Multi-physics, multi-domain, and multi-scale coupling concept for groundwater to surface-water interactions driving submarine landscape evolution

Groundwater is an important geomorphic agent, and its interactions with land and surface-water play a critical role in driving landscape evolution through highly coupled and complex processes such as internal erosion, surface erosion, slope failure and debris flows.

From a modelling point of view, a common approach has been to tackle the hydrologic and geomorphic processes separately. To date, the multi-domain and multi-physics interactions across the groundwater – surface-water interfaces remain poorly conceptualized. Furthermore, the groundwater models are typically highly simplified; the effects of thermal fluxes, diagenetic processes, multi-phase transport and phase transitions, density driven flows, etc. are either ignored, or introduced through lumped or reduced models without fully resolving the underlying process couplings. In sediments that are susceptible to chemical weathering, like carbonate rocks, there are almost no known and validated models that can handle evolving pore geometry due to carbonate dissolution in conjunction with internal erosion. Moreover, the saturated and unsaturated groundwater, surface-water, seafloor, and subaerial terrain constitute multiple distinct domains. The interactions across these domains occur at multiple length and time scales, and the landscape evolution results in changing geometries of each domain.

Due to the multi-physics, multi-scale, and multi-domain characteristics of these models, the resulting systems of equations are highly complex, nonlinear, non-standard, and computationally challenging, both in terms of numerical implementation as well as computing resources. These features not only add to the conceptual and numerical complexity, but also demand specialized and targeted solutions.

Here we present an abstract modular framework to resolve the multi-physics couplings within targeted sub-domains (i.e., 2D/3D sub-surface and1D/2D surface-water, sea-floor and subaerial terrain) and across the domain interfaces. We also introduce an algorithm to handle the changing geometries of these domains. This framework is being developed within the ERC project ‘MARCAN’ and will be used to assess the role of groundwater flow and seepage on the geomorphic evolution of continental margins.