Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Vegetation Response and Landscape Dynamics of Indian Summer Monsoon Variations during Holocene: An Eco-Geomorphological Appraisal of Tropical Evergreen Forest Subfossil Logs

Vegetation Response and Landscape Dynamics of Indian Summer Monsoon Variations during Holocene: An Eco-Geomorphological Appraisal of Tropical Evergreen Forest Subfossil Logs

  • Navnith K. P. Kumaran, 
  • Damodaran Padmalal, 
  • Madhavan K. Nair, 
  • Ruta B. Limaye, 
  • Jaswant S. Guleria, 
  • Rashmi Srivastava, 
  • Anumeha Shukla


23 Sep 2014: The PLOS ONE Staff (2014) Correction: Vegetation Response and Landscape Dynamics of Indian Summer Monsoon Variations during Holocene: An Eco-Geomorphological Appraisal of Tropical Evergreen Forest Subfossil Logs. PLOS ONE 9(9): e109455. doi: 10.1371/journal.pone.0109455 View correction


The high rainfall and low sea level during Early Holocene had a significant impact on the development and sustenance of dense forest and swamp-marsh cover along the southwest coast of India. This heavy rainfall flooded the coastal plains, forest flourishing in the abandoned river channels and other low-lying areas in midland.The coastline and other areas in lowland of southwestern India supply sufficient evidence of tree trunks of wet evergreen forests getting buried during the Holocene period under varying thickness of clay, silty-clay and even in sand sequences. This preserved subfossil log assemblage forms an excellent proxy for eco-geomorphological and palaeoclimate appraisal reported hitherto from Indian subcontinent, and complements the available palynological data. The bulk of the subfossil logs and partially carbonized wood remains have yielded age prior to the Holocene transgression of 6.5 k yrs BP, suggesting therein that flooding due to heavy rainfall drowned the forest cover, even extending to parts of the present shelf. These preserved logs represent a unique palaeoenvironmental database as they contain observable cellular structure. Some of them can even be compared to modern analogues. As these woods belong to the Late Pleistocene and Holocene, they form a valuable source of climate data that alleviates the lack of contemporaneous meteorological records. These palaeoforests along with pollen proxies depict the warmer environment in this region, which is consistent with a Mid Holocene Thermal Maximum often referred to as Holocene Climate Optimum. Thus, the subfossil logs of tropical evergreen forests constitute new indices of Asian palaeomonsoon, while their occurrence and preservation are attributed to eco-geomorphology and hydrological regimes associated with the intensified Asian Summer Monsoon, as recorded elsewhere.


The buried fossil woods and sub-fossil logs in the wetlands of coastal plains and river banks are the compact plant remains that form an important source of information on environmental changes of the geological and recent past. This compactness entails that wood resists decay and survives on land or in water for a considerably long period before fossilization. More than a century has elapsed since the occurrence of sub-fossil woods in shallow subsurface sediments of recent origin in several parts of India was discovered. One of the oldest descriptions of such occurrences was by Oldham in the early 1900s, when he published the area of recent subsidence. In several parts of India, particularly in coastal areas, the local people are familiar with the occurrence of this buried wood as it is often dug out and used as a source of fuel and timber for making furniture. However, such wood, buried in the shallow subsurface strata, has not attracted adequate attention of geoscientists or dendrochronologists. Our investigations since 1998 in the coastal plains and adjacent hinterland of Kerala State, situated in the southwestern part of India, points out several areas where logs of such fossils are found under thin overburden of Quaternary deposits [1]

Unlike the woods of older geological periods, the Quaternary woods are seldom preserved as petrifactions or permineralized forms. Here wood is fossilized as a charcoal-like material, called ‘fusains’ in the lignites and peat or remains undecayed as a sub-fossil, especially in the Holocene sediments. Considering their ubiquity in the sedimentary sequence, the fossil wood represents a unique palaeoenvironmental database. Such fossil logs do reveal detailed cellular structure and may be compared with their modern analogues. In fact, these fossils form a valuable source of climate data to alleviate a lack of contemporaneous meteorological records, provided they possess annual growth rings. As ‘growth ring formation’ is primarily related to climatic conditions, a specimen of this wood can serve as a potential sample for assessing the pattern of climatic changes in the immediate past few thousand years. This technique has already gained importance in palaeoclimatic study [2]. While studying the Quaternary stratigraphic sequence and geological events of the South Kerala Sedimentary Basin, the authors came across large quantity of tree trunks, embedded in the carbonaceous and silty clays at different stratigraphic levels at several locations along the Kollam-Kodungallur stretch. Therefore, their importance as potential archives of Late Quaternary environment is being addressed here. The natural affinity of the carbonized woods and subfossil logs with their nearest-living-relative forms (NRL), retrieved from the wetlands and associated landforms of southwestern Kerala, has been addressed while appraising the vegetation and climate dynamics of the Holocene period.

Regional Setting

The area of investigation is restricted to the Southwestern part of India, and it covers the coastal lands <8 m amsl [3] and adjoining parts of the midlands 8–75 m amsl [4] of southern Kerala. The eastern boundary of Kerala State is defined by the Western Ghats mountain range, also called ‘Sahyadri’, and to its west lies the Arabian Sea. This narrow strip of land has a 560 km long coastline. It has an area of 38,863 sq. km and extends between north latitudes 8°17′30″and 12°27′40″ and east longitudes 74°51′57″ and 77°24′47″. The land between the Sahyadri escarpment and the present coast is very narrow. Numerous small rivers originating in the plateau run for a short distance aided by the steep gradient and debouch into the Arabian Sea. Most of these rivers have developed estuaries of various geometries and orientations.

Geologically, the study area is composed of the Archean crystallines (Khondalites and Charnockite group of rocks), Neogene sediments (represented by Warkalli and Quilon Formations) and Quaternary deposits (represented by coastal sands, muds and alluvium) (Figure 1). The main part of the coastal land occurs as a curvilinear embayment, which has proved to be a landward extension of the off shore Kerala- Konkan Basin. This landward extension is called the South Kerala Sedimentary Basin (SKSB) [5]. The entire area is drained by 13 small rivers with catchment area <10,000 sq. km [6]. The river valleys in the coastal lands and adjoining midlands are broad with fairly thick alluvial sediments. The lower part of the alluvial sediments is composed of channel sands, while the upper part is made up of silty sediments. The present river channel is narrow as compared to the wide valleys occupied by flood plain sediments. The flood plain sediments of Early Holocene are deposited over riparian vegetation/forests at many stretches. In addition to river valleys, the coastal land close to wetland and areas within wetlands also contain buried woods and sub fossil logs (Figure 2). The area of investigation comprises the coastal plains and the adjoining hinterland between lat. 8°25′ and 10°15′ N. This part of coastal stretch comprises lagoons and estuaries, perennial and seasonal inland wetlands, used for rice cultivation, alluvial fans and fan deltas and the ridge-runnel systems. The coastal plains, located in stretches that are tectonically stable or undergoing uplift, suffer erosion, and consequently no sizeable thickness of sediments can be found in such plains. The Quaternary sediments, therefore, are generally confined to estuaries, abandoned river channels and river terraces.

Figure 1. Location of major wetlands in Southwestern India showing geological formations/rock type and lineaments (Modified after [85]).

Figure 2. Landward extension of Kerala–Konkan basin showing fossil wood locations associated with wetlands and major river basins of south west India.

Materials and Methods

Approval for field studies was obtained from Directors of Centre for Earth Science Studies, Thiruvananthapuram and Agharkar Research Institute Pune, India. Systematic field studies were carried out to collect data on various landform features in the coastal lands, and regions adjoining the midlands that constituted the study area. The locations of buried wood and sub fossil logs were mapped on 1∶50,000 scale Survey of India Topo base maps. Details of the location of wood samples and their retrieval along with their environment are presented in table 1. Fossil specimens studied, and the micro- preparations in the form of slides are housed in the repository of museum of the Birbal Sahni Institute of Palaeobotany (BSIP), Lucknow, India.

Table 1. Location of wood samples along with their environment and details of sample collection.

Specimen details

Dipterocarpus sp. cf. D. indicus Bedd. (BSIP Museum No. 40078 and BSIP Museum Slide No. 40078–1, 2, 3), Calophyllum sp. (BSIP Museum No. 40078 A and BSIP Museum Slide No. 40078 A–1, 2, 3, 4), Diospyros sp. cf. D. bourdilloni Brandis, BSIP Museum No. 40079 and BSIP Museum Slide No. 40079–1, 2, 3,4,5,6,7), Careya arborea Roxb. (BSIP Museum No. 40080 and BSIP Museum Slide No. 40080–1, 2, 3), Artocarpus sp. cf. A. lacucha Buch-Ham. (BSIP Museum No. 40081 and BSIP Museum Slide No. 40081–1, 2, 3), Rhizophora sp. cf. R. mangle L.(BSIP Museum No. 40082 and BSIP Museum Slide No. 40082–1, 2, 3) and Neolamarckia sp. cf. N. cadamba BSIP Museum No. 40083 and BSIP Museum Slide No. 40083–1, 2, 3).

