Groundwater quality and community health risk in Lalitpur Metropolitan City, Nepal – a geospatial analysis

Abstract

Groundwater is the main source of drinking water in the Kathmandu Valley (Nepal). To assess the groundwater quality of Lalitpur Metropolitan City (LMC) of this valley, samples were collected in winter and post-monsoon seasons from 35 shallow dug wells and 16 parameters: pH, TDS, EC, Na⁺, K⁺, Ca²⁺, Mg²⁺, Total hardness, Cl^ˉ, HCO₃ˉ, NO₃^ˉ, NH₄⁺, Fe, PO₄^3-, SO₄^2-, and Total Coliform were analysed. Hydrochemical analysis and GIS-based mapping have revealed spatial and seasonal variations in water quality. The concentration of most of the parameters was higher in the winter samples compared to the post-monsoon. Comparison with the WHO and NDWQS guidelines has revealed anomalous water chemistry. Zones hosting groundwater, suitable and unsuitable for drinking and irrigation were demarcated. However, the presence of pathogens in all the groundwater samples put the entire population of LMC under high health risk. This study bears importance in groundwater exploration, management, and prevention of water-borne diseases.

Keywords:

• Groundwater quality
• total coliform
• health risk
• Lalitpur Metropolitan City
• Nepal

1. Introduction

Water is essential for survival of life on Earth. Throughout the history of human civilisation, water has played a vital role in mankind’s development. Out of the total volume of water on the earth, 98% is saline sea water, 1.6% is groundwater and 0.004% is surface water. Groundwater, the largest available source of fresh water, is replenished by rainfall and is least polluted as it gets filtered naturally in the soil and aquifer media. Due to the steady rise in population and scarcity of surface water bodies and resources, usage of groundwater has increased manifolds. Groundwater ion chemistry is governed by both natural and anthropogenic factors. The return flow of irrigation and evaporation also influence water chemistry (Rajmohan and Elango 2006; Stewart and Aitchison-Earl 2020; Tulip et al. 2022). Owing to unplanned urbanization, rapid industrialisation, and uncontrolled use of agrochemicals, soil and groundwater resources across the world are getting polluted continuously. Mining of groundwater coupled with improper waste management in densely populated areas has resulted in the depletion of groundwater resources, degradation of water quality, and subsidence of land (Pandey et al. 2010; Rehman and Cheema 2016). Groundwater quality is also being deteriorated continuously due to an upsurge of salinity, nitrate, iron, fluoride, and heavy metals (Négrel et al. 2007; Rezaei et al. 2018; Karunanidhi et al. 2019; Tang et al. 2022) due to excessive use of fertilizers and pesticides, and disposal of untreated sewage (Gautam et al. 2013; Syafrudin et al. 2021).

The use of polluted groundwater for drinking as well as for domestic and irrigation purposes has direct and indirect adverse effects on health. Although pH does not have any direct impact, very high or low pH can affect health indirectly, since solubility and mobility of various elements are pH dependent. Highly acidic or highly alkaline water is harmful to both plant and animal growth and causes digestive problems for humans. Studies have indicated that acid-base condition in the body is influenced by drinking water which is important for the homeostasis of calcium and magnesium (Rylander 2008). Similarly, very high TDS also affects plant growth and blood circulation in human beings. Total hardness (TH) is associated with the concentration of calcium and magnesium in water. Soft water is low in calcium and magnesium content. Low calcium in drinking water may cause strokes and high blood pressure, osteoporosis, kidney stones, obesity, irregular sugar level in the blood, digestive disorder, weak tooth cavity, and rickets (Khurshid and Basheer 1998). On the contrary, consumption of hard water for a longer duration may cause gallbladder stones and kidney stones (Kozisek 2020). Besides, exceptionally high calcium concentration in blood is associated with digestive disorder, dehydration, diarrhoea, etc. Hypercalcemia is also found correlated with cardiovascular diseases (Bolland et al. 2010; Bridget and Kuehn 2013; Michaëlsson et al. 2013). Magnesium, the fourth most abundant cation in the human body plays a crucial physiological role in body functions. Magnesium deficiency in the human body may cause various gastrointestinal diseases, chronic renal failure, both Type-I and Type-II diabetes mellitus (Gullestad et al. 1992; Castiglioni 2021; Fiorentini et al. 2021), retardation of growth, weakening of muscles, osteoporosis, neuromuscular, hypertension (Rosanoff et al. 2021), and cardiovascular diseases (Swaminathan 2003; Rylander 2008; Rosanoff 2013). On the contrary, excess magnesium (>5%) leads to kidney dysfunction, and acute renal failure (Swaminathan 2003). Sodium is also a very important constituent for human health. A link has been found between sodium deficiency (hyponatremia) and diseases such as pneumonia, liver cirrhosis, congestive heart failure (CHF), acquired immune deficiency syndrome, etc. (Kovesdy 2012). Many other diseases and symptoms such as indigestion, anorexia, hormonal imbalance, arthritis, rheumatism, neuralgia, psychological disorder, and fatigue are found associated with sodium deficiency (Hurley and Johnson 2015). On the other hand, excess sodium concentration (>2.5 gm/l) in drinking water causes hypertension (Karppanen and Mervaala 2006; Cook et al. 2007; Doyle and Glass 2010), cardiovascular diseases (Strazzullo et al. 2009), and chronic kidney disease (He and MacGregor 2007). An optimum level of potassium and sodium–potassium balance is crucial for the proper functioning of human body (National Research Council 1989; Cook et al. 2009). Symptoms such as diarrhoea, vomiting, depression, muscle weakness, heart rhythm disorder, etc. are common with potassium deficiency (Rose et al. 2000). Excess chloride (>250 ppm) in drinking water not only imparts a salty taste but also causes physiological disorder. Phosphate similar to chloride and sulphate anions are acidic in nature, and it breaks the acid-base equilibrium when ingested in a high amount resulting hyper acidic condition or acidosis (Enyoh et al. 2018). Intake of high phosphate can damage bones and muscles and increase the risk of heart attacks, strokes, and kidney failures (Komaba and Fukagawa 2016). Excess nitrate concentration (>11 ppm) in adults may cause ulcers, gastric cancer (Morales-Suarez-Varela et al. 1995; Adimalla 2020), colon cancer (Ward et al. 2005), recurrent acute respiratory tract infections (Gupta et al. 2000), hypertension, and cardiovascular diseases (Napoli and Ignarro 2009). In infants, it causes methaemoglobinemia or blue baby disease (Camargo and Alonso 2006), anoxaemia, and asphyxia. Similarly, regular high intake of iron exceeding the tolerable limit (>0.5 g) through drinking water may cause adverse health impacts such as diabetes and heart diseases (Ghosh et al. 2020), and other diseases of liver, heart, and lung along with hormonal disorder and lowering of immunity (Gurzau et al. 2003). These all are possible health hazards due to the anomalous concentration of individual elements. However, the health hazards may be much more severe and complicated when anomalous high or low concentrations of multiple elements (and compounds) act together, or when a person has comorbidity.

Therefore, groundwater quality assessment and monitoring are very important to prevent health hazards. In most of the developed countries, the quality of water resources including groundwater is assessed using well-defined criteria and is regularly monitored. This is integral to water resources assessment, management, and domestic water supply planning. If water quality does not fulfil some essential criteria, or satisfy the required standard, then the water (surface water or groundwater) is treated physically and chemically before supply and use. The standards and guidelines for drinking water quality prescribed and revised by the World Health Organisation (2004, 2011) are considered and followed in countries across the globe. Besides, there are country-specific or even region-specific detailed guidelines for groundwater in general and drinking water in particular. For example, Belitz et al. (2003) of the U.S. Geological Survey presented the framework for groundwater quality monitoring and assessment for California, which is used for that region. Similarly, the European Commission proposed the Water Framework Directive (WFD) (2000/60/EC) for sustainable and integrated management of water resources across Europe. The Groundwater Working Group of the WFD further developed a policy for consistent groundwater monitoring across Europe (European Commission 2007a, 2007b). The common objectives include monitoring the status and trend of the quantity and quality of groundwater and maintaining the quality criteria of drinking water (Grath et al. 2007). Different criteria and thresholds are recommended for drinking water quality standards and environmental quality standards. Significant variations in the criteria thresholds are also found in different European countries due to local (country-wise) risk perception about the chemical constituents and pollutants (Balderacchi et al. 2013). Groundwater standards and guidelines (for drinking and irrigation purposes) are also established in most Asian countries including China, India, Japan, South Kora, and even Nepal. However, in Nepal, actual implementation of policy regarding monitoring, assessment, and management of groundwater is lacking.

