Late-Holocene climate in the Lower Narmada valley,Gujarat,western India,inferred using sedimentary carbon and oxygen isotope ratios

Abstract

We report the late-Holocene climate and vegetation of the Lower Narmada valley, Gujarat, western India as inferred from the stable carbon and oxygen isotopic composition (δ¹³C and δ¹⁸O) of sedimentary carbonates and the associated organic carbon. The alluvial plain from the surface to a depth of ~2 m consists of late-Holocene sediments, deposited during last ~3 kyr, probably by large historic and paleofloods. δ¹³C of both carbonates and organic carbon in the sediments suggest that climate was subhumid throughout the late Holocene (~ 3 kyr) as it is today, and the vegetation was of mixed C₃-C₄ type with little change in their relative proportions. The modern vegetation mostly comprises shrubs and scattered woody plants (C₃) with a little grass (C₄) in some places. The recent change in vegetation is attributable to anthropogenic disturbance: the natural grasslands (C₄) are replaced by shrubs and woody plants (C₃). Two comparatively drier events at ~2.1 ka and ~1.3 ka are observed, consistent with widespread proxy paleoclimatic records and are attributed to a weaker southwest monsoon rain. Marine influence on the isotopic compositions is observed in a cliff section of the Narmada estuary throughout its depositional history of ~4.2 kyr. The radiocarbon ages of the sediments are of the order of decades to a century at the surface and increase almost linearly with depth, being 1000–3000 yr at ~100 cm and 3200–4200 yr at ~200 cm.

Keywords

carbonate Holocene organic carbon paleoclimate radiocarbon stable isotopic composition

Introduction

The Quaternary sediments of the Gujarat alluvial plains, west India are of marine, fluvial and aeolian provenance (Merh and Chamyal, 1997). Until recently, their origin (Chamyal et al., 2003), geomorphic evolution in relation to neotectonism (Chamyal et al., 2002) and sediment characteristics (Bhandari et al., 2005) were studied extensively. However, the time frame of these sediments, especially the top few meters is not well constrained. Soils, mainly vertisols, entisols and inceptisols, have developed over varied landforms in the region (Raj, 2007). These deposits are useful for deciphering the paleoclimate of the region (e.g. Khadkikar et al., 2000; Prasad et al., 2007) the recent past of which is not well explored.

Stable isotopic composition of both carbonates and organic carbon (δ¹³C_carb, δ¹⁸O_carb and δ¹³C_oc) in soils and sediments have been widely used to infer regional paleoclimate and paleovegetation (Achyuthan et al., 2007; Bowman et al., 2004; Cerling, 1984; Laskar et al., 2010; Leier et al., 2009; Sukumar et al., 1993). Secondary soil carbonates, often found in arid, semi-arid and subhumid regions are a result of the soil-forming (pedogenesis, Khadkikar et al., 1998; Zaleha, 1997) as well as soil water processes (Khadkikar et al., 1998; Pimentel et al., 1996). They are formed mainly during the warm and dry season when evaporation or evapotranspiration exceeds precipitation making the vadose zone soil water supersaturated relative to the dissolved carbon species (Breecker et al., 2009). Usually δ¹³C_carb carries the imprint of the δ¹³C value of soil CO₂, which is in turn controlled by the δ¹³C of the local plant cover (proportion of C₃/C₄ biomass) and soil respiration. Figure 1 shows the typical values of δ¹³C for various types of plant covers, the associated soil organic matter, soil CO₂ and soil carbonates. Similarly, δ¹³C_carb and δ¹⁸O_carb in authigenic deep sediments can be used to infer paleovegetation and paleoclimate, provided that they are not altered after deposition (e.g. Achyuthan et al., 2007; Alam et al., 1997; Amundson et al., 1988; Dhir et al., 2004; Goodfriend and Magaritz, 1988; Leier et al., 2009; Pendall and Amundson, 1990). δ¹³C_oc in soils and sediments is directly controlled by the type of vegetation at the time of pedogenesis (Balesdent et al., 1993; McPhearson et al., 1993) which is mainly governed by the ambient climate (Deines, 1980; Kohn, 2010; Osmund, 1978; Quade and Cerling, 1995; Weiguo et al., 2003). Additionally, a carbon isotopic fractionation of up to 2–5‰ may occur during the conversion of fresh organic matter into humic substances and during microbial decomposition (Accoe et al., 2002; Quade and Cerling, 1995; Wynn et al., 2006). However, because of refractory nature of soil organic matter, δ¹³C_oc is widely used as a paleovegetation proxy. In the tropics, δ¹⁸O_carb is governed by the rainwater δ¹⁸O, the latter depends on the amount of rainfall called ‘amount effect’ (Dansgaard, 1964). Therefore, water residing in the vodose zone or below retains the average isotopic signature of the rain water after some evaporation at the surface. Hence ideally, the δ¹⁸O_carb may be used as a proxy of the δ¹⁸O of the local meteoric water when carbonates were formed.

Figure 1.

Typical values of carbon isotopic composition (δ¹³C) in various vegetation types, underlying soils, fossil fuel and atmosphere.

Based on earlier studies, the climate in the mainland Gujarat was found to be humid during the Late Pleistocene (Badam et al., 1986; Kale et al., 2003; Laskar et al., 2010) and the pre-industrial era climate was established prior to ~3 ka (Prasad et al., 2007). However, for the last 3 ka paleoclimate data are missing. Recently, based on a preliminary investigation using stable isotopes in a sediment profile (Kanjeta cliff section), we proposed that the present-day climate in Gujarat was established sometimes prior to ~2.1 ka (Laskar et al., 2010). Now we explore this further using sedimentary carbonates and organic matters from five more sites (Dadhal, Sengur, Dhamlai, Motalimatawada and Lohara) in the lower Narmada valley.