As most of the Quaternary sediments are associated with the wetlands, retrieval of sediment samples and fossil logs was a difficult task. However, trunks, branches and roots of trees buried under 1–5 m of sediments can be retrieved. In addition to this, such vegetal remains are also encountered buried under sand and clay in river beds and flood plains, particularly along river courses in coastal plains and lower elevation of lowlands. Although, the occurrence of carbonized wood and subfossil logs has been observed virtually all along the entire coastal stretch of Kerala, fifteen locations are associated with lake and river basins and palaeoestuaries (Figure 2; Table 12). The tree trunks are often found buried under 1–3 m clay sequence in vast areas, all along the coast from Karunagapally to Ernakulam. Along the banks of Kayamkulam and Vembanad lagoons, including Mundakan, Kari, Kuttanad and Kol lands, subfossil logs are encountered while digging canals for local navigation and irrigation. Most of the paddy fields, lakes, swamps and marshes, in the midland region contiguous to SKSB and elsewhere are reported to contain trunks of trees. Some of these wetlands have 6–10 m sediment fills. The clay in some of them is mined for manufacturing of tile-brick. These quarries too expose tree trunks at several levels. Many of the river terraces, particularly those in the terrain, which have undergone uplift in Quaternary times, are also reported to contain carbonized wood. Due to the erosion of the riverbank this wood floats down the river, and is collected to serve as fuel.

Table 2. Lithological bearing of the subfossil wood samples along with other relevant details including radiocarbon dates.

The subfossil/non-carbonized/carbonized woods were thin-sectioned using standard Reichert sliding microtome, and the sections made in transverse, tangential longitudinal and radial longitudinal planes were studied under a high power binocular microscope. Systematic affinities of the woods were initially determined by consulting reference literature [7][10] and searching the computerized wood database [11]. Subsequent comparisons were made with extant wood samples housed in Xylaria of Forest Research Institute, Dehra Dun and BSIP. Wood descriptions and measurements were taken in accordance with IAWA recommendations [12]. All the figured specimens are deposited in the museum of Birbal Sahni Institute of Palaeobotany, Lucknow, India. Radiocarbon dates of the woods were obtained from Birbal Sahni Institute of Palaeobotany, Lucknow, India. Calibrated age is based on 5570±30 yrs using state of art calib611 version under 2 sigma confidence levels. Though 25 species of woods were identified in the total assemblage, only the diagnostic anatomical details and ecological and climate potential of 7 species are dealt in the main text, and other relevant information and supporting illustrations are referred to in the supplementary data and presented in table 3.

Table 3. List of identified sub fossil logs and partially carbonized woods from the wetlands of Kerala.

Observations and Results

Mode of occurrence of fossil wood and subfossil logs

The subfossil logs and partially carbonized wood remains were excavated from trenches, and also retrieved from subsurface sediments of landforms in varied environmental settings. They have been found in wetlands adjacent to lagoonal water bodies, as well as those away from coast without any physiographic connection to the present lagoons. In both cases, tree trunks are found buried under clay, usually rich in carbon. The trunks are found with roots and branches still intact, indicating clearly that the trees grew and fell down at the same place, and got buried under sediments. Besides, no geological agent appears to have been capable of transporting such large trunks with diastemising roots and branches. The sediments in which the trunks occur are found to be of the same age or ages younger by 100–200 years. These sediments are rich in microflora, supportive of inland swamps/marshes, or in some cases of mangrove origin with salinity tolerant varieties [13][14]. The sediments with salt tolerant microflora, which mostly contain littoral micro fauna possibly washed in by tidal action, also store preserved hard tree trunks.

Tree trunks of the past vegetation are also preserved in wetlands that lie at a distance of 10–20 km away from the coast. These wetlands have no physical contiguity, in the form of embayment, with multiple digits that contain laterites or sand mounds on the sides. The clay in such wetlands is usually 2–5 m thick and rich in carbon. Carbonized wood fragments are rich in this type of environmental setting. Tree trunks with roots and branches are noticed at different locations of these sand mounds. It is apparent then that the laterites and sand mounds were covered by thick forest, yielding abundant organic matter deposited with the clay.

The sediments in the river channels and adjacent paleo-flood plains are underlain by a horizon with uprooted tree trunks. These get exposed when the 4–5 m thick sand in the river channels is mined. The paleo - flood plains also include meandering river channels where sand mining exposes buried tree trunks. This wood undergoes fragmentation on exposure with the flowing river bed, and the resultant fragments are collected by locals to serve as cooking fuel. Preservation of fossil wood and sub fossil logs is directly related to the sediments in which they are found. The best preservation with minimal carbonization and disintegration is found in cases where the wood is covered by pure clay. If the clay is highly sandy/silty, the wood undergoes carbonization at an increased pace. In fact, in most cases, the clay is sandy, resulting in various degrees of carbonization. In rare cases, wood has been found at the interface between a lower sandy layer and an upper clayey horizon. In such cases, the wood appears to have undergone complete carbonization with overburden pressure; the carbonized trunk appears highly fragmented, devoid of any internal structure. Therefore, it appears that preservation of wood is a function of the sediment which encases it. The inference, therefore, is that only a small part of the wood that grew in an area is preserved.

Field observation and chronology of subfossil logs

Considering the radiocarbon ages, the collected fossil wood and sub fossil log samples can be grouped into three categories. The oldest category belongs to Late Pleistocene- Early Holocene (13.0–10.0 k yrs BP) and the fossils are from Vettiyar and Vamanapuram River bank, west of NH 47. Their occurrence can be attributed to the intensified monsoon of ca 13 and 10 k yrs BP. The pattern of flooding and uprooting of roots along the river banks and entombment of the tree trunks occurred soon thereafter. The second category includes the wood remains of ∼8.0 to 5.3 k yrs BP retrieved from wetlands adjacent to lagoonal water bodies and abandoned meander loops. The tree trunks in this setting are embedded in clayey sand/sandy clay and exhibit evidence of abrupt flooding that led to uprooting of trees, and burial under sediments of approximately the same age. The samples from Pangod quarry, Karippuzha ‘punja’, Panthalam (Achankovil River bank) and Karamana River bank belong to this category. The fallen trees were buried by excessive sediment deposited by the rivers. Samples from Vamanapuram River bank to the east of NH 47 and Killiar bank represent the present day vegetation and climate (Figure 3).

Figure 3. Field view of subfossil logs from wetlands of Kerala.

(a) Vamanapuram River Bank, Thiruvananthapuram, ∼3 km west of Parippally. (b), (d) and (g) Karamana River bank, Thiruvananthapuram. (c) and (f) Ayiroor. (e) Partially carbonized wood from Karippuzha, Alappuzha. (h) Heap of carbonized logs in the backyard of a coastal village, Pathiyoor. (i) Vettiyar, Alappuzha. (j) Coastal lowlands of Periyar River. (k) Sacred grove of Keerikkad village, Alappuzha.

Systematic description of identified woods

Seven taxa were anatomically recognized among the 25 species of wood assemblage (Table 2). Anatomical literature and illustrations of earlier published works were used for identification, and sub fossil logs were assigned to their NRL [7][10], [15][31]. All the taxa are briefly described to show the basis of identification, affinity and ecological and climate potential. The antiquity of the described species and their present distribution has also been dealt with. Anatomical description of the identified woods and supporting illustrations are provided in the supplementary files.

Artocarpus sp. cf. A. lacucha.

Buch-Ham., Moraceae, Figure A,1–5 in File S1; Figured Specimen–BSIP Museum No. 40081.

Carbonized wood pieces were collected from Thottapally wetland (9°18′41″ N–76°24′00″ E) in Alappuzha. The wood samples are well preserved and show all the anatomical details.

The important characters of the present fossil are: presence of medium to large vessels, which usually occur alone or in radial multiples of 2–4 with abundant tyloses; large inter-vessel pits, vasicentric to aliform-confluent parenchyma, heterocellular broad rays with occasional occurrence of latex tubes and non-septate fibres. These xylotomical features indicate its close affinity with the modern woods of the genus Artocarpus Forster and Forster f. of the family Moraceae. In order to find their nearest modern counterpart, thin sections available at Birbal Sahni Institute of Palaeobotany and photographs of a number of Artocarpus species were critically examined, viz., A. chaplasha Roxb., A. dadah Miq., A. elasticus Reinw. ex Blume, A. gomezianus Wall. ex Trec., A. heterophyllus Lamarck., A. hirsutus Lamarck., A. incise Linn., A. lacucha Buch-Ham., A. lancefolius Roxb., A. nitidus Trec. A. scortechinii King, A. sericarpus Jarrett, A. sepicanus Deils and A. tomentosulus Jarrett. The present carbonized wood shows resemblance with A. chaplasha, A. gomezianus and A. lacucha which are xylotomically inseparable [20]. Amongst them, A. lacucha is found in Kerala [25]. In view of its close anatomical similarity and its occurrence in the Kerala, the sub-fossil has been assigned to A. lacucha.

The family Moraceae is a small family with 38 genera and 1100 species distributed in tropical warm regions with a few in temperate zones. Of the five tribes of Moraceae, the genus Artocarpus Forster J. R. and Forster belongs to the tribe Artocarpeae. About 50 species of this genus are distributed in Indo-Malaysian region [26]. About 7–8 species are found in India, of which A. heterophyllus Lamarck, A. hirsutus Lamarck and A. lacucha Buch-Ham. are distributed in the moist evergreen forests of Western Ghats.

The fossil history of the genus Artocarpus dates back to Maastrachtian-Danian in India, and a large number of fossils of different plant parts have been recorded from different Neogene sequences of the world [32]. The genus has continued to occur along the Kerala coast since Miocene i.e., about 15 million years ago [32][34].

Calophyllum sp., Clusiaceae, Figure 4, 1–4; (see Text S1 in File S1 for detailed description).

Figure 4. Wood anatomical details of Calophyllum sp. (1–4) and Careya arborea Roxb. (5–10).

