ISSN : 0970 - 020X, ONLINE ISSN : 2231-5039
     FacebookTwitterLinkedinMendeley

Recent Monitoring of Ground Water Quality in and Around Industrial Area of Vellore City at Two Different Monsoon Periods, South India

M. Sangeetha Priya 1, A. Thaminum Ansari2* and V. Kanchana3

1Research and Development Centre, Bharathiar University, Coimbatore, Tamil Nadu, India.

2Department of Chemistry, Government Thirumagal Mills College Gudiyattam Tamil Nadu, India.

3Department of Chemistry, Sree Sastha Institute of Engineering and Technology, Vellore Tamil Nadu, India.

Corresponding Author E-mail: thaminumansari14@gmail.com

DOI : http://dx.doi.org/10.13005/ojc/390510

Article Publishing History
Article Received on : 13 Sep 2023
Article Accepted on :
Article Published : 13 Oct 2023
Article Metrics
Article Review Details
Reviewed by: Dr. Neeta Thakur
Second Review by: Dr. Amit Yadav
Final Approval by: Dr. Luigi Campanella
ABSTRACT:

The present study focuses on the assessment of seasonal variation in groundwater quality of in and around industrial area of Vellore City. The samples were collected seasonally and are categorized as premonsoon, monsoon and post-monsoon during April 2022 and March 2023. Eighteen physicochemical parameters were assessed for forty eight different samples collected along the region of in and around industrial area of Vellore City at two different Monsoon periods. The analysis of the water quality parameters, including pH, EC, TDS, Ca2+, Mg2+, Na+, K+, Cl-, HCO3-, CO32-, SO42-, and heavy metals, was done in accordance with BIS and WHO standards. The results of these surveys were used to pinpoint the geochemical processes taking place in this area. According to the analytical findings, there were significant variations in the water quality inclinations between samples and locations. Water management and treatment policy decisions can be made with the support of water quality analysis which can also help to identify potential health issues.

KEYWORDS:

Groundwater quality; Industrial areas; premonsoon; post monsoon; Vellore city

Download this article as: 

Copy the following to cite this article:

Priya M. S, Ansari A. T, Kanchana V, Recent Monitoring of Ground Water Quality in and Around Industrial Area of Vellore City at Two Different Monsoon Periods, South India. Orient J Chem 2023;39(5).


Copy the following to cite this URL:

Priya M. S, Ansari A. T, Kanchana V, Recent Monitoring of Ground Water Quality in and Around Industrial Area of Vellore City at Two Different Monsoon Periods, South India. Orient J Chem 2023;39(5). Available from: https://bit.ly/3M153hr


Introduction

Different processes, including organic matter degradation, rock-water interactions, aerobic respiration, iron reduction, mineral dissolution, weathering, industrial discharge effluents, and mixing of fresh and salt water, have been connected to variations in groundwater quality indicators. In many places of the world, water shortage has resulted from rising water demand over time. India is currently on the verge of a groundwater disaster, primarily as a result of poor management of water resources and environmental damage. On the quality of the groundwater and water contamination in Tamil Nadu, there is scant study 1-3. The lakes provide the majority (80%) of Chennai’s drinking water. Pumping stations used to draw drinking water from wells near river basins supply around 25% of the world’s population. The purpose of the current study is to look into the hydro-chemical changes, repercussions, and appropriateness of groundwater from March 2022 to April 2023.

Materials and Methods

48 samples of groundwater (bore well) were taken over the course of a year close to an industrial sector. All reagents and solutions were made using AR grade chemicals and double distilled water. At the sampling site itself, measurements of temperature, pH, electrical conductivity, and TDS concentrations were made. The usual methods have been used to measure total hardness, chloride, calcium, magnesium, alkalinity, sulphate, and bicarbonate 4-6. Flame photometers have been used to measure sodium and potassium. WQI 7,8 and correlation analysis have also been used to assess the quality of groundwater for potable uses and the interplay of chemical trends. In order to comprehend the effects of hydro-geochemistry and human involvement on groundwater quality, graphical approaches including Piper-Trilinear, Durov, Principal Component Analysis, Factor analysis, Gibbs ratio, SAR, and Corrosive ratio have also been used.

Results and Discussion

Seasonal Variations

The results of the physico-chemical parameters of groundwater samples are presented in Tables 1 through 3. The pH values between 6.7 and 8.5 fell within those recommended for residential use, which were 6.5 to 9.0 (USEPA, 1975), 5.5 to 9.0 (ICMR, 1975), and 7.0 to 9.0 (ICMR). Seasonal oscillations show that the pH value is highest during the monsoon and lowest during the pre-monsoon. The groundwater samples seldom have an alkaline pH. It was discovered that the EC values ranged from 502 to 1217 mhos/cm.  It indicates the presence of pollutants when the EC of water abruptly rises. Dug wells are normally between 28 and 38 metres deep, although tube wells can reach depths of over 50 metres. There aren’t many differences in the EC of samples from tube wells and dug wells. The salinity of Chennai’s groundwater rises as you move south. However, there is some salinity distribution variability seen. The EC noted changes in the research area’s groundwater quality, which according to field observations are caused by companies and dumping grounds. The rise in conductivity shows how many ions the salinity values can support [8 & 9].

Table 1: Physicochemical Parameters of Groundwater Samples of Vellore City (Pre Monsoon)

Sample Code

pH

TDS

CO32-

HCO32-

EC

Free CO2

Cl

Nit.

