- Email: [email protected]
Comparison of tree sprouting in three regeneration stages of an evergreen broadleaved forest in a karst landscape, SW China
Acta Ecologica Sinica 31 (2011) 126–132
Contents lists available at ScienceDirect
Acta Ecologica Sinica journal homepage: www.elsevier.com/locate/chnaes
Comparison of tree sprouting in three regeneration stages of an evergreen broadleaved forest in a karst landscape, SW China Youxin Shen a,⇑, Guangrong Yang b, Jin Huang b a b
Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences (CAS), Kunming 650223, PR China Shilin Stone Forest Geographical Park, Kunming 652211, PR China
a r t i c l e
i n f o
Article history: Received 26 January 2010 Revised 27 January 2011 Accepted 3 March 2011
Keywords: Karst forest Recruitment Persistence Sprouting capability Sprouting intensity
a b s t r a c t Little is known about the role of tree sprouting in the regeneration of karst forest communities. In Shilin Stone Forest Geographical Park, southwestern China, all genets with the largest stem P3 cm DBH (diameter at breast height) and/or stumps P3 cm BD (basal diameter) were identified and number of sprouts counted in 10 transects (10 m 100 m) in each of three evergreen broadleaved forest stands representing three regeneration stages (about 10, 20, and 30 years old). Species with >10 genets accounted for 72.4% of the 76 species, and all of them showed evidence of sprouting. One-third to two-thirds of the genets in the three forests were sprouting, with an average of 4.0–5.7 sprouts per sprouting genet. Sprouting capability (sprouting genets/total genets) and intensity (sprouts per sprouting genet) differed significantly among the three forest stages. More than 90% of the damaged genets were sprouting. The number of sprouts in a non-damaged genet was determined by intrinsic sprouting ability, and the number of sprouts in damaged genets was determined by stump size. As the forest developed, percentage of damaged genets increased, the portion of shoots P3 cm DBH co-existing in a genet decreased, and the portion of shoots <3 cm DBH sprouted from damaged genet increased. Thus the role of sprouting changed from contributing recruitment in the young stage to persistence in the later stage. Ó 2011 Ecological Society of China. Published by Elsevier B.V. All rights reserved.
1. Introduction Trees have two modes of regeneration: sexual reproduction by seedlings and vegetative recruitment by sprouts. Sprouts grow much faster than seedlings and can quickly re-occupy their own gaps when a tree is blown over or damaged because of the already established and functioning root system [1]. In contrast, some species with apparently non-damaged trees have a sprout bank independent of disturbance, and these trees are replaced by their sprouts [2,3]. Sprouting ability can have a major impact on plant populations by reducing turnover of genets [4,5] and minimizing effects of disturbance [6,7] via recruiting new sprouts; and increasing in situ persistence [5]. The study of sprouting has been recommended as a means of extending our understanding of plant traits and functions in forest communities [3,5,8]. However, most existing observations are based on damaged or coppiced genets, and the other genets, which may play different functional roles, have been excluded. Consequently, the status and functional role of sprouting in forest dynamics are uncertain. ⇑ Corresponding author. Address: Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, 88 Xuefu Road, Kunming 650223, PR China. Tel.: +86 871 5160910; fax: +86 871 5160916. E-mail address: [email protected] (Y. Shen).
Sprouting is a result of the species’ intrinsic abilities combined with specific environmental cues. In a particular environment, some species/individuals will never sprout, while others have a strong tendency to sprout [9]. Sprouters tend to be more numerous in harsh, unstable conditions [10,11] and in less-productive sites [12]. The incidence of sprouting species varies greatly with forest types and environmental variation [13–16]. Sprouting intensity is largely determined by amount of carbohydrates stored in the roots [1] and by frequency and severity of disturbance [9,10]. On a continuum of forest development from the early stage toward the climax stage, forests differ in species composition, structure, and function [17], and individual trees, whether sprouters or not, will experience different biotic and abiotic environments and be subjected to different ongoing disturbances. Thus, incidence and intensity of sprouters, together with the functional role of their sprouts may vary. However, few of studies have focused on sprouts’ contributions to this process [3,18]. Karst landscapes and hydrological systems on limestone and other soluble rocks, which are distributed widely throughout the world [19], are highly fragile environments, and they are vulnerable to a variety of disturbances. Thus they should be suitable sites for sprouting studies. In a typical karst slope, soil and water are easily lost into the epikarst and deeper underground, where they are routed through caves to the springs at the regional boundaries.
1872-2032/$ - see front matter Ó 2011 Ecological Society of China. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.chnaes.2011.03.001
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
As a result of this, a rocky and stony land surface with bedrock pinnacles is formed. Plants growing on the remaining shallow soil in between these pinnacles may be stressed by shortage of water and soil nutrients, which cause them to die [20–22]. Thus, a karst environment may induce high frequency and low intensity of tree sprouting. Many karst areas have been densely populated by humans and thus heavily impacted for hundreds of years. Not surprisingly, modification by human activities will strongly impact regeneration of forest and water/soil runoff features. Forest regeneration after human modification of the landscape is a long process, and it consists of several stages. The vulnerable abiotic karst environment may vary greatly from one stage to the other and thus have a strong influence on sprouting capability and intensity. However, there have been relatively few studies done on sprouting regeneration by trees in such sites. The karst area in southern China is one of the largest in the world, covering more than 500,000 km2 [23]. Various types of evergreen broadleaved forests grow in this karst area across a wide range of climate conditions [20]. Although sprouting is known to occur often in karst forests in SW China [24–28], the importance of this method of regeneration has not been quantified at the community level. Thus, we studied three karst forests in different regeneration stages in a typical karst area in southern China. Evidence of sprouting of each genet in which the main stem wasP3 cm DBH (diameter at breast height) and /or the stumpP3 cm BD (basal diameter) and number of sprouts were documented and analyzed to compare the capability and intensity of tree sprouting in these forests. The purposes of our study were to: (1) determine sprouting capability and intensity of each tree species; (2) compare role of sprouting in the three stages of forest regeneration; and (3) compare sprouting in forests of southwest China to that in forests in other parts of the world.
2. Material and methods 2.1. Study sites The study was carried out in Shilin Stone Forest Geographical Park (SGP) (24°38´–24°58´N, 103°11´-103°29´E), a karst geo-park famous for its various stone forest landforms. The climate is subtropical and semi-humid. Mean annual temperature is 16.2 °C, mean maximum temperature of the warmest month (July) is 20.7 °C, and mean minimum temperature of the coldest month (January) is 8.2 °C. Average annual precipitation is 967.9 mm, 80% of which falls between May and October [29]. The moderate temperature period from late spring to early autumn coincides with the rainy season and this favors plant growth [20]. Dissolution on the exposed limestone bedrock surface produces both soil and bare rock outcroppings at the surface of upland terraces. The small size of rock gaps, rock ditches, small rock caves, and rock slots are typical surface forms on the ground, and soil is distributed in or between these various rock surface forms [29]. Water and nutrients are the main limiting factors for plant growth in these shallow soil patches, especially during the dry season. The zonal vegetation type on this karst land is semi-evergreen broadleaved forest dominated by Cyclobalanopsis glaucoides, Olea yunnanensis, and Neolitsea homilantha [20,30]. Historically, agricultural practices were rudimentary, with subsistence cultivation in SGP. Rapid growth of the human population in this area resulted in intensive clearing of forests, especially in the last century. Forest cutting for fuel wood and burning to clear land for cultivated crops and pastures occurred in this area. However, in the past decades, under the efforts of the SGP authority, local government and villagers living within SGP, some mountains and hills were closed and then forest patches regenerated naturally. Three forests (SF1, SF2,
127
SF3) with different closing and regenerating time schedules were selected for this study. The three forests have very similar abiotic environments. The altitude of the three sites is about 1900 m, slopes are 10°–20°, and soil surface/rock surface ratio is 2:3. SF3 is about 30 years old and contains two tree strata, canopy height is 5.1 m and average DBH 8.0 cm. SF2 is about 20 years old and contains only one canopy stratum; canopy height is 3.5 m and average DBH 5.7 cm. SF1, which has been generating for about 10 years, has a canopy height of 3.3 m and an average DBH of 4.1 cm. Shen et al. (2005) described the floristic composition and community structure of SF3 and SF2 in detail [31]. The SGP authority declared that while these three forest stands were strongly preserved, removal of dead and broken tree trunks was allowed, leaving the stumps in place. This natural dying or breaking of stems was treated as damage to the genets in this study.
2.2. Sprouting survey and statistical analysis Ten transects 10 m in width along the slope were established in each of the three forest stands and each transect was divided into 10 quadrats (10 m 10 m) with plastic lines. Sprouting surveys were carried out in each of those quadrats. We refer to each individual plant as a genet, even if it had only one stem, and each stem in a genet as a shoot (ramet). All shoots of the genet originated from a seed, and they were morphologically connected by roots. In a genet with multiple shoots, shoots that generated other shoots are referred to as the main shoot and the others as sprouting shoots. Once the main shoot was damaged or dead below breast height, we considered it to be a stump. In each quadrat, genets with the main shootP3 cm DBH and/or with the stumpP3 cm BD were carefully checked, DBH or BD measured, and number of sprouts counted. Connection of shoots by roots was checked by excavation whenever there was any uncertainty. If a sprouting genet was recorded for a species in the forest, this species was considered ‘‘able to sprout’’; otherwise, it was ‘‘unable to sprout’’. Since sprouting was strongly influenced by disturbance regimes and varied among study sites, a species that sprouted in one forest might not sprout in the others. For consistency, once a species sprouted at one of the three forests, we considered it ‘‘able to sprout’’ in all three of them. The sprouting rate (SR) was calculated as number of sprouting genets/total number of genets to indicate the sprouting capability of a species and forest stand. To assess the intensity of sprouting, average sprouts per genet was calculated and compared among the three forests, and then genets were divided into five categories: 0, 1–5, 6–10, 11–15, P16 sprouts per genet to check the distribution pattern. Presence of a stump was taken as an indicator of damage, and then all genets were categorized into two groups: damaged genets with stumps and non-damaged genets without stumps. Sprouting rates and intensities were compared between these two groups. Some genets may be capable of sprouting, yet left no sign of it during the survey. Therefore, we might have underestimated some sprouting rates. It was difficult to distinguish sprouts and resprouts, especially in genets that had both shoots and stumps. Thus, we treated all of them as sprouts. One-way ANOVA followed by Duncan’s Post Hoc Tests were used to test for differences in capability and intensity of sprouting among the three forest types. Differences in capability and intensity of sprouting among species were tested with likelihood-ratio Chi-square tests. All the sprouting genets that were not damaged (751) were selected and grouped into six classes to evaluate the relationship of DBH of the main shoots and average number of spouts via ANOVA. To assess effects of stump size on sprouting intensity, all genets with stumps (654) were selected and then divided into seven BD classes. One-way ANOVA followed by
128
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
among sprouting rates of species (v2 = 178.5, P < 0.001). For most species present at all three sites, such as Neolitsea homilantha and Olea yunnanensis, genet density was lowest, but sprouting rate was highest in the youngest forest stage (SF1) (Table 1). Exceptions were found in two cases (Dalbergia mimosoides and Pyrus pashia) in which SR was lower in SF1. Sprouting rate was highly significantly different among the three forests (F = 29.8, P < 0.001, n = 30). In the youngest forest (SF1), 63.8% of the genets were sprouting, but only 43.1% in SF2 and 29.9% in SF3 (Fig. 1).
