Applied Vegetation Science 12: 211–224, 2009 & 2009 International Association for Vegetation Science
211
Recovery of sandy beach and maritime forest vegetation on Phuket Island (Thailand) after the major Indian Ocean tsunami of 2004 Hayasaka, D.1,2; Fujiwara, K.1,3 & Box, E.O.4,5 1
Graduate School of Environment and Information Sciences, Yokohama National University, Hodogaya-ku, Tokiwadai 79-7, Yokohama 240-8501, Japan; 2 Current address: NIPPON KOEI Co., Ltd., Hakata-ku, Higashi Hie 1-2-12, Fukuoka 812-0007, Japan; 3 E-mail
[email protected]; 4 Department of Geography, University of Georgia, Athens, GA 30602-2502, USA; 5 E-mail
[email protected]; Corresponding author; Fax181 924 754 330; E-mail
[email protected]
Abstract Question: How rapidly has the sandy beach and maritime forest vegetation on Phuket recovered and regenerated after the impact of the major Indian Ocean tsunami of 2004? What are the characteristics of sandy beach species for regenerating their populations and the invasion patterns of originally non-sandy beach species or other newcomers after the tsunami? Location: Phuket Island, southern Thailand. Methods: Species composition of beaches was studied on the same research plots 6 months before and 9 months after the tsunami. The changes in individual species cover before and after the tsunami were determined by w2 tests. Change in community composition was analysed by detrended correspondence analysis. The relationship between species and environmental factors was analysed by canonical correspondence analysis. Results: The sites disturbed by the tsunami were often invaded by annuals, especially grasses and asteraceous plants, rather than by perennials. In contrast, species with clonal growth by stolons decreased significantly. Factors determining the species habitat differences were soil hardness (penetration resistance of sandy soil), per cent silt content, soil water content and beach management. Habitat differences among originally non-sandy beach herbaceous species that expanded their population or moved to the coast after the disaster were defined by sand accretion or erosion caused by the tsunami. Many sandy beach herbaceous communities changed into Dactyloctenium aegyptium communities because of the tsunami were originally constituted by non-sandy beach D. aegyptium with Cenchrus echinatus. Although the forest floors of most maritime forests were invaded by originally non-sandy beach Tridax procumbens, Eleusine indica or D. aegyptium because of the tsunami, this did not result in a change in the vegetation unit, because species’ loss was restricted to the understorey. In time, these forests will recover their previous community composition.
Conclusions: Our results suggest that originally non-sandy beach native species invaded the disturbed beaches rapidly after the tsunami but their habitats differ. Sites where sand accumulated on a beach because of the tsunami were invaded by D. aegyptium and E. indica, whereas soil erosion permitted invasion by Digitania adscendens. Tridax procumbens establishes rapidly on wet sites with hard soil, high per cent silt content and low beach management pressure. Sandy beach species with subterranean long rhizomes are strongly tolerant of such disasters. We concluded that the species composition of the beaches disturbed by a temporary large disaster is determined by dormancy and growth forms, with radicoid form being influential. Keywords: Habitat segregation; Invasive species; Lifeform strategy; Recolonization; Vegetation dynamic. Nomenclature: Miyawaki et al. (1994) & Simitinand (2001)
Introduction Sites such as seacoasts, sand dunes, sandy beach, mangroves, swamps, salt marshes and estuaries, and the organisms living in these habitats, are sensitive to disturbances such as wave action, droughts, tsunamis, typhoons, other storms, and consequent erosion events. There have been many studies on the relationship between coastal vegetation dynamics and natural, mostly storm, disturbances (Stoneburner 1978; Carter 1980; Gardner et al. 1992; Hayden et al. 1995; Conner 1998; Nordstrom et al. 2002; Conner et al. 2005). Although such areas provide resources, biodiversity, and habitats as well as services to humans, we have disturbed them severely for a long time. Ironically, the great human tragedy caused by the Indian Ocean tsunami of December 26 2004 may provide the stimulus for a better understanding on
212
DAISUKE, H. ET AL.
the recovery and regeneration processes of sandy beach and maritime forest vegetation after the disaster. Mazda et al. (1997) reported that reduction of wave height depends on the water depth, wave period, wave height and, if present, the species of mangrove trees, density of mangrove forest and the diameter of mangrove roots and trunks. Owing to the relative infrequency in the tropics of earthquakes above magnitude 8.0 (which have occurred only three times in recorded history: 1797, 1833 and 1861) there are few scientific data or reports on recovery and regeneration of sandy beach and maritime forest vegetation after such tsunami. The Indian Ocean tsunami ‘‘attacked’’ Phuket Island, Thailand, with waves traveling at about 200 km h–1 generated by the massive undersea earthquake, measuring 9.3 on the Richter scale. This tsunami gave an opportunity to identify how sandy beach and maritime forest vegetation reacted, to observe recovery and regeneration processes that might improve future coastal risk assessment (Bush et al. 1996), and to evaluate how coastal vegetation plays a role in protection against severe tsunami disturbance (Dahdouh-Guebas et al. 2005; Kathiresan & Rajendran 2005). Dawson (1994) reported that coastal landscapes may be greatly altered not only by direct tsunami overwash orthogonal to the shoreline but also by episodes of vigorous backwash and by currents along the coastline. He suggested that the entire physical structure of coastal vegetation was changed drastically by this catastrophic event. The original surface soil on the beach was removed or buried by the tsunami. In this paper we describe how rapidly sandy beach vegetation on Phuket recovered and how rapidly species regenerated their populations after the tsunami. The goal of the study is to clarify the indigenous sandy beach species’ ecological characteristics for recovery and regeneration after disturbance and the invasion patterns of originally non-sandy beach species or other newcomers in response to disturbance.
