Differential roles of perennial and annual grasses in semi-arid grassland response to precipitation repackaging: Insights from experiment-informed Community Land Model (CLM5.0) parameterization

Authors: HU, TIANYI - University Of Arizona; ZENG, XUBIN - University Of Arizona; Biederman, Joel; SMITH, WILLIAM - University Of Arizona; Scott, Russell; ZHANG, FANGYUE - University Of Arizona; SONG, YANG - University Of Arizona

Submitted to: Agricultural and Forest Meteorology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/10/2026
Publication Date: 2/21/2026
Citation: Hu, T., Zeng, X., Biederman, J.A., Smith, W., Scott, R.L., Zhang, F., Song, Y. 2026. Differential roles of perennial and annual grasses in semi-arid grassland response to precipitation repackaging: Insights from experiment-informed Community Land Model (CLM5.0) parameterization. Agricultural and Forest Meteorology. 380; 111079. https://doi.org/10.1016/j.agrformet.2026.111079.
DOI: https://doi.org/10.1016/j.agrformet.2026.111079

Interpretive Summary: In many dryland agroecosystems of the western US, rainfall has become less frequent but more intense since the 1970s. We lack information about how this will affect the health and productivity of dryland farms, forests, and rangelands. We operate a rainfall manipulation experiment to measure how grassland soils and plants respond to rainfall arriving in fewer, larger storms. Here, we describe our work to apply those data to improve the Community Land Model, with the aim of being better able to predict and describe how drylands respond to altered rainfall patterns. We found that fewer, larger storms tend to infiltrate the soil more deeply with water. Importantly, we structured the model to separately handle deeper-rooted perennial grasses and shallow-rooted annual grasses. This allowed the model to more accurately represent how grasslands with different plant communities either can or cannot take advantage of the deeper soil moisture from fewer, larger rainfalls. The improved model demonstrated that if a grassland has sufficient deep-rooted perennial grasses, then fewer, larger rainfalls can be expected to improve the water use efficiency of the grassland increasing forage production.

Technical Abstract: Precipitation in dryland ecosystems is becoming less frequent but more intense, and this trend is expected to continue. The positive, negative, and neutral impacts of precipitation temporal repackaging on dryland productivity have been observed, and the underlying mechanisms remain unclear. We combined observations from a precipitation manipulation experiment in a semi-arid grassland to improve the Community Land Model (CLM5.0)’s simulation of coupled soil water-vegetative growth responses to changing hydroclimate conditions. Parameterizing the distinct phenological, photosynthetic, and structural traits of annual and perennial C3/C4 grasses significantly improved the simulation of this grassland productivity and evapotranspiration under varying precipitation patterns. Incorporating soil hydraulic parameters fitted to observations notably reduced the overestimation of soil water in the Arizona semi-arid grassland. The improved model enables us to interpret the mechanisms that determine the direction and magnitude of semi-arid grassland productivity responses to changing hydroclimates, highlighting that the coexistence of diverse annual and perennial grasses with distinct functional traits and complementary water uptake behaviors, combined with coarse, highly permeable soils in the semi-arid grassland enhances the plant community’s uptake of soil water across time and depths, thereby increasing the resilience of this semi-arid grassland to more extreme hydroclimate.