities for western pine beetle were higher than Blodgett up  to emer-
gence, and then fell below Blodgett.  The SBNF  second  generation emer-
gence was extremely low. Predator emergence in  generation one was lower than
Blodgett, while generation two was higher; the  relationship between para-
sites was just the opposite.  Both parasitoids  and predators were more
numerous in the first generation, this was true for x-ray and for sticky
cartons.

     Sticky carton analysis—Analysis of the sticky carton data is
complete through the first generation of 1976.  Results from this sample
procedure show the same trends as the rearing data, Table 29-   The differ-
ential effects of environment can be seen by comparing the rearing
samples with the sticky carton samples.  The mean densities for western pine
beetle were always higher in the rearing samples than  the sticky carton
samples.  However, parasites and predators were more dense in the sticky
cartons except for the first generation predators.  The relationships

                                    107

-------
were exactly the same at Blodgett, Tables  28 and  29.   This  suggests  that one
method should be used for the bark beetle  and another  for the natural
enemies.

     Significant differences are shown for generation  and smog  rating,
but not for years for the western pine beetle.  With the natural  enemies
only year was significant for predators.

     While there is only circumstantial evidence  for the direct effect
of photochemical oxidants on the western pine beetle,  there  is  good  evidence
of an indirect effect on the brood from each of the four sampling proce-
dures. Smog damaged trees had a depressing effect on western pine beetle
populations.  The implications of this in  the forest community  in San
Bernardino are that the probability of an  outbreak is  lessened  since it
takes fewer beetles to kill diseased trees and fewer beetles are produced  in
such a system.  Also, since it takes more beetles to kill healthier  trees,
the healthier trees should be relatively safe.  The incidence of drought,
mechanical injury, root disease, and fire  could drastically alter these
generalizations for a particular site.  It remains to  be seen what this
means in terms of management in a recreation forest such as the San  Bernar-
dino.  The only conclusion that can be made now is that attempts to  control
the western pine beetle are not necessary as this is treating a symptom
rather than the cause of the problem.

Mountain Pine Beetle and Jeffrey Pine Beetle—

     The associates, parasitoids and predators reared  from trees infested
by mountain pine beetle or Jeffrey pine beetle have been listed previously
(Dahlsten, 1977).  In addition Tables giving means by  height in the  trees
and sample size for number of attacks and gallery length for each beetle
species have been recorded earlier (Dahlsten, 1977) •

     A series of analyses was made on several variables (attacks, gallery
length, total larvae, emergence, and total pupae) for  each beetle species.
For all runs, the lowest cost for a fixed variance or  the lowest variance
for a fixed cost occurred with the 500 or  1000 cm^ disk sizes.  Figure
25 shows a typical relationship for cost of sampling for fixed variances
as a function of bark sample size and Figure 26 is similar except that
variance is shown as a function of bark sample size for fixed costs.  The
feasibility of using the larger bark samples are  dubious since handling
becomes difficult beyond a certain size and consequently more measurement
error may be introduced.  The implications are, however, that the larger
bark sample sizes are most cost effective.

     Variance as a function of cost for a  fixed bark sample  size for various
numbers of samples per height are shown in Figure 27.  In nearly every
case the sampling of more than 2 units per level  did not increase cost
effectiveness.  Even in cases where 3 or more units per height  were  better
the results were nearly as good for 2 as for more sample units  per level.

     Application of Tang's procedure—Only two of the  variables measured
for the Jeffrey pine beetle were used to test the application of Tang's

                                    108

-------
   7
O
Q

-------
 UJ
 o
 o:
 
-------
        8
       7-
    LJ
    O
    z
    <  4
    tr
                                          8       10       12
                                           COST (in Thousands)
14
16
18
Figure 27.  Variance as a function of cost for different numbers of 1000 cm2 samples per
            height for gallery length of the mountain pine beetle in ponderosa pine, San
            Bernardino National Forest, 1974.

-------
procedure.  The first set of calculations was  performed  to  check the sig-
nificance of  the  tree effect on attack density,  i.e.,  to find out if there
was significant variation in attack density between  trees.   For the given
conditions  (6 trees, 3 heights, 2 measurements at  each height which were
considered  replicates), $50* the value of cj) needed to  have  a 50% proba-
bility of finding significance is 1.7.   If 18  trees  were used 4>50 would
be approximately  1.0, and for 3 trees about 2.0.   The  cj>56 are given
approximately because the values must be interpolated  from  graphs.  For
sample size 1(100 cm2 disk) ^was .871.  Hence  using  sample  size 1 there
would be little chance of finding significance even  if 18 trees were used.
On the other hand for sample size 5(1000 cm2 rectangle) cj) was 2.47.  Thus
using sample  size 5 there was an excellent chance  of finding a significant
effect with 6 or  even with  3 trees.  The reason  for  the  difference is that
sample size 1 is much smaller, so that the fluctuations  are a bigger per-
centage of  the actual measured values.  Hence  the  ratio  of  the between group
to the within group variance is smaller, and more  measurements are needed to
establish significance.

     Generalizations can be made from this example to  construct a procedure
for using Tang calculations in planning sampling schemes.   Suppose there
is a desire to determine if there is an effect due to  treatment of at
least a certain size (on the average).  Then 5Q should  be  calculated
for several plausible experimental designs, using  values  of 02 estimated
from previous experiments.  Actually it might  be preferable to calculate
  85 or  95' When the measurements are made or relevant  calculations
performed the significance  test will show (with  appropriate confidence)
whether or  not the effect of the treatment is  greater  than  a predetermined
size.

     Table  30 shows the results of calculations  for  attack  density and
gallery length density for  the tree effect.  Table 31  presents results
on the same data  for the height effect, i.e.,  the  significance of being
above, below, or  at the center of the infestation.

     It can be easily seen  from the tables that  the  chance  of finding an
effect from a given number  of trees generally  increases  with the sample
size.  Moreover, in cases where the Tang calculation predicted a high
probability of finding significance, the confidence  found was generally
high.  The  number in the "confidence found" column is  1  minus the pro-
bability of obtaining at random the F value actually found.

     Some additional tests were performed on this  data.   For selected
pairs of trees the t test was used to determine  the  significance of the
differences in attack density between trees.   The  t  test  is a special
case of the analysis of variance test in which there are  only two treat-
ments, e.g., one degree of freedom in the numerator.   The calculations
below are presented merely to illustrate the techniques  on  several pairs
of data sets.  Multiple comparisons should not be  made with a test on
several samples.  This is because if there are many  samples it is likely
that some pairs will have what appear to be significant  differences,  even if
there is no effect of treatment.  In such cases  an analysis of variance or
multiple comparison test should be used.

                                    112

-------
     In addition to the basic t test  the following  calculations  were per-
formed to add information to that supplied by  the t test:

     1.  Difference in population means that would  be just  large enough
to give a significant difference (95% confidence level) was computed and
compared to the actual difference in means.

     2.  If the means differed significantly,  then  a minimum difference
value, M was calculated.  It can be said with  confidence  (95% confidence
level) that the means differ by at least M.

     3«  A calculation was made to find the ratio of the  likelihood  that,
given the observed results, the observed means differ by  the measured mean,
to the likelihood that the means were the same.  In addition,  the ratio
of the likelihood that the observed means differ by twice the measured
mean to the likelihood that they differ by 9 was computed.   In the formula
for t(equation 1) for the difference of the two populations the  appropriate
difference was substituted and the differential probablity  computed  from
equation 2, the probability density function for the t variable.
     (1)  t  =*
X-Y

s 2
  n
         X  is the mean for the first population

         Y  is the mean for the second population

         s  is the estimate of the variance

         n  is the number of degrees of freedom in each population
     (2) f(u)  =  const. (1 + u2) = -|f + 1
     This test can provide a lot more information additional  to  that  in
a significance test because it can provide confidence levels  on  the differ-
ences in the means.  It can give the relative likelihood that the popula-
tion means actually differ by any combination of values.  If  it  is found
that the means are about as likely to differ by 0 as by the measured  means,
then the two populations are likely to be the same.  If, on the  other
hand, it is twenty times as likely that the populations differ by the mea-
sured mean as by 0, it can be concluded that they are significantly different
and also have an idea of the size of the difference.

     Table 32 shows some of the results of these calculations.   These are
for comparisons of various trees for attack density using sample size
4(500 cm^ rectangle).
                                    113

-------
Vegetation Plot Surveys—

     The results of the vegetation plot surveys through  1975  (Table  33)
show that the western pine beetle and the Jeffrey pine beetle were the two
most important species in terms of pine mortality. A more complete analysis
of the cause of tree mortality on the San Bernardino National Forest  is
currently in progress (McBride et^ al., 1977).  There is  no indication as  yet
that the smaller vegetation plot survey is representative of the mortality
occurring on the San Bernardino National Forest.
                                    114

-------
EFFECTS OF PHOTOCHEMICAL AIR POLLUTION ON THE EPIDEMIOLOGY OF FOREST TREE
PATHOGENS

Introduction

     Since the last report, several studies that were in progress have been
completed and others initiated during the period 1976-1977 have also been
completed.  The results of these studies are summarized as follows.

Susceptibility of Roots of Mature Trees—

     All live tree root inoculations have been completed.  Data (Table
34) for ponderosa pine indicate differences in susceptibility to infection
among trees from various air pollution damage categories.  Roots from
trees showing very severe to severe damage were much more susceptible
to proximal infection by JF. annosus than roots from trees with slight
air pollution injury.  The term proximal infection is used to indicate that
the portion of root between the point of inoculation and the main stem of
the tree was invaded by the organism.  Distal infection or colonization
refers to fungal invasion toward the root tip from point of inoculation.

     When distal and proximal infection are considered together, no clear
differnces among air pollution damage classes emerge.  Results for Jeffrey
pine do not provide enough basis for air pollution damage comparisons because
of the absence of severely injured trees in the sample.

     Colonization results for inoculated roots are presented in Table 35.
For ponderosa pine, data indicate definite differences in proximal coloniza-
tion rate in roots from trees showing various levels of air pollution
damage.  The trend is like that described for infection; i.e., roots from
trees severely damaged were more susceptible to colonization by the pathogen.
Distal colonization rates show no distinct air pollution related trends.

     Regression lines for comparisons between oxidant injury scores and
rates of proximal colonization are shown in Figure 28.  For both ponderosa
pine inoculation trials, relationships between these two variables were
statistically significant (P = 0.01).

     Greater infection and more rapid colonization of roots by _F. annosus
in severely injured trees could have significant effects on disease epi-
demiology in affected stands.  In such stands, greater tree mortality in
shorter time periods could be expected.

Stump Inoculation Studies—

     Studies described in the previous progress report were completed.

                                    115

-------
TABLE 34.  INFECTION OF INOCULATED PINE ROOTS WITH FOMES ANNOSUS  IN RELATION
TO THE SEVERITY OF AIR POLLUTION INJURY.
Air Pollution No. Roots Root Infection
Site Species3 Damage Inoc. A" % Bc %
Snow Valley JP

Holcomb Valley JP
Breezy Point PP


Camp Paivika PP


Moderate
Slight-No Injury
Slight-No Injury
Very Severe-Severe
Moderate
Slight-Very Slight
Very Severe-Severe
Moderate
Slight
12
12
19
16
8
16
22
17
10
5
7
10
7
5
7
15
6
5
41.7
58.3
52.6
43.7
62.5
43.7
68.2
35.3
50.0
0
4
3
5
0
1
15
5
1
0
33.3
15.8
31.3
0
6.3
68.2
29.4
10.0
aJP = Jeffrey pine; PP = ponderosa pine

°No. of roots showing proximal and/or distal infection from the point of
 inoculation

cNo. of roots showing proximal infection from the point of inoculation
                                    116

-------
TABLE 35.  COLONIZATION OF INOCULATED PINE ROOTS  BY FOMES  ANNOSUS  IN RELA-
TION
Site
Snow Valley
Holcomb Valley

Breezy Point


Camp Paivika

TO THE SEVERITY OF AIR POLLUTION INJURY.
Air Pollution
Speciesa Damage
Moderate
JP
Slight-No Injury
JP Slight-No Injury
Very Severe-Severe
PP Moderate
Slight-Vey Slight
Very Severe-Severe
PP Moderate
Slight
Distal
Colonization"
14.7
17.4
27.4
11.3
20.5
10.4
9.4
6.1
6.4
Proximal
Colonization"
0
3.1
0.3
5.6
0
0.05
2.0
0.7
0.07
aJP = Jeffrey pine; PP = ponderosa pine




 mm/month
                                    117

-------
oo
         o
         E
OTMB ^' f
o^
P o
< Z
N Z
o£
-J >-
< CD
     X  H
     O  O
     tr  o
     a.  tr
        I2J
             10-
              8-
              6-
        4-
              2-
                    \
                                     \
 LEGEND
 _ - - — Ponderosa Pine  (Breezy Point)
       Statistically Significant (P=0101)
       Regression Equation:
          y = 10.7 - 0.5 x

•in  i»—. Ponderosa Pine  (Camp  Pavika)
       Statistically Significant (P=0.01)
       Regression Equation:
          y = 3.8 - 0.2 x


 "      Jeffrey Pine (Snow Valley & Holcomb
                   Valley
       Not Statistically Significant
       Regression Equation:
          y = 0.1 + 0.03
                                   12     16   '  20  '  24  '   28  '  32
                                    OXIDANT  INJURY SCORE
                                (DECREASED INJURY
                                                                  »
     Figure 28.  Relationships between oxidant air pollution injury and proximal colonization of
                inoculated  pine roots by Fomes annosus.

-------
Also, two additional twenty-stump trials were done  (one  for  Jeffrey pine
and one for ponderosa pine).  Table 36 summarizes characteristics  of stumps
which were inoculated.

     All inoculated stumps became infected with J?.  annosus.   Stumps from
trees severely injured by air pollution were about  twice as  susceptible
to _F. annosus, based upon percent of surface area colonized  (Table 37),
as stumps from trees showing slight or no injury.   Surface colonization
percentages were similar for ponderosa pine at Barton Flats  and  Jeffrey
pine at Amphitheatre.  Much greater surface colonization occurred  in ponder-
osa pine stumps at Camp Paivika.  This may be explained  in part  by the
different _F. annosus isolate used there.

     Regression analyses comparing oxidant injury score  with surface coloni-
zation by JF. annosus were completed.  For ponderosa and  Jeffrey  pine,
as the oxidant injury score decreased (corresponding to  greater  air pollu-
tion injury) stump surface colonization increased.  This correlation was
significant for ponderosa pine at both sites tested (Barton  Flats  P=0.025;
Camp Paivika P=0.01).  The correlation for Jeffrey  pine  at Amphitheatre
was significant only at the 0.25 level.

     Table 38 presents JF. annosus colonization rate and  extension  in stumps
at least six months after inoculation.  The pathogen colonized stumps of
ponderosa pine trees severely injured by air pollution at a  30%  greater  rate
than stumps from trees showing slight or no injury.  A 50% rate  differential
was found in colonization of Jeffrey pine stumps. Stump  colonization rates
at Barton Flats were less than at the other two sites.   This is  probably
caused, in part, by season of inoculation and isolate differences  from other
trials.

     Regression analyses, comparing downward colonization and coloniza-
tion rates with oxidant injury score were done.  As oxidant  injury score
decreased, indicating greater injury colonization rate and downward fungal
extension increased. For both dependent variables,  relationships were
statistically significant at the .05 level for ponderosa pine at CP and
Jeffrey pine at the Amphitheatre.  They were significant for ponderosa pine
at BF only at the 0.25 level.

     The stump volume colonized by _F. annosus six (BF and Amphitheatre)
and ten (CP) months after inoculation are summarized in  Table 39.   At
BF, stumps from trees severely injured by air pollution  had  twice  as much
volume colonized by _F. annosus as stumps from trees slightly injured.
Amphitheatre and CP had nearly a 3:1 volume colonization ratio of  stumps
from severely injured trees to stumps from slightly injured  trees.

