Effect of disturbance on the regeneration of four dominant and ... - IISc

RESEARCH COMMUNICATIONS

Effect of disturbance on the regeneration of four dominant and economically important woody species in a broadleaved subtropical humid forest of Meghalaya, northeast India B. P. Mishra, R. S. Tripathi*, O. P. Tripathi and H. N. Pandey Department of Botany, School of Life Sciences, North-Eastern Hill University, Shillong 793 022, India

Effect of disturbance was studied on the regeneration behaviour of four dominant woody species, viz., Casearia vareca Roxb., Eurya japonica Thunb., Psychotria symplocifolia Kurz. and Rhododendron arboreum Sm., of a sacred forest which represents a subtropical wet hill forest of Meghalaya in northeast India. About one third area of this forest is undisturbed or mildly disturbed (stand I), while two thirds of this area is moderately-to-highly disturbed (stand II). Tree density, light interception and canopy cover were markedly higher in stand I when compared to stand II. E. japonica and R. arboreum are heliophilic as depicted by their greater numbers of seedlings and saplings in stand II. On the contrary, C. vareca and P. symplocifolia showed higher density of seedlings and saplings in stand I which indicates that these two species are sciophilic in nature. In all the species the per cent of saplings that grew into trees was higher than the percentage of seedlings that developed into saplings. Sprouting behaviour of four species differed greatly in two stands. The average number of sprouts per stump was higher in stand II, which could be linked to greater availability of light in this stand. E. japonica showed greater coppicing potential than the other three species. The coppicing in this species was better in stand II as compared to stand I. Analysis of variance showed a significant (P < 0.05) decrease in number of sprouts of E. japonica and R. arboreum from stand II to stand I. SEEDS and sprouts are the important means of tree regeneration in natural forests. Regeneration through seeds depends on production, dispersal and germination of seeds, and establishment and growth of seedlings. Several factors such as resource availability, pollination success, predation of flowers, fruits and leaves, genetic make-up, and age and size of plant affect seed production1. Microsite characteristics of the forest floor and microenvironmental conditions under the forest canopy also influence the regeneration of trees by seeds2. An increased availability of light stimulates germination of seeds in several forest trees and inhibits it in a few3,4. The treefall gaps cause increase in seedling recruitment and establishment of seedlings as well as saplings5. Growth of seedlings is often limited by *For correspondence. (e-mail: [email protected]) CURRENT SCIENCE, VOL. 84, NO. 11, 10 JUNE 2003

the availability of soil nutrients, especially N and P, and the forest soils of this region are poor in these nutrients6. Sprouts are the major contributors to natural regeneration where tree felling is common7,8,9. Sprouting may also be an active strategy to withstand dry environment and mild disturbance which enhances coppicing in trees10. Sprouting behaviour of tree species is highly influenced by girth and height of the stump11–14. The canopy cover determines the light penetration to the sub-canopy layer and forest floor. The low and open canopy as well as the presence of large gaps allow more light to reach the forest floor15,16 which in turn causes increase in temperature as a consequence of which sprouting may be promoted. Forest ecosystems of northeast India have been studied by several workers and considerable amount of data is available on species composition, biodiversity, litter dynamics, forest soil characteristics, and natural regeneration of forest trees. Interactive influence of girth and height of stumps on sprouting behaviour and survival of sprouts of Alnus nepalensis, Quercus dealbata, Quercus griffithii and Schima khasiana13; effect of cultural disturbance on production, dispersal and germination of seeds of Lithocarpus dealbatus, Quercus griffithii and Schima khasiana17; effect of seed weight and microsite characteristics on germination and seedling fitness in two species of Quercus2, and a comparison of the regeneration behaviour of tree species in the undisturbed areas within a sacred forest with that in the disturbed forest stands17 of the area have been studied. However, we came across a sacred forest called Swer sacred grove which comprised of forest patches exposed to varied degrees of disturbance. Thus, this sacred forest provided a good site for studying the effect of disturbance on tree regeneration. In an earlier study it was found that Casearia vareca Roxb., Eurya japonica Thunb., Psychotria symplocifolia Kurz. and Rhododendron arboreum Sm. possess high importance value index (IVI) and are the dominant woody species of the grove. Besides, they are economically important for the villagers as sources of fuelwood, fodder and small timber. Of these, E. japonica is an early successional species, whereas C. vareca and P. symplocifolia are late successional. Rhododendron arboreum, however, grows in both disturbed stand as well as in undisturbed stand representing primary forest. The present paper focuses on the effect of disturbance on natural regeneration of these woody species in the Swer sacred grove which represents the subtropical wet hill forest of Meghalaya, northeast India. The study was conducted in a sacred forest at Swer (25°25′N and 91°47′E, 1990–2035 m asl) in the East-Khasi Hills district of Meghalaya. Swer is about 28 km south of Shillong, the capital state of Meghalaya. This sacred forest is spread over a 40 ha area on a hill called Lum Swer. About 15 ha of the sacred grove is covered by dense forest, and is either undisturbed or mildly disturbed. This part of the forest has been referred to as stand I in this 1449

