06 Warrier et al. 2014 - P3 - TK Sedimentology - Lake level variations.pdf

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/authorsrights

Author's personal copy

Sedimentological and carbonate data evidence for lake level variationsduring the past 3700 years from a southern Indian lake

Anish Kumar Warrier ⁎, R. Shankar, K. Sandeep 1

Department of Marine Geology, Mangalore University, Mangalagangotri 574199, India

a b s t r a c ta r t i c l e i n f o

Article history:Received 31 March 2012Received in revised form 11 May 2013Accepted 20 May 2013Available online 28 May 2013

Keywords:Particle sizeLake sedimentsRock magnetismPaleoclimateLake levelSouthern India

Over the years, several proxies have been developed to reconstruct rainfall variability. However, most rely onindirect approaches to provide qualitative paleorainfall estimate. In an attempt to obtain a more directmeasure of paleorainfall, Shankar et al. (2006) explored the rock magnetic properties of lake sediments fromThimmannanayakanakere (TK) in tropical southern India. They proposed the use of magnetic susceptibility asa proxy for rainfall in the tropics. Warrier and Shankar (2009) provided geochemical evidence in support ofthis proposition. Here, sedimentological and carbonate data is provided as further evidence to bolster Shankaret al.'s (2006) proposition.High (low) values of χlf indicate high (low) rainfall in the region of TK during the past 3700 years. Particlesize variations suggest that the sand % was high (low) during arid (humid) periods, when the TK lake levelwas low (high). Hence, a negative correlation is documented between sand % and χlf along with other rockmagnetic parameters. HIRM (an indicator of magnetically “hard” minerals like haematite and goethite) issuggestive of a relatively arid climate; the high (low) HIRM values in TK sediments indicate arid (humid)conditions. For this reason, sand % is positively correlated with HIRM. By contrast, fine silt and clay contentsare low during low-rainfall periods and vice versa. Thus, both fine silt and clay contents are positively corre-lated with χlf and other rock magnetic parameters, but negatively correlated with HIRM. Magnetic mineralsreside principally in the fine silt fraction of TK sediments as evidenced from the positive correlation betweenfine silt content and magnetic susceptibility. Carbonate content too is indicative of paleorainfall conditions,being high (low) during arid (humid) climatic conditions. Based on the χlf, sand % and carbonate % data,we have inferred lake level variations in TK during the past 3700 years.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Determining the periodicity and amplitude of climatic variationsin the past is important for projecting the future climate scenario.Of the several parameters that constitute climate, rainfall is perhapsthe most important, as it impacts the agriculture, economy andwell-being of the people of a region. Over the past few decades,several archives like marine (Trauth et al., 2003; Zheng et al., 2010)and lacustrine (Holzhauser et al., 2005; Cherapanova et al., 2007;Dearing et al., 2008; Stansell et al., 2010) sediments, tree-rings(Raspopov et al., 2004; D'Arrigo et al., 2008), speleothems(Denniston et al., 2000; Burns et al., 2002; Lachniet et al., 2004) andice-cores (Jouzel et al., 1993) were investigated to decipherpaleoclimate/paleoenvironment. They used proxies such as stable

isotopes of oxygen and carbon, clay mineralogy, diatoms and pollen,all of which give only indirect estimate of rainfall.

In this regard, we investigated the rock magnetic parameters ofsediments from Thimmannanayakanakere (TK) — a small lake insouthern India, and proposed magnetic susceptibility as a proxy forpaleorainfall variations in the tropics (Shankar et al., 2006). Webased our proposition on the positive correlation documentedbetween magnetic susceptibility of TK sediments and instrumentalrainfall data for the past 130 years. This proposition was bolsteredby historical records of a drought and a high-rainfall period, besidesproxy records from geographically distant locations. Subsequently,Warrier and Shankar (2009) provided geochemical evidence for theuse of magnetic susceptibility as a rainfall proxy in the tropics. Theyfound a high percentage of pedogenic magnetite during high rainfallperiods (and vice versa), with metal/Al ratios also exhibiting a goodcorrelation with magnetic susceptibility. Here, we test the hypothesisthat sedimentological and CaCO3 data can provide additional proof forthe proposition that magnetic susceptibility may be used as a proxyfor rainfall in the tropics.

Our rationale for the hypothesis is as follows: Particle size of lakesediments must be related to lake level, which, in turn is coupled to

Palaeogeography, Palaeoclimatology, Palaeoecology 397 (2014) 52–60

⁎ Corresponding author at: National Centre for Antarctic & Ocean Research, Ministryof Earth Sciences, Government of India, Headland Sada, Vasco-da-Gama — 403804,India. Tel.: +91 832 2525633; fax: +91 832 2525566.

E-mail address: [email protected] (A.K. Warrier).1 Present address: Department of Post Graduate Studies and Research in Geology,

Government College Kasaragod, Kerala-671123, India.

0031-0182/$ – see front matter © 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.palaeo.2013.05.026

Contents lists available at ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology

j ourna l homepage: www.e lsev ie r .com/ locate /pa laeo


rainfall. Thus, there is a rainfall-particle size linkage. Similarly, there is acarbonate content-rainfall linkage because of carbonate precipitation inlakes under arid (low rainfall) but not humid (high rainfall) conditions.

