Using hydrological modelling and data-driven approaches to quantify mining activities impacts on centennial streamflow

Exploring the impacts of mining-related human activities on catchment streamflow time series can facilitate understanding of streamflow response to external factors, including climatic inputs. This study selects the Goulburn River catchment that is located in southeastern Australia having more than 100 years of streamflow observations and more than 90 years of mining history, and divides the streamflow period from 1913 to 2015 into four sub-periods: 1913-1920 (no mining), 1921-1950 (small-scale underground mining), 1951-1980 (commence of new underground mining) and 1981-2015 (large-scale mining of open cut and underground mining). Process-based hydrological modelling (SIMHYD), simple one-parameter hydrological model (Budyko-based method) and a data-driven approach (the double mass curve, DMC), are used to assess mining-related centennial streamflow changes and to quantify relative contributions of climate variability and mining activities on streamflow. These results show that streamflow increased during periods of underground mining, which is likely due to over-pumping to sustain the underground mining. The difference between the observed streamflow and the SIMHYD simulated was very marginal in the reference period of 1913-1920; while it increased by 3.39 mm yr(-1) in 1921-1950 and by 10.51 mm yr(-1) in 1951-1980. The DMC slope increased from 0.021 in 1913-1920 to 0.025 in 1921-1950, then 0.033 in 1951-1980. These impacts of underground mining on increased streamflow have been greatly reduced and even become negative when open cut mining was largely expanded. The average difference of streamflow during the period of 1981-2015 was reduced from 10.51 mm yr(-1) to 2.84 mm yr(-1). The DMC slope decreased from 0.033 in 1951-1980 to 0.027 in 1981-2015. The contributions between climate changes and mining activities to streamflow changes estimated using these methods are consistent in trends in different mining periods, but noticeably different in magnitudes.