Wind exposure is known to have stressful effects on plant growth, particularly at high altitudes. We studied how environmental factors affected carbon assimilation in Pinus pumila needles at a wind-exposed site. Needle gas exchange rates were determined for detached shoots in the laboratory where the needles were free from field environmental stresses, and also determined for attached shoots in the field under in situ environment. There was no difference in gas exchange characteristics determined in the laboratory between shoots from the wind-exposed and the wind-protected sites, suggesting that wind exposure did not affect the photosynthetic potential. In the field, however, the photosynthetic rate of one-year-old needles (in situ Aarea) was significantly lower at the windexposed site than that at the wind-protected site. There was a positive correlation between the in situ Aareaand the xylem pressure potential, suggesting that water deficit caused photosynthetic suppression at the wind-exposed site. The in situ Aarea was lower at the wind-exposed site, even with the same electron transport rate and the same stomatal conductance. These results suggest that CO2 assimilation is suppressed by lower mesophyll CO2 conductance at the wind-exposed site. We conclude that the carbon gain is limited by water stress in wind-exposed regions.
Introduction
With increasing altitude, the environmental conditions become increasingly unfavorable for plant growth, and the number of plants that can survive decreases (Tranquillini, 1964; Körner, 2003). Plants that are adapted to such harsh environments generally exhibit suppressed growth compared with that of plants found at lower altitudes. Although various environmental factors have been suggested as potentially stressful for plants growing at high altitudes, some of these factors are often not general but regional specific, which can be confused with altitude effects (Körner, 2007). Thus, it is necessary to specify the factors that limit plant growth in high-altitude areas.
Wind exposure has been suggested as a potentially important determinant of plant growth and vegetation at high altitudes (Körner, 2003). However, wind exposure is not a simple function of altitude, and it is affected by the microtopography (Barry, 1981). Plants that grow at wind-exposed sites are known to be shorter (Okitsu and Ito, 1984) with lower shoot growth rates (Lawton, 1982), poor vegetation cover, and lower productivity (Baptist and Choler, 2008), which results in a mosaic of vegetation patterns in high-altitude regions.
What are the environmental factors that suppress plant growth at wind-exposed sites? High wind speed may increase evapotranspiration from the leaves and evaporation from the soil, which leads to a water deficit in plants. Stomatal closure to prevent water loss may decrease the CO2 concentration at the carboxylation sites, thereby decreasing the photosynthetic rate (Larcher, 2001). However, the effects of wind on photosynthesis are not consistent in previous studies. Based on a wind exposure experiment where potted plants were placed in a wind tunnel, Caldwell (1970) found that wind exposure decreased the stomatal conductance of Rhododendron ferrugineum seedlings but not Pinus cembra at a wind speed of 15 m s-1, which suggested interspecific differences in the wind response. Cordero (1999) found that the stomatal conductance, transpiration rate, and water-use efficiency of Cecropia schreberiana saplings were not affected by a wind exposure treatment. Snow cover is often absent at wind-exposed sites, and the leaves are exposed to the snow surface where the temperature drops below the freezing point and wind-blown ice crystals may damage the leaf surface. Strong wind causes severe abrasion and removes the wax from leaf surfaces (Hadley and Smith, 1986). Kayama et al. (2009) studied the needle characteristics of reforested Picea glehnii (Sakhalin spruce) on a lowland hillside and suggested that the needles were not protected by snow cover during winter; therefore, they suffered from drought stress. These factors may decrease the instantaneous rate of photosynthesis and the potential rate.
