Two decades of irrigation in a water-limited grassland: Above and belowground C responses

Background/Question/Methods

Climate change impacts will include chronic alterations of average precipitation amounts in most terrestrial ecosystems, which will have significant impacts on production and associated ecosystem services (e.g., forage production, carbon sequestration). Although much research has focused on aboveground plant responses to these forecasted changes, relatively few studies have focused on belowground production patterns. In tallgrass prairie where annual burning is common practice, fire has been shown to increase production and thus C volatilization, which would normally reduce amounts of soil C except for fire-driven increases in root allocation. However, chronic changes in precipitation amounts may alter biomass allocation patterns potentially resulting in the depletion of soil C under annual burn regimes. In 2013, we extensively sampled various biotic and abiotic responses to chronically increased precipitation within an existing long-term (23 year) irrigation experiment – established in an annually burned watershed at the Konza Prairie Biological Station in Manhattan, KS – to address the following questions: (1) Do belowground production responses to long-term chronic increases in precipitation mirror those aboveground? And (2) how do biotic and abiotic responses to these new precipitation regimes alter long-term carbon pools under annual burning practices? 

Results/Conclusions

Leading up to our 2013 sampling effort, irrigated aboveground net primary productivity (ANPP) was significantly higher in irrigated vs. ambient plots in 16 of the 22 years (1991-2012), and in 1996 the plant community shifted to a more mesic assemblage in irrigated plots. In 2013, we found no differences between ANPP in irrigated vs. non-irrigated plots despite a ca. 275 mm increase in precipitation accompanied by a significant increase in soil moisture during the growing season. Contrastingly, we found a significant reduction in belowground net primary productivity (BNPP) in irrigated plots corresponding with a reduction in root:shoot relative to ambient. Despite this, we detected no differences in standing crop root biomass, live root C:N, or soil available nitrogen. Additionally, no change in soil C has been detected in irrigated plots throughout the experiment. The 2013 BNPP reduction under irrigation may be due to allocation differences in irrigated communities combined with legacy effects of an extremely low production year in 2012. Based on these results, we suggest that despite decreased root:shoot under chronic irrigation causing increased C volatilization via annual burning, soil C pools in this system are sufficiently large to provide surprising resistance to proportional ecosystem C loss.