全 文 ：Journal of Forestry Research (2009) 20(4): 337–342
DOI 10.1007/s11676-009-0057-4
Transpiration surface reduction of Kousa Dogwood trees during serious
water imbalance
WANG Fei1, Haruhiko YAMAMOTO2
1. The United Graduate School of Agriculture Science, Tottori University, Tottori 680-8550, Japan
2. Faulty of Agriculture, Yamaguchi University, Yamaguchi 753-8515, Japan
Abstract: The response of Kousa dogwood (Cornus kousa Buerg.) to extreme stresses was investigated by RGB image analysis in the hot,
dry and windy summer in 2007 in Yamaguch, Japan. Results show that tip and margin leaf scorch was observed on many Kousa dogwood
trees and clearly dark brown defense barrier appeared on scorched leaves. The defense barrier withdrew back from distal to proximal gradu-
ally until successful control of scorching, and left a series of unsuccessful defense traces. By responsive analysis of leaf color homogeneity
with RGB image analysis method, a sharp logistic equation was obtained for the relative green/luminance (RGL) value of scorched leaves.
By the meteorological analysis, the occurrence of dogwood leaf scorch-back was almost synchronous with the aridity peak period. It sug-
gested that during the sudden aridity increment the extreme water stresses induce the defense response of Kousa dogwood tree to shear the
excessive transpiration leaf area, and prevent the rest of the trees from further water loss. Image pixel analysis showed that 40.2% leaf area
of sampled dogwood trees was reduced through the partial leaf scorch-back by the end of August in 2007. In contrast, only 13.2% leaf area
was reduced from the same trees in 2008, for the reason of sufficient precipitation during first half year. In any case, the Kousa dogwood
trees indeed reduced their transpiration surface area and appeared a surface reduction pattern differing from those shedding leaves or with-
ering all the aboveground. Based on desiccation process analysis, it is considered that the interaction of the leaf dried back and the
self-defense response was the key of the transpiration surface reduction (TSR) of Kousa dogwood during sudden hot and droughty stresses.
Keyword: aridity peak; Kousa dogwood; leaf scorch-back; logistic responsive function; relative G/L; transpiration surface reduction
Introduction
Plants usually live in the contradict processes of capturing large
amount of carbon and energy at expense of enormous water.
High photosynthesis and carbohydrate production per land area
need additional leaf areas, which imply more water and nutrition
consumption. Most of water absorbed from soil is lost by plant
transpiration and less 5% is used in metabolism and growth.
Therefore transpiration has ever been regarded as an unavoidable
evil since it causes water deficit and injury by dehydration
(Kramer 1983). Acting as transpiration cooler to avoid leaf tem-
perature over rise, causing the ascent of sap and increasing the
absorption of minerals, transpiration is also considered to be
beneficial (Clements 1934). Plant tissues dissipate heat by three
main processes, emission of long-wave radiation, convection of
heat, and transpiration of water. Of which transpiration tends to
be the most effective process of heat dissipation for plant tissues,
Received: 2009-03-31; Accepted: 2009-04-25
© Northeast Forestry University and Springer-Verlag 2009
The online version is available at http://www.springerlink.com
Biography: Wang Fei (1959- ), male, researcher in Shandong Forestry
Research Institute, China, mainly engage in the study in the filed of
ecophysiology of trees and shrubs. E-mail: wf-126@126.com
Responsible editor: Hu Yanbo
particularly at the sunny midday. High plant temperatures
(>40°C) are almost invariably associated with the cessation of
transpiring cooling, following stomata closure in response to
drought (Fitter et al. 2002). Therefore transpiration-cooling fail-
ure during a serious drought stress seems lethal to plants. In ad-
dition, increase in temperature tends to cause an increase in the
rate of transpiration through its effect on saturation water vapor
density (Fitter et al. 2002). Under these conditions, excessive leaf
area usually causes imbalance of energy and water so as to be
dangerous to their lives. Many plant species respond the unfa-
vorably extreme hot and droughty stress by transpiration surface
reduction (TSR) to maintain the water balance of the rest parts.
TSR has been considered as a hydro-ecological factor for a long
time (Orshan 1954; 1989), as the protective response (LIU et al.,
2007) and as an approach of reducing radiation acceptation to
maintain the energy balance of plants (Kozlowski 1973). It can
be seen in various patterns; for instance, leaf or branch shedding
for many deciduous trees even evergreens (Addicott 1973; Rust
et al. 2004; Kozilowski 1976) and the death of aboveground for
most annuals and grasses etc. (Kozlowski 1973; Bhat et al. 1986).
