J For Res (2003) 8:163–168 DOI 10.1007/s10310-002-0022-y
© The Japanese Forestry Society and Springer-Verlag Tokyo 2003
ORIGINAL ARTICLE
Akira Mori · Hiroshi Takeda
Light-related competitive effects of overstory trees on the understory conifer saplings in a subalpine forest
Received: August 5, 2002 / Accepted: December 20, 2002
Abstract To clarify the mechanism by which overstory trees shade understory saplings, we investigated the relationships among light conditions of the saplings (measured as indirect site factor; ISF and direct site factor; DSF), the calculated competition effects of overstory trees on the saplings (W), and relative height growth rate of the saplings (RHGR). We calculated several W values in order to find a W value which can express the light conditions as appropriately as possible, and the results indicated that W explained only 21.9%–24.7% of the total variance of light conditions in the cases where W gave the best fit. In this study, W was calculated based on the basal areas of overstory trees. However, it is known that canopy structure also affects the light regimes in the forest understory, and this might yield the possible errors even within W representing the shading effects most adequately. Therefore, although W significantly represents the shading effect from overstory trees, a great proportion of the variance remained without being explained by W. RHGR was negatively correlated with W, and the W value which had the most adequate explanation of the shading effect also showed the best negative correlation with RHGR. This provides the evidence that the competitive effect of overstory trees on sapling growth is mediated by the shading effect, indicating that competition for light clearly exists within this forest. Such competition for light may closely relate to the well-known phenomenon of gap regeneration in subalpine forests in central Japan. Key words Abies · Competition · Local crowding · Picea · Shading effect
A. Mori (*) · H. Takeda Laboratory of Forest Ecology, Division of Environmental Science and Technology, Graduate School of Agriculture, Kyoto University, Oiwake-cho, Kitashirakawa, Sakyo-ku, Kyoto 606-8502, Japan Tel. 81-75-753-6080; Fax 81-75-753-6080 e-mail:
[email protected]
Introduction Plant neighbors influence growth, reproduction, and mortality of individual plants (Mithen et al. 1984; Weiner 1984; Pacala and Silandar 1985; Thomas SC and Weiner 1989; Takahashi 1996; Wada and Ribbens 1997; Umeki and Kikuzawa 2000). Since light, which is one of the critical determinants affecting plant performance, comes directionally from above, larger plants have suppressive effects on smaller ones by shading (Kikuzawa and Umeki 1996). Therefore, plant competition for light is generally asymmetric or one-sided (Weiner and Thomas 1986; Thomas EM and Weiner 1989; Thomas SC and Weiner 1989; Weiner et al. 1990; Weiner and Thomas 1992; Peterson and Squiers 1995), and thus researchers have suggested the significant relationships between plant growth and plant competition with reference to shading (Weiner et al. 1990; Kubota and Hara 1996). Competitive interactions have been evaluated by using local crowding models as readily available competition indices, which are calculated using positional information and neighbor size (e.g., Thomas SC and Weiner 1989; Peterson and Squiers 1995). In the consideration of the competitive effects of larger neighbors on the performance of smaller plants, calculating overstory local crowding must provide useful information. Plant performance reflects the actual resource availability so that individual response to overstory local crowding can be regarded as a response to the local light availability. However, the relationship between local crowding and the light condition of individual plants remains unclear. Evaluating this relationship contributes to clarifying how larger plants affect the local light availability of smaller plants. In developing forest, competition for light is the predominant factor (Gilbert et al. 2001). Therefore, it is necessary for the study of forest dynamics to clarify the competitive effects of overstory trees related to the understory light regime in terms of shading effects. In this study, we evaluate the relationship between local crowding of overstory trees and understory light regimes and the effects of
164
local crowding on the growth of understory saplings. From these evaluations, we aim to clarify the mechanism of competition in which overstory trees shade understory saplings.