SO42-

TH

Ca2+

Mg2+

Na+

K+

COD

S1

6.96

986

166

46

774

26

152

16

108

234

68.4

28.8

385.0

10.9

98

S2

7.17

912

212

52

721

32

134

14

98

212

64.8

24.9

262.7

61.4

112

S3

6.92

1128

154

42

918

24

176

22

182

414

127.2

48.4

401.4

55.2

134

S4

7.74

986

250

112

774

38

164

16

112

232

76.8

24.9

84.2

55.4

98

S5

7.13

896

220

78

672

28

112

12

72

184

45.6

25.9

96

15.7

94

S6

7.18

1432

214

98

1118

30

194

28

218

580

142.8

82

186.5

82.9

128

S7

6.99

1521

234

62

1206

28

198

28

224

520

136.8

71.5

243.3

25.9

146

S8

7.86

987

282

132

896

46

167

24

112

284

85.2

34

108.5

1.9

188

S9

8.02

976

296

296

889

48

142

22

104

224

75.6

23.5

201.4

2.4

132

S10

7.98

898

292

292

684

46

124

18

78

192

52.8

24.9

186.5

45.4

116

S11

8.11

1413

254

368

1196

54

182

32

212

582

147.6

80.6

719.3

46.6

94

S12

7.76

1623

202

312

1217

44

214

34

238

640

163.2

88.3

734.9

10.9

192

S13

7.02

1112

212

102

918

28

170

26

114

408

117.6

50.8

543.0

61.4

154

S14

7.23

988

278

108

743

32

166

18

118

288

91.2

32.6

367.5

55.2

96

S15

8.02

994

292

296

774

52

172

20

128

310

117.6

27.3

677.7

55.4

142

S16

7.84

1256

264

268

923

42

198

24

152

432

130.8

51.3

581.3

15.7

188

C1

7.49

912

182

54

684

28

124

14

48

192

50.4

25.9

543.2

82.9

98

C2

7.98

983

281

102

714

36

142

16

72

234

58.8

32.6

308.5

25.9

134

C3

7.25

1112

176

48

897

22

178

22

114

342

86.4

47.5

802.4

43.7

146

C4

7.37

1167

204

44

916

32

182

24

124

356

91.2

48.9

686.4

25.1

228

C5

7.13

1217

146

40

987

18

188

32

148

412

100

58.5

719.3

91.1

98

C6

7.32

1002

222

76

814

28

164

28

98

296

64.8

45.1

512

79.0

112

C7

7.26

1008

218

74

817

26

166

28

98

312

74.4

45.1

496.2

2.9

142

C8

7.16

1078

222

64

854

24

170

30

114

372

82.8

56.1

412.2

4.2

198

C9

7.98

1438

282

186

1102

52

186

36

178

486

118.8

69.1

560.0

65.9

168

C10

8.14

1457

226

386

1023

58

190

34

182

512

128.4

71.5

420.3

72.5

232

C11

8.12

994

244

354

798

58

152

18

86

256

61.2

36.9

617.7

36.1

146

C12

7.89

1530

198

298

1147

44

198

38

198

532

140.4

71

886.1

16.1

218

C13

7.21

997

178

168

788

36

152

22

86

254

58.8

37.4

622.7

114.8

196

C14

7.45

1312

250

162

996

42

178

28

146

458

117.6

62.8

813.7

152.4

178

C15

7.54

1325

304

208

1104

54

182

32

152

456

110

65.2

887.9

102.9

98

C16

8.23

1289

270

368

998

62

178

28

138

424

100

61.4

730.7

32.8

112

W1

6.84

792

148

48

593

14

134

12

48

214

52.8

30.2

70.9

1.9

88

W2

6.79

798

150

48

598

18

146

14

52

214

52.8

30.2

84.3

8.0

96

W3

6.98

886

150

62

688

26

152

24

68

252

58.8

36.9

82.0

8.4

112

W4

6.77

912

154

38

723

14

168

28

74

312

67.2

48

196.8

10.9

98

W5

7.12

884

226

92

662

32

148

24

58

246

55.2

36.9

178.3

5.5

124

W6

6.78

787

154

44

590

18

114

14

46

212

51.6

30.2

99.3

7.9

86

W7

7.42

1212

280

104

964

38

162

36

112

424

79.2

70

591.4

10.3

142

W8

7.02

1108

216

96

886

32

154

28

98

388

74.4

63.3

527.4

67.9

186

W9

7.67

986

306

142

677

42

172

32

84

324

68.4

50.4

457.0

106.3

178

W10

7.13

987

260

64

672

28

172

28

88

328

68.4

51.3

83.2

0.9

98

W11

7.24

1321

266

68

1102

30

180

40

124

496

88.8

83.52

783.5

44.2

204

W12

6.82

1486

242

56

1114

28

187

42

147

546

92.4

94.4

834

74.8

214

W13

7.81

1234

319

193

977

44

158

32

114

432

79.2

72

645.6

54.7

202

W14

7.98

799

368

218

512

46

118

16

54

244

64.8

26

613.2

54.2

98

W15

6.96

1218

194

98

993

24

164

32

126

428

81.6

70

628.4

44.6

114

W16

7.14

896

300

82

617

28

148

26

62

298

70.8

43

512.2

10.2

156

Units: All the parameters are given in ppm, excluding EC-m.mhos/cm, pH

Table 2: Physicochemical Parameters of Groundwater Samples of Vellore City (Monsoon)

Sample Code

pH

TDS

CO32-

HCO32-

EC

Free CO2

Cl

Nit.