Some tree species respond the unfavorably extreme droughty
environment with partial leaf scorch (Yapp 1912; Treshow 1970;
Vollenweider et al. 2006; Günthardt-Goerg et al. 2007) since the
special leaf structures and adaptive mechanism. This kind of
response can be remarked as a process of partial aboveground
death. The homogeneity between these trees is shrinking-back of
the living parts from distal to proximal to respond the water defi-
RESEARCH PAPER
Journal of Forestry Research (2009) 20(4): 337–342
338
cit and excessive radiation acceptance. Through partially wither-
ing leaves, these trees reach the trade-off among water loss, ra-
diation acceptance and CO2 fixation to survive the extreme hot
and droughty environment. Unlike the plants that respond to
drought stress as a whole, or tissues and organs evenly changed,
many of them showed uneven responses from distal to proximal.
As a result, partial transpiration leaf area of these trees was
sheared. Many landscape trees including bamboos (Sasa sp.),
dogwoods (Cornus sp.), and Oaks (Quercus sp.) even some suc-
culent plant species (Addicott 1982) usually appear similar
symptom.
Kousa Dogwood is a well-known landscape tree species with
showy flower, and widely cultivated along streets and around
houses in Yamaguchi city, Japan. But it often shows leaf scorch
affected by drought, after root injury and transplanting shock etc.
Similar symptoms appeared on many dogwood trees after af-
fected by dry and hot summer in 2007 in Yamaguchi city. Tip
and/or edge leaf scorch occurred on many of dogwood trees,
which made the crown of them discoloration in different scales.
Gradually drying back of living tissues during unfavorably ex-
treme drought stresses took place, accompanying with the oc-
currence of clear defense barrier. During the spring and summer
in 2008, the gradual scorch-backed leaves were also observed on
some transplanting shocked and normal dogwood trees in Ya-
maguchi. The heterogeneousness of leaf scorch-back and leaf
color variation from distal to proximal causes some difficultness
to directly measure the scorched leaves. The flexibility of RGB
image analysis makes it suitable to measure the leaves differen-
tially. In the present study, the leaf RGB images were equally
divided into ten sections from proximal to distal and the
“Switch-off” types of threshold responsive functions (Thornley
1976) were used to describe the gradually scorched leaves. The
leaf water content was also analyzed by similar differential pat-
tern. The TSR process of the dogwood trees during the abnormal
extreme drought events was described quantitatively by using the
image pixel analysis and water content measurement of scorched
leaves.
Materials and methods
Dogwood trees (7-year-old) with height of 3−4 m were observed
to study the leaf scorching response to the extreme drought
stresses in the years of 2007 and 2008. Some newly planted
dogwood saplings were also observed to research the transplant-
ing shock during the early summer in 2008. All of them were
planted around/near a park that is the ancient riverbed and the
former athletic track in Yamaguchi city, Japan. Branch dieback
occurred on many landscape tree species planted around, includ-
ing the dogwood trees. On the stems of some dogwood trees, the
trace of scale insect parasite was also found. It suggested that
improper site condition made the trees sensitive to environmental
changes.
The RGB image of leaf scorch-back for these dogwood trees
came from leaf scanning with a flat bed scanner (Canon D125u2).
Ten leaves were typically sampled from each tree and 60 plants
were mechanically sampled in total. Leaf scorch area percentage
(LSAP), which is the proportion of scorched area to overall leaf
area, was determined by image pixel method. It was measured by
getting pixels of overall leaf and the green part for each leaf with
Photoshop. The LSAP was calculated by Equation (1).
)100
leaf overallfor Pixels
leafofareagreen for Pixels
(100(%)
×−=LSAP (1)
To analyze the characteristics of leaf scorch, the same leaf
images in LSAP calculation were used to measure the G/L value,
an index in RGB image analysis (Wang et al. 2008). Before get-
ting RGB pixel data, the images were hand prepared by eraser of
Photoshop to remove the background and objects except for the
objective leaf. Then leaf images were equally divided into 10
sections from base to tip. The green (G) luminance (L) values for
each section were read from the average histogram value of
Photoshop. The G/L value of leaves was calculated by the Equa-
tion (2).
leaf theof valueluminance
leaf theof egreen valu=leafLG (2)
By repeated regression test, the relative G/L (RGL) decreased
nonlinearly from proximal to distal and could be modeled by
logistic threshold responsive Equation (3) for scorched leaves.
rnae
knRGL −+= 1)( (3)
where RGL (refer to Equation 4) stands for relative G/L. RGL(n)
is the RGL value at n section. Obtained by regression process, r
is the regression coefficient, a is a constant, and k is the maxi-
mum value that the RGL can reach. The character n (=1,2,…,10)
is the number of leaf sections.
