Short-Term Predictions of Evaporation Using SoilCover at the Near-Surface of a Mine Waste Pile following Heavy Rainfall Events

Abstract

Accurate measurements and predictions of near-surface soil drying and evaporation following heavy rainfall events are often needed for research in agriculture and hydrology. However, such measurements and predictions at mine waste pile and tailing settings are limited. The prediction of evaporation at mine waste piles is essential for many problems in geotechnical engineering, including the design of soil cover systems for the long-term closure of hazardous waste sites, and thus mitigates, for example, the generation of acid mine drainage (AMD) and metal leaching. AMD is one of mining’s most serious threats to the environment. This study investigated the short-term (8 days) and medium-term (27 days) drying rates and evaporative fluxes at the surface and near-surface of the Deilmann South waste-rock (DSWR) pile at the Key Lake uranium mine, northern Saskatchewan, using the gravimetric (GV) method and SoilCover (SC) model, respectively, during and following heavy rainfall events for the environment. The SC simulation results showed that during the weather-controlled stage (Stage I) of the first 5-day period of rainfall events, while the surface was wet, the potential evaporation (PE) was equal to the actual evaporation (AE) (i.e., AE/PE = 1). As the surface became drier on Day 6, the cumulative PE began to separate from the cumulative AE and the surface’s drying rate rapidly diverged from those at the deeper depths. This occurrence signaled the onset of the soil profile property-controlled stage (Stage II). As the drying continued, the surface became desiccated and the slow-rate drying stage (Stage III) was established from Day 7 onward. The SC-simulated AE results were compared to those measured using the eddy covariance (EC) method for the same test period at the DSWR pile in a different study. The comparison showed that the two methods yielded similar AE results, with 18% relative errors. The results of this study provided the opportunity to validate the SC model using actual data gathered under field conditions and to ascertain its ability to accurately predict the PE and AE at the surfaces of mine waste piles.