Our objectives are: (1) to estimate the time frame of deposition of the youngest alluvial sediments occurring at the surface up to a depth of a couple of meters, (2) to reconstruct paleoclimate and paleovegetation of the region during the late Holocene (last ~3 ka BP) using δ¹³C in coexisting carbonates and organic carbon, and δ¹⁸O in carbonates and (3) to determine factors controlling the isotopic composition of oxygen in sediment carbonates.

Climate and sediment characteristics of the study area

The Narmada river originating at the Amarkantak hills of Madhya Pradesh, Central India (~1150 m a.s.l.), passes through rocky areas of late Cretaceous–Eocene basaltic lava flows (part of the Deccan Trap) and travels about 1300 km westwards before draining into the Gulf of Cambay (Chamyal et al., 2002). In its lower course (Figure 2), the river passes through the alluvial plains that have overlying Quaternary sediments, mostly sand and silt. The sediments have a maximum thickness of ~800 m and are dominated by overbank deposits (Bhandari et al., 2005). Above this alluvial plain facies, there exists a reddish brown paleosol of thickness 4–5 m, formed between 15 and 10 ka BP in a comparatively wetter climate (Laskar et al., 2010). The overlying, stratified sediments forming the valley-fill terrace were deposited in the late Holocene, during historic and paleoflood events (Chamyal et al., 1997; Kale et al., 2003; Raj, 2008; Sukumaran et al., 2012). Instrumental weather records show several flood events since 1887 ce, the largest of which was in 1970 ce followed by 1968, 1973 and 1944 ce (Kale et al., 1994). As change in climate and sea level in the late Holocene (Jansen et al., 2007) was not strong enough to incise these depositional surfaces to several meters, tectonic uplift during the Holocene could be responsible for 5–50 m high-terrace surfaces, located well above the present-day tidal limits (Chamyal et al., 2002).

Figure 2.

Map of western India showing sampling sites. 1: Kanjeta-1; 2: Kanjeta-2; 3: Dadhal; 4: Sengpur; 5: Dhamlai village; 6: Motalimatawada forest; 7: Lohara. Inset shows a map of India with location of other paleomonsoon proxy records which are compared with the present results: a: Laskar et al., unpublished data (2011) using stalagmite δ¹⁸O; b: Yadava and Ramesh (2005) using stalagmite δ¹⁸O; c: von Rad et al. (1999) using varve sediments; d: Tiwari et al. (2006) using foraminiferal δ¹⁸O; e: Chauhan et al. (2009) using foraminiferal δ¹⁸O.

Soils in the uppermost Holocene alluvial deposits are weakly developed inceptisols without any observable diagnostic horizons. They are calcareous in nature, developed in transported alluvium, derived from the trappean rocks. Figure 3 shows two cliff sections of the Narmada River at Kanjeta and Lohara. Details of the soil classification at these sites are discussed elsewhere (Laskar, 2012). The carbonates are produced in situ, formed during pedogenesis and are yet to form nodules. The pedogenic origin of carbonates in the Lower Narmada Valley has been discussed elsewhere (Hedge and Switsur 1973; Khadkikar et al., 2000; Merh and Chamyal, 1997). The analysis was carried out on fine carbonates, most of which were not distinguishable by the naked eye. The source of organic carbon is mainly the vegetation present during pedogenesis, though the chance of percolation of recent organic material cannot be ruled out completely. More details on the topography, physical stratigraphy, geomorphologic settings, sediment types and their origin can be found elsewhere (Bhandari et al., 2005; Chamyal et al., 1994, 1997, 2002).

Figure 3.

(a) Kanjeta cliff-section before sample collection. A paleosol layer dated back to 15–10 kyr BP (Laskar et al., 2010) is visible at a depth of 10–15 m. (b) Lohara cliff section at Narmada Estuary.

The modern climate of the region is semi-arid to subhumid, with a mean annual rainfall of ~900 mm that occurs mainly during June to August due to the southwest monsoon. The warmest month’s maximum and coldest month’s minimum temperatures are, respectively, 44.8° and 4.8°C. The climatological parameters are obtained from the nearest meteorological site Baroda (22°18′N 73°15′E), about 70–100 km away from the study sites (India Meteorological Department, 1999). The elevation of the study sites varies from a few meters at Lohara to ~150 m at the Motalimatawada reserve forest, Rajpipla. Samples were collected from all sites as bulk sediments at 10 cm intervals from cliff sections and using an auger on plains, packed in clean plastic bags and brought to the laboratory. For the radiocarbon dating of organic carbon, samples were collected from the top, middle and bottom of each section. Along with sediments, various types of vegetation (grass, shrub and wood) were also collected at each site. Details of the individual sites, vegetation cover, soil types and carbon contents are given in Table 1.

Table 1.

Location, soil type, vegetation, carbon content and δ¹³C values of the surface vegetation in all the sites.