Materials and Methods Study site The study was carried out on Phuket Island, a near-shore island in the Andaman Sea, located at 8107 0 -7153 0 N latitude and 981190 -98124 0 E longitude. The characteristics of the four study beaches (Nai Yan, Mai Khao, Kamala and Karon Beach) are shown in Table 1. The climatic conditions at two nearby meteorological stations (Phuket airport and Phuket town) are very similar and represent a tropical monsoon climate. Mean annual temperature of the two stations is 27.41C and 28.11C, mean annual relative humidity is 75% and 80%, mean annual rainfall is 2504 and 2317 mm, and Kira’s (1977) Warmth Index (WI) is 268.7 and 277.6, respectively. The dry season is from October to April, and the wet season is from May to September. Maximum wave height during the recent Indian Ocean tsunami was 4.07 m at Nai Yan and Mai Khao Beach, 5.29 m at Kamala Beach and 4.49 m at Karon Beach (http://www.drs.dpri. kyoto-u.ac.jp/sumatra/thailand/phuket_survey.html). Species’ composition and species’ characteristics Species composition at the four study beaches was first surveyed and recorded by phytosociological releve´s (full floristic composition, separate sampling of tree, shrub and herbaceous layers) in spatially homogeneous vegetation (Braun-Blanquet 1964) by Hayasaka & Fujiwara (2005) in June 2004, 6 months before the tsunami. At that time we recorded the latitude and longitude of the research quadrats using a GPS (eTrex Legend; GARMIN Internationnal Inc., Olathe, KS, USA). Those same beaches were resurveyed in September 2005, 9 months after the tsunami, using exactly the same quadrats and quadrat size as Hayasaka & Fujiwara (2005) (Fig. 1). The number of field releve´s at the four study beaches before the tsunami was 126, ranging in size from 9 m2 (herbaceous communities)
Table 1. Beach characteristics of the study sites. Beach
Beach segment
Portions of topography
Access
Relative number of visitors
Beach structures in the sand dunes
Beach management per about every second or 3 days
Nai Yan
National park
Poor visitor access
Very few visitors
None
Garbage cleaning by manpower
Mai Khao
National park
Poor visitor access
Very few visitors
Seats
Garbage cleaning by manpower
Karon
Resort
Easy access
Not many visitors
Beach houses
Kamala
Resort
Flattish sandy beach Wave-cut sand terrace Flattish sandy beach Wave-cut sand terrace
Easy access
Many visitors
Beach houses and chairs
Garbage cleaning and weeding by manpower and machines Garbage cleaning and weeding by manpower and machines
- SPECIES ECOLOGICAL TRAITS CONTRIBUTING TO EFFECTIVE REGENERATION AND INVASION AFTER TSUNAMI - 213
Fig. 1. Landscape change of Mai Khao Beach from the same location. (a) Before the tsunami in June 2004, the abundant sandy beach herbaceous vegetation was of Vigno-Ipomoeetum pedis-caprae (vege5). (b) in September 2005, after the tsunami, most of sandy beach herbaceous vegetation was removed or buried, and maritime forest species were defoliated by the tsunami.
to 100 m2 (woody communities). The number of field releve´s after the tsunami was 127 (adding one that was investigated after the disturbance where Spinifex littoreus had colonized a site that had no vegetation before the tsunami). To assess the change in species composition after the tsunami, the new releve´ data were compared with the data from 2004. The sandy beach herbaceous and maritime forest vegetation was then classified by analytical tablework, based on Ellenberg (1956) and MuellerDombois & Ellenberg (1974). Vegetation type numbers correspond to the vegetation zonation patterns from the shoreline inland (see App. 1), with small numbers representing the vegetation on the foredunes and larger numbers that further inland. Environmental data were collected for all research quadrats: % silt content (o0.02 mm in grain diameter), % soil water content, soil hardness (kg cm 2), distance from the shoreline based on the high tide water mark (m), micro-topography (flat, slope, concave and convex), the amount of soil erosion or accretion (cm), beach management pressure (Table 1) and the amount of garbage (%). We think that soil hardness is associated with distance from the shoreline and with the degree of root penetration or soil water supply. A soil water tester (DM-18; Takeyama Electric Works, Co., Ltd, Tokyo, Japan) was used to measure surface soil water content. Soil hardness based on the penetration resistance of surface soil was measured by a soil hardness tester (Yamaoka System Hardness Tester; Fujwara Scientific Company, Japan). The amount of soil erosion or accretion after the tsunami was estimated from the high-tide water mark by measuring the change in elevation with respect to mean sea level of the study quadrats using a transit compass (Tracon LS-25; Ushikata Mfg., Co., Ltd, Tokyo, Japan). Beach management pressure was classified as follows: (1) no beach cleaning;
(2) beach cleaning by manpower; (3) weeding by machines; and (4) beach cleaning and weeding by manpower and machines. The amount of garbage on research quadrats was classified as follows: (1) under 5%; (2) 5-25%; (3) 25-50%; (4) 50-75%; and (5) above 75%. We differentiated indigenous sandy beach species from originally non-sandy beach species by the primary habitat of the species, based on field observation and published descriptions of the ranges of wild plants in Japan (Hatsusima 1975; Miyawaki et al. 1994) and Thailand (Simitinand 2001). The life-forms of the species were categorized by the dormancy forms, radicoid (subterranean organ) forms, growth forms and disseminule forms described by Raunkiear (1934) and Numata (1947, 1990). We thought that these eco-morphological traits might be key factors related to the initial recovery process of sandy beach species after the disturbance. Statistical analysis The change in individual species cover after the tsunami was analyzed by a Pearson’s w2 test, in order to show the differences between expected and observed species cover before and after the disaster. Data analysis was conducted using SPSS statistical software (ver. 11.0J; SPSS Japan, Tokyo, Japan). Changes in species composition after the tsunami, based on species cover in the same study plots in 2004 and 2005, were analysed by detrended correspondence analysis (DCA; Hill & Gauch 1980). We presumed that the sand-dune herbaceous vegetation would show large changes caused by the tsunami, while maritime forest communities would show small changes. Therefore we analysed the two vegetation types (sandy beach herbaceous and maritime forest) separately, based on clear differences in habitat and dormancy types between these two general
214
DAISUKE, H. ET AL.
vegetation classes. The dormant buds of many herbaceous species are near the surface, while those of forest trees are well above ground. The relationship between plant habitat and environmental factors was analysed by canonical correspondence analysis (CCA; ter Braak 1995) using PC-ORD statistical software (ver. 4.0; MjM Software Design, Gleneden Beach, OR, USA). The dependent variable was individual species cover on the plots before and after the tsunami, and the independent variables were the uncorrelated environmental measurements of all plots, including per cent silt content, soil water content, soil hardness, microtopography, the amount of soil erosion or accretion, beach management pressure and the amount of garbage.
