气候变化领域集成服务门户

We evaluated global soil organic carbon (SOC) stocks and turnover time predictions from a global land model (ELMv1-ECA) integrated in an Earth System Model (E3SM) by comparing them with observed soil bulk and Delta C-14 values around the world. We analyzed observed and simulated SOC stocks and Delta C-14 values using machine learning methods at the Earth System Model grid cell scale (similar to 200km). In grid cells with sufficient observations, the model provided reasonable estimates of soil carbon stocks across soil depth and Delta C-14 values near the surface but underestimated Delta C-14 at depth. Among many explanatory variables, soil albedo index, soil order, plant function type, air temperature, and SOC content were major factors affecting predicted SOC Delta C-14 values. The influences of soil albedo index, soil order, and air temperature were primarily important in the shallow subsurface (<= 30cm). We also performed sensitivity studies using different vertical root distributions and decomposition turnover times and compared to observed SOC stock and Delta C-14 profiles. The analyses support the role of vegetation in affecting soil carbon turnover, particularly in deep soil, possibly through supplying fresh carbon and degrading physical-chemical protection of SOC via root activities. Allowing for grid cell-specific rooting and decomposition rates substantially reduced discrepancies between observed and predicted Delta C-14 values and SOC content. Our results highlight the need for more explicit representation of roots, microbes, and soil physical protection in land models.

Plain Language Summary Quantifying feedbacks between the terrestrial carbon cycle and climate is important for understanding climate change. Among many factors that control terrestrial carbon cycle responses to climate, soil organic carbon (SOC) dynamics are particularly important, although highly uncertain. In addition to SOC stocks, radiocarbon is an important observational constraint for land model predictions. We evaluated, against worldwide observations of SOC stocks and radiocarbon, predictions from a new land model used for climate change analyses. We analyzed differences between model predictions and observations using a machine learning method at a large grid cell scale (similar to 200km). Among many explanatory variables, soil albedo index, soil order, plant function type, air temperature, and SOC densities were major factors affecting predicted SOC radiocarbon values. The influences of soil albedo index, soil order, and air temperature were primarily important for topsoil. Our sensitivity analysis highlights the role of plant root activity in affecting soil carbon turnover, particularly in deep soil, possibly through supplying fresh carbon and degrading physical-chemical protection of SOC. Allowing for grid cell-specific rooting and decomposition rates substantially reduced discrepancies between observed and predicted values. Our results highlight the need for more explicit representation of roots, microbes, and soil physical protection in land models.