Site name	Latitude, longitude	Major vegetation cover	Soil type	Carbonate, Organic carbon in %	Sample collection	δ¹³C_VPDB in ‰ of vegetation^a	Remarks
Kanjeta cliff section	21°55′41″N, 73°21′12″E	Shrubs and woody plants	Sandy loams	3–14, 0.05–1	Trench method	−25.4 ± 0.4 (3)	No clear diagnostic horizons present in the sediments in any of the sediment profiles
Dadhal and Sengpur	21°37′57″N, 73°03′56″E and 21°37′07″N, 73°05′37″E	Woody plants/grass	Sandy loams	2.7–13, 0.2–0.8	Trench method	−25.1 ±0.7 (4)/−15.0± 0.4 (3)
Dhamlai village	21°41′58″N, 73°13′02″E	Medium-size trees	Sandy loams	No carbonate, 0.2–0.5	Augur method	−24.0 ± 0.6 (3)
Motalimatawada forest	21°50′33″N, 73°30′43″E	Dense forest	Sandy loams	No carbonate, 0.4–1.4	Trench method	−29.9 ± 1.0 (5)
Lohara	21°39′34″N, 72°33′09″E	Shrubs	Loamy sand	10–15, 0.03–1.7	Trench method	−28.6 ± 0.6 (4)

Note: ^aAverage values along with the standard deviations at one sigma level and number of samples measured are given within parentheses.

Materials and methods

Stable isotope analysis

For stable carbon and oxygen isotopic analysis of pedogenic carbonates, dried sediment sample rich in carbonate (~5 mg CaCO₃) was put in a glass tube, evacuated and reacted with 100% orthophosphoric acid at 25°C for about 8 h. The CO₂ produced was purified by pumping it through two traps containing alcohol-liquid nitrogen slush (−80°C) to freeze moisture and other condensable gases and two liquid nitrogen traps (−196°C) to freeze CO₂. Pure CO₂ was transferred to a high vacuum bottle and analyzed using a stable isotope ratio mass spectrometer (Europa Scientific GEO 20-20) for δ¹³C_carb and δ¹⁸O_carb at Physical Research Laboratory Ahmedabad. To determine the δ¹³C_oc, a small quantity (~1 g) of well mixed sample was treated with 10% HCl for ~8 h to remove CaCO₃, washed with distilled water to obtain neutral pH, dried and heated at 800°C along with CuO powder and silver foil in a small quartz tube (inner diameter: 7–8 mm) for 1 h. The CO₂ evolved was purified cryogenically as described for carbonate and analyzed in the mass spectrometer. The concentration of organic carbon was measured by expanding the CO₂ in a precalibrated volume with an attached pressure gauge. The precision of concentration measurement was better than 0.1%. Isotopic composition was expressed as δ¹³C and δ¹⁸O in parts per thousand (‰) relative to the Vienna Pee Dee Belemnite (VPDB) standard, where

δ = (\frac{R_{sam}}{R_{std}} - 1) \times 1000

(1)

R_sam and R_std refer to ¹³C/¹²C or ¹⁸O/¹⁶O ratio in the sample and standard, respectively. The external precision for carbonates, calculated using a secondary laboratory standard was ± 0.1‰ for δ¹³C and ± 0.2‰ for δ¹⁸O (Laskar, 2012; Yadava and Ramesh, 1999). The laboratory standard is a pure calcium carbonate obtained from Makarana Marble of Rajasthan. For reproducibility check, we did a few repeat analyses of some samples and no significant difference was observed between replicates. The accuracy of the measurements for carbonates is checked occasionally by running an international standard (NBS 19 supplied by IAEA), the long-term averages (5 years) of which for δ¹³C and δ¹⁸O are 1.95 ± 0.08‰ and −2.3 ± 0.1‰, respectively, with respect to VPDB (error at 2 sigma level). The accuracy of measurements for organic carbon is checked using international standards ANU Sucrose (IAEA-CH-6) and NBS Oxalic acid II, the obtained δ¹³C values of which are −10.5 ± 0.1 and −17.7 ± 0.1‰, respectively with respect to VPDB.

Radiocarbon dating

For radiocarbon dating, about 100 to 200 g of sample, after removal of rootlets and pebbles, was treated with 10% HCl for ~8 h to remove carbonates, washed several times with distilled water to make it pH-neutral, and dried in air at 70°C for a couple of days. The dry sample was combusted at 900°C under a continuous flow of pure oxygen at a pressure of 30–40 Torr for 1 h. The CO₂ produced was purified chemically and cryogenically to prepare benzene, the radioactivity of which was counted in a liquid scintillation counter (Quantulus 1220) at Physical Research Laboratory Ahmedabad following the standard convention (Laskar, 2012; Stuiver and Polach, 1977; Yadava and Ramesh, 1999;). The radiocarbon ages were calculated using the relation

t = 8268 \ln (\frac{A_{o}}{A})

(2)

where A_o and A are the standard and sample activities, respectively. The ages were expressed as years before present (BP). The percent modern carbon (pMC) was calculated using the relation (Stuiver and Polach, 1977; Yadava and Ramesh, 1999):

pMC = (\frac{pMC A}{A_{o}}) \times 100

(3)

The radiocarbon ages were corrected for fractionation using δ¹³C values, measured on a small aliquot of CO₂ before benzene synthesis. For samples containing bomb carbon (pMC ≥ 100%), radiocarbon ages were considered to be modern. The radiocarbon dates were converted into calendar ages using calibration programme Calib 6.1.0 and IntCal09 (Reimer et al., 2009; Stuiver and Reimer, 1993).

