Risk analysis for groundwater intake in an old mining shaft with increased chloride content, Upper Silesia, Southern Poland

Water from underground intakes is the main source of potable water for people in Poland, hence the protection of its resources is of great importance for the functioning of society and the economy. A new regulation in Polish Water Law imposes the obligation of performing risk analysis for water intakes, including the assessment of health hazards, factors negatively affecting water quality which are identi ﬁ ed on the basis of hydrogeological and geological analyses. The main objective of the study was to determine the health risk for chlorides and to present an innovative approach to the health risk for non-toxic substances. In Upper Silesia, which is the most industrialized and urbanized area in Poland, old mining shafts are often used as deep wells in the water supply chain, and higher mineralization is the key feature of abstracted water which does not quite eliminate them as a source of drinking water supply. This paper proposes a new method of health risk determination as hazard index (HI). We present analysis of the health risks with increased concentration of chlorides in water which cause health effects for water consumers, especially for men, children aged 4 – 8, pregnant women and women during lactation.


GRAPHICAL ABSTRACT INTRODUCTION
The Upper Silesian Coal Basin, Southern Poland, is a region which is affected by mining activity which has been carried out for 200 years and is now in a transition process. Mine closure, restructuring of the infrastructure and the beneficial use of abandoned mines are parts of a wider mandate for the mining industry to move towards long-term environmental and socio-economic sustainability (Petritz et al. ).
Old mine shafts are often used as wells for drinking water supply, especially those in relatively water-rich areas where permeable Quaternary strata cover Carboniferous aquifers. The owners of these intakes who perform tasks in the field of water supply must comply with the quality standards of water with regard to increasing the wellbeing of society and its demand for high quality services. In addition, legal requirements, technological progress and the principles of sustainable development force the service providers of the water supply to take many actions to ensure the primary objective, i.e., continuous water supply of very good quality for human consumption.
Thus, use of water from old shafts for drinking purposes is one benefit of the transition of a coal region, but on the other hand, problems of quality and an increase in some parameters, i.e., chloride content, limits the use of water from shafts without the use of technological solutions or risk assessment for the stability of the water chemistry.

REGULATORY CONTROLS FOR DRINKING WATER FROM GROUNDWATER INTAKES
Technological and legal solutions are conducive to ensure the supply of the population with drinking water of the required quality and in the desired quantity. The entrepreneurs providing water have an impact on many factors, such as controlling the processes of water treatment, distribution, supply and monitoring of these processes. There are also legal and economic instruments for protecting water intakes. Recently, in Poland, there have been major changes in the legislative field, which brought into force the provisions of the Water Law Act of 20 July 2018, which are currently in line with the EU Water Framework Directive (2000/60/EC). One of these requirements, which concerns the protection of water intakes, is the obligation to carry out risk analysis for water intakes. This analysis, in accordance with the requirements resulting from the new regulations, should assess health hazards, including factors which negatively affect the quality of the water from the intake. Identification of these factors is to be carried out based on hydrogeological or hydrological analyses (depending on the type of intake), water quality analysis, as well as the identification of hazards resulting from the land use as well as urban and industrial impact. It is worth mentioning that this obligation is new in Polish water management. The results of the risk analysis should lead to the justification of the establishment of a protection zone, which will have specific orders, prohibitions and restrictions in the scope of land use and water use. Carrying out such analysis and then establishing a protection zone for water intakes will require the establishment of local law acts along with the implementation of restrictions, prohibitions and orders in the area of the protection zone. The impact of human activity on the natural environment, and the quality of groundwater and surface water is visible, especially in highly industrial and urban areas. In Polish legislation, only risk analysis for water intakes are obligatory, while such analysis is part of a wider water safety management tool: Water Safety Plan (WSP). WSPs have an important role to improve water safety for small systems, which commonly face challenges related to human and financial resources, training, equipment, geographic remoteness and highly variable water supply system types and management arrangements (WHO ).
In the European Union (EU), Directive / of 6 October 2015 amends Annexes II and III of the EU Drinking Water Directive, giving EU member states (among others) the option to deviate from the list of drinkingwater monitoring parameters and from the stipulated minimum monitoring frequency in case a risk assessment has been implemented as a basis for the deviation. supply systems (among other system types) in order to verify understanding of WSP principles, to discuss any barriers to WSP implementation. In Kenya, Uganda and the United Republic of Tanzania, informal WSP audits were carried out at three water utilities (one per country). Results from the audits revealed that 77% of the WSP process was well developed and implemented but gaps remained in operational monitoring and verification that could undermine WSP effectiveness (WHO ).
The audits further provided a mechanism to confirm required upgrades for ageing infrastructure to support the preparation of informed investment plans. In Poland, the results of health risk analysis only were the basis to prepare investment plan for water intakes, barely to improve treatment process of water.
In the risk analysis for groundwater intake, it is therefore necessary to carry out a fairly detailed inventory of

