A floristic study on herbs and climbing plants at Puducherry, South India: An approach to biodiversity conservation and regeneration through eco-restoration

: A qualitative floristic exploration with life form classifications and monitoring of flowering and fruiting phenology has been carried out at a restored site near Puducherry, South India in 2009 and 2010. The species were classified into three categories based on their occurrence status, namely, Naturally Occurring, Naturally Regenerated and Introduced. The present study focuses only on two life forms, the Herbaceous and the Climbing plants. The site selected for eco-restoration originally comprised of an eroded and severely degraded landscape with scattered remnant species. However, active human intervention over a thirty year period included the introduction of appropriate plant species and other physical measures to enhance soil fertility and ground water level, and regenerate and conserve the deteriorating typical Tropical dry evergreen forest (TDEF) vegetation. A large number of naturally ccurring herbaceous, climbing species (172) and a consistent number of naturally regenerated species (44) are now observed as a result of eco-restoration. Lowland herbaceous species have also established themselves as a green cover at ground level. At present, parts of the area have fertile soils and rich floristic composition with the herbaceous life form represented by 165 species encompassed in 105 genera and 37 families, and the climbing plants represented by 68 species belonging to 54 genera and 25 families. diverse have been observed flowering throughout the year and about half of the species were observed fruiting throughout the year. We emphasize that the two fold approach of land and vegetation reclamation has been very effective in helping restore the unique TDEF vegetation at the local level and the same may be extended to help regenerate and conserve the Coromandel Coastal vegetation at the regional scale.


Introduction
India occupies only 2.4% of the world's geographical area, but is endowed with a variety of bedrocks, soils, climates, flora and fauna. It represents a wide range of ecological diversity (Gupta et al. 2006). Around 69% of India's total geographical area (about 328 million hectares, further abbreviated as mha) is classified as arid (15.8%), semi-arid (37.6%) or dry sub-humid (16.5%) (Ajai et al. 2009). The Ministry of Environment and Forests of the Government of India has reported that 107.4 mha of the total geographical area is affected by desertification (MOEF 2007). About 72.1% of existing forests have lost the capacity for natural regeneration (Rai and Saxena 1997). In India, waste lands account for nearly 55.6 mha while 32.8 mha forests are degraded or open, but could be made fertile again if treated properly through eco-restoration (MOEF 2007;National Remote Sensing Centre (NRSC) and Ministry of Rural Development 2010).
'Eco-restoration is the process of assisting the recovery of an ecosystem that has been degraded, damaged or destroyed' (Society for Ecological Restoration International Science and Policy Working Group 2004). In the present scenario, land and vegetation have deteriorated severely by high anthropogenic pressure that has led to unpredictable changes in the environments, eventually increasing the socio-economic crises. Hence high priority is needed Abstract: A qualitative floristic exploration with life form classifications and monitoring of flowering and fruiting phenology has been carried out at a restored site near Puducherry, South India in 2009 and 2010. The species were classified into three categories based on their occurrence status, namely, Naturally Occurring, Naturally Regenerated and Introduced. The present study focuses only on two life forms, the Herbaceous and the Climbing plants. The site selected for eco-restoration originally comprised of an eroded and severely degraded landscape with scattered remnant species. However, active human intervention over a thirty year period included the introduction of appropriate plant species and other physical measures to enhance soil fertility and ground water level, and regenerate and conserve the deteriorating typical Tropical dry evergreen forest (TDEF) vegetation. A large number of naturally ccurring herbaceous, climbing species (172) and a consistent number of naturally regenerated species (44) are now observed as a result of eco-restoration. Lowland herbaceous species have also established themselves as a green cover at ground level. At present, parts of the area have fertile soils and rich floristic composition with the herbaceous life form represented by 165 species encompassed in 105 genera and 37 families, and the climbing plants represented by 68 species belonging to 54 genera and 25 families. 'Genus to family' and 'species to genus' ratios indicate the establishment of diverse vegetation in the study site. Nearly one third of the species have been observed flowering throughout the year and about half of the species were observed fruiting throughout the year. We emphasize that the two fold approach of land and vegetation reclamation has been very effective in helping restore the unique TDEF vegetation at the local level and the same may be extended to help regenerate and conserve the Coromandel Coastal vegetation at the regional scale.
