December 8,
2000
Growth of young Douglas-fir plantations across a gradient in Swiss needle cast severity within north coastal Oregon Douglas A. Maguire Department of Forest Resources Oregon State University Corvallis, OR, USA 97331 Alan Kanaskie and W illiam Voelker Oregon Department of Forestry Salem, OR, USA 97310 Randy Johnson USDA-Forest Serv ice Pacific Northwest Research Station Corvallis, OR, USA 97331 Greg Johnson Willamette Industries Albany, OR, USA 97321 Corresponding author : Doug Maguire
Department of Forest Resources
Oregon State Univers ity
Corvallis, OR 9733 1
Phone:
FAX:
email:
541 -737-42 1 5
541 -737-3049
doug.maguire@orst.edu
Acknowledgments This work was funded by the Swiss Needle Cast Cooperative, the Oregon Department of Forestry, and the USDA­
Forest Service Forest Health Monitoring Program.
Many individuals have contributed to the ideas, data, and
analyses presented in this paper, particularly the foresters, scientists, and students involved in the Swiss Needle Cast
Cooperative: Mark Gourley, John Washburn, Bill Atkinson, John Trobaugh, Greg Filip, Katy Kavanagh, Everett
Hansen, Jeff Stone, Mike Mc Williams, Steve Stehman, Keith Sprengel, Jerry Beatty, Dave Overhulser, LeRoy Kline,
Jack Prukop, Jim Baronek, Steve Dutton, Doug Mainwaring, Ryan Singleton, Minh Quoc Ho, Lou Beer, Chet Smith,
Rick Christian, Dan Manter, Pablo Rosso, Lori Winton, Bryan Capitano, Wendy Sutton, Matthew Goslin, Laina
Smith, Thomas VanDeusen, Nick Greiner, Connie Landis, Jerry Haan, and David Dyer. Field work for Phase I of
the Growth Impact Study was completed by Miller Timber Services: Rick Allen, John Jayne, Arnold Schaffer,
Clarence Stein, Consuello Brandis, Salvador Hernandez, Tim Miller, and Levi Miller. Cooperating landowners for
the north coast plantation surveys included Willamette Industries, Simpson Timber, Starker Forests, Stimson
Lumber, Boise Cascade, The Timber Company (formerly Georgia-Pacific), Confederated Tribes of the Grande
Ronde, Hampton Tree Farms, Miami Corporation, Longview Fiber, USDI Bureau of Land Management, USDA
Forest Service, and the Oregon Department of Forestry.
Growth ofyoung Douglas-fir.
Growth of young Douglas-fir plantations
across a gradient in Swiss needle cast severity
within north coastal Oregon
Abstract
I
2
3
During the past decade, Swiss needle cast (SNC) damage has intensified in many Douglas-fir plantations in
4
the Coast Range of Oregon, particularly along the immediate north coast.
5
growth losses and reduced tree vigor are evident, but the magnitude of growth losses associated with varying
6
intensities of damage is not known.
7
between growth losses and visual symptoms in 10- to 30-year-old Douglas-fir plantations in coastal northwestern
8
Oregon, a population totaling 75,700 ha (187,000 ac). A random sample of 70 Douglas-fir plantations was selected
9
from the population and was evaluated for Swiss needle cast severity. Seventy of these plantations were
In plantations with severe symptoms,
A growth impact study was conducted in 1997 to quantify the relationship
I0
destructively sampled (one 0.02-ha (0.05-ac) plot in each ) to reconstruct past height and basal area growth trends,
11
and to make detailed measurements of stems and foliage.
12
analyzed separately, and the SNC "effect" was assessed by comparing growth of plantations with varying degrees of
13
Swiss needle cast damage to those that were relatively unaffected by the disease, after correcting for other covariates.
14
O f numerous possible SNC indices, mean ne.edle retention (years) exhibited one of the strongest correlations with
15
both basal area and top height growth. Prior to 1990, top height growth was similar across all plantations after
16
correcting for site quality and plantation age; but, b y 1992, top height growth patterns began to diverge among
17
plantations that exhibited a range in SNC damage. In 1996 top height growth was reduced by up to 25 percent of the
18
growth of plantations with little or no SNC. Basal area growth reductions began to appear around 1990, and in 1996
19
basal area growth of the most heavily damaged plantations was 35 percent less than the growth that would be
20
expected in absence of SNC damage. The inferred volume growth loss for 1996 averaged 23% for the full 75,700-
21
h a target population, but this loss averaged as high as 52% for the most severely impacted plantations.
22
Top height growth and basal area growth trends were
Growth ofyoung Douglas-fir.
I
2
Introduction
2
3
Over 2 million hectares (5 mi llion acres ) of commercial timberland in the Pacific
4
Northwest are occupied by young stands (c<:40 yrs) of Douglas-fir, Pseudotsuga menziesii
5
((Mirb.) F ranco) (Oswald et al. 1986). The growth, yield, and overall health of thes e plantations
6
is tremendously important to the economy and environmental quality of Oregon and Was hington.
7
U nfortunately, Swiss needle cast (SNC) has recently emerged as a s erious problem for many
8
Douglas-fir plantations in coastal Oregon (Kanaskie and McWilliams 2000, Hansen et al. 2000).
9
Although the disease was firs t observed in Europe after establishment of Douglas-fir plantations
10
i n Switzerland (Anonymous 1939, Boyce 1940), the fungus caus ing SNC, Phaeocryptopus
11
gaumann ii, was s ubsequently recognized as endemic throughout the range of coas t Douglas-fir
12
(Anonymous 1939, Meinecke 1939, Boyce 1 940). Only recently has concern ar isen about the
13
potential destructive power of SNC within the natural range of Douglas-fir, primarily in res pons e
14
to the increas ing s everity i n SNC symptoms observed in coastal Oregon over the las t 6- 1 0 years .
15
The SNC problem is now regarded as s ufficiently urgent that a maj or research effort has been
16
launched to gain a better understanding of its bas ic biology, its geographic distribution, and its
17
growth impact; much of this research is coordinated through the Swiss Needle C as t Cooperative
18
(SNCC) at Oregon State Univers ity (F ilip 2000).
19
Symptoms of SNC include premature loss of foliage, abundant fungal pseudothecia on
20
needles , yellowing of foliage, s mall average needle s ize, and apparent reduction in di ameter and
21
height growth. Although a range i n symptom severity and ass umed infection intens ity can be
22
observed, the magnitude of growth loss associated with various foliar symptoms has been largely
Growth of young Douglas-fir.
3
unknown. Growth losses are quite apparent upon casual observation in severely impacted
2
plantations , but may be more subtle in zones with fewer or less severe vis ual symptoms . The
3
problem may be pa rticularly critical in young plantations ( 10- 30 years) due to limited
4
s ilvicultural options and the f inancial implications of taking corrective action. Without the
5
option of commercial harvest, re-establishment of productive stands cons isting of res is tant
6
varieties of Douglas-fir or alternative species would cons titute a net expense; yet, without
7
corrective action s ignificant yield losses may continue to accrue. Estimation of potential growth
8
losses for varying levels of infection severity would allow economic analysis of management
9
alternatives, ranging from no treatment to stand re-es tablishment. Without a quantitative link
10
between easily observed SNC symptoms and tree growth, the both local and regional growth
11
impact of SNC intens ification would be at best difficult to assess .