     Regression analysis showed that the volume of  stump colonized in-
creased with decreasing oxidant injury scores.  This correlation was
statistically significant at the .01 level for ponderosa pine at CP,
but not at BF (P = .25).  The correlation was significant at the 0.10
level for Jeffrey pine (Amphitheatre).
                                     119

-------
TABLE 36.  NUMBER OF PINE STUMPS INOCULATED WITH FOMES ANNOSUS  BY  SITE
           AND SPECIES.
Barton Flats Amphitheatre
Pollution Damage* (Ponderosa Pine) (Jeffrey Pine)
Very Severe
Severe
Moderate
Slight
Very Slight
No Injury
Total
Ave. DBH (dm)
Ave. Hgt.(m)
*Equivalent Numerical
Very Severe:
Severe:
Moderate :
Slight :
Very Slight:
No Inj ury :
5 1
5 4
0 5
3 2
3 7
4 1
20 20
1,54 2.27
8.64 11.81
Ratings:
1-8
9-14
15-21
22-28
29-35
36+
Camp Paivika
(Ponderosa Pine)
5
5
0
9
1
0
20
2.53
10.60


                                    120

-------
TABLE 37.  RELATIONSHIP BETWEEN THE SURFACE COLONIZATION OF  INOCULATED
           PINE STUMPS BY FOMES ANNOSUS AND THE SEVERITY OF  AIR POLLUTION
INJURY.
Site1
BF

BF
AMP

AMP
CP

CP
Species^
PP

PP
JP

JP
PP

PP
F. .annosus
Isolate
SV1

SV1
SV1

SV1
JL1

JL1
Air Pollution
Damage
Very Severe-
Severe

Slight-No
Injury
Very Severe-
Moderate

Slight-No
Inj ury
Very Severe-
Severe

Slight-Very
Slight
% Surface
Colonization^
36.9

17.6
33.1

15.8
86.0

45.9
Ratio

2.10


2.09


1.87

1BF = Barton Flats; AMP = Amphitheatre; CP = Camp Paivika

2pp = ponderosa pine; JP = Jeffrey pine

^One month after inoculation
                                    121

-------
TABLE 38.  DOWNWARD COLONIZATION AND COLONIZATION OF FOMES ANNOSUS  IN
           INOCULATED PINE STUMP RELATIVE TO AIR POLLUTION INJURY.
Site1
BF

BF

AMP

AMP

CP


CP

Species2
PP

PP

JP

JP

PP


PP

Air Pollution Downward
Damage Colonization3
Very Severe-
Severe , 22 Q
Slight-No 94 5
Injury
Very Severe- ,
... . jUb.l
Moderate

Slight-No
In j ury
Very Severe- ,07 _
Severe

Slight-Very
Slight Jli-4
Colonization
Rate* Ratio

20.5 U3Q
15.8

51.0
1.49
34 2

40 7

1.31
31 1

      Barton Flats; AMP = Amphitheatre; CP = Camp Paivika

 PP = ponderosa pine; JP = Jeffrey pine
     months after inoculation for Barton Flats and Amphitheatre;
 ten months after inoculation for Camp Paivika (Given in mm)

^Given in mm/month
                                    122

-------
TABLE 39.  VOLUME OF INOCULATED  STUMPS  COLONIZED BY FOMES ANNOSUS IN
           RELATION TO AIR POLLUTION  INJURY.
                           Air Pollution         % Volume
Site^        Species^          Damage           Colonized^         Ratio

  BF            PP         Very  Severe-
                           „                       1 j « o
                           Severe
                                                                   1.95
  BF            PP         Slight-No
                           Injury                  8'°

  AMP           JP         Very  Severe-
                           i,  i    ._                 f-O • 1
                           Moderate
                                                                   3.19
  AMP           JP         Slight-No               _  0
                           _  .                     o «o
                           Injury

  CP            PP         Very  Severe-            , ,  n
                           _                       bl «U
                           Severe
                                                                   3.10
  CP            PP         Slight-Very             in  _
                           Slight                  19'7
      Barton Flats; AMP = Amphitheatre;  CP  = Camp Paivika

2pp m ponderosa pine;  JP = Jeffrey  pine

^For Barton Flats  the  figures  indicate % volume colonized in the top
  130 mm of each stump  at the end  of six  months.  For Amphitheatre and
  Camp Paivika, the figures indicate colonization for the entire height
  of each stump at  the  end of six  and ten months respectively.
                                     123

-------
Effects on Growth,  Spore Production,  Spore Germination,  and  Adaptability
of Fomes Annosus—

      Photochemical  air pollutants probably influence J>  annosus  directly
during portions of  the disease  cycle, such as during spore production,
dispersal and germination.   Penetration  of stump  surfaces by the fungus
may also be  affected.

      Direct  effects  of ozone on characteristic behavior  of J>  annosus
in culture and during wood disk colonization were  investigated.   Growth
chambers were used  to control other environmental  factors.   Five character-
istics of _F. annosus potentially affected by ozone were  studied:   linear
growth, conidial production,  conidial germination, colonization  of freshly-
cut pine discs and  adaptability over  successive generations.

      Four San Bernardino Mountain isolates were used to  study  ozone effects
on J[. annosus linear growth  rate and  conidial production.  Ten-day-old
cultures were fumigated at four ozone concentrations (0.05 - 0.45 ppm = 98
- 88z ug/m3) for three days.  Growth  and conidial  production were determined
after fumigation.   As ozone  dosage increased, growth rate and  conidial
production decreased (Table  40).  Significant differences in growth rate
between the  controls and ozone-exposed culture occurred  for  ozone concentra-
tions of 0.10 ppm (196 ug/m3) or greater.  Differences in conidial produc-
tion  occurred at all ozone concentrations.

      Experiments were conducted with  2 isolates to ascertain direct ozone
effects on conidial  germination.  Spores on water  agar petri plates were
fumigated with ozone at three concentrations for various time  periods.
As ozone dosage increased, percent germination generally decreased, (Tables
41 and 42) with the  greatest  change at higher ozone levels,  such as 0.45 ppm
for 4 and 8 hours (4301.7 and 9544.2 ug/m3/hr).

      In another experiment to determine  ozone effects on conidial ger-
mination and germ tube extension, conidia were fumigated for 12  hours
at concentrations of from 0.09  - 0.72 ppm (176.4  - 1411.2 ug/m3).  As
ozone dosage increased, germination decreased (at  0.72 ppm,  no conidia
germinated), average germ tube  length decreased,  and percentage  of germ
tubes with branches  was less  (Table 43)•

     Two JF. annosus  isolates  were used to evaluate effects of  ozone on
the colonization of  wood discs.  Discs inoculated  with conidia were fumi-
gated at four concentrations  of  ozone for seven days.  Extent  of coloniza-
tion by J?. annosus over the  surface of discs were  determined after ozone
exposure.  Colonization of discs exposed to ozone  was significantly less
than the controls at ozone concentrations of 0.11  ppm (16868.2 ug/m3/hr)
and greater  (Table 44)•

     Two F. annosus  isolates  were used to investigate adaptability to
ozone through a number of successive  generations.  J?. annosus  conidia
were ozone-fumigated for three  successive generations to ascertain whether
germination percentages changed after each fumigation.   No significant
differences in conidial germination occurred during the  sequences of ozone

                                    124

-------
TABLE 40.  INFLUENCE OF OZONE ON THE LINEAR GROWTH  RATE  AND  CONIDIAL
           PRODUCTION OF FOMES ANNOSUS.

Ozone
Cone.
Isolate (ppa)
0.045
JL1 °a°
0.22
0.45
0.05
JP1 °-10
0.22
0.45
0.045
ci oao
0.22
0.45
0.05
PP1 °a°
0.22
0.45
Ozone
Dosage*
(ug/m3-hr)
2401.0
5634.0
11477.0
23814.0
2646.0
6350.4
11477.0
25401.6
2401.0
6350.4
13230.0
23814.0
2646.0
5634.0
13230.0
25401.6
Growth
Rate
(% of Controls)
77. 62
10.3
12.0
7,5
75.52
19.6
11.3
4.6
86. 82
42.6
18.9
14.8
82, 12
30.8
19.0
5.5
Conidial
Production
(% of Controls)
16.3
1.2
0.0
0.0
14.6
8.3
5.9
0.0
10.3
8.6
3.6
0.0
34.3
28.0
7.5
0.0
^•Cultures were fumigated 9 hours daily  for  3 days  (total  27  hours).

2Not statistically significant  (P=0.05) using  one-way  analysis  of
 variance comparing control and ozone fumigated  cultures.  All  other
 such comparisons were statistically different.
                                     125

-------
TABLE 41.  INFLUENCE OF OZONE ON CONIDIAL GERMINATION OF FOMES ANNOSUS  (ISOLATE:   JL1).
Ozone
Cone .
(ppm)
0.10
0.10
0.22
0.10
0.45
0.22
0.10
0.22
0.45
0.22
0.45
0.45
Exposure Ozone
Time Dosage
(hr) (ug/m3-hr)
1
2
1
4
1
2
8
4
2
8
4
8
183.9
396.9
444.7
749.2
888.0
931.6
1573.9
1770.6
2058.0
4167.5
4301.7
9544.2
Control
Ave. %
Germin . a Sxb
90.3AB
89. 6A
9 2. 2 ABC
94.6BC
89. 3A
91.7ABC
91.1ABC
93.4BC
91.5ABC
90.2AB
90.1AB
88. 8A
2.34
3.92
2.48
1.90
2.79
1.95
1.64
2.63
2.37
2.62
2.13
1.87
Ozone Fum.
Ave. %
Germin.3 Sxb
93. 4F
85.4BC
87. 1C
86. 4C
82.1ABC
85. 7C
83.8ABC
83.5ABC
79 . 7AB
78. 9A
57. 8E
29. ID
1.95
5.60
3.18
3.13
4.15
2.87
3.85
3.17
1.89
5.22
3.12
5.26
% of
Control
103.4
95.3
94.5
91.3
91.9
93.5
91.9
89.4
87.1
87.5
64.1
32.8
F
Value0
10. 2d
3.8e
16. Od
50. ld
20. 7d
29. 9d
29. 5d
57. 7d
151. 7d
40. ld
730. 7d
1143. 5d
aMeans followed by the same capital letter are not significantly different (P=0.05) using the
 Studentized Range Test for Multiple Comparisons.
^Standard Deviation
cBased on one-way analysis of variance comparing control and ozone fumigation mean germination
 values:
            Statistically Significant (P=0.01)
            eNot Statistically Significnat

-------
TABLE 42.  INFLUENCE OF OZONE ON CONIDIAL GERMINATION OF FOMES ANNOSUS (ISOLATE:  PP1).

Ozone
Cone.
(ppm)
0.10
0.10
0.22
0.10
0.45
0.22
0.10
0.22
0.45
0.22
0.45
0.45
Exposure Ozone
Time Dosage
(hr) (ug/m^-hr)
1
2
1
4
1
2
8
4
2
8
4
8
183.9
396.9
444.7
749.2
888.0
931.6
1573.9
1770.6
2058.0
4167.5
4301.7
9544.2
Control
Ave. %
Germin.3 Sxb
91. OB
91. 8B
73. 1A
90. 6B
88. IB
72. 4A
89. 4B
72. 1A
89. 6B
71. 4A
90. 2B
88. 2B
2.34
2.78
3.14
2.07
3.41
4.06
1.65
5.34
2.01
2.22
1.93
2.90
Ozone Fum.
Ave. %
Gerroin.3 Sxb
94. 1G
80 . 2EF
75.7DE
85. 2F
66.9.C
68.3CD
81.9.EF
55. 6B
56. 6B
40. 9A
54. 3B
35. 2A
1.95
4.47
3.31
6.68
4.01
5.91
4.72
5.93
2.17
13.00
2.79
2.53
% of
Control
103.4
87.4
103.4
94.0
75.9
94.3
91.6
77.1
63.2
57.3
60.2
39.9
F
Value0
6.6e
48. 6d
3.2§
6.0f
161. 9d
3.3g
22. 7d
42. 7d
1266. 7d
53. 5d
1118. 5d
1956. 7d
aMeans followed by the same capital letter are not significantly different (P=0.05) using the
 Studentized Range Test for Multiple Comparisons.

^Standard Deviation

cBased on one-way analysis of variance comapring control and ozone fumigation mean germination
 values:
           Statistically Significant (P=0.01)       Statistically Significant (?=0.05)
          ^Statistically Significant (P=0.025)
Statistically Significant

-------
     TABLE  43.   INFLUENCE OF OZONE EXPOSURE ON CONIDIAL GERMINATION AND GERM TUBE EXTENSION OF
                 FOMES  ANNOSUS.
NJ
00

Isolate


JL1




HB11



Ozone
Cone.
(ppm)
0
0.09
0.18
0.25
0.72
0
0.09
0.18
0.25
0.72

Ozone
Dosage
(yg/mP-hr)
0
2205.0
4145.4
5884.9
16993.2
0
2205.0
4145.4
5884.9
16993.2

Percent
Germination3
41.8
23.5
4.0
2.2
0
39.1
21.0
4.8
3.4
0

Ave. Germ
Tube Length*3
(mm)
0.077E
0.039CD
0.038CD
0.022B
OA
0.079E
0.04 ID
0.36CD
0.029BC
OA

S_c
X
0.051
0.020
0.025
0.018
0
0.047
0.020
0.025
0.026
0

Percent
BGT
20.0
7.0
11.2
4.0
0
13.4
2.6
5.6
4.0
0
      aAfter  12 hours  incubation
      t>Mean followed by the same capital letter not  significantly different (P=0.05)  using the
       Studentized  Range Test for Multiple Comparisons.
      GStandard Deviation
      ^Percent  of germ tubes with branches.

-------
     TABLE 44.  EFFECTS OF OZONE ON THE COLONIZATION OF PINE DISCS BY FOMES ANNOSUS.
NO
vO
Isolate

JL1


HB11

Ozone Ozone
Concentration Dosage
(ppm) (ug/m3-hr)
0.06
0.10
0.11
0.27
0.06
0.10
0.11
0.27
8736.2
12595.0
16868.2
31928.4
8736.2
12595.0
16868.2
31928.4
Disc Colonization
Control3 03 Exposed3
71. 1A
75. SAB
80.4BC
84. OC
79.0ABC
77.5ABC
76.2ABC
70. 8A
75. 6C
59 . 3AB
59.6AB
55. OA
73. 8C
67.9BC
62.6AB
57. OA
% of Control Fb
106.3
78.5
74.1
65.5
93.4
87.6
82.1
80.5
2.1
10. 7C
29. 2C
134. 6C
0.5
2,3
35, 6C
39.8°
     aGiven in percent—Means followed by the same capital letter are not significantly
      different (P=0.05) using Duncan's Multiple Range Comparison Test.

      F values based on one-way analysis of variance comparing control and ozone exposed percent
      colonization values.

     Statistically significant (P=0.01) ,

-------
fumigation  (Table 45).  Spores selected for their germination  ability in an
ozone environment did not give rise to colonies which  subsequently had
spores with improved germinative capabilities.

     In summary, ozone influenced certain cultural characteristics of
_F. annosus  under growth chamber conditions.  Growth rate,  sporulation,
spore germination and colonization of wood discs were  all  limited  by
ozone, with the most dramatic effects occurring at high dosages.   No
evidence of adaptation by the fungus to an environment with  ozone  was
found.  Sensitivity of the fungus to ozone may have little effect  on
epidemiology under field conditions since the pathogen, because of its
occurrence within host tissues and because it commonly sporulates  at
night and during moist periods, is not often exposed to the  gas during
its life cycle.

Epidemiological Model

     Development of a model to simulate _F. annosus behavior  in a pine
ecosystem impacted by photochemical air pollution was a goal of this
research.   From such a model, prediction of potential disease  buildup
and future  losses might be made.

     The model predicts tree mortality expected in stands  subjected to
high dosages of photochemical air pollution relative to stands not im-
pacted.  Experimental data and regression equations were used  at various
disease cycle stages to quantify air pollution effects.  The model pre-
dicts that _F. annosus root disease would be expected to increase at a
rate 7 times greater than in a stand with no injury.  For  example,  the
model indicates that within 50 years of initial stump  infection, 31 times
more trees would die from JF. annosus infection in stands severely  injured
by air pollution (average rating = 14) than in stands with no  injury.
This prediction needs verification by further experimentation  and/or testing
in the field.

Disease Survey

     Results of the surveys of 1974 and 1976 on all vegetation plots
were keypunched and input to the data management system.   A  print-out
was made of the discrepancies, if any, in disease observation  of each
tree between the two years.  Each discrepancy was screened to  determine
if:

     1)  An observational error was made (a highly improbably  change
         indicated this).

     2)  An incorrect code was recorded for a good observation.

     3)  An actual change in disease status was observed.

     In the case of (1) and (2) a certain number of trees  required re-
checking in the field.  This was done during the early summer  of 1977.
The list of corrections have not yet been input to the computer data file.

                                    130

-------
TABLE 45.  INFLUENCE OF OZONE ON FOMES ANNOSUS CONIDIAL GERMINATION THROUGH SUCCESSIVE GENERATIONS
OF
Isolate
Designation
JL1
JL1-1A
JL1-1B
JL1-2A
JL1-2B
HB11
HB11-1A
HB11-1B
HB11-2A
HB11-2B
EXPOSURE .
Ozone
Concentration
(ppm)
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10

Cumulative
Dosage
(ug/m3-hr)
2353.
4704.
4704.
7056.
7056.
2353.
4704.
4704.
7056.
7056.

Percentage
Controls
40. OA
40. 6A
40. 2A
39. OA
42. OA
39. 4A
39. 2A
40. 4A
36. 2A
35. 6A

Germination*
03 Exposed
27. OA
22. 4A
26. 6A
27. 4A
27. 8A
23. 4A
24. 2A
24. 8A
22. 4A
25. OA

Test
Gl
G2
G2
G3
G3
Gl
G2
G2
G3
G3
*Within each category, means followed by the same Capital letter are not significantly
 different (P=0.05) using Duncan's Multiple Range Comparison Test.

-------
CAUSE AND EXTENT OF TREE MORTALITY

Introduction

     Tree mortality is an important regulator of stand composition  and
density.  It also plays a major role in forest succession.  The  study
reported here was initiated in 1976 to determine the cause and extent
of tree mortality in the mixed conifer forest of the San Bernardino Moun-
tains. The focus of the study was to identify the major causes of mortality
and relate their occurrence to the incidence of oxidant pollutant injury  to
the forest.