Results Species sensitivity to the disturbance The total number of species at the Phuket study sites was 72, of which all were native and 36 (50%) occur widely in tropical to subtropical regions (Hatsusima 1975; Miyawaki et al. 1994; Simitinand 2001); 23 (32%) and are widespread in tropical to temperate or boreal regions (Takematsu & Ichizen 1987-1997; Miyawaki et al. 1994); 13 (18%) occur only in tropical regions (Simitinand 2001). Eighteen species (25%) showed a significant difference in frequency of occurrence before and after the tsunami, as shown by the w2 tests (Table 2). D. aegyptium, T. procumbens, E. indica, D. adscendens, C. echinatus, Sida acuta, Zoysia matrella and Ipomoea pes-caprae significantly increased the proportion of plots they occupied after the tsunami (Table 2). Zoysia matrella and I. pes-caprae are indigenous perennial sandy beach species with either geophytic or wideranging rhizomatous growth. The other species, except for S. acuta, are annual originally non-sandy beach species, including grasses and asteraceous plants dispersed by wind. Conversely, the cover of Ipomoea gracilis, Lepturus repens, Ischaemum muticum, Thuarea involuta, Remirea maritima, Alysicarpus vaginaris, Kyllinga nemoralis, Vernonia cinerea, Mimosa pudica and Eragrostis multicaulis decreased significantly after the tsunami (Table 2). Indigenous sandy beach species L. repens, R. maritima and T. involuta are perennial species with stoloniferous growth. Ischaemum muticum and I. gracilis are perennials with non-clonal growth. There was no significant change in the proportion of plots occupied by maritime forest tree species (which were 0.5-10 m high) after the tsunami. Trees
and their organs (branches and leaves) under 2.5 m were defoliated or died back, but branches recovered to produce new leaves. Physical damaged or injured parts of the trees coincided with the wave height during the tsunami. The relationship between species life-form patterns and tsunami disturbance The number of species decreased at all study sites after the tsunami. A clear change in life-forms composition on the same beach before and after the disturbance was found (Table 3). There were significant changes in dormancy and growth-form composition after the tsunami at the Mai Khao and Kamala beaches, but none for radicoid form composition (Table 3). Although change in R1 or R2 species composition of the four study beaches as a result of the tsunami was not found, a decrease in R5 species was observed after the tsunami. Mai Khao was in the most natural condition of the four study beaches, whereas Kamala was under the highest pressures due to human activity (Table 1). Community composition change after disturbance The DCA shows the change in species composition of the sandy beach herbaceous and maritime forest communities, on the same beach, after the tsunami (Figs 2 and 3). Eigenvalues for axes 1 and 2 for the herbaceous vegetation were 0.973 and 0.852, respectively, while those for maritime forest vegetation were 0.934 and 0.570. On Nai Yan beach after the major tsunami, all of vege3 (R. maritima and Hydrophylax maritima), vege5 (I. pes-caprae, Canavalia lineata and Vigna marina) and vege6a changed into vege4, constituted by originally non-sandy beach D. aegyptium with Cenchrus echinatus. However, vege7 (Wedelia biflora) did not change as a result of the disturbance (Fig. 2, App. 1). Similarly, on Mai Khao beach, all of vege3, vege6a,c, and most plots of vege5 changed into vege4, but vege7 did not change as a result of the tsunami (Fig. 2, App. 1). Although vege6b (Cyperus stoloniferus and E. multicaulis with I. muticum) did not change to a different vegetation unit, vege6a disappeared from Kamala beach after the tsunami. Vege1 (Z. matrella) newly appeared near the shoreline after the disturbance (Fig. 2, App. 1). On Karon Beach, all of vege3 and vege5 changed into vege4, and vege6a changed into vege6b after the tsunami, and vege2 (S. littoreus) appeared (Fig. 2, App. 1). Vege3 occurred under the influence of natural disturbance such as over-wash and back-wash, and
- SPECIES ECOLOGICAL TRAITS CONTRIBUTING TO EFFECTIVE REGENERATION AND INVASION AFTER TSUNAMI - 215 Table 2. Response to the Indian Ocean tsunami based on species cover using w2 test (oindicates originally non-sandy beach species) in the 127 plots ( indicates statistically significant). Abbreviation of life-forms is as follows. Dormancy forms were classified as Th (Therophyte), Ch (Chamaephyte), H (Hemicryptophyte), G (Geophyte), N (Nanophanerophyte), M (Microphanerophyte) and MM (Mesophanerophyte). Radicoid forms were classified as R1 (wide rhizomatous growth), R2 (moderate), R3 (narrow), R4 (clonal growth by stolons), R5 (non-clonal growth: monophyte), and E (epiphyte and parasite). Growth forms were classified as e (erect), b (branched), t (tussock), l (climbing or liana), p (procumbent), and r (rosette). Disseminule forms were defined as D1 (disseminated widely by wind and water), D2 (attaching to or eaten by animals and man), D3 (disseminated by mechanical dehiscence of fruits), D4 (having no special modification for dissemination), and D5 (not producing seeds as an asexual plants). Species name