SO42-

TH

Ca2+

Mg2+

Na+

K+

COD

S1

7.05

946

154

44

724

32

126

8

92

228

67.2

27

392.0

8.8

86

S2

7.23

892

180

58

707

36

108

8

78

202

61.2

24

268.2

54.2

106

S3

6.98

1108

144

38

897

26

154

8

162

398

124

46

424.2

43.1

124

S4

7.88

963

224

112

747

44

144

10

96

218

74.8

22

88.2

51.1

92

S5

7.27

846

170

98

567

30

88

8

58

178

43

25

104.2

9.8

88

S6

7.28

1403

226

76

1108

34

166

12

188

564

140

79

196.2

81.2

116

S7

7.09

1497

205

67

1082

36

168

16

192

512

134

69

256.2

25.1

138

S8

7.94

962

272

126

798

48

148

12

84

272

81.2

33

112.4

1.2

178

S9

8.23

951

354

218

792

54

124

16

88

214

72.4

23

211.1

1.8

126

S10

8.03

857

360

208

518

54

102

12

52

184

50.2

24

198.2

35.4

108

S11

8.44

1392

316

292

1106

64

146

18

190

574

142

81

749.1

41.1

88

S12

7.98

1584

306

196

1118

62

164

22

192

590

158.4

78

814.2

10.2

184

S13

7.13

1089

220

86

884

42

152

14

102

394

115.4

48

543.3

58.4

144

S14

7.48

953

260

102

693

34

138

8

106

286

89

33

398.2

51.2

94

S15

8.22

964

362

212

713

58

154

10

108

302

117.2

25

712.9

54.4

140

S16

7.91

1227

330

184

887

48

164

10

138

416

124.6

50

598.4

1.6

176

C1

7.62

892

180

48

627

34

106

8

24

184

49.2

24.4

554.2

78.9

94

C2

8.12

967

249

123

677

38

108

8

52

228

54.6

33

318.2

22.9

126

C3

7.37

1096

170

38

846

28

146

8

98

336

84.2

47

832.3

42.7

132

C4

7.52

1136

180

52

884

44

164

12

102

334

89.6

44

706.2

21.1

212

C5

7.23

1198

134

36

936

20

154

20

118

402

96.2

58

749.7

90.1

86

C6

7.39

978

172

102

774

30

148

18

72

284

63.2

43

556

77.0

104

C7

7.42

984

188

96

784

28

148

16

68

298

71.2

43

532.1

2.1

136

C8

7.33

1062

210

62

802

32

154

16

96

384

80.2

60

442.4

3.8

196

C9

8.04

1412

308

138

1012

68

162

22

156

474

104

72

584.2

59.2

164

C10

8.31

1413

310

286

1013

72

174

26

168

498

124.2

70

418.2

61.2

216

C11

8.22

973

310

264

782

74

124

14

62

242

58.4

34.8

636.6

27.1

138

C12

7.99

1456

266

212

1019

48

172

18

172

498

132

66

916.2

9.1

218

C13

7.46

972

216

108

763

44

128

12

68

236

57.2

33

637.1

98.8

192

C14

7.61

1278

280

118

934

48

142

16

122

452

114.8

62

816.6

110.4

164

C15

7.68

1313

354

142

1084

68

164

18

138

446

107.2

64

888.8

89.9

96

C16

8.54

1278

332

282

984

84

162

18

116

420

97.4

62

732.3

23.4

108

W1

6.92

787

143

39

524

24

112

6

32

210

48.2

31

81.9

0.8

86

W2

6.83

774

147

39

536

24

118

8

34

212

47

32

86.4

6.4

88

W3

7.13

867

160

44

664

28

134

8

42

244

56

36

88.1

7.5

102

W4

6.82

897

142

36

685

26

126

12

58

308

65.4

48

206.1

6.8

92

W5

7.34

861

230

72

615

38

124

8

44

232

54.2

34

179.8

4.2

116

W6

6.93

737

149

43

515

20

94

10

28

208

47.3

31

102.3

6.8

82

W7

7.63

1197

266

108

914

40

138

18

94

412

72.4

70

598.6

7.8

138

W8

7.23

1087

232

66

835

38

124

20

72

356

76.8

59

536.4

61.2

178

W9

7.82

963

194

232

623

48

148

16

68

308

64.2

48

487.1

92.5

166

W10

7.18

974

206

102

617

34

134

20

78

312

64

49

85.1

0.5

86

W11

7.39

1302

196

118

965

32

152

26

112

474

82

81

788.6

32.8

196

W12

6.98

1412

208

64

988

34

158

30

128

512

88.6

88

892.1

67.4

202

W13

7.88

1202

254

244

925

52

122

22

102

408

74.6

68

652.1

49.2

194

W14

8.24

773

268

264

502

58

96

6

36

214

61.6

27

663

46.2

86

W15

7.14

1198

204

64

976

34

132

18

108

412

78.2

68

631.7

34.5

106

W16

7.19

877

266

98

584

36

118

12

48

292

67.4

43

518.1

7.9

148

Units: All the parameters are given in ppm, excluding EC-m.mhos/cm, pH

Table 3: Physicochemical Parameters of Groundwater Samples in Vellore City (Post- Monsoon)

Sample Code

pH

TDS

CO32-

HCO32-

EC

Free CO2

Cl

Nit.