Results

Chronology of the sediments

Radiocarbon ages (in yr BP) of organic carbon at various depths are shown in Figure 4 and in Table 2. Bomb carbon (pMC>100%) is present in the surface sediments (0–10 cm). The dates in between two measured values are obtained by linear interpolation. The uncertainty in the radiocarbon ages varies between 5% and 25% (Figure 4 and Table 2).

Figure 4.

Radiocarbon ages of soil organic carbon at various depths. The ¹⁴C content at the top sediment layers (0–10 cm) is measured only in three profiles (Kanjeta 1, Dadhal and Dhamlai) and found modern with significant bomb carbon content (pMC>100%) and is assumed similar for all other sites. For the subsurface layers, the ages are calculated based on actual ¹⁴C measurements in all the profiles. The errors are at one sigma level.

Table 2.

Radiocarbon ages and corresponding δ¹³C_oc and calendar ages of the samples at various depths.

Sampling site	Depth (cm)	Laboratory code	δ¹³C_VPDB in ‰	¹⁴C age±1 σ (yr BP)	Cal. age in 1 σ range (cal. BP)
Kanjeta cliff section 1	0–10	PRL 3107	−17.60±0.01	Modern^a	Modern
	90–100	PRL 3108	−19.38±0.01	1250±100	1056–1264
	190–200	PRL-3109	−18.05±0.02	2900±400	2460–3410
Kanjeta cliff section 2	0–10	NA	−19.85±0.02	NA	NA
	90–100	PRL 3170	−18.67±0.03	1200±120	964–1182
	150–160	PRL 3171	−18.61±0.06	2150±140	1919–2303
Dadhal	0–10	PRL 3113	−17.35+0.01	Modern	Modern
	40–50	PRL 3114	−16.88±0.01	1420±130	1168–1414
	90–100	PRL 3115	−17.85±0.01	2200±270	1811–2467
Sengpur	0–10	PRL 3124	−15.31±0.05	Modern	Modern
	50–60	PRl 3125	−16.93±0.01	1450±260	1054–1569
	90–100	PRL 3126	−17.83±0.01	1960±560	1287–2491
Dhamlai	0–10	PRL 3121	−18.74±0.01	Modern	Modern
	50–60	PRL 3122	−19.83±0.01	700±110	545–698
	120–130	PRL 3123	−19.95±0.01	1900±200	1605–2063
Motalimatawada forest	20–30	PRL 3117A	−23.98±0.01	230±100	0–428
Motalimatawada forest	30–40	PRL 3117B	−24.19±0.01	630±170	473–739
Lohara	0–10	PRL 3110	−23.46±0.01	Modern	Modern
	80–90	PRL 3111	−20.48±0.01	3000±140	2919–3246
	190–200	PRL 3112	−20.23±0.02	4280±250	4380–4985

Notes: ^aPercent modern carbon >100% (see text).

Organic carbon content and stable isotopic composition

Variations of organic carbon content with depth are shown in Figure 5a. The proportion of organic carbon is low (<1%) in all the profiles except at some of the upper most sediment layers. Figure 5b shows the depth profile δ¹³C_carb, the values fall in a narrow range from −5 to −8‰, except at Lohara (a cliff section at the Narmada estuary). δ¹³C_carb at Lohara remains close to zero with little change throughout the profile. A substantial variation is observed in δ¹⁸O_carb almost in all the profiles (Figure 5c). At Lohara value of δ¹⁸O_carb is close to zero throughout the profile. Figure 5d shows depth profile δ¹³C_oc. The δ¹³C_oc at Lohara, like δ¹³C_carb and δ¹⁸O_carb, shows a deviation from δ¹³C_oc in other profiles and they fall in the range of −20 to −22‰. The present-day vegetation in almost all cases is shrubs and scattered woody plants (C₃) except at Dadhal and Sengpur where tropical grasses (C4) are also observed. δ¹³C values of shrubs and woody plants range between −24 and −30‰ and it is −15‰ for grasses (Table 1). Figure 6a and b show the variation in δ¹³C_carb and δ¹⁸O_carb with radiocarbon ages. There is hardly any variation in δ¹³C during the last ~3 kyr (excluding Lohara). δ¹⁸O, however, shows a significant deviation at ~1.3 and ~2.1 ka almost in all profiles. The difference between δ¹³C_carb and δ¹⁸O_carb (Δ) is shown in Figure 6d. Differences, observed in almost all cases but one, lie between −11 and −13‰. Figure 7 shows a scatter plot between δ¹³C_carb and δ¹⁸O_carb. The data points from Lohara cluster in the carbonates precipitated under marine influence.

Figure 5.

Depth profiles of sediment (a) organic carbon content, (b) δ¹³C in carbonate, (c) δ¹⁸O in carbonate and (d) δ¹³C in organic matter.

Figure 6.

Variation in stable isotopic composition with radiocarbon ages in the sediment profiles: (a) δ¹³C in carbonate, (b) δ¹⁸O in carbonate and (c) δ¹³C in organic matter. (d) Difference between δ¹³C values in carbonate and organic carbon (Δ= δ¹³C_carb − δ¹³C_oc).

Figure 7.

Scatter plot between δ¹³C and δ¹⁸O for the studied sediment carbonates. The encircled region shows the approximate range of δ¹³C and δ¹⁸O of marine carbonate (Zeebe and Wolf-Gladrow, 2003). The carbonates from Lohara plot within the field of marine values.