RESEARCH COMMUNICATIONS paper. The other portion of the grove which covers much larger area (≈ 25 ha) is moderately-to-highly disturbed. The villagers from the nearby area meet their timber and fuelwood requirements from this portion of the grove, and they also graze their cattle here. This portion of the forest has been referred to as stand II. Tree density and canopy cover were much lower in stand II as compared to stand I, as a result of which the intensity of light reaching the forest floor was much higher in the former stand (Table 1). The climate is monsoonic with an average annual rainfall of about 2500 mm. The mean monthly ambient temperature varies between 3°C and 22°C. The soil is leached, poor in nutrients and acidic in nature (pH 5.7). The vegetation of the sacred grove is a broad-leaved subtropical humid forest. The canopy layer is largely composed of evergreen trees (Aporusa dioica, Beilshmiedia assamica, Persea duthiei, Psychotria symplocifolia and Rhododendron arboreum) with some deciduous (Casearia vareca, Engelhardtia spicata and Glochidion khasicum) elements. The sub-canopy is composed of Actinidia spp., Aporusa spp., Casearia vareca, Daphniphyllum himalayense, Erythroxylum kunthianum, Eurya acuminata, Eurya japonica, Glochidion khasicum, Mussaenda spp., Myrica spp., Psychotria symplocifolia, Rhododendron arboreum, Saprosma tetrasperma, Symplocos spp. and Wendlandia peniculata. Casearia vareca, G. khasicum, P. symplocifolia

and R. arboreum were present in both canopy and subcanopy layers. While Rhododendron arboreum was equally important in both the layers, C. vareca and P. symplocifolia were more important in the canopy layer, and E. japonica in the sub-canopy layer. Tree density, canopy cover and light interception were determined in both stands. About one per cent area of each of the two forest stands was sampled between October 1999 and March 2001 for analysing the community attributes, which were determined by following the methods outlined by Misra18 and Mueller-Dombois and Ellenberg19. Density, frequency and basal cover of the trees (gbh > 15 cm), and density of saplings (gbh 5–15 cm) and seedlings (gbh < 5 cm) of four dominant woody species namely, Casearia vareca Roxb. (Flacourtiaceae; flowering and fruiting during March–July), Eurya japonica Thunb. (Theaceae; flowering and fruiting during June– December) and Psychotria symplocifolia Kurz. (Rubiaceae; flowering and fruiting during May–December) and Rhododendron arboreum Sm. (Ericaceae; flowering and fruiting during March–October) were determined by the quadrat method. In an earlier study, these woody species were identified as dominant in this forest on the basis of their high importance values. Thirty quadrats each for trees and shrubs, and 120 quadrats for seedlings were studied in stand I and stand II. The quadrat size was 10 × 10 m for trees, 5 × 5 m for shrubs and 1 × 1 m for seedlings. Density, height and girth of the stumps of these species were also determined in both the stands. For the study of regeneration through sprouts, observations were recorded in respect of the number of sprouts arising from stumps of different girth and height. The stumps were grouped into five girth and five height classes, viz., 0–10, 10–20, 20–30, 30–40 and > 40 cm. The number of sprouts emerging from the individual stump of each girth and height class was counted to study the sprouting behaviour of the above species. E. japonica was common in stand II, and C. vareca and P. symplocifolia in stand I. However, R. arboreum was not much affected by disturbance since it was equally common in both stands. Tree density of E. japonica and R. arboreum was much higher than C. vareca and P. symplocifolia. E. japonica showed higher density in stand II,

Table 1. Canopy cover, light intensity, tree density and number of tree stumps per hectare and disturbance level in the two stands of the Swer sacred forest Forest stands Parameters Canopy cover (%) Light interception (%) Tree density (individuals ha–1) Number of tree stumps (ha–1) *Disturbance index (%)

Stand I

Stand II

> 40 > 50 2103 ± 25 146 ± 12 7

< 40 < 50 1060 ± 38 382 ± 17 27

±, standard error. Number of tree stumps *,

× 100 Total number of trees including the tree stumps

Table 2.