Materials and methods Study site and field methods The study site is located within a subalpine forest (altitude 2050 m, 35°56N, 137°28E) on Mt. Ontake (3067 m high) in Central Japan. The species composition is shown in Table 1. This subalpine forest is dominated by two Abies species (Abies mariesii and Abies veitchii) and Picea jezoensis var. hondoensis (Table 1). A 50 145 m (0.725 ha) study plot was established in June 1999. A large gap that formed in fall 1994 is located at the western side of the plot. However, the central part of the plot was also disturbed by a typhoon in September 1999, so we chose the two undisturbed parts (called subplot A and subplot B) and only the spatial data from these subplots were used. Subplot A is 25 30 m and subplot B is 35 40 m (0.215 ha in total). The edge of the disturbed area in 1999 was about 30 m from each subplot, so that we considered that the disturbance in 1999 had no effect on the structures and analyses of the two subplots. We measured the x and y horizontal coordinates, height in 1995 and 2000, and diameter at ground level (DGL) of all saplings of the three conifers (A. mariesii, A. veitchii, and P. jezoensis var. hondoensis) in the two subplots in August and September 2000. In this study, saplings were defined as individual trees of A. mariesii, A. veitchii, or P. jezoensis var. hondoensis, which were from 5 to 200 cm in height. For all overstory trees (200 cm height) within the plot, we measured the x and y coordinates, height (H), and diameter at breast height (DBH) in September and October 2000. Height of overstory trees was measured using the IMPULSE laser range finder device (Laser Technology, Englewood, CO, USA).
photographs were taken 2 m above ground at 2.5-m intervals in mid-September 2000. Photographs were made under overcast sky conditions, using a Coolpix 910 digital camera with a FC-E8 fish-eye lens (Nikon, Tokyo, Japan). The camera was kept in a horizontal position with a leveling device. From the photographs, the indirect site factor (ISF) and direct site factor (DSF) were estimated, using HEMIVIEW canopy analysis software version 2.1 (Delta-T Devices, Cambridge, UK). After estimating the site factors, geostatistical analyses were done to interpolate the light distributions (Robertson 1987; Rossi et al. 1992). Here, we used a punctual kriging method (Robertson 1987), using GS software version 3.1.7 (Gamma Design Software, Plainwell, MI, USA) (Robertson 1998). Kriging provides a means of interpolating values for points using semivariograms that represent spatial relationships (Robertson 1987), and this interpolation method has been used in studies of forest ecology (e.g., Morris 1999; Nanos and Montero 2002). Punctual kriging is an exact interpolator, so that where interpolated points coincide with measured points, the estimated values are identical to measured values (Robertson 1987). In addition, semivariograms have been applied to the spatial analysis of light regimes in the forest understory (e.g., Nicotra et al. 1999). In a semivariogram, the semivariance (γ(h)) is the autocorrelation statistic, defined as N( h)
[
2
]
γ( h) 1 2 N ( h) Â Z ( xi ) Z ( xi h) i1
where γ(h) is the semivariance for interval distance class h, Z(xi) is the measured sample value at point xi, Z(xi h) is measured sample value at point xi h, and N(h) is the total number of sample couples for the lag interval h. We calculated the semivariance and then fit curves to the semivariograms using a spherical model (Isaacs and Srivastava 1989) defined as
[
γ(h) C 0 C 1.5(h A0 ) 0.5(h A0 )
3
γ(h) C 0 C
Estimation of light distribution To measure the spatial change in the understory light regimes within the two subplots, hemispherical fish-eye
]
for h A0 for h A0
where h is the lag distance interval, C0 is nugget variance 0, C is structural variance C0, and A0 is the range. The
Table 1. Relative density, relative basal area, and the sum of the basal areas of woody species in the plot at Mt. Ontake, central Japan Species
Abies mariesii Abies veitchii Picea jezoensis var. hondoensis Other woody species (5 spp.) Evergreen conifers Deciduous broad-leaved Total
Attainable height (m) 27.78 28.30 32.85
Relative density (%) H 2m
H 2m
Basal area (m2/ha)
Relative basal area (%) Total
H 2m
H 2m
Total
H 2m
H 2m
Total
61.3 21.1 3.9
48.6 39.9 8.7
52.4 34.2 7.3
21.4 11.0 36.3
74.9 19.9 4.0
26.0 11.7 33.6
8.2 4.2 13.8
2.7 0.7 0.1
10.9 4.9 14.0
3.5 10.2
1.2 1.7
1.9 4.2
14.1 17.2
0.8 0.4
12.9 15.8
5.4 6.6
0.1 0.1
5.4 6.6
100.0
100.0
100.0
100.0
100.0
100.0
38.1
3.8
41.7
Attainable height was estimated by using the expanded allometry equation of Thomas (1996). Basal area of overstory trees (H 2 m) was calculated from diameter at breast height (DBH), and that of understory saplings (H 2 m) was calculated from diameter at ground level (DGL). H indicates the height in 2000
165