oAchyranthes bidentata (Ach.b) oAlocasia macrorrhiza (Alo.m) oAlysicarpus vaginalis (Aly.v) oBorreria laevis (Bor.l) Calophyllum inophyllum (Calop.i) oCalotropis gigantea (Calot.g) Canavalia lineata (Can.l) Cassytha filiformis (Cas.f) Casuarina equisetifolia (Cas.e) oCenchrus echinatus (Cenc.e) oCentotheca lappacea (Cent.l) oCoccinea indica (Coc.i) Clerodendron inerme (Cle.i) Crinum asiaticum (Cri.a) Cyperus polystachyos (Cyp.p) Cyperus stoloniferus (Cyp.s) oDactyloctenium aegyptium (Dac.a) oDesmodium triflorum (Des.t) oDigitaria adscendens (Dig.a) oEleusine indica (Ele.i) oEmilia sonchifolia (Emi.s) oEragrostis multicaulis (Era.m) oEupatorium odoratum (Eupa.o) Euphorbia atoto (Euph.a) oEuphorbia geniculata (Euph.g) oEuphorbia hirta (Euph.h) Fimbristylis sericea (Fim.s) Gloriosa superba (Glo.s) oGymnopetalum integrifolium (Gym.i) oHewittia sublobata (Hew.s) Hibiscus tiliaceus (Hib.t) Hydrophylax maritima (Hyd.m) Ipomoea gracilis (Ipo.g) Ipomoea pes-caprae (Ipo.p) Ipomoea stolonifera (Ipo.s) Ischaemum muticum (Isc.m) oKyllinga nemoralis (Kyl.n) oLantana camara var. aculeata (Lan.c) Lepturus repens (Lep.r) oLonicera japonica (Lon.j) oMelinis repens (Mel.r) oMimosa pudica (Mim.p) oMomordica charantia (Mom.c) oMukia maderaspatana (Muk.m) oOplismenus compositus (Opl.c) oOxystelma esculentum (Oxy.e) oPaederia scandens (Pae.s) Pandanus odoratissimus (Pan.o) oPaspalum conjugatum (Pasp.c) oPaspalum dilatatum (Pasp.d) oPassiflora laurifolia (Pass.l) oPhyllanthus simplex (Phy.s) Pongamia pinnata (Pon.p) Premna integrifolia (Pre.i) Remirea maritima (Rem.m) oRuellia tuberosa (Rue.t) Scaevola sericea (Sca.s)
Occurrence frequency
Significance
Dormancy
Radicoid
Growth
Disseminule
Before
df 5 1
Forms
Forms
Forms
Forms
n.s. n.s.
H G H Th MM M G Ch MM Th H G N H Th H Th H Th Th Th Th H H N Th H G G G M G G G G H H N H N Th Th Th Th Ch G N M Ch H Th Th MM M H G M
R5 R3 R4 R5 R5 R5 R1 E R5 R4 R5 R5 R5 R5 R5 R2 R4 R3 R5 R5 R5 R5 R3 R5 R5 R5 R2 R3 R5 R2 R5 R4 R5 R1 R1 R5 R3 R5 R4 R3 R5 R5 R5 R5 R2 R5 R4 R5 R3 R3 R5 R5 R5 R5 R4 R5 R5
e r b b e e l l e p t l e e t e e p t t b t e b e b t e l l b p l l l p t e p l t b l l p l l b t t l e e e p b b
D2 D2 D5 D1, 3 D1, 4 D1, 4 D1, 4 D1 D1, 4 D1, 2 D1, 2 D2 D1, 4 D1, 4 D1, 2 D1, 2 D1, 2 D2 D1, 2 D1, 2 D1, 2 D1, 2 D1, 2 D1, 4 D1, 2 D1, 2 D1, 4 D1, 4 D4 D5 D1, 4 D1, 4 D1, 2 D1, 4 D1, 4 D1, 2 D1, 2 D1, 2 D1, 2 D2 D1, 2 D1, 2 D2 D2 D2
2 1 8 1 1 3 71 20 12 9 6 2 8 3 5 18 18 2 6 1 1 8 1 19 1 4 4 1 2 1 6 33 18 107 3 52 4 7 19 2 3 8 1 5 3 1 1 16 2 4 3 10 3 14 26 1 24
After 1 1 7 1 1 3 4 8 1 17 5 1 3 0 2 7 63 2 18 28 5 5 1 2 1 3 3 2 1 1 0 2 18 14 2 19 4 2 18 2 3 8 1 0 3 0 1 1 2 2 3 4 1 3 10 1 4
n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
n.s.
n.s.
n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
n.s.
n.s.
n.s. n.s.
n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
n.s. n.s.
D3 D1, 4 D2 D1, 2 D2 D1, 2 D1, 4 D1, 4 D1, 4 D3 D1, 4
216
DAISUKE, H. ET AL.
Table 2. (Continued). Species name
oSida acuta (Sid.a) oSonchus oleraceus (Son.o) oSpermacoce laevis (Spe.l) Spinifex littoreus (Spi.l) oStachytarpheta indica (Sta.i) Terminalia catappa (Ter.c) Thuarea involuta (Thu.i) oTridax procumbens (Tri.p) oVernonia cinerea (Ver.c) Vigna marina (Vig.m) oVinca rosea (Vin.r) Vitex trifolia (Vit.t) Wedelia biflora (Wed.b) oZoysia japonica (Zoy.j) Zoysia matrella (Zoy.m)
Occurrence frequency
Significance
Dormancy
Radicoid
Growth
Disseminule
Before
df 5 1
Forms
Forms
Forms
Forms
H Th Th H H MM H H Th G Th N H G G
R5 R5 R5 R4 R5 R5 R4 R4 R5 R1 R5 R5 R5 R1 R1
e pr e p b e p b b l b e l t t
D1, 2 D1 D1, 2 D1 D1, 2 D1, 4 D1, 2 D1, 2 D1, 2 D1, 4 D1, 2 D1, 4 D1 D1, 2 D1, 2
3 1 1 0 7 15 18 12 11 6 6 3 32 1 5
After 9 1 1 2 2 6 9 24 10 0 3 1 12 1 32
consequent sand movement. Vege5 widely appeared from near the shoreline to around residential areas. Vege4 originally occurred in anthropogenic areas such as in residential area, vacant land and wasteland. Vege6 established itself under the influence of anthropogenic pressure such as waste dumping on the beach. The original habitat of vege2 is similar to that of vege6. The extent of species composition change of vege7 on Nai Yan was different from Mai Khao, perhaps because the Nai Yan beach was narrower than on Mai Khao. Vege7 appeared in ecotonal zone located between sandy beach herbaceous and maritime forest vegetation zones. Although not all the assigned maritime forest communities changed to different vegetation units because of the tsunami, the forest floor of these communities was invaded by originally non-sandy beach T. procumbens, E. indica or D. aegyptium (Fig. 3, App. 2). In Mai Khao beach, most understory species, including shrubs of vege8a-1, died back or were washed away by the tsunami and produced new leaves. The change of species composition of the same plots at Nai Yan as a result of the tsunami was larger than on the other beaches.