Research Objectives

1.   To identify the causes of mortality of the conifer species  in  the
     mixed conifer forest of the San Bernardino Mountains.

2.   To determine the extent of mortality in these conifer species.

3.   To relate mortality to the incidence of air pollution injury to
     the forest.

Literature Review

     The general impact of forest insects, pathogens, and oxidant air
pollutants on the forests of the San Bernardino Mountains has been  re-
viewed by Wood (1973) and Miller and McBride (1973).  Observations  of
tree mortality and insect occurrence on the 18 permanent plots established
to observe air pollution injury on forest trees in the EPA study have
been reported by Miller (1977) and Dahlsten et al (1974).  The major insect
pests, western pine beetle (Dendroctonus brevicomis Le Conte), mountain
pine beetle (D_. ponderosae Hopkins) and Jeffrey pine beetle (I),  jeffreyi
Hopkins), and some aspects of their population dynamics in the San  Bernar-
dino Mountains have been discussed in relation to oxidant air pollutants
by Dahlsten et al (1977).  James et al (1977) has reported several  experi-
ments which explored the relationship between the pathogenicity  Fomes
annosus (Fr.) Cke. and oxidant air pollutants.

     All of these studies have focused on selected plots within  the San
Bernardino Mountains which were established in an attempt to determine
the extent and cause of tree mortality as well as air pollution  injury
(McBride, 1974; Dahlsten et al, 1974; Cobb et al, 1974).  The plots unfor-
tunately were not large enough to sufficiently measure the extent and cause
of tree mortality.
                                    132

-------
Materials and Methods

     The method used to determine  the  cause  and extent of tree mortality
was developed by Byler, Hart  and Wood  (1976).  In brief,  the method involved
the following steps:

     1.  Random selection  of  plots,  of approximately 285 acres each,  to
         provide a  15% sample of the mixed conifer and yellow pine forest
         types in the San  Bernardino Mountains as defined by the U.S.F.S.

     2.  Aerial photography of each  plot  using Ektachrome MS film.

     3.  Photointerpretation  of each plot to identify and locate all  dead
         trees.

     4.  Ground check of a sample  of dead trees reported by photointer-
         preter.  This survey was  designed to  check the  accuracy of the
         photointerpretation  and to  determine  the cause  of tree mortality.

     5.  Measurement of stand conditions  on  a  sample of  the ground check
         plots to determine stand  and  site conditions and pests.

     6.  Calculation of extent  of  mortality  by each pest complex.

Laboratory Analysis Procedure

     Verification of the field  identification  of  pathogens was made on
cultured samples of root material.   This  root  material was plated  on
agar and incubated for 14  days  under sterile conditions  in the laboratory.
Identification of fungal species was based on  the appearance of fruiting
bodies and/or hyphal configurations  examined under the microscope.

Results and Discussion

     A total of 220 dead trees were  inspected  during the ground check
portion of the study.  It  is  estimated, by projecting this sample  to  the
entire forest, that 11,243 trees died  during 1976.   An overall summary
of the mortality percentages  is shown  in  Table 46.

     Specific pest complexes  were  identified for  the major forest  species.
Only Jeffrey pine and white fir are  reported here because an insufficient
number of mortality centers for the  other species (Table 47).   Examination
of the data indicate that  an  overwhelming number  of trees have succumbed to
the combined attack of disease and forest insects rather than a single pest
species.  The combined effects of  the  Jeffrey  pine beetle and dwarf mistle-
toe were the most common cause of  mortality  in Jeffrey pine while  the fir
engraver beetle and Fomes  root and butt rot  were  most common in white
fir.

     The correlation between  tree  mortality  and air pollution injury  was
much lower than anticipated (Table 48).   The data collected reflects  the
distribution of the samples rather than what is believed to be the actual

                                     133

-------
TABLE 46.  TREE MORTALITY BY CAUSES AND  SPECIES  IN THE  SAN  BERNARDINO
           MOUNTAINS  IN  1976.
Mortality  Cause
                                  Percent of Total Mortality
   Jeffrey Pine
       Combined
   Jeffrey Pine and
      White Fir
                                                             White Fir
Pathogens and
Insects
Insects
Pathogens
Other

61
34
1
4

70
21
1
10

78
9
3
8
TABLE  47.  TREE MORTALITY BY PEST COMPLEX AND SPECIES IN THE  SAN  BERNARDINO
           MOUNTAINS IN  1976.
  Pest
Complex
                               Percent of Total Mortality
Jeffrey Pine
    Combined
Jeffrey Pine and
    White Fir
White Fir
Root Disease
and Insects             26

Root Disease,
Mistletoe and
Insects                 18*

Mistletoe and
Insects                 22*

Insects alone           19
Insects and
Mechanical Injury       15

Pathogens alone         1

No insect, path-
ogen,  or injury
apparent                4
                        39


                        16

                        15
                        14

                         7
                         1


                        10
                           53


                           19**


                            6
                            9

                            0
                            3
*Dwarf Mistletoe
**Tree Mistletoe
                                     134

-------
relationship between air pollution injury and tree mortality.   It  does
not fit the trends observed by Miller  (1977) on  the  18 permanent plots
established to observe air pollution injury.  The samples  selected for  study
fell outside of the zone of severe air pollution.  Most  of  the  samples  were,
in fact, in areas of slight oxidant concentration.   Subsequent  surveys  of
tree mortality will be stratified in order to obtain samples from  zones
of high oxidant concentration.
TABLE 48.  PERCENT OF MORTALITY CENTERS IN THE MIXED CONIFER AND YELLOW
           PINE FORESTS IN RELATION TO OXIDANT INJURY.
	Oxidant Injury	          Percent of Sampled Mortality Centers
                                    Mixed Conifer            Yellow Pine
Rating           Class                 Forest                   Forest
1- 8
9-14
15-21
22-28
29-35
sen-
very severe
severe
moderate
slight
very slight
no visible symptoms
0
1.1
3.4
8.0
20.4
67.1
0
0
7.9
7.9
38.6
45.6
                                     135

-------
EFFECTS OF PHOTOCHEMICAL AIR POLLUTION ON FOREST TREE  SEEDLING
ESTABLISHMENT

Introduction

     The purpose of this subproject is to investigate:   (1) the  rate  of
tree seedling establishment in forest stands exposed to  different  levels  of
photochemical oxidants; (2) the influences, both direct  and indirect, of
oxidants on the establishment of ponderosa and Jeffrey pine seedlings;  (3)
the individual and concurrent joint effects of biotic and physical factors
on seedling establishment, and (4) the influence of an oxidant gradient on
these biotic and physical factors.

     The investigations initiated in the 1976-1977 period included studies
(1) to determine the rate of seedling establishment in forest stands  along
an oxidant gradient; (2) to evaluate the individual and  joint effects of
vertebrates, arthropods, and pathogens on seeds and seedling establishment;
(3) to evaluate the interactions between litter depth and type and the
effects of the biotic agents; (4) to determine the relationships between
litter decomposing organisms and those organisms causing loss of seeds and
seedlings; (5) to further analyze data from studies initiated during  the
1975-1976 contract period.

     The following is a summary of studies and results obtained  during
the 1976-1977 period.

Pathogenicity Test of Litter Organisms

     This test was designed to determine which fungi isolated from pine
litter were pathogenic to pine seeds and seedlings.  Seven fungi were
tested in the initial study:  Mucor, Curvularia. Aureobasidium, Alternaria,
Ulocladium, Penicillium, and Fusarium roseum.

Summary of Test Procedure—

     Inoculum of the seven fungi was grown on a litter substrate in dishes
for 12 days; 10 seeds then planted in each plate.  The plates were watered
and incubated in a cool environment (4.4'C) for 20 days; then they were
removed during the day to room temperature (IS^C) and returned to  the
cool environment at night for an additional 57 days.  They were watered
periodically, and dying seedlings were removed and isolated from the  culture
dishes.  At the conclusion of the study remaining seeds, seedlings, and
samples from the surface and subsurface of the litter  substrate  were  iso-
lated from each of the substrate dishes.
                                     136

-------
Test Results—

     A summary of the results for each fungus and the control are  given
in Table 49.  Seed germination was observed to be lower  in all  inoculated
plates than in the controls,.but it was only substantially lower for
the Penicillium plates.  Although the fungus was not isolated from a high
percent of the nongerminated seeds.

     The Alternaria plates were observed to have a substantially higher
number of diseased seedlings than the control; however,  Alternaria was
only isolated from about half of the diseased seedlings.  The 31.7% diseased
controls contained some organisms used in the other plates, indicating some
contamination may have occured between plates.

     Final conclusions from this experiment will be drawn after a  statis-
tical analysis is performed on the data.  Initial results indicate that
Alternaria was pathogenic on seedlings.  The other organisms showed an
effect on the seedlings, but because of the contamination of the con-
trols, we have not yet been able to draw conclusions about these organ-
isms.
TABLE 49.  PATHOGENICITY TEST SUMMARY
                # Plates
                (10 seeds
Test Organism   per plate)
Control
Mucor
Curvularia
Aureobasidium
12
6
6
6
Alternaria
Ulpcladium           6
Penicillium          6
Fusarium roseum      5
                                 Mean % Di-
             Mean %    Mean %    seased Seed-
  Mean %     Seeds     Seedlings lings with
Germination  Diseased  Diseased  Test Organism
95.8
80.0
88.3
83.3
91.7
85.0
60.0
82.0
0.8
3.3
1.7
3.3
8.3
6.7
5.0
0
31.7
25.0
26.7
38.3
45.0*
33.3
21.7
32.0
NA
93.3
55.8
53.3
49.6
92.2
93.8
74.7
                                    137

-------
Field Study,  1975-1976.

     The field study of 1976-1976 involved  4 vegetation plots  located
across the air pollution gradient, CAO, BF, HB, and HV.   Seeds were planted
in 0.4 meter  square areas  (mini-plots) with 49 seeds per  square in these
plots.  The mini-plots were either open or  screened, and  some  were litter
covered while others were bare mineral soil.  The basic design is  given on
page 188 of the  1975-1976 EPA Progress Report.

Results—

     Predation by birds and small mammals was extremely heavy  on seeds
in unscreened miniplots.  Only about 3% of  the seeds on unscreened plots
without litter survived to germination, and only 1% of the  seeds survived
to germinate  on  the unscreened plots with litter.  It appears  that animals
were primarily the cause of the mortality in the plots without litter,
while in the  plots with litter a combination of animal predation and seed
pathogens depressed survival even further.  There was no  significant differ-
ence among the four vegetation plots because of high variability among
the individual mini-plots.  Results for the screened mini-plots are given
in Table 50 and  51.  Germination as checked in May 1976 averaged about 62%
of the seeds  in  the screened plots without  litter had survived to  germinate,
while only about 20% of the seeds in the screened, plots  with  litter had
survived (Table  50).  Two-way analysis of variance among  the 4 vegetation
plots and between litter and no litter showed the differences  to be non-
significant among the vegetation plots but  the differences  were highly
significant (.001 level) between litter and no-litter plots.   It is believed
that seed pathogens found mostly in the litter were primarily  responsible
for this difference in germination rate.

     In July  of  1976 the numbers of living  seedlings remaining alive had
been greatly  reduced from the number germinated.  Table 51  gives the means
per mini-plot of surviving seedlings in July.  Compared to  the number
germinated, about 16% had survived in the no-litter mini-plots while 5% had
survived in the  litter mini-plots.  Seedling damping-off  pathogens vrere
believed responsible for part of the mortality, especially  in  the  late
spring, but by mid-summer drought was the major factor.   The driest plot,
HV, had no surviving seedlings by July in the no-litter mini-plots, and
only 1 in the litter mini-plots.  The damping-off fungi appeared to have
a greater effect on the seedlings in the litter mini-plots.  An analysis
of variance showed the difference between litter and no-litter to  be
highly significant (.001 level) while the differences among vegetation
plots was marginally significant (.10 level).  The vegetation  plot differ-
ences were due primarily to the high mortality at HV because of drought.  No
smog effects were apparent; if smog effects were present, they were obscured
by the rainfall differences between plots and the small sample size.

     The results of this study pointed to the need of an  expanded  study
with higher sample numbers and the addition of a control  variable  for
soil moisture content.
                                     138

-------
TABLE 50.  1975-1976 SBNF SEEDLING ESTABLISHMENT FIELD  STUDY, GERMINATED
           SEEDS, MEANS PER SCREENED MINI-PLOT IN MAY,  1976.  FORTY-NINE
           SEEDS WERE PLANTED IN EACH MINI-PLOT.
Vegetation Plot
CAO
BF
HB
HV
Mean of Plots
No Litter

Litter
30.0   34.0   32.0   26.0

13.0   12.0    7.1    6.4
                          30.0

                           9.7
TABLE 51.  1975-1976 SBNF SEEDLING ESTABLISHMENT FIELD  STUDY,  SURVIVING
           SEEDLINGS, MEANS PER SCREENED MINI-PLOT AS OF JULY,  1976.
           FORTY-NINE SEEDS WERE PLANTED.
Vegetation Plot
CAO
BF
HB
HV
Mean of Plots
No Litter

Litter
5.9    8.1    5.6    0

0.1    1.3    0.3    0.1
                          4.9

                          0.4
                                    139

-------
Field Studies,  1976-1977.

     The  seedling establishment field  studies  of  1976-1977  consisted of
one  primary  study (Seedling Air Pollution  Study - SAPS)  and two supple-
mentary studies (Supplementary Seedling-Animal Study  -SAAS. and Supple-
mentary Litter-Seedling  Study - SLSS).

     The  objective  of  the  SAPS study was to determine air pollution and
moisture  effects on the  survival of young  seedlings in the  San Bernardino
mountains.   The first  supplementary study  (SSAS)  was  designed to determine
relative  rates  of seed predaton by small mammals  vs.  birds  on the forest
floor.  The  second  study (SLSS) was designed to test  the potential for seed
survival  and germination in different  types of litter under natural condi-
tions .

Primary design  (SAPS)—

     Healthy Jeffrey pine  seeds were planted on four  vegetation plots,
two  (BF and  CAO) with  relatively high  air  pollution and  two (HV and HB)
with low  air pollution.   Each vegetation plot  was  subsampled with four
subplots, and each  subplot had 16 mini-plots consisting  of  screened 15 x 15
inch frames  planted with 49 seeds per  frame.   Twelve  of  the 16 mini-plots
had  the litter  layer completely removed prior  to  planting,  while 4 were
placed on undisturbed, natural pine litter.  One  of the  4 subplots on each
vegetation plot was chosen as a "water" subplot.   These  subplots were
periodically (once  a month or as necessary) watered with an amount of water
necessary to simulate  a  heavy rainfall month.  This amount  was defined as
the  mean  rainfall for  each specific month  over the past  25  years at Big Bear
Lake Fire Station,  plus  one standard deviation.   On each field trip to the
study plots  (done every  3  to 4 weeks during the spring,  summer,  and fall of
1977) soil samples  were  taken from 4 mini-plots (two  with litter and two
without)  on  each water subplot, and 2 mini-plots  (one litter and one no-
litter) for  each of the  other subplots.  Moisture  content of the soil
from these samples  was determined in the lab.  On  each field trip data was
taken on  surviving  seedlings and dead  seedlings,  and  all dead seedlings were
removed.  Possible mortality causes and plot disturbances were noted.
Seedling  data was not  taken on those mini-plots used  for soil sampling.

     At the  end of  the field season, analyses  will be made  on seedling
mortality by plot,  considering the variables of soil  moisture and rain-
fall, for each  3 to 4 week period.  Rainfall and  oxidant exposure data
from other projects will also be used  to try to relate periods of greatest
mortality with  corresponding environmental conditions.

Supplementary design (SSAS)—

     Four subplots were  chosen on pine litter  covered areas of CAO, and
6 mini-plots  (15 x  15 inches) were planted with 49 sound Jeffrey pine
seeds each in the fall of  1977.  On each subplot,  three  of  the mini-plots
had  the litter  layer completely removed, while on the other three the
litter was undisturbed.  On each litter type,  one  plot was  completely
screened to  exclude all  vertebrates, one was closed except  for a narrow gap

                                    140

-------
at ground level, which would allow  small mammals  to  enter,  and one plot
was completely open.

     The following table is a summary of the  percent of  seeds  that sur-
vived to germinate:

                               No Litter       Litter

Completely screened              20.9%          3.6%

Screened top, open base           0.5%          0.5%

Open top                          1,5%          0%

     The seeds used in this test showed an  85% germination  rate under
sterile lab conditions.  The results indicate that in the case of  the
completely protected mini-plots, 65% of the additional mortality was due
to pathogens and/or arthropods.  In the case  of the  protected  litter
mini-plots, an additional 16% mortality occurred.  The results from the open
base mini-plots and open top mini-plots indicate nearly  complete mortality
(the total number of seeds surviving to germination  on these mini-plots was
only 5 of 784, or 0.6%).  Since the open base mini-plots had about the  same
survival as the open top mini-plots, it can be concluded that  small mammals
were probably most important in seed predation.  Broken  seed coats and  scat
(mouse droppings) were observed around most of these mini-plots.