One of the basic physical properties of sediments is their particlesize. Variations of particle size in sediment records suggest the chang-ing transport energy, lake levels and paleoenvironmental zones ofdeposition (Conroy et al., 2008). The transport and deposition ofclastic materials from the watershed to a depositional basin are princi-pally controlled by hydraulic conditions (Sly, 1978; Håkanson andJansson, 1983). Particle size distribution was widely used by researchersto determine paleoclimate because it is sensitive to climatic changes, notdisturbed by biological activities (Chen et al., 2004) and also because themethodology is simple, fast and inexpensive.

A survey of literature shows two diametrically opposite climaticinterpretations of particle size data:

a) Down-core variations in the particle size of lake sediments reflectfluctuations in lake level as well as general trends of warming versuscooling. A high content of coarse particles is generally considered anindicator of warm and dry climate, whereas a high content of fineparticles usually indicates a cool and wet climate (Wang et al.,2001; Chen et al., 2004; Yanhong et al., 2006; Xiao et al., 2009;Burnett et al., 2010). During dry periods, there is a drop in lake levelwith a reduction in the lake area. This results in the sampling locationbeing closer to shore line where coarse particles are easily depositedbut fine particles are still kept in suspension because of the stronghydrodynamic disturbance at low lake level. Thus, a high content ofcoarse particles is noted (Finney and Johnson, 1991; Shuman et al.,2001). During periods of high rainfall, however, the rise in lakelevel enlarges the lake area and as a consequence, the sampling loca-tion shifts far from the shore line. Hence, due to the weak hydrody-namic conditions, fine particles get deposited in the centre of thelake, whereas coarse particles are deposited near the shore(Menking, 1997; Chen andWan, 1999). In a few studies of lake sedi-ments, clastic materials like gravel, sand, silt and clay were found tooccur, in that order, from the lake shore to the centre, signifying adecreasing particle size with increasing water-depth and decreasinghydrodynamic power (Sarmiento and Kirby, 1962; Sly, 1978; Picardand High, 1981; Sun et al., 2001).

b) Another interpretation is that during periods of high rainfall, coarseparticles are transported to the centre of the lake due to high runoff,increased erosive power and large transport capacity. Duringlow-rainfall periods, on the other hand, only fine particles aretransported to deep water regions due to the low runoff, decreasederosive power and decreased transport capacity (Kashiwaya et al.,1987; Chen et al., 2004; Peng et al., 2005; Conroy et al., 2008).Hence, a coarse particle size reflects high rainfall conditions andvice versa. Percentage of carbonate in lake sediments, especiallythose of closed basins, is another useful proxy for lake-level varia-tions (Bischoff et al., 1997) and an important indicator of tempera-ture and humidity variations (Wetzel, 2001). Temperature plays animportant role in enhancing the productivity of a lake which, inturn, favours the production of autochthonous carbonate withinthe lake. High temperature leads to an increase in algal productivity,which depletes the dissolved CO2 content of lake water. This processfavours the formation of dissolved inorganic carbon. During periodsof high evaporation (warm and dry conditions), inorganic carbon-ates are precipitated in situ and deposited on the lake floor(Wetzel, 2001). Hence, a rise (drop) in the carbonate % of lacustrinesediments indicates a low (high) lake level and an arid (humid)climate.

2. Site settings

Thimmannanayakanakere (TK) – a small lake covering an area of0.17 km2 – is located at the foothills of the Chitradurga Fort, Karnataka

State (Fig. 1). Chitradurga is located in the southern part of the DeccanPlateau. The TK Lake is situated in a mountainous belt and the adjoininghills are mostly bare, stony and aligned SSE–NNW. A few streams thatdrain them transportwater and sediment to TK. Barring these hill ranges,the area is open and plain. Geologically, TK is situated in the ChitradurgaBasin of the Chitradurga group. The area is predominantly covered withporphyritic and coarse grained granitic gneiss. Other rock types belong-ing to this group are greywacke, chert, phyllite, banded ferruginouschert, volcanics, Fe–Mn formations, limestone, dolomite and phyllite(Radhakrishna and Vaidyanadhan, 1997). The average annual rainfallin the area is ~64 cm, received mostly during the SWmonsoon (Fig. 2).The lake-level rises by ~2–3 m during the monsoon season; however,during the peak of summer, the water evaporates, leaving the lake-beddry. This is evidenced by the markings of paleo-lake-level on the rocksnear the lake (~5.5 m; Plate 1). It also indicates that the lake level wasmuch higher earlier as it rarely rises to the height of the markings atpresent. The temperature is 16.6–27.9 °C during November–Februaryand 36.2–41 °C during March–May. The relative humidity is high duringJune–November, but decreases to ~30% during the other months. Strongwinds blowmostly fromSWorwest during the SWmonsoon. During therest of the year they blow fromNE and SE (Gazetteer of India, 1985). Theslopes of the hills in the study area are covered with patches of acacia,bamboos and other timber species. Previous records indicate thepresence of good timber trees, suggesting that the soil was fertile. How-ever, such vegetation is rarely seen at present due to recurring droughtsand deforestation (Gazetteer of India, 1985).

3. Materials and methods

3.1. Thimmannanayakanakere sediments and chronology

Sedimentological and inorganic carbonate studies were carriedout on samples collected (at 2-cm interval) from a 3.7-m pit dug inthe distal end of TK. The chronology of the TK sediment profile wasestablished using 14C dates obtained on the organic matter in sedi-ment samples by liquid scintillation counting (Gupta and Polach,1985). The ages were calibrated using the software CALIB, version4.3 (Stuiver et al., 1998). The age–depth model obtained with thehelp of the two 14C dates suggests a mean sedimentation rate of0.99 mm/year and each of the 2-cm thick samples represents aduration of ~22 years. Further details of sampling, 14C analysis andage–depth model are given by Shankar et al. (2006).