The relationship between species distribution and environmental factors The relationship of species to abiotic factors is shown by a CCA (Fig. 4). Eigenvalues for axes 1 and 2 were 0.611 and 0.335. Soil hardness (r 5 0.764, Po0.01), per cent silt content (r 5 0.754, Po0.01), and soil water content (r 5 0.550, Po0.01) showed a negative relationship with axis 1. Change in eleva-
n.s. n.s. n.s. n.s. n.s.
n.s. n.s. n.s. n.s. n.s.
Table 3. Change in life-forms composition (dormancy, radicoid, growth forms) on the same study beach after the tsunami using Mann–Whitney’s U-test. Significances: ,po0.05, , po0.01. Standard nomenclature of lifeforms is shown in Table 2. For the timing of tsunami, B and A indicate before and after the tsunami, respectively. Beach name
Nai Yan
Mai Khao
Kamala
Karon
Timing of tsunami Number of species
B 34
B 25
B 50
B 21
Dormancy form Th Ch H G N M MM Significance
9 1 12 7 0 0 5 n.s.
Radicoid form R1 R2 R3 R4 R5 E Significance Growth form e b t l p r Significance
A 29 2 1 0 2 0 0 0
A 18
3 2 9 5 1 0 5
2 1 2 1 0 0 1
2 0 1 0 2 0 6 7 22 17 1 0 n.s.
1 0 2 7 14 1 n.s.
7 1 7 1 4 0 9 4 6 0 1 1 n.s.
6 6 2 6 5 0 n.s.
A 32
A 19
5 1 4 2 4 0 2
5 2 1 0 7 2 4 1 1 1 0 0 3 0 n.s.
1 0 2 3 3 0
3 1 4 2 4 3 4 2 34 10 1 0 n.s.
1 0 1 1 0 0 8 1 10 2 1 0 n.s.
3 2 0 1 1 0
15 12 10 10 2 1
4 4 5 3 1 1
3 0 5 1 3 1 5 1 5 1 0 0 n.s.
15 2 12 7 6 0 8
tion caused by soil erosion or accretion after the disturbance (r 5 0.347, Po0.01) and beach management pressure (r 5 0.421, Po0.01) had a positive
- SPECIES ECOLOGICAL TRAITS CONTRIBUTING TO EFFECTIVE REGENERATION AND INVASION AFTER TSUNAMI - 217
Fig. 2. Detrended correspondence analysis (DCA) ordination of stands of sandy beach herbaceous vegetation on the same study beach before and after the tsunami. The spatial separation of the plots before and after the tsunami shows the extent of compositional change caused by the disturbance. The points at the ends of each line represent the same permanently marked plots that were sampled both before and after the disturbance. Length and direction of arrow shows the extent of species composition change after the tsunami. The convergence of the same permanently marked plots indicates that the cover of non-sandy beach species increases due to tsunami. Conversely, the divergence shows that species composition changes after the disturbance. Numbers show the pre- and post-tsunami sandy beach and maritime forest vegetation. Vegetation names of the numbers represented in this figure are given in App. 1.
relationship with axis 1, but beach management pressure had a negative relationship (r 5 0.546, Po0.01) with axis 2. The amount of garbage had no relationship with either axis. Canavalia lineata, Cassytha filiformis, Euphorbia atoto, H. maritima, I. pes-caprae, R. maritima, C. echinatus and V. marina were associated with low amounts of dry silt, with low soil hardness, and with low beach management pressure. All woody species, plus L. repens, Stachytarpheta indica, Phyllanthus simplex, V. cinerea and T. procumbens, were related to a high percentage of wet silt, to soil hardness (penetration resistance of sandy soil), and to low beach management pressure. Dactyloctenium aegyptium, E. indica, S. acuta, Z. matrella and C. stoloniferus were found on sites that had had sand added during the tsunami. Digitaria adscendens, E. atoto, T. involuta and W. biflora were found to have invaded sites that had been eroded by the tsunami (Fig. 4).
Discussion Vegetation response after the tsunami This is the first report on the recovery and regeneration of sandy beach vegetation after the Indian Ocean tsunami of 2004. Pre-tsunami vegetation structures are an important data to clarify the recovery and regeneration processes of the post-disturbance vegetation. Sites containing predisturbance vegetation types vege3, and vege6a and vege6c changed to type vege4, dominated by originally non-sandy beach D. aegyptium. This species is broadly a native to the tropics. Only three community types (vege1, vege2 and vege4) persisted through the disturbance (see App. 2). Many component species of the communities appearing after the tsunami, except for vege4, could disperse by ocean drift (thalassochory). Spinifex littoreus, which
218
DAISUKE, H. ET AL.
Fig. 3. Detrended correspondence analysis (DCA) ordination of stands of maritime forest vegetation on the same study beach before and after the tsunami. The spatial separation of the plots before and after the tsunami shows the extent of compositional change caused by the disturbance. The points at the ends of each line represent the same permanently marked plots that were sampled both before and after the disturbance. Length and direction of arrow shows the extent of species composition change after the tsunami. The convergence of the same permanently marked plots indicates that the cover of non-sandy beach species increases due to tsunami. Conversely, the divergence shows that species composition changes after the disturbance. Numbers show the pre- and post-tsunami sandy beach and maritime forest vegetation. Vegetation names of the numbers represented in this figure are given in App. 2.
