A synoptic assessment of groundwater quality in high water-demand regions of coastal Andhra Pradesh, India

Safe water, sanitation, and hygiene form the most basic needs for human health and wellbeing. The United Nations Development Programme has set a goal of achieving clean water and sanitation by 2030 under Sustainable Development Goal 6 (SDG-6). Groundwater is one of the most important and reliable resources of freshwater but it is limited. Climate change and human interference have been impacting both the quality and quantity of groundwater in many parts of the world. A recent report on the assessment of climate change over the Indian sub-continent has highlighted that monsoonal rainfall intensity has been decreasing due to climate change leading to poor recharge of ponds, lakes, and other surface water bodies. From the current water level fluctuation data, it is estimated that 37.3% of the assessment units (Blocks/Taluks/Mandals/Districts) fall under the unsafe category, i.e. semi-critical to over-exploited in India (CGWB 2017). Another major concern is the degradation of water quality by both geogenic and anthropogenic sources. Dumping untreated industrial wastes, sewage, and agricultural wastes is a major factor degrading the surface and groundwater quality in India (Khatri & Tyagi 2015). In addition, geogenic contamination such as fluoride (Keesari et al. 2007, 2021; Jha & Tripathi 2021), arsenic, and other metals (Poonia et al. 2021), as well as salinization of freshwater aquifers (Pant et al. 2020), are some major concerns. Many coastal regions have been impacted by saline water either due to seawater intrusion or the upcoming deep-seated saline pockets (Prusty & Farooq 2020). Among anthropogenic contamination, nitrate contamination is most prevalent in many parts of India, which is mainly attributed to the contribution of agricultural and industrial wastes (Jain & Sharma 2008).

The study area falls in the southern part of Srikakulam district, Andhra Pradesh state of India and extends from latitudes 18.110° to 18.412° N and longitudes 83.582° to 83.865° E encompassing a total area of 1,153 km² (Figure 1(d)). This area is agrarian and the irrigation water needs are met mainly by groundwater. The study area is categorized as semi-critical to over-exploited. The major perennial rivers are Vamsadhara and Nagavali, both of which are east-flowing rivers and drain into the Bay of Bengal. The drainage pattern of the study area is dendritic with a drainage density varying from <0.2 to 1 km/km². The study area receives rainfall of about 1,067 mm/a, in which about 70% of the rainfall occurs during the southwest monsoon and 20% during the northeast monsoon. The maximum temperature is about 34 °C during May and the minimum is about 17.5 °C during December/January (APSAC 2018).

The geology of the study area belongs to the Archaean group and alluvium. Khondalites and Charnockites of the Eastern Ghat Super Group constitute the Archaean group of rocks and the Migmatite group includes Granitic Gneisses. Khondalites and Charnockites formations have low yields in the order of 10–20 m³/d in the weathered zones while granitic gneiss formation has higher yields of about 10–40 m³/d. Sandstone formation with limited extension is also present in the Ranasthalam mandal located in the western part of the district. The main types of soils found in the district are red soil, red loams, sandy loams, sandy soil, black soil, and alluvial soil (APSAC 2018).

Samples were collected from the existing hand pumps (HP), dug wells (DW), and tube wells (TW) in the study area, with a total of 114 samples. During the premonsoon season, 32 samples were collected to get an initial overview of the groundwater quality, however, during postmonsoon season, a detailed sampling was carried out (82 numbers) covering a wider network of wells so that monsoon-induced water quality changes can be perceived clearly. In addition, power failure and drying up of wells also limited sample availability during the premonsoon season. The sample location map is provided in Figure 1(d). The physicochemical parameters (EC, pH, and TDS) were measured in situ using a multiparameter kit (Hanna Make, HI 9829 11042). Total alkalinity describes the capacity of the sample to neutralize acids and is commonly measured at the site using titrimetric analysis, where the equivalent of acid used to reach a pH of 4.2 corresponds to the total alkalinity of the sample. In natural water samples, bicarbonate and carbonate ions are the predominant ions that determine the pH of the water. Therefore, alkalinity is expressed as (i) P-alkalinity, which refers to carbonate concentration (acid titration with phenolphthalein indicator) and (ii) M-alkalinity, which refers to the sum of carbonate and bicarbonate concentration (acid titration with methyl orange indicator). Alkalinity can also be computed using total inorganic carbon (TIC) that can be determined by wet chemical analysis where a measured sample is injected with a percentage of 1 N hydrochloric acid. The carbonates are reduced to CO₂ and are detected using a non-dispersive infrared detector (NDIR). These methods are described by Michałowski & Asuero (2012), Hassoun et al. (2015), Boyd et al. (2011) and references therein. In this study, acid-base titration is applied for total alkalinity measurement. A 10 mL water sample was titrated against 0.02 N H₂SO₄ and the end point was determined using methyl orange indicator.