sill is expressed as C0 C. The nugget, the y intercept of the variogram, indicates the variance that does not appear to be spatially dependent, and is either a random error or represents spatial dependence at scales smaller than the minimum distance examined. The range indicates the distance (m) along the x axis at which the semivariogram function stops increasing. The sill indicates the ordinate value at which the variogram becomes flat. The proportion of structural variance (C) to sill (C0 C) can be used as an indicator of magnitude of spatial dependence. (See Morris 1999; Hirobe et al. 2001) This value will approach 1.00 in a strongly structured system, whereas in a system that has little structure or is structured at scales larger or smaller than those measured, the proportion by structured variance will approach 0.00. In this study, we used an active lag set of 25 m and lag distance of 2 m. As a result, each analysis of the spherical model applied to the site factors showed a highly significant fit (all P 0.0001), and had strong spatial dependence; C/(C0 C) was from 0.69 to 1.00 (Table 2). After analysis of the semivariogram, punctual kriging was done to interpolate the distributions of site factors. To assess the light conditions of each sapling, we aggregated the light conditions that were determined by the kriging method to the saplings based on their coordinates, using ArcView GIS software, version 3.2 (Environmental Systems Research Institute, Redlands, CA, USA). By using this analysis, the ISF and DSF of each individual sapling were determined. In the above method, light conditions for every sapling are determined at 2 m above ground level. However, since light intensity within the forests generally decreases from forest canopy to understory, light intensity may change even within the range of the sapling height (e.g., 200 cm to 5 cm). Therefore, to evaluate the extent of vertical change in understory light intensity, we first examined the difference in light conditions between 1 m and 2 m above ground. Our preliminary results showed that the site factors at 2 m were significantly correlated with those at 1 m (r 2 0.994, P 0.0001 for ISF and r 2 0.924, P 0.0001 for DSF, respectively), and regression slope showed no departure form the value of 1 (P 0.05 for ISF and P 0.05 for DSF, respectively). In addition, the distributions of the two data sets (2 m and 1 m) were similar (Kolmogorov-Smirnov twosample test, P 0.05 for ISF and P 0.05 for DSF, respectively). Thus, we considered that there was little vertical change in understory light availability in this study forest. Light conditions of the understory saplings were therefore
uniformly estimated at 2 m above ground irrespective of the sapling height. Calculation of local crowding To calculate the local crowding of overstory trees, we used Weiner’s (1984) neighborhood statistic. The neighborhood interference (W) of Weiner (1984) is defined as N
(
W Â S i di a i1
)
where N is the total number of neighbors, Si is the size (basal area; m2) of the ith neighbor, di is the distance (m) from each target sapling to the ith neighbor, and a is the distance power 1 (proportional) or 2 (quadratic). Here, we wanted to evaluate the shading effects of larger trees on the understory saplings so that we defined the overstory trees (2 m height) as neighbors. Here, since Takahashi (1996) demonstrated that there was little species effect on the crowding index in boreal Abies–Picea forest in northern Japan, we did not consider the species difference of overstory trees. Previous studies have argued that effects of asymmetric competition for light relate to the distance from the neighbors (Weiner 1984; Peterson and Squiers 1995; Takahashi 1996), since individual plants only compete with their immediate neighbors and do not compete with every individual within the same population or community (Harper 1977; Pacala 1986; Goldberg 1987; Tilman 1997). Many studies therefore deal with the competitive effect by crowding indices, which involve the effects of distance and plant size (Weiner 1984; Peterson and Squiers 1995; Takahashi 1996). In the study of Weiner (1984), a formulation of local crowding that involved the proportional effect to the distance from the neighbors had a similar or better fit than a formulation using the square of distance. Peterson and Squiers (1995) also showed the asymmetry of competitive effects on growth by using a formulation involving proportional distance. However, Takahashi (1996) provided evidence of the effect of local crowding on the tree growth by a formulation which involved the quadratic effect of the distance from the neighbors. Thus, we tested the local crowding with both distance power 1 (proportional) and 2 (quadratic). For the limit of distance from the target saplings, Takahashi (1996) used 8 m in his analysis, because 6.3 m is the mean distance of crown spread for canopy of Abies sachalinensis and Picea glehnii. On the other hand,
Table 2. Semivariance analysis of spatial structure of the two site factors in the two subplots Subplot
A B
Site factor