Duncan’s Post Hoc Tests were used to test for differences in intensity among the seven classes. Incidence of sprouting species/total number of species in a forest stage was also compared with other studies. To evaluate the contribution of sprouts on forest structure, shoots were separated intoP3 cm DBH and<3 cm DBH and those percentages of total shoots were calculated.
3. Results 3.1. Species and their sprouting rate
3.2. Damage and sprouting rate
A total of 6250 genets belonging to 76 taxa were identified. Fifty-five species, all of which had >10 genets, showed evidence of sprouting, and 21 species with <10 genets each (average 2.3) did not. Sprouting rate of species varied from 0% to 100% with an average of 34.7% (Table 1). Significant differences were found
Among the total of 6250 genets identified, 81.3% did not have stumps, and 18.7% (damaged genets) obviously had stumps. Damage status differed significantly among the three forests (F = 71.4, P < 0.001, n = 30). Only 2.6% of the genets obviously had stumps in the youngest forest (SF1); however, the percentage increased
Table 1 Number of genets with main stem P3 cm DBH and/or with stump P3 cm BD and sprouting rate per species in the three regenerating karst forests. Species
Family
Cyclobalanopsis glaucoides Schottky Olea yunnanensis
Fagaceae
SF1 N
Hand.-Mazz. Neolitsea homilantha Allen Pistacia weinmannifolia J.Poisoon ex Fr. Toxicodendron griffithii.(Hook.f.) O. Ktze. Carpinus mobeigiana Hand.-Mazz Machilus yunnanensis Lecomie Rhamnus leptophyllus Schneid. Dalbergia mimosoides Franch. Rhamnella martini (Levl.) Schneid. Jasminum humile Linn. Pistacia chinensis Bunge Diospyros mollifolia Rehd. et
Oleaceae
SF2 SR%
SF3
N
SR%
Total N
SR%
834
35.0
N 692
SR% 46.0
1526
35.1
25
76.0
956
35.0
385
46.0
1366
38.9
Lauraceae Anacardiaceae
345 11
71.0 54.6
291 600
39.5 40.5
328 349
17.7 18.6
964 960
43.4 32.7
Anacardiaceae Corylaceae
1
67
25.4
60 98
36.7 11.2
128 98
30.5 11.2
68
38.2
26
19.2
94
33.0
0
23
0
63
19.0
88
13.6
4.5 33.3 25.0 15.4 0
85 82 76 64 52
25.9 24.4 25.0 18.8 50.0
0
Lauraceae Rhamnaceae
2
Papilionaceae Rhamnaceae Oleaceae Anacardiaceae Ebenaceae
14 25
28.6 36.0
49 51
34.7 17.6
2 46
0 56.5
23 2
26.1 0
22 6 76 39 4
Wils. Ficus chapaensis Gagnep. Albizia mollis (Wall.) Boiv. Toxicodendron succedaneum
Moraceae Mimosaceae Anacardiaceae
1
30 1
6.7 0
50 18 45
22.0 27.7 31.1
51 48 48
23.5 14.6 33.3
(L.) O. Ktze. Zanthoxylum scandens Bl. Osyris wightiana Wall. Celtis tetrandra Roxb. Ilex macrocarpa Oliv. Dichotomanthes tristaniaecarpa Kurz Myrsine semiserrata Wall. Juniperus formosana Hayata Pittosporum brevicalyx (Oliv.) Sageretia theezans Linn. Lindera communis Hemsl. Prunus zippeliana Miq. Pentapanax henryi Harms Milletia dielsiana Harms Distyliopsis laurifolia (Hemsl.)Endress Nothapodytes tomentosa C. Y.
Rutaceae Santalaceae Ulmaceae Aquilifoliaceae Rosaceae Myrsinaceae Cupressaceae Pittosporaceae Rhamnaceae Lauraceae Rosaceae Araliaceae Papilionaceae Hamamelidaceae Icacinaceae
39
10.3
29 5
31.0 40.0
20
50.0
25 29 8 27
24.0 27.6 37.5 40.7
22
4.6 21 16 18 13 12 10 16 10
28.6 6.3 27.8 23.1 33.3 10.0 12.5 40.0
39 37 30 29 29 27 22 21 21 19 18 17 18 16 11
10.3 37.8 26.7 27.6 48.3 40.7 4.6 28.6 14.3 26.3 22.2 35.3 11.1 12.5 45.5
Wu Pyrus pashia Buch.-Ham. Ex
Rosaceae
11
63.6
D. Don Other sprouting species (23) Non-sprouting species (21) Total
100.0
2
100.0
8
62.5
1
100.0
3 1 5
66.7 0 20.0
2
0
3
33.3
5 5
40.0 0
1
100.0
6
33.3
3
22 4
54.5
22 10
31.8
50 34
44.0
94 48
43.6
528
63.8
3131
33.7
2591
29.9
6250
34.7
100
2
100
Species with >10 genets are listed. N: Total genets with the main stem P3 cm DBH and/or with stump P3 cm BD. SC: sprouting rate was computed as the number of sprouting genets/total number of genets per species or per forest.
129
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
80
a
100
60
Sprouting Rate (%)
Sprouting rate (%)
70 50 a
40
b
30 20 10 0
SF1
SF2
SF3
80 60 40 20
Ave. 0
Fig. 1. Sprouting rate (mean ± SE) of the three regenerating karst forests in SW China.
Non-damaged
3.3. Sprouting intensity Number of sprouts per genet varied from 0 to 47 (SF3, O. yunnanensis). A large proportion of genets did not have sprouts, and only a small proportion of genets had>10 sprouts. In each of the three forests, the portion of genets decreased with increase in number of sprouts, and the percentage of four categories with 1– 5, 6–10, 11–15, andP16 sprouts per genet was higher in the youngest stage (SF1) than in the oldest stage (SF3) (Fig. 4). For sprouting genets only, the number of sprouts per sprouting genet by species varied from 3.0 to 6.3 across the three forests (Table 2). There was an average of 4.7 sprouts per sprouting genet for all species. This mean number of sprouts did not differ significantly among species (v2 = 18.18, P = 0.378). However, the average number of sprouts differed significantly among forests (F = 9.4, P = 0.001, n = 30). The oldest forest type (SF3) had 5.7 ± 0.2 sprouts per sprouting genet, while SF2 had 3.9 ± 0.1 and SF1 4.7 ± 0.2. Damaged genets had 5.7 ± 0.3 sprouts per genet and non-damaged genets 3.4 ± 0.2 sprouts. Further analysis showed that the average number of sprouts per genet differed significantly among species (v2 = 237.05, P < 0.001), but not among DBH groups of the main stems (F = 1.8, P = 0.102) for non-damaged genets. However, average number of sprouts per stump did not differ among species (v2 = 39.24, P = 0.286), but it did differ among BD groups of stumps (F = 13.1, P < 0.001); the number of sprouts increased with basal diameter Fig. 5. 3.4. Sprouts and their contribution to forests Shoot density, calculated as the number of shoots per transect (Table 3) showed highly significant differences among the three
c b
a
Fig. 2. Percentage (mean ± SE) of damaged genets in the three regenerating karst forests in SW China.
Damaged Fig. 3. Sprouting rate (mean ± SE) of damaged and of non-damaged genets in the three regenerating karst forests in SW China.
Percentage of genets
to 15.5% in SF2 and 25.7% in SF3 (Fig. 2). Damage had a strong effect on sprouting rate. Almost all damaged genets sprouted again regardless of forest stage and there were no significant differences among forests (F = 1.6, P = 0.22, n = 30). However, sprouting rate of non-damaged genets differed significantly among the three forests (F = 70.6, P < 0.001, n = 30). This capability was higher in the earlier forest stage (SF1) than in the later stages (SF2 and SF3) (Fig. 3).