Pinus pumila (Pallas) Regel (Siberian dwarf pine) is a typical dominant shrub species above the forest line in Japan. It exhibits suppressed growth at wind-exposed sites compared with that at wind-protected sites (Okitsu and Ito, 1984; Takahashi, 2005). Nagano et al. (2009) investigated the traits of current-year P. pumila needles that had been grown at wind-exposed and wind-protected sites. Unexpectedly, the photosynthetic rate per unit area at the wind-exposed site was not significantly different from that at the wind-protected site. Does this mean that the growth of P. pumila at wind-exposed sites is not limited by photosynthesis? It should be noted that Nagano et al. (2009) determined the needle photosynthetic rate only in a controlled environment where the needles were free of various stresses; therefore, the photosynthetic rate in the field would not be the same at wind-exposed sites. Furthermore, current-year needles have not experienced the winter season where they might experience severe damage, particularly at wind-exposed sites. The needle lifespan was found to be shorter at wind-exposed sites (Nagano et al., 2009), which suggests that certain stresses can reduce needle longevity.
In the present study, we hypothesized that: (i) photosynthesis would be suppressed more in the field at wind-exposed sites than at wind-protected sites due to water stress; and (ii) the photosynthetic potential of one-year-old needles would be suppressed more at the wind-exposed sites than that at the wind-protected sites because the needles would have suffered greater damage during winter. We determined the gas exchange characteristics of one-year-old P. pumila needles that had grown at a wind-exposed and a windprotected site. First, we studied the potential photosynthesis of current-year and one-year-old needles in laboratory conditions to determine whether the previous experience of a winter reduced the potential photosynthetic rate. Second, we determined the gas exchange rate in the field to determine whether in situ environmental stresses suppressed the photosynthetic rate. In addition, we performed simultaneous measurements of environmental variables to identify the limiting factors for photosynthesis in the field.
Materials and Methods
STUDY SITES AND PLANT MATERIALS
The study sites are located 2770 m above sea level (a.s.l.) on Mount Norikura (36°06′N, 137°33′E; 3026 m a.s.l. at summit) in the central mountain range of Japan. We established two plots, i.e., one on a ridge (wind-exposed site) and another in a col (wind-protected site). The details of the study sites are described in Nagano et al. (2009). P. pumila dominates from 2500 to 3000 m a.s.l. on Mount Norikura. The height of P. pumila differed between the sites, i.e., 0.3–0.4 m and 1.0–1.2 m at the wind-exposed and wind-protected sites, respectively. We tested current-year needles and one-year-old needles that had been exposed to full sunlight and that faced windward, which developed on the leading shoots of mature shrubs that were considered to be more than 20 years old.
MEASUREMENTS IN THE LABORATORY
We conducted gas exchange measurements in the laboratory from 14 to 16 September 2004 using current-year needles, and from 23 to 25 July 2004 and 23 July 2005 using one-year-old needles. One leading shoot per individual shrub was selected from six individual shrubs per site. The shoots (20 cm in length) were cut in water before dawn, inserted into water vials to allow water absorption and brought to the laboratory at Norikura Observatory, Institute for Cosmic Ray Research, the University of Tokyo, located 100 m from the study sites. Photosynthetic measurements of the cut shoots were acquired within 12 h of harvesting. Before the measurements, the shoots were adapted to photosynthetic photon flux density (PPFD) of 300 µmol m-2 s-1. The gas exchange rates were determined using a photosynthetic measurement system (Li6400, Li-Cor, Lincoln, Nebraska, U.S.A.) with an artificial light source (6400-02B, Li-Cor Inc., U.S.A.) at a leaf temperature of 20 °C. The light response curve was obtained when the CO2 concentration in the air entering the leaf chamber was 360 µmol mol-1. PPFD was changed in a stepwise manner (25, 50, 100, 200, 300, 400, 500, 700, 1000, and 2000 µmol m-2 s-1), before the light was finally turned off. The initial slope of the light response curve (A/PPFD) was obtained for the slope of the linear regression at 0–100 µmol m-2 s-1 PPFD. The CO2 response curve was obtained with an increasing CO2 concentration of 50–800 µmol mol-1 (50, 100, 150, 200, 300, 380, 400, 500, 600, and 800 µmol mol-1) at 1000 µmol m-2 s-1 PPFD. The initial slope (IS) of the response curve for photosynthesis based on the CO2 concentration at intercellular space (Ci) was calculated from the slope of the regression at 50–200 µmol mol-1 Ci. For each shoot, three bundles of one-year-old needles were inserted into the chamber. After the measurements, the needles were sampled from the chamber, and their projected area in the chamber was measured using an image scanner. Subsequently, the needles were dried at 80 °C for a minimum of 2 d in an oven and their N content was determined using an elemental analyzer (Vario EL-III, Elementar Inc., Germany). The data for the current-year needles were partially reported in Nagano et al. (2009).