ISF DSF ISF DSF
Nugget
Sill
Range (m)
Proportion
C0
C0 C
A0
C/(C0 C)
0.0000 0.0001 0.0009 0.0011
0.0243 0.0531 0.0029 0.0045
30.78 45.46 33.36 23.65
1.000 0.998 0.690 0.756
A spherical model was applied to the analyses. RSS represents the sum of squares of residuals
r2
RSS
0.979 0.990 0.900 0.981
9.698 1.284 1.766 1.013
P value
106 105 107 106
0.0001 0.0001 0.0001 0.0001
166
Peterson and Squiers (1995) showed that asymmetric competition for light occurred about 2–4 m distances in white pine and aspen. Thus, we examined local crowding using several neighborhood radii (di); 2, 4, 6, 8, and 10 m. W was calculated for saplings within the center quadrat in each subplot to allow 10-m margins, which should eliminate edge effects. Because of the large number of individuals, this analysis was calculated with a program using the GNU C Compiler (Free Software Foundation, Boston, MA, USA).
Then, the observed coefficient of determination was compared with the null distributions to determine significance. Here, null hypothesis is no relationship between W and the response variable. In this study, except for analyses of geostatistics, calculation of local crowding, and randomization test, statistical analysis was performed with SPSS software version 10.0.5 (SPSS, Chicago, IL, USA).
Statistical analysis Relative growth rate is superior to absolute growth rate, since absolute growth rate may be positively correlated with the size at certain times during the life of a tree (Peterson and Squiers 1995). Therefore, height growth of saplings was evaluated by relative height growth rate (RHGR). RHGR was defined as RHGR (ln Ht 2 ln Ht 1 ) (t2 t1 ) where H indicates the sapling height for the year t (Williams et al. 1999). In this study, t1 and t2 are 1995 and 2000, respectively. To evaluate the best fit of both distance power and neighborhood radii in the analysis of Weiner’s (1984) local crowding index, regressions of ISF or DSF as dependent variables and W as an independent variable were conducted with each combination of the distance power (a) and neighborhood radii (di). These regressions were analyzed based on the coordinates of saplings (570 individuals in total for the three conifers). Again, to evaluate the effects of W on the sapling growth, regressions of RHGR as a dependent variable and W as an independent variable were conducted with each combination of a and di. These regressions were analyzed for the saplings of each conifer species. Before the regressions, dependent variables were log-transformed to improve the homogeneity. However, the significance of these regressions cannot be determined in the typical way, since the predictor variables (W) violate the assumption of independence (Mitchell-Olds 1987; Thomas SC and Weiner 1989; Peterson and Squiers 1995; Umeki 1995). In this study, null distributions of the coefficient of determination were generated by randomly assigning observed W values to the response variable (ISF, DSF, or RHGR) using 5000 permutations for each randomization test based on the observed sapling coordinates.
Results and discussion Regressions of the site factors against the local crowding of overstory trees (W) with each combination of the distance power and neighborhood radii are shown in Table 3. The results indicated that a distance power of 2 could not explain the shading effects of overstory trees on the understory saplings and that one of 1 gave a better fit. According to the coefficient of determination (r2), neighborhood radii of W with the distance power of 1 showed the best fit for 10-m distances, compared with the smaller distance classes. Canopy height is a major determinant of the effects of overstory trees on the understory light regimes (Canham et al. 1990; Clark et al. 1996). In this study forest, the attainable height of canopy trees was greater than 30 m (Table 1), so that overstory trees further away can greatly affect the light condition of understory saplings. However, although W correlated with site factors, these explained only 21.9%– 24.7% of the total variance in the cases where W gave the best fit (Table 3). This indicates that the greater proportion of the variance remained unexplained by W. Since the size of neighbors relates to competitive performance (Goldberg 1987; Gaudet and Keddy 1988), a competition index involving the tree size as a parameter can explain the shading effect to some extent. However, in the forests, canopy structure also greatly affects the understory light regimes (Clark et al. 1996). In addition, canopy structure is determined by the species-specific crown shape, and crown shape is affected by other factors such as age structure and density of stands (Kikuzawa and Umeki 1996; Song et al. 1997). Thus, the local crowding model of this study may include such possible errors, although it adequately represents the shading effect of overstory trees. Furthermore, individual trees even beyond 10-m distances may also affect the understory light regimes.
