1 Introduction

Stalagmites have enormous prospective for documenting the information of paleoclimatic conditions. The moist and warm climatic conditions are ideal for the growth of a stalagmite (Ayliffe and Veeh 1989). Moreover, high growth rates of stalagmite are observed throughout the interglacial periods (Harmon et al. 1975). The stable C and O isotopic compositions in stalagmites have been valuable proxies to reconstruct the climatic changes occurred in the past. A range of studies have been accomplished on δ18O of speleothems which establishes that the δ18O in conjunction with radiometric dating of speleothems provides very important information of past climate (Harmon et al. 1975; Ayliffe and Veeh 1989; Dorale et al. 1992; Kotlia et al. 2012, 2016; Lone et al. 2014; Joshi et al. 2017; Raza et al. 2017), but carbon isotope ratios of speleothems have acquired less attention as compared with oxygen isotope; however, quite a few studies have been established the utility of δ13C of speleothems (Hendy 1971; Quade et al. 1989; Brook et al. 1990; Dorale et al. 1992, 1998; Baskaran and Krishnamurthy 1993; Genty et al. 1998, 2003; Frappier et al. 2002; McDermott 2004; Fairchild et al. 2006; Cosford et al. 2009; de Cisneros and Caballero 2011). In their study, Quade et al. (1989) exhibit an evident link between carbon isotope ratios and vegetation changes. Moreover, Dorale et al. (1992) have also demonstrated that the vegetation type over the cave can be inferred by using the δ13C of speleothems. Furthermore, Baskaran and Krishnamurthy (1993) established that the δ13C values in speleothems can be employed to interpret the isotopic ratios of atmospheric CO2. Cosford et al. (2009) describe the utilization of δ13C of speleothems to obtain the records of paleo-environmental changes. However, several other studies have also suggested that the changes in concentration of atmospheric CO2, surface temperature, land cover and cave environment (Baldini et al. 2006; Breitenbach et al. 2015) all can influence speleothem δ13C. Therefore, stalagmite δ13C is more complex than δ18O because of various controlling factors. However, studies have revealed that stalagmite δ13C largely represents the local vegetation change (Tieszen et al. 1979a; Ku and Li 1998; Kuo et al. 2011; Zhao et al. 2015 and references therein). de Cisneros and Caballero (2011) also establishes the ability of carbon isotope values as paleoclimate indicator in their study on stalagmites from Nerja Cave, South Spain. They stated that the carbon isotope variations are associated with the terrestrial carbon cycle and thus provide information about the vegetation. In particular, it has been primarily employed to interpret C3 vs. C4 vegetation changes (Drysdale et al. 2004; Genty et al. 2006; Fleitmann et al. 2009; Rudzka et al. 2011) with lower δ13C value reflecting better vegetation cover because of dissolving more biogenic CO2 into the seepage water during the periods of better precipitation, whereas the enriched or higher δ13C values reflect less vegetation coverage due to a reduced amount of vegetative input of biogenic CO2 during the periods of arid climates (Dorale et al. 1998; Kotlia et al. 2016; Joshi et al. 2017). The temperature and precipitation mainly alter the properties of vegetation above the cave. In general, the vegetation cover increases during periods of better precipitation. Raza et al. (2017) established that the VSPM1 δ18O record on annual to centennial scales reflects Indian summer monsoon (ISM) or southwest monsoon precipitation changes with lower δ18O corresponding to wet climates. As mentioned earlier, usually vegetation development is better under wet climates. Therefore, a wet climate will show the depletion in both δ13C and δ18O with occasionally a time delay in δ13C because of delayed response of vegetation change.

2 Geological settings

Valmiki cave is situated in Kurnool district of southern India (Raza et al. 2017) (figure 1). The local climate is dominated by hot/moist conditions during the summer months and cold/arid during the winters (October–January). The mean annual rainfall and temperature are around 670 mm and 28°C, respectively. From a geological perspective, the Valmiki cave is situated in the Paleoproterozoic Vempalle Formation of stromatolitic dolomite as mentioned by Raza et al. (2017). This cave is an exceptional site of prehistoric significance and provides paleoenvironment record of the last deglacial period. The Valmiki cave comprises several hallways and the revelation of internal chambers with the outside atmosphere is restricted to a very narrow creep-in passage. The other details of the Valmiki cave have been described in Raza et al. (2017).

