1. INTRODUCTION:

Ground Water is a significant resource for human life. The chemical quality of the ground water percolating through the soil which is anthropogenically polluted has been reduced. The quality as well as the availability of fresh ground water resources in coastal regions is threatened by sea water intrusion from the sea side¹. Hence understanding the changes in coastal ground water quality in the present context of contamination problems especially affect tsunami inundation has become a major priority. Multivariate analysis is highly useful due to its relative significance in evaluating the combination of large chemical variable data set as they will become helpful analytical tools to reduce and organize large hydro geochemical data sets into particular groups with similar characteristics. The rotated factor analysis is widely employed as a statistical technique in hydro-geo chemistry. The analysis is highly useful for the interpretation of the ground water quality data and relating them to specific changes in hydro-geological processes. The factor analysis has been successfully applied to the sort out of hydro-geo chemical processes from the collected ground water quality data ^2-6.

The main purpose of such analysis for the study of the hydro geo-chemistry of an aquifer is to find a set of factors, few in number, which can explain a large amount of the variance of the analytical data. The present study employing the factor analysis has great potential to demonstrate its usefulness as a tool for the estimation of salt water intrusion problems (after the happening tsunami) in Visakhapatnam coastal aquifers system in Andhra Pradesh. The study Area is location is Uppada, Visakhapatnam district having coordinates 17°41′18.16″N 83°13′07.53″E and the study area map is presented in diagram – 1.

Diagram-1: Study area map

MATERIAL & METHODS:

Sixteen Ground water samples from bore wells and hand pumps were collected for two season’s pre and post monsoon and the details of the sampling code and sampling locations are presented in table-1.

Table-1: Details of Sampling code, Source type (BW-Bore well & OW- Open well) and Location

Sample Code/ Source Type	Location
GW-1/BW	Mangavani Peta
GW-2/BW	Kapula Uuppada-1
GW-3/BW	Kanarbalam
GW-4/OW	Kukkarlapeta-1
GW-5/BW	Kukkarlapeta-2
GW-6/OW	Peda Uppada-1
GW-7/BW	Peda Uppada-2
GW-8/BW	Nagarapalem-1
GW-9/BW	Nagarapalem-2
GW-10/BW	Somannapalem
GW-11/BW	Kapula Uppada-2
GW-12/BW	Kapula Uppada-3
GW-13/BW	Kapula Uppada-4
GW-14/BW	Kapula Uppada-5
GW-15/BW	Kapula Uppada-6
GW-16/BW	Kapula Uppada-7

The collected samples were analyzed for the parameters pH, EC, TDS, Na, K, Ca⁺², Mg⁺², Cl^-, HCO₃^-, NO₃^-, SO₄^-2and PO₄^-3 following standard procedures(APHA⁷) and the analytical data is presented in table-2&3 respectively.

Processing of Data:

The analytical data was used as variable inputs for factor analysis and performed by employing SPSS package described by Nie, the data was standardized according to criteria⁸. This procedure renders a new rotated factor varimax (Table-4&5) in which each factor was described in terms of only those variables and affords greater ease for interpretation. Factor loading is an indicator of the degree of closeness between the variables and the factor analysis provides several positive features that allow interpretation of the data set.

By verifying the factor loadings, communalities and Eigen values the variables belonging to a specific chemical process can be identified and the significance of the major parameters can be evaluated in terms of the total data set and in terms of each factor. Communality is an indicator of the error term. The factor scores for each sample and reflect the importance of a given factor at that sample site, the factor scores can be counted for each factor and for evaluating the aerial importance of the chemical process represented by that factor. Factor scores can be related to intensity of the chemical process described by each factor⁹. Extreme negative numbers (<-1) reflect areas essentially unaffected by the process and positive scores (>+1) reflect areas most affected. Near zero scores approximate areas affected to an average degree by the chemical process of that particular factor.