3.2. Particle size analysis

Based on the magnetic susceptibility data (Shankar et al., 2006), 31sediment samples representing periods of high and low rainfall wereselected to study their particle size distribution (Carver, 1971). Approx-imately 10 g of the sediment sample was taken in a pre-weighed glassbeaker and 50 ml of 10% glacial acetic acid and 20 ml of 30% hydrogenperoxide (H2O2) were added to eliminate carbonate material and or-ganic matter. The sample was washed thoroughly with double distilledwater (Millipore). Ten millimetre of 5% sodium hexametaphosphate(calgon) solution was added to the sediment sample to deflocculatethe clay particles. The sample was then wet-sieved through an ASTMsieve (mesh no. 230) to separate the sand (>63 μm) and silt + clay(b63 μm) fractions. The >63 μm fraction was transferred to apre-weighed beaker and oven-dried at 100 °C. Later 10 ml of calgon so-lutionwas added to the silt + clay fraction and the solutionwas stirredand poured into a 1000 ml measuring cylinder. The cylinder was filledwith double distilled water up to the 1000 ml mark. The sample wasstirred and, according to Stokes' law, coarse silt (CS), medium silt(MS), fine silt (FS) and clay (b2 μm) fractions were withdrawn fromthe cylinder with the help of a 20-ml pipette. The 20-ml sample solu-tions were transferred to pre-weighed beakers and dried in an oven at100 °C. Once dried, the weight was noted down. The weight of the

53A.K. Warrier et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 397 (2014) 52–60


clay fraction obtained was multiplied by 50 and a value of 1 (calgoncorrection factor) was subtracted from it to account for the weight ofdispersant (sodium hexametaphosphate). Similarly, the total weightfor coarse silt, medium silt and fine silt were calculated and the dataobtained plotted against their respective ages using SIGMAPLOT v. 11.

3.3. CaCO3 analysis

Carbonate content was determined for 35 sediment samples fromTK. Bulk sediment samples were dried in a hot air oven at 40 °C, finelyground and homogenised using an agate mortar and a pestle. About2 g of the sediment sample was taken in a beaker and 1 N dilute HCladded in small increments until effervescence stopped (Schumacher,2002). It was kept overnight at room temperature to facilitate completeremoval of carbonates. The sample was washed 3–4 times withdeionised water and dried in the hot air oven at 100 °C. Weight lossafter the HCl treatment gives the weight of CaCO3 or inorganic carbonand was expressed as percentage.

4. Results and discussion

4.1. Particle size variations in Thimmannanayakanakere sediments

Variations in the particle size of Thimmannanayakanakere sedimentsfor the past 3.7 cal. ka are shown in Fig. 3. Table 1 is the correlation ma-trix for magnetic susceptibility and particle size data. The sediments of

TK are composed mainly of fine silt and clay (b2 mm) fractions. Sandcontent is high at 3.4 cal. ka B.P. (7.87%) but low at 0.54 cal. ka B.P.(0.21%). The average sand content (%) in the sediment profile is 3.10%.The sand content is negatively correlatedwith χlf andother rockmagnet-ic parameters like χfd, χfd%, χARM and SIRM (Table 1).

However, a positive correlation (r = 0.47; p = 0.01; n = 31;Table 1) is documented between sand content andHIRM. Like sand con-tent, coarse silt (CS) fraction also exhibits a negative correlationwith χlfand other rockmagnetic parameters (Table 1). The CS fractionwas highat 3.4 cal. ka B.P. (9.65%) but low at 2.23 cal. ka B.P. (0.33%), with theaverage CS content in the sediment profile being 3.53%. A negativecorrelation is documented between CS content and χlf (r = −0.46;p = 0.01; n = 31; Table 1). Rock magnetic parameters like χfd, χfd%,χARM and SIRM also exhibit negative correlations with CS fraction(Table 1). By contrast, HIRM is positively correlated with CS fraction(r = 0.48; p = 0.01; n = 31; Table 1). The average medium silt (MS)content in the TK sediment profile is 6.20%. It is high at 2.49 cal. ka B.P.(12.05%) but low at 0.54 cal. ka B.P. (1.60%). Similar to sand and coarsesilt fractions, the medium silt fraction also exhibits a negative correla-tion with χlf (r = −0.44; p = 0.01; n = 31; Table 1) and other rockmagnetic parameters. However, it is positively correlated with HIRM.The fine silt (FS) fraction is low at 2.49 cal. ka B.P. (11.72%) but highat 0.61 cal. ka B.P. (27.39%). The average FS (%) in the TK sediments is20.33%. Clay is the principal size fraction present in TK sediments. Theaverage clay fraction is 66.84%. It is high at 0.54 cal. ka B.P. (74%) butlow at 3.41 cal. ka B.P. (57.53%). In contrast to coarse and medium silt

Fig. 1. (A) Location map of Thimmannanayakanakere (TK) (modified after Warrier and Shankar, 2009). Topographic contours are in metres. The numbers 1 to 7 in the inset map arelocations of other paleoarchives that are discussed in the text. (B) Satellite image of TK and its adjoining area. The sample location is shown by a yellow square.

Fig. 2. Instrumental rainfall data for the Chitradurga rainfall station during the past 100 years (India Meteorological Department, 2005).

54 A.K. Warrier et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 397 (2014) 52–60


Plate 1. Paleo-lake level markings measuring ~5.5 m from the lake bottom on rock boulders near TK Lake.



fractions, the fine silt fraction is positively correlated with χlf and otherrock magnetic parameters. Clay fraction exhibits a weak correlationwith all the rock magnetic parameters except HIRM (Table 1).

