This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy

Forest Ecology and Management 259 (2010) 547–554

Contents lists available at ScienceDirect

Forest Ecology and Management

journal homepage: www.elsevier.com/locate/foreco

Age structure and regeneration of subalpine fir (Abies fargesii) forests across an altitudinal range in the Mountains, China

Haishan Dang, Yanjun Zhang, Kerong Zhang, Mingxi Jiang, Quanfa Zhang *

Key Laboratory of Aquatic Botany and Watershed Ecology, Botanical Garden, the Chinese Academy of Sciences, Wuhan 430074, PR China

ARTICLE INFO ABSTRACT

Article history: Age structure and regeneration dynamics of subalpine fir (Abies fargesii) forest were studied across the Received 31 July 2009 altitudinal range in both the north and south aspects of the Qinling Mountains, China. Ages of individual Received in revised form 7 November 2009 fir trees were determined based on the number of rings counted from cores and the number of years to Accepted 9 November 2009 reach coring height estimated using age–height regression. Fir age structure and regeneration dynamics were similar in both the north and south aspects. A unimodal population age structure was found at the Keywords: low- and mid-elevations in both aspects, indicating that environmental factors might play an important Age structure role in shaping A. fargesii age structure and regeneration at those sites. There was a recruitment pulse Regeneration dynamics during the time period 1830–1890 at each altitudinal site, but no stem recruitment occurred at the low- Abies fargesii The Qinling Mountains and mid-elevations in the last century, which might be attributed to the intensive cover of understory bamboo. Fir trees were, however, persistently recruited at the upper limits during the last 150 years, and the fir tree density at the upper limits was significantly higher than that at the lower limits in both aspects. The fir population at the upper limits showed a significant increase in recruitment and stem density relative to the fir population at the low- and mid-elevations in the last century. We propose that the differences in recruitment might promote variations in stand structure and regeneration dynamics of the subalpine fir forests along the altitudinal gradient in the Qinling Mountains, China. ß 2009 Published by Elsevier B.V.

1. Introduction et al., 2004; Wangda and Ohsawa, 2006; Penuelas et al., 2007; Fyllas et al., 2008). Age structure and regeneration dynamics can be Forest dynamics, such as population age structure and used to infer population response to known environmental events recruitment pattern, are influenced by many factors including and reconstruct forest development history (Brubaker, 1986). disturbance, competitive interactions between trees (Taylor et al., Investigations of age structure and regeneration dynamics could 1996; North et al., 2004). Disturbances drive the regeneration give insights into the processes that determine population dynamics of most closed-canopy forests by creating opportunities structure and pattern over time (Veblen et al., 1980; Svensson for the establishment of new individuals through canopy opening and Jeglum, 2001). Quantitative reconstructions of population age and have been shown to play an important role in age structure and structure (i.e., the distribution and range of tree ages) could serve regeneration dynamics of many subalpine forests (Veblen et al., as a reference central to restoration and management of forest 1981; Taylor et al., 1996; Cullen et al., 2001; Parish and Antos, ecosystems (Covington et al., 1997; Fule´ et al., 1997; Mast et al., 2004). Community response to disturbance varies widely and 1999; Wang et al., 2004). Assessing and analyzing age structure depends on the types, size, severity, and frequency of disturbance and regeneration dynamics are therefore essential to understand and species’ life history attributes. Studies on stand dynamics can the long-term ecological processes of natural forests. provide a substantial information on the regeneration and Within the geographical range of a species, environmental population structure of forests (Nakashizuka, 1991; Taylor et al., conditions vary along physiographic and ecological gradients, 1996; Antos and Parish, 2002). creating the potential for spatial heterogeneity in growth-limiting Studies on regeneration dynamics in many forests have factors and in tree establishment and regeneration (Block and typically dealt with the age and size structures and seedling Treter, 2001; Peterson et al., 2002; Carrer et al., 2007; Penuelas density (Taylor and Qin, 1988; Taylor et al., 1996; Antos and Parish, et al., 2007). Some environmental factors often become limiting 2002; Goldblum and Rigg, 2002; Parish and Antos, 2004; Wang factors resulting in poor establishment of plant individuals and insufficient recruitment of the local population. Across a species distributional range, however, optimal environmental conditions * Corresponding author. Tel.: +86 27 87510702; fax: +86 27 87510251. could enable a well-developed population (Taylor and Qin, 1988; E-mail address: [email protected] (Q. Zhang). Block and Treter, 2001; Wang et al., 2004; Holz et al., 2006). As a

0378-1127/$ – see front matter ß 2009 Published by Elsevier B.V. doi:10.1016/j.foreco.2009.11.011 Author's personal copy