Density, basal area and IVI of four dominant woody species in the two stands of the Swer sacred forest Species C. vareca

E. japonica

P. symplocifolia

R. arboreum

Parameters

Stand I

Stand II

Stand I

Stand II

Stand I

Stand II

Stand I

Stand II

Tree density (individuals ha–1)

48 ± 9

18 ± 3

66 ± 18

108 ± 22

37 ± 8

23 ± 5

157 ± 14

137 ± 12

1.4 8.5

0.4 10.9

1.1 20.1

1.5 38.5

1.0 12.2

0.5 16.8

4.1 24.1

3.7 37.7

Basal area (m2ha–1) IVI ±, standard error. 1450

CURRENT SCIENCE, VOL. 84, NO. 11, 10 JUNE 2003

RESEARCH COMMUNICATIONS while the other three species had greater density values in stand I. R. arboreum was a dominant species in both the stands in terms of basal area. However, IVI value of E. japonica was greater than R. arboreum in stand II. All the four species showed a marked increase in IVI from stand I to stand II; the increase being much greater in case of E. japonica and R. arboreum as compared to the other two species (Table 2). The overall population structure of all the four species was similar in the two stands as they all showed a successive decrease in population density from seedling to adult stage. Density of seedlings and saplings of E. japonica and R. arboreum was higher in stand II as compared to stand I. The tree population of E. japonica was also larger in stand II while the number of trees of R. arboreum was higher in stand I. C. vareca and P. symplocifolia showed higher population density of seedlings, saplings and trees in stand I (Figure 1). The total stump density in stand II (382 ha–1) was about two and a half times greater than in stand I (146 ha–1). E. japonica had maximum number of stumps, followed by R. arboreum, P. symplocifolia and C. vareca in stand II. In stand I, stump density of R. arboreum was maximum among the four species. Further, the number of stumps of all girth classes was more in stand II compared to stand I. E. japonica showed higher stump density than the other three species in all girth classes in stand II, except the girth class > 40 cm, in which number of stumps was maximum in case of R. arboreum. The highest density of stumps of E. japonica was recorded in the girth class 10– 20 cm (Figure 2).

All four species showed high sprout density (number of sprouts per stump) in stand II. Sprouting was more copious in the stumps of intermediate girth classes. Coppicing was maximum in E. japonica where the number of sprouts per stump was highest in the 20–30 cm girth class. All other species also showed maximum sprouting in this girth class (Figure 3). In E. japonica and R. arboreum, the stumps of 20–30 cm height class produced more sprouts than the stumps of other height classes. However, in C. vareca and P. symplocifolia, maximum coppicing was seen in the stump height class 30–40 cm. Of all the four species, E. japonica generally showed maximum sprouting regardless of girth and height of stump (Figure 4). ANOVA showed a significant (P < 0.05) increase in number of sprouts of E. japonica and R. arboreum from stand I to stand II in the stumps of different girth classes. However, the difference in sprouting from the stumps of different height classes was insignificant between the stands in all species. The number of sprouts emerging from the stumps of all four species increased significantly with the girth class (up to 20–30 cm) in stand I. In stand II, however, this trend was observed only in case of C. vareca and P. symplocifolia. The number of sprouts per stump also increased significantly with stump height up to 30–40 cm in C. vareca and P. symplocifolia in both the stands, but in other two species the increase was not significant (Table 3). The sacred grove was composed of a large number of multiple-stemmed trees and shrubby vegetation indicating thereby that trees had been cut in the past. However,

Figure 1. Density of tree (gbh > 15 cm), sapling (gbh 5–15 cm) and seedling (< 5 cm) populations of four woody species in stand I ( ) and stand II ( ) of the Swer sacred forest. Line bar indicates ± SE.

