In part 1, I reported on my sabbatical visit to New Zealand grew into a three-way collaboration between labs in New Zealand, the US and Canada. Here in part 2, I’ll describe one conceptual framework that the Powell Center working group on soil carbon came up with to guide our thinking about the next generation models for soil carbon cycling. And in part 3 (coming soon), I’ll describe how data, code and results are fused together in ongoing projects.
Part 2: Concepts, measurements, models
I currently have the pleasure of working with 14 other great researchers on a USGS-funded Powell Center Working Group project called “What lies below? Improving quantification and prediction of soil carbon storage, stability, and susceptibility to disturbance”, led by Corey Lawrence (USGS), Kate Heckman (US Forest Service), Marco Keiluweit (UMass) and Susan Crow (University of Hawaii). According to their website, “the Powell Center serves as a catalyst for innovative thinking in Earth system science research by providing the time, creative space, and computational, data manipulation and data management resources to promote synthesis of existing information and emergent knowledge”, and our working group project seeks to “synthesize information on the processes controlling soil C storage across different spatial scales and develop new procedures to translate local measurements to the regional and global scale datasets used by models”. We met for an intensive week in early May at the Powell Center in Fort Collins, CO.
One of the conceptual frameworks that the group intends to pursue is the union of concepts, measurements and models. The premise here is that our current models of soil C cycling (like Century) have matured well, and are quite robust because they adequately balance concepts, measurements and modeled processes. There are, in a sense, balanced or equilateral triangles. However, a new generation of models seek to be more mechanistic and better able to predict soil C responses to disturbances such as climate change. These new models can be loosely grouped into several efforts that have implicit strengths in concepts, measurements or modeled processes. Analogously, these models are unbalanced triangles. For example, there is a vast (and growing) amount of data and measurements on the molecular biology of the soil microbial community, but a weaker understanding of how this genetic diversity translates to soil C cycling processes, or how these data might be incorporated into next-generation models. Or, we understand conceptually that soil mineralogy has a profound effect on the formation of organo-mineral associations that stabilize organic C, and we have a large number of measurements of C and mineralogical characteristics. However, our models frequently only use soil clay content (if anything at all) to represent mineral stabilization of organic matter. Our Powell Center Working Group has tasked itself to explore and illustrate these imbalanced triangles as a way to identify knowledge gaps, and paths forward to the development of new, robust and mature models.