Content uploaded by Marcello Biocca
Author content
All content in this area was uploaded by Marcello Biocca on Feb 09, 2015
Content may be subject to copyright.
The Persistence of Oak Decline in the Western North Carolina Nantahala Mountains
Author(s): M. Biocca, F. H. Tainter, D. A. Starkey, S. W. Oak, J. G. Williams
Source:
Castanea,
Vol. 58, No. 3 (Sep., 1993), pp. 178-184
Published by: Southern Appalachian Botanical Society
Stable URL: http://www.jstor.org/stable/4033641
Accessed: 22/01/2009 05:08
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=sabs.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.
Southern Appalachian Botanical Society is collaborating with JSTOR to digitize, preserve and extend access to
Castanea.
http://www.jstor.org
CASTANEA 58(3): 178-184. SEPTEMBER 1993
The Persistence of Oak Decline in the Western
North Carolina Nantahala Mountains
M. BIOCCA,' F.H. TAINTER,2 D.A. STARKEY,3
S.W. OAK,4 and J.G. WILLIAMS5
lIstituto Sperimentale per la Patologia Vegetale, Via C.G. Bertero, 22, 00156 Roma, Italy;
2'5Department of Forest Resources, Clemson University, Clemson, South Carolina 29634;
34USDA Forest Service, Southern Region, Forest Pest Management, Atlanta, Georgia 30367
ABSTRACT
Dendrochronological techniques were used to examine growth differences between
paired "declined" and "healthy" oak trees (subgenus Erythrobalanus) in western North
Carolina. Declined trees exhibited better growth prior to the late 1940's. Following major
growth reductions in the early 1950's, in 1961, and in the late 1960's for both health groups,
annual radial growth of declined trees continued in a downward trend while growth of
trees in the healthy class recovered, albeit at a somewhat reduced rate. By 1991, trees in
the declined group exhibited typical decline symptoms.
INTRODUCTION
During the summer of 1979, an unusually high incidence of oak decline and
death was observed on the Wayah Ranger District, Nantahala National Forest,
North Carolina. Trees were categorized as having either: 1) visibly declined crowns,
or 2) relatively healthy crowns. Investigations by Tainter et al. (1984, 1990) and
Starkey et al. (1989) correlated the onset and occurrence of decline with topo-
graphic position, soil type, and a series of temperature and precipitation extremes.
Examination of annual radial increments measured from these trees revealed
that the two populations had distinctively different radial growth patterns which
were visually separable several decades earlier. The decline study areas were last
visited in 1987 toward the end of a decade-long period of higher than normal
temperatures and lower precipitation. Since harsh conditions have continued
into the present (1991) growing season, it was decided to revisit the areas to
determine if there were any continuing long-term effects of the decline or recovery
in the surviving trees.
MATERIALS AND METHODS
In July 1991, two areas in the Wayah Bald area of the Wayah Ranger District
were revisited. Pairs of dominant and codominant "declined" and "healthy" red
oak trees were selected. First selected were declined trees which exhibited poor
178 CASTANEA VOLUME 58
Table 1. Annual radial increment means of total age of northern red and scarlet oaks,
by health condition
Total
(Scarlet and
Northern Red Scarlet Northern Red)
(mm) (mm) (mm)
Healthy 1.914 1.624 1.775 (c)
Declined 1.882 1.607 1.743 (c)
Total 1.898 (a) 1.615 (b)
(healthy and
declined)
Values with different letters are significant (P > F = 0.001).
Interaction (condition x species) is not significant.
crowns with severe dieback and sometimes with chlorotic leaves of reduced size.
A paired healthy tree of the same species and diameter was then selected in the
immediate vicinity. A total of 30 trees was sampled from the two areas, resulting
in 15 pairs of declined/healthy trees. These included 14 northern red oaks (Quer-
cus rubra L.), 11 scarlet oaks (Q. coccinea Muench), and 5 black oaks (Q. velutina
Lam.). Area 1 had relatively few declined trees and only 4 tree pairs were sampled.
Decline was much more prevalent in Area 2 and 11 tree pairs were obtained with
little difficulty. Diameter was measured at 1.4 m for all trees. Two increment
cores were taken at 1.4 m from each tree on opposite faces and perpendicular to
the slope using a 5 mm Swedish increment borer.
The cores were clamped in special mounting blocks, air-dried for several
days and then glued onto grooved blocks. A flat surface was sanded on each core
to produce a flat transverse face of xylem elements suitable for measuring. Two
cores were subsequently rejected because of defects which made them unsuitable
for measurement, leaving a total of 58 cores for examination.
Total width of each annual radial increment was measured to 0.01 mm from
1991 for the entire age of the tree or as far back as possible (generally to the
early 1900's) using a Bannister incremental measuring device. Sometimes the
pith was missed or knots or other defects close to the pith did not permit readings
in the first few years of age. The increment readings were cross-dated to ensure
that the proper time sequences were maintained. The slower than normal growth
in the early 1950's and especially during 1961 allowed accurate cross-dating of
all ring-width chronologies back to that time. For eighteen trees in which the
core intersected the pith, it was possible to accurately determine age at 1.4 m.