Groundwater is a vital source of water supply in the Kathmandu Valley since long time. About two decades ago, 45% of people of the Kathmandu Valley were depended on groundwater for drinking and domestic uses (MOPPW. 2003). In fact, in most of the cities in Nepal groundwater is the only source of drinking water supply (Pandey et al. 2010). In recent years, the use of groundwater resources in the Kathmandu Valley has increased many folds due to a steady rise in population, unplanned urbanization, rapid industrialization, and is adversely affected in terms of quality and quantity (Pradhanang et al. 2012; Bhandari et al. 2021). However, regular monitoring and strict maintenance of groundwater quality in Nepal for human consumption, agricultural and industrial activities have not been given due attention. Thus, to protect the groundwater resources of the Kathmandu Valley, it is important to map the spatial distribution of various chemical components of groundwater to identify safe and unsafe groundwater zones. A groundwater quality map prepared based on hydrochemical analysis is useful for the demarcation of zones hosting portable groundwater (Annapoorna and Janardhana 2015). Such kind of GIS-based water quality zoning maps also help in identifying the key factors governing groundwater chemistry and their relationship with local geology and land use/land cover (Zhang et al. 2007). Maps with spatial statistics not only provide information about the occurrence of portable groundwater or its pollution level but also offer a better understanding of the chemical processes governing the subsurface water chemistry (Rezaei and Hassani 2018; Yadav et al. 2018).

Considering all these facts and factors, a GIS-based groundwater quality zoning in Lalitpur was carried out in this study with the following three objectives. (1) To assess spatial and seasonal variation of groundwater quality, (2) To demarcate groundwater safe and unsafe zones, and (3) To evaluate the number and percentage of the population of this region having high health risk due to use of unsafe groundwater.

2. Study area

Kathmandu Valley is situated in central Nepal comprising of three administrative districts: Bhaktapur, Lalitpur, and Kathmandu covering an area of 694.27 km², (Ishtiaque et al. 2017). The valley in the Himalayan foothills is spread between 27°31′ to 27°50′ North, and between 85°11′ to 85°34′ East. Altitude of this valley varies between 1213 and 2714 m above the mean sea level (msl). The valley is drained by the Bagmati River and its tributaries which also serve as the main source of water for domestic and irrigation uses (Shakya et al. 2019; Chaudhary et al. 2021). In the 1980s, the region was mostly covered by natural vegetation (forest, grassland, and shrub) and agricultural land. However, due to rapid urbanisation and population growth, there was a remarkable change in land use/land cover (LULC) in the Kathmandu Valley including LMC in the last three decades when most of the vegetated and agricultural land got converted into impervious urban land (Sarif et al. 2020). At present, more than 90% area of Lalitpur is urban land (Figure 1). Encroachment of cropland and forest located at the fringes of LMC is quite obvious in near future.

Figure 1. Study area with groundwater sample sites and the land-use classes.

Climate in this region is characterized by relatively high temperature and equitably dispersed precipitation throughout the year. The average yearly maximum, minimum, and mean temperature of the basin are 24.2 °C, 12.4 °C, and 18.2 °C, respectively (DHM 2015). Likewise the average annual rainfall depth is around 1500 mm (Pokhrel 2018) in the city and about 1800 mm in the slopes and foothills (Thapa et al. 2017). Of the total annual precipitation, 75% occurs in the rainy season (June–September), 20% in the pre-monsoon (March to May), and the rest during the post-monsoon period (October–February) (Dongol et al. 2005).

Geologically, Kathmandu Valley is covered by fluvio-lacustrine deposits of Quaternary to Pleistocene ages. Rocks in this region belong to the Chandragiri Formation, the Sopyang Formation, and recent alluvium (DMG 1998). Most parts of LMC are covered by recent alluvial deposits (gravel, sand, and silt) of Quarternary age whereas rocks of the Chandragiri Formation and Sopyang Formation (limestone, quartzite, phyllite, and meta-sandstones) are present only in a small part in its west (Figure 2).

Figure 2. Geological map of Lalitpur Metropolitan City showing rock formations and groundwater sampling points.

The top sediments are composed mainly of unconsolidated sand, silt, clay, and rock fragments (Kharel et al. 1998; Sakai 2001) forming shallow (≤60 m) and deep (>60 m) aquifers (Gurung et al. 2007). The aquifers are unconfined in nature which gets replenished through direct infiltration of rainwater (Shakya et al. 2019). Groundwater is the main source of drinking water supply in the Kathmandu Valley, especially in Lalitpur. Although Lalitpur is a metropolitan city, it hosts agricultural land on its fringes. Groundwater is extracted both for domestic as well as for irrigation purposes.

The population rise has led to the urbanisation and decreased fresh water supply. With a growth rate of 4% per year, the population in the Kathmandu Valley increased to 2.5 million (World Bank 2022). With a 128% rise in population in the Kathmandu Valley during 1991 to 2011, acute deficit of portable water has resulted. Consequently, the water demand has risen from 35.1 million litres daily (MLD) in 1998 to 470 MLD in 2020 while the average water supply was only 114.2 MLD in 2020 (KUKL 2021). To overcome the deficit (355.8 MLD) in water supply, KUKL utilises groundwater (38.52 MLD) as an alternative resource for water supply (KUKL 2021).

Lalitpur is the third biggest city of Nepal after Kathmandu and Pokhara located in the south-central portion of Kathmandu Valley and Bagmati waterway on the south side. The city is constituted of 29 municipal wards covering an area of 15.43 km². By the year 2011, it hosted a population of 226,728 under 54,748 family units (CBS 2011). With rapid population growth and settlement, Lalitpur has become one of the fastest-growing cities in South Asia (Maharjan et al. 2021). However, unplanned development in terms of rapid urbanisation, rampant industrialisation, waste mismanagement and poor sanitation have caused pollution of both surface water and groundwater, and have led to water-borne diseases in the Kathmandu Valley including LMC. Since groundwater is the main source of drinking water in LMC, groundwater quality monitoring, assessment, and treatment have the maximum importance and priority.

3. Materials and methods

3.1. Sample collection and analysis

A total of 35 groundwater samples were collected from dug wells of dispersed locations in winter (January–February 2020) and post-monsoon period (October–November 2020). All these private shallow wells (depth < 55 m) are mostly used for domestic and irrigation purposes. The reason for sample collection in winter and post-monsoon was to observe seasonal variations in the groundwater chemistry. Since rainfall in the Kathmandu Valley is very low or negligible in winter, there is little influence of rainfall-induced recharge on groundwater chemistry. On the contrary, groundwater chemistry is highly influenced by the rainfall–recharge process during the monsoon season which is reflected in the post-monsoon groundwater quality. Parameters such as pH, TDS, and EC were measured in situ by multi-probe. The samples were stored in a refrigerator at 4 °C before analysis. Chemical analysis of major ions (Na⁺, K⁺, Ca²⁺, Mg²⁺, NH₄⁺, PO₄^3-, SO₄^2-, Cl^ˉ, HCO₃^ˉ, NO₃ˉ, and Fe) were carried out in the Laboratory of Central Department of Environmental Science, Tribhuvan University using standard methods recommended by American Public Health Association (APHA 2005). Of the cations, Na⁺ and K⁺ were analysed using flame atomic absorption spectrophotometry. Total hardness (TH), Ca²⁺ and Mg²⁺ were measured using EDTA (Ethylene Diamine Tetra-acetic Acid) titration. The concentration of NH₄⁺ and Fe was determined using UV-spectrophotometer. Of the anions, Cl^ˉ was measured by Argentometric Titration method, SO₄^2-, PO₄^3-, and NO₃^ˉ by spectrophotometer, HCO₃^ˉ by alkalinity concentration, and TC by incubation at 45 °C.

3.2. Reliability and error analysis of measurements

Assessment of reliability and error of the chemical analysis of groundwater samples is important for the estimation of the measurement uncertainty, the sampling bias, as well as spatiotemporal variability in water quality (Guigues et al. 2016; Wali et al. 2020). The objectives, sampling sites, the number of samples, analytical parameters, and analysis techniques were determined a priori. The laboratory analysis including pre-treatment, preservation, and chemical analysis of samples was carried out in the laboratory of the Central Department of Environmental Science (CDES), Tribhuvan University (Nepal) by applicable national and international procedure, accredited as per ISO-17015 Standard (ISO 2005). The data analysis was performed as per the procedure of APHA (2005). Analysis of the groundwater samples of both seasons was performed in triplicate. The physicochemical components’ analysis accuracy (Topol and Mitchell 1985) was evaluated through the ion-balance error (IE) which was estimated using Equation (1)(1) $$\mathrm{IE}=\frac{\mathrm{\Sigma }\mathrm{Cations}-\text{ ΣAnions}}{\text{ΣCations }+\text{ ΣAnions}}\times 100$$(1) . (1) $$\mathrm{IE}=\frac{\mathrm{\Sigma }\mathrm{Cations}-\text{ ΣAnions}}{\text{ΣCations }+\text{ ΣAnions}}\times 100$$(1)

3.3. Groundwater contamination

The degree of contamination (C_d) method developed by Rapant et al. (1995) is widely used for the estimation of heavy metal pollution levels in the water. In this study, this method was employed for the estimation of the pollution level in groundwater using basic physicochemical parameters. In this method, the water quality is evaluated by computing the contamination level through the sum of the contamination factors of each chemical component exceeding the upper permissible limit. The C_d method thus, summarizes the combined effects of a number of quality parameters considered to be unsafe in drinking water (Rubio et al. 2000; Boateng et al. 2015). As proposed by Backman et al. (1998) the degree of contamination (C_d) may be presented as: (2) $$C_d={\sum }_{i=1}^n{C}_{fi}$$(2) where C_fi is the contamination factor calculated using Equation (3)(3) $$C_{fi}=\mathrm{ }\frac{{CM}_i}{{CS}_i}$$(3) : (3) $$C_{fi}=\mathrm{ }\frac{{CM}_i}{{CS}_i}$$(3) where CM_i, and CS_i, respectively, are the analytical value and upper permissible concentration for the i^th component. Following Rubio et al. (2000), the degree of anomalous concentration of chemicals in the groundwater was categorized as, low (C_d ˂ 1), medium (C_d = 1 – 3), and high (C_d > 3).

