Different responses of denitrification rates and denitrifying bacterial communities to hydrologic pulsing in created wetlands

Hydrologic pulsing, including water level drawdown and subsequent flooding, may have a considerable impact on both biogeochemical processes and microbial communities in wetlands. Since denitrifying bacteria play a key role in water quality improvement in wetlands, changes in their activities and communities with hydrologic pulsing are an important issue. We investigated the responses of in situ denitrification rates, denitrifying bacterial community structure and their quantities using nitrite reductase (nir) S gene under different hydrological pulsing conditions in created wetlands in central Ohio USA. Average denitrification rates, measured from 4 different sampling locations, were 302, 133, 71 and 271 μg N2O–N m−2 h−1 during inundated, saturated, drying and reflooding periods, respectively. In particular, the denitrification rates in shallow water level marsh areas (SM) followed by deepwater level marsh areas (DM) showed more sensitivity and magnitude of changes to hydrologic pulsing events than did non-vegetated deepwater areas. This may have been due to the high aerobic decomposition during the drying period and nutrient flushing in shallower marsh areas after the reflooding event. In contrast, the community structure and diversity of denitrifiers based on terminal-restricted fragment length polymorphism (T-RFLP) analysis showed no significant change due to hydrologic pulsing. Instead, the presence and absence of vegetation altered denitrifying bacterial community structure. The nirS gene copy number remained relatively constant with only minor increases during water level drawdown followed by a significant decrease when a sudden reflooding event occurred. These results indicate that environmental disturbances, such as hydrologic pulsing, have a major impact on the denitrification process, but less impact on the community structures of the denitrifying bacteria. In addition, there was no relationship among the denitrification rate, the community structure, and the quantity of denitrifiers, suggesting that changes in denitrification rates during hydrologic pulsing events were not caused by the changes in microbial community structure but more by physicochemical factors, such as substrate availability and hydrology.