Supplementary design (SLSS)—

     On one vegetation plot (CAO),  seven different major types of  litter
cover were selected:

     1.  Jeffrey light litter (with PP and  Oak)
     2.  Jeffrey heavy litter (with PP and  Oak)
     3.  Oak litter (with PP and JPO
     4.  Mixed Jeffrey - Ponderosa  - Oak litter
     5.  White Fir (with JP, PP and Oak)
     6.  Bare ground with some JP,  PP needles
     7.  Bare ground with some grass

     Two areas were found for each  of the above litter types.   On  each
areas, 6 mini-plots, (.4 x .4m) were established at random and marked
by stakes at two corners.  On each  of three of the mini-plots  49 seeds
were placed, while the other three  received 16 seeds each.  Litter depth
and % cover were recorded for each  mini-plot.

     The seeds were planted on November 3,  1976.  All the plots were
rechecked on April 20, 1977.  This  check showed no visible  seedlings on
any of the mini-plots.  Further checks will be made  in the  fall of 1977
to see if any seedlings have become established.

Miscellaneous field activities—


                                    141

-------
     During the fall of 1976 cones were collected from several healthy
Jeffrey pines in the SBNF.  No cone trees with significant smog  damaged
were found at this time.  Seeds from the healthy tree cones were used
in the studies described above.

     Cones were again collected in August, 1977.  This time both smog
damaged and healthy trees were located (both Jeffrey and Ponderosa  pines).
The cones for each tree were measured, and the numbers of cones  and yield
of seeds were recorded for each tree.  Tests of the cone data, seeds,
and seedlings grown from them in the greenhouse will be made to  determine
if there is a relationship between cone, seed, and seedling quality land
the degree of smog damage of the parent tree.

     A pilot study was made in the fall of 1976 to determine if  seed drop
under natural conditions could be determined on the vegetation plots.
Very few cone bearing trees were found on any of the plots.  A few  trees
were found on CP, and a 30 x 30 meter subplot was established under these
trees.  Fifteen seed traps were placed randomly in this subplot, but no pine
seeds were recovered.  It is believed that seed drop was too small  to be
detected by the density of traps used.
                                    142

-------
CONE AND SEED PRODUCTION FOR DOMINANT CONIFER TREE REPRODUCTION

Introduction

     The two primary objectives of this study are:

     1)  To test the hypothesis that cone crop abundance and frequency
         in ponderosa and Jeffrey pines are affected by ozone injury;
         and

     2)  To describe the probability that a tree with specified charac-
         teristics will produce a cone crop in a given year.

     The rational for the cone study and the study design, methods, and tree
characteristics used to classify the trees have been previously reported
(Luck, 1977).

Recent Research Progress

     Status in acquiring plot data other than the cone counts is presented
in Table 52.  A total of 9 pots have been completed, the data punched
on 1MB cards and the values on those cards verified against the origi-
nal data.

     These data sets have been forwarded to the data management group
for entrance into the data bank.  The other ten plots are in various
stages of completion.

     The annual cone crop data, 1973-1976, obtained by visually counting
cones within the tree crowns, have been punched on IBM cards, verified
and forwarded to the data management group for entry into the data manage-
ment system.  These counts have been reported by Luck (1977). The 1977 cone
crop is currently being counted.  Ground counts have been made of the whole
cones, and of those eaten by squirrels during the 1976 cone crop. Insect
damaged cones and cones which aborted before maturing were placed in rearing
containers.  Cones from the 1975 crop that were placed in rearing are being
processed.  The insect species assciated with a particular type of damage is
being identified so that damaged cones can be identified in the future, even
though the insects causing; the damage are absent.  Curation of insects
reared from damaged cones obtained from the 1974 and 1975 cone crops has
been partially completed.  We are waiting for the return of some specimens
sent to several specialists for identification.

     As an example of the way in which the data is being analyzed, Figures
29, 30, 31 and 32 present annual cone crop data (combined visual counts for
the 1974-76 cone crops) for ponderosa and/or Jeffrey pine trees on three

                                    143

-------
TABLE 52.  STATUS IN ACQUIRING PLOT DATA ON THE 19 STUDY PLOTS LOCATED
           IN THE SAN BERNARDINO MOUNTAINS OF SOUTHERN CALIFORNIA.
Data Type
Plot
ID
UCC
HB
COO
SF
CP
DW
SV.I
TUN2
GVC
BP
NEGV
HV
DL
CAO
CA
BF
SC
SCR
BL
Tree
age
X
0
X
X
—
X
X
X
X
X
X
X
X
X
X
0
0
X
—
Crown
height
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0
0
0
0
Tree
height
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
0
0
X
0
Keen
class
X
0
X
X
0
X
X
X
X
X
X
0
0
0
—
0
0
0
0
Crown
class
X
X
X
X
—
X
X
X
X
X
X
X
0
X
X
—
0
0
X
Data verified and entered
into data management system
X

X
X

X
X
X
X
X
X








Key:  X  = complete
      — = partially done
      0  = not started
                                    144

-------
    II
         II

        k
         41
              NEGV
              No. of Trees 65
              X Age 162
                       12
              GVC
              No. of Trees 63
              X Age 148
I
                         22
                                 15
                                       13
                                           13
                                          SF
                                         I 2
                                                44
                                  35
                                                86
                                               70
                                                       V/.
                                                        66
                                 No. of Trees 113
                                 X Age 81
                                                        18
   <30  50   70   90  110   130  150  170   190  >200      % CONES

                          AGE (years)
Figure 29.  Age  class  distribution of ponderosa and/or  Jeffrey pine
            trees  on three plots.  Solid bars = %  of  trees in each age
            class; Open bars = % of cone bearing trees  in each age
            class; Horizontal hashed bars = % cones produced per age
            class; Diagonally hashed bars = % of the  combined three
            plot cone  production.
                                145

-------
                     40-
                                       NEGV
                                       No of Trees 65
                                       X Age 162
                           28
                          sT
 28


11
                      96

                                       GVC
                                       No. of Trees 63
                                       X Age 148
                                                      66
                           2728


                         d
                                       SF
                                       No. of Trees
                                       X Age 81
            15
                113
                    DO    CD    II     10    IS     SS

                              CROWN CLASS
                      % CONES
Figure 30.  Grown class distribution of  ponderosa and/or Jeffrey pine trees
            on three plots.  Solid bars  = % of trees in the six specified
            crown classes; Open bars = % of trees bearing cones in a crown
            class; Horizontally hashed bars = % of cones with respect to
            crown class of cone bearing  trees; Diagonally hashed bars = % of
            the combined three plot  cone production.  DD-Dominant; CD-Co-
            dominant; Il-Intermediate; 10-Intermediate open; IS-Intermediate
            suppressed; and SS-suppressed.
                                     146

-------
                 NEGV
                 No. of Trees 65
                 X Age 162
              14
                                                               16
                 GVC
                 No. of Trees 63
                 X Age 148
              15
                                              66
                                             1
              43
SF
No. of Trees 113
~ Age 81
  13
                   I
                                                               IB
              15    25   35   45   55   65  75   85   >90

                                  DBH (cm)
                                            %CONES
Figure 31.  Diameter class distribution of ponderosa and/or Jeffrey pine on
            three plots.   Solid bars = % of trees in specified diameter
            classes; Open bars = % of trees in a diameter class that produced
            cones; Horizontal hashed bars = % of cones with respect to diam-
            eter of cone bearing trees; Diagonally hashed bars = % of the
            combined three plot cone production.
                                    147

-------
               NEGV
                GVC
                No. of Trees 63
                X Age 148
                   16
                   I
                                   32
                SF   No. of Trees 113
                     X Age 81
                          13
        20

       MJ 10
                  No. of Trees  65
                  X Age  162
                                         27
                                      JD
                                         64
              45
                                                24
                                               13
                                                       16
                             66
                            i

                                                       18
15     25     35     45

  TREE HEIGHT (m)
                                                    % CONES
Figure 32.  Height class distribution of ponderosa and/or Jeffrey pine on
            three plots.  Solid bars = % of trees in the five height
            classes;  Open bars = % of trees in each height class that pro-
            duced cones; Horizontally hashed bars = % of cones produced
            with respect to height of cone bearing trees; Diagonally hashed
            bars = % of the combined three plot cone production.
                                    148

-------
plots, Northeast Green Valley  (NEGV), Green Valley  Creek  (GVC)  and Sky
Forest (SF).  The trees from these plots are  classified into  groups based
on 20 year age intervals (Fig. 29),  crown  classes  (Fig; 30)  (see  Luck,
1977, for a description of the crown classes),  10 cm  intervals  in diameter
(at breast height) (Fig. 31), and 10 m intervals in tree  height (Fig,
32).

     The first point to note is that NEGV  and GVG have older  trees pres-
ent on the plots (Fig* 29).  Trees 200 years  or older are absent  from  SF.
The lack of older trees on SF is due to the logging history on  that plot.
The 200+ year class of trees, however, produces more  than its share of
cones.  For example, trees in the 200* year class at  NEGV represent 35  per
cent of the trees bearing cones and  bear 44 per cent  of the cone  crop  (total
number of cones) while they make up  only 29 per cent  of the stand.   This
pattern is even more pronounced at GVC. Trees in the  200* year  class repre-
sent 45 per cent of the stand but account  for 70 per  cent of  the  trees
producing cones and bear 85 per cent of the cone crop.  The older trees,  age
classes 130, 150, 170, show the same trend at SF.   Clearly the  older trees
are the greatest source of cones and, thus, of  seed as well.

     Crown class is perhaps the most important  tree characteristic linked
to cone production (Fig. 30).  Clearly, dominant trees are the  greatest
source of seeds because they contribute more  than their share of  cones
and seeds based on their representation in the  stand.

     The larger diameter (Fig. 31) and taller trees (Fig. 32) are also
disproportionate contributors to the seed  crop.  However,  both  these
variables are correlated with tree age and crown class (dominant  trees
are usually the taller and larger diameter ones); hence,  these  variables
may explain little or no variation in crown crop abundance when they
are corrected for covariance with other variables.

     Ozone damage classes were not represented  by a sufficient  number
of trees on these plots to permit assessment  of the effect of this variable
on cone production.  It should be used to  reclassify; tree only within  a
crown class or age class.  For example, given that  a  tree belongs to the
dominant crown class, how does ozone damage effect  the cone crop  produced  by
that class?  Thus, before the effects of ozone  damage can be  assessed,  the
pattern of cone produciton as influenced by a number  of other variables,
such as age and crown class, needs to be determined first.  Furthermore,
since cone crops vary substantially  in abundance, a number of years of  data
are essential before the pattern of  cone production can be revealed.
                                     149

-------
LITTER PRODUCTION SUBSYSTEM

Introduction

     Oxidant air pollution stimulated development  of  the  abscission zone and
thus induced heavy defoliation, especially on yellow  pines.   The rate of
needle fall in  the early stages of  impact increased markedly and later
decreased to zero as  the tree was increasingly defoliated and finally
killed.  The question here is how does  the change  in  rate of needle and
branch mortality and  the consequent  change in addition  of needle and branch
litter affect the nature and amount  of  organic litter covering the soil?
Further, are there consequent effects upon:  the soil moisture and tempera-
ture; seed germination; seedling survival; and the forest floor as a habitat
for micro- and  macroorganisms?  Since defoliation  also  results from other
plant damage (pathogens and insect  attack) there is also  a question as to
the effect of air pollutants in relation to  other  causes  of  defoliation.

     As  foliage of the canopy is damaged and the density  is  altered, it is
postulated that both  chemical composition and amount  of crown drip falling
on the soil during periods of precipitation  will be altered.  Here the
question is, do changes in crown drip and throughfall precipitation affect
the soil and the forest floor; and  if so, are there consequent effects?

Research Objectives

     The objectives of this project  relative to oxidant injury gradient
were as  follows:

     1.  Continue and expand the measurement of litter  production from
individual trees on all major plots.

     2.  Continue to  measure nutrient content of needle fall and accumulate
litter under individual trees variously affected by oxidant  air pollutants.

     3.  Verify with  expanded data  the  effect of oxidant  pollutants on
needle size in  the litter.

     4.  Measure accumulated litter  under all trees for which data were
being collected.

Materials and Methods

Measurement of  litter production—

     Litter was collected as it fell on 46 cm square  screens under 50
yellow pines (P^. ponderosa, £- Jeffrey) on 16 plots.  Collections

                                     150

-------
from the screens were made in the late spring and late fall  in both  1975
and 1976, expanding the number over that of  1974 which was 39 trees.
Two to four screens were used under each tree depending upon crown size.
The litter was oven-dried, separated into needles and other  material  (twigs,
etc.) and weighed.  A large number of intact needle-fascicles were sepa-
rated, counted and weighed to obtain the mass per fascicle as a measure of
needle size. The total needle component of the litter was analysed for (N,
P, K, Ca and Mg).

     Needles were digested by the Johnson and Ulrich (1959)  procedure
in preparation for nutrient analyses.  Nitrogen was measured by a modified
Kjeldahl method from Black (1965) and phosphorous was determined by a
colorimetric method of Richards in a book edited by Jackson  (1960).  Atomic
absorption was used to measure the concentration of cations.

Results and Discussion

Pine Litter Production—

     Litter collected under pine trees during 1975 (47 trees) and 1976 (41
trees) on 18 major vegetation plots was analysed for total litter deposit,
total needle component and size of needle fascicles (mass per fascicle) in
relation to oxidant injury rating.  Needles were also analysed for content
of N, P, K, Ca and Mg.

     Amount of litter—Dry weight of needles collected during late autumn of
1975 and 1976 as related to oxidant injury score is shown in Figure 33.  The
increase in needle fall from relatively healthy trees (score >31) of 87.5
gm/m^ to 256.5 gm/m^ from severely affected  trees (score 8 to 15) was
statistically significant (P=.01) as is the  decline to 125.7 gm/m^ from
severely affected trees (score 8 to 15) is statistically significant  (p =
.01) as is the decline to 125.7 gm/m^ from very severely affected trees
(score 0 to 7).  The total amount of litter  of all kinds collected on the
screens followed almost exactly the same pattern, with litter other than
needles (twigs, branches, etc.) contributing about 33 percent of the
total.

     An important impact of oxidant air pollutant was to increase the litter
fall during the period of tree injury from the onset of clearly discernable
injury (score 30) until the tree died.  If the duration of this period is
assumed to be 8 years, then the litter fall would be almost  exactly double
that expected from unaffected pine trees.  After the first year following
death of the tree, needle fall is very small, but a vast increase in dead
branches (woody litter) may be added.  The net result is predictable,
namely that the amount of undecomposed litter on the forest  floor is
increased and thus the fire hazard increased during the injury period and
for a number of years following the death of the pine trees.

     In addition to the fire hazard created  by increase litter fall,  the
extra thickness of fresh loose litter created on the forest  floor can
be expected to have a marked detrimental influence on seed germination and
seedling survival during and following the period of oxidant injury  and  tree

                                    151

-------
    IB
     0>
     O
     CJ
     d

     UJ

     O
     UJ
     111
     z
     UJ
     z
     Q.
     U_
     O

     CO
     CO
     O
          70
60
          50
          40
30
20
           10
                  On-17
                              on=2?
                                                   Oh
                                  i
e
                      7          15         23
                        OXIDANT  INJURY SCORE
                                             31
Figure  33.  Total pine needle litter collected related to oxidant  injury
           score 1975 and 1976.
                                 152

-------
mortality:  this aspect of litter fall  is  being  investigated by others
on the project.

     Size of needles in the litter—The size  of  needles  as  measured in
grams per fascicle (cluster of 3 needles)  by  counting  and weighing a
large number of inteact fascicles separated from the total  litter collected
by the screens was determined for each  tree.   The data showed clearly that
as oxidant injury increased, size of  the pine needles  decreased.   The rela-
tionship was almost identical for 1975  and 1976  (difference between the
regression lines was not significant).   Least squares  linear regression for
the combined years is Nm = .01665 +  .0073  1R,  where Nm is mass  per
fascicle in g/m^, and IR is the oxidant impact rating  scaled with increas-
ing scores indicating decreasing injury.   The correlation r = .392 is very
highly significant (P = .001) although  there  is  considerable scatter about
the regression line (Fig= 34).  Considering the  usual  variability in biolog-
ical systems, it appears safe to conclude  that the oxidant  injury rating
system does reflect the general health  of  the pine trees.   Stratification of
the data by grouping into either high and  low rainfall plots or into pon-
derosa pine and Jeffrey pine does not give increasing  reliability, although
the correlation for the high rainfall plot was higher  (r =  0.536)  and the
regression line somewhat steeper (Nm =  0.145  + .00158  IR).   However, the
difference from the general equation does  not  appear to be  very significant.

Plant Nutrient Content of Needle Litter—

     The content of nitrogen (N), potassium (K),  phosphorous (P),  calcium
(Ca), and magnesium (Mg) in the needles collected on the screens  in 1975 and
1976 were determined and the relationship  with oxidant injury ratings and
needle size was analysed by linear regression.

