Soil systems – the challenges of complexity and scale
Soils are complex systems, in which physical, geochemical and biological processes interact in aggregate structures situated in dynamically shifting air- and water-filled spaces. It is difficult to adequately sample soil properties and to model processes related to those soil measurements. These challenges were discussed in a stimulating three-day conference on Complex Soils Systems in Berkeley a few weeks ago. Attendees came from an incredible diversity of backgrounds with a common interest in tackling issues in soil science. The need to better understand soils was motivated by the importance of soil processes in climate and for figuring out “How to feed the soil and the planet?” in the anthropocene – a question posed early on by Professor John Crawford.
Issues of scale were brought up explicitly or were evident implicitly in many of the presentations. Namely, that relevant processes in biogeochemical cycles occur over a wide range of spatial (nano- to mega-meter) and temporal (seconds to millennia) scales, but our observations are typically limited to a much narrower scope given measurement and resource constraints. These issues were elegantly summarized in the recent article “Digging Into the World Beneath Our Feet: Bridging Across Scales in the Age of Global Change” by Hinckley, Wieder, Fierer and Paul in Eos, Transactions American Geophysical Union 95 (11), 96-97. In a real sense, the scale issue presents problems when societal decisions regarding soil sustainability and ecosystem services are made using data and models derived from different (often smaller) spatial scales than are relevant to the policies and issues themselves.
One illustration of the concept of a spatially complex soil system is illustrated with the figure below by California College of the Arts (CCA) student Sakurako Gibo. The image depicts a theoretical assemblage of soil microbes with different morphologies (for instance round spores versus string-like mycelia). In the second figure, the complex system is “pulled apart” into bins that might represent the effect of a sampling strategy that subsamples components of the whole system. The information about the original complex assemblage and connections is not retained, and as a result, data and rules based off of the binned samples may be different from the case in the real intact community.