Table-2: Characteristics of Ground water

Sample Code	pH		EC(µmhos/cm)		TDS(mg/l)		Ca⁺²mg/l		Mg⁺²mg/l		HCO₃^-mg/l
Sample Code	Pre- M	Pos-M	Pre- M	Post-M	Pre- M	Post-M	Pre- M	Post-M	Pre- M	Post-M	Pre- M	Post-M
W-1	6.8	6.84	366	270	234	172.8	56	80	17	73.2	805	200
W-2	7.7	5.93	1472	290	942	185.6	32	80	35	97.6	1342	200
W-3	6.88	6.92	3830	280	2451	179.2	136	80	68	146.4	488	200
W-4	7.1	7.13	1554	890	994	569.6	112	120	2	48.8	1098	400
W-5	6.7	7.16	520	900	333	576	56	80	21	73.2	708	300
W-6	6.4	7.0	370	720	237	460.8	48	120	17	24.4	537	200
W-7	6.0	7.09	239	1280	153	819.2	44	80	15	48.8	390	500
W-8	6.2	6.36	252	510	161	326.4	BDL	80	32	48.8	293	300
W-9	6.3	6.32	557	520	356	332.8	40	80	32	73.2	610	300
W-10	7.5	7.12	1747	950	1118	608	32	120	46	48.8	1610	400
W-11	7.37	6.63	1128	810	722	518.4	64	80	27	24.4	1244	300
W-12	7.4	6.73	1470	1080	941	691.2	96	120	46	24.4	1244	400
W-13	7.1	6.88	1192	500	763	320	80	80	15	97.6	1244	400
W-14	6.7	6.28	1000	340	640	217.6	216	80	12	48.8	1147	300
W-15	6.98	6.82	1000	1370	640	876.8	280	120	BDL	48.8	1269	500
W-16	7.5	6.84	858	1440	549	921.6	112	80	44	73.2	781	500

Table- 3: Characteristics of Ground water

Sample Code	Cl^-mg/l		Na in ppm		K in ppm		Phosphate(mg/l)		Sulphate(mg/l)		Nitrate(mg/l)
Sample Code	Pre- M	Post-M	Pre- M	Post-M	Pre- M	Post-M	Pre- M	Post-M	Pre- M	Post-M	Pre- M	Post-M
W-1	751	230	16.25	4.08	12.29	0.36	0.64	BDL	14.6	15.3	8.6	12.4
W-2	234	124	18.47	5.52	20.41	0.53	0.42	BDL	12.8	16.7	6.4	8.7
W-3	794	266	27.45	5.59	38.53	0.44	0.31	BDL	11.8	13.6	7.6	9.5
W-4	305	248	19.84	15.38	20.19	1.21	BDL	BDL	56.4	64.2	5.2	8.4
W-5	71	460	11.70	16.86	9.47	1.14	BDL	BDL	48.0	53.9	9.6	14.3
W-6	BDL	336	6.83	6.24	10.18	0.86	0.21	BDL	45.2	47.9	4.2	8.6
W-7	BDL	195	5.41	19.85	10.35	1.79	BDL	BDL	48.4	52.7	7.6	11.4
W-8	28	106	5.32	7.7	16.29	0.72	0.15	BDL	38.0	52	6.4	9.2
W-9	63	230	9.13	6.81	20.02	0.47	0.10	BDL	48.2	52.3	11.2	16.3
W-10	234	284	30.27	16.64	22.91	0.36	BDL	BDL	74.6	83.3	5.4	5.1
W-11	106	17.7	15.92	10.72	27.60	1.08	BDL	BDL	38.2	40.8	6.3	5.9
W-12	212	53.2	18.61	13.43	20.22	8.13	BDL	BDL	49.6	54.9	10.5	8.8
W-13	155	106	17.18	8.05	20.53	0.5	0.24	BDL	23.2	24.5	14.2	13.2
W-14	333	BDL	19.71	5.25	21.26	0.42	0.68	BDL	8.50	13.2	8.6	7.3
W-15	439	88.6	31.17	20.4	22.57	0.26	2.10	BDL	51.42	66.3	6.4	5.4
W-16	71	BDL	6.35	19.96	19.16	0.11	1.80	BDL	47.0	55.2	11.8	13.6

