Lake water quality assessment using spatial interpolation and the development of the WQI on an educational campus, Assam, India

Natural water comes from springs, rivers, and lakes. The health of the human race and other animal species is directly impacted by the water ecosystem. Monitoring and evaluating a water body's health in terms of water quality is an essential tool. Given that the water body serves as a source for human consumption and drinking purposes, it is crucial to assess the water quality in accordance with established standards (Kazi et al. 2009). When evaluating water quality indicators, physical, chemical, and biological elements of the water are taken into consideration. Both temperature and dissolved oxygen (DO) are essential for aquatic organism viability. A water body's health can be determined by the growth of aquatic vegetation and a rise in fish populations (Akongyuure & Alhassan 2021). Lack of lake water quality management results in salinization and eutrophication. Climate change, domestic/industrial/agricultural runoff, etc., all of which have a significant impact on the water quality variables, primarily affect the water bodies. The sources affecting the water quality of a specific water body can be found by analysing the variance in water quality. The concept of the Water Quality Index (WQI), which serve as an indicator for the water quality evaluation, was initially put forth by Horton (1965) and Rajiv Das Kangabam et al. (2017).

The variables (pH, EC, temperature, and DO) are combined into a single number by a mathematical formula known as the WQI (Elshemy 2019). The WQI categorises the body of water on a scale of 0–100. Depending on what they are used for, numerous sorts of water quality indicators exist (drinking, irrigation, etc). The effectiveness of a successful plan or strategy for managing water resources is also indicated by the WQI (Hector et al. 2012).

Water resource management has used GIS (geographic information system) extensively over the years. The values at an unsampled place can be predicted with the aid of spatial interpolation methods. There are several types of interpolation, including kriging, inverse distance weighted (IDW), spline, trend, and cokriging (Khouni et al. 2021). The effectiveness of GIS as a tool for managing water quality has long been acknowledged.

Seven water quality measures, including total dissolved solids (TDS), temperature, pH, DO, ammonia, hardness, and chloride, were assessed as part of our study. Since each parameter's concentration significantly affects the water quality, these water quality parameters were chosen. (i) pH: The pH level is crucial because it lessens the hazardous effects of gases like hydrogen sulphide, dissolved metals, and ammonia. (ii) temperature: The temperature has a significant impact on the population of aquatic organisms. Additionally, it controls how soluble DO is. (iii) DO is an important component for aquatic plants and animals; a decline in DO levels can exacerbate the effects of harmful gases like H₂S. (iv) Ammonia: The most prevalent type of nitrogen in water is ammonia. Ammonia functions as a buffer, and aquatic plants use the NO₂ and NO₃ that are produced when it is broken down by bacteria. (v) TDS: TDS prevent sunlight from penetrating the lake's bottom, lowering the amount of DO that is produced there. Low TDS levels are therefore desirable since they guarantee great water quality. (vi) Hardness: The Ca²⁺ and Mg²⁺ ions are primarily responsible for the production of fish scales and bones. (vii) Chloride: An excessive amount of chloride can injure aquatic organisms by impeding their osmoregulation process, which is a biological process that allows organisms to regulate the concentration of salts and other minerals in their bodily fluids (Hunt et al. 2012).

At first, we used six parameters (pH, TDS, hardness, chloride, ammonia, DO) to compute the lake's WQI. Following the initial phase of IDW, the mean values of the water quality metrics and the WQI surrounding the lake have been spatially varied using the interpolation technique used through GIS.

Comparing Silchar's temperature to those of other Assamese cities, it was discovered to be quite high (33.41 °C, 25.37 °C). During periods of intense rainfall, different contaminants and nutrients are transported to the lake, which causes a temporary shift in the lake's composition (Li et al. 2020). In the city of Silchar, there is 2,800 mm of yearly rainfall on average.

The NIT Silchar Lake has been chosen for the current study's investigation of its water quality. It is a naturally rain-fed lake that can be found at 24°51′20.25″N and 92°47′27.57″E longitudes. Numerous aquatic species, including frogs, fish, migratory birds, and plants, have found refuge in the lake.

