Is frequency or amount of precipitation more important in controlling CO2 fluxes in the 30-year-old fenced and the moderately grazed temperate steppe?

• DNDC modeling qualified effects of altered precipitation patterns on fenced and grazed steppe.
• Ecologically effective rainfall controls ecosystem CO2 sink/source functions.
• Altered rainfall patterns had no significant effect on CO2 fluxes and plant productivity.

Climate change is causing measurable change in rainfall patterns with uncertain implications for key processes in ecosystem function and carbon cycling. We conducted a modeling analysis to identify how variation in amount and frequency of precipitation affected on CO2 fluxes and net primary productivity. The denitrification-decomposition model was used to quantify the effects of altered precipitation quantity and frequency under 12 climate scenarios over a period of 30-years on both a fenced and a moderately grazed temperate steppe in Inner Mongolia, China. The modeling results show the 12 climate scenarios had an obvious effect on gross primary productivity (GPP), ecosystem respiration (Re) and net primary productivity (NPP) in both a fenced and a grazed site. GPP, Re and NPP increased in both sites under increased precipitation scenarios called A3 (+20% precipitation) and A4 (+40%) when compared with baseline conditions, while GPP, Re and NPP declined under decreased precipitation scenarios A1 (−40% precipitation) and A2 (−20%) scenarios. The changed rainfall frequency resulted in a decline in GPP, Re and NPP compared with those parameters under the base conditions at both sites. The ecologically effective rainfall (ER), not total rainfall, controls the ecosystem CO2 sink/source function. When ER exceeded 318 mm yr−1 in the fenced site and 224 mm yr−1 in the grazed site, the steppe switched from CO2 emission to CO2 absorption. CO2 fluxes in the typical steppe which was fenced and moderately grazed are relatively responsive to changes in the amount of rainfall. However, in terms of the long-term modeling analysis, the modeled results suggest the effects of altered rainfall quantity, frequency and the interaction of rainfall quantity and frequency on CO2 fluxes and plant productivity had no significant difference because of the fluctuating interannual biotic and abiotic factors.