FIELD OBSERVATIONS
Micrometeorological Environment and Soil Water Content
A quantum sensor (IKS-27, Koito Industries, Tokyo, Japan) was connected to a data logger (KADEC UP, Kona-System, Sapporo, Japan) at the midpoint of the study sites placed 50 cm above the ground where it recorded PPFD every minute. The air temperature (Ta ) and relative humidity (RH) were measured at canopy height at each site using a HOBO data logger (H8 Pro Series, Onset Computer Corp., Bourne, Massachusetts, U.S.A.) with a lightshielding hood. The vapor pressure (VP) and vapor pressure deficit (VPD) were calculated based on Ta and RH. The soil volume water content (SWC) was continuously measured at each site using a ML-2 probe and a DL-6 data logger (Delta T devices, Cambridge, U.K.), and on a further two occasions by sampling a soil core using a 1000-mL cylinder. Two ML-2 probes were set in the soil at the wind-exposed site whereas one was set in the soil at the windprotected site. We measured the fresh mass of the soil core samples and the dry mass after drying at 120 °C for 2 d.
Measurements of Gas Exchange, Chlorophyll Fluorescence, and Water Status in the Field
The leaf gas exchange rates of one-year-old needles at the wind-exposed and the wind-protected site were measured using a portable photosynthesis system where sunlight entered a normal leaf chamber. Ambient air was introduced into the leaf chamber through a temporary air reservoir without a desiccant or a soda lime pathway. Measurements were taken on fine days from 30 July to 04 August 2006. We conducted a gas exchange measurement for one-year-old needles in the field on 18 occasions during the experimental period, i.e., 10 measurements were made in the morning (before noon), seven in the afternoon, and one after sunset. During that period, we made repeated measurements using the same needles. Needles on a shoot were selected from the canopy tops of five individual shrubs to make the measurements. The net assimilation rate per unit area (Aarea ), stomatal conductance (gs ), and intercellular CO2 concentration (Ci) were calculated based on the projected total leaf area in the leaf chamber.
The chlorophyll fluorescence yield was measured at the same time as the gas exchange rate using the same needles with a portable pulse-amplitude modulated fluorometer (PAM-2100, Heinz Walz, Effeltrich, Germany). The sets of needles were arranged on the leaf clip in a layer. Before dawn, the quantum yield of photochemistry in the dark (Fv
/Fm
) was measured from 30 July to 5 August. Fv
/Fm
was also measured on 30 July after sunset (20:00 h). During the daytime, the quantum yield was measured in light (
). The electron transport rate (ETR) was calculated as ETR (µmol m-1 s-1) = ΔF/F × PPFD × 0.83 × 0.50 (Schreiber et al., 1994).
The xylem pressure potential (XPP) of one-year-old needles was measured in a pressure chamber (DIK-7002, Daiki Rika Kogyo, Saitama, Japan) with nitrogen gas to determine the diurnal water status of the needles. The needles were cut at the base of a short shoot using sharp scissors and placed in the chamber with the cut end pointing out. XPP was determined in negative pressure by blowing the xylem liquid out of ⅗ needles.