1. Transverse section showing distribution of obliquely arranged tylosed vessels and apotracheal parenchyma bands. Scale bar  = 250 µm; (BSIP Museum Slide No. 40078 A-1). 2. Tangential longitudinal sections showing uniseriate rays and tylosed vessel elements. Scale bar  = 100 µm; (BSIP Museum Slide No. 40078 A-2). 3. Tangential longitudinal sections magnified showing uniseriate rays ands solitary crystal Scale bar  = 50 µm; (BSIP Museum Slide No. 40078 A-3). 4. Radial longitudinal section showing heterocellular rays. (BSIP Museum Slide No. 40078 A-4). 5. Transverse section showing distribution of tylosed vessels and apotracheal parenchyma lines. Scale bar  = 250 µm; BSIP Museum Slide No. 40080-1. 6. Transverse section of another sample showing distribution of tylosed vessels and apotracheal parenchyma lines. Scale bar  = 250 µm; BSIP Museum Slide No. 40080-2. 7. Tangential longitudinal section showing distribution of multiseriate rays and tylosed vessels. Scale bar  = 250 µm; BSIP Museum Slide No. 40080-3. 8. Tangential longitudinal section enlarged of another sample showing a multiseriate rays. Scale bar  = 100 µm; BSIP Museum Slide No. 40080 - 4. 9. Tangential longitudinal section showing bordered, alternate hexagonal intervessel pits. Scale bar  = 50 µm; BSIP Museum Slide No. 40080 -5. 10. Radial longitudinal section showing heterocellular ray cells. Scale bar  = 100 µm. BSIP Museum Slide No. 40080 - 6.

The characteristic features of the fossil wood are: solitary vessels with tyloses arranged in oblique radial lines, vasicentric tracheids, apotracheal parenchyma in broken bands and uniseriate rays. These characters collectively indicate affinities of the wood sample with the extant woods of the genus Calophyllum Linn. of the family Clusiaceae.

The genus Calophyllum Linn. is confined to moist tropical regions of the world mainly in Southeast Asia. About a dozen species occur indigenously in India and it is difficult to distinguish them xylotomically from each other. Hence the sample is assigned to Calophyllum sp. Out of a dozen species, Calophyllum inophyllum Linn., C. tomentosum Wight and C. wightianum Wall. are found on river banks and evergreen forests of Kerala. C. inophyllum is found along the coast above high water marks.

Careya arborea Roxb., Lecythidaceae, Figure 4, 5–10 and Figure B, 1–6 in File S1, Figured Specimen–BSIP Museum No. 40080.

The description is based on two pieces of woods retrieved from Karamana River bank (Trikkannapuram - 8°28′30″ N–77°00′00″ E) in Thiruvananthapuram and Pathiyur wetland (9°12′07″ N–76°30′37″ E) in Alappuzha.

The characteristic features of the fossil wood are: vessels mostly in radial multiples, simple perforations, intervessel pits medium-large; abundant parenchyma, both paratracheal and apotracheal; paratracheal vasicentric and apotracheal as diffuse-in-aggregate forming 1–2 seriate broken lines; rays 1- 4(5) seriate, heterocellular and fibres nonseptate. These characters collectively indicate that the fossil logs possess a close resemblance with modern woods of Careya Roxb. from the family Lecythidaceae. On surveying the anatomical literature and examining the available modern wood slides, it is found that the samples exhibit strong resemblance with the extant woods of Careya arborea Roxb. Pantropical family Lecythidaceae consists of about 24 genera and 285 species distributed across tropical rain forests, especially South America. Careya is a small genus (4 species) of trees and shrubs found in the Indo-Malaysian region. Careya arborea Roxb., which resembles the wood samples, is a moderate sized tree and widely distributed across India including West Coast (Karnataka, Kerala and Tamil Nadu States) and Myanmar. Hence it is assigned to the same species. The genus Careya is well documented from the Neogene exposures of India. The genus has continued to occur on the Kerala coast since Miocene times [35].

Diospyros sp. cf. D. bourdilloni Brandis, Ebenaceae, Figure C, 1–7 in File S1, Figured Specimen–BSIP Museum No. 40079.

The carbonized wood samples were collected from Elanjikkal wetland (9°16′40″ N–76°34′33″ E) in Alappuzha.

The diagnostic features of the wood are: presence of mostly small to medium sized vessels in radial multiples of 2–5, scanty paratracheal parenchyma, apotracheal forming uniseriate concentric lines at regular intervals; frequent uniseriate rays, rare biseriate due to pairing of cells, composed of procumbent cells in the centre with uniseriate extensions of 1–2 upright cells at both the ends and non-septate, thick walled fibres. These features suggest close affinity of the carbonized wood with the genus Diospyros L. of the family Ebenaceae. The study revealed that though xylotomical characters of most species are very similar; yet the carbonized wood under consideration closely resembles D. bourdilloni Brandis and D. varigata Kurz in all structural details, owing to the frequent occurrence of biseriate rays [10].

The small family Ebenaceae consists of only 2 genera (Diospyros L. and Euclea Murr.), and 485 species distributed across tropical and sub-tropical regions of the world. Of the 475 species of genus Diospyros, 200 occur in Indo-Malaysian region [26]. About 40 species are found in India. D. bourdilloni -the nearest modern counterpart is found in the moist tropical evergreen forests of Anamalai and hills of Tirunelveli up to an elevation of 600 m [23], while D. varigata is distributed across the north-eastern states (Assam, Meghalaya and Mizoram).The fossil woods of genus Diospyros are known to exist from Maastrachtian-Danian onwards in the world, and occur frequently in Neogene exposures in India. The genus has continued to occur in South India (Kerala and Tamil Nadu) since Miocene times [32][34].

Dipterocarpus sp. cf. D. indicus Bedd., Dipterocarpaceae, Figure D, 1–5 in File S1, Figured Specimen–BSIP Museum No. 40078.

The species is based on carbonized wood piece retrieved from river channel deposits of Manimala River - 2 (Karuthavadasserikkara: 9°23′48″ N–76°39′26″ E) Pathanamthitta.

The diagnostic features of the wood are: almost solitary tylosed vessels, vasicentric tracheids, scattered gum canals, which occur as solitary or in groups of 2–4(5), paratracheal parenchyma vasicentric, apotracheal parenchyma scanty, as few diffuse cells among fibres and in the form of short tangential bands enclosing gum canals, 1–6 (mostly 3–5) seriate, distinctly heterocellular rays. The combination of all these characters indicates that the fossil belongs to genus Dipterocarpus Gaertn. f. of the family Dipterocarpaceae. From the survey of wood slides and literature, it is found that the fossil shows close resemblance to the wood structure of Dipterocarpus indicus Bedd. (BSIP wood slide no. 308) and D. lowii Hook. f. (BSIP wood slide no. 2106) as diffuse to diffuse-in-aggregate parenchyma is almost absent in these two species. However, in other wood samples of Dipterocarpus indicus amount of parenchyma is much more.

The genus Dipterocarpus Gaertn.f. includes about 69 species which are mainly confined to Indo-Malaysian region with maximum development in Borneo, Malaysian Peninsula and Sumatra [26]. The genus ranges in its distribution from India in the west to Philippines in the East. In India, it is found in Assam, the Andamans and the Western Ghats [17]. Dipterocarpus indicus, which resembles the fossil closely, is a large tree found in Western Ghats [24] while D. 1owii grows in the Malaysian region. In India, the fossil history of genus Dipterocarpus traces back to Lower Miocene, and a large number of fossils of different plant parts are recorded from different Neogene exposures. The genus has occurred on the Kerala coast since Miocene i.e., about 15 million years ago [32], [34], [36].

Neolamarckia sp. cf. N. cadamba Rubiaceae, Figure E, 1–5 in File S1, Figured Specimen–BSIP Museum No. 40083.

The important characteristics of the present carbonized wood are: presence of mostly small to medium sized vessels which occur either as solitary or in radial multiples of 2–5, medium sized inter-vessel pits, parenchyma both paratracheal and apotracheal; paratracheal scanty, apotracheal diffuse to diffuse-in-aggregate, forming uniseriate lines; heterocellular, 1–3 seriate rays with long uniseriate extensions upright cells and non-septate, thin walled fibres. These features collectively indicate its close affinity with the modern woods of the genus Neolamarckia cadamba (Roxb.) Bosser (Synonyms: Anthocephalus cadamba (Roxb. Miq.; Nauclea cadamba Roxb.) of the family Rubiaceae. However, the controversy whether novem Neolamarckia should be used or Anthocephalus is to be retained, has not yet died down [37]. The classification and phylogeny of Nauleaea based on molecular and morphological data favours maintenance of the genus Neolamarckia [38]. The carbonized wood also shows superficial resemblance to the woods of family Sapotaceae as they possess similar type of parenchyma and rays, but can be differentiated by the presence of vasicentric tracheids and heavily tylosed vessels.

Rubiaceae (Coffee family) is the fourth largest angiosperm family in wet tropics with more than 80% arborescent genera [39] but very few also occur in temperate and arctic region [29]. It contains 680 genera and more than 10,200 species [26]. Most of the genera occur in forest under storey where their fruits, leaves etc. provide food resources for animals. The genus Neolamarckia cadamba (Roxb.) Bosser is widely distributed from India to Malaysian peninsula, Indonesia, Philippines, New Guinea and Australia [37], [40]. It grows best on deep moist alluvial soil along river banks in tropical moist deciduous and evergreen forests. In India, it occurs in sub-Himalayan tract from Nepal eastwards in Darjeeling, Assam, northern Bihar, Orissa and Eastern Ghats. It is not frequently found in southern region, but reappearing in Kadappah and Karnool. In the West coast, it occurs from North Kanara to Travancore and is also found in Andamans [29].