Discussion

Chronology of the sediments

All the sediments in the study area are exposed and a systematic study of their chronology would be important to understand their geomorphic evolution in the region. Sediments, up to a depth of 200 cm were deposited in the late Holocene (0–3000 yr BP), probably during large historic and paleofloods in the Narmada basin, as suggested by Kale et al. (2003). There is a gradual increase in the ages in all the sediment profiles with depth without any reversal indicating their systematic deposition and non-monogenetic origin. The uppermost sediment layers up to ~10 cm contain bomb carbon indicating that the origin of a significant fraction of the organic matter deposited after 1950 ce. The radiocarbon ages vary between 1300 and 3300 yr at ~100 cm depth, and between 3200 and 4200 yr at ~200 cm depth. A much higher age value at Lohara, where sediments have deposited under marine environment, is probably due to accumulation of older detrital materials.

Pedogenic origin of carbonate

There is a ¹³C enrichment of about 10‰ at 25°^oC during the precipitation of carbonates from CO₂ (Bottinga, 1969). Again the soil CO₂ is enriched in ¹³C by ~4.4‰ compared with organic carbon because of differences in the diffusivities of the two isotopic species of carbon (Amundson et al., 1998). Therefore, carbonates precipitated under isotopic equilibrium with soil CO₂ should have ¹³C enriched by ~14.4‰ with respect to organic carbon. The difference between carbonate and organic carbon δ¹³C, observed here, varies between 11 and 13‰ in most cases, smaller than expected (~14.4‰ at 25°C). This could be due to the discrimination against ¹³C during microbial decomposition. It is believed that when microbes decompose organic matter, they act preferably on isotopically lighter molecules. As a result, the residual biomass may get enriched in ¹³C up to 3–5‰ (Accoe et al., 2002; Ehleringer et al., 2000; Wynn et al., 2006) while the soil CO₂ is likewise depleted (Amundson et al., 1998). Carbonates precipitated from CO₂, depleted in ¹³C, could possibly give rise to this mismatch. Kinetic fractionation during calcite precipitation always leads to ¹³C enrichment in the residual reservoir, and hence should increase the difference between the two δ¹³C values, but our observations do not support this. Also the possibility of kinetic fractionation can be ruled out because the carbonate in soils is in isotopic equilibrium with CO₂ since the rate of change in p(CO₂) in soils is much slower than the rate considered for equilibrium isotopic fractionation (Turner, 1982). Other possibilities for the sources of carbon during calcite precipitation are the atmosphere and limestone bedrock. The mixing of atmospheric CO₂ depends on the soil respiration rate. At the higher respiration rate (>1 × 10–⁵ mole/cm² per yr), the contribution from atmospheric CO₂ is insignificant below a depth of 10 cm (Cerling, 1984). Contribution from atmospheric CO₂ (δ¹³C ~ −8‰) or limestone bedrock (δ¹³C ~ −0‰) during carbonate precipitation are not significant as they should increase the difference between δ¹³C of organic carbon and soil carbonates. Some other less-understood fractionation pathways that could cause this mismatch are reported by Monger et al. (2009).

To check if the origin of carbonate is the soil CO₂, we compared δ¹³C values of organic and carbonate carbon with the laboratory simulation results (Bottinga, 1969). Isotopic fractionation lines between δ¹³C of coexisting organic carbon and carbonate are drawn for various temperatures with which the present data are compared (Figure 8). Most of the data points lie between 25 and 40°C, which is the temperature of the warmer season in the region when the majority of the carbonates precipitate. This indicates that the source of the carbonates is the soil CO₂ from root respiration and microbial decomposition of organic carbon. This also acts as evidence against the diagenetic alteration of the carbon isotopes in the carbonates after their formation. In the nearby semi-arid region of Gujarat, pedogenic origin for carbonates was reported by Sareen and Tandon (1995) in the Sabarmati basin and Gibling and Tandon (1997) in the Rupen river basin. In the Lower Narmada valley, the pedogenic origin of carbonates is discussed in some previous studies (e.g. Hedge and Switsur, 1973; Khadkikar et al., 2000; Merh and Chamyal, 1997). Based on morphology, dimensions and the distribution of microscopic features, Khadkikar et al. (2000) ruled out the possibility that they are products of floodplain accretion. Climate fluctuation and their control of the alluvial fan in the Lower Narmada valley were reported by several others also (e.g. Chamyal et al., 1997; Prasad et al., 2007).

Figure 8.

Carbon isotopic compositions of coexisting organic matter and pedogenic carbonate. Straight lines are drawn based on laboratory experiments by Bottinga (1969) to obtain fractionation factor (α) during carbonate precipitation at different temperatures. Most of the points from the present study fall between the 25 and 40°C lines, the warm season temperature of the region. Only data for authigenic carbonates are plotted here.

Paleoclimate reconstruction

Lack of significant variation in δ¹³C_carb and δ¹³C_oc in the sediment profiles indicate that there was little change in the vegetation type for the last ~3 kyr, except very recently. The present-day vegetation in the region is quite different from that during the last ~3 kyr. It was mixed C₃-C₄ during the late Holocene. At present, it is dominantly C₃ with average δ¹³C in most cases ranging from −24 to −30‰, except at two sites where C₄ grasses are also present with average δ¹³C values of −15‰ (Table 1). The recent change is probably due to anthropogenic disturbance which caused replacement of C₄ grasses by shrubs and woody plants. Meteorological data show that there was no overall change in the mean values of temperature and precipitation in the region during the last ~150 years. Therefore the shift in vegetation is unlikely a result of a change in precipitation or temperature. The region is very much affected by industrial development. The grasses must have been destroyed and trees planted, along with the natural flourish of seasonal and non-seasonal herbs and shrubs. Shifting of grassland (C₄) to woody forests (C₃), because of anthropogenic disturbances is also reported by Caner et al. (2007) in southern India. The δ¹³C_carb and δ¹⁸O_carb at Lohara remain close to zero throughout the profile. This is probably due to marine influence during carbonate deposition. A difference of ~20‰ between δ¹³C_carb and δ¹³C_oc at Lohara is not possible for modern carbonates precipitated under isotopic equilibrium without the influence of an external source (Cerling, 1984). The possible external source is marine influence and is also supported by δ¹³C_oc as it mostly lies between −20 and −26‰ (Bentaleb et al., 1998).