100 80 60 40 20 0
Fig. 4. Proportion (mean ± SE) of genets with different numbers of sprouts in the three regenerating karst forests in SW China.
forest stages (F = 46.9, P < 0.001, n = 30). Density in the youngest forest was the lowest and the density in middle stage (SF2) was the highest. In this study, shoots<3 cm DBH were all treated as sprouts. It was difficult to trace the sprouting origin of all shootsP3 cm DBH, however, if there were two or more than two shootsP3 cm DBH in a genet, contribution of sprouts can be presumed in a genet. Thus, percentage of shootsP3 cm DBH from more than one shootP3 cm DBH genet was used as another indicator of sprouts’ contribution to forests. The higher percentage the 3 cm DBH from genet with more than one shootsP3cm DBH was, the higher contribution theP3 cm DBH sprouts made to the forest. The proportion of shootsP3 cm DBH or<3 cm DBH differed significantly among the three forest stages (F = 12.0, P < 0.001, n = 30). The proportion of shootsP3 cm DBH in SF1 and SF3 was significantly lower than the proportion in SF2, and in contrast, the proportion of shoots<3 cm DBH in SF1 and SF3 was significantly higher than the proportion in SF2. Higher portion of shootsP3 cm DBH was from single-shoot genets and a low proportion from multi-shoot genets in SF3. In the youngest forest, a high proportion of shootsP3 cm DBH co-existed in a genet. Most of the small shoots had sprouted from non-damaged genets (non-stump) in the youngest forest (SF1). Conversely, most of them had sprouted from damaged genets (with stump) in the oldest forest (SF3). 4. Discussion Sprouting was common for species in all three forest stages, and significantly more plants sprouted in the youngest forest stage. Of the 21 species that did not sprout, ability to do so was uncertain
130
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
Table 2 Mean number of shoots per sprouting genet per species (mean ± SE) in the three regenerating karst forests. Species
Family
SF1
Jasminum humile Linn. Cyclobalanopsis glaucoides Schottky Olea yunnanensis Hand.-Mazz. Neolitsea homilantha Allen Ficus chapaensis Gagnep. Dichotomanthes tristaniaecarpa Kurz Machilus yunnanensis Lecomie Myrsine semiserrata Wall. Pistacia weinmannifolia J.Poisoon ex Fr. Osyris wightiana Wall. Carpinus mobeigiana Hand.-Mazz. Rhamnella martini (Levl.) Schneid. Oxicodendron succedaneum (L.) O. Ktze. Toxicodendron griffithii (Hook.f.) O. Ktze Pistacia chinensis Bunge Dalbergia mimosoides Franch. Rhamnus leptophylus Schneid. Diospyros mollifolia Rehd. et Wils.
Oleaceae Fagaceae Oleaceae Lauraceae Moraceae Rosaceae Lauraceae Myrsinaceae Anacardiaceae Santalaceae Corylaceae Rhamnaceae Anacardiaceae Anacardiaceae Anacardiaceae Papilionaceae Rhamnaceae Ebanaceae
SF2 3.2 ± 0.2 4.2 ± 0.2 4.3 ± 0.4 19 6.0 ± 1.7 4.4 ± 0.6
6.2 ± 1.3 4.9 ± 0.2 2.0
3.0 ± 1.0 2.8 ± 0.7
4.2 ± 0.2 5.3 ± 1.3
3.6 ± 0.7 5.5 ± 1.5
3.3 ± 0.8
Average
6.3 ± 1.7 6.4 ± 0.3 6.0 ± 0.5 4.9 ± 0.5 4.1 ± 0.8 3.7 ± 1.3 5.6 ± 1.4 4.6 ± 0.4 5.3 ± 0.5
6.3 ± 1.7 4.9 ± 0.2 5.0 ± 0.2 4.8 ± 0.2 5.3 ± 1.4 5.2 ± 1.3 4.6 ± 0.5 4.6 ± 0.4 4.4 ± 0.2 4.4 ± 0.9 4.4 ± 1.1 4.2 ± 0.8 4.1 ± 0.4 3.9 ± 0.4 3.3 ± 0.8 3.2 ± 0.4 3.1 ± 0.6 3.0 ± 0.4
4.4 ± 1.1 10.5 ± 6.5 3.9 ± 0.4 4.8 ± 0.6 4.5 ± 1.4 4.0 3.1 ± 0.6
2.7 ± 0.5 2.0 ± 0.5 2.9 ± 0.5
4.0 ± 0.9
SF3
3.0 ± 0.4
Average
4.7 ± 0.2
3.9 ± 0.1
5.7 ± 0.2
4.7 ± 0.1
Species with a total of P10 sprouting genets are listed.
Average no. of shoots
12 10 8 6 4 2 0
1
2
3
4
5
6
7
Basal diameter class Fig. 5. Number of sprouts per stump with different basal diameters in SGP, SW China. Means not sharing the same letter are significantly different at p < 0.05. (1) BD 6 5 cm; (2) 5 cm < BD 6 10 cm; (3) 10 cm < BD 6 15 cm; (4) 15 cm < BD 6 20 cm; (5) 20 cm < BD 6 25 cm; (6) 25 cm < BD 6 30 cm; (7) DB > 30 cm.
Table 3 Average number of shoots per transect (10 m 100 m) and their different origins in the three karst forests representing different regeneration stages (mean ± SE). Site
SF1
SF2
SF3
Average
Average number of shoots Shoots P3 cm DBH From genets with one shoot P3 cm DBH (%) From genets with more than one shoot P3 cm DBH (%) Subtotal (%) Shoots<3 cm DBH From non-damaged-genets (%) From damaged-genets (%) Subtotal%
216.5 ± 17.1a
681.3 ± 44.4b
640.2 ± 44.4b
512.7 ± 44.3
46.9⁄⁄⁄
17.1 ± 6.8a 20.6 ± 5.4a 37.7 ± 2.8a
36.7 ± 11.7b 18.6 ± 6.4a 55.3 ± 2.3b
33.1 ± 10.3b 6.8 ± 2.5b 39.9 ± 3.2a
29.0 ± 12.8 15.3 ± 7.9 44.3 ± 2.1
11.3⁄⁄⁄ 21.7⁄⁄⁄ 12.0⁄⁄⁄
59.3 ± 3.5a 2.9 ± 1.4a 62.3 ± 2.8a
14.2 ± 2.8b 30.5 ± 2.4b 44.7 ± 2.6b
4.4 ± 0.6c 55.7 ± 3.1c 60.1 ± 3.2a
26.0 ± 4.7 29.7 ± 4.2 55.7 ± 2.1
116.7⁄⁄⁄ 124.0⁄⁄⁄ 12.0⁄⁄⁄
since their density was low (<10 genets per species). Also, information on the juvenile stage and disturbance history was lacking. Species may sprout at the juvenile stage and lose the capacity to do so at the adult stage [14,32], and sprouting may be triggered by disturbances [8,9]. Disregarding this uncertainty and considering only species that had >10 genets, the incidence of sprouting species was very high (100%) compared with other studies, most of those being based on damaged or coppiced genets. McLaren and McDonald [16] reported that 94% of species resprouted 14 months after the stumps were coppiced in a tropical dry limestone forest in Jamaica [16]. Everham and Brokaw [14] found that the incidence of sprouting species after wind throw was 35.9% in temperate forest sites and 51.5% in tropical sites [14]. However, the 29.9%–63.8% of sprouting rate in our study (Table 1) was higher than the 6% in a
F
50-ha permanent plot in a tropical moist forest on Barro Colorado Island, Panama [3] and the 3%–15% in different successional stages in the Atlantic tropical forest of southern Brazil [18]. This high incidence and high capability of sprouting in karst forests in SW China may be associated with harsh environmental conditions there. Sprouting is an important component of life-history strategy where environmental conditions are harsh or disturbance is severe [9–11]. Sprouters also tend to be more numerous in less productive sites [12]. Shallowness and patchiness is the common characteristics of karst soil [19,33]. However, trees in karst are largely rooted to the upper 2 m of the soil/bedrock profile, and they depend mostly on water stored within this layer [34]. In addition, the outcropping pinnacles that isolate soil into various patches also restrict the horizontal distribution of roots. Trees growing in these
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
poor soil conditions may experience limitations of nutrients and water. This karst environment may favor species that reproduce both sexually and asexually. However, the intensity of sprouting recorded in this study (average of 4.7 sprouts per sprouting genet and a small portion of genets with>10 sprouts) was not as high as that reported in other studies. For instance, Miller [6] reported 11 sprouts per genet 3 months after a tropical deciduous forest was slashed and burned in Mexico [6]. McLaren and McDonald [16] coppiced 476 stumps of 51 species in a tropical dry limestone forest in Jamaica and recorded an average of 23 sprouts per stump 14 months later [16]. The low intensity in our study is due partly to the effect of small DBH/BD (average 4.1 cm–8.0 cm) since below-ground carbohydrate stored in the genets is an important factor determining intensity [1,16]. The poor soil conditions may also limit carbohydrate storage in roots and shoots and thus inhibit sprouting. Thirdly, our data came from natural forests, where any new sprouts would be exposed to inter- and intra-genet competition. Some sibling sprouts might have died prior to our survey, thus, the intensity calculated using only the extant sprouts might be lower than those in sites, where all existing shoots were coppiced. Differences in capability of sprouting were found among species, as shown in other studies [3,9]. However, the differences in intensity did not coincide well with other studies. Most existing observations and theory based on damaged or coppiced genets support differences in intensity between species [16,35]. Largestem coppicing [35] or larger-sized stump [36] produced more sprouts than smaller stems or stumps. However, in our study differences were found between species for non-damaged genets (without stumps), but not for damaged genets (with stumps). Conversely, sprouting intensity differed significantly among BD groups of damaged genets, but not among DBH groups of non-damaged genets. This suggests that sprouting intensity patterns differ between damaged and non-damaged genets. Thus, we conclude that intrinsic sprouting ability, rather than size of stem (DBH), determines the number of sprouts in a non-damaged genet. However, once a genet is damaged the number of sprouts is largely determined by the size of the stump (BD). The three forests in our study can be considered a regeneration continuum after disturbance in the karst area in SW China. Thus, SF1 is the ‘‘stand initiation and regeneration’’ stage; SF2 ‘‘thinning or stem exclusion’’ stage; and SF3 the stage from ‘‘thinning or stem exclusion’’ to ‘‘transition or understory regeneration’’ [17]. Genet status and evidence for its sprouting on the continuum of forest recovery differed among stages. Consequently, sprouting rate and intensity calculated based on species composition of the genets differed from one stage to another, and the role of sprouts was also different. At the youngest regeneration stage (SF1), tree genet density is lower (528 genets, Table 1) than at SF2 or SF3 stages. A large proportion of genets sprouts (63.8%, Table 1 and Fig. 1) and expanded spatially. On the other hand, small trees were not vulnerable to damage which leads to a relatively lower ratio of genets with stump (Fig. 2). This high probability of sprouting creates both a high proportion of large sprouting shoots (more than half of shootsP3 cm DBH are from multi-shoot genets, Table 3) and of young sprouts (62.1% of the total shoots, Table 3) ready to make a positive contribution to net photosynthesis. However, a multishoot genet has a relatively lower competitive advantage since it allocates resources to a number of basal shoots and buds rather than maximizing vertical extension of a single leader, as is the case with a seeder [37,38]. Once the forest developed further, as in SF3, both inter- and intra-genet competition increases, and seeders and genets with a lower intensity of sprouting will gradually dominate the canopy layer after stem thinning. There is a decrease Table 1 and Fig. 1), and the death rate of adult and young sprouts may increase. Consequently, the ratio of stumps in the later stage forests