A floristic study on herbs and climbing plants at Puducherry, South India: An approach to biodiversity conservation and regeneration through eco-restoration highlighted anthropogenic pressures as a severe causal factor and afforestation as a priority especially needed in the Tropical dry evergreen forest (TDEF) of Coromandel Coast (Parthasarathy and Karthikeyan 1997;Sridhar Reddy and Parthasarathy 2003;Selwyn and Parthasarathy 2006;Muthumperumal and Parthasarathy 2009). The fact that our restored study site currently consists of a mosaic vegetation highlights the uniqueness of this check list that also tabulates actual observed reproductive phenology during the two years study period.
Phenology, the study of relationship between climatic factors and periodic biological phenomena in organisms, provides knowledge on the pattern of plant growth and development and selective pressures on flowering and fruiting behaviour (Opler et al. 1980;Zhang et al. 2006). Flowering of many herbaceous plants is influenced by environmental stimuli, mainly photoperiod and low temperature. Rainfall variability also plays an important role in the induction and length of flowering phenology (Sivaraj and Krishnamurthy 1989;Rivera and Borchert 2001;Borchert et al. 2004). Kramer (1997) stated that phenology and climate relationship can reveal the potential impacts of climate changes.
Ecological (Champion and Seth 1968;Blasco and Legris 1973;Meher-Homji 1973, 1974Parthasarathy and Karthikeyan 1997;Ramanujam and Kadamban 2001;Sridhar Reddy and Parthasarathy 2003), taxonomical (Kadavul et al. 2004a, b;Ramanujam et al. 2007;Udayakumar and Parthasarathy 2010) and ethno-botanical (Parthasarathy et al. 2008) studies have been carried out in the surroundings of the study area which comprises a rich and diversified vegetation, termed as TDEF showing high bio-resource values that are being fragmented severely by anthropogenic pressures and all these authors have emphasized conservation as a high priority. The added value of the floral checklist presented here stems from its uniqueness in representing a successful thirty year eco-restoration effort as well as the broad spectrum of vegetation types characterizing Puducherry and its surrounding areas.

Study site
The studied site (11°57'8.3"N, 79°45'57.2"E, and 40-50 m a.s.l.) spreads over 160 ha on the Eastern shore of the fresh water Ousteri Lake and about 10 km West of Puducherry city (Figure 1). Of the total study area, 40 ha is characterized as Cuddalore sandstone formation with red ferralitic soil. This red terrain is highly eroded as evidenced by the stratified walls of gullies and pebbles of various sizes and shapes ( Figure 2). The rest of the area is characterized by clay, sand and silt soils and a few places are alkaline in nature. In the last thirty years (1981 -2010), total mean annual precipitation of 1371 ± 354 mm (max= 2043, min= 845, median= 1333) and 56 ± 10 rainy days (max= 82, min= 37, median= 56) have been recorded. The rainfall is highly seasonal, with 63% of total precipitation occurring between October and December. During the past 6 years, as per our field records, rains occurred mainly in the nights. The monthly mean maximum and minimum temperature were 36.4°C and 21.6°C in June and January respectively ( Figure 3). Six to eight months of the year are dry. Puducherry region has no major forest or hills, but has small patches of TDEF elements, deciduous forests, scrub jungles, sacred groves and mangroves. Coastal wetland species are also found here.
The original site comprising an eroded and severely degraded landscape with scattered remnant species was taken up for eco-restoration, maintaining a "control plot" reflecting the condition, thirty years back (Figure 2A -D). Eco-restoration comprised active human intervention through the introduction of drought tolerant soil nitrogen fixing species and TDEF elements; along with native pollinators, especially Apis cerana Fabricius that were found in the canyons of the original fragmented landscape; construction of check-dams along the gullies to prevent the soil erosion and conservation of rainwater to increase the soil fertility and ground water level; simultaneous development of nurseries for propagating the collected seeds and seedlings from various regions helped increase the diversity and richness of vegetation. As a result, the introduced TDEF species have established themselves and have naturally regenerated along the thickets in the past thirty years ( Figure 2G). Lowland herbaceous species have also established themselves as a green cover at ground level, and soil texture has been transformed and the fertility increased.