The drinking water suitability was evaluated by comparing EC, TDS, and TH with the permissible limits set by WHO (2011) and BIS (2012).

The irrigation suitability was examined using SAR, Na%, PI, RSC, MH, KR, and CR.

q_i is the quality rating scale of ith parameter

C_i is the measured value of the ith WQPs; S_i stands for the permissible standard limit defined by BIS (2012) and WHO (2011) of the same parameter, C_pH is the measured pH value of a groundwater sample and S_pH is the standard value of pH.

On the basis of EWQI, groundwater can be divided into five categories, excellent (<25), good (25–50), moderate (50–100), poor (100–150), and extremely poor (>150).

A geostatistical approach was employed to understand spatio-temporal variation of the EWQI values. The Spatial analyst module of ArcGIS 10.8 software was used for this purpose. The inverse distance weighted (IDW) technique was used for preparing the spatial distribution maps. ArcGIS software has long been used for geospatial and geostatistical analysis of WQPs (Mukherjee & Singh 2022). Kriging, Ordinary Kriging, Empirical Bayesian Kriging, and IDW techniques are some of the interpolation methods used for the preparation of spatial distribution maps. IDW method is often found to be the best model for interpolation of WQPs. IDW is a type of deterministic method for multivariate interpolation with a known scattered set of points. IDW determines cell values using a linear-weighted combination set of sample points. It weighs the points closer to the prediction location greater than those farther away (Elumalai et al. 2017). IDW is generally applied to highly variable data, and it is highly possible to return to the original collection site and record a new value that is statistically different from the original value but is within the general trend for the area (Khouni et al. 2021). However, IDW assumes stationarity, i.e., the relationship between distance and influence remains constant across the study area and therefore would not provide precise information in the case where study areas exhibit large variations in water quality. For IDW interpolation power value 2 and variable search radius with 12 number of neighbor points were used. The spatial distribution maps were finally reclassified using the Raster analysis tool in ArcGIS 10.8 according to the EWQI categories.

A normality test was performed on the WQPs as well as calculating the EWQI using Shapiro–Wilk and Anderson–Darling tests. Based on the normality results, parametric one-way analysis of variance (ANOVA) or non-parametric Kruskal–Wallis tests were performed to evaluate seasonal variation of the means or medians of the WQPs and EWQI values, respectively. The statistical calculations were done using a python programming language.

EC depends on the concentration of dissolved ions in groundwater (Hem 1985). During premonsoon, the EC is in the range of 469–3,483 μS/cm with a mean value of 1,136 μS/cm, while during postmonsoon it is in the range of 245–4,400 μS/cm with a mean value of 1,413 μS/cm (Table 1, Figure 2(b)). The temporal trends in EC suggest that postmonsoon samples have slightly higher EC values compared to premonsoon. This could be attributed to the addition of salts from the soil zone. The TDS of groundwater samples is found to be in the range of 236–1,744 mg/L with a mean value of 567 mg/L in premonsoon and 123–2,200 mg/L with a mean value of 706 mg/L during postmonsoon (Table 1, Figure 2(c)). Unlike pH, large fluctuations in the TDS values (71–3,115 mg/L) are reported in earlier studies (Lal et al. 2020). Temporal trends of TDS and EC were found to be similar and the plausible reason for higher values during postmonsoon could be the leaching of salts from the unsaturated zone with the infiltrating rainwater. Total alkalinity present in groundwater during premonsoon is in the range of 195–488 mg/L with a mean value of 339 mg/L and 171–976 mg/L with a mean value of 518 mg/L during postmonsoon (Table 1, Figure 2(d)). High alkalinity gives an unpleasant taste to water (Roy et al. 2018).

