Early landscape evolution - A field and modelling assessment for a post-mining landform

- We examine sediment output from four similar 30 m by 30 m plots on a rehabilitated mine site.
- There is an initial high pulse of sediment which rapidly reduces to rates similar to that of a natural a surface.
- This plot data was compared to predictions from a calibrated landscape evolution model.
- The landscape evolution model is able to reliably predict sediment output from these field plots.

Data from field plots describing how new surfaces evolve in the first few years post-construction are scarce in the literature. Here we examine sediment output from four similar 30 m by 30 m plots on a rehabilitated mine site over a six year period. Field measurements from the trial plots found that there is an initial high pulse of sediment over the first three years which rapidly reduces to rates similar to that expected for a natural or undisturbed surface. At 6 years the sediment output is equivalent to that expected from the surrounding undisturbed landscape. This plot data was compared to predictions from a calibrated landscape evolution model. The landscape evolution model used two sets of parameters, one derived from bare waste rock and one derived from an older vegetated surface. The simulations using bare waste parameters produced sediment output that matched the plot data in the first few years while the vegetated parameters produced sediment output which compared well with the field plot data at times > 3 years. The results demonstrate that when correctly calibrated the landscape evolution model is able to reliably predict sediment output from these field plots. These results suggest that there is the potential to employ the bare waste rock dump parameters for the first 3-4 years then switch to vegetated parameters for the longer term modelling. Both the field plots and landscape evolution model simulations displayed considerable annual variability in total load. This variability is the result of different surface structure from imposed surface roughness (ripping by a bulldozer) and their unique topographic structure. Both initial DEM and model parameters have a large influence on predicted sediment load. The results here support the reliability of the model at the sub-metre grid scale.