Data collection
A qualitative floristic exploration has been conducted in 2009 and 2010. Sampling was carried out on alternate  days. All the species observed were collected and duly identified with the help of local floras (Bentham andHooker 1862 -1883;Gamble andFischer 1915 -1935;Matthew 1981Matthew -1983Nair et al. 1983;Henry et al. 1987;Henry et al. 1989) and authentic herbaria. For all plant species documented, the binominal and author citation have been checked with International Plant Name Index (IPNI) in 2009. Voucher specimens were deposited in the Herbarium of the French Institute of Pondicherry (HIFP). Flowering and fruiting phenology were monitored and life forms were also categorized simultaneously according to their habits. The species were then classified, based upon their occurrence status (OS), into three categories namely 1. Naturally Occurring (NO), referring to species growing naturally thirty years back when the eco-restoration started; 2. Naturally Regenerated (NR), those species which established themselves during restoration and 3. Introduced (IN) are the species selected and planted for restoration. The list of NO plant species is based on a survey done beginning in 1979 with the help of botanists of the French Institute of Pondicherry (Prakash Patel 1999-2000. The binomial and author citations, life forms, occurrence status, reproductive phenology and available Tamil names of the species observed are presented in Table 1. For most of the species, a complete photographic documentation is also presented (Figures 4-35).

Floristic composition
In the overall floristic survey, seven life forms, namely Herbs, Trees, Shrubs, Climbers, Grasses, Sedges and Epiphytes have been recorded. The present study focuses only on Herbaceous (H) including prostrate, procumbent, spreading herbs, sub-shrubs and under shrubs and Climbing life forms (C) including tendril and non-tendril climbers, vines, runners, lianas, stragglers, ramblers and twiners. Altogether 233 species belonging to 157 genera and 56 families were enumerated (Table 1 and 2); this comprised 165 species belonging to 105 genera and 37 families under herbaceous life form and 68 species belonging to 54 genera and 25 families under climbing plants. The present study includes 56 of the 206 families proposed in Bentham andHooker (1862 -1883) system of classification with subsequent modifications as in the flora of Tamil Nadu (Nair et al. 1983;Henry et al. 1987;Henry et al. 1989). The high resolution photographic documentation of 190 species, arranged in alphabetical order of families then by species (Figures 4-35) will facilitate field identification, as such documentation, especially for herbs which are pictorially less represented so far. Tamil vernacular names are provided in this check list with a view to make this available to the non-specialist also.
The actual number of herbs and climbers recorded in the present study vis à vis their occurrence in Puducherry and the Southern Coromandel Coastal region have been systematically compiled and summarized in table 3 from published literatures (Parthasarathy and Karthikeyan 1997;Sridhar Reddy and Parthasarathy 2003;Kadavul et al. 2004a, b;Ramanujam et al. 2007;Parthasarathy et al. 2008;Padmavathy et al. 2010;Udayakumar and Parthasarathy 2010); the table also provides information on the actual numbers of species, genera and families reported in each of the studies and details the respective site detail. We have compiled this table in order to highlight the conservation value of the eco-restoration carried out in our study site at a regional level. About 50% of species listed in these earlier works, spread over a wider geographic area, were enumerated in the present study site.
Out of the herbs and climbers observed, Fabaceae is well represented with the largest number of genera (19) and species (31), followed by Asteraceae (14, 17), Euphorbiaceae (7, 16), Amaranthaceae (11, 14), Acanthaceae (10, 12) and Convolvulaceae (6, 12); these six dominant families together comprise ~ 44% of the species in the two life forms reported here. The present study is in conformity with the earlier works of Ramanujam and Kadamban (2001), who have enumerated 169 angiosperms species around Puducherry and reported Fabaceae as the most speciose family in this area. Padmavathy et al. (2010) reported single species occurrence in 14 families. In our study, single species occurrence was found in 21 families and 116 genera. The three Monocotyledones families observed include 5 genera each with only one species per genus ( Table 2).