Rhizophora sp. cf. R. mangle L., Rhizophoraceae, Figure F, 1–6 in File S1, Figured Specimen–BSIP Museum No. 40082

The description is based on wood samples collected from Adichanalloor wetland (9°52′32″ N–76°42′36″ E) in Kollam.

Presence of small vessels with scalariform perforation plates and scalariform intervessel pits, scanty unilaterally paratracheal parenchyma, heterocellular multiseriate rays and thick walled fibres indicate its affinity with the woods of Rhizophora, particularly with R. mangle L. of the family Rhizophoraceae.

Family Rhizophoraceae consists of 15 genera, often mangroves and 120 species distributed in tropical regions of specially Old World. The genus Rhizophora L. is sometimes collectively called true mangroves. R. mangle naturally grows in subtropical and tropical regions of both the northern and southern hemispheres between 28° N to S latitudes where they exist in conditions of high salinity, extreme tides, strong winds, high temperatures and muddy and anaerobic soils [41]. The oldest record of Rhizophora in India dates back to Oligocene of Assam [42].


Taphonomy of fossil woods

Though there have been a few reports of fossil woods and partially carbonized logs, these paleosources have not attracted any scientific interest. The reported information is devoid of any description of the geological settings and taphonomy associated with them. Except a few radiocarbon dates, no details are available though they form potential archives of environmental changes and palaeoclimatic signatures. In fact, large and small trunks of trees, often with their bark, parts of the roots and branches intact, are very common in the seasonal and perennial wetlands. In some cases, the wood is partially carbonized especially in the Late Pleistocene sequences, whereas, the Holocene logs are generally hard and are usually found with the bark intact (Figure 3 a–b & e–f). These fossils and sub-fossil logs are regularly retrieved, when the local inhabitants make trenches and canals to irrigate their paddy crop during the dry season, and recover brick clay during mining. Further, the wood parts frequently encounter at intervals while retrieving the subsurface sediments with the help of Standard Penetration Test (SPT) coring. A major part of the wetlands, including the areas associated with the backwaters of Kerala, is referred to as “Karinilam”. Karinalam literally implies “black land”, which is a reference to the high content of organic carbon sediment found in this area. The area is, thereby, one of the most prominent carbon sinks in this part of Peninsular India. In fact, the wetlands and associated landforms hold immense potential for carbon sequestration, and mitigating the impact of climate change due to Global warming in the region. Therefore, the study of terrestrial organic matter particularly that of the buried and preserved woods is significant, while ascertaining the response of vegetation to climate dynamics over a period of time.

Since the region has been subjected to tectonic and hydrologic regimes during the Late Quaternary, the spatio-temporal distribution and environmental setting of these woods are brought to light, while addressing the response of the vegetation to the coastal dynamics. As the fossil logs are found in abundance and are incorporated into peat or lignite (coalified in varying degrees), the cellular details would be preserved in them, and they could be assigned to their “nearest-living-relative” (NRL) or modern analogues. With the aid of the preservation potential of these fossil logs, the growth ring studies may also be carried out as if they were modern woods in dendroclimatology. Accordingly, these ancient woods represent an important source of palaeoclimatic information, which is now being utilized as a proxy. Although, conifers and dicotyledonous woods in temperate and sub-alpine zones have been utilized for climatic reconstruction due to the presence of distinct growth rings and vessels of special arrangement patterns associated with thermal seasonality, tropical and subtropical trees with the exception of Teak (Tectona grandis) have seldom been tested. This paucity can be attributed to the lack of information on datable tree-rings of fossil, and modern woods from Indian subcontinent. In coastal areas like Kerala, variation in temperature and humidity is not so evident (non-seasonality). However, Worbes [43], [44] confirmed that in tropical regions, where an annual drought or flood occurs, the wood also exhibits annual growth rings, and the width of growth rings depend on the length of time a tree remains exposed to hydrological extremes. Our efforts over the past few years, persistent excavation, and meticulous search of these fossil woods and sub-fossil logs in the wetlands of Kerala have led to identification of about twenty five species, mainly from tropical evergreen forests, including a few mangrove taxa. These species possess the potential to be employed for dendroclimatic study. The natural affinity of these fossils has been established from anatomical details and ethnobotanical information including local/vernacular names provided by the local inhabitants, who frequently use them for making implements, and fuel due to their high calorific value. Some of these species have already been reported from the Neogene sediments (Mio-Pliocene) that belong to Warkalli Formation, and a few species are still grown, particularly in the sacred grooves, as relict representatives of the pristine forest that had thrived profusely until the Late Pleistocene - Early Holocene in the low lands of Kerala (Figure 3k; Table 1). Although the potential of a few taxa for climate reconstruction has been established, there has hardly been any attempt to utilize them in the paleoclimatic perspective. The few reports in recent years on the buried fossil logs in the wetlands and their utility in environment and culture appraisal, have been sporadic and few [5], [13], [33], [45], [46]. Cherian et al. brought to light recently the practice of usage of local timbers for wharf and boat making in the historical contacts with Mediterranean, Red Sea and Indian Ocean rims [47].

Chronology of fossil woods

The studied wood samples have yielded ages that fall mainly into two groups. Wood samples that occur very frequently have yielded ages ranging from ca 9000 to 5000 yrs BP (Table 4). This time interval is unique to the entire area, as it falls along the path of South West (SW) Monsoon or Asian Summer Monsoon. The monsoon is stated to have been considerably stronger than the present [1], [5], [48], corresponding to the Holocene Climate Optimum (HCO). The terrain under present investigation now receives 200–400 cm of annual rainfall from the SW Monsoon. This supports the thick tropical rainforest vegetation covering >35% of the states area in the highland - midland and lowland regions. The coastal plains have historically been lush with the mangroves and freshwater swamps and marshes. However, since the late forties this picture has dramatically changed, and the state now has a net forest cover of 15–20% which is fast depleting. Mangroves and other swamps and marshes are confined to less than 10% of their area of that of pre-50 s. If such a thick forest cover can be supported by the present rainfall, it is not difficult to visualize the amount of vegetation that would have been flourishing when rainfall was 2–3 times more than the present. Another sample of wood has yielded an age of ca 13000 yrs BP [13].The period 13000–12000 yrs BP is believed to have witnessed intensified monsoon activity. As compared to the Early to Middle Holocene, the terrestrial organic influx during the Late Holocene was considerably lesser. This event is well etched in most of the wetland sedimentary sequences, and is represented by a layer with very little microfloral content and few subfossil woods. The period around 2.5 k yrs BP is considered to have witnessed decreased rainfall, and aridity over large areas of Asian Summer Monsoon. The lower terrestrial organic matter input is attributed to reduced rainfall - a consequence of weak monsoon over the period. The reduced river water influx and the sediment supply to the wetlands have had a significant impact on the preservation of vegetal matter. Considerable areas and sediments of the wetlands have been exposed to subaerial oxidation by 2.5 k yrs BP. The accompanied ecological shift, and the subsequent environmental set up dominated by silty-sand/sand sequence that hardly favored deposition and preservation of tree trunks and organic matter to the extent of Early-Middle Holocene. However, this event is followed by restoration of the monsoon to the present level, but not to the extent of 8000–5000 yrs BP monsoon peak. A generalized cross section distinguishes the Pleistocene and Holocene sediments that entomb the wood remains along Kollam - Ernakulam coast (Figure 5). Palynological and palynodebris analysis of Pangod, West Kallada and Munrothuruthu areas too compliment the palaeoclimate signatures of fossil woods that show the Holocene Climatic phases (Figure 6).

Figure 5. Cross sectional view Pleistocene and Holocene sediments along Kollam - Ernakulam coast.

Figure 6. Correlation of Pangod, West Kallada and Munrothuruthu areas showing Holocene Climatic phases.

Table 4. Radiocarbon dates of the Holocene woods/peat/sediments of the wetlands of Kerala.

Fossil logs and vegetation types

Of the twenty seven investigated wood samples, twenty five were assigned to their natural taxa based on anatomical features of the nearest relative living form/comparable modern analogues and the ethnobotanical information provided by the local inhabitants (Table 3). Two of them could not be assigned to their NRL due to poor preservation. However, they were found to be dicotyledonous woods. Leaving aside three species of the mangrove, the rest of them belong mostly to the tropical wet evergreen forest types comprising Artocarpus, Calophyllum, Canarium, Dipterocarpus, Diospyros, Shorea, Toona, Holigarna, Hopea, Mimusops, Neolamarckia, Sarcostigma and Terminalia. This indicates their prevalence occurs as a result of high atmospheric humidity, warm temperature and rainfall in the range of 200–750 cm. Semi-evergreen and moist deciduous forms that require annual rainfall <200 cm are Careya, Dalbergia, Ptercocarpus and Lannea. The families of Anacardiaceae, Burseraceae, Clusiaceae, Dipterocarpaceae, Ebenaceae, Fabaceae, Icacinaceae, Moraceae and Rubiaceae are well represented in the wood assemblage. The genus Artocarpus is ubiquitous in the wetlands of Kerala and three species, viz., A. hirsutus, A. heterophyllous, and A. lacucha have been identified. Investigations have pointed out that the drowned boat excavated by archeologists at Thaikkal was constructed from the wood of A. hirsutus, which is endemic to Western Ghats, and is still grown along the Kerala coast [47], [49]. Dipterocarpus and Hopea of the Dipterocarpaceae and Calophyllum of the Clusiaceae also recur often in the studied wood samples. The other significant evergreen taxa are Canarium, Holigarna, Dalbergia and Diospyros (endemic to Western Ghats from Central and South Sahyadri). Many of these taxa have also been recorded in the Neogene sediments of southwestern India, suggesting therein that humid climate and high rainfall prevailed largely during the Early to Middle Holocene period. However, the disappearance of some Dipterocarps like Anisoptera, Dryobalanops as well as certain species of Shorea, Hopea, and other evergreen taxa along Kerala coast in the modern flora reflect the response and sensitiveness of these taxa to reduced rainfall due to weak monsoon system since Neogene [50]. Species that thrive under relatively less moist conditions (semi-evergreen type and moist deciduous) include Careya, Lannea, Pterocarpus, Tectona and Terminalia. These usually represent the youngest stratigraphical sequence of the studied wetlands, and may have been transported from hinterland by rivers. The occurrence of tree trunks of Rhizophora sp. cf. R. mangle, Rhizophora mucronata, and Sonneratia apetala,core mangrove species found in the tidal creeks and littoral forests, indicates a proximity to sea and sea level response to vegetation during the Mid-Holocene transgression, and subsequent burial [14], [51].