Figure 1
figure 1

The location of Valmiki (this study) and Kalakot caves discussed in this study are indicated here.

Full size image

3 Materials and methods

The VSPM1 stalagmite sample was cut vertically to uncover the growth bands and also to inspect the secondary adaptation. Further, a vertical section of the sample was sliced and polished (figure 2). Later, the sample was cleaned to remove the solid particles adhere to the sample. The slicing and cleaning of the sample were performed at CSIR-National Geophysical Research Institute, Hyderabad by a diamond cutter and an ultrasonic bath respectively. With an approximate length of 180 mm, the VSPM1 stalagmite sample has a maximum diameter of ~120 mm at the bottom. Consistent laminae with the shift of skinny and thick layers are clearly visible. The apparent light and dark bands show alterations in the formation. Hendy test was carried out to ensure whether or not the VSPM1 stalagmite was grown under isotopic equilibrium. This test establishes the suitability of the stalagmite sample for the paleoclimate studies. For this purpose, three different layers were selected at different depths on regular interval (figure 2). Afterward, six sub-samples were drilled from these three layers and analyzed for isotopic compositions of carbon and oxygen (δ13C and δ18O) (figure 3).

Figure 2
figure 2

VSPM1 stalagmite. Black lines represent the drilling points for δ18O measurements. Red lines represent eight 230Th dating samples. Blue lines show layers of Hendy test and green boxes mark XRD analysis points (modified after Raza et al. 2017).

Full size image
Figure 3
figure 3

Hendy test results. Red, blue and green lines represent three dissimilar test layers. (a) δ18O values, (b) δ13C values, and (c) shows the correlation between δ18O and δ13C. r2 in (c) represents that the sample formed under equilibrium conditions (modified after Raza et al. 2017).

Full size image

To assess the mineralogical composition of VSPM1 stalagmite and also to inspect any change subsequent to its deposit, the X-ray diffraction (XRD) analysis was carried out on three sub-samples (~1 gm each) by using a Philips X-ray diffractometer using nickel filtered Cu Kα radiation at the CSIR-National Institute of Oceanography (NIO), Goa, India (Kessarkar et al. 2010). Three subsamples used for detailed mineralogical investigation were drilled at ~20, 100 and 170 mm depths from top of the sample using Manix-180 drilling device. X-ray diffraction measurements revealed that the sample is made up of aragonite mineral with uniform composition throughout the sample from top to bottom. Moreover, no indication of any change in mineralogy from aragonite to calcite was noticed.

The VSPM1 stalagmite was formed between 15,610 and 13,160 yr BP (Raza et al. 2017) as determined by U–Th dating of eight sub-samples (table 1) taken from different selected layers (figure 2). U–Th dating was performed on a Thermo Fisher NEPTUNE Multi-collector Inductively Coupled Plasma Mass Spectrometer (MC-ICP-MS) at the High-Precision Mass Spectrometry and Environment Change Laboratory (HISPEC), Department of Geosciences, National Taiwan University, following the methods developed by Shen et al. (2002, 2012). The age modelling was carried out by utilizing both linear and StalAge modelling program (figure 4). The StalAge model based on Monte-Carlo simulation is described in Scholz and Hoffmann (2011). The purpose of linear modelling was to correlate it with the StalAge model.

Table 1 230Th ages of VSPM1 stalagmite sample (India) by MC-ICPMS, Thermo Electron Neptune, at NTU (Raza et al. 2017).
Full size table
Figure 4
figure 4

(a) Comparing linear and StalAge models for the ages between 13,161 and 15,607 yr BP and (b) the StalAge model. Green line represents modelled ages and red lines showing the 95% confidence limits. The black dots with vertical error bars represent eight 230Th samples with standard error (modified after Raza et al. 2017).

