Biogeochemical parameterizations of Pennsylvania dairy pasture – implications for ecosystem modeling under climate change

Background/Question/Methods

Pasture can be either a source or a sink for atmospheric carbon dioxide, depending primarily upon climate and land management. The primary data source for pasture parameterizations (i.e. a process for configuring parameters for ecosystem process models) is from grasslands in the Great Plains, where climate is more arid and less management-intensive than in the Eastern U.S. Data to model carbon dynamics under managed dairy grasslands in the humid eastern U.S. remains inadequate. This paper aims to provide TEM-Hydro – a terrestrial ecosystems model – with region-specific dairy grassland parameterizations in Pennsylvania (PA) to more realistically report humid grassland dynamics under a changing climate.  Specifically, this study provides: 1) a comprehensive and consistent calibration dataset for Pennsylvania dairy grassland, dominated by C3-orchardgrass (Dactylis glomerata L.); 2) an evaluation of ecosystem sensitivity to climate change by comparing to a semi-arid, C4 grassland-based parameterization under Pennsylvania climate; and finally we highlight the 3) importance of correct parameterization in determining proper sensitivity to climate change.

Results/Conclusions

Based on PA parameterization, our results indicate that the increase in temperature and atmospheric CO₂ concentration did not significantly affect Gross Primary Productivity (GPP) and Net Ecosystem Productivity (NEP) for Pennsylvania pasture over the 20th century, possibly due to nitrogen saturation induced by cattle fertilization.  Both GPP and NEP start to increase steadily in the 21st century, under both SRES A2 and B1 scenarios. Overall, the sensitivity of PA pasture to climate change is lower when using the xeric parameterization. Most importantly, using realistic parameterization reveals that PA dairy pastures act as a strong carbon sink under climate change (positive NEP), but soil carbon density remains low, consistent with the effect of grazing. A misleading result would emerge if the xeric parameterization was applied to PA pasture, where NEP remains positive with a high soil carbon density. This study therefore shows the importance of using the correct parameterization within a terrestrial biogeochemical model when determining sensitivity to future climate change.