     The content of N, P, and K in the  needles which fell tended  to increase
with increasing oxidant injury and declining  needle size.   Ca,  on the
other hand, decreased and Mg appeared to be variable,  but with  no con-
sistent trend.  It appears that as needle  mass decreased the carbohydrate
content and N, P and K content per fascicle remained about  constant so that
the concentration increased, a simple dilution relationship.   This can be
seen from the ratios given in Table 53  where  for  a 49-5 percent increase in
needle size in high rainfall plots, concentration of N decreased  by 50 per
cent.  The relationship was not quite as exact for phosphorous  (P),  and was
still less exact for potassium (K) especially in  the high rainfall plots.
The increase in Ca concentration with increasing  needle size appeared to be
explainable in that Ca is involved in carbohydrate in  cell  wall material so
the larger needles take in more calcium as the cells enlarge.   Relationships
between N content, versus oxidant injury score are shown in Figures 35 and
36; for K content in Figures 37 and 38;  for P in  Figure 39  and  40; and for
Ca in Figures 41 and 42.

Phosphorous in the Surface Soil—

     A discussion of phosphorous in the surface  soil is  included  here
as it is postulated that plant nutrients in the  surface soil may  be affected
by the nutrient content of the litter.   The soluble soil phosphorous in the

                                    153

-------
(JI
                  0.3
                 0.25
                 0.0
                              f
T
T
                          0 — 1978   H< 0.146 «• 0.00150
                                    r « 0.64
                          0--I976
                                   O"1978   t«0.168 1- 0.00076 F
                                            T» 0.392
                                   D —1976
                                                                      1
10       15       20     25       30
       OXIDANT INJURY SCORE (S)
                                                                              35
                                                 40
                                                 45
         Figure  34-  Size  of pine needles in  needle litter collected related  to oxidant
                      injury score.

-------
TABLE 53.  NEEDLE SIZE AND CONTENT ON N,  P,  K AND Ca IN NEEDLES COLLECTED
           ON SCREENS.
                             High       Ratio          Low        Ratio
                           Rainfall    Smaller/      Rainfall    Smaller/
                       	Plots       Larger        Plots        Larger
Oxidant Injury
    Score*	0	50	      0      50

Needle Size
gm/fascicle*           0.124   0.25     0.495     0.168   0.20     0.84

Percent N in
needles*               0.688   0.34     0.50      0.542   0.438     0.81

Percent K in
needles*               0.48    0.11     0.23      0.36    0.23     0.64

Percent P in
needles*               0.096   0.042    0.44      0.076   0.055     0.72

Percent Ca in
needles*               0.294   0.453    0.65      0.243   0.582     0.42
*NOTE:  0 oxidant injury score is very  severe  injury  and  50  is  a health
        tree with no evident injury.  The numbers  in  the  body of the
        table are average values derived from  regression  analysis.
                                     155

-------
Ul
               0.8
CO

3  0.7


H
z

UJ

ij>.6

|j_r Q
O o
               0.5
0.4
            O
            o
            §
            oc.
               0.3
                                8
                                           O
                                  a
                                                    O
                                 D
                           j	I
                                                          a
                                                               o
                                                                                 T
                                                                          T
                                                         O- 1975  N= 0.676 - 0.0067 S


                                                         Q-1976 r =0.51
                                                      J	I
                                   10       15       20     25

                                   OXIDANT  INJURY  SCORE  (S)
                                                      30      35      40      45
                                                                                  50
         Figure  35.   Nitrogen content of pine needle  litter related  to  oxidant injury

                      score—high  rain fall plots.

-------
Ui
           U|
           I  0.6

           Q.

               0.5~
O
O


ui  0.4
o
o
(C
               0.3
                              a
                                                                         "1        I       I

                                                                          O-I975 N =0.543- 0.0021 S


                                                                          D-1976  --0.43
                                          D
                                                       OO
                                  10      15      20      25

                                   OXIOANT  INJURY  SCORE  (S)
30      35     40      45
                                                                                      50
         Figure 36.   Nitrogen content  of  pine needle litter  related to oxidant  injury

                      score—low rainfall  plots.

-------
Ul

00
                0.6
            V)
            UJ

            a   0.5
            UJ



            £L
                0.4
                0.3
55   0.2


I
o
Q.

    O.I
                                    D
                                                       "1       I        I


                                                       O-I975 K= 0.48 -0.00739 S



                                                       D-1976 T=-0.59
                                    10
                                15       20      25     30

                             OXIDANT  INJURY  SCORE  (S)
35
40
        Figure 37.  Potassium content  of  pine needle litter related to  oxidant injury

                     score—high rainfall  plots.

-------
                0.6
Ul
VO
                                                                          I        I        I

                                                                    O-I975 K* O.36-0.0026 S

                                                                    D -1976 r = 0.65
                                   10
15       20     25
OXIDANT  INJURY SCORE
30
         Figure 38.   Potassium content of pine needle litter  related to oxidant injury
                      score—low rainfall plots.

-------
        0.02
                            10       \5      20      25      30
                                    OXIDANT INJURY SCORE  (S)
35
40
45
50
Figure 39.  Phosphorous  content of pine needle litter related  to  oxidant iniurv
            score—high  rainfall plots.

-------
o\

o
UJ
UJ
                  0.10
Q.
£   0.08

Z  *


g   0.06
CO
             I
             Q.
                  0.04
                  0.02
                                                                      I        I        I        I

                                                                      O-1975  P - 0.0761 - 0.000418 S

                                                                      D-1976  r=-0.52
                               O
                                 D

                                 O
                                 D
                                      no
                 I	I
                                                                   o
                                                      I
                                                                                   o
I	I
                                      10      15      20      25      30
                                             OXIDANT INJURY  SCORE (S)
                                                                                                   D
                                                               35      40      45      50
         Figure  40.  Phosphorous content of  pine needle litter related to  oxidant injury
                      score—low rainfall plots.

-------
    £0.7
   o 0.6
   tn
       0.5
   a.
   O

      0.4
      O.3
   o
O-I975  CA =0.294 + 0.00319 S
D-1976   r * 0.40
                                                                  O
                          10      15       20      25      30
                                  OXIDANT INJURY SCORE (S)
               40
45
Figure 41.  Calcium content  of  pine needle litter  related to oxidant  injury
            score—high rainfall plots.

-------
       0.7
     I
     « 0.6
! 1 i
-
1 l
0-1975
0-1976
n
1 l i 1
CA= 0.245 + 0.00654 S
r • 0.81
„ 	 •
1
^
     UJ
                          10
15      20      25      30

OXIDANT  INJURY SCORE (S)
Figure 42.  Calcium content of pine needle litter related to  oxidant  injury

            score—low rainfall plots.

-------
upper 7.5 cia of soil sampled under pine trees of varying oxidant injury
rating is shown in Figures 43 and 44.  It can be seen that as the oxidant
injury score approached zero (tree death) the soluble phosphorous in the
soil increased.  By comparing Figures 39 and 40 with Figures 43 and 44 it  is
evident that the slope of the regression lines for needle phosphorous
content with oxidant injury score are similar to the regression lines for
soluble phosphorous in soils with oxidant injury score.  There is probably a
causal relationship here, as postulated.  Similar relationships were found
for Ca, K and Mg however neither the Ca nor Mg content of needle litter
increased with increasing injury by oxidant air pollutants.
                                    164

-------
Ui
3   °'5
o
CO
u_
0   0.4
H

UJ



8 _ 0.3
             (0
             o
3
CD


O
0)
                  0.2
                 O.I
                 0.0
                                   O
                                             I
                                        I
                                                                      P» 0.288-0.0042 S

                                                                      r = 0.34
I
                                     10       15       20     25      30

                                            OXIDANT  INJURY SCORE (S)
                                                               35
               40
45
50
         Figure 43.
         Soluble phosphorous  content of surface  soils related to oxidant in-jury

         score of pine trees—high rainfall plots.

-------
o\
                  0.5
o
CO
fe   0.4
              o
                  0.3
ai
x~ 0.2
ft
                  O.I
              O
              CO
                  0.0
                                         O
                               _L
                                                    °o
1
                                                                   P= 0.254-0.00316 S
                                                                   r = 0.59
                                                                          O
1
                                                                                 O
                                     10       15      20      25      30
                                            OXIDANT INJURY  SCORE (S)
                                                              35
                               40
                       45
50
         Figure 44.   Soluble phosphorous  content of surface soils related  to  oxidant injury
                      score of pine  trees—low rainfall plots.

-------
FOLIAGE LITTER DECOMPOSITION  SUBSYSTEM:  MICROBIAL ACTIVITY AND  NUTRIENT
CYCLING

Introduction

     There is a continual turnover  of  tree biomass in  the  forest as  old
foliage, branches, and trees  die  and fall and  other  &row to replace
them.  Litter decomposition and nutrient cycling  are the means by which
the living forest recovers much of  the nutrition  incorporated in this
organic matter.  Recovery of  these  vital nutrients is  the  function
of the vast populations of litter and  soil microflora  and  microfauna.
This portion of the overall study is directed  toward determining the
influence of oxidant air pollution  on  microfloral populations and leaf
litter decomposition (primarily of  ponderosa and  Jeffrey pines).  Al-
though previous work in this  field  is  relatively  sparse, reviews are
available (Dickinson and Pugh, 1974).

Research Objectives

     Three major research approaches are being implemented in this sub-
system study.  They include the following:

     1)  To quantify needle litter  decomposition  in  the  field.   This
         will help determine  the  degree to which  oxidant air pollution
         affects (a) the quality  of needles as substrates  for decomposers
         and as sources of nutrients for cycling, and  (b)  the capacity of
         naturally-occurring  populations of litter micro-organisms to
         decompose pine needle litter.

     2)  To characterize and  quantify  microfloral inhabitant populations
         of pine needles from the period needle elongation through decom-
         position on the forest floor.  This will suggest  effects of oxidant
         air pollution on decomposer communities  and provide a basis
         for laboratory studies of  fumigation  effects  on decomposition.

     3)  To conduct laboratory fumigation experiments  on both fungal
         growth and needle decomposition.  These  are expectd to  clarify
         results obtained from field studies by eliminating such var-
         iables as moisture and temperature from  consideration.

Materials and Methods

Needle Litter Decomposition in Natural Stands—

     For this study, relatively isolated co-dominant and dominant trees
were selected.  Two each of the least  and most oxidant-injured Jeffrey

                                    167

-------
pine trees were selected on each of  two  sites.   These  sites  were Holcomb
Valley  ("no" oxidant  injury) and Camp Osceola  ("moderate"  oxidant^injury).
Similar selections of ponderosa pines were made  at Barton  Flats  ("moder-
ate" oxidant injury)  and Camp Congo  ("moderate-heavy"  oxidant  injury).
These sites represent the range of oxidant injury to each  species on
the study plos in the SBNF.  During  the  autumns  of 1974-1976,  freshly
fallen  litter was sampled randomly beneath each  selected tree  and subsam-
ples of approximately 15 gm were made at random.  One  random 20-gm subsam-
ple from each tree per year was analyzed by Drs. Gersper and Arkley for
nutrient content.  An additional random  sample was dried to  30 C, allowing
calculation of fresh/dry weight ratio for the entire sample. The remaining
subsamples were placed in labeled nylon  mesh (3  mm) envelopes  (approximately
15 cm x 30 cm) and disbursed as related  in Figure 45.   Each  arrow represents
similar treatments consisting of 5,  10,  and 30 envelopes left  in the field
for one winter, one year, and two years, respectively-  The  1974 samples
comprised the two-year experiment; the 1975 samples comprised  the one-year
and first one-winter  experiment. The first one-winter  experiment was re-
peated, and the second involved the  exchange of  ponderosa  and  Jeffrey pine
litter  between their  respective sites (Fig. 46).  The  rationale  for this is
explained in the following discussion.

     After each treatment was completed, envelopes were retrieved.  The
needles were:  (1) brushed lightly to remove excessive  inorganic and
fungal  materials; (2) dried to a constant weight at 30  C and  (3)  weighed.

     A  composite sample of the retrieved litter  for each treatment  in
the two, one and first overwinter experiments was analyzed for percent
content of N, P, K, Ca and Mg by Drs. Gersper and Arkley.  Percent  change  in
weight  and nutrient content were then calculated.

     Among the environmental variables which may influence litter de-
composition are solar radiation, litter  temperature, litter  moisture
and litter depth.  Although litter moisture has  not been measured,  it
is felt that careful  interpretation  of soil moisture depletion curves
and precipitation, soil moisture and litter temperature data will permit
the determination of  the relative litter moisture relations  between the four
plots under consideration.  Litter depth measurements  have been  taken for
comparison between trees and plots.  Solar radiation and litter  temperature
were measured once per hour, at four locations beneath each  study tree  from
sunrise to sunset on a cloudless day in  late August and early  September.
Solar radiation was measured with a Weather Measure Model  R401 Mechanical
Pyranograph.  Temperature in the upper one cm of the organic horizon was
measured with a calibrated mercury centigrade thermometer.  Each measurement
point corrsponded to the midpoint of one quarter of the arc  along which
decomposition envelopes were placed beneath each tree.

     Envelopes of litter from the two and one year experiments were
rated visually for activity in 14 categories of  signs  of decomposition.
The total length of needles on each  two  and one  year envelope  was esti-
mated by (1)  counting the number of  fascicles in each  envelope,  (2)
determining the mean length per fascicle for one envelope  per  treatment
and (3) multiplying this mean value by the number of fascicles in each

                                  168

-------
                                    Pinus Jeffrey!
       Holcomb Valley
Camp Osceola
                                    Pinus ponderosa
        Camp Oongo
Barton Flats
         1974 oxidant  score (Go868 was killed by bark beetles and replaced in
         this study by Si449.)
Figure 45.   Source and  destination  (tree tag-1976 oxidant score) of  960
             decomposition study envelopes.  Each arrow represents  30
             envelopes and points from  their source  to  their destination.
                                    169

-------
       Pinus Jeffrey!
Pinus  ponderosa
       Holcomb Valley
Barton Flats
Figure 46.   Source and destination (tree  tag-1976 oxidant score) of 160
            decomposition study envelopes.   Each arrow represents 5 en-
            velopes and points from their source to their destination.
                                 170

-------
envelope per treatment.  Because each  envelope per  treatment  constituted
a random subsample from the same source, and  each envelope  contained
approximately the same initial weight  of needles  (15  g),  it was  felt
that the estimated total length of needles per envelope  reflected  their
relative surface area.

     All of the above data types will  be examined for meaningful rela-
tionships to decomposition.

Decomposer microorganisms populations  on pine needles in natural stands
experiencing different oxidant doses—Oxidant air pollution may  alter
the rate of pine needle decomposition  by affecting  the composition of
microbial populations on senescing and fallen needles.   To  study the
succession of pine needle microorganisms, two lines of investigation were
followed in the field.  Microbial succession  in living needles was deter-
mined by isolation of fungi from surface-sterilized needles of various
ages, while succession in litter was determined by  isolating  fungi from
surface-sterilized needles that were on the forest  floor for  varying
periods of time.

     Before litter fall in 1974-1976,  one square meter of nylon  mesh
was placed beneath each of the trees involved in  the  integrated  field
needle compositon study at approximately two-thirds of the  crown radius  out
from the stem. In 1974, four  trees were tagged at each of two locations  in
the University of California  Blodgett  Experimental  Forest,  El Dorado County,
California.  They were tagged GR 1-8,  and each received  a nylon  mesh square
prior to litter-fall in 1974  and 1975. In 1975,  four Jeffrey and  four
ponderosa pines were selected for study and tagged  GR 9-16  on the  Stanislaus
National Forest, near Pinecrest, California.  Each  received mesh squares
prior to litter-fall in 1975. It was felt that pine stands  outside the SBNF
must be considered for comparison in terms of air pollution impact.  Having
separated annual increments of litter-fall in this manner,  we collected
periodic samples of litter from these  nets, surface-sterilized them, and
incubated them on water agar  in petri  dishes. The  populations of  micro-
organisms were then recorded.

     To determine the succession of microrganisms on  living pine needles,
the lowest healthy twigs on the north, south, east  and west sides  of
the stem were clipped not only from trees involved  in the integrated
field needle decomposition study, but  also from trees at Blodgett  and
Pinecrest.  The annual needle increments on each  twig were  separated in  the
field.  A subsample of each increment  was surface-sterilized  and incubated
on water agar in petri dishes.  The populations of microorganisms  were
then recorded.

     For the purposes of this study, a community will be considered to
consist of the microbial population inhabiting (1)  the needles of  a
given age on a given twig, or (2) the  needles representing  a  single
annual increment of litter-fall beneath a specific  tree.  Each community
will be characterized by an ordered pair of values  (S, e) representing
taxonomic richness and evenness (Williams, 1977)  and  a "species" list
                                   171

-------
ranked by abundance.  Taxonomic richness is represented by  the  number
of taxonomic categories isolated from each community.  Taxonomic  evenness
is defined as the distance from the point of perfect evenness to  the
point represented by a given community in n-space.  Figure  47 illustrates
the concept of evenness in a three species community.  Mathematically,
                         S     ±2  - 1
                           1-1
                              S-l          ,  where


      S  =  the number of taxonomic categories ("species")
            in the community, and
                                                      s
       TT =  the species frequency vector such that    .2 =  1-


This degree of data summarization will permit comparison of similar
communities in different situations.