In the tropics, δ¹⁸O_carb is useful as a rainfall indicator, as there is an ‘amount effect’: the higher the rain, the lower the isotopic composition, about −1.5‰ for every 100 mm increase in the monthly precipitation (Dansgaard, 1964; Yadava and Ramesh, 2005). Our observations over several years in the recent past reveal the magnitude of the amount effect in Gujarat to be −1.8‰ per 100 mm increase (Yadava et al., unpublished data, 2012). Figure 6b shows that around 1.3 and 2.1 ka BP, the δ¹⁸O_carb was significantly higher, probably a result of weakening of the southwest monsoon. Evaporation due to relatively dry conditions and/or higher temperature is another possibility. Also the marine influence during this time could shift δ¹⁸O towards more positive values, but this does not show up in δ¹³C_carb, although higher soil respiration could suppress the contribution from marine carbonate. We reconstructed the southwest monsoon for the last ~4000 years using δ¹⁸O of stalagmites from the Andaman islands(Laskar et al., unpublished data, 2011) and observed a strong reduction in the southwest monsoon during 2.0–1.8 kyr BP. Yadava and Ramesh (2005), based on stalactite δ¹⁸O from central India, reported a reduction in monsoon of ~33% around 2 ka. von Rad et al. (1999) reconstructed a high resolution paleomonsoon record using variations in the thickness of varve sediments from the northeastern Arabian sea. The two relatively arid phases around ~2.1 and ~1.3 ka BP (Figure 6b) are well documented in varved sediments and are attributed to a weaker southwest monsoon. The reduction in southwest monsoon strength during these periods is also well documented in the foraminiferal oxygen isotopic records (Chauhan et al., 2009; Tiwari et al., 2006). Considering all these evidences, we conclude that the enrichment in δ¹⁸O_carb around ~2.1 and ~1.3 ka BP is due to weaker southwest monsoon.

Conclusions

The uppermost alluvial sediments to a depth of 200 cm exposed in the Lower Narmada valley are late-Holocene deposits, probably accumulated during large historic and paleofloods. Stable isotope based paleoclimate study from the Lower Narmada valley has shown that the climate was semi-arid, established prior to ~3 ka BP, and the vegetation was of mixed C₃-C₄ type. The present vegetation is mainly shrubs and scattered woody plants (C₃) at all sites except Dhadal and Sengpur where grasses (C₄) are also present. The change in vegetation in the region is probably due to anthropogenic disturbance in recent time, C₄ grasses have been replaced in most cases by C₃ shrubs and woody plants. Carbonates present in the sediments are authigenic, formed during pedogenesis and are useful for paleoclimate reconstruction. The marine origin of carbonates and organic carbon is observed at Lohara, a cliff section of the Narmada estuary. Oxygen isotopes of sediment carbonates appear to be useful for paleomonsoon study in the region as their isotopic composition is most likely governed by the amount effect. Two relatively arid phases are observed around 2.1 and 1.3 ka BP and are consistent with a wide variety of proxy records from different geographical locations affected by the southwest monsoon.

Footnotes

Funding

We thank ISRO-GBP for funding. One of the authors (NS) thanks the DST for providing funds under the project DST no. SR/S4/ES-21/Narmada Window/P5.

References

Accoe

Boeckx

Cleemput

. (2002) Evolution of δ¹³C signature related to total carbon contents and carbon decomposition rate constants in a soil profile under grassland. Rapid Communications in Mass Spectrometry 16: 2184–2189.

Achyuthan

Quade

Loe

. (2007) Stable isotopic composition of pedogenic carbonates from the eastern margin of the Thar desert, Rajasthan, india. Quaternary International 162–163: 50–60.

Alam

Keppens

Paipe

(1997) The use of oxygen and carbon isotope composition of pedogenic carbonates from Pleistocene palaeosols in NW Bangladesh, as palaeoclimatic indicators. Quaternary Science Reviews 16: 161–168.

Amundson

Stern

Baisden

. (1998). The isotopic composition of soil and soil-respired CO₂. Geoderma 82: 83–114.

Amundson

Chadwick

Sowers

. (1988) Relationship between climate and vegetation and the stable carbon isotope chemistry of soils in the eastern Mojave Desert, Nevada. Quaternary Research 29: 245–254.

Badam

Ganjoo

Salahuddin (1986) Preliminary taphonomical studies of some Pleistocene Fauna from the Central Narmada Valley, Madhya Pradesh, India. Palaeogeography Palaeoclimatology Palaeoecology 53: 335–348.

Balesdent

Girardin

Mariotti

(1993) Site-related δ¹³C of tree leaves and soil organic matter in a temperate forest. Ecology 74: 1713–1721.