The taphonomy of woods of the above species at three different intervals: Early Holocene (10.0–6.5 k yrs BP), Middle Holocene (6.5–5.0 k yrs BP) and Late Holocene (<3.6 k yrs BP) has been found to be related to the hydrodynamic regime of the area. This classification emerges in consequence to the impact of the SW and NE monsoon systems where the climatic gradients, rainfall, duration of the dry season and temperature, determine the structural and floristic changes, enabling their distinction into different forest types. The floristic composition of preserved woods of Early to Mid - Holocene indicates a dominance of wet evergreen forest type. The dominance is influenced by the rainfall gradients, as the region is closer to the western slopes of the Western Ghats that are exposed directly to rain-bearing winds of the south-west monsoon. Therefore, the thriving of a dense forest cover during Early to Middle Holocene is mainly attributed to the heavy rainfall. However, it merits emphasis that the present wetlands, where fossil wood and sub fossil logs occur, are not ecologically suited for such a forest cover. It is apparent that such places were not wetlands when the forest cover existed in recent past. The almost catastrophic flood that occurred due to the intensive Monsoon, might have caused the destruction of the thick forest cover, and its subsequent burial in the wetlands. At the beginning of monsoon, intensification around 9000 yrs BP, the sea level was much lower. This would have led to higher stream velocity and hence extensive erosion of river channels and other environments. The sea level rose to the present level around 7000 yrs BP. Subsequently it would have risen further by 2–4 m until about 5000 yrs BP, and then receded to the present level. A sea level higher than that at present, coupled with the excessive rainfall might have paved the way for uprooting the forest vegetation, and burial of the same at its place of growth, or in other sheltered areas further downstream. Besides, the abundant supply of the excessive water that flowed down the streams would supply enough sediments to bury the wood in low lying and flooded land, before its exposure to oxidation by atmosphere, or by shallow water columns. Sediments in any environment can be of geological significance only when they are preserved by subsidence of the environment. In the absence of subsidence, particularly of shallow water bodies, the place of deposition would soon be filled up, and the sedimentary column would be of insignificant thickness. The wetlands and the rivers in the lowland seas subsided in large segments after the wood bearing sediments were deposited. This subsidence was probably accomplished through an eastward tilt of the coast that caused the subsidence of the wetlands, and preservation of these sediments with the fossil wood and sub fossil logs. The accrued data has been used to draw a modified version of the Holocene Climate models [48], [52] (Figure 7).

The accumulation of fossil woods and sub-fossil logs of large forest trees in a large quantity in the South Kerala coastal plains holds immense significance, when the ecology, depositional environment and geological history of the Late Quaternary is inferred. It is likely that extensive floods after the transgression caused massive destruction of the coastal forests, which in turn buried the huge tree trunks. Evidence of such large scale flooding and scouring in the Early Holocene from monsoon Asia has already been dealt with [53]. These fossils are found to be in situ accumulation, as roots and branches are found intact in many cases. In some cases even the bark was found intact. This phenomenon constitutes sufficient evidence of a fast burial. Such evidence indicates that these fossils did not drift too far from their place of origin, and hence are of considerable significance in determining vegetation characteristics in the catchments. The profuse growth of wet evergreen, semi-evergreen vegetation comprising many woody species, viz., Dipterocarpus, Hopea, Shorea, Diospyros, Canarium, Artocarpus, Mangifera, Pterocarpus, Toona, Leea and Cullenia (fossil pollen only) might have been the result of the intensified Asian summer monsoon.

The record of palynoflora and the subfossil logs from the peatland, wetland, palaeoestuaries, river banks and paleobeach ridges from the southwestern part of India undoubtedly prove that the entire land, west of Sahyadri up to the coast, was densely forested towards Early to Mid- Holocene (9.0–5.5 k yrs BP), and this scenario has been attributed to prevailing heavy rainfall [54]. There is a convergence over the view that the beginning of Holocene witnessed an intensification of monsoon which lasted over a few thousand years [55]. During the period from 8.5 to 5.5 k yrs BP, Asia witnessed much higher precipitation, which is now universally accepted as Holocene climatic optimum. The rainfall of this period is often stated to be three times higher than the present rate [56][57]. Further, it is interesting to note that this excessive precipitation took place not only in Asia, but also all along the path of summer monsoon. It is during this period that the deserts of Sahara too witnessed development of large lakes and prolific forest cover [58] and the Chad Lake in Central Africa expanded to three times its present state [59]. Therefore, the high rainfall during the Early Holocene was congenial for luxuriant forestation over the entire stretch of land from Sahyadri to the present coast and even beyond the coast in Peninsular India. Prior to 7000 yrs BP, there were hardly any wetlands, except isolated depressions and lakes, while the sea level was much lower [60], [61]. Even the present backwater systems of Vembanad have their origin during the regressive phases (5.0 k–3.0 k yrs BP) of the Holocene [62]. In fact, the high rainfall coupled with a rising sea level must have inundated >75% of the coastal plains until 6.0 k yrs BP. The excessive rainfall, higher than the present rate, was responsible for the development of thick forest and forest swamp cover virtually all over the terrain, including the coastal plains, midland and highlands. The forest cover seems to have extended even up to the present coastline towards Mid-Holocene [13], [46].

The excessive rainfall could have caused some inundation of the low-lying lands. The effect of this flooding was compounded by the sea level rise during the 7.0–6.0 k yrs BP. This converted 75% of the coastal plain land to a veritable lagoon - lake system; this was virtually an abrupt termination of the forest ecosystem. The lower energy level in these newly developed wetlands paved the way for the burial of the hard wooded trees, and decay and disintegration of other forest vegetation in various sediments. However, the condition of preserved wood depends on the type of sediment, and the time lag between the falling of the trees and the time of complete burial. Clay-rich sediment preserves wood and other vegetal remains better as compared to sands and other coarser sediment types. Since the sediment supply was copious due to the high rate of erosion of uplands, the hard wooded trees were entombed in various strata. The woods of relatively softer species must have decayed and became a part of the normal sediment. This is the reason for the high carbon content in the black/grey clay - silty clay sediments that preserves the wood in all the major wetlands of southwestern India.

Palaeoecological and palaeoclimate potential of fossil logs

During the last glacial event of about 18,000 yrs BP, the Arabian Sea experienced a lowering of its level to the tune of 100–120 m. The consequences of this glacial event have been of immense significance to coastal dynamics, as the landforms responded to a considerable extent such that it caused a submergence of forested areas. It also led to the development of major wetland systems including the ‘Ramsar sites’ like Ashtamudi Lake, Sasthamkotta Lake and Vembanad Lagoon [13], [60]. The occurrence of subfossil logs and carbonized woods in the sediments associated with the wetlands and river bank deposits, demonstrates clearly that the forest vegetation succumbed almost entirely to the coastal dynamics of the Early Holocene (9–7.5 k yrs BP), such that a major part of vegetation cover was lost. The rivers became much more erosive in nature due to the increased gradient, and therefore caused widespread removal of the sediments already deposited, except in inland lakes and depressions. Also, vast areas along the marine shelf got exposed to sub-aerial action. From this lower level, the sea rose in stages and reached the present stage about 7000 yrs BP. The sea level further rose by about 4–6 m inundating the low-lying areas along the coast. This marine inundation was short-lived, and the sea withdrew in stages leaving behind the lagoons and wetlands, and the typical beach deposits even far inland [63]. The coast was re-established at the present place about 3500–4000 yrs BP. The record of this marine inundation is preserved in the landward extension of the offshore basin since a major part of it was subsiding. These movements are found to be at the rate of less than a millimeter to several millimeters per year and have influenced the sedimentary environments to a considerable extent [13]. Therefore, all these factors acting in tandem have produced a complex combination of factors having their effects on ecology and shaping the landforms.