Table: 4: Percentage Variance explained by Factors for pre monsoon water samples

Rotated Factor Pattern					Communalities
	Factor-1	Factor-2	Factor-3	Factor-4	Communalities
pH	-0.11	-0.17	-0.68	-0.34	0.62
EC	0.96	0.03	0.16	-0.02	0.96
TDS	0.96	0.03	0.16	-0.02	0.96
Na	0.62	0.38	0.54	-0.30	0.91
K	0.88	0.14	0.13	-0.01	0.82
Ca⁺²	0.16	0.85	0.29	-0.02	0.84
Mg⁺²	0.70	-0.44	-0.17	0.23	0.76
Cl^-	0.65	0.55	-0.06	-0.06	0.73
HCO₃^-	0.17	0.13	0.82	-0.08	0.73
NO₃^-	-0.06	0.03	0.052	0.93	0.87
SO₄^-2	-0.34	-0.41	0.47	-0.36	0.63
PO₄^-3	-0.09	0.79	0.06	0.10	0.65
Eigen Values	4.63	2.19	1.49	1.16
% Variance	43.53	23.43	19.62	13.41

Table: 5: Percentage Variance explained by Factors for post monsoon water samples

Rotated Factor Pattern				Communalities
	Factor1	Factor2	Factor3	Communalities
pH	0.48	-0.02	0.73	0.77
EC	0.95	-0.22	0.06	0.96
TDS	0.95	-0.22	0.06	0.96
Na	0.95	-0.10	0.15	0.94
K	0.16	-0.50	-0.11	0.29
Ca⁺²	0.24	-0.79	0.40	0.85
Mg⁺²	-0.31	0.74	0.07	0.66
Cl^-	-0.15	0.17	0.93	0.93
HCO₃^-	0.93	-0.07	-0.16	0.90
NO₃^-	0.07	0.78	0.10	0.62
SO₄^-2	0.70	-0.40	0.36	0.79
Eigen Values	5.31	1.82	1.56
%Variance	52.24	27.13	20.62

RESULTS AND DISCUSSION:

Pre Monsoon:

For pre monsoon season, the first four factors show Eigen values >1, hence these four factors were considered. In the post monsoon period only three factors have Eigen values>1 and hence three factors were considered.

Factor-1 for Pre Monsoon: The pre monsoon factor-1 loaded heavily with TDS, Na, K, Mg and Cl^- (Table-3) indicating that factor-1 can be associated with the salt water inundation which leached into the aquifer system, enhances the concentrations of these ions by its percolation and longer residence time. This factor accounts for 43.53% of the variance concentration of the ground water samples and is a higher percentage than attributed to the other factors. Uppada canal mouth is highly affected by the tidal active and due to the river water seepage into the aquifer system as such the quality of ground water was depleted.

Factor-2: Pre Monsoon samples include mainly Calcium and Phosphate and this factor accounts for 23.42% of variance. The higher contribution of Ca and Phosphate indicate the excessive interaction of water with aquifer formations due to the utilization of Fertilizers for crops in surrounding areas. Factor-3of Pre Monsoon samples is represented by HCO₃^-accounts for 19.62%as the shallow aquifer is intensively used for Agricultural purposes. Factor-4of Pre Monsoon sample is represented by Nitrate and accounts for 13.4% and generally the concentration of Nitrate in sea water in much greater than in continental water. Hence Factor-4 also can be associated with the salt water inundation which leached into the aquifer system increases the concentration of the ion by its percolation and longer time of stay during high tide times.

Post Monsoon:

Factor-1 of Post Monsoon loaded largely with EC, TDS, Na, HCO₃^- and sulphate indicting the presence of higher concentrations of dissolved solids and resembles the concentrations of sea water and accounts for 52.24%of variance. The samples are represented by Na and HCO₃^- as the shallow aquifers are intensively used for agricultural purposes. Higher concentrations of Na also indicate the leaching and dissociation of secondary salts in the pore spaces.

Factor-2 of Post Monsoon accounts for 27.13% represented mainly by Mg and NO₃^-. The higher concentrations of Mg may be due to sediment water interaction and weathering process and also indicate that this can be associated with the salt water inundation during high tide. Factor-3 accounts for 20.62% of variance and indicates mainly with pH and Cl^- . Higher concentrations of chloride may be due to the salt water intrusion affecting the aquifers.