)//(
)//(100
minmax
min
LGLG
LGLGRGL ii −
−×= (4)
where G/Li is the G/L value for i section, G/Lmin is the minimum
G/L value of all sections, and G/Lmax is the maximum G/L value
of all sections.
Leaf water relation was also studied by continually measuring
water content before and after scorch-back. Small twigs were cut
from selected trees and then taken back to lab with vinyl-bags.
The water content (WC) for single leaves or leaf sections from
newly transplanted saplings and normal growing trees were
measured by rapid weighing method with an electronic balance
(1/10000 g) at room natural environment. The fresh weight of
sampled leaves, or leaf sections was weighed after sampling
from field site without delay. After obtained the dried weight of
them, the water content was calculated by Equation (5)
%100% ×−=
FW
DWFWWC (5)
Journal of Forestry Research (2009) 20(4): 337–342
339
where FW is the fresh weight of leaf samples and DW is dried
weight of them.
In order to find the relation between TSR and the extreme hot
and drought event, the daily meteorological data during 2007 and
2008 for Yamaguchi Observatory, 1.2 km from the investigated
trees, was obtained from Automated Meteorological Data Acqui-
sition System of Japan, and the first 10 records for maximum or
minimum values from 1976 to 2007 as well. The 11th day’s arid-
ity index (AD11) was calculated by Equation (6) for the mete-
orological data from April 1st to August 31, in both 2007 and
2008, respectively.
∑∑
=
+
=
+=
10
0
10
0
11
j
ji
j
jii PRMTAD (6)
where i=1,2,…,153 and i=1 on the April 1st. MT is the daily
maximum temperature and PR is daily precipitation.
Results and discussion
TSR of Kousa dogwood after persistent droughty weather in
2007
Visible symptoms of the Kousa dogwood caused by extreme hot
and droughty event varied significantly from trees to trees and
among leaves (Fig. 1-a). The threshold responsive equation for
image RGL value of leaves with different scorch area (Fig. 1-a
leaf 2, leaf 3 and leaf 4) presented distinct inverse logistic curves
(Fig. 1-b. leaf 2, leaf 3, leaf 4). This indicated that the injury did
not evenly distribute on the leaves and the scorched area bound
from distal to proximal, which is typical scorch-back characters.
Carefully observed the scorched leaf, apparent defense barrier
existed on the leaf surface and the barrier lines also arranged
from distal to proximal catastrophically (Fig. 1-a, Fig. 1-d). The
shape of responsive function varied from inverse sigmoid shape
to rectangular hyperbola shape (refer to Fig. 1-b) as the scorch
became severe. Meanwhile the leaf area was reduced through
scorching the part outside the barrier. According to the color
analysis of scorched part, only one major defense barrier can be
observed on most of the leaves (Fig. 1-c). For seriously injured
leaves, two (Fig. 1-d) even three or more defense barriers can be
seen. It indicated that the defense barrier withdrew back from
distal to proximal gradually until successfully resist scorching
and left a series of unsuccessful defense traces (Fig. 1-a, Leaf 3,
Leaf 4). The threshold responsive curves for both non-scorched
leaf and entirely scorched leaf (Fig. 1-a Leaf 1, Leaf 5) appeared
a tendency of straight lines slightly slanted and laid on top and
bottom of the coordinate separately (Fig. 1-b), which should be
the characteristics of the trees with no scorched leaves and with
dead leaves separately.
Most of the leaves, appeared multi defense traces, contained
two belts differently colored and separated by two defense barri-
ers (Fig. 1-d). Two scorched periods occurred from the sprouting
of the leaves and showed different responsive function curves.