The area of investigation presently receives annual rainfall ranging from 200 to >500 cm; about 70% of it is from the SW monsoon that is known as the Asian Summer Monsoon. Palynological data of the Kerala coast [63][65] too complements the accrued wood assemblage for palaeoclimate appraisal. Palynodebris and sediment characteristics suggest that there were two periods of abnormal high rainfall: one before the last glacial maximum (LGM), and another during Early Holocene [63], [64]. Many of the landforms in the coastal plains and several of the landforms in the hinterland are proved to contain partial to complete record of the period from Late Pleistocene to Holocene geo-environmental records [5]. Besides, there are positive indications that some of these terrains were sub-aerially exposed lands, which were once thickly forested. As a result of abnormal high rainfall coupled with tectonics, and the antecedent landform characteristics, the trees that grew there were possibly uprooted and buried in situ. This is particularly seen in the seasonal wetlands peripheral to the two main lagoonal water bodies namely the Vembanad and Kayamkulam lagoons [60], [64]. Further, the sediments in the freshwater lakes and abandoned river channels as well as in many river terraces contain tree trunks and carbonized wood. In many cases, the trees were buried under variable thickness of sediments before being destroyed by atmospheric processes due to the continued influx of terrigenous sediments under the transgressive phases of the sea. The events are well reflected in the δ13Corg and δ15N values estimated for the Pangod borehole core (Figure 8). Though, the upper yellowish brown mud rich layer records low organic carbon content (0.12%) compared to the underlying carbonaceous clay (6.05%), the δ13Corg exhibits an opposite trend [63]. This clearly indicates a gradual change in the depositional regime from terrestrial (δ13C −28.17‰) to marine entity (−19.56‰). The sediments of marine origin generally contain higher δ13Corg values, as a major part of it is derived from marine phytoplankton with higher δ13Corg values [66], [67]. The 14C age of upper most part of the organic carbon rich layer, just below the yellowish brown silt and clay layer at 3 m bgl is 14C dated as 5260±120 yrs BP. The low δ13Corg values of −28.17‰ to −26.88‰ show that the organic input in the carbonaceous clay is from C3 plants that flourished in the hinterlands during Early Holocene in Pangod section (Figure 9). The δ15N values vary from 3.92‰ to 8.85‰, with highest values recorded for the top yellowish brown layer [63]. The enhanced level of δ15 N isotope in the surface sediments as compared to the lower organic rich layers points to degradation and preferential consumption of lighter isotopes, and subsequent enrichment of heavier δ15N, a feature revealed elsewhere earlier [67]. Comparatively lower δ15N values in the layer also points to the preservation potential of organic matter, derived from terrigenous sources possibly aided by abnormal rainfall as recorded in Muthukulam core of SKSB during the Early to Mid-Holocene period [68].

Figure 8. Sand, mud, organic carbon and nitrogen contents in the sediments of Pangod quarry along with the concentration of δ15 N and δ 13C org.

Figure 9. Litholog of Pangod quarry showing carbonized and non carbonized subfossil logs at two different stratigraphic levels along with radiocarbon dates.

(a) Sub fossil log. (b) Partially carbonized wood. (c) Cross section showing indistinct growth rings. (d) Enlarged sector of growth rings. (e) and (f) Carbonized wood along with bark showing disintegration of wood.

The area of investigation had landforms supporting dense forest vegetation prior to Holocene transgression of sea, part of which could be preserved in suitable environments. This is a factor unique to southwestern coast of India and as such it has significance while addressing the palaeoclimate potential of the past forest remains. The proposed eco-geomorphological model displays the modifications of the coastline and associated landforms and forests since Early Holocene (Figure 10). However, the forest vegetation preserved is an infinitesimally small part of the one that thrived, but constitutes a major terrestrial palaeoclimate proxy from the Indian subcontinent. The evidence of plant macrofossil archive in the form of buried forests for Mid-Holocene Thermal Maximum in southwestern India is consistent with earlier reports on intensified and prolonged Asian Monsoon from other parts of west coast of India [55], [64], [65], [69][77], the Deccan Trap region [78], the Nilgiri hills in south India [57], [79], Ganga Plains [80], [81], Southern Oman [82], the northwestern Pacific [83] and the Indus delta [84]. In addition, there are numerous locales that were favorable for development of fresh water swamps and marshes and lagoonal and marginal marine mangrove swamps, which have been eventually converted into land in the recent past.

Figure 10. Ecogeomorpholgical evolution of southwest coast of India during Holocene.

(a) 9–8 k yrs BP; (b) 7–6 k yrs BP; (c) 3 k yrs BP.


Considering the scarcity of data for Holocene climate variability from lower latitudes, the accrued plant macrofossil and pollen proxies from the sedimentary archives of southwest India imply a Mid-Holocene Thermal Maximum related to intensified Asian Summer Monsoon, as recorded elsewhere. The degree of preservation and relative abundance of the wood remains varies considerably. The variation is dependent mainly on the lithological characteristics of the landform units, and the associated hydrological regimes under which they are deposited. All the fossil wood and sub-fossil logs retrieved, though insufficient for generalization, have yielded ages prior to the Holocene transgression (7.0–6.5 k yrs BP). Evidence suggests that the entire terrain west of Sahyadri (Western Ghats) was thickly forested during the Holocene climatic optimum (9.0–6.0 k yrs BP), when the region had witnessed a spell of heavy precipitation, ∼2–3 times more than the present. The flooding of the forest probably occurred as a result of intensified and prolonged Indian Summer Monsoon coupled with sea level rise to the present level or slightly above. This led to a drastic increase in the sluggishness of the river flows that resulted in abrupt flooding of the forest habitat, where the trees thrived. The increased fluvial sediment supply by the rivers enabled some of the tress to be buried and preserved as such, even before the formation of the major backwater systems including the Vembanad Lagoon. Further, the present ecology is unsuitable to support evergreen forests. It can be concluded then that coastal plains and associated landforms were covered by thick tropical evergreen forests, which got destroyed by flooding towards Middle Holocene. Though the plant macrofossil record represents a fraction in terms of taxa of the wet evergreen forests, their relative abundance along with dissolved carbon makes the wetlands and associated landforms one of the best carbon sinks in India. Besides acting as a carbon sink, the wetlands can also serve as a source of carbon, in that they may supply significant amounts of carbon to adjacent coastal ecosystems. This in turn, plays a vital role in coastal dynamics and overall productivity of the region. Thus, the wetlands of southwest India hold immense potential for palaeoecological reconstruction of long-term landscape, and vegetation changes.

Supporting Information

File S1.

Supporting Information. Text S1: Systematic description of sub fossil logs. Figure A: Anatomical details of Artocarpus sp. cf. A. lacucha Buch-Ham. Figure B: Anatomical details of Careya arborea Roxb. Figure C: Anatomical details of Diospyros sp. cf. D. bourdilloni Brandis. Figure D: Anatomical details of Dipterocarpus sp. cf. D. indicus Teysm. ex Miq. Figure E: Anatomical details of Neolamarckia sp. cf. N. Cadamba. Figure F: Anatomical details of Rhizophora sp. cf. R. mangle L.




The authors thank the Directors, Agharkar Research Institute, Pune; Centre for Earth Science Studies (CESS), Thiruvananthapuram and Birbal Sahni Institute of Palaeobotany (BSIP), Lucknow for extending facilities, encouragement and logistic support. KPNK and RBL acknowledge the CSIR for ES and SRA awards respectively. Dr. B. Sekar of the Radiocarbon Laboratory, BSIP provided necessary input while providing the calibrated dates of all the dated woods, peat and associated sediments. Prof. M. Sivadasan, Department of Botany & Microbiology, College of Science, King Saud University, Saudi Arabia provided vernacular names of the subfossil logs. The comments of anonymous reviewers are appreciated.

Author Contributions

Conceived and designed the experiments: NKPK DP RBL MKN. Performed the experiments: DP JSG RS AS. Analyzed the data: KPNK DP RBL RS. Contributed reagents/materials/analysis tools: DP JS RS AS. Wrote the paper: NKPK DP RBL. Eco-geomorphological model: NKPK DP RBL.


