Contrasting responses of heterotrophic and autotrophic respiration to experimental warming in a winter annual-dominated prairie.

Understanding how soil respiration (Rs) and its source components respond to climate warming is crucial to improve model prediction of climate-carbon (C) feedback. We conducted a manipulation experiment by warming and clipping in a prairie dominated by invasive winter annual Bromus japonicas in Southern Great Plains, USA. Infrared radiators were used to simulate climate warming by 3 °C and clipping was used to mimic yearly hay mowing. Heterotrophic respiration (Rh) was measured inside deep collars (70 cm deep) that excluded root growth, while total soil respiration (Rs) was measured inside surface collars (2-3 cm deep). Autotrophic respiration (Ra) was calculated by subtracting Rh from Rs. During 3 years of experiment from January 2010 to December 2012, warming had no significant effect on Rs. The neutral response of Rs to warming was due to compensatory effects of warming on Rh and Ra. Warming significantly (P < 0.05) stimulated Rh but decreased Ra. Clipping only marginally (P < 0.1) increased Ra in 2010 but had no effect on Rh. There were no significant interactive effects of warming and clipping on Rs or its components. Warming stimulated annual Rh by 22.0%, but decreased annual Ra by 29.0% across the 3 years. The decreased Ra was primarily associated with the warming-induced decline of the winter annual productivity. Across the 3 years, warming increased Rh/Rs by 29.1% but clipping did not affect Rh/Rs. Our study highlights that climate warming may have contrasting effects on Rh and Ra in association with responses of plant productivity to warming.