548 H. Dang et al. / Forest Ecology and Management 259 (2010) 547–554 result, age structure and regeneration dynamics of plant popula- along an altitudinal gradient; (2) to examine the differences of its tion, especially those with long-lived species, could act as an age structure and regeneration pattern across the altitudinal range indictor of environmental changes across the distributional range. in the north and south aspects; and (3) to evaluate the feasibility of In the subalpine region, environmental conditions differ greatly using age structure and regeneration pattern across an altitudinal with altitude, thus, the limiting factors related to seedling gradient to identify forest dynamics related to environmental establishment and tree recruitment might vary with changes in changes. altitude (Block and Treter, 2001; Wangda and Ohsawa, 2006; Penuelas et al., 2007). The understanding of environmental 2. Methods influences on population structure and regeneration dynamics of the natural forests could be improved by studies that sample 2.1. Study area across the distributional range from extreme to moderate environmental conditions (Veblen, 1989; Wang et al., 2004; This study was conducted in the Foping and Zhouzhi National Wangda and Ohsawa, 2006; Penuelas et al., 2007; Lingua et al., Nature Reserves (338330–338570N; 1078390–1088190E), located 2008). respectively in the north and south aspects of the Qinling Mountains The forest of subalpine fir, Abies fargesii, is one of the most of Province, China (Fig. 1). The Qinling Mountains run east– important vegetations in the biodiversity hotspot Qinba Moun- west and form an important watershed divider between China’s two tains (including the Qinling Mountains and the Daba Mountains), great rivers, the and the . The Qinling China. The subalpine fir forests occur over a wide elevation range Mountains are situated in the transitional zone between two above 2300 m in the Qinling Mountains, and have been undis- macroclimatic regimes (i.e., subtropical and warm-temperate turbed by human activities for more than a century due to zones), making it a biologically rich area and sensitive to climatic relatively difficult accessibility (Zhang, 1989). However, very little change in China (Chen, 1983; Yan, 2006; Dang et al., 2007). Elevation is known about the age structure and regeneration dynamics of the in the study area ranges from 980 to 2900 m a.s.l. The south-facing subalpine fir forests across the altitudinal range. An understanding slope is of subtropical characteristics with wet summers and warm of age structure and regeneration patterns over the past 200–300 winters, while the north-facing slope belongs to warm-temperate years would provide a context to interpret present patterns of tree zone with relatively dry summers and cold winters. Annual establishment, and this type of ecological knowledge is funda- precipitation ranges from 950 to 1200 mm, most of which falls mental for conservation and sustainable utilization of the between July and September. Snow cover usually lasts 5 or more mountain ecosystems (Cullen et al., 2001; Antos and Parish, months (from November to March), and annual mean temperature 2002; Goldblum and Rigg, 2002; Wangda and Ohsawa, 2006). ranges from 6 to 11 8C below 2000 m and from 1 to 6 8C above In this study, age structure and regeneration pattern of the 2000 m a.s.l. (Chen, 1983). subalpine fir forest were examined by ageing trees and dating The natural vegetation types are deciduous broad-leaved periods of past tree establishment across an altitudinal range, forests (below 1800 m), mixed conifer and deciduous forests including upper limits, interior forest areas, and lower limits of the (1800–2300 m), conifer forests dominated by A. fargesii (2300– subalpine fir forest, in both the north and south aspects of the 2800 m) and subalpine meadow (above 2800 m) along the Qinling Mountains, China. The objectives of this study are (1) to altitudinal gradient in the study area. Umbrella bamboo (Fargesia reconstruct the age and regeneration structure of the subalpine fir qinlingensis) is a common understory species above 2300 m. The

Fig. 1. Locations of the study sites (~) in the Qinling Mountains of Shaanxi Province, China. NL, lower distribution limit in the north aspect; NM, middle distribution zone in the north aspect; NU, upper distribution limit in the north aspect; SL, lower distribution limit in the south aspect; SM, middle distribution zone in the south aspect; SU, upper distribution limit in the south aspect. Author's personal copy

H. Dang et al. / Forest Ecology and Management 259 (2010) 547–554 549

Table 1 Species composition of the subalpine fir (Abies fargesii) forests in the Qinling Mountains, China.

Species Density (stems/ha) Dominance (m2/ha) Relative frequency Relative density Relative dominance Importance value

A. fargesii 317.5 40.85 22.73 61.06 59.80 47.86 Betula albo-sinensis 115.0 12.75 18.18 22.12 18.67 19.66 Acer maximowiczii 22.5 6.44 13.64 4.33 9.43 9.13 Maddenia hypoxantha 17.5 5.76 11.36 3.37 8.44 7.72 Prunus tomentosa 20.0 2.14 11.36 3.85 3.14 6.12 Crataegus kansuensis 12.5 0.27 9.09 2.40 0.40 3.96 Abelia macroptera 7.5 0.04 6.82 1.44 0.06 2.77 Sorbus koehneana 5.0 0.03 4.55 0.96 0.04 1.85 Corylus tibetica 2.5 0.01 2.27 0.48 0.02 0.92

Totals 520.0 68.31 100 100 100 100

Table 2 Main characteristics of A. fargesii population at the different altitudinal sites in the north and south aspects of the Qinling Mountains, China.

