Experimentally derived nitrogen critical loads for northern Great Plains vegetation.

The critical load concept facilitates communication between scientists and policy makers and land managers by translating the complex effects of air pollution on ecosystems into unambiguous numbers that can be used to inform air quality targets. Anthropogenic atmospheric nitrogen (N) deposition adversely affects a variety of ecosystems, but the information used to derive critical loads for North American ecosystems is sparse and often based on experiments investigating N loads substantially higher than current or expected atmospheric deposition. In a 4-yr field experiment in the northern Great Plains (NGP) of North America, where current N deposition levels range from ~3 to 9 kg N·ha-1 ·yr-1 , we added 12 levels of N, from 2.5 to 100 kg N·ha-1 ·yr-1 , to three sites spanning a range of soil fertility and productivity. Our results suggest a conservative critical load of 4-6 kg N·ha-1 ·yr-1 for the most sensitive vegetation type we investigated, badlands sparse vegetation, a community that supports plant species adapted to low fertility conditions, where N addition at this rate increased productivity and litter load. In contrast, for the two more productive vegetation types characteristic of most NGP grasslands, a critical load of 6-10 kg N·ha-1 ·yr-1 was identified. Here, N addition at this level altered plant tissue chemistry and increased nonnative species. These critical loads are below the currently suggested range of 10-25 kg N·ha-1 ·yr-1 for NGP vegetation and within the range of current or near-future deposition, suggesting that N deposition may already be inducing fundamental changes in NGP ecosystems.

Climatic controls of aboveground net primary production in semi-arid grasslands along a latitudinal gradient portend low sensitivity to warming.

Although climate models forecast warmer temperatures with a high degree of certainty, precipitation is the primary driver of aboveground net primary production (ANPP) in most grasslands. Conversely, variations in temperature seldom are related to patterns of ANPP. Thus forecasting responses to warming is a challenge, and raises the question: how sensitive will grassland ANPP be to warming? We evaluated climate and multi-year ANPP data (67 years) from eight western US grasslands arrayed along mean annual temperature (MAT; ~7-14 °C) and mean annual precipitation (MAP; ~250-500 mm) gradients. We used regression and analysis of covariance to assess relationships between ANPP and temperature, as well as precipitation (annual and growing season) to evaluate temperature sensitivity of ANPP. We also related ANPP to the standardized precipitation evaporation index (SPEI), which combines precipitation and evapotranspiration to better represent moisture available for plant growth. Regression models indicated that variation in growing season temperature was negatively related to total and graminoid ANPP, but precipitation was a stronger predictor than temperature. Growing season temperature was also a significant parameter in more complex models, but again precipitation was consistently a stronger predictor of ANPP. Surprisingly, neither annual nor growing season SPEI were as strongly related to ANPP as precipitation. We conclude that forecasted warming likely will affect ANPP in these grasslands, but that predicting temperature effects from natural climatic gradients is difficult. This is because, unlike precipitation, warming effects can be positive or negative and moderated by shifts in the C3/C4 ratios of plant communities.