Full size image

Continuous sub-sampling was carried out along the growth profile of VSPM1 stalagmite for high resolution measurement of carbon isotope ratios (δ13C). A total number of 263 sub-samples were analyzed for isotopic ratios. The sub-samples were reacted with saturated H3PO4 at 70°C in a vacuum system and evolved CO2 was analyzed for the measurement of δ13C according to the procedure described by Ahmad et al. (2008, 2012). The δ13C analyses of all the sub-samples were performed at CSIR-National Geophysical Research Institute, Hyderabad, India using MAT-253 Isotope Ratio Mass Spectrometer (IRMS) attached to a Kiel IV automatic carbonate device. The internationally recognized carbonate reference materials, NBS-18 and NBS-19 were analyzed after every eight samples to ascertain the reproducibility of the instruments. The measured δ13C values are articulated against VPDB (Vienna Pee Dee Belemnite) and the precision of the instruments was calculated to be ≤±0.06‰ for δ13C. The detailed procedures regarding the Hendy test, mineralogical investigation by XRD, U–Th dating and isotopic measurements of VSPM1 stalagmite sample are described in Raza et al. (2017).

4 Results and discussion

Prior to the discussion of the carbon isotope signatures of VSPM1 stalagmite, we need to confirm about the efficacy of this sample. The foremost requirement for any stalagmite sample for paleoclimate reconstruction studies is to ensure its formation under isotopic equilibrium condition. The stalagmite grown under isotopic equilibrium should exhibit almost similar δ18O values of each growth layer (Hendy 1971). The VSPM1 stalagmite was grown under isotopic equilibrium as evident by Hendy test results (figure 3). The details of Hendy test results have been provided in Raza et al. (2017). The assessment of carbon and oxygen isotopic records along with the changing growth rates of VSPM1 stalagmite reveals a wide range of δ13C and δ18O values indicating absence of any relationship.

Figure 5 shows δ13C variations of VSPM1 stalagmite sample during last deglaciation compared with its δ18O record (Raza et al. 2017). Previously published δ18O record of VSPM1 stalagmite has shown rainfall variations, with depleted and enriched δ18O values corresponding to wet and dry episodes respectively (Raza et al. 2017). The δ13C values of VSPM1 range from −2.42‰ to −6.02‰ with an average value of −4.54‰ (vs. VPDB) (figure 5). Earlier studies have demonstrated approximately 2–3‰ higher δ13C values in aragonitic stalagmites than calcitic samples (Morse and Mackenzie 1990; McMillan et al. 2005). A higher difference of 5–7‰ in δ13C values between calcite and aragonite samples has been reported by Zhang et al. (2004). Since the δ13C values of speleothems depend on soil CO2, fractionation process and the degree of interchange between drip water and source rock, the lower δ13C values suggest warm or wet climate (Dorale et al. 1992; Fairchild et al. 2006; Baldini et al. 2008; Kotlia et al. 2016). The large variations in δ13C values of stalagmites are perceived on different timescales (Frappier et al. 2002). Moreover, the source rock plays an important role in δ13C values of stalagmite due to its mineralogical and chemical composition. In addition, evaporation can also influence δ13C values of a stalagmite due to the loss of water from the solution layer and the related increase of Ca2+ ion concentrations. This may lead to elevated precipitation rates and larger isotope fractionation. A change in pCO2 due to ventilation effect can also influence stalagmite δ13C values because of the effect of pCO2 on the equilibrium concentrations of the CO2–H2O–CaCO3-system.

Figure 5
figure 5

The variations in δ13C (this study) and δ18O (Raza et al. 2017) values of VSPM1 stalagmite is presented in this graph.

Full size image

The deployment of δ13C for reconstruction of past climate is based on comprehensively diverging ratio of carbon isotopes (13C/12C) of C3 and C4 plants categorized on the basis of their photosynthesis pathways of carbon fixation. The δ13C value of C3 and C4 plants ranges from −26 to −28‰ and −11 to −14‰ respectively (Smith and Epstein 1971). In addition, the δ13C values of C3 and C4 plants are −24 to −33‰ and −10 to −16‰, respectively as reported by O'Leary (1988). Moreover, the δ13C values of C4 plants are ~17‰ higher than C3 plants and the majority of C4 plants are grasses. Therefore, environmental shift from forest to grassland can cause an increase in δ13C (Dorale et al. 1998). It can be observed from all above mention studies that the δ13C values of C3 plants are supposed to be far lower than C4 plants. The C3 plants dominate in the areas of elevated soil moisture whereas the C4 plants dominate in arid, semi-arid climates and regions of low soil moisture (Tieszen et al. 1979b). Consequently, when the C3 plants grow well under relatively warmer environment, the δ13C is influenced by decomposition and deprivation of C3 plants results in lesser δ13C values of stalagmites (Zhang et al. 2004; Fairchild et al. 2006).