Laboratory tests of the effect of oxidant  dose on decomposers and de-
composition—This phase of the project is  designed to determine who
air pollution affects  (1) growth and reproduction of microbial agents
of litter decomposition, and  (2) rates of  needle decomposition by major
microorganisms.

     Two clear and twelve opaque plexiglass fumigation chambers were
constructed and installed inside two walk-in Percival growth chambers
on the Oxford Tract, UCB.  These walk-in chambers have been renovated  to
permit control of light, temperature, and  relative humidity.  The plexi-
glass chambers permit control of ozone concentration.

     Species of fungi isolated from litter samples will be fumigated
at a number of ozone concentrations.  The  fungus will be inoculated
onto sterilized pine needle sections placed on cellophane-covered cellu-
lose agar (Eggins and Pugh, 1962) in petri dishes.  The effects of ozone
on such factors as (1) colony growth rate, (2) spore production,  (3)
spore germinability, and (4) cellulose decomposition will be quantified.

     The ability of a microflora to affect decomposition might be altered
by oxidant air pollution.  To test this possibility, preweighed,  steri-
lized pine needles from the Sanislaus N.F. were inoculated by placement on
a mixture of the organic horizons from BF, CAO, COO, HV, and the  Stanis-
laus N.F.  Moistened by constant subirrigation, one-half of the experiment
was exposed to ozone-enriched air.  Weight loss was determined after
14 weeks.

     Over a prolonged period of exposure, microbial populations might  be
altered by oxidant air pollution in ways which affect their ability  to
                                   172

-------
                       (0,1,0)
     (0,0,1)
                                                          (1,0,0)
Figure 47.  Evenness (£} in a three species community.
                                173

-------
 decompose  litter.   In  this  experiment,  preweighed  sterilized pine needles
 from the Stanislaus N.F.  were  inoculated  by  placement  on mixtures of
 the  organic  horizons from BF,  CAO,  COO, HV or  the  Stanislaus N.F.
 Moistened  by subirrigation,  the  entire  experiment  was  conducted in fil-
 tered air.   Weight  loss was  determined  after 22 weeks.

      Experiments have  been  designed which will test  (1)  the  ability^of
 a diverse  microflora to decompose needles produced by  healthy and di-
 seased trees and  (2) the  effects of moisture and temperature on litter
 decomposition.

 Results and  Discussion

 Needle Litter Decomposition  in Natural  Stands—

      Quantification of integrated needle  litter decomposition in the
 field has  been  concluded.  Data  sets  are  complete  for weight losses
 incurred by  320, 320,  160, 320 mesh litter envelopes over two years,
 one  year,  one winter and  the next winter,  respectively.   The 32 treat-
 ments involved  in each of these  experiments  are diagrammed in Figure
 45.   The 32  interspecific treatments  involved  in the second  overwinter
 experiment are  diagrammed in Figure 46.

      Ponderosa  pine litter lost  more  weight  («; = 0.001)  during the
 two  year,  one year, and first  overwinter  experiments than did Jeffrey
 pine litter  (Figure 48).  Within each species, greater weight loss occur-
 red  on the site receiving the  greatest  oxidant air pollution dosage.

      To determine whether the  difference  in  decomposition rate between
 ponderosa  and Jeffrey  pines  was  due to  some  species-specific factor or
 an environmental factor,  litter  envelopes  were placed  on plots of the
 opposite species to decompose.   Data  from this overwinter experiment
 (Fig.  49)  showed that  the species and source of litter was not signifi-
 cantly related  to decomposition  while the site of  decomposition was.  It
 is felt that the overwinter  weight  losses  experienced  in these experiments
 were relatively light, probably  reflecting the below-normal  precipitation.
 The  sparse literature  relevant to this  point suggests  that weight and
 nutrient losses by  pine needle litter during its first winter on the
 ground are relatively  great  (Stark, 1972,  1973; Millar,  1974).

     Change  in  nutrient status (per cent  N,  P, K,  Ca,  Mg on  a dry weight
basis) has been calculated for a pooled sample from each treatment in
the  two year, one year, and  first overwinter experiment.  Though both species
ultimately gained calcium and  nitrogen, ponderosa  pine litter consistently
acquired both nutrients at a greater  rate than did Jeffrey pine litter.
This may correspond  to the greater  fungal  activity and subsequent decomposi-
tion observed in ponderosa pine  litter.

     Measurement points at HV  received  the greatest total daily radiation
and temperature input  (Table 54), while those  at COO received the least.
CAO and BF lie  intermediate, CAO received  more radiation but approximately
the same temperature input as  BF.   This gradient matched the oxidant air

                                  174

-------
            JEFFREY PINE
         PONDEROSA PINE
CAMP OSCEOLA          HOLCOMB VALLEY
 -MODERATE               -HEALTHY
CAMP CONGO
 -SEVERE
BARTON FLATS
 -MODERATE
2 YEARS
10.84 > 9.41
1 YEAR
11.61 >* 9.43
0.5 YEAR
2.80 > 2.10
2 YEARS
19.99 >**
1 YEAR
14.61 >*
0.5 YEAR
8.86 >**

14.64

11.62

5.79
Figure 48.  Percent weight loss (30 C) incurred by needle litter on four
            plots, representing the range of air pollution impact on
            ponderosa and Jeffrey pine.  *-significant at 10% level;
            **-significant at 5% level; and ***-significant at 1% level.
                                   175

-------
            JEFFREY PINE                              PONDEROSA PINE

                                  0.5 YEAR

            HV COO (8.85)                              BF  COO (9-86)

                BF (3.69)                                  BF (4.27)

               CAO (7.08)                                 CAO (6.40)

                HV (4.07)                                  HV (4.27)




           CAO COO (8.89)                             COO  COO (9.60)

                BF (3.57)                                  BF (5.57)

               CAO (3.58)                                 CAO

                HV (2.45)                                  HV (5.27)
Figure 49.  Percent weight loss (30 C) incurred by needle litter  following
            transfer between and within species.  Transfers made  according
            to plant shown in Figure 48.
                                   176

-------
TABLE  54.  ESTIMATED TOTAL RADIATION  (R) AND TEMPERATURE  (T) ACCUMULATED
            DURING ONE CLEAR DAY AT MEASUREMENT POINTS BENEATH THE  INTE-
            GRATED FIELD DECOMPOSITION STUDY TREES.
Species Plot
Jeffrey HV




Jeffrey CAO




Ponderosa BF



Ponderosa COO




X
Tree
1537
1561
1574
1598
X
1865
1934
1964
807
X
2600
2625
2755
X
852
875
894
449
X

R
107.53
292-14
253.91
400.57
263.54
167.59
213.93
211.12
159.14
187.95
182.03
258.96
126.06
189.02
202.04
158.33
79.47
133.44
143.32
166.17
T
15724.4
22919.4
21113.6
24984.1
21185.4
16755.5
18626.5
15466.0
15750.0
16649.5
15988.9
20362.2
14181.12
16844.08
17864.9
17477.1
13917.4
16978.6
16559.5
16701.8
                                    177

-------
 pollution, decomposition and species gradients determined for these four
 plots.  This suggests that radiation (and temperature) might be inversely
 correlated with weight loss.  Correlation analysis, however, showed that the
 only significant within-plot relationship (<* = 0.05) between litter weight
 loss and either radiation or temperature occured at HV.  Further, at HV
 litter weight loss was directly proportional to both incident radiation and
 temperature.  The elevation of HV is the highest of the four plots studied
 and depends heavily on winter precipitation for moisture.  Perhaps points
 receiving greater radiation at HV were exposed for a greater length of time
 to temperature and moisture conditions conducive to decomposition.

 Decomposer Microorganism Populations on Pine Needles in Natural Stands
 Experiencing Different Oxidant Doses—

      One complete experiment during late summer of 1975 provided information
 on microbial succession in pine foliage in the SBNF, Blodgett Forest, and
 the Stanislaus N.F.  Data from a similar experiment in the SBNF during
 spring, 1976, will be analyzed shortly.  One additional experiment of this
 type is in progress.  In this experiment, it is hoped that nutrient status
 will be determined by Drs. Arkley and Gersper for each annual increment of
 foliage.  Results of the first experiment are tentative, requiring further
 analysis and confirmation.  Collection of data on microbial succession in
 litter continues.  The value of the data increases from each successive
 sampling; there are now four annual increments of litter-fall separated
 (except where nets have been vandalized) in the SBNF.  All data collected
 from these studies will be interpreted in the light of the existing environ-
 mental data.

      Through the study of annual increments of living foliage and strati-
 fied litter, successions of microfloral populations are being determined
 for individual trees representing plots and regions impacted by varying
 amounts of smog.  Comparisons among these population successions will
 help to explain patterns in (1) litter decomposition and (2) the inci-
 dence of fungus-caused damping-off of pine seeds and seedlings (under
 study by the seedling establishment investigators).  Species of fungi
 for which population data have been collected, to date, include phyco-
 mycetes,  ascoymcetes,  basidiomycetes, and fungi imperfect!.  A method
 for the culture of fungi on microscope slides has been employed for the
 grouping  and identification of important isolates (Riddell, 1950).

 Laboratory Tests of  the Effect of Oxidant Dose on Decomposer and Decom-
 position—

      Two growth  chamber decomposition experiments have been completed.
The data from  these  experiments will be analyzed shortly and both will
be  repeated.   The  first experiment tested the ability of BF, COO,  CAO,
HV, and Stanislaus N.F.  litter microorganisms to decompose a standard
sterile needle litter  in filtered air.   The second experiment tested
the effect of  ozone  on  the  decomposition of standardized litter by the
organisms  in a standardized  mixture of  the organic horizons from BF, COO,
CAO, HV and the  Stanislaus N.F.
                                    178

-------
                                REFERENCES

Asher, J. E.  1956.  Observations and theory on "X" disease or needle
     dieback.  File Rep.  Arrowhead Dist., San Bernardino Natl. For.,
     Calif.

Bingham, G. E. and P. I. Coyne.  1977.  A Portable, Temperature Control-
     led Steady State Porometer for Field Measurements of Transpiration
     and Photosynthesis.  Photosynthetica  11(1):148-160.

Black, C. A.  (ed.).  1965.  Methods of Soil Analysis Agronomy No. 9,
     Part II.  Madison, Wis., Amer. Soc. Agronomy, Inc. pp. 771-1572.

Botkin, D. B., J. F. Janak and J. R. Wallis, 1972-  Some Ecological Conse-
     quences  of a Computer Model of Forest Growth.  Journ Ecol. 60:849-872*

Braun-Blanque, J.  1932*  Plant sociology.  Hafner.  N.Y.  pp.  22-25.

Byler, J. W., D. R. Hart and R. E. Wood.  1976.   Project plan:  Biologi-
     cal evaluation of insect and disease caused  mortality in the San
     Bernardino National Forest.  U.S.F.S. Region 5.  San Francisco,
     CA  Mimeo 68 p.

Cobb, F. W.,  Jr., D. L. Wood, R. W. Stark, and J. R. Parmeter, Jr. 1968.
     Photochemical oxidant injury and bark beetle (Coleoptera: Scolytidae)
     infestation of ponderosa pine.  IV. Theory on the relationship between
     oxidant  injury and bark beetle infestation.  Hilgardia  39(6):141-152.

Cobb, J. W.,  Jr., et al.  1974.  Effects of photochemical air pollutants
     on the epidemiology of root pathogens in the mixed conifer forests.
     In:  Oxidant air pollutant effects on a western coniferous forest
     ecosystem.  Annual Progress Report 1973-74.  0. C. Taylor (ed.)
     Statewide Air Pollution Research Center.  Univ. of Calif. Riverside,
     CA  Section III-E.

Corn, M., R.  W. Dunlap and L. A. Goldmuntz.  1975.  Photochemical oxidants:
     sources, sinks and strategies.  J. Air Poll. Contr. Assoc.  25(1):
     16-18.

Coyne, P. I.  and G. E. Bingham.  1978.  In Annual Progress Report on the
     Studies  of Air Pollution Effects on Vegetation, Lawrence Livermore
     Laboratory,  pp. 12-20.

Dahlsten, D. L.  1974.  The role of bark beetles, Dendroctonus spp., in
     the succession of the mixed conifer ecosystem exposed to photochemical
     air pollutants, Section III-C.  In:  Annual  Progress Report 1973-74,

                                   179

-------
References Continued

     for the U. S. Environmental Protection Agency, Corvallis, Or.   10
     pp. and Appendix.

Dahlsten, D. L., C. J. DeMars, D. L. Rowney, and P- A. Felch.  1974.
     Sampling techniques and preliminary analyses for the western pine
     beetle in the central Sierras of California.   (Unpublished manu-
     script, in preparation).

Dahlsten, D. L., and D. L. Rowney.  1974.  The influence of Verticicladiella
     wagenerii on western pine beetle, Dendroctonus brevicomis, populations.
     (Unpublished progress report, Western Regional Project W-110).  5
     pp.

Dahlsten, D. L.  1977.  Western pine bark beetle population dynamics—
     stand tree mortality subsystem, pp. 159-176.   In:  Photochemical
     oxidant air pollutant effects on a mixed conifer forest ecosystem—
     a progress report.  P. R. Miller and M. J. Elderman, editors.  Con-
     tract Nos. 68-03-0273 and 68-03-2442.  U. S. Environmental Protection
     Agency, Corvallis Environmental Research Laboratory, Oregon 338-
     pp.

Dickinson, C. H., and G. J. F. Pugh.  1974.  Biology of Plant Litter
     Decomposition, Vols. I and II.  Academic Press, New York.  775 pp.

Dixon, W. J.   1975.  BMDP Biomedical Computer Programs.  University of
     California Press, Berkeley, California  792 p.

Duff, G. H. and N. J. Nolan.  1953.  Growth and morphogenesis of the
     Canadian forest species.  I.  The controls of  cambial and apical
     activity in Pinus resinosa.  Can. J. Bot.  31:471-513.

Eggins, H. 0. W. and G. J. F. Pugh.  1962.  Isolation of Cellulose-decompos-
     ing Fungi from the Soil.  Nature  193:94.

Fry, K. E. and R. B. Walker.  1967.  A Pressure Infiltration Method for
     Estimating Stomatal Opening in Conifers.  Ecology  48:155-157.

Horton, J. S.  I960.  Vegetation types of the San Bernardino mountains.
     U.S. For. Ser., U.S.D.A. Tech. Paper No. 44.   29 pp.

Jackson, N. L.  I960.  Soil Chemical Analysis.  Prentice-Hall Inc., Engle-
     wood Cliffs, N.J.  498. pp.

James, R. L. et al.  1977.  Root pathogen dynamics  and stand mortality
     subsystem.  In:  P. R. Miller and M. J. Elderman (eds.).  Photochemical
     oxidant air pollutant effects on a mixed confier forest ecosystem—
     a progress report.  Contract Nos. 68-03-0273 and 68-03-2442 (EPA),
     Univ- of Calif., Riverside, CA.  pp. 177-185.

Johnson, C. M. and A. Ulrich.  1959.  II.  Analytical Methods for Use

                                   180

-------
References Continued

     in Plant Analysis.  Calif* Agr. Exp.  Sta. Bull.   766.  77 pp.

Kickert, R. N.  1977.  Definition of the Conifer Forest Ecosystem as  a Group
     of Coupled Ecological Models.  In:  Photochemical Oxidant Air Pollutant
     Effects on a Mixed Conifer Forest Ecosystem.  Miller, Paul R. (ed.)«
     U.S. Environmental Protection Agency, Corvallis, Oregon.  71-105.

Kickert, R. N., P. R. Miller, 0. C. Taylor, J. R. McBride, J. Barbieri,
     R. Arkley, F. W. Cobb, Jr., D. L. Dahlsten, W. W. Wilcox, J. M.  Wenz,
     J. R. Parmeter, Jr., R. F. Luck and M. White.  1976.  Photochemical
     air pollutant effects on mixed conifer ecosystems:  a progress report.
     U.S. Environmental Protection Agency  Office of Res. and Dev.  Corvallis
     Environ. Res. Lab., Corvallis, Oregon.  CERL-026.  275 pp.

Kozlowski, T. T* and C. H. Winget.  1964.  The Role of Reserves in Leaves,
     Branches, Stem and Roots on Shoot Growth of Red Pine.  Amer. J.  Botany
     51:522-529.

Larsen, R. A. and W. W. Heck.  1976.  An Air Quality Data Analysis System
     for Interrelating Effects Standards and Needed Source Reductions.
     Part 3.  Vegetation Injury.  J. Air Pollut* Contrl Assoc.  26:325-333.

Luck, R. F.  Cone and seed production for  dominant conifer tree repro-
     duction.  In:  P. R- Miller and M. J. Elderman (eds.).  Photochemical
     Oxidant Air Pollutant Effects on a Mixed Conifer Forest Ecosystem.
     a progress report*  U. S. Environmental Protection Agency, Office of
     Research and Development.  Corvallis  Environmental Research Laboratory
     CERL EPA-600/3-77-104: 338 pages.

MATHTECH.  1976.  Computer Simulation Model for Analyzing Mobile Source
     Air Pollution Control Strategies.  U. S. Environmental Protection
     Agency, 600/5-76-010, Washington, D.C.  180 p.