The normality test results of the WQPs suggest that the majority of the parameters during both seasons do not follow normal distribution except F⁻ ion during premonsoon season (Table 2). Non-parametric Kruskal–Wallis (ANOVA) test was performed to understand the seasonal variation in median values of WQPs. The results of the test are presented in Table 3. The table suggests that the majority of the parameters do not show significant seasonal variation (p-value > 0.05) in the median values except the Ca²⁺ (p-value 0.006312) and K⁺ ions (p-value 0.04359) which show significant seasonal variation. This can be attributed to processes such as silicate weathering and cation exchange operating in the subsurface (Hem 1985).

Table 2

Results of normality test on the premonsoon and postmonsoon samples

.	N .	Shapiro–Wilk W .	p (normal) .	Anderson–Darling A .	p (normal) .
Ca2+_post	82	0.8617	3.13 × 10−7	3.631	3.82 × 10−9
Ca2+_pre	32	0.7803	1.77 × 10−5	1.743	1.43 × 10−4
Cl−_post	82	0.8185	1.21 × 10−8	4.190	1.65 × 10−10
Cl−_pre	32	0.7951	3.30 × 10−5	2.319	5.15 × 10−6
EWQI_post	82	0.9079	2.16 × 10−5	1.907	6.60 × 10−5
EWQI_pre	32	0.7730	1.32 × 10−5	2.496	1.85 × 10−6
F−_post	82	0.8402	5.79 × 10−8	3.456	1.03 × 10−8
F−_pre	32	0.9650	3.73 × 10−1	0.3729	3.98 × 10−1
K+_post	82	0.5974	1.18 × 10−13	11.480	8.34 × 10−28
K+_pre	32	0.4312	4.99 × 10−10	7.159	6.19 × 10−18
Mg2+_post	82	0.9222	9.99 × 10−5	1.970	4.60 × 10−5
Mg2+_pre	32	0.9310	4.17 × 10−2	0.894	1.98 × 10−2
Na+_post	82	0.9012	1.09 × 10−5	2.281	7.85 × 10−6
Na+_pre	32	0.8120	6.90 × 10−5	1.560	4.10 × 10−4
_post	82	0.8360	4.24 × 10−8	4.779	6.16 × 10−12
_pre	31	0.7524	7.61 × 10−6	2.452	2.35 × 10−6
_post	82	0.8431	7.23 × 10−8	3.684	2.84 × 10−9
_pre	32	0.7289	2.42 × 10−6	2.586	1.10 × 10−6

.	N .	Shapiro–Wilk W .	p (normal) .	Anderson–Darling A .	p (normal) .
Ca2+_post	82	0.8617	3.13 × 10−7	3.631	3.82 × 10−9
Ca2+_pre	32	0.7803	1.77 × 10−5	1.743	1.43 × 10−4
Cl−_post	82	0.8185	1.21 × 10−8	4.190	1.65 × 10−10
Cl−_pre	32	0.7951	3.30 × 10−5	2.319	5.15 × 10−6
EWQI_post	82	0.9079	2.16 × 10−5	1.907	6.60 × 10−5
EWQI_pre	32	0.7730	1.32 × 10−5	2.496	1.85 × 10−6
F−_post	82	0.8402	5.79 × 10−8	3.456	1.03 × 10−8
F−_pre	32	0.9650	3.73 × 10−1	0.3729	3.98 × 10−1
K+_post	82	0.5974	1.18 × 10−13	11.480	8.34 × 10−28
K+_pre	32	0.4312	4.99 × 10−10	7.159	6.19 × 10−18
Mg2+_post	82	0.9222	9.99 × 10−5	1.970	4.60 × 10−5
Mg2+_pre	32	0.9310	4.17 × 10−2	0.894	1.98 × 10−2
Na+_post	82	0.9012	1.09 × 10−5	2.281	7.85 × 10−6
Na+_pre	32	0.8120	6.90 × 10−5	1.560	4.10 × 10−4
_post	82	0.8360	4.24 × 10−8	4.779	6.16 × 10−12
_pre	31	0.7524	7.61 × 10−6	2.452	2.35 × 10−6
_post	82	0.8431	7.23 × 10−8	3.684	2.84 × 10−9
_pre	32	0.7289	2.42 × 10−6	2.586	1.10 × 10−6

