4.2. Legacy effects of DWC on soil carbon mineralization
Both 80-40% and 100-20% WHC treatments increased C mineralization in OF soil but decreased it in BF and BF+ soils compared to the mean constant-moisture control treatment during the extended period (Fig. 4b). This result indicated that the repeated DWC could cause a legacy effect on soil C mineralization, and this legacy effect can last for at least 38 d (including 10-d stabilization period and 28-d extended period). Butterly et al. (2011) implied that this legacy effect can last for more than 35 d, and Li et al. (2018) believed that after 4 times of DWC, this legacy effect can last for at least 60 d.
Compared to the mean constant-moisture treatment (60% WHC), the MBC content at the start of the extended period (100 d) was higher in the 100-20% WHC treatment for OF soil (Fig. 5a), while was lower in both 80-40% and 100-20% WHC treatments for BF and BF+ soils (Fig. 5a). Moreover, there was a positive correlation between MBC and cumulative C mineralization during the extended period across the three soils (Fig. 6a). These results suggested that the changes in microbial biomass affected the legacy effect of DWC on soil C mineralization. Additionally, DOC was also positively correlated with cumulative C mineralization during the extended period across the three soils (Fig. 6b), indicating that the amount of microbial biomass and labile substrates (i.e. DOC) both contributed to the legacy effect of DWC on soil C mineralization. Notably, the fact that both MBC and DOC were positively correlated with cumulative C mineralization during the extended period is likely due to the positive linear relationship between MBC and DOC (Fig. S4). Moreover, the DWC-driven changes of soil microbial community composition may also contribute to this legacy during the extended period (Meisner et al., 2021). Together, these results are consistent with the finding of Li et al. (2018) who showed that the legacy effect during the extended period is attributed to the amount of plant litter (as substrate) and microbial biomass.
Taking together, both mild and strong DWC treatments had a minor effect on SOC decomposition in OF and BF+ soils, respectively, but decreased SOC decomposition in BF soil by 12% and 14%, respectively (Fig. 4c). However, it should be noted that these results need to be interpreted with caution, because we did not measure the C release during the stabilization period (90-100 d) and the extended period (100-128 d, 28 days) may not be long enough. Future studies can use a longer extended period and measure microbial community composition for a more mechanistic understanding of the legacy effect of DWC on soil C release.