131
will increase (Fig. 2), and the proportion of shootsP3 cm DBH from multi-shoot genets will decrease (Table 3). Numerous ‘‘stumps with sprouts’’ are created (61.4%, Table 3), and thus significant sprout banks are produced that persist in the forest (Table 3). This suggests that sprouts play different roles in different developmental stages of karst forests by contributing shoots for recruitment in the younger stage and merely persisting in the older forests. However, our study examined only three forests representing different regeneration stages. A wide range of forests representing the full stages of forest regeneration would further improve our understanding of sprouting behavior and sprout dynamics in karst areas. Long-term investigations on the survival and growth rate of sprouts are required to fully understand the dynamics of sprouting and to assess its importance to structure and persistence of these forests. Acknowledgements This research was supported by ‘‘Light in Western China’’ Program of Chinese Academy of Sciences and ‘‘Key technology integration for soil and water conservation measures and its demonstration in karst mountain areas’’ (No. 2007BAD53B02). We thank Shilin Stone Forest Geographical Park for providing logistical assistance during the fieldwork period; and Jerry and Carol Baskin, Douglas Schaefer for comments on the manuscript and help with the English. References [1] Y. Iwasa, T. Kubo, Optimal size of storage for recovery after unpredictable disturbance, Evol. Ecol. 11 (1997) 41–65. [2] P.J. Bellingham, T. Kohyama, S. Aiba, The effects of a typhoon on Japanese warm temperate rainforests, Ecol. Res. 11 (1996) 229–247. [3] C.J. Paciorek, R. Condit, S.P. Hubbell, R.B. Foster, The demographics of resprouting in tree and shrub species of a moist tropical forest, J. Ecol. 88 (2000) 765–777. [4] L. Kammesheidt, Forest recovery by root suckers and above-ground sprouts after slash-and-burn agriculture, fire and logging in Paraguay and Venezuela, J. Trop. Ecol. 15 (1999) 143–157. [5] W.J. Bond, J.J. Midgley, Ecology of sprouting in woody plants: the persistence niche, Trends Ecol. Evol. 16 (2001) 45–51. [6] P.M. Miller, Coppice shoot and foliar crown growth after disturbance of a tropical deciduous forest in Mexico, For. Ecol. Manage. 116 (1999) 163–173. [7] J.P. Grime, J. Hillier, The contribution of seedling regeneration to the structure and dynamics of plant communities, ecosystems and large units of landscape, in: M. Fenner (Ed.), Seeds: The Ecology of Regeneration in Plant Communities, second ed., CAB International, Wallingford, UK, 2001, pp. 361–374. [8] P.J. Bellingham, A.D. Sparrow, Resprouting as a life history strategy in woody plant communities, Oikos 89 (2000) 409–416. [9] P.A. Vesk, M. Westoby, Sprouting ability across diverse disturbances and vegetation types worldwide, J. Ecol. 92 (2004) 310–320. [10] S.K. Pandey, R.P. Shukla, Regeneration strategy and plant diversity status in degraded sal forests, Curr. Sci. India 81 (2001) 95–102. [11] A.F. Deiller, J.M.N. Walter, M. Tremolieres, Regeneration strategies in a temperate hardwood floodplain forest of the Upper Rhine: sexual versus vegetative reproduction of woody species, For. Ecol. Manage. 180 (2003) 215– 225. [12] J.J. Midgley, Why the world’s vegetation is not totally dominated by resprouting plants: because resprouters are shorter than reseeders, Ecography 19 (1996) 92–95. [13] D.C. Le Maitre, J.J. Midgley, Plant reproductive ecology, in: R.M. Cowling (Ed.), The ecology of fynbos: nutrients, fire and diversity, Oxford University Press, London, 1992, pp. 135–174. [14] E.M. Everham, N.V.L. Brokaw, Forest disturbance and recovery from catastrophic wind, Bot. Rev. 62 (1996) 113–185. [15] P. Negreros-Castillo, R.B. Hall, Sprouting capability of 17 tropical tree species after overstory removal in Quintana Roo, Mexico, For. Ecol. Manage. 126 (2000) 399–403. [16] K.P. McLaren, M.A. McDonald, Coppice regrowth in a disturbed tropical dry limestone forest in Jamaica, For. Ecol. Manage. 180 (2003) 99–111. [17] T.T. Kozlowski, Physiological ecology of regeneration of harvested and disturbed forest stands: implications for forest management, For. Ecol. Manage. 158 (2002) 195–221. [18] C.G. Simoes, M.C.M. Marques, The role of sprouts in the regeneration of Atlantic rainforest in southern Brazil, Restor. Ecol. 12 (2007) 53–59. [19] I. Gams, Origin of the term ‘‘karst,’’ and the transformation of the Classical Karst (kras), Environ. Geol. 21 (1993) 110–114.
132
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
[20] Z.Y. Wu, Vegetation of China, Science publisher, Beijing, 1980 (in Chinese). [21] S.J. Wang, Concept deduction and its connotation of karst rocky desertification, Carsologica Sinica 21 (2002) 101–105 (in Chinese, with English abstract). [22] S.Q. Zhu, Ecological research of karst forest, 3, Guizhou Science and Technology Press, Guiyang, 2002 (in Chinese). [23] D.X. Yan, D.H. Zhu, J.T. Weng, X.W. Zhu, X. JHan, X.Y. Wang, G.E. Cai, Y.F. Zhu, G.Z. Cui, Z.Q. Deng, Karst in China, Geographical Publishers, Beijing, 1994 (in Chinese). [24] S.Y. Ding, J. Peng, Study on aboveground biomass of Cyclobalanopsis glaucoides sprouted shrub community in the central part of Yunnan Plateau, J. Yunnan. Univ. 14 (1992) 191–197 (in Chinese, with English abstract). [25] J. Peng, S.Y. Ding, B.R. Wang, W.H. Su, Studies on evergreen broadleaved sprouted shrub communities and their biomass in Kunming region, J. Yunnan. Univ. 14 (1992) 167–178 (in Chinese, with English abstract). [26] X.K. Zhu, S.Q. Zhu, Study on asexual reproduction ecology of Itea ilicifolia in karst stone mountain-root morphology and biomass allocation, J. Mount. Agric. Biol. 19 (2000) 6–9 (in Chinese, with English abstract). [27] L.F. Yu, S.Q. Zhu, J.Z. Ye, L.M. Wei, Z.R. Chen, Evaluation on degradation of karst forest community and human disturbance, Chinese J. Appl. Ecol. 13 (2002) 529–532 (in Chinese, with English abstract). [28] S. Nanami, H. Kawaguchi, R. Tateno, C.H. Li, S. Katagiri, Sprouting traits and population structure of co-occurring Castanopsis species in an evergreen broadleaved forest in southern China, Ecol. Res. 19 (2004) 341–348. [29] F.M. Zhang, H. Geng, Y.H. Li, Y.N. Liang, Y.H. Yang, J. Ren, F.C. Wang, H.L. Tao, Z.D. Li, Study on the Lunan Stone Forest Karst, China, Yunnan Science and Technology Press, Kunming, 1997 (in Chinese).
[30] Z.Z. Jin, J. Peng, Vegetation of Kunming, Yunnan Science and Technology Press, Kunming, 1998 (in Chinese). [31] Y.X. Shen, W.Y. Liu, Y.H. Li, J.W. Cui, Community ecology study on karst semihumid evergreen broadleaved forest at the central part of Yunnan, Guihaia 25 (2005) 321–326 (in Chinese, with English abstract). [32] T. Yamada, Y. Kumagawa, E. Suzuki, Adaptive significance of vegetative sprouting for a tropical canopy tree, Scaphium longiflorum (Sterculiaceae), in a peat swamp forest in Central Kalimantan, Ecol. Res. 16 (2001) 641–647. [33] W.C. Su, W.X. Zhu, The eco-environmental fragility in karst mountain regions of Guizhou Province, J. Mount. Sci. 18 (2000) 429–434 (in Chinese, with English abstract). [34] J.I. Querejeta, H. Estrada-Medina, M.F. Allen, J.S. Jiménez-Osornio, Water source partitioning among trees growing on shallow karst soils in a seasonally dry tropical climate, Oecologia 152 (2007) 26–36. [35] J.M. Espelta, J. Retana, A. Habrouk, Resprouting patterns after fire and response to stool cleaning of two coexisting Mediterranean oaks with contrasting leaf habits on two different sites, For. Ecol. Manage. 179 (2003) 01–414. [36] E.F. Nzunda, M.E. Griffiths, M.J. Lawes, Sprouting by remobilization of aboveground resources ensures persistence after disturbance of coastal dune forest trees, Funct. Ecol. 22 (2008) 577–582. [37] L.M. Kruger, J.J. Midgley, R.M. Cowling, Resprouters vs reseeders in South African forest trees: a model based on forest canopy height, Funct. Ecol. 11 (1997) 101–105. [38] L.M. Kruger, J.J. Midgley, The influence of resprouting forest canopy species on richness in Southern Cape forest, South Africa, Glob. Ecol. Biogeogr. 10 (2001) 567–572.