  1. 1. Nair KM, Padmalal D (2003) Quaternary sea level oscillations, geological and geomorphological evolution of South Kerala Sedimentary Basin. Project Report, DST, Govt. of India (Unpublished).
  2. 2. Wils THG, Sass-Klaassen UGW, Eshetu Z, Brauning A, Gebrekirstos A, et al. (2010) Dendrochronology in the dry tropics: the Ethiopian case. Trees doi:10.1007/s00468-010-0521-y. Springer.
  3. 3. Thrivikramji KP, Joseph S, Narayana AC (2007) Coastal wetlands of Kerala: Origin and evolution. Proceedings of Kerala Environment Congress 2007, Trivandrum: 36–48.
  4. 4. CESS (1984) Resource Atlas of Kerala. Centre for Earth Science Studies, Thiruvananthapuram, 39p.
  5. 5. Nair KM, Padmalal D, Kumaran KPN (2006) Quaternary geology of South Kerala Sedimentary basin- An outline. Journal of Geological Society of India 67: , 165–179.
  6. 6. Milliman JD, Syvitski JPM (1992) Geomorphic/tectonic control of sediment discharge to the ocean: the importance of small mountains rivers. Journal of Geology 100: 525–544.
  7. 7. Metcalfe CR, Chalk L (1950) Anatomy of the dicotyledons. 2 vols. Clarendon Press: Oxford. 1500p.
  8. 8. Pearson RS, Brown HP (1932) Commercial Timbers of India. I, 2. Calcutta.
  9. 9. Ilic J (1991) CSIRO atlas of hardwoods. Bathurst, Australia: Crawford House Press, 525.
  10. 10. Gupta S (2007) Atlas of Indian Hardwoods, their photomicrographs, and anatomical features. I. Forest Research Institute, Indian Council of Forestry, Research & Education, Dehradun.
  11. 11.;jsessionid=ae13cbe19093e8a61b6b904e6c73 Accessed 27 February, 2013.
  12. 12. IAWA Committee (1989) IAWA list of microscopic features for hardwood identification. In: Wheeler EA, Baas P, Gasson P, IAWA Bull. ISSN: New Series 10(3): 219–332.
  13. 13. Nair KM, Kumaran KPN, Padmalal D (2009) Tectonic and hydrologic control on Late Pleistocene - Holocene landforms, palaeoforest and non forest vegetation: Southern Kerala, Project Completion Report, Kerala State Council for Science, Technology and Environment, Thiruvananthapuram, pp. 1–83.
  14. 14. Kumaran KPN, Limaye Ruta B, Padmalal D (2013) Mangrove Responses to Climate Change along the Southwestern Coast of India during Holocene: Evidence from Palynology and Geochronology. In: Sundaresan J, Sreekesh S, Ramanathan AL, Sonnenschein L, Boojh R, Climate Change and Island and Coastal Vulnerability. Springer, Dordrecht, The Netherlands and Capital Publishing Company, New Delhi, India, pp. 217–238, doi:10.1007/978-94-007-6016-5_15.
  15. 15. Janssonius HH (1926) In: Moll JW, Janssonius HH, Micrographie des Holzer der auf Java vorkommenden Baumarten. Vol IV. E.J. Brill, Leiden.
  16. 16. Anonymous (1958) Indian Woods - their identification, properties and uses-1. Manager of publications, Govt of India, Delhi/Dehradun.
  17. 17. Chowdhury KA, Ghosh SS (1958) Indian Woods - Their Identification, Properties and Uses. I. Manager of Publications. Govt. of India. Delhi.
  18. 18. Kribs DA (1959) Commercial foreign woods on the American market. Pennsylvania.
  19. 19. Hayashi S, Kishima T, Lau LC, Wong TM, Menon PKB (1973) Monographic Atlas of Southeast Asian Timbers. Wood Research Institute, Kyoto, Japan.
  20. 20. Purkayastha SK, Juneja KBS, Husain Kazmi SM (1976) Anatomy of more important Andaman commercial timbers (With notes on their supply, Properties and Uses). Indian Forest Rec. 2(1): , 1–48. Indian Council of Forestry Research and Education, Dehra Dun.
  21. 21. Miles A (1978) Photomicrographs of world woods. Building Research Establishment, Princes Risborough Laboratory; London: Her Majesty's Stationery Office. 233.
  22. 22. Purkayastha SK (1982) Indian Woods–Their Identification, Properties and Uses. IV. Manager, Govt., of India Press, Controller of Publications, Delhi.
  23. 23. Kazmi HSM (1982) Family Ebenaceae. Their Identification, Properties and Uses. IV: . 122–131, Manager, Govt., of India Press, Controller of Publications, Delhi.
  24. 24. Negi BS, Raturi RD (1992) Anatomy of some of the West Coast timbers. Indian Forest Records 4(1): 1–62.
  25. 25. Raturi RD, Chauhan L, Gupta S, Vijendra Rao R (2001) Indian Woods –Their Identification, Properties and Uses. VI. Indian Council of Forestry Research and Education, Dehradun.
  26. 26. Mabberley DJ (2005) The Plant Book. A Portable Dictionary of Vascular Plants. Cambridge University Press, Cambridge.
  27. 27. Lens F, Baas P, Jansen S, Smets EA (2007) A search for phylogenetically informative wood characters within Lecythidaceae s.l. American Journal of Botany 94(4): 483–502.
  28. 28. Koek-Noorman J (1977) Systematische Holzanatomie einiger Rubaceen. Berichete der Deutschlands Botanisches Gesselschaft 90: 183–190.
  29. 29. Purkayastha SK, Chauhan L (1982) Family Rubiaceae. Indian Woods –Their Identification, Properties and Uses. IV: , 66–94.
  30. 30. Jansen S, Robbrecht E, Beeckman H, Smets E (1997) Wood anatomy of the predominantly African representatives of the tribe Psychotrieae (Rubiaceae-Rubiodeae). IAWA Journal 18(2): 169–196.
  31. 31. Lim SC, Gan KS, Thi BK (2005) Identification and utilization of lesser known commercial timbers in Peninsular Malaysia. 4. Kelempayan, Melembu, Membuloh and Mempari. Timber Tech. Bull. 32: 1–8.
  32. 32. Srivastava R (1998) Fossil wood of Artocarpus from Warkalli Formation of Kerala Coast, India. Phytomorphology 48(4): 391–397.
  33. 33. Guleria JS, Sekar B, Nair MV (2004) A report on the identification of wood samples of ship excavated from Thaikkal, Alappuzha District, Kerala. Science & Culture 70: 169–170.
  34. 34. Srivastava R, Guleria JS (2006) A Catalogue of Cenozoic (Tertiary) plant Megafossils from India (1989–1005). B.S.I.P. Diamond Jubilee Spec-Publ:1–76.
  35. 35. Awasthi N, Srivastava R (1992) Additions to the Neogene flora of Kerala Coast, India. Geophytology 20(2): 148–154.
  36. 36. Shukla A, Guleria JS, Mehrotra RC (2012) A fruit wing of Shorea Roxb. from Early Miocene sediments of Kachchh, Gujarat and its bearing on palaeoclimatic interpretation. Journal of Earth System Science 121(2): 195–201.
  37. 37. Ridsdale CE (1978) A revision of tribe Naucleeae s.s (Rubiaceae). Blumea 24: 307–366.
  38. 38. Razafimandimbisan SG, Bremer B (2002) Phylogeny and classification of Naucleeae s. L. (Rubiaceae) inferred from molecular (ITS, rBCL and tRNT-F) and morphological data. American Journal of Botany 89(7): 1027–1041.
  39. 39. Robbrecht E (1988) Tropical woody Rubiaceae. Charateristic features and progressions. Contributions to subfamilial classification. Opera Bot. Belg. 1: 1–271.
  40. 40. Wong KM (1989) Rubiaceae- In: F.S.P Ng, Tree flora of Malaysia. A manual for Foresters. 4: : 381–382. FIRM and Ministry and Primary Industries, Malaysia.
  41. 41. Kathiresan K, Bingham BL (2001) Biology of mangroves and mangrove ecosystems. Advances in Marine Biology 40: 81–251.
  42. 42. Awasthi N, Mehrotra RC (1995) Oligocene flora from Makum Coalfield, Assam, India. Palaeobotanist 44: 157–188.
  43. 43. Worbes M (1985) Structural and other adaptations to long-term flooding by trees in Central Amazonia. Amazonia 9: 459–484.
  44. 44. Worbes M (1989) Growth rings, increment and age of trees in inundation forests, savannas and a mountain forest in the Neotropics. IAWA Bull. n.s. 10: 109–122.
  45. 45. Guleria JS, Srivastava R, Kumar BA, Satheesh R (2010) Holocene woody vegetation and environment of Meenachil River Basin, Kottayam District, Kerala, India. Geophytology 38: 1–14.
  46. 46. Kumaran KPN, Nair KM (2005) Tracing palaeoclimatic significance in fossil woods and sub-fossil logs of Kerala, Southwestern India. Pages News 13: 15–17.
  47. 47. Cherian PJ, Raviprasad GV, Koushik D, Dinesh KR, Selvakumar V, et al. (2009) Chronology of Pattanam: a multi-cultural port site on the Malabar Coast. Current Science 97(2): 236–240.
  48. 48. Prell WL, Kutzbach JE (1987) Monsoon variability over the past 150,000 years. Journal of Geophysical Research 92: 8411–8425.
  49. 49. Nair MV, Selvakumar V, Gopi PK (2004) Excavation of a Unique Sailboat at Kadakkarappally. Current Science 86(5): 709–712.
  50. 50. Awasthi N, Srivastava R (2005) Neogene flora of Kerala coast and its palaeoecological and phytogeographical implications. In: Reddy, R et al. Gleanings in Plant Sciences, Prof. Ramanujam Commem. Vol. Dttasons, Nagpur, pp. 265–277.
  51. 51. Limaye, Ruta B, Kumaran KPN (2012) Mangrove vegetation responses to Holocene climate change along Konkan coast of South-western India. Quaternary International 263: 114–128.
  52. 52. Kale VS (2005) Late Quaternary Climate changes: The Indian Scene. Key note address. Proceedings of 18th Convention of IGI & National Seminar on Quaternary Climate Changes: 15–28.
  53. 53. Kale VS, Gupta A, Singhvi AK (2004) Late Pleistocene - Holocene palaeohydrology of monsoon Asia. Journal of Geological Society of India 64: 403–417.
  54. 54. Kumaran KPN, Ruta B Limaye, Padmalal D (2012) Holocene vegetation dynamics in southwestern India: palynological and palaeoecological appraisal of terrestrial and marine archives. Japanese Journal of Palynology 58 (Special issue XIII-IPC & IX–IOPC 2012) 117.