Altitudinal site Elevation (m) DBH (cm) Basal area (m2/ha) Density (stems/ha) Canopy coverage (%)

North aspect Lower limit 2300–2400 41.0 2.1 (66.9)a 28.9 7.3a 265 45.4a 43.4 8.8a Middle elevation 2400–2700 35.3 1.7 (51.6)b 38.8 6.9b 375 72.8ab 53.8 9.9b Upper limit 2700–2800 21.3 3.9 (35.1)c 21.1 6.8a 470 218.9b 48.8 9.7ab

South aspect Lower limit 2300–2400 40.8 12.0 (63.7)a 29.9 12.4a 225 58.6a 49.6 6.7a Middle elevation 2400–2650 33.7 3.6 (52.9)a 44.9 8.4b 418 94.4ab 59.8 4.4a Upper limit 2650–2750 20.4 5.2 (39.5)b 23.1 13.2a 535 329.6b 53.6 4.3a

Note: Data are presented in mean S.E. The values of maximum DHB are given in parentheses. The different letters indicate the significance at P 0.05 level.

research focuses on the conifer forests dominated by A. fargesii, i.e., for a very long period (Antos et al., 2005). As a result, fir seedlings above the elevation of 2300 m. Besides A. fargesii, the subalpine fir and saplings with severe release or suppression events (Nowacki forests are comprised of other eight species (Table 1), each of which and Abrams, 1997) were excluded from developing age–height only has a low relative abundance (Dang et al., 2009). Historically, regression. the subalpine fir forests in the Qinling Mountains were selectively cut several times during the period 1790–1870 (Zhang, 1989; Dang 2.3. Data analyses et al., 2009). In the past century, however, human activities are rare in the subalpine forest areas of the Qinling Mountains due to the The increment cores were dried, mounted and sanded using relatively difficult accessibility, and some of the only remaining progressively finer sandpaper until growth ring boundaries were intact forests occur in the region (Liu et al., 2002). Thus, the study clearly visible. The ages of cores were read using the Windendro only examines regeneration dynamics of A. fargesii based on its image-analysis system (Regent instruments Inc., Quebec, Canada), population structure and seedling/sapling density across its and then crossdated by the program COFECHA (Holmes, 1983). For altitudinal range in the Qinling Mountains, China. the increment cores missing the pith, the number of rings required to reach the pith was estimated geometrically (Duncan, 1989; 2.2. Field sampling Szeicz and Macdonald, 1995). The correlation between age (y, year) and height (x, cm) of the In the summer of 2005, we measured age structure and fir seedlings is statistically significant (y = 0.2374x + 3.274, regeneration dynamics in the deciduous to conifer forests R2 = 0.9335, P < 0.0001, n = 38) (Fig. 2), and it takes the fir trees transitional zone at the lower elevation, the interior forests about 36 years to reach breast height. Due to the absence of fir dominated by A. fargesii at the middle elevation, and the treeline seedlings and saplings at the mid- and low-elevations, the age– environment at the upper elevation in both the north and south height regression developed at the upper limits was used to adjust aspects of the Qinling Mountains by sampling five plots fir ages for all the sites. Ages of individual fir trees were estimated (20 m 20 m) at each elevation zone (Table 2). The plots were by adding the number of 36 rings on each core, the estimated selected based on the criteria that there were similar habitats and that the stands should represent the fir forest structure at each altitudinal site. Within each plot, the DBH (diameter at breast height, 1.37 m above ground level) of each fir tree was measured, and fir trees taller than 2 m were cored at breast height. One sound core per tree was extracted in the direction parallel to the slope contour using increment borers. In total, 412 fir trees were cored. Within each plot, five 5 m2 subplots were established to determine seedling (<0.5 m in height) and sapling (0.5–2 m in height) density. Thirty-eight saplings and seedlings with normal growth form were destructively sampled to create an age–height regression to estimate the time to reach breast height. Shade- tolerant species can vary greatly in growth rate, and some species Fig. 2. Correlations between age and height of fir seedling/sapling (<2 m in height) of Abies can form seedling banks in which individuals can persist in the Qinling Mountains, China. Author's personal copy

550 H. Dang et al. / Forest Ecology and Management 259 (2010) 547–554 number of years for a fir to reach coring height. The age distribution had only small fractions of the total individuals. At the upper limits, was presented at 10-year interval to eliminate possible errors in small fir trees with DBH <15 cm accounted, respectively, for over the age determination process, and all the trees were also grouped 22% and 21% in the north and south aspect, while no fir trees were into size classes at 5 cm intervals according to DBH in each site. The represented in the large diameter classes more than 35 and 40 cm cross-sectional area at breast height (, cm2) was calculated using in the north and south aspect, respectively. In addition, the size the DBH data. ANOVAs and post-hoc Tukey HSD tests were applied distribution pattern indicated that large fir trees with DBH greater to test differences in basal area, tree density and age structure of fir than 40 cm gradually decreased, and small fir trees with DBH less populations (not including the seedlings and saplings) at the three than 40 cm increased with increasing altitude in both aspects different altitudinal sites in both the north and south aspects. (Fig. 3).