3.4. Mapping spatial distribution and groundwater quality zoning

The spatial distribution of the groundwater quality parameters was mapped using ArcGIS 10.2 software. Point values of the parameters at the sample locations were interpolated using Inverse Distance Weight (IDW), which is a deterministic interpolation technique. The IDW interpolation works with a simple logic: the influence of data points is inversely proportional to their distance from the target location, and hence closer data points have similar values (and vice-versa). Thus, its computation is also straightforward (Chen and Liu 2012). With well-dispersed data points, the IDW interpolation technique produces reasonably accurate interpolated values as the mathematical equation passes through the actual data points (Yang et al. 2015). Thus, the use of IDW interpolation seems logical and appropriate to generate thematic maps for different hydro-chemical parameters. The basic formula of IDW interpolation is: (4) $$v^{\mathrm{*}}=\frac{w_1{x}_1+\mathrm{ }{w}_2{x}_2+w_3{x}_3+.\dots .\mathrm{ }+\mathrm{ }{w}_n{x}_n}{w_1+\mathrm{ }{w}_2+w_3+\dots +\mathrm{ }{w}_n}$$(4) where v* is the unknown value to be determined at any location; w is the weight and x is the value at the known point.

The weight is an inverse distance of a point to each known point value that can be calculated using Equation (5)(5) $$w_1=\mathrm{ }\frac{1}{d_{ix\mathrm{*}}^p}$$(5) as shown below. (5) $$w_1=\mathrm{ }\frac{1}{d_{ix\mathrm{*}}^p}$$(5)

In Equation (5)(5) $$w_1=\mathrm{ }\frac{1}{d_{ix\mathrm{*}}^p}$$(5) , p is the power, and the higher the power; the lower will be the weight. An optimum value between 1 and 2 is generally used for power p. In this study, power 1 was used. Despite errors generated by interpolation, the integration of spatial distribution maps of various chemical components helps in the demarcation of zones with groundwater quality suitable and unsuitable for drinking and irrigation (Bhuiyan and Champati Ray 2017; Rezaei and Hassani 2018).

Concentrations of the groundwater quality parameters were compared with the permissible limits for drinking purposes set by World Health Organization (WHO 2011) and Nepal Drinking Water Quality Standard (NDWQS 2005). A groundwater quality zoning map was generated through integration of the chemical parameters with suitable ranks (Table 1). Weights (cumulative ranks) were obtained through addition of ranks of the constituent chemical parameters. To obtain a drinking water quality map, the weights were classified into five classes viz. ‘Very Good’ (weight: 16 with all 16 parameters having rank 1), ‘Good’ (weight: 17 to 24 with majority parameters having rank 1), ‘Moderate’ (weight: 25 to 32 with most of the parameters having rank 2), ‘Bad’ (weight: 33 to 40 with many parameters having rank 3), and ‘Very bad’ (weight: 40 to 47 with most of the parameters having rank 3) (Table 2). In the same way, the groundwater contamination map was generated using the C_d values and was classified following the scheme proposed by Rubio et al. (2000) as discussed.

Table 1. Weights assigned to groundwater parameters for drinking purpose.

Parameter	Unit	Categories	Weight assigned	Parameter	Unit	Categories	Weight assigned
pH		≤6.5	1	Cl-	mg/L	<100	1
		6.6-8.5	2			100-250	2
		>8.5	3			> 250	3
TDS	mg/L	≤500	1	HCO3^-	mg/L	≤150	1
		501-1000	2			151-300	2
		>1000	3			>300	3
EC	μS/cm	≤1000	1	NO3^-	mg/L	≤11.2	1
		1001-1500	2			11.3-50	2
		>1500	3			>50	3
Na^+	mg/L	≤100	1	NH4^+	mg/L	≤1.2	1
		101-200	2			1.3-1.5	2
		>200	3			>1.5	3
K^+	mg/L	≤50	1	Fe	mg/L	≤0.05	1
		51-100	2			0.06-0.3	2
		>100	3			>0.3	3
Ca^2+	mg/L	≤100	1	PO4^3-	mg/L	≤0.5	1
		101-200	2			0.6-1.0	2
		>200	3			>1.0	3
Mg^2+	mg/L	≤50	1	SO4^2-	mg/L	≤ 125	1
		51-100	2			126-250	2
		>100	3			>250	3
TH	mg/L	≤250	1	TC	MPN	< 1	1
		251-500	2			≥ 1	2
		>500	3

Table 2. Groundwater quality zoning for drinking through classification of weightage.

Total weightage	Quality classified
16	Very good
17–24	Good
25–32	Moderate
33–40	Bad
40–47	Very bad

Similarly, suitability of groundwater for irrigation was evaluated through different ionic species present in groundwater and was classified following the guidelines by Wilcox and Durum (1967) as well as by Ayers and Westcot (1985). Total dissolved solid (TDS) reflects the state of inorganic contamination and is a measure of ionized species in the aquatic environment (Debels et al. 2005). TDS and Electrical conductivity (EC) were used to measure the water salinity (Fipps 2003), along with chloride (Cl^ˉ) and are indicators of salinity hazard, the most important water quality parameter for plant production (Salifu et al. 2017). The sodium percentage (Na %) (Hussein 2004), sodium adsorption ratio (SAR) (Chen et al. 2017) and permeability index (PI) (Li et al. 2016) were calculated by using the standard formulae (Equations 6–8). Suitability of groundwater for irrigation also depends on the equilibrium state of magnesium and calcium since a high amount of magnesium imparts adverse effects on soil and plant due to high alkalinity (Doneen 1964; Al-Shammiri et al. 2005). Hence, the magnesium adsorption ratio (MAR) (Rengasamy and Marchuk 2011) was calculated using the standard formula (Equation 9(9) $$\mathrm{MAR}=\frac{\mathrm{Mg}}{\mathrm{Ca}\mathrm{ }+\text{ Mg }}*100$$(9) ). (6) $$\mathrm{Na}\mathrm{\%}=\frac{\left(\mathrm{Na}+\mathrm{K}\right)}{(\mathrm{Ca}+\mathrm{Mg}+\mathrm{Na}+\mathrm{K})}*100$$(6) (7) $$\mathrm{SAR}=\frac{\mathrm{Na}}{\surd (\mathrm{Ca}+\mathrm{Mg})/2}$$(7) (8) $$\text{PI \%}=\frac{\mathrm{Na}+\sqrt{\mathrm{H}\mathrm{CO}3-}}{\mathrm{Ca}+\mathrm{Mg}+\mathrm{Na}}$$(8) (9) $$\mathrm{MAR}=\frac{\mathrm{Mg}}{\mathrm{Ca}\mathrm{ }+\text{ Mg }}*100$$(9)

The ranks and weights were assigned to groundwater parameters for irrigation (Table 3) appropriately. The integrated map was reclassified as ‘Very good’, ‘Good’, ‘Moderate’, ‘Bad’, and ‘Very bad’ to obtain the final irrigation zoning map (Table 4).

Table 3. Weights assigned to groundwater parameters for irrigation purpose.

Classification patterns	Categories	Ranges	Weight	Winter samples	Post monsoon sample
Salinity hazard (TDS)	Excellent	<175	1	2	2
	Good	175–525	2	22	24
	Permissible	525–1400	3	11	9
	Doubtful	1400–2100	4
	Unsuitable	>2100	5
Chloride (mg/L)	Excellent	<135	1	28	32
	Good	136–355	2	7	3
	Fair	>355	3
Sodium absorption rate (SAR)	Excellent	0–10	1	35	35
	Good	10–18	2
	Fair	18–26	3
	Poor	>26	4
Sodium percentage (Na %)	Excellent	0–20	1	20	13
	Good	20–40	2	14	20
	Permissible	40–60	3	1	2
	Doubtful	60–80	4
	Unsuitable	>80	5
Permeability index (PI)	Class I	>75	1
	Class II	25–75	2	6	2
	Class III	<25	3	29	33
Magnesium adsorption ratio (MAR)	Suitable	<50	1	35	33
	Unsuitable	>50	2		2

Table 4. Groundwater quality zoning map for irrigation.