Contents lists available at ScienceDirect
Acta Ecologica Sinica journal homepage: www.elsevier.com/locate/chnaes
Comparison of tree sprouting in three regeneration stages of an evergreen broadleaved forest in a karst landscape, SW China Youxin Shen a,⇑, Guangrong Yang b, Jin Huang b a b
Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences (CAS), Kunming 650223, PR China Shilin Stone Forest Geographical Park, Kunming 652211, PR China
a r t i c l e
i n f o
Article history: Received 26 January 2010 Revised 27 January 2011 Accepted 3 March 2011
Keywords: Karst forest Recruitment Persistence Sprouting capability Sprouting intensity
a b s t r a c t Little is known about the role of tree sprouting in the regeneration of karst forest communities. In Shilin Stone Forest Geographical Park, southwestern China, all genets with the largest stem P3 cm DBH (diameter at breast height) and/or stumps P3 cm BD (basal diameter) were identified and number of sprouts counted in 10 transects (10 m 100 m) in each of three evergreen broadleaved forest stands representing three regeneration stages (about 10, 20, and 30 years old). Species with >10 genets accounted for 72.4% of the 76 species, and all of them showed evidence of sprouting. One-third to two-thirds of the genets in the three forests were sprouting, with an average of 4.0–5.7 sprouts per sprouting genet. Sprouting capability (sprouting genets/total genets) and intensity (sprouts per sprouting genet) differed significantly among the three forest stages. More than 90% of the damaged genets were sprouting. The number of sprouts in a non-damaged genet was determined by intrinsic sprouting ability, and the number of sprouts in damaged genets was determined by stump size. As the forest developed, percentage of damaged genets increased, the portion of shoots P3 cm DBH co-existing in a genet decreased, and the portion of shoots <3 cm DBH sprouted from damaged genet increased. Thus the role of sprouting changed from contributing recruitment in the young stage to persistence in the later stage. Ó 2011 Ecological Society of China. Published by Elsevier B.V. All rights reserved.
1. Introduction Trees have two modes of regeneration: sexual reproduction by seedlings and vegetative recruitment by sprouts. Sprouts grow much faster than seedlings and can quickly re-occupy their own gaps when a tree is blown over or damaged because of the already established and functioning root system [1]. In contrast, some species with apparently non-damaged trees have a sprout bank independent of disturbance, and these trees are replaced by their sprouts [2,3]. Sprouting ability can have a major impact on plant populations by reducing turnover of genets [4,5] and minimizing effects of disturbance [6,7] via recruiting new sprouts; and increasing in situ persistence [5]. The study of sprouting has been recommended as a means of extending our understanding of plant traits and functions in forest communities [3,5,8]. However, most existing observations are based on damaged or coppiced genets, and the other genets, which may play different functional roles, have been excluded. Consequently, the status and functional role of sprouting in forest dynamics are uncertain. ⇑ Corresponding author. Address: Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, 88 Xuefu Road, Kunming 650223, PR China. Tel.: +86 871 5160910; fax: +86 871 5160916. E-mail address: [email protected] (Y. Shen).
Sprouting is a result of the species’ intrinsic abilities combined with specific environmental cues. In a particular environment, some species/individuals will never sprout, while others have a strong tendency to sprout [9]. Sprouters tend to be more numerous in harsh, unstable conditions [10,11] and in less-productive sites [12]. The incidence of sprouting species varies greatly with forest types and environmental variation [13–16]. Sprouting intensity is largely determined by amount of carbohydrates stored in the roots [1] and by frequency and severity of disturbance [9,10]. On a continuum of forest development from the early stage toward the climax stage, forests differ in species composition, structure, and function [17], and individual trees, whether sprouters or not, will experience different biotic and abiotic environments and be subjected to different ongoing disturbances. Thus, incidence and intensity of sprouters, together with the functional role of their sprouts may vary. However, few of studies have focused on sprouts’ contributions to this process [3,18]. Karst landscapes and hydrological systems on limestone and other soluble rocks, which are distributed widely throughout the world [19], are highly fragile environments, and they are vulnerable to a variety of disturbances. Thus they should be suitable sites for sprouting studies. In a typical karst slope, soil and water are easily lost into the epikarst and deeper underground, where they are routed through caves to the springs at the regional boundaries.
1872-2032/$ - see front matter Ó 2011 Ecological Society of China. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.chnaes.2011.03.001
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
As a result of this, a rocky and stony land surface with bedrock pinnacles is formed. Plants growing on the remaining shallow soil in between these pinnacles may be stressed by shortage of water and soil nutrients, which cause them to die [20–22]. Thus, a karst environment may induce high frequency and low intensity of tree sprouting. Many karst areas have been densely populated by humans and thus heavily impacted for hundreds of years. Not surprisingly, modification by human activities will strongly impact regeneration of forest and water/soil runoff features. Forest regeneration after human modification of the landscape is a long process, and it consists of several stages. The vulnerable abiotic karst environment may vary greatly from one stage to the other and thus have a strong influence on sprouting capability and intensity. However, there have been relatively few studies done on sprouting regeneration by trees in such sites. The karst area in southern China is one of the largest in the world, covering more than 500,000 km2 [23]. Various types of evergreen broadleaved forests grow in this karst area across a wide range of climate conditions [20]. Although sprouting is known to occur often in karst forests in SW China [24–28], the importance of this method of regeneration has not been quantified at the community level. Thus, we studied three karst forests in different regeneration stages in a typical karst area in southern China. Evidence of sprouting of each genet in which the main stem wasP3 cm DBH (diameter at breast height) and /or the stumpP3 cm BD (basal diameter) and number of sprouts were documented and analyzed to compare the capability and intensity of tree sprouting in these forests. The purposes of our study were to: (1) determine sprouting capability and intensity of each tree species; (2) compare role of sprouting in the three stages of forest regeneration; and (3) compare sprouting in forests of southwest China to that in forests in other parts of the world.
2. Material and methods 2.1. Study sites The study was carried out in Shilin Stone Forest Geographical Park (SGP) (24°38´–24°58´N, 103°11´-103°29´E), a karst geo-park famous for its various stone forest landforms. The climate is subtropical and semi-humid. Mean annual temperature is 16.2 °C, mean maximum temperature of the warmest month (July) is 20.7 °C, and mean minimum temperature of the coldest month (January) is 8.2 °C. Average annual precipitation is 967.9 mm, 80% of which falls between May and October [29]. The moderate temperature period from late spring to early autumn coincides with the rainy season and this favors plant growth [20]. Dissolution on the exposed limestone bedrock surface produces both soil and bare rock outcroppings at the surface of upland terraces. The small size of rock gaps, rock ditches, small rock caves, and rock slots are typical surface forms on the ground, and soil is distributed in or between these various rock surface forms [29]. Water and nutrients are the main limiting factors for plant growth in these shallow soil patches, especially during the dry season. The zonal vegetation type on this karst land is semi-evergreen broadleaved forest dominated by Cyclobalanopsis glaucoides, Olea yunnanensis, and Neolitsea homilantha [20,30]. Historically, agricultural practices were rudimentary, with subsistence cultivation in SGP. Rapid growth of the human population in this area resulted in intensive clearing of forests, especially in the last century. Forest cutting for fuel wood and burning to clear land for cultivated crops and pastures occurred in this area. However, in the past decades, under the efforts of the SGP authority, local government and villagers living within SGP, some mountains and hills were closed and then forest patches regenerated naturally. Three forests (SF1, SF2,
127
SF3) with different closing and regenerating time schedules were selected for this study. The three forests have very similar abiotic environments. The altitude of the three sites is about 1900 m, slopes are 10°–20°, and soil surface/rock surface ratio is 2:3. SF3 is about 30 years old and contains two tree strata, canopy height is 5.1 m and average DBH 8.0 cm. SF2 is about 20 years old and contains only one canopy stratum; canopy height is 3.5 m and average DBH 5.7 cm. SF1, which has been generating for about 10 years, has a canopy height of 3.3 m and an average DBH of 4.1 cm. Shen et al. (2005) described the floristic composition and community structure of SF3 and SF2 in detail [31]. The SGP authority declared that while these three forest stands were strongly preserved, removal of dead and broken tree trunks was allowed, leaving the stumps in place. This natural dying or breaking of stems was treated as damage to the genets in this study.
2.2. Sprouting survey and statistical analysis Ten transects 10 m in width along the slope were established in each of the three forest stands and each transect was divided into 10 quadrats (10 m 10 m) with plastic lines. Sprouting surveys were carried out in each of those quadrats. We refer to each individual plant as a genet, even if it had only one stem, and each stem in a genet as a shoot (ramet). All shoots of the genet originated from a seed, and they were morphologically connected by roots. In a genet with multiple shoots, shoots that generated other shoots are referred to as the main shoot and the others as sprouting shoots. Once the main shoot was damaged or dead below breast height, we considered it to be a stump. In each quadrat, genets with the main shootP3 cm DBH and/or with the stumpP3 cm BD were carefully checked, DBH or BD measured, and number of sprouts counted. Connection of shoots by roots was checked by excavation whenever there was any uncertainty. If a sprouting genet was recorded for a species in the forest, this species was considered ‘‘able to sprout’’; otherwise, it was ‘‘unable to sprout’’. Since sprouting was strongly influenced by disturbance regimes and varied among study sites, a species that sprouted in one forest might not sprout in the others. For consistency, once a species sprouted at one of the three forests, we considered it ‘‘able to sprout’’ in all three of them. The sprouting rate (SR) was calculated as number of sprouting genets/total number of genets to indicate the sprouting capability of a species and forest stand. To assess the intensity of sprouting, average sprouts per genet was calculated and compared among the three forests, and then genets were divided into five categories: 0, 1–5, 6–10, 11–15, P16 sprouts per genet to check the distribution pattern. Presence of a stump was taken as an indicator of damage, and then all genets were categorized into two groups: damaged genets with stumps and non-damaged genets without stumps. Sprouting rates and intensities were compared between these two groups. Some genets may be capable of sprouting, yet left no sign of it during the survey. Therefore, we might have underestimated some sprouting rates. It was difficult to distinguish sprouts and resprouts, especially in genets that had both shoots and stumps. Thus, we treated all of them as sprouts. One-way ANOVA followed by Duncan’s Post Hoc Tests were used to test for differences in capability and intensity of sprouting among the three forest types. Differences in capability and intensity of sprouting among species were tested with likelihood-ratio Chi-square tests. All the sprouting genets that were not damaged (751) were selected and grouped into six classes to evaluate the relationship of DBH of the main shoots and average number of spouts via ANOVA. To assess effects of stump size on sprouting intensity, all genets with stumps (654) were selected and then divided into seven BD classes. One-way ANOVA followed by
128
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
among sprouting rates of species (v2 = 178.5, P < 0.001). For most species present at all three sites, such as Neolitsea homilantha and Olea yunnanensis, genet density was lowest, but sprouting rate was highest in the youngest forest stage (SF1) (Table 1). Exceptions were found in two cases (Dalbergia mimosoides and Pyrus pashia) in which SR was lower in SF1. Sprouting rate was highly significantly different among the three forests (F = 29.8, P < 0.001, n = 30). In the youngest forest (SF1), 63.8% of the genets were sprouting, but only 43.1% in SF2 and 29.9% in SF3 (Fig. 1).