As the WQI expression requires plenty of physiochemical and biological parameters, hence further studies can be made on the lake by involving heavy metals, enumeration of faecal indicator bacteria, etc. Though faecal coliform is not itself pathogenic, its presence indicates the presence of other pathogenic bacteria. The impact of the water quality parameters on the faecal coliform can also be studied by developing regression models. Other indices can also be calculated such as trophic state index and heavy metal index.

Locations A, B, C, and D are the four convenient sampling locations that have been chosen; they are illustrated in Figure 1. The co-ordinates of the sites A, B, C, and D are (24°45′21.29″ N, 92°47′27.14″ E); (24°45′20.00″ N, 92°47′25.31″ E); (24°45′19.48″ N, 92°47′26.97″ E); and (24°47^′19.98″ N, 92°47′29.74″ E), respectively.

Grab sampling was used to obtain samples from a depth of 0.5–1 m. The containers were completely cleaned and dried before sampling. Before examination, the containers were protected from sunlight to prevent any changes to their chemical content. Every week in the months of February and March 2022, samples were taken.

The aforementioned water quality metrics were chosen because they each have a different effect on the ecology of the water body and the availability of resources.

The WQI is determined by taking into account six different aspects of water quality: pH, total dissolved solids (TDS), chloride, DO, ammonia, and total hardness. The so-called ‘weighted arithmetic index approach’ (Brown et al. 1972) was used.

S_n is the default desirable value for the nth parameter. The parameter's real value in pure water is known as v_o.

For most parameters v_o = 0 except for pH and DO (pH = 7, DO = 14).

Table 1 classifies the WQI as per the categorization.

Further we have compared the Horton's WQI with another method. Das Kangabam et al. (2017) had developed a WQI model by taking 11 parameters, each of them has been assigned a parameter ranging from 1.46 to 5 based on expert's opinion from previous studies. The highest weight has been assigned to most significant parameters such as DO and lowest weights have been assigned to less significant parameters.

• 1.
  The relative weight was calculated using the following equation:

  

  (4)

  where RW refers to the relative weight, AW refers to the assigned weight of each parameter, and n refers to the number of parameters.
• 2.
  A quality rating scale (Q_i) for each parameter is computed by dividing its concentration in each water sample by its respective standard according to the guidelines of World Health Organisation. If WHO guidelines are not available, then the Indian Standard BIS is used and the result is multiplied by 100 using Equation (2):

  

  (5)

  while the quantity rating for pH and DO was calculated on the basis of

  

  (6)

Q_i is the quality rating, C_i refers to the value of the water quality parameters obtained from the analysis, S_i refers to the value of the water quality parameters obtained from WHO and BIS parameters, V_i refers to the ideal value for pH = 7.0, and DO = 14.6.

• 3.
  Next the sub-indices have been calculated to compute the WQI:

  

  (7)

Table 2 classifies the WQI generated from the above equation as per (Yadav 2010).

In order to generate surfaces and determine the predictions of unmeasured points, IDW interpolation relies on two statistical and mathematical techniques, the most common of which is the use of a weighted linear aggregate set of pattern points.

The lake on the academic campus is a naturally rain-fed body of water; no other sources of inflow have been found. The parameters for measuring water quality might vary, and the mappings produced using GIS are shown below.

In this study, we have measured seven parameters however only six parameters have been used in the WQI equation. As temperature and DO are highly correlated to each other, during summer the solubility of oxygen in water decreases with increase in temperature. Sunlight helps in self-purification, by adding oxygen during the process of photosynthesis. Further, temperature has also been found to effect other water quality parameters apart from DO. Hence temperature has been measured for establishing its relationship with other water quality parameters (Talukdar 2023). The coefficient of correlation was performed by using Python 3.11.3 packaged by Anaconda, Inc.

The month of March has the highest temperature (30.5 °C), and the month of February has the lowest temperature (22.4 °C). Since temperature readings are taken after sunrise, they might be rather accurate representations of the day's lowest temperature. Because there are no aquatic plants or other sunlight-blocking objects at position A, the spatial variation map of mean temperature at the four sites reveals that location A has a noticeably somewhat greater range of temperatures.

