Abstract
[1] Most global ecosystem models assume that controls over soil organic matter are alike in climatically similar regions. In this study, we tested the generality of controls over soil organic carbon (SOC) and soil organic nitrogen (SON) in temperate grasslands. We measured organic matter pools in Inner Mongolia, China, along the Northeast China Transect, and analyzed the relationship of SOC and SON to climate, soil texture, and land use variables. We then compared our data to values simulated by a regression model developed in the U.S. Great Plains and also to Century model simulations. We found that, as in the U.S. Great Plains, climate and soil texture variables could explain a large proportion of variation in observed SOC and SON, but a regression model developed in the Great Plains overestimated SOC and underestimated SON in Inner Mongolia. Using Century, we found that simulated SOC and SON values were sensitive to both inclusion of altered land use and changes in N deposition and that the model that best fit our data included higher-intensity grazing and N deposition values higher than that in the Great Plains. This model also produced aboveground net primary production (ANPP) values comparable with values observed in the literature for Inner Mongolian grasslands, but these values were higher than ANPP predicted by previously published regression models. These results suggest that different controls over SOC and SON cycling in Inner Mongolia may affect our ability to predict SOC and SON pool sizes using relationships in other regional models.
Highlights
[2] A central challenge in global biogeochemical modeling is developing a generalizable structure that accurately captures variation among ecosystems
In contrast to other modeling approaches that focus on parameter optimization and testing of mechanisms, by testing the generality of a simple model and a highly parameterized model, we could better explain discrepancies that arise between predicted values and values we measured across the Inner Mongolia transect, and improve our understanding about how ecosystem dynamics may differ between the two regions
Because we knew that this region may have undergone significant land use intensification in the 1950s [Sneath, 1998; Xiong et al, 2008], and that recent estimates report N deposition levels higher than that in the Great Plains [Lu and Tian, 2007], we focused on N deposition and land use as possible influences of simulated Soil organic carbon (SOC) and soil organic nitrogen (SON)
Summary
[2] A central challenge in global biogeochemical modeling is developing a generalizable structure that accurately captures variation among ecosystems. In an attempt overcome this variation, most global ecosystem models assume that in climatically similar regions, such as grasslands, relationships between SOC and its environmental controls are the same, despite regions evolving independently The extrapolation of these relationships in ecosystem models allows us to predict ecosystem dynamics in the future and over large regions for which we have little data, and improve our understanding of these systems. EVANS ET AL.: SOC AND SON IN CHINESE AND US GRASSLANDS ecological convergence and recent global change Considering these potential influences, it is important to continue to test the generality of grassland models developed in one region for their application for all regions of similar climate. We use a grassland regression model developed in the Great Plains [Burke et al, 1989] to test its applicability to Chinese grasslands, and the more highly parameterized Century model [Parton et al, 1987] to investigate which parameters are most important for simulating predictions comparable to our data
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
