Soil erosion predictions from a landscape evolution model - An assessment of a post-mining landform using spatial climate change analogues

摘要

Landscape Evolution Modelling (LEM) technologies provide a means by which it is possible to simulate the long-term geomorphic stability of a conceptual rehabilitated landform. However, simulations rarely consider the potential effects of anthropogenic climate change and consequently risk not accounting for the range of rainfall variability that might be expected in both the near and far future. One issue is that high resolution (both spatial and temporal) rainfall projections incorporating the potential effects of greenhouse forcing are required as input. However, projections of rainfall change are still highly uncertain for many regions, particularly at sub annual/seasonal scales. This is the case for northern Australia, where a decrease or an increase in rainfall post 2030 is considered equally likely based on climate model simulations. The aim of this study is therefore to investigate a spatial analogue approach to develop point scale hourly rainfall scenarios to be used as input to the CAESAR -Lisflood LEM to test the sensitivity of the geomorphic stability of a conceptual rehabilitated landform to potential changes in climate. Importantly, the scenarios incorporate the range of projected potential increase/decrease in rainfall for northern Australia and capture the expected envelope of erosion rates and erosion patterns (i.e. where erosion and deposition occurs) over a 100 year modelled period. We show that all rainfall scenarios produce sediment output and gullying greater than that of the surrounding natural system, however a 'wetter' future climate produces the highest output. Importantly, incorporating analogue rainfall scenarios into LEM has the capacity to both improve landform design and enhance the modelling software. Further, the method can be easily transferred to other sites (both nationally and internationally) where rainfall variability is significant and climate change impacts are uncertain.