Radial increment data were analyzed using SAS (Statistical Analysis Sys-
tem) analysis of variance procedures for differences in health condition, location,
and species, and the time series by decade (from ca. 1900 to 1991). In order to
estimate differences in the pattern of growth between declined versus healthy
oaks, the annual increment chronologies were divided into two visually distinct
chronology segments. Because the overall chronologies tended to increase until
about 1948 and then decrease after that time, 1948 was designated as a breaking
SEPTEMBER 1993 179
Table 2. Slopes of linear regressions of annual radial increments for two oak species
Period 1900 to 1948* Period 1949 to 1991
Northern Red Scarlet Northern Red Scarlet
Slope 0.028 0.011 -0.028 -0.032
R
2 0.391 0.106 0.372 0.624
* The slopes of the two regression lines are significantly different (P > F = 0.0001).
point to separate the chronologies into two separate data sets for the subsequent
linear regression analyses.
RESULTS
The average diameter of all trees at 1.4 m was 36.7 cm. The average age of
declined trees for both areas was 79 years and for healthy trees was 90 years.
Analysis of variance of mean annual increments of total age of scarlet oaks versus
northern red oaks revealed a significant difference for means of total annual
increment for each species. The difference between declined and healthy mean
total increments of both species combined was not significant, nor was the species
condition interaction (Table 1). Northern red oak always grew better than scarlet
oak but growth of both species declined similarly in the period 1949 to 1991
(Table 2).
A comparison of mean annual radial increment by decade between declined
and healthy trees for the two areas revealed significant growth differences in
certain decades and for the interaction between condition and area (Table 3).
Periods of maximum and minimum annual radial increment were closely
Table 3. Differences between annual radial increment means by decade'
Condition2 Area3 Interaction
Decade (mm) (mm) (Condition x Area)
1900-094 0.252 n.s. -0.008 n.s. *
1910-19 -0.157 n.s. -0.010 n.s. n.s.
1920-29 0.147 * 0.568 ** n.s.
1930-39 0.303 * -0.086 na. **
1940-49 0.303 ** -0.005 n.s. **
1950-59 -0.056 n.s. 0.460 * n.s.
1960-69 -0.129 n.s. 0.381 ** n.s.
1970-79 -0.285 ** 0.026 na. **
1980-89 -0.594 ** 0.222 ** **
1990-915 -0.737 * -0.039 n.s. n.s.
I * = significant at P = 0.05; ** = significant at P = 0.001; n.s. = not significant.
2Condition = (radial increment mean, trees declined) - (radial increment mean, trees healthy).
3Area = (radial increment mean, trees in Area 1) - (radial increment mean, trees in Area 2).
4Number of observations are >20.
5Number of observations are 4.
180 CASTANEA VOLUME 58
4)4
C.)
bl)
-
$-4
4-)
c)
-t 0
b4)
WW
Nl H10IM UNI-
SEPTEMBER 1993 181
Table 4. Slopes of annual radial increment trends obtained by linear regression
Period 1900-48* Period 1949-91 **
Declined Healthy Declined Healthy
Slope 0.031 0.020 -0.044 -0.025
(declined trees grew faster than (growth of declined trees slowed
healthy trees) faster than growth of healthy
trees)
R
2 0.424 0.390 0.734 0.488
* The slopes of the two regression lines are significantly different (P > F = 0.008).
** The slopes of the two regression lines are significantly different (P > F = 0.001).
correlated for declined and healthy trees on both areas, with noticeable growth
minima in 1911, 1925-26, 1930-33, 1952-54, 1961, and 1984-86 (Figure 1). Start-
ing in approximately 1920 and continuing through the 1940's, declined trees
experienced a somewhat better annual radial growth than did their healthy
counterparts. A separation of growth chronologies between declined and healthy
trees was the most striking after 1971 although some distinction appeared as
early as 1958. Both healthy and declined populations experienced abruptly re-
duced growth in the early 1950's for a 3-year period. Healthy and declined pop-
ulations rapidly recovered, however, and grew similarly, although at a decreasing
rate, until 1961, when growth rate of both populations dropped even more pre-
cipitously. This growth reduction appears to have been the climax of an entire
decade of stress and appears to have been critical in the subsequent formation
of the two major health groups. The decline group recovered at a somewhat lower
rate compared with the healthy group, but in the late 1960's both populations
suffered another decrease in growth from which the declined trees never recov-
ered. After 1971, growth of healthy trees appeared to hold at approximately 1.5
mm/year. In declined trees, however, annual growth steadily decreased to just
over 0.5 mm/year by 1990. The 1991 increment cannot be considered as it rep-
resented only the early portion of that growing season.