3. Results 3.3. Age structure

3.1. Subalpine fir forest structure At the lower limits of both aspects, almost all the fir trees were over 120 years old (Fig. 4), and the oldest individual was more than Basal area, tree density, canopy coverage and DBH varied 300 and 230 years in the north and south aspect, respectively. At among the three different sites in both the north and south aspects the mid-elevations, the majority of the fir trees were between 110 (Table 2). The fir population at the mid-elevations in both aspects and 170 years old. Those individuals that exceeded 200 years old had the largest basal area and canopy coverage. A. fargesii density were scarce, and only a few fir trees were younger than 100 years increased with the increase in elevation, and the fir population at old. At the upper limits, however, most of the fir trees were the upper limits had significantly higher density than that at the between 60 and 150 years old, and 13.8% and 13.1% of fir trees were lower limits in both aspects (ANOVA, P 0.05). DBH declined with younger than 60 years in the north and south aspect, respectively increasing elevation, and DHB was significantly greater at lower (Fig. 4). limits than at upper limits in both aspects (ANOVA, P 0.05) Age structure analysis indicated that fir trees’ mean age (Table 2). decreased with the increase in elevation in both aspects (Table 3), and the fir trees at the upper limits were significantly 3.2. Size structure younger than those at the lower elevations (ANOVA, P 0.05). Age structure distribution was similar to that for the size distribution of The fir population showed a bell-shaped size distribution the fir population, displaying a decrease in the number of mature pattern (5-cm DBH classes) at each elevation (i.e., lower limit, fir trees and an increase in the number of young fir trees along the middle elevation, and the treeline environment) in both aspects altitudinal gradient in both aspects of the Qinling Mountains (Fig. 3). At the lower limits, however, large fir trees with DBH (Figs. 3 and 4). >40 cm accounted respectively for 71.1% and 73.3% of the total fir trees in the north and south aspect, while small fir trees were rare 3.4. Regeneration dynamics and no fir trees with DBH <15 cm were found. At the middle elevations, A. fargesii mainly occurred in the middle diameter The fir populations at the low- and mid-elevations demon- classes from 25 to 45 cm. The smaller and the larger DBH classes strated a unimodal regeneration type, while there was a sustained

Fig. 3. Size class distribution of the fir trees at different elevations in the north aspect (a) and south aspect (b) of the Qinling Mountains, China. Author's personal copy

H. Dang et al. / Forest Ecology and Management 259 (2010) 547–554 551

Fig. 4. Age structure by 10-year interval for the fir trees at different elevations in the north (a) and south aspect (b) of the Qinling Mountains, China.

Table 3 Age statistics of A. fargesii population at the different altitudinal sites in the north and south aspects of the Qinling Mountains, China.

Altitudinal site Elevation (m) Mean age (year) S.D. Range (year) Seedling and sapling (no./ha)

North aspect Lower limit 2300–2400 162.11a 34.28 121–302 0a Middle elevation 2400–2700 149.11a 24.79 108–182 0a Upper limit 2700–2800 127.88b 38.48 43–203 1200b

South aspect Lower limit 2300–2400 165.05a 32.97 105–239 0a Middle elevation 2400–2650 134.88b 26.65 66–214 0a Upper limit 2650–2750 108.70c 37.55 32–222 900b

Note: The different letters indicate the significance at P 0.05 level. regeneration of fir tree during the last 150 years at the upper limits 4. Discussion in both aspects (Fig. 4). In the north aspect, there was a broad period of stem recruitment with two weak peaks during 1900– Population dynamics of plant species, especially those with 1950 and 1840–1880 at the upper limit. At the low- and mid- long lifespan, could be considered as an indictor of vegetation elevations, the majority of the fir trees were recruited from 1830 to dynamics as well as environmental changes across their distribu- 1890, and none during the last century (Fig. 4(a)). In the south tional ranges (Brubaker, 1986; Wang et al., 2004; Camarero et al., aspect, the highest establishment occurred during the period 2005). The long lifespan of trees makes it infeasible to witness their 1860–1900, followed by lower and decreasing establishment in the whole life history, so a static investigation on age structure and last 100 years at the upper limit. There had been consistent regeneration of population is often accepted in population regenerations of fir trees during the 19th century at the middle dynamics estimates (Brubaker, 1986; Svensson and Jeglum, elevation and the lower limit, but no regeneration of fir trees 2001; Wangda and Ohsawa, 2006). However, care must be taken occurred at the lower limit during the 20th century (Fig. 4(b)). The when explaining static age structure data because of difference in regeneration reconstruction in both aspects, however, did not species mortality with various age and stand-history events. include the past four decades because the breast height cores were Interpretation of tree population dynamics requires informa- taken. In addition, fir seedlings and saplings with a density of 1200 tion on the temporal patterns of tree establishment, often inferred and 900 stems/ha for the north and south aspect were recorded at from age and size distributions (Cullen et al., 2001). Age and size the treeline environments, but none at the low- and mid- studies along an altitudinal gradient are helpful for understanding elevations in both aspects (Table 3). the relationship between environmental factors and population Author's personal copy