Total weightage	Quality classified
Up to 9	Very good
10	Good
11	Moderate
12	Bad
>12	Very bad

3.5. Groundwater induced health risk

To assess groundwater-induced health risk in LMC, a municipality ward-wise population map was prepared using the demographic data obtained from the National Population and Housing Census of Nepal (2011). Since the recent population census (of year 2021) is not yet published, the projected population in LMC was computed using the population growth-rate data obtained from the World Bank (World Bank 2022). Groundwater induced health risk was assessed through visual comparison and correlation of drinking and irrigation water quality maps with the projected municipality population and population density maps.

4. Results and observations

4.1. Spatial and seasonal variation in groundwater quality

The difference was noticed in the concentration of the chemical parameters in the winter and post-monsoon samples of groundwater. The concentration of majority of the chemicals was higher in the winter samples (Table 5).

For winter, $$\mathrm{IE}=\frac{\text{ΣCations }\left(10802.19\right)\hspace{0.17em}-\hspace{0.17em}\text{ΣAnions }\left(9808.44\right)}{\text{ΣCations }\left(10802.19\right)+\text{ΣAnions }\left(9808.44\right)}\times 100=\frac{2993.75}{20610.63}\times 100 =\pm \hspace{0.17em}4.82$$

For the post monsoon, $$\mathrm{IE}=\frac{\text{ΣCations }\left(5514.91\right)\hspace{0.17em}-\hspace{0.17em}\text{ΣAnions }\left(5028.47\right)}{\text{ΣCations }\left(5514.91\right)+\text{ΣAnions }\left(5028.47\right)}\times 100=\frac{486.44}{10543.37}\times 100 =\pm \hspace{0.17em}4.61$$ So, the mean ion-balance error of the samples for winter and post-monsoon seasons was respectively, ± 4.82 % and ± 4.61 % (as shown above) indicating the reliability of the measurements and little sampling bias. However, this also reflects seasonal variation in the groundwater quality.

Statistical analysis (correlation analysis) revealed an association among several chemical parameters in the groundwater of LMC, which also indicates seasonal variation. The association was identical in nature (directly or inversely proportional) between certain parameters both in winter and the post-monsoon, despite variation in their strength of association. For example, an association between EC and TDS, between iron and ammonia, and of TH with calcium and magnesium were strong and positive in both the seasons. Similarly, a strong inverse relation between pH and TC in both the seasons. However, diverse and exclusive associations could be noticed among some other parameters either in winter or in the post-monsoon as is shown in Table 6. For example, a statistically significant association was noticed for iron with magnesium as well as nitrate in winter but not during the post-monsoon. The strong NO₃^ˉ was most likely form the agricultural input (Wei et al. 2017; Li et al. 2018). Although, the use of fertilizers not only adds NO₃^ˉ in groundwater but also add Ca²⁺, Mg²⁺, Cl^ˉ, PO₄^3- and K⁺ (Hagedorn and Cartwright 2009), which can lead to a change in the groundwater geochemistry (Li et al. 2018). Similar season-specific associations were noticed among some other parameters which need further study, detailed analysis, and reasoning.

Geospatial mapping of hydro-chemical parameters obtained through interpolation of the point values also revealed the spatial variation in the groundwater chemistry of LMC, as discussed in the following sub-sections.

Table 5. Statistics of groundwater chemistry in Lalitpur Metropolitan City.

Parameters		pH	TDS	EC	Na^+	K^+	Ca^2+	Mg^2+	TH	Clˉ	HCO3ˉ	NO3ˉ	NH^4+	Fe	PO4^3-	SO4^2-	TC
Unit			mg/l	μS/cm	mg/l	mg/l	mg/l	mg/l	mg/l	mg/l	mg/l	mg/l	mg/l	mg/l	mg/l	mg/l	MPN/100 ml
Minimum	W	8.38	137	276	33.2	0.84	64	2.4	100	22.72	20	0.04	< 0.1	< 0.01	0.23	5.52	56
	PM	7.28	112	226	13.9	0.27	38.47	2.4	80	20.45	15	0.04	< 0.1	< 0.01	0.10	4.42	68
Maximum	W	9.54	1063	2127	70.3	41.5	478	61.68	612	241.4	450	0.14	0.99	10.16	3.58	77.7	254
	PM	9.54	1051	2021	79.8	50.3	389	81.87	485	220.1	105	0.3	0.46	4.19	3.65	63.22	315
Mean	W	9.15	491.71	983.63	53.65	16.21	218.4	22.21	306	103.71	148.29	0.06	0.16	1.02	0.78	21.02	123.09
	PM	7.97	466.35	913.63	55.85	15.87	158.93	22.14	253.03	80.87	39.29	0.07	0.13	0.42	0.45	16.47	151.48
SD	W	0.41	217.69	436.53	10.27	13.08	105.62	11.3	126.55	52.55	87.04	0.02	0.17	2.42	0.86	20.93	53.66
	PM	0.44	198.15	393.5	17.9	13.63	83.3	18.18	97.14	45.767	21.73	0.05	0.08	0.97	0.64	15.69	67.41
CV	W	4.48	44.27	44.38	19.14	80.69	48.36	50.87	41.36	50.67	58.7	33.07	105	236.48	109.64	99.54	43.6
	PM	5.56	42.49	43.07	32.05	85.87	52.42	85.32	38.39	56.59	55.32	79.4	60.02	230.28	140.71	95.24	44.5
NDWQS (2008)	Range	6.5 − 8.5	1000	1500	----	----	200	----	500	250	----	11.2	1.2	0.3	----	250	95% in sample
WHO (2011)	Range	6.5 − 8.5	1000	1500	200	100	100	50	500	250	300*	50	1.5	-----	1	250	95% in sample

Note: W winter, PM post-monsoon

Table 6. Association among various chemical parameters in the groundwater of Lalitpur Metropolitan City in Winter and Post-monsoon as revealed by the Pearson moment correlation coefficient.

Winter samples			Post-monsoon samples
Parameters		R	Parameters		R
pH	TC	−0.56**	pH	TC	−0.53**
EC	TDS	1.00**	EC	TDS	0.99**
TH	Ca^2+	0.92**	TH	Ca^2+	0.66**
TH	Mg^2+	0.57**	TH	Mg^2+	0.55**
Fe	NH4^+	0.62**	Fe	NH4^+	0.64**
Fe	Mg^2+	0.38*	pH	SO4^2-	0.035*
Fe	NO3^-	0.68**	TH	Na^+	0.42*
SO4^2-	K^+	0.36*	Cl^-	K^+	0.48**
Mg^2+	K^+	−0.34*	Cl^-	Mg^2+	0.39*
NH4^+	Mg^2+	0.49**	NO3^-	K^+	0.45**
NH4^+	TH	0.39*	HCO3^-	TDS	0.57**
NH4^+	NO3^-	0.45**	HCO3^-	EC	0.57**
			HCO3^-	Na^+	0.34*
			TC	Cl^-	0.43**
			TC	SO4^2-	−0.34*

Note: ^**Correlation is significant at the 0.01 level (2-tailed).

* Correlation is significant at the 0.05 level (2-tailed).

4.2. Physicochemical parameters

pH: Normal pH of groundwater varies between 6.5 and 8.0 (Islam et al. 2014). However, at LMC, groundwater was alkaline, as the pH values of most of the samples vary in the range of 8.38–9.54 (mean: 9.15) in winter and within 7.28–9.54 (mean 7.97) in the post-monsoon (Figure 3). While pH of only five post-monsoon samples was above the WHO and NDWQS limit (8.5), pH was above this threshold in as many as 30 out of 35 winter-samples (>85%). A slight decrease in groundwater pH during the post-monsoon period was probably due to the mixing of fresh rainwater. Thus alkaline nature of groundwater samples is attributed mainly to the lithology of the study area consisting of quartzite, argillaceous limestone, and dolomitic limestone of the Chandragiri, Sopyang, and recent formations (Acharya and Paudyal 2019).

Figure 3. Spatial distribution map of pH of study area.

Total dissolved solid (TDS): TDS is an important attribute in determining the taste of water (Christensen et al. 2018). In natural water, TDS consists of degraded gases, as well as decomposable and non-decomposable compounds (Demirel 2007). In this study, TDS in the groundwater samples varied between 137 and 1063 mg/L (mean: 491.71 mg/L) in winter and between 112 and 1051 mg/L (mean: 466.35 mg/L) in the post-monsoon indicating significant dissolution of minerals in the aquifers (Figure 4).

Figure 4. Spatial distribution map of Total Dissolved Solids of study area.

The interpolated maps show that TDS in major parts of LMC was within the permissible limit for domestic use (500 mg/L). Only at two locations, TDS was above the maximum permissible limit of 1000 mg/L in both seasons. Such isolated high values are most likely due to the occurrence and dissolution of minerals from the heterogeneous geological formations constituting the aquifers (Nagarajan et al. 2010). Furthermore, high TDS concentration in groundwater may originate from anthropogenic activities like sewage discharge (Das and Nag 2017).