View Large

Table 3

Results of Kruskal–Wallis non-parametric ANOVA test

Parameter .	H (χ2) .	Hc (tie corrected) .	p (same) .	Remark .
F−	1.655	1.656	0.1982	Not significant
Cl−	0.6929	0.6929	0.4052	Not significant
	1.8	1.8	0.1797	Not significant
	0.9007	0.9008	0.3426	Not significant
Na+	0.6364	0.6365	0.425	Not significant
K+	4.072	4.073	0.04359	Significant
Ca2+	7.456	7.459	0.006312	Significant
Mg2+	0.7087	0.709	0.3998	Not significant
EWQI	0.4468	0.4468	0.5038	Not significant

Parameter .	H (χ2) .	Hc (tie corrected) .	p (same) .	Remark .
F−	1.655	1.656	0.1982	Not significant
Cl−	0.6929	0.6929	0.4052	Not significant
	1.8	1.8	0.1797	Not significant
	0.9007	0.9008	0.3426	Not significant
Na+	0.6364	0.6365	0.425	Not significant
K+	4.072	4.073	0.04359	Significant
Ca2+	7.456	7.459	0.006312	Significant
Mg2+	0.7087	0.709	0.3998	Not significant
EWQI	0.4468	0.4468	0.5038	Not significant

View Large

Na⁺ ion concentration in groundwater ranges from 2.82 to 597 mg/L (mean value: 158 mg/L), and 19.6–504 mg/L (mean value: 160 mg/L) during pre- and postmonsoon seasons, respectively (Table 1). The concentration of Na⁺ is slightly higher during postmonsoon compared to premonsoon period (Figure 3(c)). The major sources of Na⁺ ions in this region are minerals like feldspars and clays (APSAC 2018). K⁺ ion concentration is in the range of 0.54–315 mg/L (mean value: 25.3 mg/L) in premonsoon and 0.12–246 mg/L (mean value: 26.7 mg/L) during postmonsoon (Table 1, Figure 3(d)). The major sources of K⁺ ions in the groundwaters are feldspar minerals as well as fertilizers (Hem 1985). Excessive Na⁺ ion intake is a matter of concern for the person suffering from heart disease while high K⁺ ion concentration impacts human digestive and nervous systems (WHO 2011). There are no health-based standards prescribed for Na⁺ and K⁺ ions and they are not considered as drinking water indicators (BIS 2012).

Cl⁻ ion concentration varies from 11.2 to 740 mg/L (mean value: 169 mg/L) and 10.6–995 mg/L (mean value: 194 mg/L) during pre- and postmonsoon seasons, respectively (Table 1, Figure 4(b)). High Cl⁻ ion concentration i.e. more than 250 mg/L is observed in 25% and 28% of the samples in pre- and postmonsoon seasons, respectively (WHO 2011; BIS 2012). High Cl⁻ ion concentration in waters can interrupt the microbial process of denitrification and also cause hypertension, asthma, kidney disease, and dehydration (Raviprakash & Krishna 1989). The ion concentration ranges from 3.9 to 220 mg/L (mean value: 44 mg/L) and 4.54–200 mg/L (mean value: 49.1 mg/L) in pre- and postmonsoon seasons, respectively (Table 1, Figure 4(d)). As per permissible limits ascribed by BIS (2012) and WHO (2011), only 3% of samples are sulfate-rich during premonsoon, while none of the samples fall above limits in postmonsoon (Table 4).