STUDY AREA
The groundwater intake 'Jarosław Dą browski' from a Carboniferous formation is located in the former shaft of the 'Sobieski' mine. This shaft was designed as a ventilation

HEALTH RISK ANALYSES: METHODOLOGY
The determination of hazards for the water being used is important to define the factors that increase health risks due to the deterioration of its quality. Consuming low-quality water carries a risk of disease, such as cancer and diseases of the circulatory or digestive system. Highly The World Health Organization (WHO) in the 'Guideline for drinking water quality' has determined the lethal dose (LD 50 ) for rats. The data are collected in Table 1.
According to WHO recommendations, the daily intake of chlorides by adults should not exceed 5 g/d, and a dose of 0.6-3.6 kg/d is considered fully safe and has a non-harmful effect. The average daily intake of chlorides with food in the case of a salt-free diet is around 0.1 g/day (WHO b). In order to analyse the health risk for the studied underground water intake in an old mine shaft located in a region of high pressure of mining plants on the water level, the following values of chloride concentrations were considered: The assessment of health risk and exposure was based on the recommendations of the US Environmental Protection Agency US EPA (US EPA ). In line with these recommendations, the risk assessment should first and foremost characterize the hazard, and then determine the value of the dose taken (Equation (1)), i.e., the amount of harmful substance that the body is exposed to in a given way of exposure per day per 1 kg of body weight: where: I is the dose taken of the substance, the estimated daily intake in conditions of chronic exposure [mg· (kg À1 d À1 )]; C w is concentration of a chemical substance in water [mg/L water]; K w is magnitude of the exposure to a given environmental medium per unit of time [L water/d]; F l is uncertainty factorthis number is defined within the range 0-1, which determines what part of the actual consumption comes from a contaminated source; EF is exposure factor; and MC is body weight (kg).
The exposure factor (EF) represents daily exposure to the contaminant. The EF is calculated by multiplying the exposure frequency by the exposure duration (ED) and dividing by the time period during which the dose is to be averaged (Equation (2)) (PHA ): where: F is frequency of exposure and duration of exposure [days/year]; ED is exposure duration (years); AT is averaging time (ED × 365 days/year).
The values used for the calculation are listed in Table 2.
The frequency of exposure was assumed by taking into account the number of days off work in a calendar year, which is approximately 114 days. It is assumed that we spend half of this time away from home, so the frequency of exposure in a year is 308 days.
Subsequently, the health risk of chlorides supplied with drinking water was estimated by comparing the upper intake level (UL) with the value of the estimated daily intake (EDI) of chlorides with drinking water. UL is the dose of the substance that an adult person can safely be exposed to on a daily basis throughout life, without harmful impacts on health, according to current knowledge. UL values for individual age groups are summarized in Table 3.
The daily intake (EDI) was estimated according to Equation (3)): where: F is water consumption [L/person d À1 ] and R is concentration of chlorides in water from intake [mg/L].
According to the recommendations of the US EPA, while it is impossible to accurately estimate the amount of water consumed from a given source, the largest possible value is estimated (US EPA ). The amounts of water consumed daily for a given age group and gender recommended by the Polish Experts Group (Wos´et al. ) (Table 4) were used for calculations. In addition, it was assumed that 100% of drinking water is supplied from the intake.
In the last stage of calculations, the value of reference dose of chlorides in drinking water was compared with the actual collected dose by determining the hazard index (HI) (Equation (4)): where: HI is hazard index; I is the dose taken of the substance (mg·kg À1 ·d À1 ); Rfd is reference dose (mg·kg À1 ·d À1 ); UF 1-4 is uncertain factor; MF is modification factor (takes into account other uncertainties) (value 1-10); UF 1 takes into account species variability when extrapolating from animals to humans (value 10); UF 2 takes into account the individual variability within the human population (value 10); UF 3 is used in the extrapolation of data from the study of subchronic conditions to chronic conditions (value 10). In this study, it was assumed that the value of a UF is 100 (UF1·UF2·1) and MF is 1.
The RfD is a benchmark dose operationally derived from the NOAEL or the LOAEL by consistent application of generally order-of-magnitude uncertainty factors (UFs) that reflect various types of data sets used to estimate RfDs. The RfD is determined by use of Equation (5): The calculated value of LOAEL by EChA for NaCl was 2,533 mg·kg À1 ·d À1 (EChA ). It was assumed that if HI >1, there is a possibility of negative health effects as a result of long-term exposure to water consumption with a defined (higher) chloride content. In the opposite case (HI <1), the health risk is very low.