Despite the characteristic of the defense barrier and leaf
scorched area varied significantly, the total living area of leaves
commonly reduced. Based on the calculation, the image LSAP
was about 40% of sampled leaves in total. The first and second
phases respectively accounted for 2.7% and 38.6% for all of the
sampled trees. Even if the precipitation during the first nine
months was coincident about 40% less than that of normal years,
the relevance between the leaf scorch of Kousa dogwood trees
and less precipitation should be less doubt. It is clear that Kousa
dogwood trees manifested a grass-like response to the summer
drought in 2007, and showed serious leaf scorch-back under the
insufficient water supply. Leaf scorch became serious as the
stresses increase and resulted in decreasing the leaf areas of liv-
ing part, which indirectly decreased the water or precipitation
requirement and received less radiant energy for the living parts
of entire tree. Meanwhile, the green parts of scorched leaves
maintained active status and as the environment became favor-
able they restored vigorous immediately. In the same city, the
green part of scorched leaves of some Japanese blue oak (Quer-
cus glance Thunb.) trees hit by typhoon 0613, maintained normal
function more than two years was observed.
TSR of some Kousa dogwood during transplanting shock in
2008
After transplantation, the successful survival of trees mostly
depends on rapidly establishing the perfect root system. If not,
the new sprouting leaves usually dried out or scorched-back
under sudden drought environment for the reason of serious wa-
ter imbalance. Sufficient precipitation, 116% of the normal, dur-
ing the first half year in 2008 resulted in almost no appearance of
leaf scorch symptoms during early summer days in 2008 on the
same dogwood trees observed in 2007 (Fig. 2a-1). Only some
newly complementarily planted trees showed the gradually leaf
scorch symptoms during the sudden temperature increase and
persistent no rain in the first days of May (Fig. 2a-2, 2a-3).
Leaves from normal growing trees appeared a level curve of
RGL value in Fig. 2b-1; while under the transplanting shock, the
leaves desiccated from tip to base and the uneven appearance
from distal to proximal can be seen from Fig. 2a-2. A black shade
layer, as described by Whitehead (1963), between dried and
non-dried area was observed, and the responsive curve slanted at
tail end (Fig. 2b-2). In this situation, although the leaf tip had
dried out, the color of it still remained green, which indicated
that water loss was too fast to change the chlorophyll. Three days
later, the leaf tips became deep gray (Fig. 2a-3) and a typical
RGL responsive function of inverse logistical curve or scorch
symptom occurred (Fig. 2b-3). During the shock, a lot of seri-
ously hit leaves dried out after several days of temperature in-
crease and no rain weather at the beginning of the May in 2008.
Soon after, the coming of the Japanese rainy season and about
350 mm monthly precipitation in June promoted the new sprout-
ing of small leaflets with long and narrow tips on the tree. Meas-
ured by image pixel method, single leaf area of the new leaves
was only 38.6% of that before the shock. After the end of the
Japanese rainy season in the beginning of July and about ten
Journal of Forestry Research (2009) 20(4): 337–342
340
days of persistent drought and hot weather, the remained leaves
and small new sprouting leaves scorched-back once again (Fig.
2a-4, 2b-4). Some of them also presented two barriers after two
periods of shock (Fig. 2a-5-A, B; 2b-5,). The RGL responsive
lines showed a similar tendency as the first shock during May.
Fig. 1 Variant characteristics of Kousa dogwood leaf scorch from
proximal to distal and defense barriers. Fig. 1-a presented five leaves
from different trees with different leaf scorch area percentage (LSAP), 0, 27.7,
36.9, 59.8 and 100, respectively. Image G/L value ranges from the maximum
(1.258) to the minimum (0.744); Fig. 1-b was the responsive curves of relative
G/L (RGL) for these leaves, with the characteristics of typical logistic curves
for scorched leaves (Leaf 2, Leaf 3 and Leaf 4), and straight lines for overall
green and entire brown leaves (Leaf 1 and Leaf 5); Fig. 1-c showed a leaf with
one defense barrier and its RGL threshold responsive curve; Fig. 1-d presented
a leaf with two barriers constructed in May and August 2007, and related
threshold responsive curve of RGL value. In addition, the threshold function
in Fig. 1-d was established by refilling method or after normally making the
second threshold curve, the first one was established by filling the scorched
area between the first and second barrier with average green value of the leaf.