Table 3. The values from r2 of regressions between the two site factors, ISF and DSF and local crowding (W) of each combination of neighborhood radius and distance power Distance power
1 2
Site factor
ISF DSF ISF DSF
Neighborhood radius 10 m
8m
6m
4m
2m
0.219*** 0.247*** 0.003 0.007
0.132*** 0.166*** 0.002 0.006
0.051*** 0.080*** 0.002 0.006
0.027*** 0.055*** 0.002 0.005
0.015** 0.033*** 0.001 0.005
n 570 ** P 0.01, *** P 0.001. Significance levels are based on randomization tests
167
RHGR of saplings was negatively correlated with W (Fig. 1). This inhibited height growth can be regarded as a response to the shading by overstory trees, since saplings of Abies and Picea generally show reduced height growth under the shaded conditions in order to form the flatshaped crowns which are acclimated to shading (Kohyama 1980; King 1997; Messier et al. 1999). Here, among several W values, W using the distance power of 1 and 10-m limit distance again showed the largest r2 values in the correlation with RHGR, irrespective of species (Table 4); that is, local
Fig. 1. Correlation of relative height growth rate (RHGR) of saplings with local crowding of overstory trees (W), using a 10-m limit distance and proportional distance effect. Regression lines are shown. *** P 0.001
0
A. mariesii r2 = 0.123***
–1
–2
–3
log RHGR
0
A. veitchii r2 = 0.200***
–1
–2
–3 0
P. jezoensis var. hondoensis r2 = 0.163***
–1
–2
–3
0
0.2
0.4
0.6
0.8
1
W
crowding which represents the shading effects most adequately could also explain such sapling growth inhibition by the overstory trees most adequately. This provides strong evidence that the competitive effect of overstory trees on the sapling growth is mediated by the shading effect, indicating that competition for light clearly exists within this study forest. If W using the distance power of 2 showed a better fit with sapling growth, nearer neighbors would have larger competitive effects on the sapling growth. However, W using the distance power of 2 hardly explained any of the sapling growth variation, and W using the power of 1 showed a significant fit with sapling growth (Table 4), indicating that the competitive effect of overstory trees on the sapling growth is inversely proportional to distance. Weiner (1984) suggested that neighbor effects decrease with distance when competition occurs for light, whereas the effect of neighbors decreases with the square of its distance when competition is for nutrients and water. This also suggests the importance of light resource in the competitive interactions within this subalpine forest. In this study, overstory trees affect the understory light availability and thus regulate the growth of understory saplings. However, in addition to the tree responses to actual light availability (e.g., King 1997; Poorter and Werger 1999), individual trees grow according to the surrounding conditions such as the R : FR ratio, available space for crown expansion, and neighbor crown display (e.g., Tremmel and Bazzaz 1993; Ballaré 1994; Umeki and Kikuzawa 2000). Such local variability might yield additional errors, resulting in a decrease in r2 values in the correlation with sapling growth greater than that with site factors (Tables 3 and 4). This suggests that, except for shading from the overstory, other factors such as among-sapling competition for available space (Mori and Takeda 2003) also relates to the sapling growth, although light is a major resource which individual saplings compete for. Importance of gap formation for tree regeneration has been broadly recognized in subalpine forests in central Japan (Yamamoto 1993, 1995, 2000; Narukawa and Yamamoto 2001). Gap formation excludes the competitive effects from overstory trees, then accelerates the sapling growth, and thus promotes tree regeneration. Consequently, the competitive effects of overstory trees on the
Table 4. The values from r2 of regressions between the relative height growth rate (RHGR) and the local crowding (W) of each combination of neighborhood radius and distance power Distance power