  55. 55. Singhvi AK, Kale VS (2009) Palaeoclimate Studies in India Last Ice Age to the Present. In: IGBP-WCRP-SCOPE-Report Series, 4, Indian National Science Academy, New Delhi: 1–34.
  56. 56. Bryson RA, Swain AM (1981) Holocene variations of monsoonal rainfall in Rajasthan. Quaternary Research 16: 135–145.
  57. 57. Geetha R, Sukumar R, Ramesh R, Pant RK, Rajagopalan G (1997) Late Quaternary vegetational and climatic changes from tropical peats in southern India - An extended record up to 40,000 years BP. Current Science 73: 60–63.
  58. 58. Petit-Maire N (1995) Changes and the tropical/semiarid belt in the North Africa. Journal of Coastal Research. Spec. Issue, 17, Holocene cycles, climate, sea levels and sedimentation. pp. 87–92.
  59. 59. Ghienne JF, Schuster M, Bernard A, Duringer P, Brunet M (2002) The Holocene Giant Chad revealed by digital elevation models. Quaternary International 87: 81–85.
  60. 60. Padmalal D, Kumaran KPN, Nair KM, Baijulal B, Limaye RB, et al. (2011) Evolution of the coastal wetland system of SW India during the Holocene: Evidence from marine and terrestrial archives of Kollam coast, Kerala. Quaternary International 237: 123–139.
  61. 61. Padmalal D, Kumaran KPN, Nair KM, Limaye Ruta B, Vishnu Mohan S, et al. (2013) Consequences of Sea level and Climate changes on the Morphodynamics of a Tropical Coastal Lagoon during Holocene: An evolutionary model. Quaternary International doi:10.1016/j.quaint.2013.12.018.
  62. 62. Ajaykumar B, Mahesh M, Shylesh Chandran MS, Jayasooryan KK, Unni KS, et al. (2010) Mechanism and geochemical implications of the Late Holocene organic carbon burial event: an example from the southwest coast of India. Environment Earth Science 59: 1409–1416.
  63. 63. Padmalal D, Nair KM, Kumaran KPN, Sajan K, Vishnu Mohan S, et al. (2013) Climate and Sea Level Changes in a Holocene Bay Head Delta, Kerala, Southwest Coast of India. In: Sundaresan J, Sreekesh S, Ramanathan AL, Sonnenschein L, Boojh R, Climate Change and Island and Coastal Vulnerability. Springer, Dordrecht, The Netherlands and Capital Publishing Company, New Delhi, India, pp. 191–208. doi:10.1007/978-94-007-6016-5_13.
  64. 64. Nair KM, Padmalal D, Kumaran KPN, Sreeja R, Limaye RB, et al. (2010) Late Quaternary evolution of Ashtamudi - Sasthamkotta lake systems of Kerala, South west India. Journal of Asian Earth Sciences 37: 361–372.
  65. 65. Ajaykumar B, Sreedharan K, Mahesh M, Joby P, Thomas AP, et al. (2012) Evaluation of the Holocene environmental changes of the southwest coast, India: A palaeopalynological approach. Journal of Earth System Science 121(4): 1093–1103.
  66. 66. Fischer G (1991) Stable carbon isotope ratios of plankton carbon and sinking of organic matter from the Atlantic sector of the Southern Ocean. Marine Chemistry 35: 581–596.
  67. 67. Jennerjahn TC, Ittekkote V, Arz HW, Berling H, Patzold J, et al. (2004) Asynchronous terrestrial and marine signals of climate change during Henrich events. Science 306: 2236–2239.
  68. 68. Jayalakshmi K, Nair KM, Kumai H, Santosh M (2004) Late Pleistocene-Holocene palaeoclimatic history of the South Kerala Basin, Southwest India. Gondwana Research 7(2): 585–594.
  69. 69. Borole DV, Rao KK, Krishnamurthy RV, Somayajulu BLK (1982) Late Quaternary faunal change in coastal Arabian Sea sediments. Quaternary Research 18(2): 236–239.
  70. 70. Ramesh R (2001) High resolution Holocene monsoon records from different proxies: An assessment of their consistency. Current Science 81(11): 1432–1436.
  71. 71. Kale VS (2007) Fluvio–sedimentary response of the monsoon-fed Indian rivers to Late Pleistocene - Holocene changes in monsoon strength: reconstruction based on existing 14C dates. Quaternary Science Reviews 26: 1610–1620.
  72. 72. Thamban M, Rao VP, Schneider RR, Grootes PM (2001) Glacial to Holocene fluctuation in hydrography and productivity along the western continental margin of India. Palaeogeography, Palaeoclimatology, Palaeoecology 165(1–2): 113–127.
  73. 73. Kumaran KPN, Nair KM, Shindikar M, Limaye RB, Padmalal D (2005) Stratigraphical and palynological appraisal of the Late Quaternary mangrove deposits of the west coast of India. Quaternary Research 64: 418–431.
  74. 74. Nair KM, Kumaran KPN (2006) Effects of Holocene climatic and sea level changes on ecology, vegetation and landforms in coastal Kerala. Project Completion Report, Kerala State Council for Science, Technology and Environment, Thiruvananthapuram, pp. 1–41.
  75. 75. Kumaran KPN, Limaye RB, Nair KM, Padmalal D (2008) Palaeoecological and Palaeoclimate potential of subsurface palynological data from the Late Quaternary sediments of South Kerala Sedimentary Basin, southwest India. Current Science 95: 515–526.
  76. 76. Kumaran KPN, Limaye Ruta B, Padmalal D (2012) India's fragile coast with special reference to Late Quaternary Environmental Dynamics. Proceedings of Indian National Science Academy 78(3): 343–352.
  77. 77. Kumaran KPN, Limaye Ruta B, Punekar Sachin A, Rajaguru SN, Joshi SV, et al. (2013) Vegetation response to South Asian Monsoon variations in Konkan, western India during the Late Quaternary: Evidence from fluvio-lacustrine archives. Quaternary International 286: 3–18.
  78. 78. Pawar NJ, Kale VS, Atkinson TC, Rowe PJ (1988) Early Holocene waterfall tufa from semi-arid Maharastra Plateau, India. Journal of Geological Society of India 32: 513–515.
  79. 79. Sukumar R, Ramesh R, Pant RK, Rajagopalan G (1993) A δ13C record of late Quaternary climate change from tropical peats in southern India. Nature 364: 703–706.
  80. 80. Sharma S, Joachimski M, Sharma M, Tobschall HJ, Singh IB, et al. (2004) Late glacial and Holocene environmental changes in Ganga plain Northern India. Quaternary Science Reviews 23(1–2): 145–159.
  81. 81. Sinha R, Sarkar S (2009) Climate-induced variability in the Late Pleistocene-Holocene fluvial and fluvio-deltaic successions in the Ganga plains, India. Geomorphology 113: 173–188.
  82. 82. Fleitmann D, Burns SJ, Mudelsee M, Neff U, Kramers J, et al. (2003) Holocene forcing of the Indian monsoon recorded in a stalagmite from southern Oman. Science 300: 173.
  83. 83. Morimoto M, Kayanne H, Abe O, McCulloch MT (2007) Intensified mid-Holocene Asian monsoon recorded in corals from Kikai Island, subtropical northwestern Pacific. Quaternary Research 67(2): 204–214.
  84. 84. Ansari MH, Vink A (2007) Vegetation history and palaeoclimate of the past 30 kyr in Pakistan as inferred from the palynology of continental margin sediments off the Indus Delta. Review of Palaeobotany and Palynology 145(3–4): 201–216.
  85. 85. Arun PR (2006) Sedimentology and hydrogeochemistry Aruvikkara and Peppara reservoir basins, Kerala India. Ph D thesis, Cochin University of Science and Technology, Kochi, 241p.
  86. 86. Tomalin V, Selvakumar V, Nair MV, Gopi PK (2004) The Thaikkal-Kadakkarapally boat: an Archeological example of Medieval shipbuilding in the western Indian Ocean. International Journal of Nautical Archaeology 33: 253–263.
  87. 87. Meena D, Nagarajan B, Jesubalan D (2012) Future prospects for the critically endangered medicinally important species, Canarium strictum Roxb. A review. International Journal of Conservation Science 3: 231–237.
  88. 88. Troup RS (1986) Indian Woods and their uses, Soni Reprints Agency,Delhi.
  89. 89. Padmalal D, Maya K, Narendrababu K, Baijulal B (2010) Environmental appraisal and sand auditing of Manimala river, Kerala, India. Report- Government of Kerala, Centre for Earth Science Studies, Thiruvananthapuram. 163p.
  90. 90. Rajimol TR, Ramdev PR, Baburaj B, Maya K, Padmalal D (2013) Coastal plain rivers of Thiruvananthapuram district, Kerala–River characteristics, human interventions and management strategies. 23rd Kerala Swadeshi Science Congress, MG University, Kottayam, pp. 336–340.
  91. 91. Rajendran CP, Rajagopalan G (1989) Narayanaswamy (1989) Quaternary geology of Kerala: evidence from radiocarbon dates. Journal of Geological Society of India 33: 218–222.
  92. 92. Shajan KP (1998) Studies on Late Quaternary sediments and sea level changes of central Kerala coast, India. Ph.D. Thesis, Cochin University of Science and Technology, Kochi.
  93. 93. Agarwal DP, Gupta SK, Kusumgar S (1970) Radiocarbon dates of Quaternary samples. Current Science 30: 219–222.
  94. 94. Padmalal D, Maya K, Naredra Babu K, Mini SR (2005) Tile and brick clay mining and related environmental problems in the Chalakudy basin, Central Kerala - Case study IV. In: Nair KN, Chattopadhyay S, Water resources of Kerala: Issues and case studies). Kerala research programme on local level development, CDS, Thiruvananthapuram. pp. 80–82.
  95. 95. Pawar SD, Venkataramana B, Mathai T, Mallikarjuna C (1983) Systematic geological mapping around Cherthalai–Vaikom–Alleppey–Kottayam and Panthalam in parts of Alleppey and Kottayam districts/Kerala state. Report - Geological Survey of India.
  96. 96. Soman K (2002) Geology of Kerala. Geological Society of India, Bangalore,335p.