Spatial interpolation of water quality index based on Ordinary kriging and Universal kriging

Water is a very vital needed substance in order to maintain the important activities of humans. However, contaminated water can transmit diseases such as typhoid, dysentery, diarrhea, cholera and polio. Pakistan is a highly affected country due to the scarcity of safe and healthy water sources. The current study mainly focuses to determine water quality in the selected area. For this purpose, an integrated surface water quality index (SWQI) based on 18 hydrochemical parameters is employed. SWQI substantially correlates with pH, EC, SO4, HCO3 and heavy metals. Therefore, these parameters are utilized in the calculation of SWQI. The results of SWQI show that about 69.23% of samples are 'very good quality'. Moreover, 30.77% of the total samples are of poor quality and are identified as unsuitable for drinking. Further, Ordinary kriging (OK) and Universal kriging (UK) are used to predict the SWQI at unobserved locations and map the SWQI in the study area to explore the spatial distribution. The prediction performance of the two kriging methods is assessed through cross-validation in terms of root mean square prediction error (RMSPE). It is concluded that the prediction performance of the UK is better than OK in terms of RMSPE.

Automated summary

Pakistan is highly affected by the scarcity of safe water sources. A study was conducted to evaluate water quality in the selected area using a surface water quality index (SWQI) based on 18 hydrochemical parameters. The results showed that 30.77% of the samples were unsuitable for drinking. Furthermore, Ordinary kriging (OK) and Universal kriging (UK) were used to predict the SWQI and map its spatial distribution, with UK outperforming OK in prediction performance.