Regional and temporal variation in net primary production and nitrogen mineralization in grasslands

Spatial variability that occurs at large scales has long been used by ecologists as a tool to examine the controls over ecosystem structure and function. Correlations of control variables such as climatic factors and response variables such as vegetation and soil carbon storage across broad regions have played a crucial role in predicting the response of ecosystems to global climate change. Despite the importance of these large-scale space-for-time substitutions, there are substantial limitations. One of these limitations is that many of the possible control factors covary with one another, and only some of the important control factors actually exist in large-scale databases. Thus, the true proximal controls may be difficult to identify. A second limitation is that models of spatial variability may not be appropriately applied to temporal variability. In this paper, we utilize a new approach to determine the extent to which N availability may constrain aboveground primary productivity in the Central Grassland region of the U.S. The strong relationship between average annual primary production and average annual precipitation found in spatial patterns in ecosystems globally has often been interpreted as evidence of a fundamental water limitation. However, temporal variation in annual aboveground net primary production (ANPP) indicates that other factors constrain production. We generated a spatial and temporal database for annual aboveground net primary production and annual net N mineralization by linking a database of input variables (precipitation, temperature, and soils) with predictive models. We generated independent data sets of aboveground net primary production and net N mineralization by using regression models to predict aboveground net primary production, and the Century model to simulate net N mineralization. Our analyses indicate that net primary production and net N mineralization both increase with mean annual precipitation; thus, it is not possible to separate the extent to which ANPP is controlled by water or N availability. Nitrogen use efficiency (NUE) increased with increasing precipitation across the region. Aboveground net primary production decreased with increasing temperature across the region, while N mineralization increased slightly, leading to decreasing (NUE) with increasing temperature. At high precipitation levels, aboveground net primary production increased and N mineralization decreased slightly with increasing soil fineness. Nitrogen use efficiency generally increased with increasing pools of soil organic matter, likely because in grasslands, the proportion of recalcitrant organic matter increases with the total organic matter pools. A comparison of interannual variation in net N mineralization with average spatial variation indicated a high degree of inertia in the response of N availability to precipitation levels. Our simulation results as well as field results of Lauenroth and Sala (1992) raise important questions about the applicability of space-for-time substitutions when dealing with ecosystem function. The structure of the systems appears to provide important constraints on the temporal variability that are not evident in an analysis of spatial variability.