As shown in figure 5, a long-term trend is observed between δ13C and δ18O records of VSPM1 stalagmite on different time periods, despite some time lag. This time lag between these two tracers may be due to the time delay in δ13C (carbon isotopes incorporation into the system) because of delayed response of vegetation change. Laskar et al. (2011) suggest that a short-term correlation between δ13C and δ18O records is somewhat problematic because of the presence of old carbon in deeper soil horizons. The absence of a strong relationship between δ13C and δ18O values of VSPM1 stalagmite could be attributed to some contribution arising from the old soil carbon with a different isotopic composition. It appears that the hydroclimate may have influenced both oxygen and carbon isotopes. It may be noted that the vegetation cover over the Valmiki cave was probably influenced by the climatic activity as reflected in VSPM1 stalagmite δ13C record. An increase in summer temperature and rainfall can lead to enhancement in vegetation density and production of soil CO2, resulting in lower δ13C values. Also, the increase in drip rate and reduction in degassing time due to improved rainfall results in decreasing and vice-versa. Although, δ13C values of a stalagmite are influenced by various factors as discussed above, the vegetation changes are generally considered as the dominating factor that controls δ13C of a stalagmite. The enrichment trend from 15,568–15,460 and 15,105–14,722 yr BP is clearly seen in VSPM1 δ13C record. The δ13C increases steadily with small fluctuations during these periods. The δ13C increase of ~2.75‰ from 15,568 to 15,460 yr BP suggests a decline in vegetation cover during this period with poorest vegetation environment around 15,473 yr BP manifested by the most enriched δ13C value. This severe drop in vegetation coverage shown by the most enriched δ13C at ~15,473 yr BP may have caused by some local vegetation change at this time. A gradual shift from 15,460 to 15,105 yr BP, marked by 3.55‰ δ13C decrease due to change in vegetation density has been attributed to strengthened ISM activity (Raza et al. 2017). A gradual increase in δ13C is observed from 15,105 to 14,722 yr BP followed by decrease in δ13C from 14,722 to 14,597 yr BP showing reduction in vegetation density from 15,105 to 14,722 yr BP followed by enhanced vegetation coverage from 14,722 to 14,597 yr BP. Moreover, the shaded bar in figure 5 represents the coincident variability in δ13C (this study) and δ18O (Raza et al. 2017) values of VSPM1 stalagmite at around ~14,722 and 14,750 yr BP, respectively, during Termination 1a (i.e., the onset of late-glacial interstadial around 14,800 yr BP). This simultaneous reduction and enhancement in vegetation density are also demonstrated by a decline in ISM activity followed by strengthening of ISM precipitation based on δ18O values during these periods (Raza et al. 2017). The δ13C values are relatively depleted than the average δ13C value between 14,597 and 13,161 yr BP, indicating a good vegetative cover during this period. Raza et al. (2017) also reported an enhancement in ISM activity during this period, except from 14,656 to 14,369 yr BP. The VSPM1 δ13C record shows that δ13C values are mostly lighter than the average between 14,597–13,161 yr BP and 15,404–4,872 yr BP, whereas heavier between 14,872–14,597 yr BP and 15,607–15,404 yr BP indicating vegetation density was relatively better from 14,597 and 13,161 yr BP and 15,404 and 14,872 yr BP, while there was deprived vegetation environment from 14,872–14,597 yr BP to 15,607–15,404 yr BP. The better vegetation coverage during 14,597–13,161 yr BP is also evident by intensified ISM activity marked by gradual decrease in δ18O during this period except from 14,656 to 14,369 yr BP.

