Heavy Metal Distribution in Sediments of Krishna River Basin, India R. RAMESH were analyzed for heavy metals (V, Cr, Mn, Fe, Co, Ni, Cu, Department of Geological Sciences Zn, and Pb) by the thin-film energy dispersive x-ray fluores- McGill University cence technique. There is considerable variation in the con- Montreal, PQ, Canada H3A 2A7 centration of elements towards downstream, which may be V. SUBRAMANIAN due to the variation in the subbasin geology and various de- School of Environmental Sciences grees of human impact. Suspended particles are enriched in Jawaharlal Nehru University heavy metals throughout the basin relative to bed sediments. New Delhi, 110 067, India The heavy metals are enriched in coarse size fractions R. VAN GRIEKEN (10-90 i~m) throughout the Krishna River except its tributary Department of Chemistry Bhima, where finer fractions (2 #m) dominate. Transition ele- University of Antwerp (UIA) ments correlate very well with each other. There is a striking B-2610 Antwerp-Wilrijk, Belgium similarity between the bed sediments of Krishna River and the Indian average. When the annual heavy metal flux carried by the Krishna River was estimated, and viewed in relation to ABSTRACT/Suspended and bed sediments collected from the other major riverine transport, the Krishna is seen to be a the entire region of the Krishna River and its major tributaries minor contributor of heavy metals to the Bay of Bengal. Introduction world oceans (Milliman and Meade 1983). Therefore, the concentrations of heavy metals in the Asian fiver The fate of heavy metals in the aquatic environ- sediments assume importance in their global budget. ment is of extreme importance because of their impact Recently, Milliman and others (1987) discussed the on the ecosystem. The metals in such an environment human effects on sediment discharge by Asian rivers can be accommodated in three basic reservoirs: water, with Yellow river as an example. Subramanian and sediment, and biota. Sediments are usually regarded others (1987) reported on the heavy metals distribu- as the ultimate sink for heavy metals discharged into tion in the sediments of the Ganges and Brahmaputra the environment (Gibbs 1973). In a river, suspended rivers. In this report, we present the various factors sediments contain significantly higher levels of heavy regulating the heavy metal composition of suspended metals than the dissolved phase: For example, an and bed sediments from the entire Krishna basin, average world river contains 40 and 8.2 ~g/l of Fe and which is the fifth largest river (in terms of catchment Mn, respectively, in the dissolved phase and 48,000 area) in the Indian subcontinent. and 1,050 ~g/g, respectively, as particulates (Martin The Krishna River drains mineralized areas up- and Whitfield 1981). Thus a large part of the anthro- stream, while it is extensively used for agriculture pogenic discharge of heavy metals into the environ- downstream. It has a drainage area of 258,945 km 2, ment becomes part of the suspended matter in rivers, covering a distance of 1,400 km from Mahabaleshwar which acts as an efficient scavenger for these metals. in the western ghats to the river mouth at the Bay of Martin and Meybeck (1979) estimated the average Bengal. Bhima, Tungabhadra, Gataprabha, and Mala- river particulate matter composition on a global scale, prabha are its major tributaries. The river flows based on analysis of 40 elements in the Amazon, through several big cities, namely Raichur, Kurnool. Congo, Ganges, Magdalens, Mekong, Parana, and Ori- Srisailam and Vijayawada. From the data shown ;n noco rivers and compilation of published data on 13 Table 1, it is evident that Krishna River is one of the other major world rivers. These 20 major rivers repre- most intensively utilized rivers in India, and it may be sent 25% of the world drainage area and 15% of the polluted as a result of various degrees of human im- world rivers sediment discharge. However, the present pact. The river carries very little sediment load (<4 understanding of sediment chemistry is limited be- million ffyr) (Ramesh and Subramanian, 1988) per- cause of restricted information available on the large haps because of the predominance of precambrian sediment-carrying rivers of Asia. Asian rivers con- hard rocks, which cover nearly 80% of the basin area tribute about 50% of the global sediment input to (Fig. 1). Environ Geol Water Sci Vol. 15, No. 3, 207-216 © 1990 Springer-Verlag New York Inc. 208 R. Ramesh et al. Table 1. Water, land, and population data of Kirshna River basin a Water resources (million cubic meters) Potential Utilized Population Land Surface Ground Surface Ground Total Density % rural Culturable Net irrigated runoff water runoff water Total (millions) (km2) population (1000ha) (1000 ha) 62,784 9,628 52,437 6,513 58,950 38.50 149 80.9 20,299 1,819 dAfter Chaturvedi 1973. N I ' HY 0 CR'A-BZ-O". £'-L'_'- i ~k.