The overall 'genus to family' (G/F) ratio in the present study is observed to be 2.80 ± 3.54 (max= 19, min= 1, mean= 2.80 and median= 1), it is 2.84 ± 3.43 (max= 14, min= 1, mean= 2.84 and median= 1) for herbaceous and 2.16 ± 2.08 (max= 8, min= 1, mean= 2.16 and median= 1) for climbing life forms. Similarly the 'species to genus' ratio (S/G) is marked by 1.48 ± 0.97 (max= 6, min= 1, mean= 1.48 and median= 1) for all species observed. It is observed to be 1.57 ± 1.05 (max= 6, min= 1, mean= 1.57 and median= 1) for herbs and 1.26 ± 0.71 (max= 4, min= 1, mean= 1.26 and median= 1) for climbing plants. It means that relatively diverse vegetation is established in the study site, as suggested by Pielou (1975) and Magurran (2004) who opine that, in intuitive terms, hierarchical (taxonomic) diversity will be higher in an area in which the species are divided amongst many genera as opposed to one in which most species belong to the same genus, and still higher as these genera are divided amongst many families as opposed to a few. The genus, Euphorbia is found here to be the one with the most species (6), followed by Heliotropium (5), Cleome, Crotalaria, Hedyotis, Indigofera, Ipomoea, Jasminum, Leucas, Mollugo, Phyllanthus and Sida (4 species each). According to occurrence status, 172 species were found to be NO, 44 NR and 17 IN (Table 1 and 2). Among these, a large number of NO herbaceous, climbing species and a consistent number of NR species are now observed as a result of eco-restoration. The relative family (RF), genus (RG) and species (RS) values obtained are higher for herbs as compared to the climbing life form. Therefore, these findings suggest that the prevailing environmental conditions of the present study site favour the herbaceous taxa.
Species communities generally vary along ecological conditions, such as soil type and moisture content (Sollins 1998). Past phytosociological studies within the Pondicherry region have revealed that the microgeographic variation in species richness and abundance particularly determine the species composition (Dabholkar 1962;Marlange and Meher-Homji 1965;Antony 1982). In this study Striga densiflora was found to be established well in association with grasses especially Chrysopogon fulvus in the red ferralitic soil. Dugje et al. (2006) stated that same species occupied 27 -60% of millet fields in the savanna zones of northeast Nigeria. In the present eco-restoration site, Boerhavia diffusa and Trianthema portulacastrum flourish in clay soil with coarse sand. Sharma (1981) stated both of the above species show similar distribution in the semi-arid regions of India. In our study Ammannia baccifera, Biophytum sensitivum, Phyla nodiflora, Polygala erioptera and P. javana are found in stagnant water bodies or moist shady places. Interestingly, these are also reported in different parts of India (Kumar and Narain 2010;Murty and Venkaiah 2011) and other countries (Chmaitelly et al. 2009 andHarun-or-Rashid et al. 2009) in coastal and wetland habitats. We found that sandy-silt soil favours species like Crotalaria medicaginea, Evolvulus alsinodes, Hedyotis spp., Mollugo spp., and Polycarpaea corymbosa; Enicostema axillare grows well in alkaline soil; Euphorbia rosea and Tiliacora acuminata are common in Casuarina equisetifolia plantations and Mangifera indica groves respectively. Similar observation was done in the broad spectrum survey from the region of Tamil Nadu (Gamble andFischer 1915 -1935;Matthew 1981Matthew -1983. The analysis of species distribution patterns along environmental gradients is important for understanding the diversity and ecology of plants and species response to climate change, but detailed information are surprisingly scarce for the tropics (Wong and Whitmore 1970;Sollins 1998;Chaturvedi et al. 2011;Toledo et al. 2012). However, small scale regional studies, continuously over a long period, are needed to understand the dynamics of species distribution with reference to spatial and temporal factors.