Figure 2. Stump density per hectare of four woody species in stand I ( ) and stand II ( ) of the Swer sacred forest.


1451

RESEARCH COMMUNICATIONS tree cutting was not uniform throughout the grove, cut trees were less frequent in comparison to certain portions such as in stand I than stand II. This is evident from the lower (7%) disturbance index (percent of tree stumps to the total number of trees including the tree stumps in the

Figure 3. Effect of stump girth on coppicing behaviour (average number of sprouts per stump) of four woody species in stand I ( ) and stand II ( ) of the Swer sacred forest. Line bar indicates ± SE.

stand) in stand I than in stand II (27%). R. arboreum was predominant in both the stands, while E. japonica, a secondary successional species, was common in stand II. C. vareca and P. symplocifolia were abundant in stand I which was undisturbed/mildly disturbed. The seedlings and sprouts play a vital role in natural regeneration of woody species and their densities are largely dependent upon the prevailing micro-environment in the forest13. The change in population structure of these woody species from stand I to stand II may be attributed to the variation in microclimatic conditions such as temperature, humidity and light intensity caused by disturbance. The presence of large number of seedlings of P. symplocifolia, and their better growth under close canopy in stand I suggest shade-tolerant nature (sciophilic) of this species. Presence of larger numbers of seedlings of E. japonica and R. arboreum in stand II where the light intensity was high, indicated that light stimulates seed germination of these two light-demanding (heliophilic) species. The poor survival and growth of seedlings and saplings of all four species in stand II reveals that the increased level of disturbance adversely affected them at the juvenile stage. In stand I, 4–28% seedlings of these species grew into saplings, whereas in stand II the percentage did not exceed 16. However, percentage of saplings that grew into trees ranged between 18 and 64 in stand I, and between 26 and 49 in stand II (Figure 5). The presence of large number of seedlings and smaller sapling population of E. japonica in stand II (Figure 1) was mainly due to the removal of saplings by the villagers for meeting their requirement of fuelwood. In both stands, few seedlings of P. symplocifolia attained the sapling stage. In stand I this may be attributed to the arrested growth of seedlings due to low light intensity because of close canopy while in-

Table 3.

Figure 4. Effect of stump height on coppicing behaviour (average number of sprouts per stump) of four woody species in stand I ( ) and stand II ( ) of the Swer sacred forest. Line bar indicates ± SE. 1452

Analysis of variance of the data presented in Figures 3 and 4 Figure 3

Figure 4

Sprouting behaviour

F value

F value

a. Stand I × stand II C. vareca E. japonica P. symplocifolia R. arboreum

1.045** 15.923* 4.356** 9.424*

b. Stump girth/height class × stands C. vareca Stand I Stand II

10.76* 9.257*

E. japonica

Stand I Stand II

P. symplocifolia

Stand I Stand II

R. arboreum

Stand I Stand II

5.652* 0.961** 10.042* 7.696* 9.094* 4.436**

3.434** 1.592** 1.002** 3.756** 11.524* 9.362* 3.653** 1.71** 11.752* 9.293* 4.83** 2.237**

*, Significant and **, not significant at P < 0.05. CURRENT SCIENCE, VOL. 84, NO. 11, 10 JUNE 2003

RESEARCH COMMUNICATIONS a

b

sprouting11,22,23. Rapid decay of the low-cut stumps may reduce sprouting23. This is particularly true in the northeast India where microbial decomposition is fast due to high humidity and favourable temperature conditions. 1. 2. 3. 4. 5. 6. 7.

Figure 5. Regeneration efficiency of four woody species in terms of per cent conversion of (a) seedlings into saplings and (b) saplings into trees in stand I ( ) and stand II ( ) of the Swer sacred forest. 8.

