College of Engineering | Apply to CSU | Disclaimer | Equal Opportunity Statement | Privacy | Search CSU

Impact of Drought on Grassland Productivity Across the Wet-Dry Gradient in the U.S. Great Plains in 2010-2012

You must be on the CSU network—either physically or using VPN—to watch this or any of the videos on this site.

October 18, 2016
Renee Curry
Hosted by Scott Denning (advisor), Melinda Smith (Graduate Degree Program of Ecology), Lori Peek (Department of Sociology)


Severe, prolonged droughts are predicted to occur more frequently due to global
climate change. Since grasslands already grow in regions that are water limited, they are particularly vulnerable to changes in precipitation. Climate models are used to investigate how grasslands will respond to climate change; however, current land surface models have difficulty in simulating grasslands and their response to drought. The main objective of this research project was to investigate the dominant relationships between grassland productivity and precipitation and to see if this behavior could be predicted in a model. To do this, we focused on both climate (dry to wet gradient) and drought (climate anomalies) using a combination of data and the Simple Biosphere Model Version 4 (SiB4). In order to have a better understanding of the relationship between grassland productivity and precipitation on a regional scale, this research studies nine sites across the U.S. Great Plains over which there is a significant precipitation gradient. In addition to the climatic gradient in precipitation, we took advantage of a natural experiment from 2010 through 2012, during which a significant drought occurred in this region.

Observed west-to-east gradients in grassland productivity were generally well-captured by the model: there was an increase in leaf area index (LAI) with increasing precipitation, with a nearly identical linear relationship in both the observations and the model. SiB4 overestimated the magnitude of the seasonal-mean LAI; however, this bias was constant across the precipitation gradient. The drought decreased grassland productivity: both the observations and the model showed reduced LAI and a shorter growing season due to drought, and an analysis of the standardized anomalies in LAI and precipitation demonstrated that both the observations and the model have a nearly identical linear response to drought (difference in slope < 10%). Although SiB4 has a bias in the magnitude of seasonal-mean LAI, it has the same precipitation responses as seen in the data, thus showing the ability to capture the behavior of grasslands both across a dry-wet gradient and for a specific drought event.