12
Damage fr om SNC has been frequently observed on Douglas-fir grown outs ide of its
13
natural range, f irst, as mentioned above, in E urope in the early part of the century (Boyce 1 940) , 14
and more recently i n both Christmas tree plantations throughout North America (Michaels and
15
Chastagner 1984, Bergdahl and French 1976, Merrill and Longnecker 1 973) and timber stands in
16
New Z ealand (Hood and Kers haw 1 975, Hood et al. 1 990). Although some European research
17
was motivated by concern about growth losses, relatively little work has been accomplis hed to
18
date on es timation of growth impact, especially within the natural range of Douglas-fir and
19
across a gradient in infection severity. Height growth of SNC- affected Douglas -fir in the Pacific 20
Northwest has been estimated at 50-70% of normal, corresponding to as much as a 30% volume
21
growth loss (Hansen et al. 2000) . In New Z ealand, basal area growth of stands experien cing
22
" Douglas-fir decline" was as low as 50% of the bas al area growth of stands exhibiting no decline
Grow1h ofyoung Douglas-fir.
4
1
(Beekhuis 1978). Forest scientists in New Zealand at that time pointed out that, although the
2
Swiss needle cast fungus (Phaeocryptopus gaeumannii) is associated with declining Douglas-fi r
3
stands, it was uncertain as to whether this pathogen was the cause of the decline or an after-effec t
4
of other predisposing factors (see discussion of Decline in Growth, FRI 1978). Another needle
5
cast disease of Douglas-fir, caused by Rhabdocl ine pseudotsugae, was reported to impose radial
6
growth reductions that increased with increasing foliage loss and fluctuated over time, but ranged
7
between 25-75% (Kurkela 1 9 8 1 )
Given the urgent concern about Swiss needle cast in coastal Oregon and the dearth of
8
9
knowledge about the range and variation in growth losses, a retrospective growth analysis was
10
initiated in 1997 to address the fo llowing objectives:
11
1)
12
13
and corresponding growth losses;
2)
14
15
To screen various tree and foliage attributes that may serve as indices of SNC severity
To estimate the current losses of stem volume growth across the observed -range in SNC
severity; and 3)
To gain some insight into past growth trends of plantations that are currently exhibiting 16
varying levels of SNC ( while remaining cognizant of the limitations and pitfalls of
17
retrospective analysis) .
18
19
20
21
22
Methodology
Target population
The target population for the growth impact study was defined by geographical location,
species composition, and plantation age. The approximate geographic center of the region
Growth of young Douglas-fir.
5
1
exper iencing the most severe SNC was Tillamook, Oregon, USA (N45° 25' , W123°50' ). Rainfall
2
in the study area var ies fr om appr oximately 1 80 t o 300 cm per year, and January mean min imum
3
and J uly mean maximum temperat ures r ange fr om -2 to +2° C an d fr om 20 to 28° C, respect ively.
4
Patterns in precipit at ion and t emper at ure are strongly correlated with elevation an d pr oximity to
5
the coast . The st an ds of gr eat est concern were relatively young plantations, due bot h t o the
6
relative sever it y of symptoms t hat were obser ved an d to the limited silvicultur al opt ion s in t hese
7
sub-merchantable st ands. Hence, the target population for in itial growt h impact wor k was
8
d efined as:
9
1)
Douglas-fir plant ations between 10 and 30 year s of t otal age in 1 996;
10
2)
Plant at ion s located within 29 km ( 1 8 miles) of the coast, n orth of Newport (N44°35',
11
W 124°00' ) and sout h of Ast or ia (N46° 10' ,W123°50' ) (see Fig 1).
12
A list of plant at ion s meet ing t hese cr iteria was assembled, and plantation s wer e select ed fr om
13
this list wit h probabilit y pr oportional to size (area). The t arget population included 4,504
14
plant ations covering 75,929 ha ( 1 87,545 ac). The sample included 70 plant at ion s covering 2,620
15
ha (6,472 ac). Each plantation wa s assessed for SNC inten sity and was measured int en sively to
16
establish the link bet ween SNC ratings and growth loss.
17
18
Field work
19
Stands were first visited an d r ated for SNC dur ing March and Apr il, 1997. A f ive-point
20
tr ansect was fir st est ablished fr om a r andom locat ion along an access road. Sample points were
21
separted by 1 5 m, result ing in a 75-m tran sect . Color attr ibutes an d needle retent ion were
22
estimated on the two dominant or codomin ant Doug las-fir trees nearest t o each sample point an d
Growth ofyoung Douglas-fir.
_
6
on opposite sides of the transect line. Needle retention was estimated by dividing the live crown
2
into thirds, with the base of the live crown defined as the lowest live bran ch, an d by examining
3
secondary or lateral branches on a primary or main bran ch in the center of each third. The
4
average number of needle age classes was estimated for each third of the live crown to neat est
5
0.5 years. Likewise, discoloration in each third of the live crown was rated on a scale of 0-3 for
6
each tree, with 0 represen ting n o discoloration and 3 repesenting the most severe discoloration .
7
Slope, aspect, elevation and other site information were also compiled for all plantation s.
8
9
The third sample poin t on the sample tran sects was flagged, marked with a PVC stake,
an d subseq uen tly used as the center of a 0.02-ha plot (0.05- ac; radius=S.O 1 m or 26 .3 ft) on
10
which all trees were measured for dbh (diameter outside bark at 1 .37 m, n earest 0. 1 cm). The
11
four Douglas-fir trees with the latgest dbh and two with the smallest dbh were fe lled, as were the
12
two Douglas-fir closest to the mid-range of the diameter distribution . Each fe lled tree, with the
13
exception of the two smallest Douglas-fir, was measured (nearest 0.01 m) for total height, height
14
to live crown base, an d ann ual cumulative height g rowth (height of each bud scale scar, at or just
15
above each annual bran ch whorl). I dentification of annual whorls was verified by cutting the
16
stem an d coun ting ring s below every fifth whorl. Two stem disks were removed from each of the
17
felled Doug las-fir trees, one at breast height an d one at crown base. These disks were labeled
18
an d stored in a plastic bag for tran sport to the lab. Finally, a sample branch was collected from
19
the fourth largest Doug las-fir tree on each plot. This sample branch was identified as the
20
southern-most branch in the fifth whorl from the top of tree. After measuring the diameter of the
21
branch, i t was cut from the tree flush with the stem and one of the secondary branches in the third
22
whorl down from the primary branch tip was removed, placed in a plastic bag , stored on ice for
Growth of young Douglas-fir.
7
up to 4 days, and then frozen in the lab. The remaining portions of the branch were placed in a
2
plastic bag and transported to the lab for drying and weighing.
3
All initial field work was completed in the spring of 1997, prior to bud break.
4
Supplementary field work on a subset of plots was completed during the following summer
5
(June-September, 1997). During this supplementary field work, plots were revisited and all live
6
Douglas-fir trees were measured for height to crown base and total height (nearest 0.01 m),
7
ignoring any 1997 leader growth. Past height growth (nearest 0.01 m) of the two smallest trees
8
felled during the initial field work was also reconstructed from branch whorls and bud scale
9
scars. An increment core was then removed from each standing Douglas-fir tree to allow an
10
estimate of annual radial growth.
The side of the tree from which this core was extracted was
11
defined by an imaginary line from plot center to the center of the tree at breast height.
12
13
14
Lab analysis
Breast height stem disks were brought to the lab and measured for annual radial growth
15
(nearest 0.5 mm) on the two radii forming the long axis of the disk and on the two radii forming
16
the axis perpendicular to the longest. Four sapwood radii (nearest 0.5 mm) were recorded on the
17
same two axes. Increment cores from the supplementary field work were likewise transported to
18
the laboratory and measured for annual radial growth and sapwood radius (nearest 0.5 mm).