Electrical Conductivity (EC): EC is an indicator of the TDS content in water (Hem 1985). In LMC, the EC values varied between 276 and 2127 μS/cm (mean of 983.6 μS/cm) in the winter, and between 226 and 2021 μS/cm with a mean of 913.63 μS/cm during the post-monsoon. High EC values are found in the northern, eastern, and central parts of LMC, owing to a higher concentration of disintegrated ionic particles (Figure 5). Additionally, metal contents, ammonium, and nitrate are also responsible for high EC in the groundwater (Dzwairo et al. 2006). This could be due to the seepage of agrochemicals and leaching of the sewage in the aquifers.

Figure 5. Spatial distribution map of Electrical Conductivity of study area.

Total Hardness (TH): Ions such as HCO₃^ˉ, CO₃²⁻, SO₄²⁻, and Cl^ˉ of Ca²⁺ and Mg²⁺ are the determining factors for TH in water (Jain et al. 2010). Industries are not recommended for the utilization of hard water due to its corrosion effect on metals. A high degree of hardness can damage the distribution pipeline, may cause high expenditure on washing substances, hardening of tissues due to the deposition of insoluble calcium compounds in the blood circulation and urinary system, and diseases related to the digestive system (CPCB 2008). TH as calcium carbonate was found to vary between 100 and 612 mg/L (mean: 306 mg/L) during winter and between 80 and 485 mg/L (mean: 253.03 mg/L) in the post-monsoon. Figure 6 reflects spatial distribution of TH in LMC, which was mostly within the acceptable range for drinking. Only three winter samples exceeded the limit of 500 mg/L. TH in the groundwater is attributed to the carbonate rocks such as limestone, which is dominantly present in the Kathmandu Valley (Acharya and Paudyal 2019).

Figure 6. Spatial distribution map of Total Hardness of study area.

4.3. Cations

The estimated values of the cations could be arranged in descending order as: Ca²⁺ > Na⁺ > Mg²⁺ > K⁺ > NH₄⁺ for both winter and post-monsoon.

Calcium (Ca²⁺): Calcium was found to vary between 25.6 and 191.2 mg/L (mean: 87.36 mg/L) in winter and between 15.38 and 155.6 mg/L (mean: 63.57 mg/L) in the post-monsoon period. Calcium in 14.28% of the winter samples and 25.71% of the post-monsoon samples was below the NDWQS threshold (200 mg/L). Relatively high Ca²⁺ was found in some north-western and south-eastern parts of the study area although the concentration was lower than the threshold (Figure 7). This is possibly due to the dissolution of carbonate rocks (Ravish et al. 2019).

Figure 7. Spatial distribution map of Calcium ion of study area.

Magnesium (Mg²⁺): Magnesium in the collected groundwater samples varied between 2.4 and 61.68 mg/L (mean: 22.21 mg/L) in winter and between 2.4 and 81.87 mg/L (mean: 22.14 mg/L) in the post-monsoon. Thus, the spatial distribution of Mg²⁺ was also within the permissible limit (100 mg/L) of WHO and NDWQS (Figure 8).

Figure 8. Spatial distribution map of Magnesium ion of study area.

Sodium (Na⁺): Sodium is crucial in determining the groundwater chemistry for the use in agricultural purposes since it is directly proportional to soil’s hardness and inversely proportional to its permeability (Tijani 1994). Sodium content in the groundwater samples was quite high both in winter (33.2 to 70.3 mg/L, mean: 79.8 mg/L) and post-monsoon (13.9 to 79.8 mg/L, mean: 53.65 mg/L) but was within the maximum permissible limit (200 mg/L) for drinking purpose. Comparatively higher Na⁺ (>50 mg/L) in the study area may be due to multiple geogenic and anthropogenic influences (Tian et al. 2016; Ravish et al. 2019). The difference in the spatial variation of Na⁺ in winter and the post-monsoon was the effect of monsoon rainfall (Figure 9).

Figure 9. Spatial distribution map of Sodium of study area.

Potassium (K⁺): Potassium was found to vary from 0.84 to 41.5 mg/L (mean: 16.21 mg/L) in winter and from 0.27 to 50.3 mg/L (mean: 15.87 mg/L) in the post-monsoon. K⁺ concentration of all the samples falls under the permissible limit of NDWQS (100 mg/L). However, spatial and temporal variations in K⁺ in the LMC is evident (Figure 10). Relatively, higher K⁺ concentration is most likely due to the mixing of weathered carbon and silica-enriched rocks, seepage of agrochemical residue, and outflow of the sewage from the settlement and industries (Karim and Veizer 2000; Zhang et al. 2007; Selemani et al. 2017).

Figure 10. Spatial distribution map of Potassium of study area.

Ammonium (NH₄⁺): Ammonium (NH₄⁺) content in all the samples was low both in winter (< 0.1 to 0.99 mg/L, mean: 0.46 mg/L) and post-monsoon (< 0.1 to 0.46 mg/L, mean: 0.13 mg/L). The ammonium (NH₄⁺–N) concentration did not exceed the WHO permissible limit (1.5 mg/L). However, the interpolated maps of NH₄⁺ have shown sporadic higher concentration (>0.3 mg/L) in the winter (Figure 11). This is similar to the findings of Sbarbati et al. (2018).

Figure 11. Spatial distribution map of Ammonium in LMC.

4.4. Anions

Anion’s dominance in the groundwater of LMC was observed in the descending order as HCO₃^ˉ > Cl^ˉ > SO₄^ˉ > NO₃^ˉ > PO₄^3- in both winter and the post-monsoon samples.

Bicarbonate (HCO₃^ˉ): Bicarbonate exists in the aqueous solution as components of calcium and carbonate combined with carbon dioxide (Jain et al. 2010). Bicarbonate values in the groundwater of LMC varied from 20 to 450 mg/L (mean: 148.29 mg/L) in winter and between 15 and 105 mg/L (mean: 39.29 mg/L) in the post-monsoon. The spatial distribution map of bicarbonate shows a relatively low concentration in winter and a higher concentration in the post-monsoon. Besides, HCO₃^ˉ in groundwater during the post-monsoon was within the permissible limit (300 mg/L) but was higher in the winter season (Figure 12).

Figure 12. Spatial distribution of Bicarbonate in the groundwater of LMC.

Chloride (Cl^ˉ): Chloride is a steady component of groundwater, accumulated either from the dissolution of rocks and minerals or from agrochemicals and sewage (Hem 1985; Ritzi et al. 1993; Townsend and Whittemore 2005; Prasanth et al. 2012). The chloride level in LMC varied in the range of 22.72 to 241.4 mg/L (mean: 103.71 mg/L) in winter and from 20.45 to 220.1 mg/L (mean: 80.87 mg/L) during the post-monsoon, and thus was within the permissible limit of WHO (250 mg/L) in both the seasons (Figure 13). Only a very few locations had a higher concentration of chloride, which was probably due to the leaching of chloride-containing rocks and minerals in the aquifers.

Figure 13. Spatial distribution of Chloride in the groundwater of LMC.

Sulphate (SO₄^2-): Sulphate exists in nature in dissolved form along with sodium, magnesium, and calcium salts. The concentration of sulphate in groundwater may vary seasonally due to infiltration of rainwater and runoff water (Samantara et al. 2017). Sulphate concentration in LMC varied between 5.52 and 77.70 mg/L (mean: 21.02 mg/L) in winter and between 4.42 and 63.22 mg/L (mean: 16.47 mg/L) in the post-monsoon. Although the sulphate concentration was high in the central and eastern parts during the post-monsoon, it was under the acceptable range of the WHO standard (250 mg/L) in both seasons (Figure 14).

Figure 14. Spatial distribution of Sulphate in the groundwater of LMC.

Nitrate (NO₃^-): Nitrate levels in all groundwater samples were found to vary between 4 and 13.7 mg/L (mean: 6.43 mg/L) in winter and between 4.4 and 29.6 mg/L (mean: 6.58 mg/L) in the post-monsoon and was within the WHO desirable limit (50 mg/L). A slightly higher concentration of nitrate was found in the southern part of LMC during the post-monsoon compared to that during the winter season (Figure 15). The high nitrate concentration is probably due to the disposal of human and animal waste and the leaching of septic tanks (Townsend and Whittemore 2005; Shakya et al. 2019).

Figure 15. Spatial variation of Nitrate in the groundwater of LMC.

Phosphate (PO₄^3ˉ): Phosphate levels in the groundwater samples from LMC varied between 0.23 and 3.58 mg/L (mean: 0.78 mg/L) in winter and between 0.104 and 3.65 mg/L (mean: 0.45 mg/L) in the post-monsoon. The interpolated maps revealed low phosphate concentration in most parts of LMC (Figure 16). However, about 20% of the winter samples and 11.42% of the post-monsoon samples exceeded the permissible limit of WHO (1 mg/L). The high values may be due to contamination of groundwater through the leaching of manure and chemical fertilizers (Domagalski and Johnson 2011).

Figure 16. Spatial distribution map of Phosphate of study area.