SO42-

TH

Ca2+

Mg2+

Na+

K+

COD

S1

7.02

952

138

62

731

26

128

12

96

202

63

23

412.2

8.1

82

S2

7.19

898

172

72

709

32

114

12

82

185

58

21

269.4

53.1

102

S3

6.95

1112

144

42

902

24

162

18

166

366

112

43.2

448.3

43

118

S4

7.79

974

236

112

752

40

146

14

102

198

72

18.7

102.2

49.6

84

S5

7.21

861

192

84

581

28

96

10

66

154

38

21

121.1

9.2

76

S6

7.24

1413

210

94

1109

32

172

16

192

532

136

73

216.3

76.3

102

S7

7.07

1503

196

82

1114

28

172

20

202

493

126

67.9

259.4

19.2

124

S8

7.89

974

270

134

823

46

152

18

96

248

78

40.8

132.2

0.8

164

S9

8.14

959

344

238

836

50

128

18

92

388

68

65.7

242.1

0.8

114

S10

8.03

869

354

218

577

48

104

14

64

158

48

18.7

199.1

23.4

94

S11

8.18

1402

326

288

1147

62

152

20

198

556

136

79

782.8

38.2

82

S12

7.83

1598

314

192

1136

48

174

26

208

562

156

72

827.1

9.8

176

S13

7.04

1094

210

98

892

30

152

18

106

362

112

42

583.1

52.1

132

S14

7.35

967

248

126

705

32

144

14

108

262

87

28

401.4

49.8

84

S15

8.08

979

292

284

728

52

158

16

112

286

112

24

723.9

51.4

132

S16

7.85

1234

260

262

906

44

166

16

142

398

116

49

601.2

0.8

164

C1

7.54

896

140

88

639

28

106

12

32

152

48

17.2

565.2

75.2

88

C2

8.03

978

240

138

692

36

112

10

56

202

53.4

27

323.4

21.2

116

C3

7.32

1102

131

83

854

24

148

12

98

308

78

42.7

838.3

39.8

128

C4

7.49

1142

151

85

898

32

168

16

112

312

88

40

718.4

18.7

202

C5

7.19

1203

136

38

952

18

168

24

122

378

94

53

755.1

87.2

78

C6

7.38

989

180

102

789

28

152

20

74

268

61

40

578.5

75.2

94

C7

7.32

992

156

132

791

26

152

18

68

272

67.2

38.4

538.5

1.8

128

C8

7.27

1066

160

118

817

26

154

20

102

362

78.4

55.6

448.1

2.9

182

C9

8.03

1421

290

162

1057

54

164

28

163

448

92

70.8

532.3

57.1

144

C10

8.23

1429

304

298

1015

62

174

30

170

452

112

63.8

421.1

59.3

208

C11

8.17

979

354

228

789

62

128

16

62

228

52

34

648.1

26.2

132

C12

7.96

1478

376

108

1097

44

176

28

174

462

126

60

918.9

8.7

208

C13

7.32

983

202

124

779

36

132

18

72

212

54.6

29

754.1

98.2

182

C14

7.52

1287

270

132

952

42

156

20

126

434

112

59

878.8

110

158

C15

7.61

1314

356

148

1097

58

164

22

140

424

102.4

60

945.2

88

88

C16

8.44

1281

322

294

986

68

164

20

122

396

94.5

56.4

743.2

22.1

98

W1

6.88

790

142

46

554

22

116

10

36

206

47

30.4

91.1

0.7

82

W2

6.81

782

136

56

552

22

122

12

36

208

46

31.6

91.2

6.2

78

W3

7.04

874

148

58

678

24

138

16

46

232

52

34.8

89.3

7.1

98

W4

6.81

908

133

47

696

22

132

24

62

298

64.3

45.8

223.2

6.1

88

W5

7.29

869

204

102

634

32

128

12

44

218

51.2

31

184.2

3.8

102

W6

6.86

758

140

54

544

18

102

10

32

196

44.4

29.2

112.2

6.2

76

W7

7.54

1204

232

`146

932

38

142

20

98

398

66.2

69

606.2

7.1

124

W8

7.05

1094

198

104

848

36

126

22

82

324

62.4

53

537.1

58.2

162

W9

7.71

972

232

198

646

46

154

22

74

292

63.6

44

488.1

91.4

152

W10

7.15

981

200

112

637

30

142

22

80

294

62.8

45

86.3

0.4

84

W11

7.29

1308

193

123

988

30

156

30

114

446

78

75.8

796.2

32.1

184

W12

6.93

1432

182

98

992

30

164

36

132

492

86

83

894.2

66.8

188

W13

7.83

1218

266

232

944

46

132

24

104

384

72

63

658.1

49

187

W14

8.11

779

280

264

508

48

98

10

40

202

57

25.4

677.1

45.8

78

W15

7.04

1204

142

130

979

26

134

26

112

392

74

44

635.1

33

98

W16

7.15

886

225

143

591

30

126

20

52

278

63.2

41.5

542.1

7.7

132

Units: All the parameters are given in ppm, excluding EC-m.mhos/cm,pH

Correlation Analysis

By calculating the correlation coefficient, one may anticipate how an ion will explain the properties of other ions10. Between water quality metrics, the correlation coefficient (r) has been calculated 15 (Tables 1 to 3). It shows a strong association between the various metrics of water quality. Ions are strongly connected when the Correlation coefficient11 value is either +1 or -1. The ions are not correlated if the correlation coefficient is 0, and are said to be well correlated if the ratio is larger than 0.7 and moderately correlated if the ratio is 0.7 to 0.5. With the exception of bicarbonates and carbonate, total dissolved solids are shown to have good season-to-season correlation with cations and anions. Pre-monsoon has the highest pre-monsoon correlation coefficient for cationic concentration vs. total dissolved solids and the lowest monsoon correlation value. Total hardness was correlated with calcium, magnesium, chloride, Sulphates, carbonate, and bicarbonate, with correlation coefficients of 0.87, 0.95, 0.86, 0.88, 0.06, and 0.34, respectively, indicating that permanent hardness predominated in the study area throughout all seasons. Only nitrate and chloride have a moderate correlation with chemical oxygen demand (COD). Between overall hardness and electrical conductivity, a very strong positive association (0.95) was found. 10-12.

Piper and Durov analysis

It is commonly known that interpretive diagrams can be used to better understand the nature and origin of various water quality. In this instance, the relationship between various points in the systems and potential drivers can be expressed using the Durov12 and Piper diagram. The quality of the groundwater in the research area is depicted by the Durov diagram in Figure 1. The fact that the water in later boreholes has a higher Na-K-HCO3 character than calcium, magnesium, or Sulphates dominations could mean that sodium and chloride are neutralising the acidity in the subsurface throughout pre-, monsoon, and post-monsoon. By graphing the percentages of chemical elements in a Piper diagram, groundwater is further assessed to identify the facies 13-14. The seasonal plot shows a sporadic distribution with slight differences in their chemical properties (Fig.2). Although the number of samples varied, the groundwater was of the kinds Na-CO3, NaCl, Ca-MgCO3 and Ca-MgCl, as shown by the groundwater samples S4, C5 and W10. However, there were considerable differences in the percentage of samples that belonged to different types of water. There are only a few samples (S4, C5, and W10) that fall into both Ca-CO3 and Na-Cl sub-blocks in the figure. Plots provide evidence that the groundwater was mixed type and that several processes contributed to its evolution 15-18. Plots also showed that Na is the most abundant cation in groundwater, followed by Ca and Mg, and Cl is the most abundant anion. The main sources of ions are Na2CO3 and Na2SO4, which are extensively employed in the paper industry and in the production of small-scale dyes at various stages of the process. The study area’s groundwater types were identified and categorized according on where they fell on a Piper diagram. The Na-CO3 dominated facies was clearly visible in the majority of the sample.

Table 4: Correlation Coefficient matrix for ground waters of Vellore city, South India (Pre-monsoon).

Variables

pH

TDS

CO32-

HCO32-

EC

Free

CO2

Cl

NO32

SO42-

TH

Ca2+

Mg2+

Na+

K+

COD

pH

1

TDS

0.255

1

CO32-

0.644

0.099

1

HCO32-

0.858

0.379

0.454

1

EC

0.227

0.959

0.062

0.341

1

Free CO2

0.906

0.372

0.668

0.895

0.348

1

Cl

0.147

0.841

0.004

0.252

0.812

0.226

1

NO32-

0.100

0.774

0.181

0.190

0.752

0.220

0.740

1

SO42-

0.257

0.916

0.039

0.387

0.905

0.353

0.845

0.593

1

TH

0.145

0.951

0.084

0.308

0.909

0.271

0.858

0.825

0.883

1

Ca2+

0.289

0.853

0.080

0.436

0.825

0.370

0.846

0.550

0.945

0.871

1

Mg2+

0.025

0.895

0.058

0.180

0.852

0.165

0.764

0.891

0.736

0.952

0.680

1

Na+

0.302

0.589

0.220

0.367

0.571

0.375

0.482

0.578

0.424

0.580

0.464

0.565

1

K+

0.108

0.270

0.127

0.083

0.244

0.227

0.187

0.195

0.218

0.240

0.205

0.226

0.468

1

COD

0.230

0.532

0.170

0.231

0.476

0.285

0.505

0.569

0.377

0.496

0.386

0.504

0.453

0.142

1

Table 5: Correlation Coefficient matrix for ground waters of Vellore city, (monsoon).