19
Secondary branches were kept frozen until they could be analyzed for specific leaf area by
20
measuring projected area (nearest 0.001 cm2 ) and dry weight (nearest 0.0001 g) of 50-needle
21
subsamples of 1-, 2-, and 3-yr-old foliage separately. Pseudothecia of P. gaeumannii were also
22
counted on 10 needles from each 50-needle subsample, and length of each of these 1O needles
Growth ofyoung Douglas-fir.
1
was recorded (n earest 0.1 mm). Pseudot hecia counts represented the number of occluded
2
stomates out of the cent er-most 100 in the secon d row to the right of the n eedle midrib, with the
3
pet iole pointing down ward. Because each of t hese secondary branches was removed fr om the
4
third whorl down f rom the branch tip, each cont ained three different age classes of shoot s an d
5
needles.
6
8
The remain ing bran ch samples were clipped by age cla ss, dried, an d separat ed into foliage
7
and wood+bark. The samples were t hen redried for at least 24 hours an d weighed (nearest 0.01
8
g).
9
10 Data Analysis
11
Patterns in pa st basal a rea and height g rowt h were reconstructed from the annual height
12
increments and stem sect ion ra dial gr owth measurements. The cross-sect ions of heartwood an d
13
the ent ire stem bot h in side and outside bark were assumed t o be elliptical in shape, an d basal area
14
g rowth an d sapwood area were comput ed accordingly fr om cumulative radial g rowths an d
15
sapwood radii. In the case of t he single increment cores ext ract ed from st anding trees, cross-
16
sect ion s were a ssumed t o be circular.
17
Initial heig ht an d heig ht growth for each annual growth period were reconst ructed for all
18
f elled Dougla s-fir t rees. Similarly, initial diameter outside bark (dob) and in side-bark ba sal area
19
growth for each annual g rowt h period were est imat ed on all Douglas-fir t rees from which a
20
breast height disk or in crement core was removed. Key st an d-level va riables were t hen
21
computed for eac h plot, under the following a ssumpt ions:
22
1)
•
No t rees died fr om suppression mort ality or were removed by thinn ing since the year in
Growth ofyoung Douglas-fir.
which the stands reached breast height; and
1
2
9
2)
Relative basal area growth of each species group was proportional to relative basal area
growth of the sectioned and cored Douglas-fir trees on that plot.
3
4
The analysis was not expected to suffer seriously from the assumption of no mortality in these
5
young plantations, since very little evidence of mortality, especially in Douglas-fir, was observed
6
in these intensively-managed plantations. However, landowner records indicated that at least
7
half of the sampled plantations had been thinned. As a result, some of the past stand densities
8
computed by backdating diameters of trees alive in 1997 are underestimates. Possible
9
ramifications of this latter bias in estimated stand densities are discussed briefly in the
10
Conclusions section. It is also quite possible that the basal area of different species developed
11
more rapidly or more slowly than Douglas-fir basal area, causing an over- or under-estimate of its
12
contribution to basal area in the past.
13
Backdating of stand basal areas was based on the trend in basal area growth of the
14
Douglas-fir trees from which disks or cores were sampled. All species were assumed to conform
15
to the pattern in relative basal area growth observed in these Douglas-fir measurement trees.
16
Specifically,
17
relBAk = relative cumulative basal area growth of measured Douglas-fir trees to year k
18
19
20
21
22
or
I;' llBA;
basal area in year k
2;'996 llBA;
basal area in 1996
Growth ofyoung Douglas-fir.
where LIBA;
1O
basal area growth of all Douglas-fir measure trees on the
=
plot in year i (m2/ha/yr)
2
3
4
The initial basal area of each species group for each annual growth period was computed as
5
6
7
BA,
],
=
reIBA k-1 BAi 1996
•
where BAi
.
.
k
=
initial basal area for species group j in year
8
k, or at the end of year k-1 (m2/ha)
9
relative cumulative basal area growth of
measured Douglas-fir trees through year k-1
10
11
BAi.1996
=
basal area for species group j at end of 1996
(m2/ha)
12
13
Stand density measures for the total stand and for species groups were computed as initial
14
conditions for each annual growth period.
15
In addition to the SNC indices measured in the field (discoloration and needle retention),
16
the tree and sample branch measurements from the fixed-area growth plots allowed computation
17
of many alternative indices of SNC intensity.
18
Height growth and basal area growth, the two major components of volume growth, were
19
analyzed separately since it was evident early in the study that the onset of growth reductions in
20
these two components were quite possibly asynchronous and that their relative magnitudes were
21
different. The four largest trees on each plot were assumed to represent the site tree component
22
of the stands. Site index was estimated yearly for each site tree by solving Bruce's (1981)
Grou1th of yowig Douglas-fir.
11
1
equatio ns iteratively for each height-age pair. Top height growth o f the plo ts was analyzed by
2
co mputing potential predictor variables, including basal area o f co mpeting species, elevatio n and
3
slope-aspect transformatio ns (Stage 1976), and subjecting the data to an all-subsets regressio n
4
analysis with the log arithm o f annual height growth as the respo nse variable. Models were
5
screened bo th for all years co mbined and for 1996 g ro wth only.
6
exhibited bio log ically rational behavior, that was consistent with Bruce's (1981) height growth
7
mo dels, and that pro vided stro ng predictive power was selected for further analysis.
8
9
A subset o f models that
Basal area g ro wth data from all Douglas-fir measure trees, whether estimated from disks
o r cores, were co mbined to assess the "eff ect" o f SNC. As with top height g ro wth, all-subsets
IO
reg ression analyses were run to screen for the potentially mo st po werful variables for predicting
11
the logarithm o f plot basal area growth. Predictor vari ables included basal area of co mpeting
12
species, and site variables such as elevation and slope-aspect transformatio ns (Stage 1976). The
13
mo st pro mising mo dels were selected based on their ability to explain variation in basal area
14
g ro wth and on their co nsistency with current principles about stand density effects o n tree and
15
stand diameter g rowth. Screening runs were implemented on bo th the entire dataset representing
16
all years combined and o n the 1996 g rowth data o nly. The final mo del was fitted on a year-by-
17
year basis, and the " effect" o f SNC was depicted as the g rowth o f plo ts acro ss the range in the
18
SNC severity, relative to tho se plots with the lo west SNC severity.
19
In the final models, the "effect" o f SNC was assessed by fitting the g ro wth mo del o n an
20
annual basis, co mputing the % o f expected g rowth for the range in SNC severity, and g raphing
21
this % o f expected g ro wth fo r all years between 1978 and 1996 in which the SNC index was a
22
statistically significant variable in the reg ression model ( <X=0.05). In this approach g ro wth o f the
Growth of young Douglas-fir.
12
plantations with the least severe SNC ratings serve as the expected growth rate, after correcting
2
for the effects of other covariates in the regression model (that is, holding all other variables such
3
as site quality, stand density, and initial size constant).
4
5
Results
6
7
The 70 growth impact plots ranged between 8 and 30 years of total age, and all were
8
9
Douglas-fir plantations with varying amounts of naturally regenerated western hemlock and other
10
conifer and hardwood species (Table 1 ). Sampled plantations covered a range in total conifer
11
stand density, individual tree condition, and SNC status (Tables 1 and 2).