Duncan’s Post Hoc Tests were used to test for differences in intensity among the seven classes. Incidence of sprouting species/total number of species in a forest stage was also compared with other studies. To evaluate the contribution of sprouts on forest structure, shoots were separated intoP3 cm DBH and<3 cm DBH and those percentages of total shoots were calculated.
3. Results 3.1. Species and their sprouting rate
3.2. Damage and sprouting rate
A total of 6250 genets belonging to 76 taxa were identified. Fifty-five species, all of which had >10 genets, showed evidence of sprouting, and 21 species with <10 genets each (average 2.3) did not. Sprouting rate of species varied from 0% to 100% with an average of 34.7% (Table 1). Significant differences were found
Among the total of 6250 genets identified, 81.3% did not have stumps, and 18.7% (damaged genets) obviously had stumps. Damage status differed significantly among the three forests (F = 71.4, P < 0.001, n = 30). Only 2.6% of the genets obviously had stumps in the youngest forest (SF1); however, the percentage increased
Table 1 Number of genets with main stem P3 cm DBH and/or with stump P3 cm BD and sprouting rate per species in the three regenerating karst forests. Species
Family
Cyclobalanopsis glaucoides Schottky Olea yunnanensis
Fagaceae
SF1 N
Hand.-Mazz. Neolitsea homilantha Allen Pistacia weinmannifolia J.Poisoon ex Fr. Toxicodendron griffithii.(Hook.f.) O. Ktze. Carpinus mobeigiana Hand.-Mazz Machilus yunnanensis Lecomie Rhamnus leptophyllus Schneid. Dalbergia mimosoides Franch. Rhamnella martini (Levl.) Schneid. Jasminum humile Linn. Pistacia chinensis Bunge Diospyros mollifolia Rehd. et
Oleaceae
SF2 SR%
SF3
N
SR%
Total N
SR%
834
35.0
N 692
SR% 46.0
1526
35.1
25
76.0
956
35.0
385
46.0
1366
38.9
Lauraceae Anacardiaceae
345 11
71.0 54.6
291 600
39.5 40.5
328 349
17.7 18.6
964 960
43.4 32.7
Anacardiaceae Corylaceae
1
67
25.4
60 98
36.7 11.2
128 98
30.5 11.2
68
38.2
26
19.2
94
33.0
0
23
0
63
19.0
88
13.6
4.5 33.3 25.0 15.4 0
85 82 76 64 52
25.9 24.4 25.0 18.8 50.0
0
Lauraceae Rhamnaceae
2
Papilionaceae Rhamnaceae Oleaceae Anacardiaceae Ebenaceae
14 25
28.6 36.0
49 51
34.7 17.6
2 46
0 56.5
23 2
26.1 0
22 6 76 39 4
Wils. Ficus chapaensis Gagnep. Albizia mollis (Wall.) Boiv. Toxicodendron succedaneum
Moraceae Mimosaceae Anacardiaceae
1
30 1
6.7 0
50 18 45
22.0 27.7 31.1
51 48 48
23.5 14.6 33.3
(L.) O. Ktze. Zanthoxylum scandens Bl. Osyris wightiana Wall. Celtis tetrandra Roxb. Ilex macrocarpa Oliv. Dichotomanthes tristaniaecarpa Kurz Myrsine semiserrata Wall. Juniperus formosana Hayata Pittosporum brevicalyx (Oliv.) Sageretia theezans Linn. Lindera communis Hemsl. Prunus zippeliana Miq. Pentapanax henryi Harms Milletia dielsiana Harms Distyliopsis laurifolia (Hemsl.)Endress Nothapodytes tomentosa C. Y.
Rutaceae Santalaceae Ulmaceae Aquilifoliaceae Rosaceae Myrsinaceae Cupressaceae Pittosporaceae Rhamnaceae Lauraceae Rosaceae Araliaceae Papilionaceae Hamamelidaceae Icacinaceae
39
10.3
29 5
31.0 40.0
20
50.0
25 29 8 27
24.0 27.6 37.5 40.7
22
4.6 21 16 18 13 12 10 16 10
28.6 6.3 27.8 23.1 33.3 10.0 12.5 40.0
39 37 30 29 29 27 22 21 21 19 18 17 18 16 11
10.3 37.8 26.7 27.6 48.3 40.7 4.6 28.6 14.3 26.3 22.2 35.3 11.1 12.5 45.5
Wu Pyrus pashia Buch.-Ham. Ex
Rosaceae
11
63.6
D. Don Other sprouting species (23) Non-sprouting species (21) Total
100.0
2
100.0
8
62.5
1
100.0
3 1 5
66.7 0 20.0
2
0
3
33.3
5 5
40.0 0
1
100.0
6
33.3
3
22 4
54.5
22 10
31.8
50 34
44.0
94 48
43.6
528
63.8
3131
33.7
2591
29.9
6250
34.7
100
2
100
Species with >10 genets are listed. N: Total genets with the main stem P3 cm DBH and/or with stump P3 cm BD. SC: sprouting rate was computed as the number of sprouting genets/total number of genets per species or per forest.
129
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
80
a
100
60
Sprouting Rate (%)
Sprouting rate (%)
70 50 a
40
b
30 20 10 0
SF1
SF2
SF3
80 60 40 20
Ave. 0
Fig. 1. Sprouting rate (mean ± SE) of the three regenerating karst forests in SW China.
Non-damaged
3.3. Sprouting intensity Number of sprouts per genet varied from 0 to 47 (SF3, O. yunnanensis). A large proportion of genets did not have sprouts, and only a small proportion of genets had>10 sprouts. In each of the three forests, the portion of genets decreased with increase in number of sprouts, and the percentage of four categories with 1– 5, 6–10, 11–15, andP16 sprouts per genet was higher in the youngest stage (SF1) than in the oldest stage (SF3) (Fig. 4). For sprouting genets only, the number of sprouts per sprouting genet by species varied from 3.0 to 6.3 across the three forests (Table 2). There was an average of 4.7 sprouts per sprouting genet for all species. This mean number of sprouts did not differ significantly among species (v2 = 18.18, P = 0.378). However, the average number of sprouts differed significantly among forests (F = 9.4, P = 0.001, n = 30). The oldest forest type (SF3) had 5.7 ± 0.2 sprouts per sprouting genet, while SF2 had 3.9 ± 0.1 and SF1 4.7 ± 0.2. Damaged genets had 5.7 ± 0.3 sprouts per genet and non-damaged genets 3.4 ± 0.2 sprouts. Further analysis showed that the average number of sprouts per genet differed significantly among species (v2 = 237.05, P < 0.001), but not among DBH groups of the main stems (F = 1.8, P = 0.102) for non-damaged genets. However, average number of sprouts per stump did not differ among species (v2 = 39.24, P = 0.286), but it did differ among BD groups of stumps (F = 13.1, P < 0.001); the number of sprouts increased with basal diameter Fig. 5. 3.4. Sprouts and their contribution to forests Shoot density, calculated as the number of shoots per transect (Table 3) showed highly significant differences among the three
c b
a
Fig. 2. Percentage (mean ± SE) of damaged genets in the three regenerating karst forests in SW China.
Damaged Fig. 3. Sprouting rate (mean ± SE) of damaged and of non-damaged genets in the three regenerating karst forests in SW China.
Percentage of genets
to 15.5% in SF2 and 25.7% in SF3 (Fig. 2). Damage had a strong effect on sprouting rate. Almost all damaged genets sprouted again regardless of forest stage and there were no significant differences among forests (F = 1.6, P = 0.22, n = 30). However, sprouting rate of non-damaged genets differed significantly among the three forests (F = 70.6, P < 0.001, n = 30). This capability was higher in the earlier forest stage (SF1) than in the later stages (SF2 and SF3) (Fig. 3).