Variables

pH

TDS

CO32-

HCO32-

EC

Free

CO2

Cl

NO32

SO42-

TH

Ca2+

Mg2+

Na+

K+

COD

pH

1

TDS

0.246

1

CO32-

0.635

0.079

1

HCO32-

0.852

0.367

0.456

1

EC

0.218

0.958

0.038

0.328

1

Free CO2

0.905

0.364

0.666

0.891

0.339

1

Cl

0.151

0.845

0.004

0.237

0.818

0.226

1

NO32-

0.085

0.771

0.157

0.184

0.747

0.210

0.744

1

SO42-

0.260

0.920

0.040

0.382

0.910

0.355

0.842

0.598

1

TH

0.134

0.950

0.063

0.300

0.907

0.264

0.861

0.823

0.887

1

Ca2+

0.296

0.859

0.090

0.434

0.833

0.376

0.842

0.560

0.946

0.879

1

Mg2+

0.003

0.895

0.020

0.170

0.851

0.151

0.776

0.889

0.747

0.953

0.696

1

Na+

0.288

0.585

0.188

0.339

0.567

0.360

0.496

0.564

0.426

0.571

0.470

0.553

1

K+

0.104

0.264

0.124

0.089

0.237

0.225

0.179

0.192

0.216

0.236

0.205

0.221

0.451

1

COD

0.233

0.518

0.256

0.244

0.453

0.281

0.512

0.532

0.379

0.470

0.375

0.466

0.459

0.088

1

Table 6: Correlation Coefficient matrix for ground waters of ground waters of Vellore city, (post- monsoon).

Variables

pH

TDS

CO32-

HCO32-

EC

Free

CO2

Cl

NO32

SO42-

TH

Ca2+

Mg2+

Na+

K+

COD

pH

1

TDS

0.261

1

CO32-

0.859

0.340

1

HCO32-

0.822

0.219

0.757

1

EC

0.250

0.956

0.300

0.181

1

Free CO2

0.912

0.352

0.895

0.857

0.334

1

Cl

0.160

0.792

0.177

0.132

0.796

0.218

1

NO32-

0.077

0.718

0.208

0.158

0.663

0.219

0.630

1

SO42-

0.257

0.913

0.353

0.239

0.903

0.349

0.802

0.544

1

TH

0.176

0.926

0.306

0.234

0.910

0.300

0.775

0.725

0.872

1

Ca2+

0.260

0.831

0.360

0.289

0.820

0.332

0.815

0.448

0.928

0.841

1

Mg2+

0.135

0.838

0.264

0.167

0.822

0.269

0.661

0.769

0.711

0.938

0.630

1

Na+

0.298

0.558

0.251

0.299

0.550

0.305

0.442

0.486

0.398

0.501

0.455

0.427

1

K+

0.064

0.220

0.002

0.080

0.214

0.111

0.192

0.218

0.178

0.144

0.179

0.087

0.415

1

COD

0.231

0.520

0.234

0.230

0.462

0.255

0.507

0.577

0.381

0.442

0.364

0.464

0.467

0.097

1

Figure 1: Durov Diagram of WQPs of Vellore city, South India.

Click here to View Figure

Figure 2: Piper-Trilinear Diagram of WQPs of Vellore City, South India.

Click here to View Figure

Geochemical Process

It is crucial to recognize and comprehend the hydro-geochemical processes in order to assess the reasons for changes in groundwater quality and plan for groundwater protection. Gibbs plot14 (fig. 4.3) was used to pinpoint the mechanisms governing the chemistry of the groundwater. The majority of the data points, with the exception of a few in the evaporation zone, are plotted in the rock dominance zone, indicating that the interaction of aquifer material and water is the primary process regulating the chemistry of groundwater.

The Na/Cl ratio plot and the Na/Cl ratio against EC plot (fig. 4.4 a and b) demonstrate that evaporation is not a significant process. Assuming that no mineral species precipitate, concentration by evaporation would leave the ratio of Na/Cl constant. Another Na/Cl versus EC diagram would result in a horizontal line15. Groundwater has a wide range of Na/Cl ratios (fig. 4a). The connection between the Na/Cl ratio and EC (fig. 4b) is slightly sloped, which suggests that evaporation is not the major process. Na levels in groundwater are slightly higher, which suggests that silicate weathering is more likely to be the main cause than evaporation. 19-25.

Corrosive ratio and ion exchange reaction

The corrosive ratio of a groundwater sample is greater than 1, regardless of the season. If the CR is less than 1, the water is not corrosive; if the CR is greater than 1, the water is. This is brought on by the interaction of surface moieties and industrial wastewater. The significant geochemical processes that regulate the occurrence and distribution of ions in groundwater are known as cation exchange reactions. The rise in sodium in a gneissic environment is probably caused by ion exchange or industrial or agricultural contaminations 17. Cation exchange reactions are demonstrated by a high concentration of Na relative to Cl or a depletion of Na relative to Cl. [18]. Ca is kept in the aquifer material during a typical ion exchange event, while Na is discharged into the water. Cl does not counteract the excess Na produced by the ion exchange reaction; instead, alkalinity or SO4 do. Similar to this, in a reverse ion exchange, Ca is released to water while Na is kept by aquifer minerals. In this instance, Ca and Mg balance off the excess Cl over Na. In light of this, an excess of Na over Cl or Cl over Na is a reliable indicator of ion exchange processes. The depletion of Na values relative to Cl in this region (fig. 4a) is indicative of an ion exchange reaction. Every other sampling site groundwater sample uses the ion-exchange reaction with a slow rate of seasonal fluctuations, with the exception of this location (S4, S11, W2, W7, W8, W10, W11). (fig 4b).

Principal Component Analysis

Pre-monsoon season causes factor 1 to be very highly loaded with TDS, TH, and strongly loaded with EC, chloride, NO3, SO4, Ca, and Mg (Table 4.7 and fig. 4.9). The moderate loading of Cu, Zn, Pb, and COD accounts for 46.03% of the data set’s variability. Anthropogenic pollution and a decline in the groundwater table are the two processes that are suggested. Without recharge, the groundwater table lowers during the summer because significant concentrations of chloride and Sulphates were observed. Due to the presence of NO3 in this factor, anthropogenic pollution is proposed as the additional contributing process. Livestock waste and municipal landfills may be sources of nitrogen. Factor 2 accounts for 14.64% of variability and contains the variables PH, alkalinity, and free CO2. If PH and free CO2 have somewhat higher positive loadings than alkalinity, this indicates that groundwater in the research area is primarily contaminated by wastewater discharge on a regular basis. Because the pH rises as a result of the creation of acids caused by the decomposition of organic material, this component is known as the degradation factor. 25-29.

