Abstract
Due to the rising pressure from governmental regulations, the water recycling rate has increased significantly in mining operations over the last decades, resulting in a high variation of process water quality, which could potentially impact the plant performance. The current effort to assess water quality in mining is shifting from managing water to fulfill environmental regulations (focus on the effluents) to controlling water quality to maintain the operating performance (focus on the water within the process). However, minimal effort has been made to design a dedicated sampling procedure for process water. This study investigates the use of multivariate variography and principal component analysis (PCA) for improving the process water sampling procedure at the Kevitsa Mine, Finland. The aim is to design a sampling procedure for evaluating water quality using two different types of datasets and illustrating the impact of the dataset structure on the sampling design. The results showed that the common spot sampling procedure generated a very high sampling error and was not the best practice for process water. The weekly sampling frequency used at the mine site, suitable for fulfilling environmental regulations was too low to capture the process water variation. Therefore, it is not recommended to use environmental water datasets for operating control purposes. The multivariate variographic analysis revealed the hidden cyclic variation through its ability to summarize the time variations and the correlation between multiple variables that were not visible through the classical univariate variogram approach. However, the number of increments recommended by the global multivariogram became impractically high. Hence, an alternative approach combining PCA to the mutivariogram was used to filter noise from the data and keep the relevant information. This study highlights the benefits of using multivariate variography to improve water sampling procedures in the mining industry and to reduce both operational and environmental risks associated with water quality variability. Thus, this method has the potential to be used in worldwide mining operations as a standard procedure for sampling water to provide reliable results.
Highlights
Water quality assessment is a critical aspect of mining operations in fulfilling environmental regulations and controlling plant performance (Mudd, 2010; Kinnunen et al, 2021)
This study investigates the use of multivariate variography and principal component analysis (PCA) for improving the process water sampling procedure at the Kevitsa Mine, Finland
This study highlights the benefits of using multivariate variography to improve water sampling procedures in the mining industry and to reduce both operational and environmental risks associated with water quality variability
Summary
Water quality assessment is a critical aspect of mining operations in fulfilling environmental regulations and controlling plant performance (Mudd, 2010; Kinnunen et al, 2021). It is clear that for an environ mental approach, a suitable monitoring program is essential for reducing the risks of pollution related to water (Bezuidenhout, 2009; Muniruzzaman et al, 2018). For such objectives, the data should be collected regularly in order to identify long- term trends, periodic changes, and fluctuations in rates of changes (Canada Environment, 2009). When performed for process performance purposes, the sampling frequency needs to be high enough to capture the variability of process water properties that need to be monitored. The latter are defined according to their relevance to the process and are often case-specific
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
