MINErosion 4: A user-friendly catchment/landscape erosion prediction model for post mining sites in Central Queensland

Abstract

Soil erosion from post mining landscapes subjected to significant rainstorm events is believed to have adverse effects on the surrounding environment, as well as the mining processes. The research project reported in this thesis was encouraged by the fact that, to the best of the author’s knowledge, there is a lack of catchment/landscape scale erosion/deposition model, to deal with the special conditions of post-mining rehabilitation sites.

Previous research has produced the hillslope erosion model MINErosion 3.01 which was useful to determine the parameters (slope gradient, length and vegetation cover) required to design a postmining landscape that meets the criteria of acceptable erosion rates (<40 t/ha/y). However, MINErosion 3.01 was not suitable to determine the erosion rates from whole catchments or whole of mine landscapes and a model is needed to provide the environmental officers with a tool to allow them to manage the whole of mine landscape, specifically in relation to offsite and onsite discharges of water and sediment. MINErosion 3.01 was found to be a suitable for upscaling to produce a new user-friendly catchment scale model named MINErosion 4. As an initial stage, MINErosion 3.01 was revised to fix some of its errors and add some new features and a new version MINErosion 3.1 was developed and validated against plot data from a previous project (Postmining Landscape Parameters for Erosion and Water Quality Control, 1992-1998). The agreement between predicted (Y1) and measured (X1) annual average soil loss is good with a regression equation of Y1 = 0.8 X1 + 0.005 and an R2 = 0.70; while predicted (Y2) and measured (X2) rainstorm erosion events have a regression of Y2 = 0.867 X2 with an R2 of 0.68.

As the new model should be a spatial distribution model, the variability of media properties of Central Queensland coal mines need to be determined. Ninety three soil and spoil samples were collected from six selected coalmines. They were analyzed and represented in both MINErosion 4 model database and as a standalone database file for Central Queensland coalmines media properties. These properties were used to derive the rill and interrill erodibilities of soil and spoil. These values are generally in agreement with the values in the embedded MINErosion 3.1 database as determined by Sheridan (2001). This work shows large variability in soil and spoil erodibilities which should result in large variability in erosion rates across a minesite. A knowledge of the hotspots in advance will assist the mine to manage and allocate suitable resources across the postmining landscape.