previously did not occur on the side of Phuket Island facing the Andaman Sea (Simitinand 2001), was found after the tsunami (Fig. 2, App. 1) as type vege2, which occurs widely from tropical to subtropical Asia (Miyawaki & Suzuki 1976; Nakamura & Suzuki 1984; Suzuki et al. 2005). We suggest that plants with thalassochory, other newcomers such as Z. matrella (vege1) and S. littoreus (vege2) carried by sea waves, or those able to germinate from their seed banks, got a chance to establish because of substrate disturbance by the tsunami. Abiotic factors contributing to re-establishment or invasion of species after the tsunami In this study, beach soil conditions, such as soil hardness as an index of location stability, per cent soil water content and per cent silt content, and beach management pressure, were important factors determining the niche differences among species (Fig. 4). Many researchers reported the same patterns on sand dunes (Willis et al. 1959a, b; Jones & Etherington 1971; Kachi & Hirose 1979a, b). Most species of sandy beach vegetation are very different from those growing on nearby inland sites of the same region. This suggests that the environmental conditions of sandy beaches are too harsh for originally nonsandy beach plants to become established. Nevertheless, the relative importance of these natural and anthropogenic environmental conditions differs even across sandy beach zones, and differences among species in their tolerance of these factors can be a
determinant of community zonation (Ishikawa et al. 1995; Wilson & Sykes 1999). In New Zealand, for example, Sykes & Wilson (1990) reported that, although annual sandy beach species can be abundant along disturbed sandy beaches, they might not be as tolerant of sand burial. Conversely, perennial dune plants may grow better and more vigorously as sand accumulates (Woodhouse 1982; Maun & Baye 1989; Hesp 1991; Greipsson & Davy 1996). In particular, perennial graminoids usually respond to sand burial by stem elongation (Seliskar 1990; Sykes & Wilson 1990). At Phuket we found that, pre-tsunami, sandy beach herbaceous species growing near the shoreline, such as I. pes-caprae, V. marina, R. maritima, H. maritima and C. lineata, tolerate beach management pressures fairly well (Fig. 4). However, the occurrences, and therefore the habitats, of many other sandy beach herbaceous species were not clearly related to either sand accretion or erosion caused by the tsunami. Many sandy beach herbaceous species have a high tolerance to sand movement or have rapid recolonization ability, with wide rhizomatous or stoloniferous growth. Such disturbance also allows originally non-sandy beach species to occupy the coast; for example, D. aegyptium, E. indica and S. acuta invaded sites that accreted sand during the tsunami, and D. adscendens became established on sites that were eroded (Fig. 4). Soil water content is a major determinant controlling the re-establishment of sandy beach species. Indeed, soil water content often determines the vegetation zonation (Willis et al. 1959a, b; Jones & Etherington 1971; Keddy 1984; Wilson & Keddy 1986), as do soil
- SPECIES ECOLOGICAL TRAITS CONTRIBUTING TO EFFECTIVE REGENERATION AND INVASION AFTER TSUNAMI - 219 Euph.a Dig.a Cle.i
Hib.t Thu.i
Wed.b
Hyd.m dt Rem.m Ipo.p Cenh.e Cas.f
Axis2(0.335)
Sca.s Ter.c
Pan.o
sh scswc gb
Lep.r Tri.p Cas.e
Can.l
Pre.i Sid.a Sta.i
Vig.m Isc.m Dac.a bm Zoy.m. Ele.i
Cyp.s
Lan.c Ver.c
Phy.s
Axis1(0.611)
Fig. 4. Canonical correspondence analysis (CCA) ordination diagram for both the sandy beach and maritime forest species above 5% occurrence in all plots before and after the tsunami and environmental factors (bold characters). Abbreviations of species names are shown in Table 2. Species that occurred in fewer than 5% of all plots before and after the tsunami were excluded from the analysis. Before the CCA was carried out, environmental variables that were mathematical combinations of others were excluded (% fine gravel, % coarse sand, % fine sand, distance from the shoreline and microtopography), as was one member of each highly correlated variable pair (Pearson r40.65), in order to reduce multi-colinearity. Extracted environmental variables: sh, soil hardness; sc, % silt content; swc, soil water content; dt, disturbance type (change in elevation by soil erosion or accretion); gb, amount of garbage; bm, beach management pressure.
hardness and nutrient levels (Kachi & Hirose 1979a, b; Tsuyuzaki 1997). At Phuket, the habitat difference between sandy beach herbaceous species and maritime forest species after the tsunami is related to soil hardness, per cent silt content and soil water content (Fig. 4). Although most maritime forest communities were invaded by originally non-sandy beach T. procumbens, E. indica or D. aegyptium after the tsunami, not all of these communities were then assigned to a different vegetation unit (Fig. 3, App. 2). We suggest that maritime forest played an important role in protecting against soil disturbance by the tsunami. The same phenomenon occurs with typhoons and hurricanes; Gardner et al. (1992) reported that salt-induced foliage discoloration and defoliation were evident in the surge-inundated area of hurricane Hugo (South Carolina, USA). Ecological factors contributing to effective regeneration or invasion Species traits effective for recovery and regeneration or invasion after disaster were identified in this
research. After the tsunami, many sandy beach herbaceous communities were invaded by originally nonsandy beach species such as C. echinatus, D. aegyptium, D. adscendens, E. indica and T. procumbens (Fig. 2, App. 1). The occurrence frequency of originally nonsandy beach species in remaining vegetation (vege6b and 7) was also higher after the tsunami than before. In terms of dormancy forms, in this study, sites disturbed by the tsunami were often invaded by annuals, such as wind-dispersed grasses and asteraceous plants, rather than by perennials (Table 2), at least in the first stages of succession. Annual grasses and asteraceous plants are effective invaders of bare sites after large disturbances, as they can disperse over longer distance. We concluded that these originally non-sandy beach annual species have a high invasion ability for gaps in disturbed vegetation. We confirmed that when the gaps formed by the tsunami were 420% of the quadrat, originally non-sandy beach species rapidly invaded. We could not, however, clarify negative effects of originally non-sandy beach or invasive species on indigenous sandy beach communities, as a result of the tsunami, other than as changes in community composition. Indigenous sandy beach species such as I. pes-caprae, C. lineata and Z. matrella regenerated and re-established rapidly on the disturbed areas. We suggest that these species would re-establish dominance over the originally non-sandy beach species over time. Adaptive strategies of radicoid forms to the brief large disturbance were clarified. Rapid colonizers usually have subterranean long rhizomes (R1 and R2) that can tolerate disturbance better than close-set rhizomes (Tsuyuzaki 1989). While the pattern of radicoid forms on the same study beach did not appear to differ before and after the tsunami (Table 3), species with subterranean long rhizomes have strong tolerance for heavy tsunami disturbance: seven of eight sandy beach herbaceous species with subterranean long rhizomes (R1, R2) regenerated their population rapidly on both eroded and accreted areas. Species with clonal growth by stolons decreased significantly in frequency after the tsunami (Table 2). Thus, we concluded that the variation in proportion of vegetation types after a temporary large disturbance is determined by dormancy form and growth form, with radicoid form being influential (Table 3). In addition, regardless of whether resort or national park beaches, we found a significant change in the pattern of life-form composition on the same beach as a result of the tsunami on wave-cut beach terraces (Mai Khao and Kamala) than on the flattish sandy beaches (Nai Yan and Karon) (Tables 1 and 3). Therefore,
220
DAISUKE, H. ET AL.