creased grazing pressure may have been a contributory factor in stand II. The greater light intensity in stand II favoured sprouting in E. japonica and R. arboreum as is evident from commonly occurring multiple-stemmed trees in this stand. This suggests that disturbance in the form of clear felling of trees plays a positive role in regeneration of forest trees through coppicing, which is in agreement with a study in the Appalachian forest20. The regeneration behaviour of the four tree species under investigation varied between stand I (undisturbed/mildly disturbed stand) and stand II (moderately-to-highly disturbed stand). In C. vareca and P. symplocifolia, regeneration through seeds was better in stand I, while their coppicing was better in stand II. In case of E. japonica and R. arboreum, however, increased level of disturbance favoured regeneration both by seeds as well as by coppice. Thus the regeneration behaviour of forest trees seems to be closely linked with the level of disturbance. However, in order to ascertain the exact relationship between the level of disturbance in the forest ecosystem and regeneration of the various component tree species, a detailed investigation would be required. In both stands, the stumps of intermediate girth and height classes produced maximum number of sprouts, which may be related to change in tree physiology with age14. Reduced sprouting from the thick stumps indicates arrest of vegetative propagation in old trees. In fact, vegetative propagation predominates in the juvenile phase, whilst sexual reproduction in the adult phase21. With the increase in tree age, the number of dormant buds which subsequently give rise to sprouts is reduced due to their death. The low population of dormant or trace buds in the tree stumps having low girth and height causes decreased


9.

10.

11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23.

Winn, A. A. and Werener, P. A., Ecology, 1987, 68, 1224–1233. Tripathi, R. S. and Khan, M. L., Oikos, 1990, 57, 289–296. Plummer, J. A. and Bell, D. T., Aust. J. Bot., 1995, 43, 93–102. Bell, D. T., Rochin, D. P., McChesney, C. J. and Plummer, J. A., Science, 1995, 6, 797–806. Hubbel, S. P. et al., Science, 1999, 283, 554–557. Arunachalam, A., John, B., Maithani, K., Pandey, H. N. and Tripathi, R. S., Oecology, 1997, 16, 253–264. Beck, D. E., Proc. 10th Forestry and Wildlife Forum: Regenerating the Southern Forest (eds McElwee, R. L. and Bell, N. H.), Virginia Polytechnic Institute and State University, Blacksburg, VA, 1980, pp. 132–142. Kellison, R. C., Federick, D. J. and Gardner, W. E., Proc. 10th Forestry and Wildlife Forum: Regenerating the southern forest (eds McElwee, R. L. and Bell, N. H.), Virginia Polytechnic Institute and State University, Blacksburg, VA, 1980, pp. 49–61. Smith, H. C., Regenerating Oaks in Upland Hardwood (eds Holt, H. A. and Fischer, B. C.), Purdue University, Lafayette, IN, 1979, pp. 30–41. Rundel, P. W., Response of Plants to Multiple Stress (eds Mooney, H. A., Winner, W. E. and Pell, E. J.), Academic Press, New York, 1991, pp. 345–370. Harrington, C. A., Can. J. For. Res., 1984, 14, 357–361. Jornson, P. S., For. Sci., 21, 413–418. Khan, M. L. and Tripathi, R. S., For. Ecol. Manage., 1986, 17, 199–209. MacDonald, J. E. and Powell, G. R., Can. J. For. Res., 1993, 13, 390–394. Dunphy, K. K., Murphy, P. G. and Lugo, A. E., Trop. Ecol., 2000, 41, 161–167. Rao, P., Barik, S. K., Pandey, H. N. and Tripathi, R. S., Aust. J. Ecol., 1997, 22, 136–145. Barik, S. K., Tripathi, R. S., Pandey, H. N. and Rao, P., J. Appl. Ecol., 1996, 33, 1551–1560. Misra, R., Ecology Work Book, Oxford Publishing Company, Calcutta, 1968. Mueller-Dombois, D. and Ellenberg, H., Aims and Methods of Vegetation Ecology, John Wiley and Sons, USA, 1974. Boring, L. R., Monk, C. D. and Swank, W. T., Ecology, 1981, 62, 1244–1253. Hartmann, H. T. and Kester, D. E., Plant Propagation: Principles and Practices, Prentice-Hall, Englewood Cliffs, NJ, 1975, p. 559. Belanger, R. P., South J. Appl. For., 1976, 3, 101–103. El Houri Ahmed, A., Sudan Silva, 1977, 3, 90–105.

ACKNOWLEDGEMENTS. The facilities provided by the Head, Department of Botany, North-Eastern Hill University, Shillong are thankfully acknowledged. This study was supported by the DST, New Delhi in form of a research project under the Scheme for Young Scientists and the University Grants Commission, New Delhi through Special Assistance Programme in Botany at NEHU, Shillong. Received 24 December 2002; revised accepted 29 March 2003

1453