4.2. Paleoclimate in the TK catchment based on particle size variations inTK sediments

From 3.7 cal. ka B.P. to ~2 cal. ka B.P., the sand content is high(Fig. 3), suggesting low rainfall or arid conditions in the catchment.Under such climatic conditions, the lake area would have shrunk;the sampling location would have been closer to the shore wherecoarse particles are deposited (Finney and Johnson, 1991; Shumanet al., 2001). Besides, due to the strong hydrodynamic conditionshere, fine particles would not have been deposited. This interpreta-tion of climate (discussed in the Introduction Section) based on thesand content of TK sediments is supported by the χlf data. Magnetic

susceptibility is low during 3.7 cal. ka B.P.–~ 2 cal. ka B.P. periodbecause the production of pedogenic magnetite, which is the maincarrier of the magnetic signal in TK sediments, was low due tolow rainfall or arid conditions (Shankar et al., 2006; Warrier andShankar, 2009). Another rock magnetic parameter, HIRM (whichindicates the concentration of magnetically “hard” minerals likehematite and goethite) was high during this period, suggesting aridclimatic conditions (Shankar et al., 2006). This is further substantiat-ed by the negative correlation exhibited by the sand content with χlfand a positive correlation with HIRM (Table 1). The two samplesrepresenting historical records of a drought (1876 AD) and ahigh-rainfall event (1741 AD) also show variations in their sand andclay contents (Fig. 3). The sand content was high during the1876 AD drought and the value is similar to the ones documentedfor the pre-1.8 cal. ka B.P. arid period. By contrast, during the1741 AD high-rainfall event, the sand content was low and the clay

Fig. 3. Magnetic susceptibility (χlf) and particle-size variations plotted against age for Thimmannanayakanakere sediments.

Table 1Pearson's correlation coefficient matrix for particle size data and rock magnetic parameters for TK sediments. (N = 31; Bold = 99% statistical significance; underlined = 98%statistical significance).

χlf χfd χfd% χARM SIRM SIRM/χlf S-ratio HIRM Sand Coarse silt Medium silt Fine silt Clay (b2 mm)

χlf 1.00χfd 1.00 1.00χfd% 0.80 0.83 1.00χARM 0.98 0.98 0.84 1.00SIRM 0.99 0.98 0.81 0.99 1.00SIRM/χlf −0.62 −0.64 −0.57 −0.56 −0.53 1.00S-ratio 0.71 0.70 0.66 0.74 0.72 −0.47 1.00HIRM −0.74 −0.73 −0.67 −0.77 −0.75 0.51 −0.99 1.00Sand −0.54 −0.56 −0.59 −0.58 −0.55 0.42 −0.45 0.47 1.00Coarse silt −0.46 −0.47 −0.58 −0.53 −0.50 0.13 −0.48 0.48 0.72 1.00Medium silt −0.44 −0.46 −0.47 −0.49 −0.47 −0.01 −0.42 0.43 0.36 0.55 1.00Fine silt 0.65 0.65 0.61 0.67 0.68 −0.21 0.57 −0.57 −0.56 −0.70 −0.66 1.00Clay (b2 mm) 0.26 0.28 0.41 0.32 0.27 −0.13 0.27 −0.30 −0.74 −0.73 −0.50 0.21 1.00



content high. From 2 cal. ka B.P. towards the present, the sand con-tent decreased, suggesting a high lake level due to high-rainfall condi-tions. In this climatic scenario, the lake level increased, the lake areaexpanded and the sampling location shifted towards the centre ofthe lake where only fine particles were deposited while the coarseones would have been deposited close to the lake shore (Menking,1997; Chen and Wan, 1999). This scenario is generally valid for TKas fine silt and clay fractions are low from 3.7 cal. ka B.P. to2 cal. ka B.P. but increase from 2 cal. ka B.P. onwards to the present(Fig. 3). Another possible interpretation could be that during periodsof aridity, there was less rainfall due to which vegetation cover in thecatchment was less. Under low rainfall conditions, weathering ofcatchment soils was less intense because of which the relativeconcentration of coarse particles was high when compared to fineparticles. Hence, rainfall in the catchment, particularly the intenseones, transported coarse particles to the lake. The vice versa wouldbe expected during periods of low aridity or high rainfall. The positivecorrelation between χlf and the fine silt % (Fig. 4) indicates that themagnetic minerals reside principally in the fine silt fraction. Such arelationship between magnetic susceptibility and silt fraction wasreported previously by Colman et al. (1990) and Walden andAddison (1995).

4.3. Carbonate content and its relationship with χlf

The carbonate material in TK sediments ranges from 2.71 to33.71% with an average of 20.13% (Fig. 5). The inorganic carbonatein lake sediments is present in detrital and authigenic carbonates.The detrital carbonate may be derived from the products ofweathering of catchment rocks, and transported to the lake by smallstreams. In the TK sediments, the detrital carbonate content is lowas carbonate rocks are absent in the catchment (Radhakrishna andVaidyanathan, 1997). Hence, it may be argued that the carbonatesin TK sediments are principally authigenic. Authigenic carbonateforms in lake sediments depending on biological activity and physico-chemical conditions (discussed in the Introduction section). For theTK sediments, the carbonate content is negatively correlated with χlf(r = −0.75; p = 0.01; n = 35). Hence, it may be reasoned that ahigh carbonate content in TK sediments may indicate a dry and aridclimate (Wang, 1993). A negative correlation between carbonate con-tent and rainfall is documented in many other studies (Chen et al.,2002; Jenny et al., 2002; Luo et al., 2008). Luo et al. (2008) inferreda warm and dry climate in the Lop-Nur region, China, based on thehigh carbonate content in sediments. Fig. 5 is a bi-plot of the

carbonate content vs. χlf of TK sediment samples representing periodsof drought, high rainfall and historical records of 1876 AD droughtand 1741 AD high-rainfall event. The drought and high-rainfall sam-ples plot in two different clusters. This confirms that during periodsof aridity (low χlf = low rainfall), production of authigenic carbonatewas high in the lake and vice versa.