552 H. Dang et al. / Forest Ecology and Management 259 (2010) 547–554 dynamics. An upsurge in tree recruitment and establishment cutting from the 1790s to 1870s) in its development history (Dang coupled with increase in tree density is often an indication of et al., 2009). regeneration response to changes in environmental and/or The absence of fir seedlings and saplings and low recruitment of climatic conditions (Harcombe, 1987; Motta and Nola, 2001; fir tree in the last century at the lower limits and mid-elevations in Penuelas et al., 2007). In this survey of population structure, the both aspects (Fig. 4) might be attributed to the influences of distribution of individuals over the range of tree sizes might understory bamboo (F. qinlingensis) on recruitment success (Taylor represent the change in the rate of fir tree regeneration over time and Qin, 1988; Nakashizuka, 1991; Wang et al., 2006; Dang et al., across the altitudinal range in the Qinling Mountains. 2007). Bamboos, which are common understory plants in the The Qinling Mountain range constitutes a critical boundary for temperate and tropical subalpine forests, appear particularly climate and vegetation distribution in eastern central mainland effective in reducing seedling recruitment and tree regeneration China owing to its importance as a geographic demarcation line. when they achieve a high degree of dominance (Taylor and Qin, Although climate conditions between the southern and northern 1988; Taylor et al., 1996; Narukawa and Yamamoto, 2002; Holz slopes are quite different (Yan, 2006), the dynamics of the fir et al., 2006). At the lower limits and mid-elevations in the study regeneration showed no significant differences between the region, the understory bamboo with a height of 1.5–3.0 m and a north and south aspects (Fig. 4). Climatic conditions show large density of 70–140 stems/m2, usually covers more than 85%, spatial variations in mountainous areas, while the microclima- sometimes 100% of the understory of the subalpine fir forest (Ren, tology of montane landscapes is dependent on latitude, con- 1998; Wang et al., 2006). The lack of fir recruitment at the lower tinentality and topography (Barry, 1992). In this study, the limits and mid-elevations is likely to be linked to intensive cover of sampling location in the north aspect is just adjacent to that in the bamboo (F. qinlingensis) which limited germination and seedling south aspect within a small geographical region (Fig. 1), which survival and development in the last century in the Qinling might be responsible for the similar fir regeneration dynamics in Mountains (Taylor and Qin, 1988; Dang et al., 2007), which is both aspects. similar to other subalpine forests in China, Japan and South Low temperature at the upper treeline ecotones could inhibit America (Taylor and Qin, 1988; Nakashizuka, 1991; Narukawa and tree growth and development, while low precipitation and low soil Yamamoto, 2002; Holz et al., 2006). Like most other bamboos in moisture at the low-elevations might also restrict tree growth and tropical and temperate regions, F. qinlingensis is also a perennial development (Whittaker, 1956; Block and Treter, 2001). At the monocarpic species which is known for its long period of mid-elevation distributional area of a woody population, which vegetative growth followed by mast seeding, and it flowers ca. provides relatively more suitable environmental conditions, the every 50 years (Taylor and Qin, 1988; Wang et al., 2006). In the intrinsic biological traits of the tree species rather than environ- subalpine fir forests of western China, the coexisting species, such mental factors would play a larger role in the structure and as Betula sp., preferred different seed-beds than Abies for regeneration dynamics of the population (Wang et al., 2004). establishment and regeneration, and seedlings of the coexisting Age structure and size distribution of long-lived tree popula- species established more frequently on raised surfaces than Abies tions growing under near optimum conditions often show a (Taylor et al., 2006). This led to more successful regeneration of reverse-J-shape due to the initially high mortality of juvenile trees coexisting species than Abies on raised surfaces and little in the smallest size class (Svensson and Jeglum, 2001; Penuelas regeneration of Abies on the forest floor which was occupied by et al., 2007). In this study, however, the fir populations at the low- bamboo (Taylor and Qin, 1988; Taylor et al., 2006). Without large- and mid-elevations demonstrated a unimodal regeneration type scale disturbances such as fire or bamboo mass-flowering to (Figs. 3 and 4). This size and age structure is probably indicative of reduce the intensity of the understory bamboo in the future, very the impacts of past disturbances (such as unfavorable climate, pest little opportunity will exist for recruitment in the subalpine fir and pathogen outbreak, intensive understory cover) with peak forests at the low- and mid-elevations in the Qinling Mountains presumably corresponding to periods of high recruitment after (Wang et al., 2006). disturbance (Taylor and Qin, 1988; Penuelas et al., 2007). In most At the upper limits in both aspects, a large number of stem subalpine forests, age structure and regeneration dynamics of a recruitment occurred during the last century (Table 3; Fig. 4). This plant population can be influence by many factors such as might be partly due to the lower level of the understory bamboo disturbance and climatic and environmental conditions (Brubaker, cover at the upper limits in the subalpine regions (Taylor and Qin, 1986; Antos and Parish, 2002; Parish and Antos, 2004, 2006; Wang 1988; Taylor et al., 1996; Ren, 1998; Holz et al., 2006). However, et al., 2004), and a distinct pulse of tree establishment evident in an the lack of stems that were recruited after 1960 at the upper limits age-class distribution is often an indication of regeneration in is partly an artifact of not coring young stems <2 m in height that response to canopy-opening disturbance (Cullen et al., 2001; would likely occupy the <40 years age classes (Fig. 4). Hence it Parish and Antos, 2004). appears that there was an upsurge in stem recruitment in the Disturbances are common in subalpine forests and they period 1850–1960 that has not been sustained in recent decades at influence community structure and dynamics (Taylor et al., the upper limits in both aspects. 1996). Natural disturbances drive the regeneration dynamics of Apart from the reduction in intensity of the understory bamboo most closed-canopy forests by creating opportunities for the (F. qinlingensis), climatic variability might also play an important establishment of new individuals through canopy opening and role in the age structure and regeneration dynamics of the fir have been shown to play a role in the regeneration dynamics populations at the upper limits (Camarero and Gutierrez, 1999; (Veblen et al., 1981; Cullen et al., 2001). In the study region, the Daniels and Veblen, 2004; Penuelas et al., 2007). Tree populations subalpine fir forests were selectively cut several times during the at altitudinal treeline ecotones, where low temperature limits tree time period from the 1790s to 1870s (Zhang, 1989; Dang et al., regeneration and growth, are considered to be very sensitive to 2009), leading to the recruitment pulse of the fir trees from 1830 to climate warming. The age structure and regeneration dynamics of 1890 across the altitudinal range (Fig. 4). Disturbance agents such trees at subalpine treelines can reflect the dynamics (e.g., shift or as fire and livestock grazing that have played an important role in invasion of species) in relation to climate change and variability tree regeneration (Vale, 1981; Elliott and Baker, 2004) are absent in (Payette and Filion, 1985). Many studies have demonstrated that the Qinling Mountains (Wang, 1961), and previous research has upper treelines often respond to climate warming with increases in illustrated that the A. fargesii forest experienced frequent small- recruitment and tree density as well as upward advances over the scale disturbances and few large-scale disturbances (i.e., selective last 50–100 years (Payette and Filion, 1985; Rochefort et al., 1994; Author's personal copy