we suggest that after a temporary large disturbance beach condition and vegetation dynamics can be predicted by using ecological traits such as the life form and portions of topography of the beaches before the disturbances. Acknowledgements. We are grateful to Thawatchai Santisuk and Thawatchai Wongprasert of The Forest Herbarium, Royal Forest Department, Thailand, for identifying plants in Thailand. We are indebted to Takao Kikuchi and Shinichi Meguro for technical advice. The paper benefited from the constructive comments of Milan Chytry as editor, and Karel Prach, Jill Rapson and another anonymous reviewer as referees.
References Braun-Blanquet, J. 1964. Pflanzensoziologie. 3rd edn. Springer, Vienna , Austria. Bush, D.M., Pilkey Jr., O.H. & Neal, W.J. 1996. Living by the rules of the sea. Duke University Press, London, UK. Carter, R.W.G. 1980. Vegetation stabilization and slope failure of eroding sand dunes. Biological Conservation 18: 117–122. Conner, W.H. 1998. Impacts of hurricanes on forests of the Atlantic and Gulf coasts. In: Laderman, A.D. (ed.) Coastally Restricted Forests. pp. 271–277. Oxford University Press, New York, NY, U.S.A. Conner, W.H., Mixon II, D.W. & Wood, G.W. 2005. Maritime forest habitat dynamics on Bulls Island, Cape Romain National Wildlife Refuge, SC, following Hurricane Hugo. Forest Ecology and Management 212: 127–134. Dahdouh-Guebas, F., Jayatissa, L.P., Di Nitto, D., Bosire, J.O., Lo Seen, D. & Koedam, N. 2005. How effective were mangroves as a defence against the recent tsunami? Current Biology 15: 443–447. Dawson, A.G. 1994. Geomorphological effects of tsunami run-up and backwash. Geomorphology 10: 83–94. Ellenberg, H. 1956. Grundlagen der Vegetationsgliederung Part 1. In: Walter, H. (ed.) Einfu¨hrung in die Phytologie. 136pp. Eugen Ulmer, Stuttgart, Denmark. Gardner, L.R., Michener, W.K., Williams, T.M., Blood, E.R., Kjerve, B., Smock, L.A., Lipscomb, D.J. & Gresham, C. 1992. Disturbance effects of hurricane Hugo on a pristine coastal landscape: North Inlet, SC, USA. Netherlands Journal of Sea Research 30: 249–263. Greipsson, S. & Davy, A.J. 1996. Sand accretion and salinity as constraints on the establishment of Leymus arenarius for land reclamation in Iceland. Annals of Botany 78: 611–618. Hatsusima, S. 1975. Flora of the Ryukyus. Biological Education Society of Okinawa, Okinawa, Japan. (in Japanese). Hayasaka, D. & Fujiwara, K. 2005. Species composition and environmental factors, including human impacts,
on coastal sand-dunes and maritime strand-forests in Southern Thailand. Tropics 14: 245–254. Hayden, B.P., Marcio, C.F.V.S., Shao, G. & Kochel, R.C. 1995. Geomorphological controls on coastal vegetation at the Virginia Coast Reserve. Geomorphology 13: 283–300. Hesp, P.A. 1991. Ecological processes and plant adaptations on coastal dunes. Journal of Arid Environments 21: 165– 191. Hill, M.O. & Gauch, H.G. 1980. Detrended correspondence analysis: an improved ordination technique. Vegetatio 42: 47–58. Ishikawa, S., Furukawa, A. & Oikawa, T. 1995. Zonal plant distribution and edaphic and micrometeorological conditions on a coastal sand dune. Ecological Research 29: 259–266. Jones, H.E. & Etherington, J.R. 1971. Comparative studies of plant growth and distribution in relation to waterlogging. IV. The growth of dune slack plants. Journal of Ecology 59: 793–801. Kachi, N. & Hirose, T. 1979a. Multivariate approaches to the plant communities related with edaphic factors in the dune system at Azigaura, Ibaraki Pref. I. Associationanalysis. Japanese Journal of Ecology 29: 17–27. Kachi, N. & Hirose, T. 1979b. Multivariate approaches to the plant communities related with edaphic factors in the dune system at Azigaura, Ibaraki Pref. II. Ordination. Japanese Journal of Ecology 29: 359–368. Kathiresan, K. & Rajendran, N. 2005. Coastal mangrove forests mitigated tsunami. Estuarine, Coastal and Shelf Science 65: 601–606. Keddy, P.A. 1984. Plant zonation on lakeshores in Nova Scotia: a test of the resource specialization hypothesis. Journal of Ecology 72: 797–798. Kira, T. 1977. A climatological interpretation of Japanese vegetation zones. In: Miyawaki, A. & Tu¨xen, R. (eds.) Vegetation science and environmental protection. pp. 21–30. Maruzen Co., Ltd., Tokyo, Japan. Maun, M.A. & Baye, P.R. 1989. The ecology of Ammophila breviligulata Fern on coastal dune systems. Aquatic Sciences 1: 661–681. Mazda, Y., Magi, M., Kogo, M. & Hong, P.N. 1997. Mangrove on coastal protection from waves in the Tong King Delta, Vietnam. Mangroves and Salt Marshes 1: 127–135. Miyawaki, A. & Suzuki, K. 1976. Vegetation der Du¨nen und der Korallenbauten auf den Ryukyu-Inseln. Bulletin, Institute of Environmental Science and Technology, Yokohama National University 10: 75–111. Miyawaki, A., Okuda, S. & Fujiwara, R. 1994. Handbook of Japanese vegetation. Shibundo Co., Ltd., Publishers, Tokyo, Japan (in Japanese.). Mueller-Dombois, D. & Ellenberg, H. 1974. Aims and methods of vegetation ecology. Wiley, New York, NY, USA. Nakamura, T. & Suzuki, K. 1984. Studies on the plant communities of the Yap Islands – plant ecological studies of the Micronesia. The Journal of Phytogeography and Taxonomy 32: 19–30.