4.5. Iron (Fe)

Iron is a dominant metal content in groundwater. The iron content in the groundwater samples of LMC was found to vary between < 0.01 and 10.16 mg/L (mean: 1.02 mg/L) in winter, and between < 0.01 and 4.19 mg/L (mean: 0.42 mg/L) in the post-monsoon. Fe content in ∼66% of the winter samples and 80% of the post-monsoon samples were below the permissible limit (0.3 mg/L), whereas Fe in ∼34% of the winter samples and 20% of the post-monsoon samples exceeded 0.3 mg/L. Iron was found with relatively higher concentrations in the winter samples (Figure 17). The interpolated maps show ‘high’ Fe in a major part of LMC compared to the NDWQS limit. This is to be noted that WHO does not have a drinking water guideline for iron as iron content even of 2 mg/L does not have any health-hazard (WHO 2011, 2017). Iron in the groundwater of LMC was mostly below 2 mg/L. This is most likely due to absence of iron-bearing formation in the shallow aquifers in this region (Samantara et al. 2017).

Figure 17. Spatial variation of Iron in the groundwater of LMC.

4.6. Microbial parameters

Among the microbial parameters, concentration of coliform bacteria Escherichia coli (E-coli) was found high in both seasons in all the 35 groundwater samples. The E-coli values varied between 56 and 254 most portable number (MPN) (mean: 123.09 MPN/100 mL) in winter and between 68 and 315 MPN/100 mL (mean: 151.48 MPN/100 mL) during the post-monsoon. Common reasons behind the high coliform content in groundwater include open defecation (Sorensen et al. 2016; Kapembo et al. 2019), leaky sanitation system (Pujari et al. 2007; Graham and Polizzotto 2013; Sorensen et al. 2016; Bhallamudi et al. 2019; Dasgupta et al. 2021), and inadequate and faulty sewage management structures (Arnade 1999; Lu et al. 2008; Bhandari et al. 2021). Open defecation, leakage, and underground leaching of pollutants into the aquifers impart acute health risk. Contrary to the claim by the Nepal Government that the Kathmandu Valley is free from open defecation, in reality, open defecation is a common practice in the LMC due to lack of awareness, practice, and inadequate public toilets (Ojha 2019). The interpolated maps show a high concentration of total coliform (> 1 MPN/100 mL) in the entire LMC in both winter and post-monsoon (Figure 18).

Figure 18. Spatial distribution of total coliform in the groundwater of LMC.

4.7. Drinking water quality

The drinking water quality zoning map was obtained through integration of weighted parameters and their classification. For drinking purposes, the groundwater was found of ‘moderate’ quality almost in the entire LMC and ‘good’ only in three pockets in the east, west, and south during winter. During the post-monsoon period, groundwater quality for drinking was ‘good’ for western, central, and southern parts of LMC, and was of ‘moderate’ quality in the rest of the LMC (Figure 19).

Figure 19. Groundwater quality for drinking in the municipality wards of LMC with population density.

However, the actual scenario of groundwater quality was totally different which must be explained. Groundwater quality for drinking was obtained by classification of the weightage values produced through the integration of the 15 chemical parameters. Since the total weightage values lie between 23 and 32 in winter and between 20 and 27 during post-monsoon, groundwater quality was classified as ‘good’ or ‘moderate’ based on arbitrary classification of the total weights obtained (Table 2). This however does not portray the true status of groundwater quality. Therefore, groundwater contamination maps (Figure 20) were generated based on C_d values of the winter and post-monsoon samples (Table 7). These maps have revealed significant variations in spatial and seasonal groundwater quality in LMC in terms of contamination level. The degree of contamination in groundwater was ‘high’ in northern, eastern, and central parts of LMC in winter, while in the post-monsoon, ‘high’ contamination level was restricted only in some small northern, eastern, and central pockets of LMC. It is noticeable that in major parts of LMC, groundwater contamination was either ‘high’ or ‘moderate’ in winter but was ‘low’ to ‘moderate’ during the post-monsoon once again confirming the dilution effect on pollutants due to rainfall-induced recharge of the aquifers (Figure 20).

Figure 20. Spatial and seasonal variations of contamination level in the groundwater at the sampling sites (wells).

Table 7. Contamination level (C-level) of groundwater in the municipality wards of Lalitpur Metropolitan City during winter and post-monsoon as determined by the contamination index (C_d).

Sample No.	Winter		Post-monsoon
	Cd	C-level	Cd	C-level
W1	< 0.1	Low	< 0.1	Low
W2	< 0.1	Low	< 0.1	Low
W3	0.5	Low	0.2	Low
W4	0.4	Low	0.4	Low
W5	29.3	High	0.1	Low
W6	< 0.1	Low	< 0.1	Low
W7	33.8	High	13.0	High
W8	0.2	Low	< 0.1	Low
W9	0.1	Low	< 0.1	Low
W10	3.0	Medium	2.3	Medium
W11	0.1	Low	< 0.1	Low
W12	6.3	High	0.4	Low
W13	0.1	Low	0.3	Low
W14	2.7	Medium	< 0.1	Low
W15	1.8	Medium	0.5	Low
W16	1.0	Medium	0.6	Low
W17	0.1	Low	< 0.1	Low
W18	2.7	Medium	2.6	Medium
W19	0.2	Low	< 0.1	Low
W20	0.1	Low	< 0.1	Low
W21	7.2	High	3.2	High
W22	0.1	Low	< 0.1	Low
W23	0.1	Low	0.6	Low
W24	6.1	High	0.4	Low
W25	18.9	High	7.3	High
W26	0.2	Low	< 0.1	Low
W27	2.7	Medium	2.6	Medium
W28	3.6	High	1.1	Medium
W29	0.5	Medium	0.1	Low
W30	0.7	Medium	< 0.1	Low
W31	0.6	Medium	0.1	Low
W35	< 0.1	Low	< 0.1	Low

Since contaminated groundwater with poor quality has a direct bearing on community health (Rahman et al. 2018; Shukla et al. 2021), municipality ward-wise population count and population density in the LMC were correlated and compared. As per 2011 population census (CBS 2011), the LMC had total of 70,256 households and the total population of the LMC was 284,922 with 145,924 males and 138,998 females. Out of the total 29 wards, ward number 14 had the largest population of 21,232, while ward number 29 had the least number of populations with 21,232. Ward number 14 had the most (5,438) households, while ward number 27 had the least number (1,020) of households (Table 8). With the general increasing trend of population growth in this region, it may be assumed that both the total population and the population density increased further by now. Since the 2021 population census data of Nepal is not yet published, population in the LMC was projected using the year-wise growth rate of Nepal as obtained from the World Bank (World Bank 2022). Municipality ward-wise projected population density and total population values are shown in Table 8. The population density values were projected over the groundwater contamination level map to get an estimate of the vulnerable population in the LMC due to water-borne diseases (Figure 21A). It is evident that contamination risk was comparatively less during the post-monsoon, but in winter, ward number 1, 2, 3, 8, 11, 14, 17, 20 and 25 had a high risk of contamination through groundwater since many of the constituents (chemical and bacterial) had a concentration above the maximum permissible limit.

Figure 21. A: Groundwater contamination level and population density in the municipality wards of LMC. B. Ward wise total population in the LMC.

Table 8. Municipality ward-wise projected population of Lalitpur Metropolitan City in the year 2021 (based on population growth rate in Nepal by the World Bank).

Municipality ward	Well No.	Households	Population in 2011	Population in 2021*	Percent of total population
1	W10	2221	8434	9249	2.96
2	W25	4839	19061	20903	6.69
3	W24	3528	14082	15443	4.94
4	W17, W34, W23	3913	15367	16852	5.39
5	W29	1516	6404	7023	2.25
6	W8	1563	6780	7435	2.38
7	W6	1839	7849	8618	2.75
8	W9	2816	11400	12502	4
9	W13	3484	13908	15252	4.88
10	W11	1729	6554	7187	2.3
11	W12	2460	10109	11086	3.55
12	W7	2352	10349	11349	3.63
13	W18	3772	14867	16304	5.22
14	W28	5438	21232	23284	7.45
15	W31, W30	3480	13858	15197	4.86
16	W14	2058	10139	11119	3.56
17	W5	2678	10644	11672	3.74
18	W22	2007	8146	8933	2.86
19	W15	1774	7385	8099	2.59
20	W16	3121	12380	13576	4.35
21	W20, W33	1056	4927	5403	1.73
22	W32, W19	1304	5966	6542	2.09
23	W2	1854	7002	7679	2.46
24	W1, W35	1324	5676	6224	1.99
25	W21	2996	11575	12693	4.06
26	W27	1377	5813	6375	2.04
27	W26	1020	4279	4692	1.5
28	W4	1667	6577	7213	2.31
29	W3	1070	4159	4561	1.46
Total	35	70056	284922	312464	100

(Source: CBS, 2012)

* Projected population

However, the main concern regarding groundwater quality in the LMC is not due to anomalous concentration of the chemical components but due to the presence of coliform bacteria in a high concentration. Instead of equal importance to all the parameters, if the highest importance was given to total coliform, then groundwater quality of the entire LMC would be ‘very bad’. The presence of E-coli alone makes the groundwater resources of the LMC unfit for drinking. Comparing and correlating ward-wise total population and groundwater quality (Tables 7 and 8, Figures 18 and 21B), it may be inferred that with a higher population count, municipality wards 14, 2, 4, 13, 3, and 9 (red and orange in Figure 21B) have a higher health risk due to coliform. In fact, the entire population of Lalitpur is vulnerable to water-borne diseases due to the presence of TC!