H. Dang et al. / Forest Ecology and Management 259 (2010) 547–554 553

Szeicz and Macdonald, 1995; Woodward et al., 1995; Cullen et al., Brubaker, L.B., 1986. Responses of tree populations to climatic change. Plant Ecol. 67, 119–130. 2001; Germino et al., 2002; Fagre et al., 2003; Kullman, 2007; Bu, H.M., Dang, H.S., Zhang, Q.F., 2009. Climate change in the Jinshui river basin of Penuelas et al., 2007). In the Qinling Mountains, average annual the upper in recent 50 years and its impacts on ecological environ- temperatures have increased by 0.6 8C since 1950 (Bu et al., 2009). ment. Resour. Environ. Yangtze Basin 18, 459–465 (in Chinese with English abstract). The warmer climate could translate into better conditions for Camarero, J.J., Gutierrez, E., 1999. Structure and recent recruitment at alpine forest- recruitment and growing of subalpine fir at its upper limits and pasture ecotones in the Spanish central Pyrenees. Ecoscience 6, 451–464. facilitate its upward expansion (Penuelas et al., 2007). As a result, Camarero, J.J., Gutie´rrez, E., Fortin, M.J., Ribbens, E., 2005. Spatial patterns of tree tree density and stem recruitment of the fir populations at the recruitment in a relict population of Pinus uncinata: forest expansion through stratified diffusion. J. Biogeogr. 32, 1979–1992. upper limits had significantly increased relative to the fir Carrer, M., Nola, P., Eduard, J.L., Motta, R., Urbinati, C., 2007. Regional variability of populations at the lower and middle elevations during the recent climate–growth relationships in Pinus cembra high elevation forests in the Alps. decades in both aspects in the Qinling Mountains (Tables 2 and 3; J. Ecol. 95, 1072–1083. Chen, M.R., 1983. Climate and Agriculture in the Qinling Mountains. Shaanxi Fig. 4). People’s Press, Xi’an (in Chinese). A seedling bank of long-lived subalpine fir appears critical to the Covington, W.W., Fule´, P.Z., Moore, M.M., Hart, S.C., Kolb, T.E., Mast, J.N., Sackett, S.S., dynamics of the subalpine forests, because it can be the major Wagner, M.R., 1997. Restoration of ecosystem health in southwestern ponder- osa pine forests. J. For. 95, 23–29. source of new canopy trees following mortality from disturbances Cullen, L.E., Stewart, G.H., Duncan, R.P., Palmer, J.G., 2001. Disturbance and climate such as windthrow or insect outbreaks (Parish and Antos, 2006). warming influences on New Zealand Nothofagus tree-line population dynamics. The sustained establishment of fir seedling at the upper limits J. Ecol. 89, 1061–1071. Dang, H., Jiang, M., Zhang, Q., Zhang, Y., 2007. Growth responses of subalpine fir might promote variations in stand structure and regeneration (Abies fargesii) to climate variability in the Qinling Mountain, China. For. Ecol. dynamics of the subalpine fir forests along the altitudinal gradient Manage. 240, 143–150. in the Qinling Mountains, China. Dang, H., Jiang, M., Zhang, Y., Dang, G., Zhang, Q., 2009. Dendroecological study of a subalpine fir (Abies fargesii) forest in the Qinling Mountains, China. Plant Ecol. 201, 67–75. 5. Conclusions Daniels, L.D., Veblen, T.T., 2004. Spatiotemporal influences of climate on altitudinal treeline in northern Patagonia. Ecology 85, 1284–1296. Duncan, R.P., 1989. An evaluation of errors in tree age estimates based on increment Age structure analysis showed a unimodal age structure of the cores in kahikatea (Dacrycarpus dacrydioides). N. Z. Nat. Sci. 16, 31–37. fir population at low- and mid-elevations, indicating that Elliott, G.P., Baker, W.L., 2004. Quaking aspen (Populus tremuloides Michx.) at environmental factors might play an important role in shaping treeline: a century of change in the San Juan Mountains, Colorado, USA. J. Biogeogr. 31, 733–745. its age structure and regeneration dynamics at those sites. Fir Fagre, D.B., Peterson, D.L., Hessl, A.E., 2003. Taking the pulse of mountains: ecosys- regeneration dynamics in the north aspect were similar to those in tem responses to climatic variability. Clim. Change 59, 263–282. the south aspect. The majority of the subalpine fir trees were Fule´, P.Z., Covington, W.W., Moore, M.M., 1997. Determining reference conditions for ecosystem management of southwestern ponderosa pine forests. Ecol. Appl. recruited into the forest in the time period 1830–1890 at each 7, 895–908. altitudinal site, which might be triggered by selective cutting Fyllas, N.M., Dimitrakopoulos, P.G., Troumbis, A.Y., 2008. Regeneration dynamics of events from the 1790s to 1870s. At the low- and mid-elevations, a mixed Mediterranean pine forest in the absence of fire. For. Ecol. Manage. 