The system's pH is significantly controlled by lakes. The pH of naturally occurring rocks that contain carbonates or bicarbonates rises while falling in the presence of mineral or organic acids. The pH was found to be moderate, ranging from 7.9 to 8.1 and changing in the somewhat alkaline range. Overall, it was found that the system's pH was somewhat alkaline. According to BIS (1991) Standards and World Health (1993) recommendations, the ideal pH range for residential consumption is 6.5–8.5. Since rain is the only source of pollution entering the lake, the algae's activities, such as photosynthesis, which consumes carbon dioxide, are what cause the pH to vary (Gandaseca 2011).

TDS fluctuations were seen every seven days in the months of February and March.

At position A, the figure recorded was 56 mg/l, while at location B, it was 39 mg/l.

During these months, it has been observed that intense rainfall followed by runoff washes soil particles from the neighbourhood into the lake, temporarily causing turbidity to be evident.

The geology of the watershed mostly determines how hard the water is. According to Boyd & Tucker (1998), natural lakes can have a hardness as low as 5 mg/l of CaCO3 (1998). Locations A, B, C, and D had recorded average hardness values of 73, 81, 85, and 89 mg/l, respectively. Gandhi (2012) recommended a minimum CaCO₃ concentration of 50 mg/l for the better survival of fish in water bodies.

The average readings at the four sites (A, B, C, and D) are 10.4, 10.2, 8.8, and 10.8 mg/l. At position C, the DO value was measured to be 8 mg/l and the highest temperature was recorded to be 30 °C. In the winter months, levels of DO rose as a result of the drop in water temperature. For the maintenance of higher life forms, DO must be at least 2 mg/l, and for game fish, it must be at least 4 mg/l (Peavy et al. 1985).

52, 48, 48, and 42 mg/l of chloride on average were observed. The values, though, are below the accepted ranges. If there is an increase in chloride, fish will become poisoned and die.

There were measurements of average ammonia concentrations of 0.15, 0.1, 0.15, and 0.15 mg/l. The values that were seen were within the allowed tolerance of 0.5 mg/l. Eutrophication, a major harm to aquatic life, can result from a higher concentration of nutrients in water bodies (Khouni et al. 2021).

Both the WQI models have shown results ranging from 26 to 50. The WQI based on relative weight shows all the locations in good condition (Table 3), while the former one shows location B as excellent and others as good. According to our analysis, we would like to conclude that the latter one is more beneficial, as we can vary the weights because in real scenario, the pollutants or water quality parameters vary with time. However, both of the methods have proved to be excellent in water quality assessment, we can say that the weights can be assigned by quantifying the uncertainties or randomness in the datasets.

This study aims to evaluate the water quality of NIT Silchar Lake by using ‘weighted arithmetic index method’ and also mapping of the water quality parameters and the WQI by the IDW method in GIS environment. Based on the results obtained in this research, we can conclude that the lake is suitable for aquatic life. The water quality parameters along the four locations of the lake were found to be within the standards according to the USEPA (United States Environmental Protection Agency). The selected water quality parameters were pH, TDS, hardness, DO, chloride, and ammonia. Weighted Arithmetic Index Method classified location A as good, B as excellent, C as good, and D as good, thereby giving the lake as ‘A grade’ water quality. The relative weight method classified the four locations as ‘Good’. Furthermore, a comparison has also been made between the two methods demonstrating the importance of varying weights. Further analysis has to be done, such heavy metal index, trophic state index for declaring whether the lake water can be used for agricultural, domestic purpose, and it is good for survival of fishes. The IDW interpolation method has helped in mapping the water quality parameters at the unsampled location thereby giving a spatial distribution of the measured water quality parameters. WQI models such as Canadian Council of Ministers of the Environment Water Quality Index (CCMEWQI), National Sanitation Foundation Water Quality Index (NSFWQI), and other interpolation methods such as kriging, Moran's I on GIS can be used and compared with each other.

In this research, all water quality parameters have not been taken into account such as heavy metals, which can be further added into the WQI equation to get better results.

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

The authors declare there is no conflict.