4.4. Lake level variations at TK during the past 3700 cal. years based onχlf, carbonate % and sand %

Fig. 6 exhibits the down-profile variations of magnetic susceptibil-ity, carbonate % and sand % of TK sediments. From 3.7 cal. ka B.P. to~2 cal. ka B.P., χlf values are remarkably low, suggesting that aridand dry conditions prevailed during that period. However, carbonate% and sand % are high for the same period, suggesting a low lake level.From 2 cal. ka B.P. to around 1 cal. ka B.P., climatic conditionschanged to sub-humid as shown by a moderate increase in χlf valuesand a decrease in the carbonate % and sand %. These data indicate thatthe lake-level during this period was high compared to thepre-2 cal. ka B.P. From 1 cal. ka B.P. to the present, the lake level atTK rose sharply as can be seen from the massive increase in the mag-netic susceptibility values and low values of carbonate % and sand %.Such characteristics (i.e., low carbonate content and fine particlesize, indicating a high lake level and vice versa) were also document-ed by Menking (1997) in the Owens Lake sediment core.

4.5. Comparison with other paleomonsoon/paleoclimate records

Fig. 7 compares the paleorainfall variations during the past3.7 cal. ka B.P. deduced from particle size and carbonate data of TKwith proxy records from varied geographic settings like the ArabianSea sediment cores (Sarkar et al., 2000; Anderson et al., 2002;Staubwasser et al., 2003; Thamban et al., 2007), peat deposits(Sukumar et al., 1993), cave deposits (Yadava et al., 2004) andtree-ring record (Borgaonkar et al., 2010). Prior to 2.5 cal. ka B.P.,the climate was arid with very low rainfall as indicated by the highsand percentage and carbonate content but low fine silt content(Fig. 7a, b, c). The δ13C record of Nilgiri peat deposits exhibited lessnegative values during 4–2 cal. ka B.P., indicating a dry climate(Fig. 7f). Low magnetic susceptibility values were documented in asediment core from the eastern Arabian Sea (Thamban et al., 2007;Fig. 7g). A low Globigerina bulloides % was documented by Gupta etal. (2003) during the same period (Fig. 7h). Similarly, Staubwasseret al. (2003) also reported a weakening of the southwest monsoonduring this period as evident from the less depleted δ18O values for

Fig. 4. Biplot of fine silt % vs. χlf for TK sediments. Note the significant correlationbetween the two, suggesting that the magnetic minerals in TK sediments reside mainlyin the fine silt fraction.

Fig. 5. Relationship between carbonate content and χlf for TK sediments. Note that thehigh- and low-rainfall period samples plot in separate clusters.



Fig. 6. Lake level variations and climatic changes in the TK catchment during the past 3700 cal. years based on the comparison of χlf, carbonate % and sand % data.

Fig. 7. Comparison of (a) sand %, (b) fine silt % and (c) carbonate data of TK profile for the past 3700 cal. yearswith (d) δ18O of Akalagavi speleothem (1Yadava et al., 2004), (e) ringwidthindex of a teak tree from Kerala (2Borgaonkar et al., 2010), (f) carbon isotopic data of Nilgiri peat deposits (3Sukumar et al., 1993), (g) magnetic susceptibility data of a marine sedimentcore along the western continental margin (4Thamban et al., 2007), (h) G. bulloides abundance from a sediment core off Oman (5Gupta et al., 2003), (i) δ18O of G. ruber from a sedimentcore off the Sindhu (Indus) River mouth (6Staubwasser et al., 2003) and (j) δ18O of G. sacculifer from a sediment core off the southwest Indian margin (7Sarkar et al., 2000).



Globigerinoides ruber from the sediments off the Sindhu (Indus)river mouth (Fig. 7i). Less depleted oxygen isotopic values ofGlobigerinoides sacculifer (Fig. 7j) from continental margin sedimentsof the eastern Arabian Sea were also documented by Sarkar et al.(2000). Similarly, arid conditions were deciphered for the 3600–2000 cal. years B.P. period based on pollen records of Rajasthan lakesediments (Bryson and Swain, 1981). The onset of arid conditions at~3.5 cal. ka B.P. was also documented by Caratini et al. (1991) andNaidu and Malmgren (1996). From 2 cal. ka B.P. to the present,there is an increase in fine silt % and a decrease in the carbonate con-tent and sand %, indicating a high lake level and an increase in rainfall.Similar inferences may be drawn from other paleoclimatic recordsfrom the region (Fig. 7f–j). The 1876–1877 AD historically document-ed drought in southern India is well recorded in the particle size andcarbonate data of TK, the Akalagavi speleothem δ18O (Yadava et al.,2004; Fig. 7d) and Kerala tree-ring width record (Borgaonkar et al.,2010; Fig. 7e). However, the 1741 AD high-rainfall event is not con-spicuous in the particle size and carbonate data which may be dueto the coarse temporal resolution of the data. The close correlationamong proxy records from different archives suggests that the behav-iour of the southwest monsoon was near similar across a widegeographical region.