~,,. ~ JRNOOL ARABIAN SEA [Z2~ o ..... T,op, g (Lava Flows ) ( Sed,menlarle$} Early to M,dOIe I P r ec ° m br ~°r~l (FJr °ntt~s i I I Figure 1. Geological map of the Krishna River basin with sampling locations. Materials and Methods p.m) fractions by the conventional pipet method (Guy 1977). Freshly deposited bed sediments from wet por- Figure 1 shows the geology of the bedrock, general tions near the river banks were likewise collected using flow direction, and location of sampling sites. During a small, stainless-steel pipe dredge at 14 locations (Fig. August 1984, 1-1 water samples were collected in wide- 1) including six samples from the tributaries. mouth polyethylene bottles at 16 locations along the Chemical analysis was performed by the thin-film river as well as its tributaries. The locations were energy dispersive x-ray fluorescence (XRF) technique. chosen so as to represent all the regions of the river The suspended sediment filters were fitted onto basin, major urban areas, dams, etc. Suspended sedi- Teflon rings, which were attached to the XRF unit, ments were separated by filtration of water through while the bed sediments, after grinding and suspen- 0.45-p.m pore-size membrane filters. In addition, 5 1 of sion, were loaded onto a thin Mylar film and mounted water samples were collected in few locations and then likewise for XRF measurements. Details of this separated into day (2 p.m), silt (20 ~m), and sand (90 method are presented elsewhere (Van Grieken and Heavy Metals in Krishna River, India 209 10,000 106 P ..... "4 i i Ronge i i SS: suspended sediments ~r i I r BS' bed sediments i I I I f t I L I 1,000 - ,I 1", 3" "10 5 --r .L I I ! ' i IT W' i J, Jl ! i I I T ~ T i Jl i r I ! I t I I I T I ~ 1" .I.I i I "1- j , I ! I ! : I II "1" -- I ~ I I I ,,r I J¢~ , t ' ' .1. ' ' [ ~" I I ~ J. I J. ~, 100 ~, , ~ ~ "104 I i I I :i, ,,,', ,~ T ', , I I I 1 11 i I ..L I i i -4- i , ~ ~ I I I Ii iI I I i i I I I J, .L I ', ) 10 I ,I -103 ,, I I I * I J- I I I I 1 I I i I 1 '1 ' : lo2 Mn Fe V Co Or Zn Cu Ni Pb Figure 2. Range and mean concentration of heavy metals in the suspended and bed sediments of the Krishna River basin. others 1979). The accuracy was checked by analysis of cally in Figure 2. Suspended sediment concentrations Soil-5 from the International Atomic Energy Agency vary within the basin by a factor of 2-4 for all the and BCR-1 from the U.S. Geological Survey. The XRF elements except Pb and Zn, which vary by a factor of 6 unit consisted of an HV generator, a Kevex 0810 and 11, respectively. On the other hand, the concen- system with tungsten-x-ray tube, a set of interchange- tration of heavy metals for the bed sediments shows able secondary fluorescers and filters, a 16-position wide variation within the basin. For example, the en- automatic sampler changer, a 30-mm 2 Si(Li) detector richment factor for Fe, Mn, and Cr is approximately connected to a multichannel analyzer, and a magnetic 12, 17, and 24, respectively. Figure 3 shows the down- tape recorder. The final computer program took into stream profile in the heavy metal content for the sus- account the various matrix effects (Van Dyck and Van pended sediments (3A) and bed sediments (3B) for the Grieken 1980). Krishna basin. The points of deflection in the down- stream profile are not uniform for both suspended and bed sediments. The variation in concentration to- Results and Discussion wards downstream may be due to the (1) change in relative contribution of sediments draining through Heavy Metals in Suspended and Bed Sediments different geological formations, (2) size differentiation The heavy metal contents of the suspended and (sorting) during sediment transport processes, and (3) bed sediments in Krishna River are presented graphi- human influence. Krishna drains very densely popu- 210 R. Ramesh et al. SS BS lOO,OOO l I I i ,IoT IO,OOOI i i 10~ A B IO,OOC l0 s 00 105 fi c& \Mn .,,I I-" I00( I0 e O0 10 + L.L.I _ At:z; E E o o <::£ -.1. i,.i.i "Co 3- it. ~Cr IJ_ ,00 :Ni t04 t0 l0~ Ni i I i0 i , i 10~ %00 .0 .o to o ,+,ao ,4oo 103 400 600 800 1010O 12100 14100' DISTANCE DOWNSTREAM, km DISTANCE DOWNSTREAM, km Figul~ 3.