The presence of dissolved solids changes the taste of water, and consumption of high TDS water may lead to laxative and constipation effects in humans (WHO 2011). The suitability of groundwater for drinking based on TDS values is provided in Table 5. It is observed that, as per the TDS values 56.3% and 30.1% of groundwater samples fall in the desirable category and 31.2% and 54.2% of the samples fall in the permissible category during pre- and postmonsoon seasons, respectively. Based on the above observations, it can be inferred that most of the groundwater samples are suitable for drinking. Based on BIS (2012) guidelines, about 56% and 31% of the samples are suitable for drinking purposes during premonsoon and postmonsoon seasons, respectively. As per WHO limits about 87% and 84% of the samples are suitable for drinking purposes during premonsoon and postmonsoon seasons, respectively (Table 4).

The summary of indicators for irrigation water quality suitability is provided in Table 6. In addition, USSL, Wilcox diagrams, and Doneen's chart are also used in this study for better presentation of groundwater suitability for irrigation. EC classifies the irrigation suitability of groundwater into five different categories, in the excellent category (250 μS/cm) only 1.2% of samples fall in postmonsoon. About 34.4% and 15.4% of samples fall in the good (250–750 μS/cm) category during pre- and postmonsoon, respectively. Most of the samples fall in the permissible (750–2,000 μS/cm) category, premonsoon (53.1%) and postmonsoon (67.5%) as shown in Table 6. SAR measures the relative concentration of Na⁺ over Ca²⁺ and Mg²⁺ ions. The calculated SAR values of the groundwater range from 0.09 to 14.3 (mean value: 4.3) and 0.8 to 20.3 (mean value: 4.01) in pre- and postmonsoon seasons, respectively. About 93.8% and 97.6% of samples fall in the excellent category for irrigation in pre- and postmonsoon samples, respectively (Table 6). About 6.2% and 2.4% of samples fall in the good category in pre- and postmonsoon seasons, respectively (Table 6). From SAR values, it can be inferred that groundwater quality is excellent for irrigation purposes.

Percent sodium is also one of the very important indicators for checking the irrigation suitability of water. The use of Na⁺ ion-rich water for irrigation leads to changes in the pH of soil. Interaction of Na⁺ ion with CO₃²⁻ and HCO₃⁻ ions results in high alkalinity (pH > 7), while its reaction with Cl⁻ ions increases the salinity of soil. Binding and compaction due to the interaction of ions reduce water movement capacity in soil. Percent sodium of premonsoon water samples is in the range of 3.6–81.1 (mean value: 53.2), and 36.4–94.1 (mean value: 60.4) in postmonsoon. Only 6% of the groundwater samples fall in the excellent category in premonsoon, but none of the samples fall in the excellent category during postmonsoon. Groundwater samples in the good category are 22% and 7.3% for premonsoon and postmonsoon, respectively (Table 6). In order to reduce the Na% measures such as mulching of soil, drip or sprinkle irrigation, crop rotation and use of organic manures can be recommended. These measures will help in improving the water movement capacity and permeability of soils.

RSC is an essential parameter for the determination of the suitability of water for irrigation as it impacts the growth of crops because of soil sodication (Eaton 1950). RSC of the groundwater samples was in the range of –6.6 to 5.4 meq/L (mean value: 0.75 meq/L) and −7.7 to 24.2 meq/L (mean value: 7.5 meq/L) in pre- and postmonsoon seasons, respectively. According to the findings, 56.2% of premonsoon samples and 7.3% of postmonsoon samples fall in the good category while about 25% and 7.3% of samples are in the doubtful category during pre- and postmonsoon seasons, respectively (Table 6).