Fig. 2 Different leaf scorch-back of Kousa dogwood during trans-
planting shock period in 2008; the characteristics of normal leaf (a1),
initial scorched leaf (a2, a4), post scorched leaf (a3), scorched leaf with
single barrier line (a4) and dual barrier lines (a5-A, B). The threshold
responsive curves of relative G/R (RGL) value of normal leaf (b1), initial
scorched leaf (b2, b4), post-scorched leaf (b3), single barrier (b4) and
dual barriers (b5). The responsive threshold curves were made by the
average values at each point.
Water relation during TSR of Kousa dogwood
Leaf scorch-back of normal Kousa dogwood trees extended a
prolonged process. During this process, leaves usually appeared
different degrees of scrolling for a long time and maintained
lower water content than those on normal growing trees (Fig.
3-a). Although there was a tendency of lower distal to proximal
ratio of water content as the hot and droughty conditions per-
sisted (Fig. 3-b), the value of this ratio remained above 1.0 (Fig.
3-b) until a few leaves showed significant leaf scorch-back on a
tree during the sudden hot and dry weather (Fig. 4-a). During this
period of time, some of the leaves appeared evident tendency of
water content variance between leaf tip and base with a regres-
sion line of water content slightly down slanted from proximal to
distal (Fig. 4-a). But the logistic threshold response curve (Fig.
4-b) could not be seen till serious leaf scorch-back occurred on
the tree. It suggests that the ratio of distal to proximal would not
significantly change until it reaches the threshold point of water
content.
Fig. 3 The variation tendency of water content for normal growing
leaves and scrolled leaves (Fig. 3-a) and the water content ratio be-
tween distal and proximal of normal Kousa dogwood trees during
seriously hot and dry summer weather conditions and at the interval
of two times of leaf scorch occurrence (Fig. 3-b). In Fig. 3-b the distal to
proximal ratio was the proportion of the water content between distal and
proximal part of the leaves that were divided into three parts, distal, middle
and proximal. The data in Fig. 3-b was the average value from five trees and
ten leaves for each tree.
Fig. 4 The variation tendency of water content at the threshold status
of leaf scorch-back during the persistently dry and hot summer days
in August 2008. It showed the water content for the leaves before the serious
scorch-back (Fig. 4-a), in which it contained early stage of scorched leaves
(◊–◊, ■–■), seriously wilted leaf (▲─▲) and wilted leaves (∇─∇, ◪─◪).
After the significant scorch-back occurred the significant defense barrier
appeared (Fig. 4-b), which contained the leaves with dual defense barriers
(○–○, ∆–∆) and single defense barrier (●–●). In this figure leaves were
hoof-shapely cut into seven or eight (according to the leaf size) sections from
proximal to distal to reflect the water gradient.
Journal of Forestry Research (2009) 20(4): 337–342
341
According to a large amount of measurement, Kousa dog-
wood is characterized by higher leaf water loss speed than many
other deciduous or evergreen tree species in detached condition.
However, proper interconnection of leaf venation is sufficient to
counteract this shortcoming. It is observed that the main vein
cutting from leaf base cannot suffice to result in leaf water im-
balance of attached Kousa dogwood leaves. The local injury to
the leaf vascular system does not necessarily cause the water
transport obstacle (Kramer 1983). Therefore, the whole sectional
barrier is necessary to interrupt the persistent water loss during
the extreme water imbalance. According to observation, the de-
fense barrier usually appeared during night, which suggests it is a
response to water stress that cannot be completed without ade-
quate water (Kozlowski 1976). Some transplanted dogwood trees
did not appeared the barrier until a rainy day. In some situations,
not only one barrier but also two even more unsuccessful defense
traces can be seen on the same leaf (Fig. 1a-leaf 3, leaf 4). It
seems that the establishment of the defense barrier was an active
protection process. This means that under the serious tension of
enlarged water gradient, leaves responded to the summer drought
by actively shearing the terminal part to protect main body from
further water loss and reduce the transpiration.
Fig. 5 Responsive curves of water content to the leaf sections with
different distances from the leaf proximal (Fig. 5-a, n=5) for de-
tached leaves with different water loss percentages; 12.3% (○−○),
29.5% (□─□) and 45.3% (△─△). It can be seen that detached leaves
also dried from distal to proximal especially by the line of 45.3% water loss.
The visible image of dried leaf and the responsive curves of RGL value for
normal leaves (◇–◇) and natural dried detached leaves (▽−▽)(Fig. 5-b).
