Abstract
The time interval between the prediction of pit lake water quality and the determination of actual pit lake water quality following pit lake filling can be decades to centuries. This presents a significant challenge to pit lake modelers, mining companies, and other stakeholders who wish to validate pit lake water quality predictions. To address this challenge, Eary and Castendyk (2009) recommended the construction of a database of pit lake data to “…allow modelers to anticipate future water quality based on reported trends, compare observed values to initial predictions, and scrutinize models to identify accuracies and inaccuracies in modeling approaches.” The International Network for Acid Prevention (INAP) sponsored the development of a searchable database of published water quality from existing pit lakes in 2010 (presented at ICARD 7). Since then, 2,584 water quality observations from 254 pit lakes and 20 ore body types (using the USGS Geo-Environmental designation) have been added, collected from multiple depths down to 198 meters. The cloud-based database includes values for up to 49 parameters for each water sample, plus a reference to the data source as well as raw data available for download from a given pit lake or ore body type. Time series plots, Ficklin Diagrams, concentration vs. depth plots, box and whisker plots, and Piper Diagrams can all be generated using the PowerBI solution. The applications of this pit lake database are numerous, including the validation of pit lake models using data collected from similar ore bodies, and potentially integrating the compiled data into meaningful water quality predictions using machine learning and artificial intelligence.