Flowering and fruiting phenology
Nearly one third of the total observed species were flowering through the year; with most flowering having peaked in November and December. More rainfall, a shorter photoperiod and lower (mean maximum 30˚C) temperatures were recorded in this season. This study reveals that, of the observed species, most among the herbaceous life form do not show well defined vegetative and reproductive phenological phases. Although the peak flowering of most of the herbaceous species was observed after rainfall (personal observation), a considerable number of species did flower during the peak of summer. Several species initiated flowering from March onwards. Our findings on phenology are generally in agreement with those of Lang (1965); Sivaraj and Krishnamurthy (1989); Rivera and Borchert (2001); Ramirez (2002); Sridhar Reddy and Parthasarathy (2003). Climbing plants, mostly belonging to Asclepiadaceae, Menispermaceae and Vitaceae recorded flowering peak in the dry months between March and August. The results of the present study are in conformity with earlier ones obtained from other regions (Kephart 1987;Morellato and Leitao-Filho 1996). Seasonal variation in the availability of water and light should place physiological and thus evolutionary constraints on the phenologies of tropical forest plants (van Schaik et al. 1993;Wright 1996;Lieberman 1982;Borchert et al. 2004;Hamann 2004;Zimmerman et al. 2007        Long term surface data and remote sensing measurements indicated that plant phenology has been advanced by 2 -3 days in spring and delayed by 0.3 -1.6 days in autumn per decade in the past 30 -80 years, resulting in extension of the growing season (Parmesan and Yohe 2003). In our two year study, it was observed that some species were flowering twice a year. Andrographis paniculata, Heliotropium bracteatum, H. indicum, H. zeylanicum and Ventilago maderaspatana were blooming in July -November and December -March. The studies of Yadav and Yadav (2008) reported a similar observation. The peak flowering season of Capparis sepiaria, Cissus quadrangularis, C. vitiginea, Cocculus hirsutus, Derris ovalifolia, Enicostema axillare, Euphorbia rosea, E. serpens, Indoneesiella echioides, Leucas aspera, Merremia emarginata, Orthosiphon thymiflorus, Phyla nodiflora, Polygala erioptera, and Wattakaka volubilis was found to be summer (April -June). Selwyn and Parthasarathy (2006) Kumar and Narain (2010) reported similar phenology in low land areas; whereas it was winter (November -January) for Abrus precatorius, Biophytum sensitivum, Cissampelos pareira, Crotalaria medicaginea, Cyanotis tuberosa and Desmodium triflorum (Table 1). Morellato and Leitao-Filho (1996) and Selwyn and Parthasarathy (2006) also found parallel findings. Long-term monitoring and studies are essential to validate the extent of the growing season for individual species. About 50% of the species studied were found to be fruiting through the year. Demarcation between flowering and fruiting phenology is more obscure in the herbaceous than the climbing life form. Further studies on fruits and seed dispersal are proposed to be carried out.

Conclusion
The present study reports the diversity and reproductive phenology of herbs and climbing plants of an ecorestoration site. 233 species belonging to 157 genera and 56 families were enumerated. It reveals that 172 species are NO and 44 are NR. Among them, Aristolochia indica, Derris ovalifolia and Sarcostemma intermedium are rare and endangered. The rare medicinal plants, Enicostema axillare and Gloriosa superba are well established in the site. Cyclea peltata, Digera muricata, Euphorbia serpens, Gisekia pharnaceoides, Heliotropium zeylanicum, Leucas biflora, Spermacoce pusilla and Trianthema triquetra are reported for the first time in this area. About 50% of the total species reported by earlier works from the Southern Coromandel Coastal vegetation occur in our study site. Phenologically, the distinction between flowering and fruiting phenomena remained less in herbs as compared to climbers. This check list provides important baseline information on the successful plants that may be introduced in a damaged ecosystem or barren area for expanding the green cover in comparable regions. It will serve as a genetic resource for future references since there is a threat of genetic erosion due to global warming and climate change. As nearly 50% of the dry lands of India are devoid of forest, they can be reclaimed if treated properly by eco-restoration to develop a sustainable ecosystem and help to deal with the environmental crisis and enhance natural resources.                                Acknowledgments: Our sincere thanks to Mrs Sieglind D' Arcy for providing financial support through a PhD fellowship to R. Ponnuchamy. We are thankful to Mr S. Prasad and Drs B. R. Ramesh, Vincent Bonhomme and A. Stephen for taking time to give valuable suggestions on the first draft of this manuscript. We are thankful to Mr Barathan Ravi for helping with the herbarium specimens and Mr R. Sivarajan for preparing the study area map. We acknowledge Mr. G. Jayapalan for helping with proof reading the manuscript. We are grateful to the two reviewers for their feedback and suggestions to improve the manuscript.