McBride, J.  1973.  Vegetation of principle study sites in the San Bernar-
     dino mountains.  In:  Oxidant air pollutant effects on a western
     coniferous forest ecosystem.  Task C  Report.  Statewide Air Pollution
     Research Center, Univ. of Calif-, Riverside, CA.

McBride, J. R.  1974.  Annual report of the vegetation sub-committee.
     In: Oxidant air pollutant effects on  a western coniferous forest
     ecosystem.  Annual Progress Report 1973-74.  0. C. Taylor (ed.).
     Statewide Air Pollution Research Center, Univ. of Calif., Riverside,
     CA  Section III-B.

McBride, J. R., V. Semion, and P. R. Miller. 1975.  Impact of air pollution
     on the growth of ponderosa pine.  Cal. Agr.  29(12):8-10.

McBride, J. R.  1977.  Tree population dynamic subsystem.  In:  P- R.
     Miller and M. J. Elderman (ed.)«  Photochemical oxidant air pollutant
     effects on a mixed conifer forest ecosystem—a progress report.

                                   181

-------
 References  Continued

      Contract  Nos.  68-03-0273  and  68-03-2442  U.S.  Environmental Protection
      Agency, Univ.  of  Calif.,  Riverside,  CA  pp.  106-121.

 McCutchan,  M.  H.  and M.  J.  Schroeder.   1973.   Classification of Meteoro-
      logical Patterns  in Southern  California  by Descriminant Analysis.
      J.  Appl.  Meteorol.   12:571-577.

 Millar,  C.  S.   1974.   Decomposition of  Coniferous  Leaf Litter.   In:   Biology
      of  Plant  Litter Decomposition.  Dickinson, C. H. and  C. J. F.  Pugh
      (eds.) Academic Press, New York.   pp.  105-128.

 Miller,  J.  M., and  F.  P. Keen.  1960.   Biology and control of the western
      pine beetle.  A summary of the first fifty years of research.   U.
      S.  Department  of  Agriculture  Forest  Service,  Washington, D.C.  Misc.
      Publ.  No. 800.  381 pp.

 Miller,  P.  R., J. R. Parmeter, Jr., 0.  C. Taylor  and E. A. Cardiff.   1963.
      Ozone  injury to the foliage of ponderosa pine.  Phytopath.  53:1072-
      1076.

 Miller,  P-  R.   1973.   Oxidant  damage to conifers  on selected sites,  1972.
      In: Taylor, 0. C.   Oxidant air pollution effects on  a western coni-
      ferous forest  ecosystem.  Task C Report.  Statewide Air Pollution
      Research  Center.  Univ. of Calif*, Riverside, CA.

 Miller,  P-  R.   1973.   Susceptibility to ozone of  selected  western conifer.
      2nd Inter. Cong.  Plant Pathology,  Minneapolis, Abstract 0579.

 Miller,  P-  R.  and J. R.  McBride.   1973.   Vegetation committee report
      (Sec.  A). In:  Oxidant air pollutant  effects on a western coniferous
      forest ecosystem.   Task B Report.  Statewide  Air Pollution Research
      Center.   Univ. of Calif., Riverside,  CA.

 Miller,  P.  R.  and J. R.  McBride.   1975.   Effects  of air pollution on
      forests,  pp.   175-235.  In:   Mudd, J.  B.  and  T. T. Kozlowski (eds.),
      Responses of Plants  to Air Polluton.   Academic Press, N.Y.

 Miller,  P.  R.  1977.   Oxidant flux canopy response subsystem.  In:   P. R.
      Miller and M. J. Elderman (eds.).  Photochemical oxidant air pollutant
      effects on a mixed  conifer forest  ecosystem—a progress report.
      Contract  Nos. 68-03-0273 and  68-03-2442.  U.  S. Environmental  Protec-
      tion Agency.  Univ.  of Calif., Riverside, CA pp. 122-138.

Miller, P. R.,  R. N. Kickert, 0. C. Taylor, R. J.  Arkley,  F. W. Cobb,
      Jr., D. L. Dahlsten, P. J. Gersper,  R. F. Luck, J. R. Parmeter, Jr., J.
     M. Wenz,  M. White and W. W. Wilcox.   1977.   Photochemical  oxidant
     air pollutant effects on a mixed conifer forest ecosystem—a progress
     report, 1975-1976.   U.S. Environmental Protection Agency Contracts Nos.
     68-03-0273 and 68-03-2442.  Corvalis Ecological Research Laboratory,
     Corvallis, Oregon EPA-600/3-77-104:   338 pp.

                                   182

-------
References Continued

Minnich, R. D. et al.  1969-  Mapping montane vegetation  in  southern
     California.  Status Report 3, Contract No.  14-08-0001-10674.  U.
     S. Dept. Interior, Tech. Report 3, Dept. Geog. Univ.  Calif.,  Riverside,
     CA  40 pp.

Minnich, R. A.  1974.  The impact of fire  suppression  on  southern  California
     conifer forests:  a case study of the Big Bear fire,  November 13-16,
     1970.  In:  Symposium on living with  the chaparral proceedings.   Sierra
     Club.  San Francisco, CA  pp. 113-126.

MuKammal, E. I.  1965.  Ozone as a Cause of Tobacco Injury.  Agr.  Meteorol.
     2:145-165.

Nie, N. H., C. H. Hull, J. G. Jenkins, K.  Steinbrenner and D. H. Bent.
     1975.  SPSS Statistical Package for the Social Sciences.  McGraw-
     Hill Book Co., New York.  675 p.

Parmeter, J. R., Jr., R. V. Bega and T. Neff.  1962.   A chlorotic  decline
     and death of ponderosa pine in southern California.   Plant Dis. Rep.
     46:269-273.

Reed, K. L. and R. S. Waring.  1974.  Coupling of Environment to Plant
     Response:  A Simulation Model of Transpiration.   Ecology  55:62-72.

Reed, K. L.  1976.  Prediction of Tree Growth Response Surfaces in an
     N-Dimensional Habitat Coordinate System.  (in prep.).

Reed, K. L. and S. G. Clark.  1976.  Succession  SIMulator:   A Coniferous
     Forest Simulator.  Model Documentation.  Coniferous  Forest Biome
     Bulletin No. 11, University of Washington,  Seattle,  Wash.  73 p.

Riddell, R. W.  1950.  Permanent Stained Mycological Preparations  Obtained
     by Slide Culture.  Mycologia  42:265-270.

Schussel, G.  1977.  The role of the data  dictionary.  Datamation.  23(6):
     129, 133, 136, 142.

Sneath, P. M. A. and R. R. Sokal.  1973.   Numerical taxonomy.  W.  H. Freeman
     and Company.  San Francisco.  573 p.

Sokal, R. R. and F. J. Rohlf.  1969.  Biometry.  W. F. Freeman and Co.
     San Francisco.  776 pp.

Sollins, P., R. H. Waring, and D. W. Cole.  1974.  A Systematic Framework
     for Modeling and Studying the Physiology of a Coniferous Forest
     Biome.  Bulletin No. 5.  Waring, R.H. and R. L.Edmonds  (eds.).  Univer-
     sity of Washington, Seattle, Wash.  p. 7-20.

Stark, R. W., P. R. Miller, F. W. Cobb, Jr., D. L. Wood,  and J. R. Parmeter.
     1968.  Photochemical oxidant injury and bark beetle  (Coleoptera;

                                   183

-------
References Continued

     Scolytidae) infestation of ponderosa pine.   I.  Incidence  of  bark beetle
     infestation in injured trees.  Hilgardia 39(6):121-126.

Stark, N.  1972.  Nutrient cycling pathways and litter  fungi.   Bioscience
     22:335-360.

Stark, N.  1973.  Nutrient cycling in a Jeffrey pine ecosystem.   Montana
     Forest and Conservation Experiment Station.  389 pp.

Thompson, D. R. and T. M. Hinckley.  1977.  A simulation of water relations
     of white oak based on soil moisture and atmospheric evaporative
     demands.  Canad. J. Forest. Res.  7:400-409.

Westman, W. E.  1977.  How much are nature's services worth?   Science.
     197:960-964.

Wilkinson, B.  1977.  An application audit.  Datamation.   23(8):51-54..

Williams, C. B., Jr.  1967.  Spruce budworm damage symptoms related  to
     radial growth of grand fir, Douglas-fir and  Engelman  Spruce.  For.
     Sci.  13(3):274-285.

Williams, F. M.  1977.  Model-free evenness:  An  alternative to diversity
     measures.  Paper S23-6.  4 pp. presented 8/8/77 to Second International
     Ecological Congress, Satellite Program in Statistical Ecology,  Satel-
     lite A:  NATO-Advanced Study Institute and ISEP Research Workshop.

Wood, D. L.  1971.  The impact of photochemical air pollution on  the
     mixed conifer forest ecosystem arthropods.   In:  Oxidant air  pollutant
     effects on a western coniferous forest ecosystem.  Task B Report:
     Historical Background and Proposed Systems Study of the San Bernardino
     Mountain Area-  U.S. Environmental Protection Agency, Corvallis,
     Oregon, Section C.   26 pp.

Wood, D. L.  1973.  The impact of photochemical air pollution on  the mixed
     conifer forest ecosystem arthropods.  In:  Oxidant air pollution
     effects on a western coniferous forest ecosystem.  Task B Report.
     Historical Background and Proposed Systems Study of the San Bernardino
     Mountain Area.  U.S. Environmental Protection Agency, Corvallis,
     Oregon.  Section C.   26 pp.
                                   184

-------
APPENDIX 1.  PART II OF THE  SBNF  DATA DICTIONARY:   DEFINITION OF DATA-
             SET NAMES.
Dataset
 name
              Definition of data content
SBNFPUBS
PLOTINDX
SXSCAT
STEXOM
TREE
 STAGE
 SHRUB
 TREEVEG
 SHRUBVEG
 PLOTREGN
 FIRETREE
 FIRESTAG
Bibliography of publications, manuscripts in preparation,
and intended manuscripts from scientists conducting re-
search under EPA contracts 68-02-0303, 68-03-0273, 68-03-
2442, R805410-01 from 1972 through present.

For each vegetation plot, gives year established, eleva-
tion, geographic coordinates, length, width, azimuth,
general hillslope aspect, tree tag colors, lowest and
highest tree tag numbers, forest type, and number of
tagged trees;

Soil exchangeable and soluble cations, various soil depths
on each vegetation plot;

Soil texture and organic matter; percent of soil in
various texture classes, pH, and organic carbon and nitro-
gen content at various depths on each vegetation plot;

Tree identification; locations and species of tagged
trees on each vegetation plot;

Stand age data; number of trees by ten year age class,
by species, on vegetation plots;

Locations of shrub cover, by species, along a transect
across each vegetation plot;

Tree vegetation data; density, basal area, species com-
position of trees, greater than, or equal to 10 cm diameter
at breast height, on each vegetation plot;

Shrub vegetation data; frequency, density, percent cover,
by species, on each vegetation plot;

Plot regeneration data; age of tree seedlings, saplings,
poles, and age classes of trees in larger size-classes,
by cover types, on each vegetation plot;

Tree data on 85 special study plots which had burned
at various dates;

Fire stand age data; similar to STAGE  (above) but on
85 special study plots which had burned  at various dates;
                                   185

-------
 APPENDIX 1.  PART II OF THE SBNF DATA DICTIONARY:  DEFINITION OF DATA-
              SET NAMES.  (CONTINUED)
 Dataset
  name
               Definition of  data content
 FIRESHRUB
 CTREE
 FSMTINDX
 FSMET
 HMET
 HPREC
 OXIDINDX
 OXIDANT
PLOTMET
PLOTPREC
PLOTOXID
OZFLUX
 Fire  shrub  data;  similar  to  SHRUBVEG (above),  but on
 85  special  study  plots  which had  burned at various dates;

 Crown tree  data,  such as  Keen crown class, crown position
 in  canopy,  gross  geometrical crown  volume, tree height,
 and crown ratio,  for each tagged  tree on each  vegetation
 plot;

 Forest Service meteorological data  index;  days, by month
 and year, in  FSMET  dataset,  which have valid usable data;

 Forest Service meteorological data  which were  telemetered
 from  3 sites  near the vegetation  plots;  net radiation,
 air temperature at  4 ft and  6 ft, relative humidity,  wind
 direction,  wind speed, hourly resolution;

 Historical  meteorological data, incuding relative humidity,
 wind  direction, wind speed,  air temperature maximum and
 minimum on  a  daily  basis,  for sites  at various distances
 from vegetation plots;

 Historical  precipitation  data on  monthly and annual basis
 for 23 sites  operated by  the San  Bernardino County Flood
 Control District, and ranging as  far back  as 92 years;

 Oxidant index; days, by month and year,  which  have valid
 data in OXIDANT dataset;

 Hourly data on oxidant and ozone  concentration at sites
 in  the vicinity of vegetation plots;

 Plot meteorological data  on  air temperature and relative
 humidity from hygrothermographs located  on the vegetation
 plots;

 Plot precipitation data from snow gauges and summer rain
 gauges located on each vegetation plot;

 Plot oxidant  ambient concentration  data  obtained during
various time  periods on vegetation  plots;

Ozone  flux  data for sample conifer  trees near  a vegeta-
 tion plot;
                                   186

-------
APPENDIX 1.  PART II OF THE  SBNF DATA DICTIONARY:   DEFINITION OF DATA-
             SET NAMES.   (CONTINUED)
Dataset
 name
              Definition of data content
STOMRES
INJBIWK
TREE
TREEMORT

SAPTREE



SAPGRO


SAPSURF

PNFALL



TREELIT



PLITR


LITMAS


TREEGRO



TREESOIL
Stomatal resistance data for sample conifer trees near
a vegetation plot;

Air pollution- conifer tree foliar injury on a biweekly
basis for sample trees near a vegetation plot; needle
retention, needle length, injury length;

Annual foliar injury observations on each tagged tree on
each vegetation plot; needle retention score, needle
condition score, branch mortality score, needle length
score, total smog injury score, disease observation, insect
risk category, tree diameter at breast height;

Tree mortality data for each vegetation plot;

Sapling tree data, analogous to the tree dataset (above),
but for sapling plots located very near the vegetation
plot;

Sapling height growth data on an annual basis on the
sapling plots;

Sapling foliage surface data;

Pine needle fall from tree crowns on an annual basis,
beneath selected pine trees on all but 2 of the vegetation
plots;

Tree litter thicknesses on the ground in 4 directions
at various distances from the tree trunk, under selected
trees on selected plots;

Plot litter thickness at 2 m intervals down the center-
line of each vegetation plot in 1973;

Litter mass (dry weight) at various distances from  trunks
of selected trees on selected vegetations plots;

Tree growth data, cumulative radial, for conifer trees
at least 10 cm diameter, breast height, on 6 vegetation
plots representing extremes of air pollution exposure;

Soil type, hillslope gradient, and hillslope aspect for
each tree in the TREEGRO dataset;
                                   187

-------
 APPENDIX 1.  PART II OF THE SBNF DATA DICTIONARY:  DEFINITION OF DATA-
              SET NAMES.  (CONTINUED)
 Dataset
  name
               Definition of  data content
 TREEGR02
 BOGRO
 MOIST
 MATRIC
 ISURV
 BTREE
 EGG
REAR
STIK
XRAY
DISU
FASP
 Amalgamation of  TREESOIL  dataset  with TREEGRO, and con-
 verted to  give annual  radial  stem growth,  at breast
 height,  as far back as year  1920;

 Black oak  growth data,  similar  to  TREEGRO  dataset des-
 cribed above;

 Soil  water percentage,  by weight,  at  various depths at
 weekly,  or biweekly, intervals  for 22 sites, including
 the vegetation plots;

 Soil  water MATRIC potential at  various  depths for data
 from  the MOIST dataset, converted  by  means of soil water
 retention  curves determined in  the lab,  using respective
 field soil samples;

 Insect survey data  taken  in early  summer and late fall,
 from  tagged trees on vegetation plots;

 Beetle tree data; infestation heights and  stem circum-
 ferences at sampling heights for western pine bark beetle
 on killed  trees;

 Western  pine bark beetle  attack densities,  gallery lengths,
 egg counts,  from bark disc samples  off  of  killed  trees;

 Emergent bark beetle densities, parasites,  and predator
 data  from  bark field samples reared under  laboratory
 conditions;

 Emergent bark beetle densities, parasites,  and predator
 data  from  bark into sticky cartons  on tree trunks  under
 field  conditions;

 Potential  bark beetle brood, as determined from XRAYs
 of bark  sample discs;

Disease  survey data; plant diseases found  on tagged
 trees  on vegetation plots;

Fomes annosus spread plots data;
                                  188

-------
APPENDIX 1.  PART II OF  THE  SBNF DATA DICTIONARY:
             SET NAMES.   (CONTINUED)
                                                    DEFINITION OF DATA-
Dataset
 name
                                Definition of data content
SPRMORT1
SPRMORX
 STNDSITE
 TREEPEST
 SPRXTREE
 SPRXRGNS
 SPRXRGNC
 SPRXGRND
 SPRXSHRB
 SPRMORT2
 CONE