12
The final model for top height growth was based on Bruce's (1981) model for top height
13
growth in second-growth Douglas-fir. In his model, expected top height growth is a function of
14
current breast height age and site index. Site index for 1990 was selected as the measure of site
15
quality for all analyses in an effort to maximize the probability that this index was not influenced
16
by SNC. Bruce's height growth model was then further modified to illustrate the apparent
17
relationship between SNC severity and trends in top height growth, where average years of
18
foliage retention (retx) proved to be one of the best SNC indices among those tested (Table 2):
19
20
[ 1]
ln(LiH)
=
. <lo+ a, HG + az ln(retx)
·
·
21
22
where In( )
·
=
natural logarithm
Growth ofyoung Douglas-fir.. ..
H
=
annual top height growth (m/yr) 2
retx
=
average years of foliage retention 3
HG
=
Bruce's (1981) predicted top height growth (m) =
0.3048 · ln(hite b
4
·
2
•
b
3
•
13
(piage+13.5-(S4J0.3048)/20)<h3 ·1l) 5
piage
=
average initial breast height age at the start of the growth period 6
hite
=
Sig,,· exp(b2 ((piage+13.25-(S4J0.3048)/20)b3 •
(63.25-(S4J0.3048)/20)b3))
7
8
=
-0.447762 - 0.894427· ((S4J0.3048)/100) +
0.793548 ((S4J0.3048) /100)2 -
9
·
0.171666 ((S4J0.3048)/100)3
10
·
11
=
In(4.5/(S4J0.3048))/((13.25-(S4J0.3048)/20)b3 (63.25 -(S4J0.3048)/20)b3)
12
SI90
13
=
Bruce's (1981) site index computed from plot data backdated to
1990 (m at 50 years)
14
15
16
and
17
a,,, a1,and a, are regression parameters estimated from the data for each year of
18
backdated height growth
19
20
Across years the regression model accounted for 14-63% of the variation in top height growth,
21
and the foliage retention variable (retx) became statistically significant in the top height growth
22
model in 1992 and 1996 only. The plantations implied to be most severely impacted by SNC
Growth ofyoung Douglas-fir.
.
..
14
according to the foliage retention index (retx) were growing at only 75% of the rate observed in
2
"unaff ected" plantations in 1996; that is, growth was reduced 25%. Top height growth trends
3
prior to 1992 appeared totally unrelated to the 1997 SNC ratings. The equation for 1996 was:
4
=
In(AH)
5
6
7
8
9
10
11
12
ln(retx)
-0.738497 + 1 .4265 1 0 ·HG+ 0.21 3098
(0. 1 13089) (0.2 18829) (0.646926)
·
R2 = 0. 136 rmse= 0.221
The final model for plot basal area growth contained fe atures that were typical of growth
13
and yield equations developed for pure even-aged plantations i n other parts o f the world (Clutter
14
et al. 1 983). Variables representing Douglas-fir growing stock (BA0p), competition from other
15
species (BATOT), and stand age (BRAGE) comprised the basic growth model, and the marginal
16
effect o f SNC was evaluated by including mean foliage retention:
17 18
[2]
=
In(ABA) c BRAGE+ c4 ln(retx)
3
19
•
•
R2 = 0.74 - 0.97
20
21
where In( )
=
natural logarithm
22
ABA
=
Douglas-fir basal area growth for the plot (m2/ha/yr)
23
BADF
=
Douglas-fir basal area (m2/ha)
24
BATOT
=
total basal area (m2/ha) 25
BRAGE
=
average initial breast-height age (years) ·
I
I
I
Growth of young Douglas-fir.
2
15
average foliage retention (years)
retx
1
. ..
and
3
c0, c1, c2, c3, and c4 are regression parameters estimated from the data for each year
4
of backdated basal area growth
5
Site index was not included in the model because it was never statistically significant at a=0.05.
6
The basal area growth model accounted for 74-97% of the variation in plot basal area growth in
7
any given year. As expected, basal area growth increased with higher initial Douglas-fir basal
8
area (greater growing stock), but decreased with higher total basal area (competition from other
9
species). Increasing initial plot age was associated with reduced growth, as was a reduction in
10
foliage retention. Those plantations with lower foliage retention in 1997 started to slow in basal
11
area growth around 1990. Growth losses have been as high as 48% relative to expected growth;
12
for example, in 1992, plantations with only a single age class of needles (retx=l) grew 52% of
13
the basal area expected for similar plantations with "normal" foliage retention (retx= 3Y2 years). 14
The equation for 1996 was:
15
16
17
18
19
20
21
22
23
24
ln( BA)
=
-0.082462 + 1.014086 ln(BA0p)
(0.17856986) (0.09746455)
·
-
0.404472 ln(BArnT)
(0.10595974)
·
-
0.096863 BHAGE + 0.300946 ln(retx)
(0.13186040)
(0.01007467)
·
·
R' = 0.737
rmse = 0.257
Volume growth is controlled predominantly by basal area growth and height growth.
25
Although stem form changes are possible due to the reduced foliage density imposed by SNC,
26
they would be very subtle compared to changes induced by large differences in live crown length
Growth ofyoung Douglas-fir.
16
as may result, for example, from variation in stand density. Hence, changes in stem form would
2
not substantially change the growth loss estimated from diameter and height growth effects
3
alone. Taking the values of 75.9% and 67.7% for the percentages of expected height and basal
4
area growth, respectively, associated with the most severe SNC symptoms observed in 1997, the
5
corresponding volume growth for 1996 was estimated at approximately 51% of expected growth:
6
7
D.V
k· D.BA D.H
·
8
k . (0.677 . D.BAexp} (0.759 . D.Hexpl
9
0.513· k · D.BAexp D.Hexp
·
10
11
where D.BAexp and D.Hexp are expected plot basal area growth and top height growth, respectively,
12
at maximum foliage retention. In other words, plantations retaining only a single age class of
13
needles were estimated to have experienced a 49% volume growth loss in 1996 (Table 3).
14
Growth losses prior to 1990 are assumed to have been negligible based on the retrospective
15
analysis, since neither basal area growth nor top height growth varied significantly with foliage
16
retention. Cubic volume growth losses estimated for individual years between 1990 and 1996
17
ranged from 27% to 60% of expected growth (Fig. 2 and Table 3).
18
19
20
Discussion
The mechanisms by which P. gaeumann ii exerts a growth impact may be diverse, and it
21
is possible, if not likely, that the different SNC indices reflect differing aspects of the SNC
22
"effect." For example, it is conceivable that variables such as mean foliage retention could
G rowth ofyoung Douglas·fir..
..
17
1
indicate effects o f fo liage lo ss in the g ro wth analysis, while pseudothecia co unts may signal
2
direct physiolog ical disruption o f needle gas exchange by the presence o f fungal hyphae o r
3
pseudothecia. Alternative SNC indices should therefore no t be co nsidered mutually exclusive.
4
Results fro m this initial analysis suggested, ho wever, that the lo ss o f fo liag e was the predominant
5
mechanism by which SNC reduces g ro wth.
6
Caution must be exercised in attributing growth lo sses so lely to SNC in this retro spective
7
appro ach, since intensity o f SNC may be co nfounded with site o r o ther factors contributing to
8
reduced tree vigor. For example, although co nventional site descripto rs were included in the
9
analysis, site q uality is co mplex in both a static and dynamic sense and, hence, may not be well-
10
11
represented simply by site index and slope-aspect transformations.