High levels of magnesium negatively impacts the soil with increasing alkalinity, making the soil unsuitable for irrigation, and also decreases crop yield, hence MH is used as an indicator for irrigation water quality. MH of groundwater samples range from 27 to 94.3 (mean value:72.6) and 19.7 to 90.9 (mean value: 54) during pre- and postmonsoon seasons, respectively. It is observed that 6.2% of groundwater samples are suitable (MH < 50) and 93.8% are unsuitable (MH > 50) for irrigation in the premonsoon season. About 24.4% of groundwater samples are suitable and 75.6% are unsuitable for irrigation in the postmonsoon season (Table 6).

A combination of two quality indicators can provide a better representation of the water quality required for irrigation. USSL diagram is used for the assessment of water quality for irrigation purposes (USSL 1954). This diagram is plotted between SAR/sodium hazard and EC/salinity hazard and 16 classes are identified in the diagram. Based on the USSL classification (Figure 5(a) and 5(b)), water samples fall under the medium to high salinity hazard and low to medium sodium hazard categories in premonsoon while medium to very high salinity and medium sodium hazard categories during postmonsoon. Previous data from the northern region of the Srikakulam district indicated that groundwater samples fall under the low to medium-hazard category (Keesari et al. 2020). This trend suggests that the quality of groundwater is further degrading in the study area compared to the northern part of the district, and this would pose a serious concern considering the high demand for groundwater in this region.

The WQI is a popular and valuable rating model that allows aggregation and conversion of the positive and negative footprints of different WQPs into single index (Kumar & Maurya 2023). In this study, water quality evaluation of the samples was carried out using the EWQI. The relative weights of different parameters such as TH, ⁠, ⁠, Ca²⁺, Cl⁻, Mg²⁺, F⁻, Na⁺, K⁺, alkalinity, pH, and TDS are calculated based on their validity for water quality indexing. Parameters with relative weight values greater than 0.10 are considered critical while weight values between 0.05 and 0.10 are considered semi-critical. According to this classification, TH in both seasons, in premonsoon, and in postmonsoon are found to be critical parameters while TDS, alkalinity, F⁻, Cl⁻, Mg²⁺, Ca²⁺ and pH are identified as semi-critical parameters. Based on the importance, the parameters can be arranged in the following order: TH > > Cl⁻ > > F⁻ > TDS > alkalinity > Ca²⁺ > Mg²⁺ > pH during the premonsoon and TH > > > Ca²⁺ > Cl⁻ > TDS > Mg²⁺ > F⁻ > pH > alkalinity during the postmonsoon (Table 7). The calculated EWQI values range from 22.6 to 116 (mean value: 43.26) in the premonsoon season and 7.05 to 139 (mean value: 43.03) in postmonsoon season. From the EWQI values, it can be inferred that most of the groundwater samples fall in excellent to poor quality during both seasons but during postmonsoon some samples fall in unsuitable quality (Table 8). The error estimates of IDW interpolation are provided in Table 9. From the table it can be found that the IDW errors do not show significant variations with power value; however, considering the spatial variations observed in the groundwater quality of the study area, interpolations by IDW might not be very precise and care should be taken to infer the results.

Table 7

Relative weight of hydrochemical parameters for EWQI

Parameters .	Relative weight (Wi) .
	Premonsoon (n = 32) .	Postmonsoon (n = 82) .
TDS	0.076	0.068
Alkalinity	0.076	0.048
F−	0.079	0.062
Cl−	0.091	0.077
	0.083	0.125
	0.121	0.092
Na+	0.085	0.086
K+	0.074	0.081
Mg2+	0.063	0.067
Ca2+	0.065	0.088
TH	0.127	0.152
pH	0.060	0.053
ΣWi	1	1

Parameters .	Relative weight (Wi) .
	Premonsoon (n = 32) .	Postmonsoon (n = 82) .
TDS	0.076	0.068
Alkalinity	0.076	0.048
F−	0.079	0.062
Cl−	0.091	0.077
	0.083	0.125
	0.121	0.092
Na+	0.085	0.086
K+	0.074	0.081
Mg2+	0.063	0.067
Ca2+	0.065	0.088
TH	0.127	0.152
pH	0.060	0.053
ΣWi	1	1