In general, the structure and function of plants or trees are
usually identical. By differential measurement of leaf tip and
base of some landscape trees, it showed a tendency that the
proximal part accumulated more biomass than that of distal part,
which showed a ratio of tip/base less than 1.0, some of them
even less than 0.8. This kind of difference should affect the
physical and physiological characters of them, although it ap-
pears the genetic specific. According to measurement, detached
Kousa dogwood leaves showed a tendency gradually dried from
distal to proximal (Fig. 5-a). It suggests that there is a faster
evapotranspiration speed at leaf tip than at base, which was
proved by our half leaf water loss experiment. Meanwhile, the
distal part is usually the farthest from water source of them.
Therefore, it is not surprise that the phenomenon of leaf tip part
of Kousa dogwood dried first was found. No significantly visible
variation appeared on the detachedly dried Kousa dogwood
leaves and their images (Fig. 5-b). Under this kind of acute water
loss situation, there is no time for the leaves to response and no
ability to establish the defense barrier as the intact leaves. It also
indicates that the defense barrier occurred only on the intact
leaves, and the interaction of the leaf dried back and the
self-defense response was the key of the TSR of Kousa dogwood.
The interaction apparently manifested itself in the fact of species
specific and genetic variance. By observation, flower dogwood
(Cornus florida L.) usually appeared premature red leaf during
the extreme summer drought in 2007 and 2008. Only fewer of
red leaf florida dogwood trees showed the leaf necrosis after
persistent hot and drought stresses and almost no defense barrier
had been found on their leaves. However the living part of Kousa
dogwood leaves showed persistent green till late autumn. The
annual leaf necrosis on Kumazasa bamboo (Sasa Veitchii Carr.)
in early winter is characterized with leaf chlorosis from tip to
proximal firstly, and then necrosis started from the seriously
chlorotic leaf tip. Significant defense barrier is usually estab-
lished on their leaves as the same as the Kousa dogwood. Some
flowerbed planted Datura meteloides appeared a typical example
of response process from chlorosis to leaf shedding, which is
different from that of Kousa dogwood and Kumazasa bamboo.
Under the persistent water stresses, tip started leaf wilt, and sub-
sequently scroll and chlorosis were shown in turn. The water
content of the leaves just starting necrosis at the leaf tip showed a
linear decline function from leaf proximal to distal too. However
the tip necrotic leaves gradually fell off and no gradually with-
drawing defense barriers were found.
Triggering weather condition for TSR of Kousa dogwood
Leaf surface area shearing occurs when water absorption cannot
compensate for transpiration losses seriously. It appeared on
many Kousa dogwood trees during the abnormal extreme climate
in 2007. The meteorological environment in 2007 in Yamaguchi,
Japan was characterized by drought and hotness. The annual
precipitation was lower and uneven, which was 71.6% of normal
year and only 60.1% for the first nine months in 2007. The an-
nual mean temperature in 2007 accounts for 107.3% of the nor-
mal year. Particularly, it also made many new records of high
temperature and low precipitation during past 41 years (from
1967 to 2007) for Yamaguchi meteorological observatories. The
protracted drought led to insufficient soil water supply to plants,
especially during sudden hot and droughty days. For example, in
May and August in 2007, two peak periods of the AD11 in
Yamaguchi from April to August were observed (Fig. 6-a) and
consistent well with the period of Kousa dogwood leaf
scorch-back. During the first peak period of sudden increasing of
the aridity, the accumulative precipitation from April to Jun was
only 47.4% of the normal and the precipitation from May 12,
2007 to Jun. 12, 2007 was only 17.6% of total precipitation from
April to Jun. In the second peak period, about 15 days of no rain
and persistent high temperature and low humidity as well as the
high wind speed led to a foehn like weather. Although the pre-
cipitation reached a height of making new record of maximum
Journal of Forestry Research (2009) 20(4): 337–342
342
monthly precipitation in December 2007 in Yamaguchi, it was
late for the utilization of plant growing and out of the growing
season. Various symptoms appeared on many landscape trees,
especially the trees planted on the poor soils and the sites with
root growing limitation etc. During the special weather event of
high temperature and less precipitation in 2007, large amount of
Japanese red pines (Pinus thunbergii L.) in mountain area died.
Many tree species fell off partial leaves to reduce the transpiring
surface. Some sasanqua camellias shed all of leaves to evade the
extreme hot and droughty environment.