Another weak ness o f the described g ro wth analysis is that relatively little is k nown about
12
the natural variation of needle retentio n on Do ug las-fir. Do uglas-fir in the Co ast Range o f
13
Oregon is thought normally to hold four to five years o f fo liage, yet no ne o f the plantations
14
sampled in the SNC growth impact study held foliag e lo nger than four years. On eight sites in
15
the Coast Range and Cascade foo thills, Mitchell (1974) only rarely found young Douglas-fir with
16
more than five years o f needles, yet freq uently observed seven years o f needles o n older trees.
17
Although this finding suggests that fo liage retention may be co nfounded with tree age, the extent
18
to which o ther differences, such as variatio n in nutrient availability, may have influenced the
19
results is not clear. Fo r example, it is impo rtant to no te that many if not mo st o f the yo ung trees
20
in his study were essentially o pen-gro wn, while the branches o n the o lder tree that held seven
21
years o f needles as repo rted by Mitchell (1974) were apparently shaded. Standarizatio n o f
22
foliag e retentio n rating s by crown thirds, as applied in the SNC growth impact study, may
Growth ofyoung Douglas-fir.
I
therefore be quite important. Similarly, inclusion of stand and tree covariates are essential in
2
determining the relationship between the ratings and growth behavior.
18
In contrasting studies from British Columbia, Brix (1981) documented at least six years
3
4
of foliage retention on 24-yr-old Douglas-fir at Shawnigan Lake, and Silver (I 962) found foliage
5
up to 10 years old on four 80-year-old Douglas-fir trees near Victoria. Although 90% of the
6
foliage on trees examined by Silver (1962) consisted of five-year-old or younger needles, the
7
amounts of six- and seven-year-old foliage were considerable. Given the relatively low
8
photosynthetic efficiency of older foliage, however, the relative importance of this older foliage
9
to each tree's total photosynthetic capacity is probably low (Brix 1971, Woodman 1971, Salo
IO
1974). Woodman (1971) measured net photosynthetic rates (mg C02 g-1 h-1) for 1-, 2-, 3-, and 4-
11
yr-old needles that were 72, 50, 30, and 12% of the rates observed for current foliage in Douglas-
12
fir. In a growth modeling context, Mitchell (1975) assumed that 2-, 3-, 4-, and 5-yr-old foliage
13
had relative photosynthetic efficiencies that were 86, 75, 63, and 53%, respectively, of 1-year-old
14
foliage. These efficiencies were incorporated into weights applied to conceptual shells of crown
15
volume representing relative amount of foliage by age class, and yielded reasonable growth
16
predictions. Others have found the relationship between measured assimilation rates over short
17
periods and cumulative diameter or height growth to be rather ambiguous (e.g., Sulzer et al.
18
1993).
19
Artificial defoliation studies seem to support indicate the relatively greater importance of
20
younger foliage, although many of these studies do not distinguish between differences in total
21
amount of foliage and net photosynthetic rates among the age classes. For example, Linzan
22
(1958) removed entire age classes of needle from small saplings of eastern white pine and then
Growth ofyoung Douglas-fir.
l9
I
monitored growth over two subsequent growing seasons. Based on both diameter and height
2
growth, he concludes that !-year-old foliage in white pine is more important to growth than
3
current or 2-year-old foliage. However, probable differences in the total amount of needles in
4
each age class were not addressed, making it difficult to assess the relative influence of net
5
assimilation rate. Also, diameter growth rather than basal area growth was analyzed, with no
6
correction for initial tree diameter. On the other hand, to the extent that we can extrapolate from
7
eastern white pine to Douglas-fir, the treatment in which 2-year-old foliage was removed and the
8
treatment in which both 2-year-old and !-year-old foliage were removed would mimick patterns
9
of foliage loss from Swiss needle cast. For these treatments, the growth reductions relative to
IO
control trees were 55% and 91%, respectively, for basal area growth (estimated from the
11
diameter and diameter growth averages presented), and 45% and 64%, respectively, for height
12
growth. As was the case for Swiss needle cast responses, basal area and height growth responded
13
differently to the defoliation treatments. Osman and Sharrow (1993) also demonstrated the
14
relative insensivity of height growth to artificial defoliation of current-year needles in Douglas-
15
fir. Basal diameter growth, in contrast, declined with increasing intensity of defoliation.
16
A tremendous amount of literature also documents the relationship between differing
17
degrees of insect defolation and subsequent tree growth. Macl,ean ( 1985) provided an excellent
18
summary of published growth losses caused by spruce budworm defoliation. As he points out, a
19
thorough assessment of stand growth loss requires some knowledge of the losses relative to three
20
aspects of defoliation: I) degree and time sequence of defoliation; 2) relative defoliation among
21
constituent tree species; and 3) relative defoliation among constituent age classes and among
22
differing stand structures. Because the spruce budworms feed predominantly on new or current
Growth ofyoung Douglas-fir.
20
foliage, severe defoliation is often associated with complete consumption of current foliage,
2
although backfeeding on older foliage can also be observed during severe outbreaks. Macl..ean
3
(1985) discusses at length two challenges to estimating growth loss that are common to Swiss
4
needle cast as well: measurement of foliage loss (degree of defoliation) in any given year and
5
estimation of potential growth in absence of defoliation. Although numerous approaches have
6
been taken to deal with each, the visual estimate of foliage retention and growth comparison
7
among differing levels of foliage retention (or foliage loss) as applied in the SNC growth impact
8
study has also been applied successfully for assessing spruce budworm growth impact (for
9
example, Baskerville and Macl..ean 1979, Nichols 1988). The biggest difference is the additional
10
challenge associated with estimating foliage loss from SNC since the older age classes are lost
11
first; in contrast, foliage loss from spruce budworm is considerably less ambiguous since it is not
12
complicated by variation in background levels of foliage longevity.
13
In his review of 13 studies, MacLean (1985) found that typical levels bf defoliation
14
during spruce budworm outbreaks led to at least 50-75% reduction in either radial increment or
15
volume growth in balsam fir. In general, growth losses increased as the outbreak progressed; for
16
example, the first year of severe defoliation caused 15-20% growth loss, second-year severe
17
defoliation caused 25-56% growth loss, and further severe defoliation ultimately reduced growth
18
75-90%, or else led directly to tree mortality. Alfaro et al. (1982, 1985) estimated volume
19
growth loss from four successive spruce budworm outbreaks to be 17%, 15%, 8%, and 13% in
20
Douglas-fir, with a cumulative growth loss of 44%. Light defoliation (<20% of current foliage)
21
for five years led to negligible growth loss, whereas severe defoliation (90-100% o f current
22
foliage) caused an average annual volume growth loss of 60%. Maximum volume growth
Growth of young Douglas-fir.
21
I
reductions in balsam fir under severe spruce budworm defoliation have been estimated at 47%
2
(Solomon 1984) and 54% (Archambault 1983), but as mentioned above the reductions depend on
3
both the severity and timing (including duration) of defoliation. One can conclude that the
4
estimated growth losses from SNC seem relatively low compared to spruce budworm impacts,
5
perhaps because in SNC the less photosynthetically active older age classes of needles are lost.
6
Conversely, P. gauemannii mycelia and pseudothecia cause appreciable physiological disruption
7
of gas exchange and photosythesis in the surviving foliage of Douglas-fir (Manter et al. 2000), in
8
tum causing some additional impact on growth.