5. Conclusions

The particle size and carbonate data for Thimmannanayakanakere(TK) sediments obtained in this study enable us to draw the followingconclusions:

• A high (low) sand % is indicative of arid (humid) climatic conditionswhen the lake level at TK was low (high) and vice versa.

• Sand % is negatively correlated with χlf and other rock magneticparameters like χfd, χfd%, χARM and SIRM, as the low (high) valuesfor the latter parameters are indicative of arid (humid) conditions.However, sand % is positively correlated with HIRM, as both theparameters are suggestive of arid climatic conditions.

• Fine silt and clay (b2 mm) fractions are low during low-rainfallperiods and vice versa. This behaviour is in tune with the interpre-tation based on magnetic susceptibility data. Fine silt as well as claycontents are positively correlated with χlf, suggesting that magneticminerals reside principally in the fine silt fraction of TK sediments.

• Carbonate content was high during arid periods (low rainfall) andvice versa.

• From 3.7 cal. ka B.P. to ~2 cal. ka B.P., carbonate % and sand % arehigh with low fine silt % suggesting a low lake level. From2 cal. ka B.P. to around 1 cal. ka B.P., climate was sub-humid with amoderate increase in fine silt content and a decrease in the carbonate% and sand %. From 1 cal. ka B.P. to the present, the lake level at TKrose sharply as exhibited by high fine silt content and low values ofcarbonate % and sand %.

• Paleo-lake-level variations for the past 3.7 cal ka B.P. obtained fromthe TK particle size and carbonate data are broadly similar to whatis indicated by proxy records from varied geographical regions.

• Particle size and carbonate data corroborate the proposition that χlfmay be used as a proxy for rainfall in the tropics.

Acknowledgements

We thank the Department of Space, Government of India, forsupporting this work through a research project to RS (“Paleoclimateof the past few centuries from lake- and tank-bed sediments ofSouthern India”; ISRO-GBP/WG-1 letter No. 9/5/2/2004-II) underthe Geosphere–Biosphere Program. AKW and KS thank the Depart-ment of Space, the Council of Scientific and Industrial Research(CSIR), and the University Grants Commission (UGC), Governmentof India for Junior and Senior Research Fellowships. We also thank

Mr. Sarathchandraprasad for his help with particle size analysis andDr. Avinash Kumar for preparing the location map. We are gratefulto the two anonymous referees and the Guest Editors, whosecomments helped us in improving the manuscript.

References

Anderson, D.M., Overpeck, J.T., Gupta, A.K., 2002. Increase in the Asian Southwestmonsoon during the past four centuries. Science 297, 596–599.

Bischoff, J.L., Fitts, J.P., Fitzpatrick, J.A., 1997. Responses of sediment geochemistry toclimate change in Owens Lake sediment: an 800-k.y. record of saline/fresh cyclesin core OL-92. Geological Society of America Special Papers 317, 37–48.

Borgaonkar, H.P., Sikder, A.B., Ram, S., Pant, G.B., 2010. El Niño and related monsoondrought signals in 523-year-long ring width records of teak (Tectona grandis L.F.)trees from south India. Palaeogeography, Palaeoclimatology, Palaeoecology 285,74–84.

Bryson, R.A., Swain, A.M., 1981. Holocene variations of monsoon rainfall in Rajasthan.Quaternary Research 16, 135–145.

Burnett, A.P., Soreghan, M.J., Scholz, C.A., Brown, E.T., 2010. Tropical East African cli-mate change and its relation to global climate: a record from Lake Tanganyika,Tropical East Africa, over the past 90+ kyr. Palaeogeography, Palaeoclimatology,Palaeoecology 303 (1–4), 155–167.

Burns, S., Fleitmann, D., Mudelsee, M., Neff, U., Matter, A., Mangini, A., 2002. A 780-yearannually resolved record of Indian Ocean monsoon precipitation from aspeleothem from south Oman. Journal of Geophysical Research 107 (D20), 4434.http://dx.doi.org/10.1029/2001JD1281.

Caratini, C., Fontugne, M., Pascal, J.P., Tissot, C., Bentaleb, I., 1991. A major change at ca.3500 years BP in the vegetation of the Western Ghats in north Kanara, Karnataka.Current Science 61 (9&10), 669–672.

Carver, R.E., 1971. Procedures in Sedimentary Petrology. JohnWiley and Sons Inc., NewYork.

Chen, J., Wan, G., 1999. Sediment particle size distribution and its environmental signif-icance in Lake Erhai, Yunnan Province. Chinese Journal of Geochemistry 18 (4),314–320.

Chen, J., Wan, G., Wang, F., Zhang, D.D., Huang, R., Zhang, F., Schmidt, R., 2002. Environ-mental records of carbon in recent lake sediments. Science in China Series D: EarthSciences 45 (10), 875–884.

Chen, J., Wan, G., David, D.Z., Zhang, F., Huang, R., 2004. Environmental records oflacustrine sediments in different time scales: sediment grain size as an example.Science in China Series D: Earth Sciences 47 (10), 954–960.

Cherapanova, M.V., Snyder, J.A., Grette-Brigham, J., 2007. Diatom stratigraphy of thelast 250 ka at Lake El'gygytgyn, northeast Siberia. Journal of Paleolimnology 37,155–162.