STATISTICAL ANALYSIS
The traits obtained by laboratory measurements were analyzed using a two-by-two factorial design for the site (wind-exposed and wind-protected) and needle age (current-year and one-year-old) to evaluate the interactive effects of site and needle age on the needle gas exchange characteristics of Pinus pumila. The effects of site and needle age on the response variables (Aarea , Amass , gs , and Ci ) were analyzed using a generalized linear model (GLM) with R (R Development Core Team, 2009). The response variables were assumed to follow a Gaussian distribution, and a link function was designated as the identity. The statistical tests of the field measurements of gas exchange and physiological traits were performed using StatView ver. 5.0 (SAS Institute Inc., Cary, North Carolina, U.S.A.). The values were analyzed using a Student's t-test, which confirmed the statistical difference between the windexposed and the wind-protected site. Relationships between the environmental factors and the photosynthetic rate, and between Aarea and other physiological traits, were analyzed using a GLM with R.
Results
PHOTOSYNTHETIC TRAITS IN LABORATORY CONDITIONS
The net CO2 assimilation rate per unit area (Aarea ) measured in the laboratory was not significantly different in the wind-exposed and wind-protected sites, irrespective of the needle age (Fig. 1). The stomatal conductance and initial slopes of the light- and CO2response curves for photosynthesis were also not significantly different in the two sites or at the two ages (Table 1). The net assimilation rate per unit mass was higher at the wind-exposed site than at the wind-protected site (Table 1). The photosynthetic nitrogen use efficiency (assimilation rate per unit leaf nitrogen) was marginally different between the sites, but it did not change with age (Table 1). The needle nitrogen concentration per unit mass did not differ between the sites, but it decreased with age (Table 1). The needle mass per area was lower at the wind-exposed site, and it decreased with age. The nitrogen content per unit area was higher at the wind-protected site, and it decreased with age (Table 1). These results suggest that although some needle traits were different at the two sites, the photosynthetic potential was similar irrespective of the site and age when the rate was expressed on a needle area basis.
FIGURE 1.
Photosynthetic rate per unit area (Aarea ) at 2000 µmol m-2 s-1 photosynthetic photon flux density for current-year and one-year-old needles from the wind-exposed site (▪) and the windprotected site (□). The measurements were conducted in the laboratory (see text). Means and SDs are shown (n = 4 for currentyear needles, and n = 6 for one-year-old needles). The P-values are the statistical significance based on generalized linear model (GLM) analysis. Data for the current-year needles are derived from Nagano et al. (2009).
MICROMETEOROLOGICAL ENVIRONMENTS AND THE SOIL WATER CONTENT
Clear weather prevailed throughout the measurement period. The maximum PPFD reached 2500 µmol m-2 s-1 (Fig. 2, part a). The air temperature (Ta ) did not differ between the sites during the early period but it was higher at the wind-exposed site during the later period (Fig. 2, part b). VP did not differ between the sites for most of the measurements, but there was a large decrease at the wind-exposed site at midday on 1, 4, and 5 August (Fig. 2, part c). Because of the differences in Ta and VP, VPD at midday was greater at the wind-exposed site, particularly during the later period (Fig. 2, part d). SWC decreased with time and was consistently lower at the wind-exposed site (Fig. 2, part e).
PHOTOSYNTHETIC RATE, CHLOROPHYLL FLUORESCENCE, AND WATER STATUS IN THE FIELD
The gas exchange rates varied with the solar irradiance and other environmental factors. In the morning, the in situ Aarea
tended to be lower at the wind-exposed site, whereas it was similar at the two sites during the afternoon (Fig. 3, part a). The mean photosynthetic rates for all measurements were 3.3 ± 1.2 (mean ± SD) and 4.3 ± 2.2 µmol m-2 s-2 at the wind-exposed site and at the wind-protected site, respectively. The stomatal conductance (gs
) tended to be higher at the wind-exposed than at the wind-protected site at midday on 30 July and 3 and 4 August. The intercellular CO2 concentration (Ci
) tended to be higher at the wind-exposed site in the morning of 30 and 31 July, and 2, 3, and 4 August (Fig. 3, part c). The pre-dawn quantum yield (Fv
/Fm
) did not differ between the sites on 30 July, whereas it was lower at the windexposed site on other days. The quantum yield in light (ΔF/F) tended to be lower at the wind-exposed site on 30 July and 2, 3, and 4 August (Fig. 3, part d). XPP was consistently lower at the wind-exposed site during the daytime and pre-dawn (Fig. 3, part e).