Fig. 6-a gave the 11-day moving average aridity (AD11) from April
1st to the end of August in 2007. It clearly presented two peak periods
of AD11, which one occurred during last 10 days in May, and the other
during middle ten days in August. Fig. 6-b was the AD11 index from
April 1st to the end of August in 2008. It also clearly presented two peak
periods. One occurred at the beginning of May and the other from July
4th to August 4th.
In 2008, some newly complementarily planted trees, without
perfect root system established, showed the gradually leaf
scorch-back symptoms. It also occurred during two sudden hot
and dry peak periods (Fig. 6-b). For the normal Kousa dogwood
trees no similar leaf scorch-back occurred until August 2th after
almost one-month persistent drought and high temperature and it
was also consistent well with a persistent peak period of AD11 in
2008 (Fig. 6-b). Although the weather condition and water status
of the Kousa dogwood trees varied, almost all of the scorch-back
symptoms occurred during the sudden hot and dry period (Fig.
6-a, 6-b) and fewer water supplies in common. The sudden in-
creasing of the aridity under the persistent less water supply con-
dition should be the triggering factor for the leaf scorch-back of
the dogwood trees during the dual events of transpiration surface
reduction in the years of 2007 and 2008. Therefore it may be
usable to spray water on the surface of leaves or deep water the
trees before serious drought and hot event occurrence to decrease
the extreme water stress and maintain the normality of the tran-
spiring cooler system as well as support more leaf area to dog-
wood trees.
Acknowledgements
The gratitude will be expressed to the Japan Student Services
Organization for its support.
Reference
Addicott FT. 1973. Physiological ecology of abscission. In: “Shedding of
plant parts”, (T.T. Kozlowski, ed.), New York: Academic Press, 103−104.
Addicott FT. 1982. Abscission. London: University of California Press,
205−207.
Bhat By KV, Sunrendran T, Swarupanandan K. 1986. Anatomy of branch
abscission in Lagerstroemia Macrocarpa Wight. New Phytol, 133: 177−183.
Clements HF. 1934. Significance of transpiration. Plant physiol, 9: 165−172.
Fitter AH, Hay RKM. 2002. Environmental physiology of plants. Landon:
Academic Press, 162−169.
Günthardt-Goerg MS, Vollenweider P. 2007. Linking stress with macroscopic
and microscopic leaf response in trees, New diagnostic perspectives. Envi-
ron Pollu, 147: 467−488.
Kozlowski TT. 1973. Shedding of plant parts. New York: Academic Press,
1−117.
Kozlowski TT. 1976. Water supply and leaf shedding. In: “Water deficits and
plant growth vol 4”, (T.T. Kozlowski, ed.). New York: Academic Press,
191−222.
Kramer PJ. 1983. Water relation of plants. New York: Academic Press,
187−213.
Liu YB, Zhang TG, Li XR & Wang G. 2007. Protective mechanism of desic-
cation tolerance in Reaumuria soongorica: Leaf abscission and sucrose ac-
cumulation in the stem, Science China Ser C: Life Science, 50(1): 15−21.
Orshan G. 1954. Surface reduction and its significance as a hydroecological
factor. J Ecol, 42: 442−444.
Orshan G. 1989. Plant pheno-morphological studies in Mediterranean type
ecosystems. Dordrecht: Kluwer Academic Publisher, 398−399.
Rust S, Roloff A. 2004. Acclimation of crown structure to drought in Quercus
robur L. Fintra- and inter-annual variation of abscission and traits of shed
twigs. Basi App Eco, 5: 283–291.
Thornley JHM. 1976. Mathematical models in plant physiology. Landon; New
York: Academic Press, 48–50.
Treshow M. 1970. Environment and plant response. New York: Mcgraw-Hill
Publications in the Agricultural science, P22–34.
Vollenweider P, Guünthardt-Goerg MS. 2006. Erratum to “Diagnosis of
abiotic and biotic stress factors using the visible symptoms in foliage”, En-
viron Pollu, 140: 562–571.
Wang F, Yamamoto H, Ibaraki Y. 2008. Measuring leaf scorch and chlorosis
of bamboo induced by typhoon 0613 with RGB image analysis. Journal of
Forestry Research, 19(3): 225–230.
Whitehead FH. 1963. Experimental studies of the effect of wind on plant
growth and anatomy. New Phytol, 62: 80–85
Yapp RH. 1912. Spiraea Ulmaria and its bearing on the problem of xeromor-
phy in marsh plants. Ann Bot, 26: 815–70.