4.8. Irrigation water quality

Irrigation water quality was determined using various thresholds for TDS, Cl, SAR, Na%, PI, and MAR (Table 3). Based on the TDS, majority of the winter samples (68.57%), as well as of the post-monsoon samples (62.86%) were found ‘good’ for irrigation, and 25.71% and 31.42% of samples were ‘permissible’ during winter and the post-monsoon, respectively. Only two winter and post-monsoon samples with very low TDS were found ‘excellent’ for irrigation (Figure 22). However as per the criteria, all the groundwater samples were suitable for irrigation (FAO 2006).

Figure 22. Groundwater quality of samples for irrigation based on TDS concentration.

Again, according to Ayers and Westcot (1985), up to 135 mg/L chloride concentration is good for irrigation. Figure 23 indicates 80% of the winter samples and 91.42% of the post-monsoon samples had Cl^ˉ below 135 mg/L.

Figure 23. Groundwater quality for irrigation based on chloride concentration.

SAR was computed for determining the alkali/sodium hazards to crops, as it expresses the degree of cation exchange reactions in soil when sodium replaces the absorbed magnesium and calcium (Wilcox 1955). Based on grading by Richards (Subramani et al. 2005), SAR values less than 18 have low sodium hazard and are suitable for irrigation. Analysis indicated that all the groundwater samples collected in this study were ‘excellent’ for irrigation (Figure 24).

Figure 24. Sodium Adsorption Ratio (SAR) in winter and post-monsoon samples.

Besides SAR, sodium concentration also plays an important role in groundwater quality for irrigation (Haritash et al. 2008). When Na⁺ concentration in irrigation water is high, sodium replaces hardness factor like Mg²⁺ and Ca²⁺ ions, thereby decreasing soil permeability and reducing air and water circulation (Tijani 1994; Logeshkumaran et al. 2015), causing adverse impacts on crop health (Bouderbala and Gharbi 2017). Analysis revealed that the sodium percentage in groundwater was ‘very good’ in most parts of the LMC (Figure 25).

Figure 25. Spatial distribution map of Sodium percentage.

The permeability index (PI) proposed by Doneen (1964) helps to build the relationship between major cations and bicarbonate and assess the suitability of groundwater for cropland irrigation (Panneerselvam et al. 2021). The PI was used to classify soil affected by Na⁺, Mg²⁺, Ca²⁺, and HCO₃^ˉ into three classes (Table 3). Water with >75% PI (Class I) is considered good for irrigation, 25–75% PI (Class II) as suitable, and <25% PI (Class III) as unsafe since a low PI has an adverse impact on the development of vegetation (Rawat et al. 2018). In this case, six winter samples and two post-monsoon samples fell in Class II, while 29 winter samples and 33 post-monsoon samples were in Class III, pointing to their non-suitability for irrigation (Figure 26).

Figure 26. Spatial distribution map of Permeability Index.

Magnesium Adsorption Ration (MAR) was also used in this study to assess irrigation water quality. MAR values for the majority of the groundwater samples were found suitable for irrigation. Only two winter samples (W23 and W26) had unsuitable MAR (>50) while 94.28% of the samples were fit for irrigation. However, all the post-monsoon samples were suitable for irrigation with MAR values below 50 (Figure 27).

Figure 27. Magnesium Adsorption Ratio (MAR) in winter and post-monsoon samples.

The final zoning map for irrigation thus prepared based on the aforesaid chemical parameters (Figure 28) shows that for irrigation purposes, groundwater quality in most parts of the LMC was moderate in winter but was good in the post-monsoon. According to the weightage value (Table 4), only two groundwater samples in winter, one each in the northern and central parts were bad for irrigation.

Figure 28. Groundwater quality for irrigation in the municipality wards of LMC.

4.9. Hydro-geochemical processes

Hydro-geochemical processes and the geogenic and anthropogenic factors governing the groundwater chemistry in the Kathmandu Valley including the LMC could be identified in this study. The exchange of ions that occurs in the groundwater system is partially controlled by the local geological set-up influencing rock-water interactions (Wu et al. 2015; Li et al. 2018). This is reflected in the TDS, as with the variation in the ion concentration, the TDS values also changed (Figures 29 and 30). Among the major cations, Ca²⁺ and Na⁺ contribute more compared to Mg²⁺, while HCO₃^ˉ contributes most among the anions. Additionally, in bulk of the winter and post-monsoon samples Ca²⁺ and Na⁺ exhibited a counterpoise (Figures 29a,b), and in the winter samples, HCO₃^ˉ exhibited a falling trend in the TDS range of 1000 to 1200 mg/L (Figures 30a,b). The association between TDS and the majority of the ions demonstrated that the breakdown of minerals containing Ca²⁺ and Na⁺ and HCO₃^ˉ augment the majority of water-rock interactions, which may include calcite, dolomite, or albite. On the other hand, gypsum and halite dissolution may be the source of SO₄^2- and Cl^ˉ (Li et al. 2018). But in the case of Kathmandu valley, halite deposition is never reported thus Cl^ˉ source might be from anthropogenic contamination (Nakamura et al. 2014).

Figure 29. Scatter plots showing variations of the principal cations versus TDS in winter (a), and during the post-monsoon (b).

Figure 30. Scatter plots showing variations of the principal anions versus TDS in winter (a), and during the post-monsoon (b).

The silicate weathering is the dominant geochemical process governing the major ionic chemistry in groundwater (Mackenzie and Garrels 1965; Li et al. 2018; Zhu et al. 2019). Halite and silicate rocks are the main sources of sodium in the groundwater. Generally, sodium to chloride ratio in groundwater is approximately 1 if the halite dissolution process releases the sodium ions, whereas if Na to Cl ratio is greater than 1, then silicate weathering is responsible for sodium ions in water (Meybeck 1987; Wu et al. 2014; Li et al. 2016; Wali et al. 2021). Alternatively, when sodium ions are released from silicate weathering, HCO₃^ˉ occurrence will be most ample other than sodium in the water (Lakshmanan et al. 2003) as is understood by Equation (10)(10) $${2\text{NaAlSi}}_3{\mathrm{O}}_8{+2\mathrm{H}}_2{\text{CO}}_3{+9\mathrm{H}}_2{\mathrm{O}=2\text{Na}}^{+}{+\text{Al}}_2{\text{Si}}_2{\mathrm{O}}_5{\left(\text{OH}\right)}_4{+4\mathrm{H}}_4{\text{SiO}}_4{+2\text{HCO}}_3{}^{ˉ}$$(10) . (10) $${2\text{NaAlSi}}_3{\mathrm{O}}_8{+2\mathrm{H}}_2{\text{CO}}_3{+9\mathrm{H}}_2{\mathrm{O}=2\text{Na}}^{+}{+\text{Al}}_2{\text{Si}}_2{\mathrm{O}}_5{\left(\text{OH}\right)}_4{+4\mathrm{H}}_4{\text{SiO}}_4{+2\text{HCO}}_3{}^{ˉ}$$(10)

In this study, the Na/Cl ratio varied from 0.24 to 4.56 in the winter and between 0.40 and 4.89, in the post-monsoon samples with the mean values 1.09 and 1.43, respectively. The Na/Cl ratio in 88.6% of the groundwater samples was greater than or equal to 1 (Figure 31). Again, the values of HCO₃^ˉ were higher than the values of Na⁺ in the groundwater samples as shown in Figure 32. These two facts together suggest that the source of sodium in the majority of the winter samples was silicate weathering while that of the post-monsoon samples was carbonate weathering. According to the previous studies, if the silicate weathering was the source of the sodium then HCO₃^ˉ will be the abundant anion in the water sample (Rogers 1989; Rajmohan and Elango 2004). In this study, HCO₃^ˉ is the dominant anion.

Figure 31. Scatter plot showing seasonal variation of Na⁺ versus Cl^ˉ in the groundwater.

Figure 32. Scatter plot for seasonal variation of HCO₃^ˉ versus Na⁺ in the groundwater.