256, 1552–1559. however, there were no fir seedlings and saplings, and no stem Germino, M.J., Smith, W.K., Resor, A.C., 2002. Conifer seedling distribution and recruitment occurred in the last century, which might probably be survival in an alpine-treeline ecotone. Plant Ecol. 162, 157–168. attributed to the influence of the dense cover of understory Goldblum, D., Rigg, L.S., 2002. Age structure and regeneration dynamics of sugar bamboo on seedling establishment. At the upper limits, sustained maple at the deciduous/boreal forest ecotone, Ontario, Canada. Phys. Geogr. 23, 115–129. fir recruitment was found in the last 150 years, and seedlings and Harcombe, P.A., 1987. Tree life tables. Bioscience 37, 557–568. saplings with a density of 1200 and 900 stems/ha in the north and Holmes, R.L., 1983. Computer-assisted quality control in tree-ring dating and south aspect were also recorded in the recent decades, leading to measurement. Tree-Ring Bull. 43, 69–78. Holz, C.A., Veblen, T.T., Franklin, J., 2006. Tree regeneration responses to Chusquea the significant raise in tree density at the treeline ecotones. The montana bamboo die-off in a subalpine Nothofagus forest in the southern Andes. sustained fir seedling establishment at the upper limits would J. Veg. Sci. 17, 19–28. probably result in variations in stand structure and regeneration Kullman, L., 2007. Tree line population monitoring of Pinus sylvestris in the Swedish Scandes, 1973–2005: implications for tree line theory and climate change dynamics of the subalpine fir forests along the altitudinal gradient ecology. J. Ecol. 95, 41–52. in the Qinling Mountains, China. Lingua, E., Cherubini, P., Motta, R., Nola, P., 2008. Spatial structure along an altitudinal gradient in the Italian central Alps suggests competition and facili- tation among coniferous species. J. Veg. Sci. 19, 425–436. Acknowledgements Liu, H., Tang, Z., Dai, J., Tang, Y., Cui, H., 2002. Larch timberline and its development in . Mt. Res. Dev. 22, 359–367. This research was supported by the Funds of the Chinese Mast, J.N., Fule, P.Z., Moore, M.M., Covington, W.W., Waltz, A.E.M., 1999. Restoration of presettlement age structure of an Arizona ponderosa pine forest. Ecol. Appl. 9, Academy of Sciences for Key Topics in Innovation Engineering 228–239. (KSCX2-YW-Z-1023) and the National Natural Science Foundation Motta, R., Nola, P., 2001. Growth trends and dynamics in sub-alpine forest stands in of China (30900199). We thank Mr. Gaodi Dang and Xinping Ye for the Varaita Valley (Piedmont, Italy) and their relationships with human activi- ties and global change. J. Veg. Sci. 12, 219–230. the assistance in field work, Mr. Xu Pang and Ms. Xiuxia Chen for Nakashizuka, T., 1991. Population dynamics of coniferous and broad-leaved trees in assistance with tree core processing, and two anonymous reviewers a Japanese temperate mixed forest. J. Veg. Sci. 2, 413–418. for their useful suggestions. We also thank the Foping National Narukawa, Y., Yamamoto, S., 2002. Effects of dwarf bamboo (Sasa sp.) and forest floor microsites on conifer seedling recruitment in a subalpine forest, Japan. For. Nature Reserve for the logistical support during the field work. Ecol. Manage. 163, 61–70. North, M., Chen, J., Oakley, B., Song, B., Rudnicki, M., Gray, A., Innes, J., 2004. Forest References stand structure and pattern of old-growth western hemlock/Douglas-fir and mixed-conifer forests. For. Sci. 50, 299–311. Nowacki, G.J., Abrams, M.D., 1997. Radial-growth averaging criteria for reconstruct- Antos, J.A., Guest, H.J., Parish, R., 2005. The tree seedling bank in an ancient montane ing disturbance histories from presettlement-origin oaks. Ecol. Monogr. 67, forest: stress tolerators in a productive habitat. J. Ecol. 93, 536–543. 225–249. Antos, J.A., Parish, R., 2002. Structure and dynamics of a nearly steady-state Parish, R., Antos, J.A., 2004. Structure and dynamics of an ancient montane forest in subalpine forest in south-central British Columbia, Canada. Oecologia 130, coastal British Columbia. Oecologia 141, 562–576. 126–135. Parish, R., Antos, J.A., 2006. Slow growth, long-lived trees, and minimal disturbance Barry, R.G., 1992. Mountain Weather and Climate, 2nd ed. Routledge, London. characterize the dynamics of an ancient, montane forest in coastal British Block, J., Treter, U., 2001. The limiting factors at the upper and lower forest limits in Columbia. Can. J. For. Res. 36, 2826–2838. the mountain-woodland steppe of Northwest Mongolia Joachim Block and Uwe Payette, S., Filion, L., 1985. White spruce expansion at the tree line and recent Treter. In: Proceedings of the International Conference on Tree Rings and climate change. Can. J. For. Res. 15, 241–251. People. Davos, pp. 22–26. Author's personal copy