9
It has also been clearly demonstrated that the greatest reduction in basal area or radial
10
increment from spruce budworm defoliation is in the midcrown and least is near the base of the
11
tree (for example, Mott et al. 1957). This observation may be a direct consequence of the fact
12
that the greatest concentration of current foliage is near midcrown and the fact that basal area
13
increment generally increases down from the tip of the tree in proportion to the increase in
14
cumulative foliage above that point (Kershaw and Maguire 2000). Because the highest
15
concentration of older foliage would be lower in the tree (Hall 1966, Jensen 1976, Kellomliki et
16
al. 1980, Siemon et al. 1980, Kershaw 1993, Kershaw and Maguire 1996), the greatest basal area
17
growth reduction in SNC-affected Douglas-fir would likely be somewhat lower than mid-crown.
18
In addition, the net change in crown form seems to be a general reduction in needle density and a
19
more uniform distribution of foliage vertically, similar to Douglas-fir trees relegated to lower
20
crown classes in unimpacted stands (Maguire and Bennett 1996), as well as lengthening of live
21
crown due to the deeper penetration of solar radiation. These changes may cause stem form to
22
shift from a parabolic to slightly more conic shape under severe SNC. This possible decline in
Growth ofyoung Douglas-fir.
.
..
22
1
form factor would suggest that volume growth losses estimated from height growth and basal
2
area growth at breast height are conservative. Another relevant consequence of this vertical re-
3
distribution of growth is that a lag time of 2-5 years after defoliation can be observed in growth
4
· losses measured at breast height (Macl.. ean 19 8 1 ). Hence, the conclusion that basal area growth
5
losses from SNC began to appear by 1990 may be interpreted to mean that upper stem growth
6
losses could have started occurr ing earlier.
7
In both the top height and basal area growth analyses, pseudothecia counts (p 1 , p2, p3)
8
were not able to account for a significant portion of growth variability when other SNC indices
9
were already in the statistical models. This result may not be too surprising given that a majority
IO
of the growth loss can probably be attributed to premature foliage loss rather than to direct effects
11
of the fungus and its fruiting bodies on surviving foliage, and given that pseudothecia counts are
12
also correlated with foliage retention. Given that " disea se" is defined a s the disruption of normal
13
host functioning or, fr o m a timber production viewpoint, the disr uption of normal height and
14
basal area increment, it may be preferable to mea sure SNC severity by indices such as average
15
foliage retention (retx) . Some other indices performed as well as or slightly better than average
16
foliage retention, for example, the percent of total foliage mass i n 2-yr-old fo liage (pfm2) for top
17
height growth and percent of total fo liage mass in I-yr-old fo liage (pfml) for basal area growth;
18
however, given the meager gain from these variables and the fact that destructive sampling is
19
required to estimate them, foliage retention was judged preferable from a practical standpoint.
20
By convention, foliage retention is assessed just prior to bud break (approximately May 1 in the
21
Coast Range of Oregon). Swiss needle cast i s most detectable from both the ground and air at
22
this time because discoloration reaches its peak and foliage retention is typically at its lowest;
Growth ofyoung Douglas-fir.
23
hence, ratings at this time of year have maximum discriminating power. However, the timing of
2
SNC assessment underscores the fact that the resulting foliage retention is best regarded as an
3
index rather than a quantification of foliage available for photosynthesis during the growing
4
season.
Finally, it is import ant to note that the growth losses associated with a given level of
5
6
foliage retention have fluctuated over the 7-yr period ( 199 0-1996) during which this SNC index
7
was significantly correlated with growth . Growth losses after 1996 a re therefore likely to vary as
8
well, inpart due to interactions with annual weather fl uctuations and in part due to the
9
cumulative eff ects of past losses in foliage, growth vigor, and the a natomical and physiological
10
changes imposed by the disea se.
II
Conclusions
12
13
14
I.
high as 4 8%, and basal a rea growth reductions appeared to have start ed in 1990.
15
16
2.
17
18
Top height growth losses in D ouglas-fir plantations with the most severe SNC were as
high as 25%; significant top height growth reductions appeared to have started in 1992.
3.
19
20
Basal area growth losses in D ouglas-f ir plantations with the most severe SNC were as
Plantations with the most severe SNC lost on average 49 % of their potential volume
growth in 1996.
4.
Failure to detect statistically significant SNC effects prior to 1990 may reflect the recent
21
emergence of SNC as a primary determinant of growth; however, possible art ifacts of the
22
applied retrospective approach cannot be totally dismissed:
Growch ofyoung Douglas-fir.
.
24
I
I.
Uncertainty in reconstructed initial stand conditions for annual growth periods; 2
IL
Uncertainty and possible changes in intensive plantation management practices that were not accounted for in the present analysis; 3
iii.
4
possible shifts in SNC intensity in any one plantation over time; 5
6
Lack of any knowledge about SNC intensities prior to the spring of 1997, and 5.
Thinning effects have not been addressed explicitly in the analyses, but because most of 7
the plantations were thinned within the last 12 years, the reconstructed initial stand
8
densities almost certainly were underestimates. It is also possible that the thinnings could
9
have promoted a temporary decline in stand vigor to some degree (thinning shock, sensu
10
Harrington and Reukema 19 83), or an alteration in microclimate, thereby predisposing
11
Douglas-fir at least temporarily to P. gaeumannii and SNC under the prevailing climatic
12
conditions and inoculum levels.
13
6.
Height and diameter growth of stands exhibiting severe SNC symptoms were tested only
14
against those stands in the sample with no apparent symptoms, that is, those with the
15
longest needle retention. Needle retention is likely influenced by factors other than SNC, 16
so the relative effect of SNC per se on growth as estimated in this analysis is tentative.
17
Also, SNC growth impacts would be additive to other (positive or negative) departures of
18
the "unaffected" plantations from a given landowner's growth expectations, regardless of
19
the cause of the departures or the source of the expectation. 20 21 Growth ofyoung Douglas-fir.
. ..
25
Literature Cited
1
2
3
Alfaro, R.I ., A.J. Thompson, and G.A. Van Sickle. 1985. Q uant ification of D ouglas-fir growt h
4
losses caused by western spruce budworm defoliation using stem analysis. Can. J. For.
5
Res. 15:5-9.
6
Alfaro, R.I ., G.A. van Sickle, A.J. Thompson, and E. Wegwit z. 1982. Tree mort ality and radial
7
growth losses caused by the west ern spruce budworm in a D ouglas -fir stand in British
8
Columbia. Can. J. For. Res . 1 2:780-787.
9
IO
11
12
Anonymous . 1939 . The Swiss leaf-cast of D ouglas -fir. Forestry Abstracts 1 :69 -70.
Archambault , L. 1983. I mpact of defo liation cuased by the spruce budworm on volume growt h
in t hree fir st ands . Can. For. Serv., Ott awa. Research Notes 3 : 17- 19 .
Bas kerville, G.L. and D .A. Macl.-e an. 1979 . Budworm-caused mort ality and 20-yr recovery in
13
immature balsam fir stands . Canadian Forestry Service, Maritimes Forest Research
14
Centre, Frederict ion, N.S. Information Report M-X - 1 02.
15
Beekhuis , J . 1978. Growt h decline in D ouglas-fir. Pp. 1 19- 125 i n R.N. James and E.H. Bunn
16
(eds ) . A review of D ouglas-fir in New Z ealand. Forest Research I nstitute, Rot orua, New
17
Z ealand. FRI Symposium No. 15.
18
19
20
21
22
Bergdahl, D. R. and D .W. French. 1976. Swiss Needlecast of Douglas-Fir in Minnesot a. Plant
Dis. Rept r. 60: 148.
Boyce, J.S. 1940. A needle-cast of D ouglas-fir ass ociated with Adelopus gaumanni.
Phytopathology 3 0: 649-659 .
Brix, H. 19 7 1 . Effects of nitrogen fertilization on phot osynt hesis and respirat ion in Douglas-fir.