Gypsum, dolomite, limestone, and anhydrite leaching are the main sources of calcium and magnesium ions in groundwater, and the cation exchange process is another important source of calcium ions (Han and Liu 2004; Wu et al. 2015; Wali et al. 2021). The origin of the calcium and magnesium ions can also be determined by comparing the Ca/Mg molar ratio of (Ca²⁺ + Mg²⁺) to (HCO₃^ˉ + SO₄²_ˉ). This means that dolomite should typically dissolve if the Ca/Mg molar ratio is closer to 1 (i.e. the ratio is 1) and that calcite contributes more when the ratio is higher (Hundal 2011) and more silicate minerals contribute when the Ca/Mg molar ratio is higher than 2. Generally, the scatter plot ratio of (Ca²⁺ + Mg²⁺) vs (HCO₃^ˉ + SO₄^2-) should be close to the 1:1 line if the dissolution of gypsum, dolomite, and calcite are prominent in a system (Datta and Tyagi 1996; Fisher and Mullican 1997; Sonkamble et al. 2012). The scatter plots ratio tends to shift to the right when the ion exchange process is prominent in a system (Chang and Wang 2010; Jeevanandam et al. 2012). The scatter ratios will move to the left if reverse ion exchange is prevalent. The Ca/Mg ratios in this study varied from 2.61 to 45.75 for the winter samples and between 0.60 and 40.67 for the post-monsoon samples with an average value of 9.15 and 8.62, respectively (Figure 33). This indicates that the dissolution of calcite, dolomite, and silicate minerals in the alluvium and the aquifers was the primary cause of the calcium and magnesium ions in the groundwater (Saleh et al. 1999; Li et al. 2018; Wali et al. 2021).

Figure 33. Scatter plot showing seasonal variation of Ca/Mg molar ratio.

In the present study, all the post-monsoon samples and 80% of the winter samples have higher (Ca²⁺ + Mg²⁺) than the (HCO₃^ˉ + SO₄^2-), inferring that carbonate weathering is the dominant hydro-geochemical process in the groundwater (Figure 34) as explained by equation 11(11) $$\text{CaMg}{\left({\text{CO}}_3\right)}_2{+2\mathrm{H}}_2{\mathrm{O}+2\text{CO}}_2{=\text{Mg}}^{2+}{+\text{Ca}}^{2+}{+4\text{HCO}}_3{}^{ˉ}$$(11) . Fewer samples fall near and below the isoline to indicate that the main mechanisms causing calcium and sodium ions in groundwater samples were silicate weathering and ion exchange. (11) $$\text{CaMg}{\left({\text{CO}}_3\right)}_2{+2\mathrm{H}}_2{\mathrm{O}+2\text{CO}}_2{=\text{Mg}}^{2+}{+\text{Ca}}^{2+}{+4\text{HCO}}_3{}^{ˉ}$$(11)

Figure 34. Scatter plot of (Ca²⁺ + Mg²⁺) versus (HCO₃^ˉ + SO₄^2ˉ) depicting seasonal variation in silicate weathering and carbonate weathering.

5. Discussion and conclusion

The hydrochemical studies help in determining the suitability of groundwater for domestic and agricultural purposes (Ganiyu et al. 2018). Moreover, it reveals the change in water quality due to rock-water interaction and human influence, if any (Wilcox 1948). GIS further helps in visual interpretation of spatial variations of various chemical constituents in the groundwater. In this study, the spatial distribution of the chemical parameters and coliform bacteria in the groundwater of Nepal’s Lalitpur Metropolitan City was mapped using GIS. Lower concentration of most of the chemical constituents in the groundwater during the post-monsoon compared to that in winter was the effect of monsoon-driven recharge of the aquifers and dilution of the chemicals during the monsoon season (Lasagna et al. 2013). Comparison of their concentration with the WHO and NDWQS standards and guidelines revealed that most of the groundwater samples are unsuitable for drinking but can be used for domestic and irrigation purposes. Anomalous high concentrations were noticed for several parameters at several sample locations either in winter or in the post-monsoon or in both seasons. The mean value of certain parameters such as pH during winter and calcium and iron in both seasons were much higher than the permissible limits. Besides pH, calcium, and iron, the maximum values for some other parameters such as TDS, EC, TH, magnesium, bicarbonates, and phosphate were even much higher than the permissible limits. Such high concentration of these parameters is most likely due to agrochemicals (fertilisers and pesticides), leakage of solid wastes, and percolation of domestic and industrial effluents (Wei et al. 2017; Li et al. 2018; Shakya et al. 2019). Furthermore, the nitrate concentration found higher in the post-monsoon most likely due to irrigation return flow in the subsurface and septic tank leakage (Gao et al. 2012). Besides the anthropogenic activities, the groundwater chemistry was affected by the geological environment. The cation and anion concentrations changed with the TDS variation (Wu et al. 2015; Li et al. 2018). The weathering of silicate and carbonate rocks has significant influence on groundwater chemistry (Zhu et al. 2019). The high concentration of TDS, EC, HCO₃^ˉ, Cl^ˉ, and SO₄^2- in the groundwater reflects the rock-water interaction factor in groundwater quality (Chen et al. 2019). Thus, geogenic contamination and rock-water interaction both have significant influence on the groundwater quality. Anomalous high concentration of these parameters in groundwater is definitely a matter of concern considering their potential adverse impacts on the health of plants, animals, and humans (Shakya et al. 2019).

However, the biggest concern for human health in the LMC is definitely due to high concentration of faecal coliform bacteria in all the groundwater samples in both seasons. Previous studies (Warner et al. 2008; Haramoto et al. 2011; Pant 2011; Shrestha et al. 2014; Inoue et al. 2015; Bhandari et al. 2021; Sarkar et al. 2022) also reported the occurrence of total coliform, faecal coliform, and other pathogenic bacteria and viruses in the groundwater of the Kathmandu Valley. The presence of faecal coliform bacteria in an aquatic environment is an indication of contamination of the water with faecal materials of animals and humans. In such cases, the chance of contamination by pathogens i.e. disease-producing viruses and bacteria is high (Shrestha et al. 2015). Therefore, the occurrence of faecal coliform in water enhances the risk of water-borne pathogenic diseases such as diarrhoea, cholera, typhoid, bacterial gastroenteritis, and hepatitis-A through contact or consumption of the contaminated water (Kumar and Puri 2012). Moreover, the increasing presence of antibiotic-resistant bacterial pathogens in groundwater is a matter of emerging global concern (Gallert et al. 2005; Bradford and Harvey 2017).

Open defecation and leaky sanitation systems are the main reasons that put the entire population of the LMC at health risk. As part of its Sustainable Development Goals 6, the United Nations have set targets such as: (1) achieving universal and equitable access to safe and affordable drinking water for all by the year 2030, and (2) achieving access to adequate and equitable sanitation and hygiene for all and end open defecation by the year 2030 (UNDP 2018). Although the Nepal Government declared all districts to be open defecation-free zones, in reality, open defecation is rampant even in the major urban centers such as Kathmandu, Bhaktapur, and Lalitpur (Ojha 2019). With very few public toilets and only one faecal sludge treatment plant, management of faecal sludge is a big challenge in the LMC. Due to inadequate sewage treatment facility, most of the sewerage ends up in rivers without any treatment (Aawaaj News 2021; Bhandari et al. 2021), and subsequently infiltrates the subsurface. Thus, both soil and water in the LMC are highly polluted with faecal bacteria leading to a high risk of water-borne community diseases.

The contamination risk of the pathogens is high as groundwater is the main source of drinking water. Numerous cases of waterborne outbreaks were reported from across the globe due to widespread pathogenic contamination of groundwater (Gallay et al. 2006; Fong et al. 2007; Craun et al. 2010; Cho et al. 2014; Guzman-Herrador et al. 2015). Similar pathogenic cases were reported in the past also in the Kathmandu Valley (Pandey et al. 2002). If attention is not given and the groundwater is not properly treated before the municipality supply, water-borne diseases such as diarrhoea, dysentery, cholera, and hepatitis may spread across the Kathmandu Valley and may even lead to an epidemic.

Therefore, this study suggests that the legislators and regulators take appropriate steps and actions to provide clean water to the population of LMC. Great importance is attached to the purification of the groundwater. According to the WHO, unsafe water and unhygienic sanitation are the main causes of more than 80% of all diseases in developing countries. The same are the reasons behind the high rates of morbidity and mortality of children and infants in Nepal. Since faecal coliform bacteria imparts greater health risk and threat from waterborne diseases, to prevent an epidemic, immediate attention must be given to the blanket banning of open defecation and to the imposition of strict penalty for violation of the rules. At the same time, the Government of Nepal must ensure the availability of clean potable water, sufficient public toilets with the proper hygienic facility, and an adequate number of faecal sludge treatment plants.

Authors’ contributions

Conceptualization: MG; Field work and chemical analysis: MG and TR; Methodology, Data analysis and mapping: MG; Supervision: SPK; Writing original draft: MG; Rewriting: MG, TR, SPK and CB; Review, editing and final shape: CB.

Acknowledgements

The authors are thankful to the five anonymous reviewers for their valuable suggestions for the betterment of this article.

Data availability statement

All the datasets used to support the findings of this study are available from the corresponding author upon request.

Disclosure statement

It is hereby declared that there is no conflict of interest that could have influenced the result and that there is no conflict in the order in which the authors’ names appear.

Additional information

Funding

Fund was received from University Grants Commission (UGC), Nepal. Award No: MRS-76/77 and Center of Research for Environment, Energy and Water (CREEW)-2019/20.