Therefore, based on our findings, the climatic patterns can be categorized in three major phases: relatively wet period with better vegetative coverage from 15,607 to 15,105 yr BP revealed by VSPM1 δ13C values fluctuating above the average value of δ13C, a dry period with low vegetation cover between 15,105 and 14,722 yr BP followed by gradually strengthened vegetation activity between 14,722 and 14,597 shown by increase and decrease in δ13C values respectively and lastly a moderately wet period with good vegetation cover from 14,597 to 13,161 yr BP represented by decreasing trend in δ13C values of VSPM1 stalagmite (figure 5). These changes in vegetation cover and density based on δ13C values of VSPM1 stalagmite (this study) from ~15,607 to 13,161 yr BP is also verified by the variations in Indian summer monsoon (ISM) activity based on stable oxygen isotope record of the same sample (Raza et al. 2017). An almost synchronous variability appears between δ13C and δ18O records of VSPM1 stalagmite (figure 5).

5 Comparison of VSPM1 δ 13C record with another stalagmite record

We have compared δ13C variations of VSPM1 stalagmite with another published δ13C record of a stalagmite sample KL-3 in figure 6 from the northern India (Kotlia et al. 2016). The locations of VSPM1 (15°09′N: 77°49′E) and KL-3 (33°13′19′′N: 74°25′ 33′′E) stalagmites lies at an elevation (altitude) of 420 and 826 m above sea level respectively. The objective was to understand the response of hydroclimate and vegetational changes for the studied time interval. The δ13C values of VSPM1 stalagmite show greater δ13C values (ranging from −2.42 to −6.02‰ with an average value of −4.54‰), whereas δ13C of KL-3 stalagmite show significantly lower values (ranging from −8.03‰ to −10.40‰ with an average value of −9.24‰) (figure 6). Although, the comparison between δ13C records of VSPM1 and KL-3 stalagmites does not show a strong one-to-one relationship, broad synchronous variations appear in both records on multidecadal to centennial time scales within age uncertainties (figure 6). Moreover, the amplitude variations in δ13C values for the two records (VSPM1 and KL-3) after around 13,700 yr BP is almost similar (~1.5‰). A gradual δ13C depletion in both these records from around 14,300 to 13,700 yr BP is clearly evident in figure 6. The amplitude difference of ~1.23‰ in δ13C records of VSPM1 and KL-3 stalagmites may be due to the difference in vegetation changes in both these regions. The average δ13C value of KL-3 is ~4.7‰ lower than the average δ13C value of VSPM1 stalagmite. There are many factors that may be responsible for this large difference in δ13C values, such as the hydroclimatic conditions, dissimilar mineralogy of source rocks, vegetation type (C3 and C4 plants), etc. It may be noted that the host carbonate rock composition plays a vital role in δ13C values of stalagmites. The host rock for KL-3 stalagmite is limestone, whereas VSPM1 stalagmite was accumulated by overlying dolomite. In addition, the contribution of C3 and C4 plants might have played a role in δ13C values of these two stalagmites. Moreover, the amount of organic and inorganic carbon contribution in both the samples cannot be ruled out which could have an effect on the δ13C values of stalagmites. Usually in higher altitudes, the erosion is relatively more responsible for any change in environmental system than chemical weathering. It seems that there could be more inorganic carbon content in VSPM1 stalagmite because of somewhat more chemical weathering of host rock in southern India due to higher rainfall as compared with KL-3 stalagmite from the north Indian region. The δ13C values of KL-3 stalagmite are more negative relative to VSPM1 δ13C values due to more organic carbon contribution. This could be because of more organic carbon supply to cave overlying soil from fallen leaves and relatively less inorganic carbon contribution from the host rock.

Figure 6
figure 6

Graph shows the synchronous variability between δ13C values of VSPM1 (this study) and KL-3 (Kotlia et al. 2016) stalagmites.

Full size image

6 Conclusions

This study demonstrates the importance of stable carbon isotopes in understanding hydroclimate and paleo-vegetation changes during the last deglaciation based on variations in δ13C values of VSPM1 stalagmite from southern India. The δ13C fluctuations suggest rapid changes in vegetation controlled by ISM activity between 15,607 and 13,161 yr BP. Three major phases were noticed during the time window of this study, first a wet period of good vegetative coverage from 15,607 to 15,105 yr BP, followed by a dry period with low vegetation cover between 15,105 and 14,722 yr BP and finally another wet event from 14,722 to 13,161 yr BP. These changes in vegetation cover, based on δ13C record of VSPM1 stalagmite was controlled by local hydroclimate as evident by its δ18O record. This study establishes the utility of stable carbon isotopes in stalagmites as an important tracer of paleo-vegetation activity.