100 80 60 40 20 0
Fig. 4. Proportion (mean ± SE) of genets with different numbers of sprouts in the three regenerating karst forests in SW China.
forest stages (F = 46.9, P < 0.001, n = 30). Density in the youngest forest was the lowest and the density in middle stage (SF2) was the highest. In this study, shoots<3 cm DBH were all treated as sprouts. It was difficult to trace the sprouting origin of all shootsP3 cm DBH, however, if there were two or more than two shootsP3 cm DBH in a genet, contribution of sprouts can be presumed in a genet. Thus, percentage of shootsP3 cm DBH from more than one shootP3 cm DBH genet was used as another indicator of sprouts’ contribution to forests. The higher percentage the 3 cm DBH from genet with more than one shootsP3cm DBH was, the higher contribution theP3 cm DBH sprouts made to the forest. The proportion of shootsP3 cm DBH or<3 cm DBH differed significantly among the three forest stages (F = 12.0, P < 0.001, n = 30). The proportion of shootsP3 cm DBH in SF1 and SF3 was significantly lower than the proportion in SF2, and in contrast, the proportion of shoots<3 cm DBH in SF1 and SF3 was significantly higher than the proportion in SF2. Higher portion of shootsP3 cm DBH was from single-shoot genets and a low proportion from multi-shoot genets in SF3. In the youngest forest, a high proportion of shootsP3 cm DBH co-existed in a genet. Most of the small shoots had sprouted from non-damaged genets (non-stump) in the youngest forest (SF1). Conversely, most of them had sprouted from damaged genets (with stump) in the oldest forest (SF3). 4. Discussion Sprouting was common for species in all three forest stages, and significantly more plants sprouted in the youngest forest stage. Of the 21 species that did not sprout, ability to do so was uncertain
130
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
Table 2 Mean number of shoots per sprouting genet per species (mean ± SE) in the three regenerating karst forests. Species
Family
SF1
Jasminum humile Linn. Cyclobalanopsis glaucoides Schottky Olea yunnanensis Hand.-Mazz. Neolitsea homilantha Allen Ficus chapaensis Gagnep. Dichotomanthes tristaniaecarpa Kurz Machilus yunnanensis Lecomie Myrsine semiserrata Wall. Pistacia weinmannifolia J.Poisoon ex Fr. Osyris wightiana Wall. Carpinus mobeigiana Hand.-Mazz. Rhamnella martini (Levl.) Schneid. Oxicodendron succedaneum (L.) O. Ktze. Toxicodendron griffithii (Hook.f.) O. Ktze Pistacia chinensis Bunge Dalbergia mimosoides Franch. Rhamnus leptophylus Schneid. Diospyros mollifolia Rehd. et Wils.
Oleaceae Fagaceae Oleaceae Lauraceae Moraceae Rosaceae Lauraceae Myrsinaceae Anacardiaceae Santalaceae Corylaceae Rhamnaceae Anacardiaceae Anacardiaceae Anacardiaceae Papilionaceae Rhamnaceae Ebanaceae
SF2 3.2 ± 0.2 4.2 ± 0.2 4.3 ± 0.4 19 6.0 ± 1.7 4.4 ± 0.6
6.2 ± 1.3 4.9 ± 0.2 2.0
3.0 ± 1.0 2.8 ± 0.7
4.2 ± 0.2 5.3 ± 1.3
3.6 ± 0.7 5.5 ± 1.5
3.3 ± 0.8
Average
6.3 ± 1.7 6.4 ± 0.3 6.0 ± 0.5 4.9 ± 0.5 4.1 ± 0.8 3.7 ± 1.3 5.6 ± 1.4 4.6 ± 0.4 5.3 ± 0.5
6.3 ± 1.7 4.9 ± 0.2 5.0 ± 0.2 4.8 ± 0.2 5.3 ± 1.4 5.2 ± 1.3 4.6 ± 0.5 4.6 ± 0.4 4.4 ± 0.2 4.4 ± 0.9 4.4 ± 1.1 4.2 ± 0.8 4.1 ± 0.4 3.9 ± 0.4 3.3 ± 0.8 3.2 ± 0.4 3.1 ± 0.6 3.0 ± 0.4
4.4 ± 1.1 10.5 ± 6.5 3.9 ± 0.4 4.8 ± 0.6 4.5 ± 1.4 4.0 3.1 ± 0.6
2.7 ± 0.5 2.0 ± 0.5 2.9 ± 0.5
4.0 ± 0.9
SF3
3.0 ± 0.4
Average
4.7 ± 0.2
3.9 ± 0.1
5.7 ± 0.2
4.7 ± 0.1
Species with a total of P10 sprouting genets are listed.
Average no. of shoots
12 10 8 6 4 2 0
1
2
3
4
5
6
7
Basal diameter class Fig. 5. Number of sprouts per stump with different basal diameters in SGP, SW China. Means not sharing the same letter are significantly different at p < 0.05. (1) BD 6 5 cm; (2) 5 cm < BD 6 10 cm; (3) 10 cm < BD 6 15 cm; (4) 15 cm < BD 6 20 cm; (5) 20 cm < BD 6 25 cm; (6) 25 cm < BD 6 30 cm; (7) DB > 30 cm.
Table 3 Average number of shoots per transect (10 m 100 m) and their different origins in the three karst forests representing different regeneration stages (mean ± SE). Site
SF1
SF2
SF3
Average
Average number of shoots Shoots P3 cm DBH From genets with one shoot P3 cm DBH (%) From genets with more than one shoot P3 cm DBH (%) Subtotal (%) Shoots<3 cm DBH From non-damaged-genets (%) From damaged-genets (%) Subtotal%
216.5 ± 17.1a
681.3 ± 44.4b
640.2 ± 44.4b
512.7 ± 44.3
46.9⁄⁄⁄
17.1 ± 6.8a 20.6 ± 5.4a 37.7 ± 2.8a
36.7 ± 11.7b 18.6 ± 6.4a 55.3 ± 2.3b
33.1 ± 10.3b 6.8 ± 2.5b 39.9 ± 3.2a
29.0 ± 12.8 15.3 ± 7.9 44.3 ± 2.1
11.3⁄⁄⁄ 21.7⁄⁄⁄ 12.0⁄⁄⁄
59.3 ± 3.5a 2.9 ± 1.4a 62.3 ± 2.8a
14.2 ± 2.8b 30.5 ± 2.4b 44.7 ± 2.6b
4.4 ± 0.6c 55.7 ± 3.1c 60.1 ± 3.2a
26.0 ± 4.7 29.7 ± 4.2 55.7 ± 2.1
116.7⁄⁄⁄ 124.0⁄⁄⁄ 12.0⁄⁄⁄
since their density was low (<10 genets per species). Also, information on the juvenile stage and disturbance history was lacking. Species may sprout at the juvenile stage and lose the capacity to do so at the adult stage [14,32], and sprouting may be triggered by disturbances [8,9]. Disregarding this uncertainty and considering only species that had >10 genets, the incidence of sprouting species was very high (100%) compared with other studies, most of those being based on damaged or coppiced genets. McLaren and McDonald [16] reported that 94% of species resprouted 14 months after the stumps were coppiced in a tropical dry limestone forest in Jamaica [16]. Everham and Brokaw [14] found that the incidence of sprouting species after wind throw was 35.9% in temperate forest sites and 51.5% in tropical sites [14]. However, the 29.9%–63.8% of sprouting rate in our study (Table 1) was higher than the 6% in a
F
50-ha permanent plot in a tropical moist forest on Barro Colorado Island, Panama [3] and the 3%–15% in different successional stages in the Atlantic tropical forest of southern Brazil [18]. This high incidence and high capability of sprouting in karst forests in SW China may be associated with harsh environmental conditions there. Sprouting is an important component of life-history strategy where environmental conditions are harsh or disturbance is severe [9–11]. Sprouters also tend to be more numerous in less productive sites [12]. Shallowness and patchiness is the common characteristics of karst soil [19,33]. However, trees in karst are largely rooted to the upper 2 m of the soil/bedrock profile, and they depend mostly on water stored within this layer [34]. In addition, the outcropping pinnacles that isolate soil into various patches also restrict the horizontal distribution of roots. Trees growing in these
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
poor soil conditions may experience limitations of nutrients and water. This karst environment may favor species that reproduce both sexually and asexually. However, the intensity of sprouting recorded in this study (average of 4.7 sprouts per sprouting genet and a small portion of genets with>10 sprouts) was not as high as that reported in other studies. For instance, Miller [6] reported 11 sprouts per genet 3 months after a tropical deciduous forest was slashed and burned in Mexico [6]. McLaren and McDonald [16] coppiced 476 stumps of 51 species in a tropical dry limestone forest in Jamaica and recorded an average of 23 sprouts per stump 14 months later [16]. The low intensity in our study is due partly to the effect of small DBH/BD (average 4.1 cm–8.0 cm) since below-ground carbohydrate stored in the genets is an important factor determining intensity [1,16]. The poor soil conditions may also limit carbohydrate storage in roots and shoots and thus inhibit sprouting. Thirdly, our data came from natural forests, where any new sprouts would be exposed to inter- and intra-genet competition. Some sibling sprouts might have died prior to our survey, thus, the intensity calculated using only the extant sprouts might be lower than those in sites, where all existing shoots were coppiced. Differences in capability of sprouting were found among species, as shown in other studies [3,9]. However, the differences in intensity did not coincide well with other studies. Most existing observations and theory based on damaged or coppiced genets support differences in intensity between species [16,35]. Largestem coppicing [35] or larger-sized stump [36] produced more sprouts than smaller stems or stumps. However, in our study differences were found between species for non-damaged genets (without stumps), but not for damaged genets (with stumps). Conversely, sprouting intensity differed significantly among BD groups of damaged genets, but not among DBH groups of non-damaged genets. This suggests that sprouting intensity patterns differ between damaged and non-damaged genets. Thus, we conclude that intrinsic sprouting ability, rather than size of stem (DBH), determines the number of sprouts in a non-damaged genet. However, once a genet is damaged the number of sprouts is largely determined by the size of the stump (BD). The three forests in our study can be considered a regeneration continuum after disturbance in the karst area in SW China. Thus, SF1 is the ‘‘stand initiation and regeneration’’ stage; SF2 ‘‘thinning or stem exclusion’’ stage; and SF3 the stage from ‘‘thinning or stem exclusion’’ to ‘‘transition or understory regeneration’’ [17]. Genet status and evidence for its sprouting on the continuum of forest recovery differed among stages. Consequently, sprouting rate and intensity calculated based on species composition of the genets differed from one stage to another, and the role of sprouts was also different. At the youngest regeneration stage (SF1), tree genet density is lower (528 genets, Table 1) than at SF2 or SF3 stages. A large proportion of genets sprouts (63.8%, Table 1 and Fig. 1) and expanded spatially. On the other hand, small trees were not vulnerable to damage which leads to a relatively lower ratio of genets with stump (Fig. 2). This high probability of sprouting creates both a high proportion of large sprouting shoots (more than half of shootsP3 cm DBH are from multi-shoot genets, Table 3) and of young sprouts (62.1% of the total shoots, Table 3) ready to make a positive contribution to net photosynthesis. However, a multishoot genet has a relatively lower competitive advantage since it allocates resources to a number of basal shoots and buds rather than maximizing vertical extension of a single leader, as is the case with a seeder [37,38]. Once the forest developed further, as in SF3, both inter- and intra-genet competition increases, and seeders and genets with a lower intensity of sprouting will gradually dominate the canopy layer after stem thinning. There is a decrease Table 1 and Fig. 1), and the death rate of adult and young sprouts may increase. Consequently, the ratio of stumps in the later stage forests