RESULTS AND DISCUSSION
Studies have shown that there is a relationship between chloride dose and the age and gender of people in terms of health risk assessment. In the case of boys and men, the consumption of chlorides with drinking water is much higher than in the case of girls and women of the same age. It was estimated that consumption of water with a chloride concentration in the range of 232-304 mg/L by boys aged 9-18 is associated with intake from 488 to 1,004 mgCl À /person d À1 , corresponding to 27 and 56% of the value of the upper tolerable daily consumption of chlorides with food ( Figure 3). In this age group and for this gender there is a risk of exceeding the upper intake level (UL), as a result of consuming water with an increased concentration of chlorides, which may result in the development of cardiovascular disease, the development of stomach cancer or the occurrence of a stroke. The amount of chloride taken together with drinking water increases with age. Men aged 19-70 are especially exposed to excessive supply of chlorides with drinking water. The estimated daily consumption of chlorides with water from the intake may be above 126 mg/person d À1 , which is 63% of the UL value. When leading an inadequate lifestyle and a chlorine-rich diet, there is a high risk of UL being exceeded in this age group and for this gender, which in turn, will be associated with the development of diseases.
Men at retirement age are less exposed to excessive chloride intake, due to the lower demand for drinking  water in the diet. Consumption of water with a chloride content of 249 mg/L (value close to the allowable taste threshold of 250 mg/L) by a group of men aged 19-70 is associated with the consumption of half of the daily limit value of chloride. Consumption of water with chloride concentration above the taste threshold means that in the younger age group, i.e., boys aged 13-15, significantly increased chloride intake with drinking water can be observed ( Figure 3).
In the case of girls and women, the amount of chloride taken with drinking water from the intake will be approximately 441 mg/L to 700 mg/L, which is 25 and 39% of the UL value, respectively. Pregnant women and women during lactation are among those women most exposed to excessive chloride intake. The consumption of chlorides with drinking water may amount to mg/ person d À1 , which corresponds to 39-64% of the UL value ( Figure 4).
In the last stage of the research, the value of the hazard index (HI) was determined. In the case of water with a chloride concentration of 232 mgCl À /La value below the acceptable threshold of 250 mgCl À /L, the highest value of the hazard index was observed for the group of children aged 1-12 years (HI was in the range 0.33-0.49). In the case of adults of working age, no adverse health effects are observed as a result of drinking water with a chloride content below the acceptable taste threshold (HI <0.25) ( Figure 6).
The increase in chlorides' concentration in drinking water to 250 mgCl À /L is not a threat to people consuming this water (Figure 7).
Consumption of water with concentration of chlorides above the taste threshold level could pose a risk to children aged 4-8. The HI value for children was 0.64 ( Figure 8).
It is important to note that risk assessment was calculated with the general assumption that water with increased chlorides' concentration covers total daily consumption, and the sensitivity analysis of this approach is required. The sensitivity analysis was performed on the assumption that both the chloride concentration in drinking water and the amount of water consumed from an analysed water intake affect the hazard index. The analysis was carried out assuming that the amount of water consumed from the water intake would be from 25 to 100% of the recommended daily intake. The sensitivity analysis was performed for the following chloride concentrations in water: 250 mgCl À /L, 300 mgCl À /L, 350 mgCl À /L, 400 mgCl À /L, 650 mgCl À /L. Concentration of 250 mgCl À /L in consumed water from concerned intake in the amount of 25-75% of K w value will not pose a health threat to people. The high HI value (>0.8) was observed in the population of children of the age 1-8 when the K w was 100% (Figure 9(a)). The concentration of 350 mgCl À /L poses a health hazard to children (1-8 years old) for whom the only source of water would be the water consumed from the intake (100% of K w value).
Children and the elderly will be not exposed to health effects when consuming less than 50% of the recommended daily intake. The high HI value (>0.8) was observed in the population of children in the age 1-12 and adults when the K w was 75% (Figure 9(b)). Further increasing the concentration of chlorides (!350 mgCl À /L) could pose a health risk when consuming even 75% of the recommended daily intake ( Figure 9(c) and 9(d)). The results of sensitivity analysis showed that HI >1 will be observed for the adult population when the chloride concentration reaches the value 650 mg/L (Figure 9(e)).

