2.4.2. Innovative Modelling of Mantle-to-Lithosphere-to-Surface Processes

The evolution of surface topography and morphology strongly depends on the interplay of subsurface and surface processes. Erosion unloads growing topography whereas sedimentation accelerates basin subsidence. This is clearly demonstrated by the strong correlation between denudation and tectonic uplift rates in zones of active deformation. During collision, surface processes contribute towards the localization and growth of mountain belts and fault zones, and ensure stable growth of topography (Fig. 24). During crustal extension, syn-rift erosion contributes towards widening of the rifted basin, so that apparent extension coefficients can increase by a factor of 1.5 – 2 (Fig. 25). Poly-phase subsidence and other deviations from thermal subsidence models can be also controlled by feedback between surface and subsurface deformation.

A new generation of 3-D and 4-D tectonically realistic models is required for an understanding of dynamic feedbacks between tectonic and surface processes, providing new insights into the evolution of tectonically active systems and related surface topography:

• Morphologically and tectonically consistent collision and exhumation models
• Basin modelling, synthetic stratigraphy
• Climate-coupled modelling

Fig. 25. a) Syn- and post-rift feedback conceptual model; b) Numerical model (Burov and Poliakov, 2001) of rift evolution with and without active surface erosion, for the same boundary and initial conditions. Erosion results in much stronger crustal thinning and a wider basin than in the case without erosion.

The topographic reaction to surface loading and unloading depends on the mechanical strength of the lithosphere as well as on the strength partitioning between the crust and lithospheric mantle. Consequently testing different rheological profiles in areas where the data on denudation/sedimentation rates are well constrained may provide new possibilities for constraining the long-term rheology of the lithosphere (e.g. Burov and Watts, 2006).

Reliable information on (de)coupling processes at the crust-mantle and lithosphere-asthenosphere boundaries and at the two principal phase transitions within the deeper mantle (at about 410 and 660 km depth) will be of fundamental importance for modelling surface topography. The quantification of dynamic depth-to-surface relationships is a major challenge, requiring innovative approaches to 4-D modelling. The principles of available conventional fluid-dynamic modelling are robust, but require greatly increased computer power to provide adequate resolution of a convection system characterized by thermal boundary layers, slabs and plumes of complex structure that may evolve rapidly. New approaches need to incorporate yielding rheologies of crustal and mantle materials, integrated modelling of material flow and elastic deformation (also crucial for predicting realistic topography evolution), crustal and lithospheric weakness zones and/or faults. To account for elastic and plastic deformation may actually require modifying available large-scale mantle dynamics models to solve, at least for the lithospheric part, full stress equations with free upper surface boundary conditions instead of flow approximations (Fig. 24c). Mantle models need to be constrained by mantle tomography, geodetic and electromagnetic data. The latest geo-modelling tools are able to consistently treat homogeneous and inhomogeneous deformation with realistic faults, so that the magnitude of uplift, subsidence, fluid flow and other types of deformation (derived from geological markers or GPS, stress in boreholes and earthquakes) can be linked and interpreted quantitatively. The goal of 4-D modelling is to quantify the dynamic evolution of Solid-Earth boundaries and phase transitions and associated surface deformation, and to define the present state of surface deformation, including its space-time gradient (a prerequisite for geoprediction). To achieve this goal, very high-resolution at temporal and spatial scales (e.g. 50-100 years, 5-10 km) is required.