- SPECIES ECOLOGICAL TRAITS CONTRIBUTING TO EFFECTIVE REGENERATION AND INVASION AFTER TSUNAMI - 221 Nordstrom, K.F., Jackson, N.L., Bruno, M.S. & de Butts, H.A. 2002. Municipal initiatives for managing dunes in coastal residential areas: a case study of Avalon, New Jersey, USA. Geomorphology 47: 137–152. Numata, M. 1947. Ecological judgement of grassland condition and trend: I. Judgement by biological spectra. Grassland Science 11: 20–33. (In Japanese with English summary). Numata, M. 1990. The ecological encyclopedia of wild plants in Japan. Zenkoku Noson Kyoiku Kyokai, Tokyo, Japan. (in Japanese). Raunkiear, C. 1934. Life forms of plants and plant geography. Oxford University Press, London, UK. Seliskar, D.M. 1990. The role of waterlogging and sand accretion in modulating the morphology of the dune slack plant Scirpus americanus. Canadian Journal of Botany 68: 1780–1787. Simitinand, T. (Royal Forest Department) 2001. Thai Plant names. Royal Forest Department of Thailand, Bangkok, Thailand. Stoneburner, D.L. 1978. Evidence of hurricane influence on barrier island slash pine forests in the northern Gulf of Mexico. American Midland Naturalist 99: 234–237. Suzuki, K., Laongpol, C. & Sridith, K. 2005. Phytosociological studies on vegetation of coastal dunes at Narathiwat, Thailand. Tropics 14: 229–244. Sykes, M.T. & Wilson, J.B. 1990. An experimental investigation into the response of New Zealand sand dune species to different depths of burial by sand. Acta Botanica Neerlandica 39: 171–181. Takematsu, T. & Ichizen, N. 1987-1997. Weeds of the world. Vol. I-III, Zenkoku Noson Kyoiku Kyokai, Tokyo, Japan. (in Japanese.).
ter Braak, C.J.F. 1995. Canonical correspondence analysis and related multivariate methods in aquatic ecology. Aquatic Sciences 57: 255–289. Tsuyuzaki, S. 1989. Analysis of revegetation dynamics on the volcano Usu, northern Japan, deforested by 1977-78 eruptions. American Journal of Botany 76: 1468–1477. Tsuyuzaki, S. 1997. Wetland development in early stages of volcanic succession. Journal of Vegetation Science 8: 353–360. Willis, A.J., Folkes, B.F., Hope-Simpson, J.F. & Yemm, E.W. 1959a. Braunton Burrows: the dune system and its vegetation. Part I. Journal of Ecology 47: 1–24. Willis, A.J., Folkes, B.F., Hope-Simpson, J.F. & Yemm, E.W. 1959b. Braunton Burrows: the dune system and its vegetation. Part II. Journal of Ecology 47: 249–288. Wilson, J.B. & Sykes, M.T. 1999. Is zonation on coastal sand dunes determined primarily by sand burial or by salt spray? A test in New Zealand dunes. Ecology Letters 2: 233–236. Wilson, S.D. & Keddy, P.A. 1986. Species competitive ability and position along a natural stress/disturbance gradient. Ecology 67: 1236–1242. Woodhouse Jr, W.W. 1982. Coastal sand dunes of the US. In: Lewis III, R.R. (ed.) Creation and restoration of coastal plant communities. pp. 1–44. CRC Press, Boca Raton, FL , U.S.A.
Received 15 October 2007; Accepted 19 August 2008. Co-ordinating Editor: M. Chytry´
App. 1. Summary table of sandy beach herbaceous vegetation on Phuket Island, southern Thailand before and after the Indian Ocean tsunami. oindicates originally nonsandy beach species. Roman numerals for each species indicate the constancy class and arabic numerals in parentheses show dominance rate. NY, Nai Yan Beach; MK, Mai Khao Beach; KM, Kamala Beach; and KR, Karon Beach. Additional species occuring once in releve´ vege type4 (after): Crinum asiaticum r(1), Hibiscus tiliaceus r(2), oPaspalum dilatatum r(1), oCoccinea indica r(1), oBorreria laevis r(1), oOxystelma esculentum r(1), type6a (before): Premna integrifolia I(1), oKyllinga nemoralis 1(1), oMimosa pudica 1(1), oOplismenus compositus 1(1), Pongamia pinnata 1(1), Clerodendron inerme 1(1), Calophyllum inophyllum 1(1), oLonicera japonica 1(1), oDesmodium triflorum 1(1), oAlocasia macrorrhiza 1(1), type6b (before): oPaederia scandens 1(1), type6b (after): oStachytarpheta indica 1(1), oHewittia sublobata I(1), oEuphorbia hirta 1(1), type7 (before): oCentotheca lappacea II(1), oMomordica charanria I(1).
222 DAISUKE, H. ET AL.
- SPECIES ECOLOGICAL TRAITS CONTRIBUTING TO EFFECTIVE REGENERATION AND INVASION AFTER TSUNAMI - 223
224
DAISUKE, H. ET AL.