Colman, S.M., Jones, G.A., Forester, R.M., Foster, D.S., 1990. Holocene palaeoclimaticevidence and sedimentation rates from a core in southwestern Lake Michigan.Journal of Paleolimnology 4, 269–284.

Conroy, J.L., Overpeck, J.T., Cole, J.E., Shanahan, T.M., Steinitz-Kannan, 2008. Holocenechanges in eastern tropical Pacific climate inferred from a Galápagos lake sedimentrecord. Quaternary Science Reviews 27, 1166–1180.

D'Arrigo, R., Allan, R., Wilson, R., Palmer, J., Sakulich, J., Smerdon, J.E., Bijaksanae, S.,Ngkoimanif, L.O., 2008. Pacific and Indian Ocean climate signals in a tree-ring re-cord of Java monsoon drought. International Journal of Climatology 28, 1889–1901.

Dearing, J.A., Jones, R.T., Shen, J., Yang, X., Boyle, J.F., Foster, G.C., Crook, D.S., Elvin,M.J.D., 2008. Using multiple archives to understand past and present climate–human–environment interactions: the lake Erhai catchment, Yunnan Province,China. Journal of Paleolimnology 40, 3–31.

Denniston, R.F., Gonzalez, L.A., Asmerom, Y., Sharma, R.H., Reagan, M.K., 2000.Speleothem evidence for changes in Indian summer monsoon precipitation overthe last 2300 years. Quaternary Research 53, 196–202.

Finney, B.P., Johnson, T.C., 1991. Sedimentation in Lake Malawi (East Africa) during thepast 10000 years: a continuous paleoclimatic record from the southern tropics.Palaeogeography, Palaeoclimatology, Palaeoecology 85, 351–366.

Gazetteer of India, 1985. Karnataka State— Chitradurga District. Government of Karnataka,Bangalore.

Gupta, S.K., Polach, H.A., 1985. Radiocarbon Dating Practices at ANU — Handbook, Ra-diocarbon Dating Laboratory. Research School of Pacific Studies, ANU, Canberra.

Gupta, A.K., Anderson, D.M., Overpeck, J.T., 2003. Abrupt changes in the Asian south-west monsoon during the Holocene and their links to the North Atlantic Ocean.Nature 421, 354–357.

Håkanson, L., Jansson, M., 1983. Principles of Lake Sedimentology. Springer-Verlag,Berlin.

Holzhauser, H., Magny, M., Zumbühl, H.J., 2005. Glacier and lake-level variations inwest-central Europe over the last 3500 years. The Holocene 15 (5), 789–801.

India Meteorological Department, 2005. Rainfall Data for Chitradurga Station. Governmentof India.

Jenny, B., Valero-Garces, B.L., Urrutia, R., Kelt, K., Veit, H., Appleby, P.G., Geyh, M., 2002.Moisture changes and fluctuations of the Westerlies in Mediterranean CentralChile during the last 2000 years: the Laguna Aculeo record (33°50′S). QuaternaryInternational 87, 3–18.

Jouzel, J., Barkov, N.I., Barnola, J.M., Bender, M., Chappellaz, J., Genthon, C., Kotlyakov,V.M., Lipenkov, V., Lorius, C., Petit, J.R., Raynaud, D., Raisbeck, G., Ritz, C., Sowers,T., Stievenard, M., Yiou, F., Yiou, P., 1993. Extending the Vostok ice-core record ofpaleoclimate to the penultimate glacial period. Nature 364, 407–412.



Kashiwaya, K., Yamamoto, A., Fukuyama, K., 1987. Time variations of erosional forceand grain size in Pleistocene lake sediments. Quaternary Research 28, 61–68.

Lachniet, M.S., Burns, S.J., Piperno, D.R., Asmerom, Y., Polyak, V.J., Moy, C.M.,Christenson, K., 2004. A 1500-year El Nino/southern oscillation and rainfall historyfor the Isthmus of Panama from speleothem calcite. Journal of GeophysicalResearch 109, D20117.

Luo, C., Yang, D., Peng, Z., Zhang, Z., Liu, W., He, J., Zhou, C., 2008. Multi-proxy evidencefor Late Pleistocene–Holocene climatic and environmental changes in Lop-Nur,Xinjiang, Northwest China. Chinese Journal of Geochemistry 27, 257–264.

Menking, K.M., 1997. Climatic signals in clay mineralogy and grain-size variations inOwens Lake core OL-92, southeast California. Geological Society of America SpecialPapers 317, 25–36.

Naidu, P.D., Malmgren, B.A., 1996. A high-resolution record of late Quaternary upwell-ing along the Oman Margin, Arabian Sea based on planktonic foraminifera.Paleoceanography 11, 129–140.

Peng, Y., Xiao, J., Nakamura, T., Liu, B., Inouchi, Y., 2005. Holocene East Asian monsoonalprecipitation pattern revealed by grain-size distribution of core sediments ofDaihai Lake in Inner Mongolia of north-central China. Earth and Planetary ScienceLetters 233, 467–479.

Picard, M.D., High, L.R., 1981. Physical stratigraphy of ancient lacustrine deposits. In:Ethridge, F.G., Flores, R.M. (Eds.), Recent and Non-marine Depositional Environments:Models for Exploration: Society for Economic Paleontologists and Mineralogists,Special Publication, 31, pp. 233–259.

Radhakrishna, B.P., Vaidyanadhan, R., 1997. Geology of Karnataka. Geological Society ofIndia, Bangalore.