View Large

A comprehensive water quality evaluation was undertaken in parts of the coastal Srikakulam district, which has witnessed high water demand in recent times. Water samples were collected covering both spatial and temporal variations in the study area. The indicators employed for evaluating drinking water suitability were TH, TDS, EC, and chemical species including and F⁻, while irrigation suitability was examined using SAR, RSC, MH, PI, Na%, CR, and KR indicators. In addition, diagrams such as Wilcox, USSL, and Doneen's chart were used to further evaluate the water quality for irrigation use. Statistical tests were conducted on the obtained data using Shapiro–Wilk W and Anderson–Darling A tests and it was found that chemical parameters do not follow normal distribution except F⁻ ions in premonsoon. Chemical data indicate that a significant percentage of groundwater samples showed higher concentrations in Cl⁻ and TH above drinking water permissible limits, whereas Ca²⁺, Mg²⁺, and Na⁺ ions were marginally higher. High concentrations of F⁻ and ions were present in 12% and 28% of samples and the contamination levels increased during postmonsoon season. The statistical significance of seasonality in chemical ion concentration was verified using a non-parametric Kruskal–Wallis (ANOVA) test, which concluded that the majority of the parameters do not show significant seasonal variation (p-value > 0.05) except Ca²⁺ and K⁺ ions (p-value < 0.05). Based on SAR (premonsoon range: 0.09–14.3 and postmonsoon range: 0.8–20.3), groundwater falls in the excellent category during both seasons, whereas, based on RSC (premonsoon range: −6.6 to 5.4 and postmonsoon range: −7.7 to 24.2) and Na% (premonsoon range: 3.6–81.1 and postmonsoon range: 36.4–94.1), groundwater falls in the excellent to doubtful category during premonsoon and permissible to unsuitable category during postmonsoon. PI values signify that water quality is better during the premonsoon season (range: 46.4–120) compared to postmonsoon season (range: 57–201). On the other hand, based on MH (premonsoon range: 27–94.3 and postmonsoon range: 19.7–90.9) and KR (premonsoon range: 0.03–4 and postmonsoon range: 0.5–16), indicators demonstrate unsuitable nature of groundwater during premonsoon but suitable to marginally unsuitable nature during postmonsoon. CR values are higher during premonsoon (range: 0.04–3.1) but reduce during postmonsoon (range: 0.08–2.35). Based on the above observations, it can be inferred that the overall water suitability for irrigation is better in premonsoon than in postmonsoon. Shannon's entropy technique was used to estimate the EWQI to provide better appraisal of water quality. It was found that TH was the critical parameter for both seasons, followed by and ions. Spatial contours of the EWQI were prepared using the IDW interpolation method by ArcGIS software. The contours suggest that the southern part consists of good quality water compared to the northern parts during premonsoon. However, during postmonsoon the water quality is excellent to poor throughout the study area, which can be attributed to recharge by rainwater infiltration. Spatial trends in the EWQI clearly suggest that excess use of fertilizers and pesticides, as well as domestic waste discharges, play a critical role in impacting the groundwater quality of the study area. Adoption of traditional organic farming and treatment of domestic and industrial waste can improve the groundwater quality, thereby achieving sustainable groundwater management in this region.

We express our sincere gratitude to Dr S. Kannan, former Director, and Dr P.K. Mohapatra Associate Director RC& I Group, BARC, Mumbai for their constant support and encouragement during the preparation of this manuscript. Authors also thank Shri H. Kurma Rao, Shri Murali, Shri Srinivas Rao and all the field officers of District Water Management Agency, Srikakulam for their support and cooperation during the field sampling. Mr Ajay Jaryal and Mr S.N. Kamble of Isotope Hydrology Section (IRAD, BARC) for their support in measurements.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.