1 2
Species
Abies mariesii Abies veitchii Picea jezoensis var. hondoensis Abies mariesii Abies veitchii Picea jezoensis var. hondoensis
Neighborhood radius 10 m
8m
6m
4m
2m
0.123*** 0.200*** 0.163*** 0.003 0.007 0.006
0.096*** 0.122*** 0.086 0.003 0.005 0.003
0.065*** 0.075*** 0.007 0.003 0.004 0.001
0.048*** 0.041** 0.000 0.003 0.003 0.000
0.027** 0.012 0.007 0.002 0.001 0.000
n 290 (Abies mariesii), 238 (Abies veitchii), 42 (Picea jezoensis var. hondoensis) ** P 0.01, *** P 0.001. Significance levels are based on randomization tests
168
understory saplings, as demonstrated in this study, may be closely associated with this property of gap regeneration as a maintaining mechanism of subalpine forests in central Japan. Acknowledgments We thank T Ando, K Kurumado, N Miyamoto, S Hasegawa, M Hirobe, S Nanami, and N Ohte for their support in the field research, statistical assistance, and valuable advice. The programming to calculate the local crowding index was done by S Hasegawa, and the randomization test was programmed by K Umeki. We also thank H Barclay for reading through the draft manuscript and members of the Laboratory of Forest Ecology, Graduate School of Agriculture, Kyoto University for their valuable advice and helpful discussions. We are also grateful to K Seiwa, K Umeki, and the two anonymous reviewers for many helpful suggestions on this manuscript.
Literature cited Ballaré CL (1994) Light gaps: sensing the light opportunities in highly dynamic canopy environments. In: Caldwell MM, Pearcy RW (eds) Exploitation of environmental heterogeneity by plants: ecological processes above- and below-ground. Academic Press, San Diego, CA, pp 73–110 Canham CD, Denslow JS, Platt WJ, Runkle JR, Spies TA, White PS (1990) Light regimes beneath canopies and tree-fall gaps in temperate and tropical forests. Can J For Res 20:620–631 Clark DB, Clark DA, Rich PM, Weiss S, Oberbauer SF (1996) Landscape-scale evaluation of understorey light and canopy structure: methods and application in a neotropical lowland rain forest. Can J For Res 26:747–757 Gaudet CL, Keddy PA (1988) A comparative approach to predicting competitive ability from plant traits. Nature 334:242–243 Gilbert IR, Jarvis PG, Smith H (2001) Proximity signal and shade avoidance differences between early and late successional trees. Nature 411:792–795 Goldberg DE (1987) Neighborhood competition in an old field plant community. Ecology 68:1211–1223 Harper JL (1977) Population biology of plants. Academic Press, London Hirobe M, Ohte N, Karasawa N, Zhang G, Wang L, Yoshikawa K (2001) Plant species effect on the spatial patterns of soil properties in the Mu-us desert ecosystems, Inner Mongolia, China. Plant Soil 234:195–205 Isaacs E, Srivastava R (1989) Applied geostatistics. Oxford University Press, Oxford Kikuzawa K, Umeki K (1996) Effects of canopy structure on degree of asymmetry of competition in two forest stands in northern Japan. Ann Bot 77:565–571 King DA (1997) Branch growth and biomass allocation in Abies amabilis saplings in contrasting light environments. Tree Physiol 17:251–258 Kohyama T (1980) Growth pattern of Abies mariesii saplings under conditions of open-growth and suppression. Bot Mag 93:13–24 Kubota Y, Hara T (1996) Allometry and competition between saplings of Picea jezoensis and Abies sachalinensis in a sub-boreal coniferous forest, northern Japan. Ann Bot 77:529–537 Messier C, Doucet R, Ruel J, Claveau Y, Kelly C, Lechowicz MJ (1999) Functional ecology of advance regeneration in relation to light in boreal forests. Can J For Res 29:812–823 Mitchell-Olds T (1987) Analysis of local variation in plant size. Ecology 68:82–87 Mithen R, Harper JL, Weiner J (1984) Growth and mortality of individual plants as a function of “available area”. Oecologia 62:57– 60 Mori A, Takeda H (2003) Architecture and neighborhood competition of understory saplings in a subalpine forest in central Japan. Ecoscience 10:217–224 Morris SJ (1999) Spatial distribution of fungal and bacterial biomass in southern Ohio hardwood forest soils: fine scale variability and microscale patterns. Soil Biol Biochem 31:1375–1386