Figure 3: Gibbs diagram for groundwater with respect to anion and cation.

Click here to View Figure

Figure 4: (a). Relation between Na (meq/l) and Cl (meq/l). (b)  Relation between Na/Cl versus EC (mho/cm).

Click here to View Figure


Figure 5: % of Na in groundwater, Vellore city, South India       

Click here to View Figure

Figure 6: SAR in groundwater, Vellore city, South India.

Click here to View Figure

Figure 7: Corrosive ratio of groundwater in Vellore city, South India

Click here to View Figure

Figure 8: Chloro-Alkaline indices of groundwater in Vellore city, South India.

Click here to View Figure

Aluminium has a modest link with factor 3 and mercury and cadmium have a moderate correlation. Factor 3 contributes 11.6% of variability. These numbers suggest that the concentration of heavy metals decreases over the summer. Factor 4—organic matter degradation/iron reduction process—explains 6.34% of the variability and includes only a moderate association of ferrous. Iron reduction is connected to the microbial breakdown of organic substances in the aquifer. (1).

Table 7: PCA studies of Physic-chemical parameters (Pre-Monsoon).

Variable

F1

F2

F3

F4

pH

0.369

0.747

-0.474

0.032

TDS

0.936

-0.230

-0.155

-0.006

TA

0.477

0.749

-0.332

0.211

EC

0.894

-0.273

-0.221

-0.004

Free CO2

0.502

0.727

-0.378

0.106

Chloride

0.843

-0.312

-0.105

-0.050

Nit.

0.841

-0.181

0.291

0.047

SO42-

0.847

-0.273

-0.371

-0.040

TH

0.925

-0.317

-0.071

0.056

Ca

0.829

-0.216

-0.376

-0.021

Mg

0.866

-0.351

0.129

0.091

Cu

0.529

0.368

0.149

-0.308

Zn

0.605

0.238

0.071

-0.376

Al

0.278

0.426

0.488

-0.385

Fe

0.359

0.183

0.159

0.610

Pb

0.559

0.227

0.537

-0.080

Hg

0.553

0.224

0.560

0.027

Cd

0.394

0.106

0.565

0.561

COD

0.635

0.111

0.225

-0.228

Eigen value

8.747

2.782

2.209

1.205

Variability (%)

46.037

14.640

11.628

6.341

Cumulative %

46.037

60.676

72.304

78.645

Figure 9: PCA distribution diagram of physicochemical parameter of groundwater in Vellore City (Pre-monsoon) South India.

Click here to View Figure 

The factor 1 is extremely significantly connected with TDS during monsoon season (Table 8 and fig. 10), while EC and TH are strongly correlated with chloride, nitrate, Sulphates, calcium, and magnesium. Cu, Zn, and COD have a moderate association and account for 46.03% of the data set’s variability. Since chloride, nitrate, and Sulphates have lower positive correlations than the other two seasons and consequently have a negative effect on TDS, this may be owing to the recharge effect of rainwater. For this reason, this component is referred to as a solid factor. Factor 2 accounts for 14.72 percent of the data set’s variability and adds to the modest correlation of Al, Pb, and Hg. This element is thought to be a heavy metal dissolution element. The dissolving of metal during the aquifer’s recharge by rainfall may be the cause of the trace amounts of Al, Pb, and Hg that are released into groundwater. Factor 3 is responsible for 12.47 percent of the variability and includes strongly positive loadings of pH, alkalinity, and free CO2. The water’s pH may alter, most commonly as a result of continuous water inflow, and this variation may have an impact on the free CO2 level. When monsoon season arrives, the pH changes as a result of an abrupt influx of fresh rainwater in the research area. Alkalinity is directly impacted by pH variation. This element is known as the pH factor. Factor 4 explains 7.4% of the variability and has a moderately positive loading of Fe and Cd; this may be because ferrous metal dissolves during microbial degradation with the help of organic matter derived from waste water in the environment. The factor 1 is very significantly linked with TDS, EC, and TH in the post-monsoon (table 9 and fig. 11), as well as with strong positive loadings of chloride, nitrate, Sulphates, calcium, and magnesium. Strong association between Cu and Pb and moderately positive loading of Zn, Hg, and COD explain 44.18% of the data set’s variability.

Table 8:  PCA studies of Physico-chemical parameters (Monsoon).

Variable

F1

F2

F3

F4

pH

0.408

0.366

0.775

-0.137

TDS

0.926

-0.304

-0.023

0.008

All.

0.505

0.391

0.723

0.080

EC

0.890

-0.350

-0.006

-0.011

Free CO2

0.558

0.371

0.686

0.031

Chloride

0.798

-0.301

-0.051

-0.138

Nit.

0.798

0.210

-0.302

0.165

SO42-

0.831

-0.487

0.122

-0.065

TH

0.906

-0.339

0.119

0.075

Ca

0.790

-0.492

0.198

-0.074

Mg

0.851

-0.189

-0.307

0.168

Cu

0.552

0.497

-0.187

-0.287

Zn

0.640

0.125

0.082

-0.328

Al

0.246

0.681

-0.296

-0.225

Fe

0.336

0.176

0.192

0.690

Pb

0.557

0.537

-0.346

0.001

Hg

0.539

0.467

-0.404

-0.112

Cd

0.335

0.307

-0.202

0.732

COD

0.636

0.219

-0.170

-0.177

Eigen value

8.521

2.811

2.369

1.414

Variability (%)

44.845

14.793

12.470

7.440

Cumulative %

44.845

59.638

72.108

79.547

Figure 10: PCA distribution diagram of physicochemical parameter of groundwater in Vellore City (Monsoon), South India.