Raspopov, O.M., Dergachev, V.A., Kolström, T., 2004. Periodicity of climate conditionsand solar variability derived from dendrochronological and other palaeoclimaticdata in high latitudes. Palaeogeography, Palaeoclimatology, Palaeoecology 209,127–139.

Sarkar, A., Ramesh, R., Somayajulu, B.L.K., Agnihotri, R., Jull, A.J.T., Burr, G.S., 2000. Highresolution Holocene monsoon record from the eastern Arabian Sea. Earth andPlanetary Science Letters 177 (34), 209–218.

Sarmiento, R., Kirby, R.A., 1962. Recent sediments of Lake Maracaibo. Journal ofSedimentary Petrology 32, 698–724.

Schumacher, B.A., 2002. Methods for the Determination of Total Organic Carbon (TOC)in Soils and Sediments. Ecological Risk Assessment Support Center, Office ofResearch and Development, US Environmental Protection Agency.

Shankar, R., Prabhu, C.N., Warrier, A.K., Vijaya Kumar, G.T., Sekar, B., 2006. A multidecadalrock magnetic record of monsoonal variations during the past 3700 years from atropical Indian tank. Journal of the Geological Society of India 68, 447–459.

Shuman, B., Bravo, J., Kaye, J., Lynch, J.A., Newby, P., Webb III, T., 2001. Late Quaternarywater-level variations and vegetation history at Crooked Pond, SoutheasternMassachusetts. Quaternary Research 56, 401–410.

Sly, P.G., 1978. Sedimentary processes in lakes. In: Lerman, A. (Ed.), Lakes: Chemistry,Geology, Physics. Springer-Verlag, Berlin, pp. 65–89.

Stansell, N.D., Abbott, M.B., Rull, V., Rodbell, D.T., Bezada, M., Montoya, E., 2010. AbruptYounger Dryas cooling in the northern tropics recorded in lake sediments from theVenezuelan Andes. Earth and Planetary Science Letters 293 (1–2), 154–163.

Staubwasser, M., Sirocko, F., Grootes, P.M., Segl, M., 2003. Climate change at the4.2 ka BP termination of the Indus valley civilization and Holocene south Asianmonsoon variability. Geophysical Research Letters 30, 1425. http://dx.doi.org/10.1029/2002GL016822.

Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S., Hughen, K.A., Kromer, B.,McCormac, G., Van Der Plicht, J., Spurk, M., 1998. INTCAL-98 radiocarbon agecalibration 24,000–0 cal yr BP. Radiocarbon 40, 1041–1083.

Sukumar, R., Ramesh, R., Pant, R.K., Rajagopalan, G., 1993. A δ13C record of late Quater-nary climate change from tropical peats in southern India. Nature 364, 703–706.

Sun, Q.L., Zhou, J., Xiao, J.L., 2001. Grain-size characteristics of Lake Daihai sedimentsand its paleoenvironmental significance. Marine Geology and Quaternary Geology21 (1), 93–95.

Thamban, M., Kawahata, H., Rao, V.P., 2007. Indian summer monsoon variability duringthe Holocene as recorded in sediments of the Arabian Sea: timing and implications.Journal of Oceanography 63, 1009–1020.

Trauth, M.H., Deino, A.L., Bergner, A.G.N., Strecker, M.R., 2003. East African climatechange and orbital forcing during the last 175 kyr BP. Earth and Planetary ScienceLetters 206 (3–4), 297–313.

Walden, J., Addison, K., 1995. Mineral magnetic analysis of a ‘weathering’ surface with-in glacigenic sediments at Glanllynnau, North Wales. Journal of Quaternary Science10, 71–83.

Wang, Y., 1993. The carbonate sedimentation and environmental change in DaihaiLake. Oceanologia et Limnologia Sinica 24, 31–35.

Wang, H., Liu, H., Cui, H., Abrahamsen, N., 2001. Terminal Pleistocene/Holocenepalaeoenvironmental changes revealed by mineral-magnetism measurements oflake sediments for Dali Nor area, southeastern Inner Mongolian Plateau, China.Palaeogeography, Palaeoclimatology, Palaeoecology 170, 115–132.

Warrier, A.K., Shankar, R., 2009. Geochemical evidence for the use of magnetic susceptibil-ity as a paleorainfall proxy in the tropics. Chemical Geology 365 (3–4), 553–562.

Wetzel, R.G., 2001. Limnology: Lake and River Ecosystems. Academic Press, San Diego.Xiao, J., Chang, Z., Wen, R., Zhai, D., Itoh, S., Lomtatidze, Z., 2009. Holocene weak mon-

soon intervals indicated by low lake levels at Hulun Lake in the monsoonal marginregion of northeastern Inner Mongolia, China. The Holocene 19 (6), 899–908.

Yadava, M.G., Ramesh, R., Pant, G.B., 2004. Past monsoon rainfall variations in Peninsu-lar India recorded in a 331-year old speleothem. The Holocene 14 (4), 517–524.

Yanhong, W., Lucke, A., Zhangdong, J., Sumin, W., Schleser, G.H., Battarbee, R.W.,Weilan, X., 2006. Holocene climate development on the central Tibetan Plateau:a sedimentary record from Cuoe Lake. Palaeogeography, Palaeoclimatology,Palaeoecology 234, 328–340.

Zheng, Y., Kissel, C., Zheng, H.B., Laj, C., Wang, K., 2010. Sedimentation on the inner shelf ofthe East China Sea: magnetic properties, diagenesis and paleoclimate implications.Marine Geology 268 (1–4), 34–42.


Documents

06 Warrier et al. 2014 - P3 - TK Sedimentology - Lake level variations.pdf