Nanos N, Montero G (2002) Spatial prediction of diameter distribution models. For Ecol Manage 161:147–158 Narukawa Y, Yamamoto S (2001) Gap formation, microsite variation and the conifer seedling occurrence in a subalpine old-growth forest, central Japan. Ecol Res 16:617–625 Nicotra AB, Chazdon RL, Iriarte SVB (1999) Spatial heterogeneity of light and woody seedling regeneration in tropical wet forests. Ecology 80:1908–1926 Pacala SW (1986) Neighborhood models of plant population dynamics. 2. Multi-species models of annuals. Theor Popul Biol 29:262– 292 Pacala SW, Silandar JAJ (1985) Neighborhood models of plant population dynamics. 1. Single-species models of annuals. Am Nat 125:385–411 Peterson CJ, Squiers ER (1995) Competition and succession in an aspen-white-pine forest. J Ecol 83:449–457 Poorter L, Werger MJA (1999) Light environment, sapling architecture, and leaf display in six rain forest tree species. Am J Bot 86:1464–1473 Robertson GP (1987) Geostatistics in ecology: interpolating with known variance. Ecology 68:744–748 Robertson GP (1998) Geostatistics for the environmental sciences. Gamma Design Software, Plainwell, MI Rossi RE, Mulla DJ, Journel AG, Franz EH (1992) Geostatistical tools for modeling and interpreting ecological spatial dependence. Ecol Monogr 62:277–314 Song B, Chen J, Desanker PV, Reed DD, Bradshaw GA, Franklin JF (1997) Modeling canopy structure and heterogeneity across scales: from crowns to canopy. For Ecol Manage 96:217–229 Takahashi K (1996) Plastic response of crown architecture to crowding in understory trees of two co-dominating conifers. Ann Bot 77:159– 164 Thomas EM, Weiner J (1989) Local density variation may mimic effects of asymmetric competition on plant size variability. Ecology 70:1188–1191 Thomas SC (1996) Asymptotic height as a predictor of growth and allometric characteristics in Malaysian rain forest trees. Am J Bot 83:556–566 Thomas SC, Weiner J (1989) Including competitive asymmetry in measures of local interference in plant populations. Oecologia 80:349– 355 Tilman D (1997) Mechanism of plant competition. In: Crawley MJ (ed) Plant ecology. Blackwell Science, Oxford, pp 239–261 Tremmel DC, Bazzaz FA (1993) How neighbor canopy architecture affects target plant performance. Ecology 74:2114–2124 Umeki K (1995) Importance of crown position and morphological plasticity in competitive interaction in a population of Xanthium canadense. Ann Bot 75:259–265 Umeki K, Kikuzawa K (2000) Patterns in individual growth, branch population dynamics, and growth and mortality of first-order branches of Betula platyphylla in northern Japan. Ann For Sci 57:587–598 Wada N, Ribbens E (1997) Japanese maple (Acer palmatum var. Matsumurae Aceraceae) recruitment patterns: seeds, seedlings, and saplings in relation to conspecific adult neighbors. Am J Bot 84:1294–1300 Weiner J (1984) Neighborhood interference amongst Pinus rigida individuals. J Ecol 72:183–195 Weiner J, Thomas SC (1986) Size variability and competition in plant monocultures. Oikos 47:211–222 Weiner J, Thomas SC (1992) Competition and allometry in three species of annual plants. Ecology 73:648–656 Weiner J, Berntson GM, Thomas SC (1990) Competition and growth form in a woodland annual. J Ecol 78:459–469 Williams H, Messier C, Kneeshaw DD (1999) Effects of light availability and sapling size on the growth and crown morphology of understory Douglas-fir and lodgepole pine. Can J For Res 29:222– 231 Yamamoto S (1993) Gap characteristics and gap regeneration in a subalpine coniferous forest on Mt. Ontake, central Houshu, Japan. Ecol Res 8:277–285 Yamamoto S (1995) Gap characteristics and gap regeneration in subalpine old-growth coniferous forests, central Japan. Ecol Res 10:31–39 Yamamoto S (2000) Forest gap dynamics and tree regeneration. J For Res 5:223–229
