Processes and modeling of initial soil and landscape development: A review
The December 2016 issue of Vadose Zone Journal contains a review of soil and landscape evolution—the bio-hydrological processes during the initial development phases of geosystems or developments following an ecosystem disturbance. Initial development phases are observed, for instance, on glacial retreat areas due to climate-related changes. Other disturbances can be human induced such as agriculture, construction, or mining or caused naturally by landslides or volcanism.
The initial geo-hydrological system is basically a porous medium consisting of mineral particles characterized by the near absence of living matter. The impacts of external forcing (precipitation, wind, and insolation) and internal consolidation processes change the system over the course of time towards a more mature landscape with biota and developed soil horizons. The initial development of a landscape is also presumably highly dependent on the sequence and relative magnitude of events. In initial landscapes, development time is a limiting resource that may lead to similar features of mature systems, e.g., semi-arid landscapes show features (patterned vegetation and surface crusting) that can be observed during initial landscape development.
Another important factor of initial soil landscape development is the primary spatial heterogeneity. The nature of the initially present material (i.e., size, texture, and mineral composition) supposedly governs further development pathways to a great extent. The importance of the layer reaching from the bottom of the aquifer to the top of the vegetation for support of life has earned it the title “critical zone” and is now the focus of many scientific studies. Many areas around the world have established critical zone laboratories, and the results of these experiments have been published in numerous journals, including Vadose Zone Journal. Initial changes of the pore space in the uppermost layer of the critical zone affect soil hydraulic properties and, hence, the movement of water through the soil. Chemical (due to leaching and precipitation of minerals) and biological (due to the influence of microorganisms and roots) processes can change the pore size and distribution in the soil, thereby influencing infiltration and evapotranspiration.
Initial landscape development is not easy to model, as many (and often not very well known) interactions on various spatial and temporal scales need to be considered. However, for certain important aspects, well-established model from various disciplines exist. The depiction of the initial structural setup is of special importance as one of the most determining primary system components. Approaches that imitate structure-forming processes seem to be the most promising for a detailed structural model. Initial topographic evolution and the evolution of the soil surface is captured by landscape evolution or soil evolution models, which both have specific advantages and disadvantages regarding initial phase modeling. Generally, the difficulty of describing structures and processes during the initial phase is its dynamic, non-static behavior and the need to consider many continuously changing feedback loops.
An integrative model framework for the initial phase thus must take into account the relevant processes and the feedbacks between structure and structure-changing processes. It also needs to address the lack of data for model calibration and validation, adaptation of rapidly changing spatial patterns and process-structure interactions, standardization of spatial scales, and coordination of modeling time-steps. There are also many challenges in representing the spatial and temporal variability and proper coupling of model components. Validation of models using data from only a small number of test sites limits their applicability to other climates or structural conditions. A viable way to deduce a more general understanding of initial system evolution could be the introduction of three-dimensional functional units with a more or less homogeneous behavior (process domains).
Read the full article in VZJ. Free preview Feb 3 - Feb 10