Click here to View Figure

Since the content of chloride and Sulphates is decreased, this process is responsible for the dilution of groundwater. The substantial positive loading of nitrate, which is very low compared to other seasons, may be due to the monsoon’s end’s recharge effect on rainwater. The weathering or evaporation of groundwater reduces the calcium concentration, and the change in overall hardness affects the calcium level. Alkalinity, pH, and free CO2 have a moderately favorable association and account for 14.09% of the variability. The post-monsoon season is when free CO2 levels are at their highest. The presence of free CO2, which changes pH, has an impact on the level of alkalinity. One parameter, the moderately correlated Cd, was produced by factor 3 and accounts for 12.7% of the variability. This is the process that causes heavy metal to dissolve. Factor 4 contributes 7.4% of the variability, resulting in a moderately positive ferrous loading; the process attributed may be the process of iron reduction or the decomposition of organic matter.

Factor Analysis

The assessment criterion for groundwater environmental quality is applied, and the standard evaluation indexes are produced using PCA as well (19–25). The total score for each standard level is displayed in (Table 10 and fig. 12). The quality of the groundwater in samples W1, W3, W4, W7, W11, W13, S1, S5, S8, S9, S10, S11, S13, S14, S15, and S16 is good, and W2, W5, W16, S2 and S4 are better; the groundwater there satisfies the requirements of the II and III water function zones W6, W8, W9, W14, W15, and S3.

Table 9:  PCA studies of Physico-Chemical parameters (Post-Monsoon)

Variable

F1

F2

F3

F4

pH

0.413

0.601

-0.616

0.025

TDS

0.927

-0.261

-0.030

0.012

TA

0.515

0.621

-0.506

0.179

EC

0.904

-0.297

-0.088

-0.009

Free CO2

0.549

0.603

-0.494

0.175

Chloride

0.816

-0.332

-0.012

-0.121

Nit.

0.781

0.003

0.458

0.084

SO42-

0.849

-0.390

-0.251

-0.025

TH

0.911

-0.309

0.033

0.159

Ca

0.808

-0.378

-0.308

-0.075

Mg

0.850

-0.213

0.204

0.259

Cu

0.495

0.423

0.055

-0.433

Zn

0.623

0.173

-0.084

-0.381

Al

0.060

0.474

0.380

-0.287

Fe

0.395

0.238

-0.001

0.606

Pb

0.486

0.366

0.482

-0.109

Hg

0.572

0.345

0.530

-0.103

Cd

0.189

0.304

0.634

0.562

COD

0.623

0.173

0.238

-0.347

Eigen value

8.395

2.677

2.425

1.425

Variability (%)

44.184

14.091

12.763

7.498

Cumulative %

44.184

58.275

71.039

78.536

Table 10: Factor analysis of groundwater in Vellore city.

Sample code

F1

F2

Factor
score

Rank

Grade

S1

-1.692

0.736

-0.956

1

II

S2

-1.543

0.529

-1.014

2

III

S3

-1.17

0.545

-0.625

1

II

S4

-0.818

1.087

0.269

1

II

S5

-1.116

-0.182

-1.298

2

III

S6

-2.182

0.244

-1.938

3

III

S7

0.238

-0.021

0.217

1

II

S8

0.155

0.651

0.806

3

IV

S9

-0.526

-1.186

-1.712

3

III

S10

-0.485

-0.083

-0.568

1

II

S11

1.026

1.043

2.069

4

IV

S12

1.756

1.421

3.177

4

V

S13

0.496

-0.82

-0.324

1

II

S14

0.864

-0.009

0.855

3

IV

S15

0.518

1.441

1.959

3

IV

S16

-0.569

-0.647

-1.216

2

III

C1

-0.674

0.927

0.253

1

II

C2

-0.937

-0.059

-0.996

2

III

C3

-0.037

1.976

1.939

3

IV

C4

-0.581

-0.396

-0.977

2

III

C5

-1.785

-0.242

-2.027

1

I

C6

0.915

1.822

2.737

4

V

C7

0.85

1.43

2.28

4

V

C8

-0.154

-0.464

-0.618

1

II

C9

-0.303

-1.767

-2.07

1

I

C10

-1.313

-2.441

-3.754

1

I

C11

1.192

-0.771

0.421

1

II

C12

2.324

0.473

2.797

4

V

C13

-0.068

0.365

0.297

1

II

C14

-0.51

-0.209

-0.719

1

II

C15

-0.543

-1.977

-2.52

1

I

C16

0.473

-0.68

-0.207

1

II

W1

-1.758

-0.09

-1.848

3

III

W2

-1.251

-1.02

-2.271

1

I

W3

-0.063

1.251

1.188

3

IV

W4

0.267

0.936

1.203

3

IV

W5

0.597

1.974

2.571

4

V

W6

-0.258

0.543

0.285

1

II

W7

0.059

-0.005

0.054

1

II

W8

0.117

0.343

0.46

1

II

W9

1.154

-0.301

0.853

3

IV

W10

1.748

-1.643

0.105

1

II

W11

-0.109

-1.868

-1.977

1

I

W12

2.108

0.008

2.116

3

IV

W13

0.096

-0.725

-0.629

1

I

W14

1.065

-0.126

0.939

3

IV

W15

1.485

-0.443

1.042

3

IV

W16

0.945

-1.571

-0.626

1

I

Figure 11: PCA distribution diagram of physicochemical parameter of groundwater in Vellore City (Post-monsoon), South India.

Click here to View Figure

Figure 12: Factor analysis of physicochemical parameter of groundwater in Vellore City (Post-monsoon), South India.

Click here to View Figure

Conclusion

Studies on the trends in groundwater quality are being conducted in the Vellore area close to industrial areas. Na, Ca, Mg, and K were the four cations that predominated in the research region, whereas CO3, Cl, HCO32-, and SO42- were the four anions. The cation exchange process regulates the chemistry of groundwater. The correlation between the different water quality parameters (TDS, TH, Ca2+, Mg2+, and Cl, SO42-, etc.) was good. Ion exchange reactions along the groundwater flow direction result in the release of Ca2+ and the adsorption of Mg and Na, according to hydro chemical modeling. Based on the primary ion chemistry of groundwater, four hydro chemical facies have been discovered, with Na-K-HCO3 and Na-CO3 being the two major facies. A small number of groundwater samples from coastal areas in the pre-monsoon season (April 2023) exhibit the Na-Cl facies. However, during the monsoon, these samples eventually become diluted to Ca-Mg-HCO3 or Na-K-HCO3 facies.  The study locations’ potential for agricultural activities is constrained by high SAR and Na%. The pollutant load was definitely higher during the summer and lower during the monsoon, according to the PCA study. According to factor analysis, the water function zones for the groundwater quality in 29 places are II and III, while the other 19 locations are IV and V.