 PLOTSEED
                  Super  plot  mortality dataset #1;  data on dead and severely
                  damaged  trees,  obtained in cooperation with the Pest
                  Damage Inventory,  U.S.  Forest Service, Region 5, San
                  Francisco;

                  Super  plot  mortality extra data not part of the USFS/PDI
                  (above); smog score components, basal area detail, addi-
                  tional insect related symptoms, recent annual stem radial
                  growth rate,  height to  lowest live branch bearing needles,
                  all on dead/damaged tree mortality centers;

                  Stand  site  data for trees around each dead/damaged tree
                  mortality center in a superplot;

                  Tree morphological and  growth data, and insect and disease
                  data,  for each tree in the stand surrounding a mortality
                  center;

                  Super  plot  extra tree data for trees within a 30 m x 30
                  m plot around each mortality center;

                  Super  plot  extra regeneration tree data in square plots
                  (10 m x  30 m) upslope from each mortality center;

                  Super  plot  extra regeneration tree data in circular plots
                  (6.5 m diameter) around mortality centers and paired
                  living trees;

                  Super  plot  extra ground cover data in 3 plots (10 m x
                  30 m)  upslope from each mortality center;

                  Super plot extra shrub intercept data along 2 lines
                  (30 m) upslope from each mortality center;

                  Super plot mortality dataset #2; data on dead and damaged
                  trees  obtained from superplots located near air quality
                  and meteorological monitoring sites; done separately
                  from SPRMORT1 (above);

                  Annual cone counts from trees on vegetation plots;

                  Average seed production per  cone for  cone-bearing trees
                  on each vegetation plot;
                                   189

-------
APPENDIX  1.
PART II OF THE SBNF DATA DICTIONARY:
SET NAMES.  (CONTINUED)
                                                    DEFINITION OF DATA-
Dataset
 name
                  Definition of data content
 SAPS
LSOIL
 SSAS
SLSS
FLDECOMP
LITRKEM
DRIP
SFCSOLKM
    Seedling air pollution study; numbers of emerged pine
    seedlings and seedling mortality from various ground
    cover and wildlife exclusion treatments in mini-plot
    on selected vegetation plots;

    Laboratory determined soil moisture percent by weight
    for soil samples taken from seedling mini-plots;

    Seedling supplementary animal study; data on proportional
    loss of seeds and seedlings associated with different
    kinds of wildlife exclusion treatments in mini-plots
    on select vegetation plots;

    Supplementary litter seedling study; data on seedling
    emergence and mortality associated with different litter
    depth treatments;

    Field decomposition data for pine needles in net bag
    samples under select trees on select vegetation plots,
    over 1, 2,  and 3 year intervals;

    Litter chemistry; elemental content of pine needle fall
    at various  distances from tree stems,  beneath select
    sample trees on select vegetation plots;

    Intercepted precipitation crown drip elemental content,
    at various  distances from tree stems,  beneath select
    sample trees on select vegetation plots;

    Surface soil chemistry;  elemental content of surface
    soil at various distances from tree stems,  beneath select
    sample trees on select vegetation plots.
                                  190

-------
APPENDIX 2.  ON-LINE DATASET PROGRESS CHECKLIST FOR SBNF DATA BASE OF AUGUST 31, 1977.

IDENTIFICATION










DATASET
NAME


CRMppITTJO
ouiNr JT uoo
PT OTTWTIY
L Lt\J i. J.1NLJ4V
SXSCAT
STEXOM
TRID
STAGE
SHRUB
TREEVEG
SHRUBVEG
PLOTREGN
FIRETREE
FIRESTAG
FIRESHRUB
*CTREE-FD
CTREE- RD










PRINCIPAL
INVESTI-
GATOR

PROTVfT
IT jxv/ij EJ\J j.
PPD TETT
JTJXVJU Ej\_*.L
ARKLEY
ARKLEY
MCBRIDE
MCBRIDE
MCBRIDE
MCBRIDE
MCBRIDE
MCBRIDE
MCBRIDE
MCBRIDE
MCBRIDE
LUCK
LUCK










DATA
(MONTH)
YEAR


HKFPTr
WiN V_*Ij
ONCE
ONCE
ONCE/ 73
ONCE
ONCE
ONCE
ONCE
1977
ONCE
ONCE
ONCE
ONCE/ 75
ONCE/ 75
NEW
DATA




D A
A P
T F P
A 0 R
R 0
F M V
0 A E
R T D
M ?










Y










D
E
S E
C N
R T
I E
P R
T E
0 D
R ?




9
N
N
N
N
N
N
N
N
N
Y
Y




K
E
Y
P
U
N
C
H
E
D
?


Y
1
«
•
•
N
N
RPT
RPT
12/77
N
N
N
M10
M10

VERIFICATION
D
A
U T
N A
V
E E
R N
I T
F E.
I R
E E
D D
?



Y
Y
Y
10/77







UCR

V
E
R D
I E
F B
I U
C G
G
P E
R D
0 ?
G
R
























DATA
LISTING
TO
P.I,
(.DATE)



Y
Y
5/77








AT UCR


C
0
R E
R N
E T
C E
T R
I E
0 D
N ?
S




N
N









M10
N10

ANALYSIS



R
E D
D E
U P F
C L I
TAN
I N E
0 D
N
?



N
N











N
R
E
D
U
C
T
R.
P A
R N
0 ?
G
R

















INFORMATION
TRANSFER





ANALY-
SIS
OUTPUT

TO

P.I.
?
















D
0
C
U
M M U
0 E P
D N D
ETA
L A T
T E
I D
0 ?
N


















I
N
T
E
W R R
R P E
IRC
TEE
T T I
E A V
N T E
I D
0
N
















-------
APPENDIX 2.  ON-LINE DATASET PROGRESS CHECKLIST FOR SBNF DATA BASE AS OF AUGUST 31, 1977 (CONTINUED)
FSMTINDX
FSMET
HMET
HPREC


a^VT T^T'NT'n V
UAJ-UIINJUA
OXIDANT
PLOTMET
PLOTPREC

PLOTOXID
OZFLUX
STOMRES
INJBIW1C

TREE




TREEMORT
SAPTREE
SAPGRO
SAPSURF
PNFALL



TREELIT
PLITR
LITMAS

*TREEGRO


MILLER
MILLER
MILLER
ARKLEY


A/fTT T T?T3
WlLiijJiK
MILLER
MILLER
MILLER

MILLER
MILLER
MILLER
MILLER

MILLER




MILLER
MILLER
MILLER
MILLER
ARKLEY



ARKLEY
ARKLEY
ARKLEY

OHMART



75-76
73-76
TO 9/75
10/75-
9/77


1967-76
MAY 77
WINTER? 5
WINTER? 6
MAY 77
JUN 77
?
JUN 76
77
FALL 7 3
74
75
76
77
?
ONCE
1967-76
77
736J4
75
76
77
73/75
ONCE 73
73/74
rris\ o / "7 £
TOo/ 76
m/-\Q / -7 "7
TOo/ / /
— _ — — —


Y





N


N
N
N
N






N
N

N








•


N
N
N
N




N
N
N

N
N
N
N
N
N




N
N
.
N
•
Y
Y
Y
N
N
N

•


N
Y
t
.
N



»
N
,
.
N
N
N
Y
N
Y
Y
Y
Y
N
8/77
N
•
9/77
»
N
N
N
.
•
*

*



Y
(
•
UCR



•
UCR
.
*



UCR
UCR
UCR
UCR
UCR
UCR
UCR
UCR
UCR
UCR
UCR
Y
N


Y
Y
Y

•



	
	
	













	
	
	
	
	


	
	

N






»



AT UCR
AT UCR
,
AT UCR




AT UCR





AT UCR
AT UCR
AT UCR
AT UCR
AT UCR
AT UCR
AT UCR
AT UCR
AT UCR
AT UCR
AT UCR
Y
N


Y
Y
Y

•



Y
Y
Y
UCR



Y

Y
Y



N

N
N
N
N
N


Y
UCR
N



N
N
N

"



N
N
N




N

N
N





N
N
N
N



N









•













































































-------
APPENDIX 2.  ON-LINE DATASET PROGRESS CHECK LIST FOR SBNF DATA BASE AS OF AUGUST 31,  1977 (CONTINUED)
ATRFF^OTT
.L iX.CjJ_i C? \J JL JLt
TREEGR02

*BOGRO-FD
*CMOIST
*POINTS
*GRAV
&MHT QT— TPD
iviu j_ o i r u

MOIST-RD

*H20CURVS
MATRIC
ISURV




BTREE



EGG



REAR



STIK



OHMART
wniiruxj.
OHMART

LAVEN
ARKLEY
ARKLEY
ARKLEY
APVT T7V
/\r\Jxljij j.

ARKLEY

ARKLEY
ARKLEY
DAHLSTEN




DAHLSTEN



DAHLSTEN



DAHLSTEN



DAHLSTEN



fVMPF
VJ1N v»J_j
T08/76
T08/77
12/77
ONCE
ONCE
7
7 *} 7 A
/ -J— / D
77
73-76
77
ONCE
73-76
73
74
75
76
77
73-1/2
74-1/2
75-1/2
76-1/2
73-1/2
74-1/2
75-1/2
76-1/2
73-1/2
74-1/2
75-1/2
76-1/2
73-1/2
74-1/2
75-1/2
76-1/2



.

	
































Y
.
„
N

•
	
„
Y
.
N
N




N



N



N



N





N
N
.
.
N

•
N
.
N
.
N
B
.
^
,
N
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y


N

%
,


•
N
.
N
,

B
,
(
.
N


















N

B
.


*
	
Y
N


.
,
%
*



























10/77



a
B
B
.

.
^
.
.
.
.
*
.
i
.
B
.
B
#
4
•




Y
Y



Y
N
N


Y
Y
Y
Y


.

.


*

.
.
.

.
.
.
•




Y
Y



Y
N
N
Y

N
N
N
N

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y










N






?
7
?
7
?
7
7
?
7
7
7
7
7
7
7
7

SOME


































6/77





































































-------
     APPENDIX 2.  ON-LINE DATASET PROGRESS CHECKLIST FOR SBNF DATA BASE AS OF AUGUST 31, 1977 (CONTINUED)
vO
XRAY



DISU

FASP
SPRMORT 1

SPRMORX
STNDSITE
TREEPEST
SPXTREE
SPRXRGNS
SPRXRGNC
SPRXGRND
SPRXSHRB
SPRMORT2

CONE

GCONE
PLOTSEED
SAPS
LSOIL
SSAS
SLSS
FLDECOMP
LITRKEM

DRIP
SFCSOLKM
DAHLSTEN



COBB

COBB
MCB/D/C

MCB/D/C
MCB/D/C
MCB/D/C
MCBRIDE
MCBRIDE
MCBRIDE
MCBRIDE
MCBRIDE
MCB/D/C

LUCK
.
LUCK
LUCK
COBB
COBB
COBB
COBB
PARMETER
ARKLEY

ARKLEY
ARKLEY
73-1/2
74-1/2
75-1/2
76-1/2
JUN 74
JUN 76
?
76
77
77
77
77
77
77
77
77
77
76
111
73-76
77
SPRING78
76
77
77
77
77
?
73-75
76??
74-75
75






10/77
















Y
N
Y
Y
N




N



N

N
Y
N
N
N
N
N
N
N
N
N
Y
N
,

N
W
N
N
N
N
N
N

•
N
Y
Y
Y
Y
.
•
N
*
N
N
N
N
N
N
N
N
N
*
N
•
N
N
RPT
N
N
N
N
N
Y
N
.
Y




•
•
N
,
N
N
N
N
N
N
N
N
N
.
N
UCR
UCR
UCR

N
N
N
N
N
Y
N
.
Y




Y
Y

	









	

	

	





N





f
9
•
4/77
kill

Y









Y

AT UCR
AT UCR
AT UCR






Y

Y
Y

f
,
•
N
N

7









7

Y
UCR
4/78






N

N
N
Y
Y
Y
Y
N
N

N









N

N








Y

Y
Y
7
?
?
?
























N

N
N

































































































-------
APPENDIX 2.  ON-LINE DATASET PROGRESS CHECKLIST FOR SBNF DATA BASE AS OF AUGUST 31, 1977 (CONTINUED).


LEGEND FOR HEADINGS:

   "VERIF PROGR" IS VERIFICATION PROGRAM (COMPUTER),
   "P.I." IS PRINCIPAL INVESTIGATOR,
   "REDUCT PROGR" IS (DATA) REDUCTION PROGRAM (COMPUTER),

LEGEND FOR COLUMN ENTRIES:

   "*" PRECEDING DATASET NAME INDICATES ORIGINAL DATA USED TO DERIVE SUBSEQUENT DATASET, SO ORIGINAL
       MAY NOT BE DIRECTLY MEANINGFUL TO OTHER INVESTIGATORS.
   "N" IS NO; THIS NEEDS TO BE DONE, BUT HAS NOT BEEN COMPLETELY FINISHED YET (AS FAR AS IS KNOWN),
   "Y" IS YES, THIS STEP HAS BEEN ACCOMPLISHED,
   "RPT" MEANS THAT THIS IS TO BE GOTTEN OUT OF AN EXISTING REPORT,
   "?" MEANS THE STATUS OF THE INFORMATION ITEM HAS NOT BEEN DETERMINED YET,
   "ONCE" MEANS THE DATA ARE ONLY COLLECTED ONCE, NOT REPEATEDLY,
   "UCR" MEANS THAT THIS STEP IS BEING HANDLED AT THE UC RIVERSIDE CAMPUS,
   "8/77" IS THE APPROXIMATE DATE WHEN THE P.I. PLANS TO COMPLETE THE STEP,
   "M10" MEANS THAT DATA FOR SOME PLOTS HAVE BEEN RECEIVED, BUT WE ARE STILL "MISSING 10" PLOTS'
         DATA WHICH P.I. IS STILL PROCESSING,
   "." MEANS PROGRESS HAS GONE BEYOND THIS STEP; LOOK AT COLUMNS TO THE RIGHT	>
   "	" MEANS THIS STEP IS NOT APPLICABLE IN THE MODELLING/DATA MGMT SUBPROJECT; BEING DONE
         ELSEWHERE.

-------
                                  TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 . REPORT NO.
 EPA-600/3-80-002
                                                           3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE

 Photochemical  Oxidant Air Pollution Effects  on  a Mixed
 Conifer Forest Ecosystem
             5. REPORT DATE
              January 1980  issuing  date
             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
O.C.  Taylor, Editor; Principal authors:  R.N.  Kickert,
                                                           8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 University of California, Riverside 92521
 and
 University of California, Berkeley, 94720
             10. PROGRAM ELEMENT NO.

             1AA602    	
             11. CONTRACT/GRANT NO.

             Contract  #  68-03-2442
12. SP.ONSORING AGENCY NAME AND ADDRESS        ,,.    _„
Environmental Research Laboratory-Corvallis,  OR
Office of Research and Development
U.S.  Environmental Protection Agency
Corvallis, Oregon. 97330
             13. TYPE OF REPORT AND PERIOD COVERED
              extramural.  final
             14. SPONSORING AGENCY CODE
              EPA-600/02
15. SUPPLEMENTARY NOTES
 Project Officer: R.G. Wilhour, Environmental  Research  Laboratory, Corvallis, OR 97330
                  FTS 420-4634 (503-757-4634
16. ABSTRACT
          EPA contract 68-03-2442 provided support  for three years of the studies to de-
 termine the chronic effects of photochemical  oxidant air pollutants on a western mixed
 conifer forest ecosystem.  Progress reports were published for years 1974-75 and 1975-7!:
 This  report deals with the year 1976-77 and is  the final  publication on EPA contract
 68-03-2442.  A computer data bank was partially developed in the early years of the
 study at the Lawrence Livermore Laboratory and  was subsequently revised and moved to th<
 computer at the University of California, San Fransisco.   Verification and auditing of
 datasets is underway and several sets are ready for cross-disciplinary analysis for
 modeling.   Computer simulation programs have  been  written for some of the subsections.
 Subsystems which received greatest attention  during this study were: major tree species
 response to oxidant dose, tree population dynamics, tree growth, moisture dynamics, soi
 chemical and physical properties, tree mortality relative to disease, insects and other
 factors, epidemiology of forest tree pathogens  with emphasis on Fomes annosus, cone and
 seed  production, tree seedling establishment, litter production and litter decompositior
 relative to microfloral decomposer populations.  Progress is being made in preparation
 of models  for the purpose of describing the behavior of interlinked subsystems.  Since
 much  progress has been made in verifying accuracy  of data and of identifying informatior
 in the data bank the study of subsystems interaction should be accelerated.
17.
                               KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
COS AT I Field/Gioup
plant  ecology
ecological  succession
plant  growth
forest land
plant  reproduction
forest trees
pine trees
 photochemical oxidants
 conifer ecosystems
 interdisciplinary invest!
   gations
 ecological responses
2/F
6/F
6/C
 8. DISTRIBUTION STATEMENT
    Release to public
19. SECURITY CLASS (ThisReport)
 unclassified
                                                                         21. NO. OF PAGES
20. SECURITY CLASS (Thispage)

 unclassified	
                                                                         22. PRICE
EPA Form 2220-1 (Rev. 4-77)
                                            196

-------