Bentaleb

Fontugne

Desccolas-Gros . (1998) Carbon isotopic fractionation by plankton in the Southern Indian Ocean: Relationship between δ¹³C of particulate organic carbon and dissolved carbon dioxide. Journal of Marine Systems 17: 39–58.

Bhandari

Maurya

Chamyal

(2005) Late Pleistocene alluvial plain sedimentation in Lower Narmada Valley, Western India: Paleoenvironmental implications. Journal of Asian Earth Sciences 24: 433–444.

10.

Bottinga

(1969) Calculated fractionation factors for carbon and hydrogen isotope exchange in the system calcite-carbon dioxide-graphite-methane-hydrogen-water vapor. Geochimica et Cosmochimica Acta 33: 49–64.

11.

Boutton

(1991) Stable carbon isotopes of natural materials: II. Atmospheric, terrestrial, marine and fresh environments. In: Coleman

Fry

(eds) Carbon Isotope Techniques. New York: Academic Press, pp. 173–185.

12.

Bowman

DMJS

Cook

Zoppi

(2004) Holocene boundary dynamics of a northern Australian monsoon rainforest patch inferred from isotopic analysis of carbon, (¹⁴C and δ¹³C) and nitrogen (δ¹⁵N) in soil organic matter. Austral Ecology 29: 605–612.

13.

Breecker

Sharp

McFadden

(2009) Seasonal bias in the formation and stable isotopic composition of pedogenic carbonate in modern soils from central New Mexico, USA. Geological Society of America Bulletin 121: 630–640.

14.

Caner

Lo Seen

Gunnel

. (2007) Spatial heterogeneity of land cover response to climatic change in the Nilgiri highlands (Southern India) since the last glacial maximum. The Holocene 17: 195–205.

15.

Cerling

(1984) The stable isotopic composition of modern soil carbonate and its relationship to climate. Earth and Planetary Science Letters 71: 229–240.

16.

Chamyal

Khadkikar

Malik

. (1997) Sedimentology of Narmada alluvial fan, western India. Sedimentary Geology 107: 263–279.

17.

Chamyal

Maurya

Bhandari

. (2002) Late Quaternary geomorphic evolution of the lower Narmada valley, Western India: Implications to neotectonic activity along the Narmada-Son Fault. Geomorphology 46: 177–202.

18.

Chamyal

Maurya

Raj

(2003) Fluvial systems of the drylands of western India: A synthesis of Late Quaternary environmental and tectonic changes. Quaternary International 104: 69–86.

19.

Chamyal

Sharma

Merh

. (1994) Significance of blank material at Tilakwada in Lower Narmada Valley. Current Science 66: 306–307.

20.

Chauhan

Vogelsang

Basavaiah

. (2009) Reconstruction of the variability of the southwest monsoon during the past 3 ka, from the continental margin of the southeastern Arabian Sea. Journal of Quaternary Science DOI: 10.1002/jqs.1359.

21.

Dansgaard

(1964) Stable isotopes in precipitation. Tellus 4: 436–468.

22.

Deines

(1980) The isotopic composition of reduced organic carbon. In: Fritz

Fontes

(eds) Handbook of Environmental Isotope Geochemistry, 1. The Terrestrial Environment. Amsterdam: Elsevier, pp. 329–406.

23.

Dhir

Tandon

Sareen

(2004) Calcretes in the Thar deserts: Genesis, chronology and paleoenvironment. Proceedings of Indian Academy of Science (Earth and Planetary Science) 113(3): 473–515.

24.

Ehleringer

Buchmann

Flanagan

(2000) Carbon isotope ratios in belowground carbon cycle processes. Ecological Applications 10(2): 412–422.

25.

Gibling

Tandon

(1997) Erosional marks on consolidated banks and slump blocks in the Rupen river north west India. Sedimentology 44: 339–348.

26.

Goodfriend

Magaritz

(1988) Palaeosols and Late Pleistocene rainfall fluctuations in the Negev Desert. Nature 332: 144–146.

27.

Hedge

KTM

Switsur

(1973) Radiocarbon dates on the buried soil in the Lower Narmada valley. Current Science 42: 607–609.

28.

India Meteorological Department (1999) Climatological Tables of Observatories in India 1951–1980. New Delhi: National Data Centre, pp. 327–328.

29.

Jansen

Overpeck

Briffa

. (2007) Palaeoclimate. In Solomon

Qin

Manning

. (eds) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, New York: Cambridge University Press, pp. 435–497.

30.

Kale

Ely

Enzel

. (1994) Geomorphic and hydrologic aspects of monsoon floods on the Narmada and Tapi Rivers in central India. Geomorphology 10: 157–166.

31.

Kale

Mishra

Baker

(2003) Sedimentary records of palaeofloods in the bedrock gorges of the Tapi and Narmada rivers, Central India. Current Science 84: 1072–1079.

32.

Khadkikar

Chamyal

Ramesh

(2000) The character and genesis of calcrete in late Quaternary alluvial deposits, Gujarat, wesetrn India, and its bearing on the interpretation of ancient climates, Palaeogeography Palaeoclimatology Palaeoecology 162: 239–261.

33.

Khadkikar

Merh

Malik

. (1998) Calcretes in semi-arid alluvial systems formative pathways and sinks. Sedimentary Geology 116: 251–260.

34.

Kohn

(2010) Carbon isotope compositions of terrestrial C₃ plants as indicators of (paleo)ecology and (paleo)climate. PNAS 107: 19,691–19,695.

35.