TABLE 1
Characteristics of current-year needles in September (partially excerpted from Nagano et al., 2009) and 1-year needles in July at the windexposed site and the wind-protected site. Abbreviations are as follows: Amass , photosynthetic rate at 2000 µmol m-2 s-1 photosynthetic photon flux density (PPFD) per unit mass; gs , stomatal conductance at 2000 µmol m-2 s-1 PPFD; Ci , intercellular CO2 concentration; A/PPFD, apparent quantum yield; IS, carboxylation efficiency; PNUE, photosynthetic nitrogen use efficiency; Narea , nitrogen content per unit area; Nmass , nitrogen content per unit mass; NMA, needle mass per unit area. Means with SD in parentheses are shown (n = 10 for Narea, Nmass and NMA for current-year needles; n = 4 for other characteristics of current-year needles; and n = 6 for all characteristics of one-year needles). Values were analyzed using a two-by-two factorial design for the site (wind-exposed and wind-protected) and needle age (current-year and one-year-old) to evaluate the interactive effects. Results of generalized linear model (F-value) performed to assess the effect of sites and needle age and their significance are also shown. Significant levels: ***, P < 0.001; **, P < 0.01; *, P < 0.05; +, P < 0.10; ns, P > 0.10.
The in situ Aarea tended to be low with low PPFD and low temperature (data not shown). We analyzed the relationships between the physiological parameters and the in situ Aarea at high light (>500 µmol m-2s-1) and a high leaf temperature (>20 °C). The in situ Aarea decreased with XPP at the two sites (Fig. 4, part a). The in situ Aarea was not correlated with ETR, and it was lower at the wind-exposed site when compared at the same ETR (Fig. 4, part b). The in situ Aarea was positively correlated with the gs at each site (GLM, P < 0.001), and it was lower at the windprotected site compared with that at the same gs (Fig. 4, part c). XPP was negatively correlated with Ta , VPD, and SWC (GLM, P < 0.001; Fig. 5).
Discussion
In laboratory conditions, the gas exchange characteristics of Pinus pumila current-year and one-year-old needles did not differ between the wind-exposed and the wind-protected site, when expressed on a needle area basis (Fig. 1; Table 1), which suggested that the potential photosynthesis did not differ between the two sites. The effect of leaf age was also not significantly different (Table 1), which did not support the hypothesis that needles suffer greater damage during winter in wind-exposed habitats compared with those at wind-protected sites. In the field, however, Aarea tended to be lower in the wind-exposed site than that at the windprotected site (Figs. 3, 4). In both sites, Aarea was positively correlated with XPP (Fig. 3). These results suggest that, although the potential was similar, the photosynthetic carbon gain was suppressed more because of water deficiency in the wind-exposed sites than that at the wind-protected sites.
The lower XPP at the wind-exposed site is explained well by the higher VPD and lower SWC (Fig. 5). A high wind velocity may enhance the evapotranspiration potential from the leaves due to a reduction in the boundary layer resistance (Campbell and Norman, 1998), as well as evaporation from the soil; thereby, leading to a decrease in SWC. A decrease in SWC reduced the latent heat flux from the soil, which led to an increase in the air temperature (Kim and Entekhabi, 1998; Dai et al., 1999). Furthermore, we found there was a considerable decrease in water vapor at the windexposed site (Fig. 2). These changes in the air temperature and water vapor resulted in a large increase in VPD at the wind-exposed site, particularly during the later period of our experiment (Fig. 2). Thus, wind-exposed habitats in high-altitude regions influence plant photosynthesis through dehydration.