CONCLUSION:

The result of multivariate statistical analysis as applied to the chemical analytical data set of ground water in the present study coastal areas provides an insight into the underlying factors controlling hydro geochemical processes in the region. Four factors in Pre Monsoon period indicating factor -1, (EC, TDS, Na, K, Mg, and Cl^- ), Factor -2 (Ca and Phosphate), Factor-3 (HCO₃^-) and Factor-4 ( Nitrate) expected from the data set represent the signatures of salt water intrusions, dissociation of secondary precipitates related compounds in the ground water. Factor-1 represents the Parameters with dominant concentrations and contributors to the ground water salinity.

These factors in Post Monsoon season including factor-1 (EC, TDS, Na, HCO₃^-, SO₄^-2), factor-2 (Mg and Nitrate), factor-3 (pH and Cl^-) resulted from the data set also indicate the signatures of salt water intrusion, dissociation of secondary precipitate related compounds in ground water decrease in the loads of parameters EC, TDS, SO₄^-2, Ca, NO₃^- The concentrations of Na, Mg, HCO₃^-, and Cl^- increases and pH has also become significant in factor-3 explaining its nature. The present study demonstrated usefulness of factor analysis in interpreting the hydro geochemistry data and relating those data to salt water intrusion, percolation and leaching processes occurring in general in this coastal aquifer. In the present study the salt water intrusion into the ground water in these two periods under study (Pre Monsoon and Post Monsoon) are very well represented by the factors with the loading of Cl, Ca, MG, Na, HCO₃^-. The increase in Bicarbonate concentration during hot season may be attributed to the fact that the increase in temperature accelerates the Organic content accessible to bacterial decomposition, where HCO₃^- is the final product of this decomposition¹⁰. The anthropogenic signatures are also well demarcated by NO₃and phosphate content. This technique can be extended to all coastal aquifer as a complement to standard hydro geochemical methods. Further the numerical analysis can help to resolve ambiguities and provide unique hydro geochemical information.

REFERENCES:

1 Bear, J., Chng, A.H.-D., Sorek, S., Ouazar, D. and Herrera, I (Eds.), (1999). Seawater Intrussion in Coastal Aquifers – Concepts, Methods and Practices, Kluwer Academic Publishers, The Netherlands, 625p.

2 Hitchon B, Billings GK & Kolvan JE, Geochemistry and origin of formation waters in the Western Canada sedimentary basin. 111. Factors controlling chemical composition. Geochim Acta 35: (1971), 567-598

3 Lawrence FW, Upchurch SB, Identification of recharge areas using geochemical factor analysis. Ground water 20, 1971, 680-687.

4 Seyhan ER, Van De Caried AA & Engelen GB, Multivariate analysis and interpretation of the hydrochemistry of a dolomite reef aquifer, Northern Italy. Water Resou Res 21, 1985, 1010-1024.

5 Razack M Dazy J, Hydro chemical characterization of groundwater mixing in sedimentary and metamorphic reservoirs with combined use of piper’s principal and factor analysis. J Hydrol 114, 1990, 371-393.

6 Jeyakumar R & Siraz L, Factor analysis in hydro geochemistry of coastal aquifers – A preliminary study, Environmental Geology 31, 1996, 174-177.

7 APHA, Standard methods for the examination of water and waste, 16^th edition. Apha, Awwa wef, DC. 1989

8 Davis JC, Statistics and data analysis in geology. Wiley, New York, 1973, pp 550.

9 Dalton MG & Upchurch SB Interpretation of hydro chemical facies by factor analysis. Groundwater 16, 1978, 228-233.

10 Abdo, M.H., Environmental Studies on Rosetta Brach and some chemical applications on the area extended from El-Kanater El-Khyria to Kafr El-Zayat city. Ph.D. Thesis. Faculty of Science Ain Shams Univ. Cairo, Egypt, 2002.

Received on 03.09.2013 Modified on 21.09.2013

Asian J. Research Chem. 6(12): December 2013; Page 1103-1106