Laskar

(2012) Stable and radioactive carbon in Indian soils: implications to soil carbon dynamics. Ph D thesis, Department of Physics, Mohanlal Sukhadia University.

36.

Laskar

Sharma

Ramesh

. (2010), Paleoclimate and paleovegetation of Lower Narmada Basin, Gujarat, Western India, inferred from stable carbon and oxygen isotopes. Quaternary Interanational 227: 183–189.

37.

Leier

Quade

DeCelles

. (2009) Stable isotopic results from paleosol carbonate in South Asia: Paleoenvironmental reconstructions and selective alteration. Earth and Planetary Science Letters 279: 242–254.

38.

McPhearson

Boutton

Midwood

(1993) Stable carbon isotope of soil organic matter illustrates vegetation change at the grassland/woodland boundary in south-eastern Arizonz, USA. Oecologia 93: 95–101.

39.

Merh

Chamyal

(1997) The Quaternary Geology of Gujarat Alluvial Plains. New Delhi: Indian National Science Academy.

40.

Monger

Cole

Buck

. (2009) Scale and the isotopic record of C₄ plants in pedogenic carbonate: From the biome to the rhizosphere. Ecology 90(6): 1498–1511.

41.

Osmund

(1978) Crassulacean acid metabolism: A curiosity in context. Annual Review of Plant Physiology 29: 379–414.

42.

Pendall

Amundson

(1990) The stable isotope chemistry of pedogenic carbonate in an alluvial soil from the Punjab, Pakistan. Soil Science 149: 199–211.

43.

Pimentel

Wright

Azevedo

(1996) Distinguishing early groundwater alteration effects from pedogenesis in ancient alluvial basins: Examples from the Paleogene of southern Portugal. Sedimentary Geology 105: 1–10.

44.

Prasad

Phartiyal

Sharma

(2007) Evidence of enhanced winter precipitation and the prevalence of a cool and dry climate during the mid to late Holocene in mainland Gujarat, India. The Holocene 17: 889–896.

45.

Quade

Cerling

(1995) Expansion of C₄ grasses in the late Miocene of Northern Pakistan: Evidence from stable isotopes of paleosols. Palaeogeography Palaeoclimatology Palaeoecology 115: 91–116.

46.

Raj

(2007) Strike slip faulting inferred from offsetting of drainages: Lower Narmada basin, western India. Journal of Earth System Science 116: 413–421.

47.

Raj

(2008) Occurrence of volcanic ash in the Quaternary alluvial deposits, lower Narmada basin, western India. Journal of Earth System Science 117: 41–48.

48.

Reimer

Baillie

MGL

Bard

. (2009) IntCal09 and Marine09 radiocarbon age calibration curves, 0–50,000 years cal BP. Radiocarbon 51: 1111–1150.

49.

Sareen

Tandon

(1995) Petrology, micromorphology and granulometry of mid–late Quaternary continental deposits of the semi-arid Sabarmati basin, western India. In: Wadia

Korisettar

Kale

(eds) Quaternary Environments and Geoarchaeology of India. Bangalore, pp. 258–276.

50.

Stuiver

Polach

(1977) Reporting of ¹⁴C data. Radiocarbon 19(3): 355–363.

51.

Stuiver

Reimer

(1993) Extended ¹⁴C data base and revised CALIB 3.0 ¹⁴C age calibration program. Radiocarbon 35: 215–230.

52.

Sukumar

Ramesh

Pant

. (1993) A δ¹³C record of late Quaternary climate change from tropical peats in southern India. Nature 364: 703–705.

53.

Sukumaran

Sant

Krishnan

. (2012) High resolution facies record on late Holocene flood plain sediments from lower reaches of Narmada Valley, western India. Journal of Geological Society of India 79: 41–52.

54.

Tiwari

Ramesh

Yadava

. (2006) Is there a persistent control of monsoon winds by precipitation during the late Holocene? Geochemistry Geophysics Geosysystem 7(3): 1–7.

55.

Turner

(1982) Kinetic fractionation of carbon-13 during calcium carbonate precipitation. Geochimica et Cosmochimica Acta 46: 1183–1191.

56.

von Rad

Schaff

Michele

. (1999) A 5000-yr record of climate change in varved sediments from the oxygen minimum zone off Pakistan, Northeastern Arabian Sea. Quaternary Research 51: 39–53.

57.

Weiguo

Zisheng

Weijian

. (2003) Carbon isotope and C/N ratios of suspended matter in rivers – An indicator of seasonal change in C₄/C₃ vegetation. Applied Geochemistry 18: 1241–1249.

58.

Wynn

Harden

Fries

(2006) Stable carbon isotope depth profiles and soil organic carbon dynamics in the lower Missisippi Basin. Geoderma 131: 89–109.

59.

Yadava

Ramesh

(1999) Speleothems – useful proxies for past monsoon rainfall. Journal of Scientific and Industrial Research 58: 339–348.

60.

Yadava

Ramesh

(2005) Monsoon reconstruction from radiocarbon dated tropical Indian Speleothem. The Holocene 15: 48–59.

61.

Zaleha

(1997) Siwalik paleosols (Miocene, northern Pakistan): Genesis and controls on their formation. Joural of Sedimentary Research 67: 821–839.

62.

Zeebe

Wolf-Gladrow

(2003) CO₂ in Sea Water: Equilibrium, Kinetics, Isotopes. Elsevier Oceanography Series 65, Amsterdam: Elsevier, pp. 167–188.