FIGURE 2.
Temporal changes of (a) photosynthetic photon flux density (PPFD); (b) air temperature (T); (c) vapor pressure (VP); (d) vapor pressure deficit (VPD); and (e) soil volume water content (SWC) in the field. Data were obtained at the wind-exposed site (dotted line) and at the wind-protected site (solid line), although PPFD was measured at an intermediate point between the two sites. For SWC (e), ML2 probe data for WE1 and WE2 at the wind-exposed site are indicated by a dotted line and a dashed line, respectively. The open circles represent the soil core samples from the wind-exposed site. The bold solid line represents WP of the ML2 probe at the windprotected site, whereas the solid circles represent the soil core samples at the wind-protected site.
What was the limiting step during photosynthesis at the windexposed site? We found that the quantum yield of photochemistry was often significantly lower in the needles at the wind-exposed site compared with those at the wind-protected site (Fig. 3, part d). However, its difference tended to be smaller than that in the Aarea. Furthermore, there was no significant correlation between Aarea and ETR (Fig. 4, part b). These results suggest that photoinhi- bition was not a major cause of the decrease in Aarea at the windexposed site. Indeed, in situ Aarea was lower at the wind-exposed site even when it was compared at the same ETR (Fig. 4, part b). At the wind-exposed site, therefore, the reducing power produced by thylakoid electron transport was utilized less for CO2 assimilation and it was consumed more by other processes. If we assume that ETR is proportional to the RuBP (ribulose-1,5-bisphoshate) production rate, this implies that RuBP was consumed more by oxygenation (photorespiration) than by carboxylation. This may occur when the CO2 concentration at the carboxylation site is low (Farquhar et al., 1980). Therefore, the CO2 supply may have limited the photosynthetic rate at the wind-exposed site. However, we also found that the stomatal conductance was not necessarily lower at the wind-exposed site (Fig. 3), and the in situ Aarea was lower when compared at the same stomatal conductance (Fig. 4, part c). These results suggest that stomatal limitation was not a cause of lower assimilation rates. Therefore, it is suggested that CO2 conductance from the intercellular space to the carboxylation site (mesophyll conductance) may have limited photosynthesis in the wind-exposed site. This is consistent with recent studies, which show that mesophyll conductance is reduced by water stress (Flexas et al., 2002, 2008; Niinemets et al., 2009).
FIGURE 3.
Diurnal changes in the (a) net photosynthetic rate per area (in situ Aarea
); (b) stomatal conductance (gs
); (c) intercellular CO2 concentration (Ci
); (d) quantum photochemical yield (ΔF/F in the daytime and Fv/Fm in the nighttime); and (e) xylem pressure potential (XPP) for one-year-old needles of P. pumila at the wind-exposed site (
) and at the wind-protected site (
). Means and SDs (n = 5) are shown. The white and black bars denote the daytime and nighttime, respectively. Asterisks indicate statistical differences between the sites according to the Student's t-test (**, P < 0.01; *, P < 0.05; +, P < 0.10; n, P> 0.10).
An alternative explanation for the low in situ Aarea with a given ETR at the wind-exposed site is that the reducing power was consumed by the water-water cycle, i.e., electrons were transferred from photosystem I to oxygen, and the active oxygen species produced were scavenged using the reducing power (Asada, 1999). This cycle contributes to the dissipation of excess energy under stressful conditions such as high light and water stress (Biehler and Fock, 1996; Yamazaki et al., 2003). The water-water cycle is known to consume up to 15% of the reducing power produced by thylakoid electron transport in watermelons (Miyake and Yokota, 2000) and 10–20% in a tropical tree (Lovelock and Winter, 1996). It is probable that the needles at the wind-exposed site dissipated their excess energy through the water-water cycle in water stress conditions.
