Assessment of Groundwater Quality Using GIS - A Case Study of the Churu District of Rajasthan - page 05

Fig05a
Figure 5a: Spatial Distribution of Water Quality Index (case II) in the study area for Pre Monsoon duration.

 

Fig05b
Figure 5b: Spatial Distribution of Water Quality Index (case II) in the study area for Post Monsoon duration.

 

Tab05

 

The groundwater quality index assessed in Case I from the groundwater quality data values range from 0-543 and 0- 789 during pre and post monsoon respectively (Table 5). Based on the standard classification (Table 4), the groundwater quality status ranges from good to very bad. Figures 3a and 3b classify the WQI values into three groundwater quality zones (good, bad and very bad) and provide the spatial extent of it. The areas covered by the very good water quality lie in the western portion of the study area and are confined to a very local portion. The bad quality water is present in small patches in the rest of the area. The remaining area have very bad water quality. After rain, a local patch of good water quality in the western portion developed. The distribution of wells in the value range of the quality index is visible in Figure 4 and displays the changes during 2 different seasons.

The water quality index assessed in Case II from the groundwater quality data; values ranged from 0-2729 and 0-3279 during pre and post monsoon respectively (Table 5). In this case the water quality is categorized into good, bad and very bad zones. Figures 5a and 5b showing the spatial extent of the WQI, indicating that the higher values of WQI are associated with the Churu district, and have been found to be mainly due to nitrate and total dissolved solids. During pre monsoon, the western part of the study area has good quality water. Some very local patches of good WQI are randomly present in the area. The bad quality index has a negligible area and only makes a thin boundary between the good and very bad WQI region. The rest of the area has very bad water quality. While, during post monsoon the local patches of good water quality developed in NE of the study area. For the rest of region conditions remained very similar to that of the pre monsoon. The bad quality area slightly increases from pre monsoon to post monsoon. Figure 6 show the variation of distribution of wells in the study area and where they stand on the WQI scale.

The hierarchal cluster analysis was applied to identify the processes controlling groundwater chemistry. The dendrogram (Figure 8) displayed two clusters. Cluster 1 comprised of turbidity, colour and temperature showing close linkage and Cluster 2 comprised of TDS, Nitrate, Nitrite, Fluoride and pH and joined with the same Cluster 1.

This cluster is interpreted as anthropogenic contamination of water, which might be related to natural processes.

The value range for water quality index categories has been assigned by checking the weight-age given to wells in the quality index calculation i.e. the WQI =0 to 50 means the water is safe for human consumption and the values exceeding 50 are not safe for consumption as all parameters are crossing specified limits of water quality for human consumption.

The results in Table 5 show that every method would generate a different value range of water quality indexes. It would be difficult to standardize the value range of WQI. In this work we have shown that one equation reduces the value range while the other increases it.

The comparison of the 2 different WQI derived from two different sets of equations shows that the pattern is the same for water quality during pre and post monsoon from one set of eq. (The location having low values of WQI from one set of eq. also has low WQI from another set of eq.; but there is offset present between them). The location having a high WQI value in Case 1 also showed high WQI values in Case 2 (Fig 7a and 7b).

Table 6 indicated positive correlation among the physico-chemical parameters ranging from 0.0059 to 0.956. Clustering indicates that turbidity is associated with colour, TDS is associated with Nitrate and Fluoride is associated with pH. The negative correlation in cluster might indicating the different origin. The correlation matrix shows that there is a direct relation between the following, viz. Turbidity and Colour, TDS and Nitrate and Fluoride and pH. Other parameters have either low correlation or negative correlation. From this high correlation of parameters we can suggest the elements of correlation matrix as having a value of more than 0.35 which are responsible for the bad water quality of water in the study area. Only 3 elements of the correlation matrix from 8 are responsible for bad water quality.

 

Fig06
Figure 6: Water Quality Index (case II) of groundwater samples location for Pre & Post Monsoon duration.

 

Fig07a
Figure 7a: Water Quality Index (case 1 & Case 2) of groundwater samples location for Pre Monsoon duration.

 

Fig07b
Figure 7b: Water Quality Index (case 1 & Case 2) of groundwater samples location for Post Monsoon duration.

 

Tab06

 

Fig08
Figure 8: Dendrogram of groundwater samples Parameter.

 

Conclusion

The calculated WQI from the empirical formulas for the groundwater samples range from 0 to 789 in one case, and in another case the value ranges from 0 to 3,279 and falls within the Good (Acceptable), Bad and Very Bad (Rejected) class. Every method generated a different index range. The geographical information system delineated the area into good to very poor potential areas of groundwater quality and therefore proved to be a useful tool in mapping groundwater quality and analyzing the vulnerability of the population associated with bad quality. Hierarchal cluster analysis identified anthropogenic contamination and natural mineralization as the major processes controlling groundwater chemistry. It also helped in minimizing the parameters responsible for water quality. It is recommended that groundwater quality monitoring should be encouraged in order to ensure groundwater quality protection and conservation. The pattern of WQI for the study area is very similar from the 2 different set of equations used.