Effect of grazing intensity on net ecosystem CO2exchange
Grazing can shift the balance between vegetation being a carbon source or sink in steppe grasslands (de la Motte et al. 2018; Ondieret al. 2021). Our finding that NEE, ER, and GEP decreased with increasing grazing intensity is consistent with previous results from desert steppe (Jin et al. 2023; Wang et al. 2023). Our finding that heavy grazing had a stronger inhibitory effect on ER than GEP is consistent with the results of Peng et al. (2007). This may have resulted because livestock feeding reduces the aboveground biomass which cannot be compensated by regrowth (Zhang et al. 2018; Zhanget al. 2023), such that the effective amount of leaf area available for both photosynthesis and respiration is reduced so that the net CO2 exchange rate decreases (Oba et al. 2000; Shi et al. 2022).
Although our finding of the positive correlation between productivity and NEE is consistent with many previous studies (Danielewska et al. 2015; Xu et al. 2022), we also found a positive correlation between aboveground biomass and NEE, primarily driven by shrubs and semi-shrubs and perennial forbs, which is consistent with previous work (Zhang et al. 2023). Our finding that shrubs and semi-shrubs and perennial forbs were strongly influenced by grazing, while grasses were less so is consistent with the idea that shrubs and semi-shrubs and perennial forbs are more palatable and have higher nutritional value than grasses, mainly stipa breviflora , which are not preferred by livestock. Shrub roots can reach up to 70 cm deep into the soil layer, allowing them to better utilize deeper water and nutrients to maintain a high carbon fixation capacity and a high net carbon uptake capacity (Niuet al. 2023). which can explain why their loss dramatically influenced NEE.
Our finding that plant N content is negatively correlated with net ecosystem CO2 exchange is inconsistent with previous findings that loss of leaf N attenuates ecosystem carbon cycling (Wang et al. 2014 in Chinese)(Gong et al. 2021), This may be because altered plants allocated more N to non-photosynthetic proteins to increase their compensatory growth in response to grazing, but with reduced photosynthetic capacity (Onoda et al. 2004), resulting in a decrease in net CO2 exchange rate (Zhang et al.2006). A study by Wu et al. (2021) showed that N addition in desert steppe increased the net CO2 exchange rate, while You et al. (2016) showed that high levels of N addition inhibited NEE, but moderate levels promoted NEE. This suggests that the changes of nitrogen absorbed and used by plants are complex and require further investigation (Schimel et al. 2001).
Although NEE decreased in response to increasing grazing intensity, it is of interest that there was no significant difference in NEE rates between the LG and MG treatments in our study (Fig. 4a, bars), because although short-flowered needlegrass was a well-established species and widely distributed in our experimental sample plots, livestock did not prefer it, resulting in no significant difference in vegetation stock and cover between the LG and MGC treatments and the non-grazed areas, so their net CO2 exchange rates were not significantly different from those of ck. The net CO2 exchange rate was not significantly different from that of CK (p > 0.05, Fig. S1a).
Although NEE decreased in response to increasing grazing intensity, we found no difference in NEE rates between the light and moderate grazing treatments. This was likely because less preferred grasses dominated both treatments.