Acknowledgment

The author thanks Bharathiar University, Coimbatore for providing opportunity to completing this research works. The author also thanks to the Editors of this journal for considering my research manuscript.

Conflict of Interest

The authors declare no conflict of interest.

References

  1. Krishnamoorthy, N.; R. Thirumalai,; M. Lenin Sundar,; M. Anusuya ; P. Manoj Kumar, E. Hemalatha; M. Mohan Prasad, and Neha Munjal; Urban Climate.,  (2023), 49, 101436.
  2. Kistan, A. ; Kanchana V.,; Indian J. of Environmental Protection., 2020, 40, 5, 527
    – 532.
  3. Senthilkumar, Mohanavelu, and Devadasan Gnanasundar.;  Environmental Earth Sci., 2022, 81, no. 6 (2022): 187
  4. Reddy, D. V., Devender Kumar, and Kiran Kumar Bhukya.; Environmental Earth Sci.,2023,  82, no. 13 (2023): 335.
  5. Anjum, Rukhsar, Sk Ajim Ali, Mansoor Alam Siddiqui, Farhana Parvin, Zainab Khan, Nishat Khan, Zeba Khanam, and Mohammad Nafees. Chemical Engineering Journal Advances.; 2023, 16, 100547.
  6. Srihari, S., T. Subramani, V. N. Prapanchan, and Peiyue Li. Hum. Ecol. Risk Assess., 2022,  29, 1-12.
  7. Gogoi, Ashini, Shaik Mahamad, Dipjyoti Gogoi, and G. Bhaskaran.  Arabian Journal for Science and Engineering., 2023, 48(6), 7985-8000.
  8. Sangaré, Lamine Ousmane, Sidy Ba, Oumou Diallo, Diakalia Sanogo, and Tong Zheng. Environmental Monitoring and Assessment. 2023, 195(6), 1-19.
  9. Libina, R. S., K. Prakash, R. Jegankumar, S. P. Dhanabalan, and M. A. Arya. In Adv., in Water Resource Planning and Sustainability, Singapore: Springer Nature Singapore, 2023, pp. 203-221.
  10. Rajendiran, Anbuchezhian.; Envi.,  Monitoring and Assessment., 2023,  195(7), 883.
  11. Subramanian, Keerthivarman G., Manikandavelu Dhanushkodi, Aruna Satyapriyan, Muralidharan Nagarajan, and Govarthanan Muthusamy. Molecular Biotechnology,. 2023,  1-12.
  12. Manju, A., K. Kalaiselvi, V. Dhananjayan, M. Palanivel, G. S. Banupriya, M. H. Vidhya, K. Panjakumar, and B. Ravichandran; Air Quality, Atmosphere & Health 11., 2018,  1179-1189.
  13. Kistan, A., V. Kanchana, L. Sakayasheela, J. Sumathi, A. Premkumar, and A. Selvam.;
    Oriental J., of Chem., 2018,  34(2), 1000-1010,. http://dx.doi.org/10.13005/ojc/340250   
  14. Senthilkumar, S., K. Vinodh, G. Johnson Babu, B. Gowtham, and V. Arulprakasam.; Applied Water Science. 2019, 9, 1-20.
  15. Dhamodharan, A., S. Shanthakumar, and G. P. Ganapathy.; Global NEST Journal., 2016,  18, 3 ,  527-545.
  16. Chaurasia, Abhishek Kumar, H. K. Pandey, S. K. Tiwari, Ram Prakash, Prashant Pandey, and Arjun Ram. J. of the Geological Soci of India, 2018,  92,  76-82.
  17. Mohanakavitha, T., K. Shankar, R. Divahar, T. Meenambal, and R. Saravanan. Archives of Agriculture and Environmental Science., 2019, 4(4),  379-387.
  18. Madhavan, Shanmugam, Srinivasa Raju Kolanuvada, Vinothkumar Sampath, Priyadarsi Debajyoti Roy, Prabhakaran Moorthy, Logesh Natarajan, and Lakshumanan Chokkalingam.; Environmental Monitoring and Assessment.,2023, 195(2), 272.
  19. Packialakshmi, S., and N. K. Ambujam. “The peri-urban to urban groundwater transfer and its societal implications in Chennai, south India-A case study.” Indian Journal of Agricultural Research., 2017, 51(2), 135-141.
  20. A Kistan, Andiyappan, V. Kanchana, , Asian J. Chem., 2023, 35(8), 1899-1906., https://doi.org/10.14233/ajchem.2023.27792    
  21. Sathish, S., and L. Elango.; Arabian J.,  of Geosciences., 2016, 9, 1-17.
  22. Debsarma, Chitrangada, Paulami Sahu, and M. H. Kalubarme.;  Envir., Monitoring and Assessment;,2023, 195(9),1-24.
  23. Ramachandran, A., R. R. Krishnamurthy, M. Jayaprakash, and A. Shanmugasundharam.; Acta Ecologica Sinica.,2019, 39(2), 125-132.
  24. A Kistan, A., A. Premkumar & V. Kanchana, ;  Asian J. Chem.; 2022, 34(7), 1698-
    1702 https://doi.org/10.14233/ajchem.2022.23699.
  25. Gnanachandrasamy, G., T. Ramkumar, S. Venkatramanan, S. Vasudevan, S. Y. Chung, and M. Bagyaraj.; Applied water sci.,2015,   5,  39-55.
  26. Natarajan, Surendar;  Nature Environment & Pollution Technology 22, no. 3 (2023).
  27. Alagarsamy, Vijaya Anandan Veerasamy, Kistan Andiyappan, Saral Abdul Kadar Avuliya,
    and Thaminum Ansari Abubacker;  Oriental Journal of Chemistry., 2018,  34(1), 381.
    http://dx.doi.org/10.13005/ojc/340141 
  28. Senthilkumar, M., D. Gnanasundar, and E. Sampath Kumar. Clean and Sustainable Groundwater in India.,  (2018),  25-38.
  29. Guruge, Keerthi S., P. Goswami, I. Watanabe, S. Abeykoon, V. P. Prabhasankar, K. R. Binu, D. I. Joshua, K. Balakrishna, Masato Akiba, and N. Munuswamy.; Int.,  J., of Environmental Sci and Tech., 2017, 14, 1-18.

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

About The Author