CONCLUSIONS
In this paper, we presented an approach to health risk analysis for drinking water, in which the chloride concentration ranged between 232 and 304 mgCl À /L and was, on average, 249 mg/L. In the first step, the hydrogeological factors and relationship of chloride content in water from the old mine shaft were identified.
Conducting a broad environmental study for the needs of the risk analysis of a water intake is the first phase of work, which should then focus on analysing the quality parameters of raw water from the intake in order to track trends in water chemistry changes, and then carry out health risk analysis.
Research revealed that the concentration of chlorides due to the impact of industrial, hydrogeological and mining factors would increase in the water from the intake. Health risk analysis is carried out as a part of legal An important element of the risk analysis was the determination of the value of the hazard index. It has been found that children aged 1-8 years are most exposed to the negative effects of chlorides consumed with drinking water. In their case, the value of the hazard index was high even in relation to the consumption of waters with chloride concentration close to the permissible taste threshold. Older children and adults are less exposed to the health effects caused by drinking water with chloride 250-300 mg/L. This research has shown how important it is, in the case of water intakes in highly urbanized and industrial areas, to recognize the factors that affect changes in water quality, which in turn, leads to the estimation of the health risk associated with the consumption of water from an intake with deteriorated quality. In the case of the unfavourable assessment for health risk, proposals should be made for actions in the resource area (protection zone) to limit the causes of negative changes in the quality of drinking water.
Studies have shown that there is a need to carry out a health risk analysis each time also for substances for which concentration monitoring is not required and there are no specific limit values. In the case of chloride, consuming water with a concentration close to the threshold value will have health effects for young children (1-8 years old).
The methodology for determining the health risk presented in the paper and the results of the analysis have shown that it is easy to determine the health risk also for substances for which the hazard parameters have not been clearly defined.
The main objective of the study was to determine whether it is necessary to perform the health risk for substances for which concentration monitoring in raw water is not required and there are no threshold values, e.g., chlorides.
The paper also presents an innovative approach to determining the health risk for non-toxic substances for which there are no clearly defined hazard parameters.
The results of the sensitivity analysis showed the importance of health risk assessment also for all non-carcinogenic substances in a concentration above the limit value. In addition, the results of study have shown that risk analyses should also be assessed for substances without a threshold value, when there is a suspicion of health hazards.
Due to the intake's location in a former coal field area, it is not possible to limit the impact of mining drainage and the introduction of saline waters into the environment for the protection of resources of abstracted drinking water. Its owner has no influence on external factors that are out of the protection zone of the considered intakein this case,  consistent with the requirements in this regard. The preliminary analysis of the health risk for water with an increased content of chloride ions carried out for the purpose of the intake was helpful in making the decision to implement the necessary technological solutionin this case reverse osmosisin the water treatment plant. Risk analysis for groundwater intakes limited only to intake itself seem to be first step to protect consumers against health risks and poor quality of delivered water.

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