Pedogeomorphometry: An Approach to Model Soil Across Spatial and Temporal Scales.

Pedogeomorphomometry integrates soil and landscape modelling in space and time. This concept attempts to combine pedometrics, digital soil mapping, and soil dynamic modelling for a better representation of soil properties, processes and entities. While digital soil mapping has been successfully used to predict the spatial distribution of soil, estimates are based on empirical relationships that are in general non-transferrable and involve no underlying mechanisms. Furthermore, the change in dynamics of soil properties such as soil carbon over time requires us to consider the time dimension in our maps. On the other hand, mechanistic or process-based models have been used effectively to predict the state of change in soil properties (soil carbon) at points in the soil landscape. Most process-oriented soil carbon models are just an austere representation of a topsoil layer with no spatial component. In addition, they need calibration with ‘real world’ data, and also require the knowledge of the initial state (or condition) of the soil. Digital soil maps which provide the spatial distribution of C are usually fed into a mechanistic model to predict the future change of C under different scenarios. However, we need to reverse this role and use our process-based understanding of a mechanistic model to help us make a better prediction of the spatial distribution of C.  This paper will therefore discuss pedogeomorphometry as an approach of dynamic spatiotemporal modelling of soil processes.