131
will increase (Fig. 2), and the proportion of shootsP3 cm DBH from multi-shoot genets will decrease (Table 3). Numerous ‘‘stumps with sprouts’’ are created (61.4%, Table 3), and thus significant sprout banks are produced that persist in the forest (Table 3). This suggests that sprouts play different roles in different developmental stages of karst forests by contributing shoots for recruitment in the younger stage and merely persisting in the older forests. However, our study examined only three forests representing different regeneration stages. A wide range of forests representing the full stages of forest regeneration would further improve our understanding of sprouting behavior and sprout dynamics in karst areas. Long-term investigations on the survival and growth rate of sprouts are required to fully understand the dynamics of sprouting and to assess its importance to structure and persistence of these forests. Acknowledgements This research was supported by ‘‘Light in Western China’’ Program of Chinese Academy of Sciences and ‘‘Key technology integration for soil and water conservation measures and its demonstration in karst mountain areas’’ (No. 2007BAD53B02). We thank Shilin Stone Forest Geographical Park for providing logistical assistance during the fieldwork period; and Jerry and Carol Baskin, Douglas Schaefer for comments on the manuscript and help with the English. References [1] Y. Iwasa, T. Kubo, Optimal size of storage for recovery after unpredictable disturbance, Evol. Ecol. 11 (1997) 41–65. [2] P.J. Bellingham, T. Kohyama, S. Aiba, The effects of a typhoon on Japanese warm temperate rainforests, Ecol. Res. 11 (1996) 229–247. [3] C.J. Paciorek, R. Condit, S.P. Hubbell, R.B. Foster, The demographics of resprouting in tree and shrub species of a moist tropical forest, J. Ecol. 88 (2000) 765–777. [4] L. Kammesheidt, Forest recovery by root suckers and above-ground sprouts after slash-and-burn agriculture, fire and logging in Paraguay and Venezuela, J. Trop. Ecol. 15 (1999) 143–157. [5] W.J. Bond, J.J. Midgley, Ecology of sprouting in woody plants: the persistence niche, Trends Ecol. Evol. 16 (2001) 45–51. [6] P.M. Miller, Coppice shoot and foliar crown growth after disturbance of a tropical deciduous forest in Mexico, For. Ecol. Manage. 116 (1999) 163–173. [7] J.P. Grime, J. Hillier, The contribution of seedling regeneration to the structure and dynamics of plant communities, ecosystems and large units of landscape, in: M. Fenner (Ed.), Seeds: The Ecology of Regeneration in Plant Communities, second ed., CAB International, Wallingford, UK, 2001, pp. 361–374. [8] P.J. Bellingham, A.D. Sparrow, Resprouting as a life history strategy in woody plant communities, Oikos 89 (2000) 409–416. [9] P.A. Vesk, M. Westoby, Sprouting ability across diverse disturbances and vegetation types worldwide, J. Ecol. 92 (2004) 310–320. [10] S.K. Pandey, R.P. Shukla, Regeneration strategy and plant diversity status in degraded sal forests, Curr. Sci. India 81 (2001) 95–102. [11] A.F. Deiller, J.M.N. Walter, M. Tremolieres, Regeneration strategies in a temperate hardwood floodplain forest of the Upper Rhine: sexual versus vegetative reproduction of woody species, For. Ecol. Manage. 180 (2003) 215– 225. [12] J.J. Midgley, Why the world’s vegetation is not totally dominated by resprouting plants: because resprouters are shorter than reseeders, Ecography 19 (1996) 92–95. [13] D.C. Le Maitre, J.J. Midgley, Plant reproductive ecology, in: R.M. Cowling (Ed.), The ecology of fynbos: nutrients, fire and diversity, Oxford University Press, London, 1992, pp. 135–174. [14] E.M. Everham, N.V.L. Brokaw, Forest disturbance and recovery from catastrophic wind, Bot. Rev. 62 (1996) 113–185. [15] P. Negreros-Castillo, R.B. Hall, Sprouting capability of 17 tropical tree species after overstory removal in Quintana Roo, Mexico, For. Ecol. Manage. 126 (2000) 399–403. [16] K.P. McLaren, M.A. McDonald, Coppice regrowth in a disturbed tropical dry limestone forest in Jamaica, For. Ecol. Manage. 180 (2003) 99–111. [17] T.T. Kozlowski, Physiological ecology of regeneration of harvested and disturbed forest stands: implications for forest management, For. Ecol. Manage. 158 (2002) 195–221. [18] C.G. Simoes, M.C.M. Marques, The role of sprouts in the regeneration of Atlantic rainforest in southern Brazil, Restor. Ecol. 12 (2007) 53–59. [19] I. Gams, Origin of the term ‘‘karst,’’ and the transformation of the Classical Karst (kras), Environ. Geol. 21 (1993) 110–114.
132
Y. Shen et al. / Acta Ecologica Sinica 31 (2011) 126–132
[20] Z.Y. Wu, Vegetation of China, Science publisher, Beijing, 1980 (in Chinese). [21] S.J. Wang, Concept deduction and its connotation of karst rocky desertification, Carsologica Sinica 21 (2002) 101–105 (in Chinese, with English abstract). [22] S.Q. Zhu, Ecological research of karst forest, 3, Guizhou Science and Technology Press, Guiyang, 2002 (in Chinese). [23] D.X. Yan, D.H. Zhu, J.T. Weng, X.W. Zhu, X. JHan, X.Y. Wang, G.E. Cai, Y.F. Zhu, G.Z. Cui, Z.Q. Deng, Karst in China, Geographical Publishers, Beijing, 1994 (in Chinese). [24] S.Y. Ding, J. Peng, Study on aboveground biomass of Cyclobalanopsis glaucoides sprouted shrub community in the central part of Yunnan Plateau, J. Yunnan. Univ. 14 (1992) 191–197 (in Chinese, with English abstract). [25] J. Peng, S.Y. Ding, B.R. Wang, W.H. Su, Studies on evergreen broadleaved sprouted shrub communities and their biomass in Kunming region, J. Yunnan. Univ. 14 (1992) 167–178 (in Chinese, with English abstract). [26] X.K. Zhu, S.Q. Zhu, Study on asexual reproduction ecology of Itea ilicifolia in karst stone mountain-root morphology and biomass allocation, J. Mount. Agric. Biol. 19 (2000) 6–9 (in Chinese, with English abstract). [27] L.F. Yu, S.Q. Zhu, J.Z. Ye, L.M. Wei, Z.R. Chen, Evaluation on degradation of karst forest community and human disturbance, Chinese J. Appl. Ecol. 13 (2002) 529–532 (in Chinese, with English abstract). [28] S. Nanami, H. Kawaguchi, R. Tateno, C.H. Li, S. Katagiri, Sprouting traits and population structure of co-occurring Castanopsis species in an evergreen broadleaved forest in southern China, Ecol. Res. 19 (2004) 341–348. [29] F.M. Zhang, H. Geng, Y.H. Li, Y.N. Liang, Y.H. Yang, J. Ren, F.C. Wang, H.L. Tao, Z.D. Li, Study on the Lunan Stone Forest Karst, China, Yunnan Science and Technology Press, Kunming, 1997 (in Chinese).
[30] Z.Z. Jin, J. Peng, Vegetation of Kunming, Yunnan Science and Technology Press, Kunming, 1998 (in Chinese). [31] Y.X. Shen, W.Y. Liu, Y.H. Li, J.W. Cui, Community ecology study on karst semihumid evergreen broadleaved forest at the central part of Yunnan, Guihaia 25 (2005) 321–326 (in Chinese, with English abstract). [32] T. Yamada, Y. Kumagawa, E. Suzuki, Adaptive significance of vegetative sprouting for a tropical canopy tree, Scaphium longiflorum (Sterculiaceae), in a peat swamp forest in Central Kalimantan, Ecol. Res. 16 (2001) 641–647. [33] W.C. Su, W.X. Zhu, The eco-environmental fragility in karst mountain regions of Guizhou Province, J. Mount. Sci. 18 (2000) 429–434 (in Chinese, with English abstract). [34] J.I. Querejeta, H. Estrada-Medina, M.F. Allen, J.S. Jiménez-Osornio, Water source partitioning among trees growing on shallow karst soils in a seasonally dry tropical climate, Oecologia 152 (2007) 26–36. [35] J.M. Espelta, J. Retana, A. Habrouk, Resprouting patterns after fire and response to stool cleaning of two coexisting Mediterranean oaks with contrasting leaf habits on two different sites, For. Ecol. Manage. 179 (2003) 01–414. [36] E.F. Nzunda, M.E. Griffiths, M.J. Lawes, Sprouting by remobilization of aboveground resources ensures persistence after disturbance of coastal dune forest trees, Funct. Ecol. 22 (2008) 577–582. [37] L.M. Kruger, J.J. Midgley, R.M. Cowling, Resprouters vs reseeders in South African forest trees: a model based on forest canopy height, Funct. Ecol. 11 (1997) 101–105. [38] L.M. Kruger, J.J. Midgley, The influence of resprouting forest canopy species on richness in Southern Cape forest, South Africa, Glob. Ecol. Biogeogr. 10 (2001) 567–572.