Using linear regression analyses (Table 4), tests of comparison of slope of
the regression lines showed that the regression lines for both declined and healthy
chronologies were significantly different (P < 0.008) before 1948 and after 1949
(P < 0.001). Slope of the regression line for declined trees increased at a greater
rate before 1948 and decreased at a greater rate after 1949 when compared to
healthy trees. The coefficient of determination of declined trees (R2 = 0.73)
showed relative homogeneity in the response of these trees.
DISCUSSION
Relative health of these oak trees was easily discernible based on signs and
symptoms of the crowns (Starkey et al. 1989); thus, healthy versus declined trees
could be visually determined. The growth chronologies are in general agreement
with earlier ones generated from paired trees growing in similar areas (Tainter
et al. 1984, 1990). Although both healthy and declined populations in Area 1 had
182 CASTANEA VOLUME 58
been experiencing a general growth decline since 1971, there is some evidence
from the chronologies that a new declined population formed in response to an
additional stress period after 1985. These trees experienced an increased exac-
erbation in 1986, apparently in response to some stress possibly in 1984 or 1985.
A severe regional drought and record high temperatures occurred during part of
that period (Burnett 1987). From December 1, 1985 to July 31, 1986, for example,
total precipitation at the Asheville, North Carolina weather station was 42% of
normal.
Why two distinct populations of quite different relative health should form
in response to stress is not easy to explain. Perhaps some root pathogen such as
Armillaria spp. gained advantage of the declined trees during this time. Surveys
have not addressed this possibility although field observations have routinely
identified specific signs of Armillaria root disease on dead trees in this area.
Another possible explanation is that declined trees are occupying microsites
with poor fertility or soil moisture characteristics. In these cases, the selection
of the nearest paired tree may be inadequate to estimate differences in potential
soil productivity.
A comparison of growth increments suggests that decline affects scarlet oak
and northern red oaks in the same manner. Although the different behavior of
the two species may not be related to their different genetic backgrounds, it is
possible that the differing growth patterns between these two species was based
on characteristics not closely related to the decline syndrome. In fact, radial
growth in trees is not a good indicator of their genetic characteristics, but it is
an excellent index of the response of the trees to the environment (Kramer and
Kozlowski 1960). Tree age is another factor that may play a role in the decline
complex. Analysis of the range of ages of the two populations shows that the
declined population is almost even-aged and a decade younger than the healthy
population. This even-aged group could have experienced several cycles of stress-
es in the past at approximately the same age. The healthy population shows a
wider range of ages perhaps because they survived earlier stressful periods when
they were not at a sensitive age. This is very plausible for younger trees but is
less likely for older trees that are still healthy, because the effects of the recent
droughts should have especially affected these old trees if they were at a more
sensitive age. This group of healthy and older trees may form a third group of
trees that tolerate drought because of different genetic characteristics or microsite
conditions. Healthy trees during favorable periods (i.e., during the late 1930's
and 1940's) show less growth than that of declined trees at the same period. It
is possible that during this period healthy trees have "chosen" a resources-
allocation policy oriented toward support of defenses rather than growth. This
might be because: 1) they were growing on sites favorable for light competition;
2) they were surviving a mild stress that acted as a "hardening" factor; or 3)
they showed different genetic adaptation. The advantage experienced by healthy
trees during more favorable growth years in terms of resistance or defense may
have proven to be useful in the long-term. In fact, in the first period of drought
(the 1950's), the healthy and declined trees show the same growth. The 1950's
drought permanently "altered" the declined trees, and although they were able
SEPTEMBER 1993 183
to tolerate this first important period of crisis nearly as well as the healthy trees,
further crises during the 1960's and 70's revealed their sensitivity.
Another "stress" was created following the arrival of chestnut blight during
the 1930-1960's. Chestnut blight was very severe in the Nantahala Mountains.
The death of chestnuts created new spaces for oaks that quickly replaced the
dying chestnuts and in many instances on sites where oaks were not well adapted.
So, the present decline may be a consequence of the short life cycle of oak trees
growing on sites not entirely suitable for high-density composition of old oak.
LITERATURE CITED
BURNETT,
H. 1987. The great drought of 1986. Amer. For. May/June:22-25, 79.
KRAMER,
P.J. and T.T. KOZLOWSKI. 1960. Physiology of trees. McGraw-Hill Book Com-
pany, New York. 642 p.
STARKEY, D.A., S.W. OAK, G.W. RYAN, F.H. TAINTER, C. REDMOND, and H.D. BROWN.
1989. Evaluation of oak decline areas in the South. USDA For. Serv., Prot. Rept.
R8-PR17, 36 p.
TAINTER, F.H., S.W. FRAEDRICH, and D.M. BENSON. 1984. The effect of climate on growth,
decline, and death of northern red oaks in the western North Carolina Nantahala
Mountains. Castanea 49:127-137.
TAINTER, F.H., W.A. RETZLAFF, D.A. STARKEY, and S.W. OAK. 1990. Decline of radial
growth in red oaks is associated with short-term changes in climate. Eur. J. For.
Path. 20:95-105.
Received September
25, 1992;
Accepted March 15, 1993.
184 CASTANEA VOLUME 58