Growth ofyoung Douglas-fir.
1
2
3
4
5
6
7
.
26
For. Sci. 1 7:407-414. Brix, H. 1981. Effects of thinning and nitrogen fe rtilization on branch and foliage production in
D ouglas-fir. Can. J. For. Res. 11:502- 51 1 .
Bruce, D . 19 8 1 . Consistent height-growth and growth-rate estimates for remeasured plots.
Forest Science 27:71 1 -725.
Clutter, J.L., J.C. Fortson, L.V. Pienaar, G.H. Brister, and R.L. Bailey. 1983. Timber
Management: A Quantitative Approach. John Wiley & Sons, New York. 333 p.
8
Curt is, R.O. 1982. A simple index of stand density for D ouglas-fir. Forest Sci. 28:92-94.
9
Filip, G.M. 2000. Swiss Needle Cast Cooperative Annual Report. Forest Research Lab, College
10
11
of Forestry, Oregon State University, Corvallis, OR. 52 p.
Forest Research Institute. 1978. D iscussion on decline in growth. Pp. 23-36 in R.N. James and
12
E.H. Bunn (eds). A review of D ouglas-fir in New Zealand. Forest Research Institute,
13
Rotorua, New Z ealand. FRI Symposium No. 15.
14
Hall, G.S. 1966. Age distribution of needles in red pine crowns. Forest Sci. 12:369-37 1 .
15
Harrington, C.A. and D .L. Reukema. 1983. Initial shock and long-term stand development
16
f ollowing thinning in a D ouglas-fir plantation. Forest Science 29 :33-46.
17
Hansen, E.M., J.K. Stone, B.R. Capitano, P. Rosso, W. Sutton, and L. Winton, A. Kanaskie, and
18
M . McWilliams. 2000. Incidence and impact of Swiss needle cast in f orest plantations of
19
Douglas-fir in coastal Oregon. Plant. D is. 84:773-778.
20
21
22
Hood, I. A. and D . J. Kershaw, D . J. 1975. D istribution and infection period of Phaeocryptopus
gaeumannii in New Z ealand. New Z ealand J. For. Sci. 5(2):201 -208.
Hood, I. A., C.J. Sandberg, C.W. Barr, W.A. Holloway, and P.M. Bradbury, 1990. Changes in
Growth of young Douglas fir.
27
needle retention associated with the spread and establishment of Phaeocryptopus
2
3
4
5
6
7
gaeumannii in planted D ouglas-fir. Eur. J. For. Path. 20:4 18-429 .
Jensen, E.C. 1976. The crown structure of a single codominate D ouglas-fir. M.S. Thesis,
College of Forest Resources, University of Washington, Seattle, WA.
Kanaskie, A. and M. McWilliams. 2000. 2000 Swiss Needle Cast Aerial Survey. Oregon
D epartment of Forestry, Salem, Oregon. 3 p. + maps.
Kellomiiki, S., P. Hari, M. Kanninen, and P. Ilonen. 1980. Eco-physiological studies on young
8
Scots pine stands: IL D istribution of needle biomass and its application in approximating
9
light conditions inside the canopy. Silva Fennica 14:243-257.
10
Kershaw, J.A., Jr. 199 3 . Crown str ucture and stem form development in young stands of
11
western hemlock. Ph.D. dissertation, College of Forest Resources, University of
12
Washington, Seattle, WA. 237 p.
13
Kershaw, J.A., Jr., and D .A. Maguire. 1996. Crown structure in western hemlock, Douglas-fir,
14
and grand fir in western Washington: Horizontal distribution of foliage within branches.
15
Can. J. For. Res. 26: 1 28- 142.
16
Kershaw, J.A., Jr., and D .A. Maguire. 2000. Inf luence of vertical foliage structure on the
17
distribution of stem cross-sectional area increment in western hemlock and balsam fir.
18
Forest Science 46: 86-94.
19
20
21
22
Kurkela, T. 19 8 1 . Growth reductions in Douglas fir caused by Rhabdocline needle cast.
Commicationes Instituti Forestalis Fenniae 102: 1 - 1 6.
Linzan, S.N. 1958. The effect of artificial def oliation of various ages of leaves upon white pine
growth. For. Chron. 34:49-56.
Growth ofyoung Douglas fir.
28
MacLean, D.A. 1 9 8 1 . Impact of defoliation by spruce budworm populations on radial and
. 2
3
4
volume growth of balsam fir: A review of present knowledge. Mitt. Forst!. Bundes. Wien
142:293- 306.
MacLean, D.A. 1985. Effects of spruce budworm outbreaks on forest growth and yield. Pp.
5
148- 175 in C.J. Sanders, R.W. Stark, E.J. Mullins, and J. Murphy (eds.). Recent
6
Advances in Spruce Budworms Research. Canada-United States Spruce Budworms
7
Program, Canadian Forestry Service, Ottawa, Canada.
8
9
10
Maguire, D.A. and W.S. Bennett. 1996. Patterns in vertical distribution of foliage in young
coastal Douglas-fir. Can. J. For. Res. 26:1991-2005.
Manter, D.K., B.J. Bond, K.L. Kavanagh, P. Rosso, and G.M. Filip. 2000. Timing and
11
mechanism of impact of the Swiss needle cast fungus, Phaeocryptopus gaeumann ii, on
12
Douglas-fir needle gas exchange and rubisco activation. New Phytologist, in press.
13
14
15
16
17
18
19
Means, J.E. and T.E. Sabin. 1989. Height growth and site index curves for Douglas-fir in the
Siuslaw National Forest, Oregon. West. J. Appl. For. 4: 1 36-142.
Meinecke, E.P. 1939. The Adelopus needle cast of Douglas fir on the Pacific coast. California
Department of Natural Resources, Division o f Forestry. Sacramento. 6 p.
Merrill, W. and J. Longnecker. 1973. Swiss Needlecast of Douglas-fir in Pennsylvania. Plant
Dis. Reprtr. 57:984.
Michaels, E. and G.A. Chastagner. 1984. Distribution, severity, and impact of Swiss needle cast
20
in Douglas-fir christmas trees ii:J western Washington and Oregon. Plant Disease
21
68:939-942.
22
Mitchell, R.G. 1 974. Estimation of needle populations on young, open-grown Douglas-fir by
I
Growth ofyoung Douglas-fir. . . .
1
2
3
4
5
6
7
8
9
10
29
regression and life table analysis. USDA-FS Res. Pap. PNW- 1 8 1 . 14 p.
Mitchell, K.J. 1975. Dynamics and simulated yield of Douglas-fir. Forest Science Monograph
17. 39 p.
Mott, D.G., L.D. Nairn, J.A. Cook. 1957. Radial growth in forest trees and effects of insect
defoliation. For. Sci. 3:286-304.
Nichols, T.J. 1988. The relationship between western spruce budworm defoliation levels and
growth of individual Douglas-fir and grand fir trees. Forest Sci. 34:496-504.
Osman, K.A. and S.H. Sharrow. 1 993. Growth response of Douglas-fir (Pseud otsuga menziesii
[Mirb.] Franco) to defoliation. For. Ecol. & Manage. 60: 105-1 17.
Oswald, D.D., F. Hegyi, and A . Becker. 1986. The current status of coast Douglas-fir timber
11
resources in Oregon, Washington, and British Columbia. Pp. 26- 3 1 in C.D. Oliver, D.P.
12
Hanley, and J.A. Johnson (eds). Douglas-fir: Stand Management for the Future. College
13
of Forest Resources, University of Washington, Seattle, WA. Institute of Forest
14
Resources Contribution No. 55.