Kayama et al. (2009) compared photosynthesis in needles from the upper and lower branches that had been exposed to a prevailing wind during the winter and those that were protected by snow cover in reforested Picea glehnii on a lowland hillside; they showed that photosynthesis was suppressed in the needles from upper branches. They argued that the photosynthetic suppression was caused mainly by stress during the winter. However, our study shows that the photosynthetic potential was not decreased by the winter conditions; instead the rate decreased in response to drought stress during the growing season. This implies that Pinus pumila, which grows at higher altitudes, is better adapted to a winter environment than Picea glehnii.
FIGURE 4.
Relationships between (a) the net photosynthetic rate per area in the field (in situ Aarea
) and the xylem pressure potential (XPP); (b) the in situ Aarea
and electron transport rates (ETR); and (c) the in situ Aarea and stomatal conductance (gs
) of one-year-old needles at the wind-exposed site () and the wind-protected site (). The in situ Aarea at high light (>500 µmol m-2 s-2) with a high leaf temperature (>20 °C) is shown. The regression lines (bold line: pooled; dashed line: the wind-exposed site; and solid line: the wind-protected site) are shown where the effect of the independent variables was significant (P < 0.05) in the GLM analysis.
In our study, the mean photosynthetic rate at the wind-exposed site was 76% of that at the wind-protected site (Fig. 3). Furthermore, the needle lifespan was shorter at the wind-exposed site compared with the wind-protected site (Nagano et al., 2009). Thus, the photosynthetic production may be highly suppressed at windexposed sites. This could explain why the annual shoot growth (Kajimoto, 1993; Takahashi, 2005; S. Nagano, personal observation) and plant height (Okitsu and Ito, 1984) of P. pumila are lower at wind-exposed sites. In the past decade, many studies have demonstrated the significance of sink limitation for plant growth at high altitudes in the context of tree line formation (e.g., Hoch et al., 2002). Sink development may be limited by low temperature, thereby leading to the suppression of plant growth, although the photosynthetic production rate could be higher than the actual growth. In our study, however, the air temperature was similar throughout the season at both sites (data not shown). Therefore, sink limitation was not likely to have been a cause of reduced growth at the wind-exposed site. Thus, our results suggest that the growth of P. pumila at wind-exposed sites was limited by source production rather than sink development. This also suggests that sink or source limits on plant growth may depend on the microtopography even at the same altitude. This may partly explain the discrepancies among studies of sink/source limitations on plant growth at high altitudes (Hoch et al., 2002; Susiluoto et al., 2010).
FIGURE 5.
Relationships between (a) the xylem pressure potential (XPP) and air temperature (Ta
); (b) XPP and vapor pressure deficit (VPD); and (c) XPP and soil water content (SWC) at the wind-exposed site () and the wind-protected site (). Regression lines (bold line: pooled; dashed line: the wind-exposed site; and solid line: the wind-protected site) are shown where the effect of the independent variables was significant (P < 0.05) in the GLM analysis.
Acknowledgments
We thank two anonymous reviewers for their valuable comments on this manuscript. We thank Dr. Masato Takita, Mr. Yoshiaki Agematsu, and the staffs for permission to use the Norikura Observatory, Institute for Cosmic Ray Research, the University of Tokyo; and Dr. Masako Mitamura, Dr. Tsuyoshi Sakata, Dr. Takefumi Ikeda, Mr. Tetsuo Mutsuji, Mr. Teppei Miyake, Dr. Mitsumasa Kubota, and Dr. Jun-ya Yamazaki for technical support. This study was supported in part by the Sasakawa Scientific Research Grant from the Japan Science Society to Nagano (18–212), and by KAKENHI from the Japan Ministry of Education, Culture, Sports, Science and Technology, the Global Environment Research Fund (D-0904) from the Japan Ministry of the Environment, and the Global COE Program “Center for ecosystem management adapting to global change (J03)” of the Ministry of Education, Culture, Sports, Science and Technology of Japan to Hikosaka.
