Land degradation and soil erosion are perceived as important problems facing humanity in the next 50 years. Many spatially distributed soil erosion models have been developed for small spatial units (e.g. individual plots or fields). Their application at a regional scale (e.g. large drainage basins) was hitherto not very successful since most models are specific to particular space and time scales. Often the solutions used to address these problems are relatively crude, and in many cases lack of input data is also a limiting factor.

For example, there is clear evidence that sediment yield rates are generally less for larger areas due to, among other things, soil patchiness, the duration of intense bursts of rainfall and imperfect connectivity; and at larger scales due to the size of storm cells. Some of these effects are being incorporated into models through considering sediment connectivity within catchments, and through specifically addressing issues of up and down scaling, together with the appropriate representation of processes across the range of scales. The complexity and dynamics of erosion and sediment system in catchments are therefore key issues that all researchers are facing, either explicitly or by default.

Particular challenges identified in improving our understanding of soil erosion include:
1. The complexity and large number of relevant interactions, with the emergence of hierarchical levels of organization that make it difficult to link the dynamics across a broad range of scales.
2. The high level of spatial heterogeneity and how to incorporate it in a description of the interaction and connectivity network, including size selectivity and the stochastic nature of particle deposition and consideration of the anastomosing flow pathways on hillsides
3. The temporal resolution of rainfall intensity, since the dynamics of sediment transport depend on high resolution data that is not widely available outside research catchments, so that models may have to rely on, for example, daily rainfall totals.
4. The uncertainties in data, spatial parameterisation and process understanding, and their implications for reliable forecasting, particularly at the scale of small to medium catchments.
5. Difficulties in linking quantitative and qualitative knowledge.


Selected presentations will be considered for publication in a special issue of an international journal.

Funding provided by EGU YSTA support and by Spanish Ministerio de Ciencia e Innovación (MICINN) through project no. AGL2010-21501/AGR is greatly appreciated.