15
16
17
18
19
Reineke, L.H. 1933. Perfecting a stand-density index for even-aged forests. Journal of
Agricultural Research 46:627-638.
Salo, D.J. 1974. Factors affecting photosynthesis i n Douglas-fir. University of Washington,
Seattle, WA. Unpublished Ph.d. Thesis.
Siemon, G.R., W.J. Muller, G.B. Wood, and W.G. Forrest. 1 980. Effect of thinning on the
20
distribution and biomass of foliage in the crown of radiata pine. N. Z . J. For. Sci. I 0:46 1-
21
475.
22
Silver, G.T. 1962. The distribution of Douglas-fir foliage by age. Forestry Chronicle 3 8:433-
Growth ofyoung Douglas-fir .
1
2
. ..
30
438_ Solomon, D.S- 1984_ Foliage weight production and bole growth of balsam fir trees defoliated 3
by the spruce budworm. Pp- 25-34 in D.P. Ostaff, M.D. Cameron, and E.J. Mullins
4
(eds.). Proceedings of the Damage Assessment Working Group, CANUSA Spruce
5
Budworms Program, Oct. 25-26, 1983, Bangor, ME. Canadian Forestry Service, Ottawa,
6
Canada.
7
8
Stage, A.R. 1976. An expression for the effect of aspect, slope, and habitat type on tree growth.
Forest Science 22:457-460.
9
Sulzer, A.M., M.S. Greenwood, and W.H. Livingston. 1993. Early selection of black spruce
10
using physiological and morphological criteria. Can. J. For. Res. 23:657-664.
11
12
13
14
Woodman, J .N. 1 97 1 . Variation of net photosynthesis within the crown of a large forest-grown
conifer. Photosynthetica 5:50-54.
Growth ofyoung Douglas-fir_ .
Table 1 .
Attributes of the 70 plots and 788 Douglas-fir trees sampled in 1996 for
retrospective growth impact analysis.
2
3
4
Attribute
Mean sd
Minimum
Maximum
5
6
Plot-level
7
Douglas-fir basal area growth (m2/ha/yr)
1 .72
0.8 1
0.49
3.95
8
Douglas-fir basal area (m2/ha)
15.77 9.02
2.03
37.24
9
Douglas-fir trees per ha
688 148 1 927 347
1 7 . 8 1 6.63
10
Douglas-fir quadratic mean dbh (cm)
11
Basal area in other conifers (m2/ha/yr)
2.39
6.14
0
43.37
12
Basal area in hardwoods (m2/ha/yr)
1 .09
2.63
0
15.31 13
Total basal area (m2/ha)
19.25
2.84
54.97
14
Average breast-height age (years)
12.2
4.9
3.3
28.33 15
Average foliage retention (years)
2.19
0.53
1 .09
3.65
16
•
Site index (height at 50 years, in m)
4 1 .0
5.1
24.4
52. l 17
Conifer relative density!
4.52
2.50
1 .04
1 5.48
18
Conifer stand density index*
426.7 234.6
90.7
1430.2
19
20
21
10.96
5.52 33.56
31
Growth ofyoung Douglas-fir.
1
Table l .
Continued.
2
3
Attribute
Mean sd
Minimum
Maximum
4
5
Tree level
6
Dbh (cm) 14.8 7.7
0.4
7
Height (m)
1 1 .28 4.83
2.09 3 1 .08
8
Height to crown base (m)
2.70
2.75
9
Crown length (m)
8.58
3.62
Crown ratio (%)
78.8
18.1
10
11
12
13
'
Bruce ( 1 981)
14
1Curtis ( 1982)
15
*Reineke ( 1 933); trees per ha @ mean dbh of 25.4cm
16
17
18
0
40.3 14.16
0. 1 6 20.34
1.5
100
32
Growth ofyoung Douglas-fir .
Table 2.
fudices of Swiss needle cast considered in growth impact analysis.
2
fudex
Mean
Range
5
Ave. length of I -yr-old needles (mm)
23.3
13.9-30.9
6
Ave. length of 2-yr-old needles (mm)
24.1
16.9-33.6
7
Specific leaf area of I -yr-old needles (g/cm2)
8
96.7
9
Specific leaf area of 2-yr-old needles (g/cm2)
3
4
69.4
56.5-
61.4
46.5-
IO
94.0
11
Foliage mass on I -yr-old shoots (g)
91.71
7.74-255.70
I2
Foliage mass on 2-yr-old shoots (g)
48.89
0.50-143.18
13
Foliage mass on 3-yr-old shoots (g)
20.08
0-90.38
I4
Foliage mass on I -5-yr-old shoots (g)
162.17
25.38- 439.21
I5
Percent foliage mass in I -yr-old needles 59.4
17.5-98.9
I6
Percent foliage mass in 2-yr-old needles
28.6
0.8-45.5
I7
Percent foliage mass in 3-yr-old needles
10.9
0-38.4
IS
Pseudothecia count on I -yr-old needles
4.0
0-28.6
19
Pseudothecia count on 2-yr-old needles
35.5
0- 80.2
20
Ratio of branch foliage mass to branch basal area (g/mm2)
41.7
10.6-!08.7
21
22
33
Growth of young Douglas-fir. .
1
Table 3 .
Estimated volume growth losses 1990 through 1996 by level of 1996 foliage
retention.
2
3
4
Foliage
5
retention
6
in 1996
7
(yrs)
1 990
1 99 1
1992
1993
1994
1995
1996
9
3.65
0
0
0
0
0
0
0
0
10
3.0
8
8
13
7
7
5
10 8
11
2.5
15
15
23
13
13
9
18
15 12
2.0
23
23 35
20
20 13
27 23 13
1.5
32
32
47
28
28
19 .
37
32 14
1 .0
42
42
60
38
37
27 49
42 Estimated % volume growth loss by year
ave. 1 990-96
8
15
16
34
•
•
•
•
•
OdM8N
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
\
•
•
•
•
•
,,
•
•
1 1!1
•
>1C)C1v v·1·-1 ! ..L
•
•
•
•
•
•
•
•
•
•
•
I
•
•
•
•
-----
Growth ofyoung Douglas-fir . . .
Figure 1 .
35
Locations of sample plantations included in the Swiss needle cast growth impact
study.
3
4
Figure 2.
Trends in estimated cubic volume growth loss (%) for the range in average foliage
5
retention (years) observed on 70 sample plots in 1 996. Growth was not significantly
6
related to foliage retention in any year prior to 1990.
7
8
9
.
I'
Figure 2
· .
��
-0 -r--...... cf?..
--
0
30
20
10
-1-�����l���---i
-i--����J
0
•
2. 5 ·2 0
2. 0 :::I -30
(.)
15
:J
() -60
,
,...-.,,
:"''>--<'""'
\\\\ "" " ·"'
..
..
, ... •;\
. . ..,
,. . ..,,
. . .,.,
..
I
I
'" . .
""
1. 5 -40
- 50
I
•
30
-
0
>
retention "
0
"- - 1 0
O> E
foliage 1.0 _j
1
_
_
_
_
_
_
_
_
_
_
_
-70 -t-�r---,-�-r-�-,.-�.,---,-�-,-�--,-�.,--.�-,-�-.-�-,-�r-,-�-,--j
1 980 1 981
1 982 1 983 1 984 1 985 1 986 1 987 1 988 1 989 1 990 1 99 1
Year
1 992 1 993 1 994 1 995 1 996