554 H. Dang et al. / Forest Ecology and Management 259 (2010) 547–554

Penuelas, J., Ogaya, R., Boada, M., Jump, A.S., 2007. Migration, invasion and decline: Veblen, T.T., 1989. Tree regeneration responses to gaps along a transandean gradi- changes in recruitment and forest structure in a warming-linked shift of ent. Ecology 70, 541–543. European beech forest in Catalonia (NE Spain). Ecography 30, 829–837. Veblen, T.T., Donoso, C., Schlegel, F.M., Escobar, B., 1981. Forest dynamics in south- Peterson, D.W., Peterson, D.L., Ettl, G.J., Morrison, S.F., Forbes, G.J., Young, S.J., O’Neill, central Chile. J. Biogeogr. 8, 211–247. K.P., Kasischke, E.S., Richter, D.D., Cruickshank, M.G., 2002. Growth responses of Veblen, T.T., Schlegel, F.M., Escobar, B., 1980. Structure and dynamics of old-growth subalpine fir to climatic variability in the Pacific Northwest. Can. J. For. Res. 32, Nothofagus forests in the Valdivian Andes, Chile. J. Ecol. 68, 1–31. 1503–1517. Wang, C.W., 1961. The Forests of China with a Survey of Grassland and Desert Ren, R., 1998. Vegetation of Giant Panda’s Habitat of the Qinling Mountains. Shaanxi Vegetation. Harvard University, Cambridge. Science and Technology Press, Xi’an (in Chinese). Wang, T., Liang, Y., Ren, H., Yu, D., Ni, J., Ma, K., 2004. Age structure of Picea Rochefort, R.M., Little, R.L., Woodward, A., Peterson, D.L., 1994. Changes in sub- schrenkiana forest along an altitudinal gradient in the central Tianshan Moun- alpine tree distribution in western North America: a review of climatic and tains, northwestern China. For. Ecol. Manage. 196, 267–274. other causal factors. Holocene 4, 89–100. Wang, W., Franklin, S.B., Ren, Y., Ouellette, J.R., 2006. Growth of bamboo Fargesia Svensson, J.S., Jeglum, J.K., 2001. Structure and dynamics of an undisturbed old- qinlingensis and regeneration of trees in a mixed hardwood-conifer forest in the growth Norway spruce forest on the rising Bothnian coastline. For. Ecol. Qinling Mountains, China. For. Ecol. Manage. 234, 107–115. Manage. 151, 67–79. Wangda, P., Ohsawa, M., 2006. Structure and regeneration dynamics of dominant Szeicz, J.M., Macdonald, G.M., 1995. Recent white spruce dynamics at the subarctic tree species along altitudinal gradient in a dry valley slopes of the Bhutan alpine treeline of North-Western Canada. J. Ecol. 83, 873–885. Himalaya. For. Ecol. Manage. 230, 136–150. Taylor, A.H., Qin, Z.S., 1988. Regeneration patterns in old-growth Abies–Betula Whittaker, R.H., 1956. Vegetation of the great smoky mountains. Ecol. Monogr. 26, forests in the Wolong Natural Reserve, , China. J. Ecol. 76, 1204–1218. 1–80. Taylor, A.H., Qin, Z.S., Jie, L., 1996. Structure and dynamics of subalpine forests in the Woodward, A., Schreiner, E.G., Silsbee, D.G., 1995. Climate, geography, and tree Wang Lang Natural Reserve, Sichuan, China. Plant Ecol. 124, 25–38. establishment in subalpine meadows of the Olympic Mountains, Washington, Taylor, A.H., Shi Wei, J., Lian Jun, Z., Chun Ping, L., Chang Jin, M., Jinyan, H., 2006. USA. Arct. Alp. Res. 27, 217–225. Regeneration patterns and tree species coexistence in old-growth Abies–Picea Yan, J.P., 2006. Comparable Studies on Environmental Changes in Southern and forests in southwestern China. For. Ecol. Manage. 223, 303–317. Northern Qinling Mountains. China Science Press, Beijing (in Chinese). Vale, T.R., 1981. Tree invasion of montane meadows in Oregon. Am. Midl. Nat. 105, Zhang, Y.Q., 1989. Shaanxi Forest. Xi’an Science and Technology Press, Xi’an 61–69. (in Chinese).