Evaluation of optimal irrigation scheduling and groundwater recharge at representative sites in the North China Plain with SWAP model and field experiments

•The SWAP model was established in representative sites in the North China Plain.•Optimal irrigation schedulings of wheat and maize were compared among the sites.•Groundwater recharge changes under optimal irrigation were compared among the sites.•Precipitation, soil and water table effects on groundwater recharge were clarified.

The application of water saving irrigation is essential to alleviate the rapid depletion of groundwater resources in water crisis areas such as the North China Plain (NCP). The integrated effects of climate, soil and groundwater conditions on water balance and crop response need to be further studied. It is imperative to investigate spatial differences of irrigation scheduling and groundwater recharge within the NCP for regional water management. In this study, three representative sites including Luancheng (LC), Tongzhou (TZ) and Yucheng (YC) with different precipitation, soil and water table conditions were selected. The SWAP model was established at each site to compare optimal irrigation scheduling and groundwater recharge of the winter wheat-summer maize double cropping system under different hydrological years. Largest optimal irrigation amount and additional irrigation at the winter-dormancy stage of wheat were required for the TZ site in each hydrological year. The optimal irrigation scheduling was almost same at the three sites for summer maize, except that no irrigation was needed at filling stage in dry year at the YC site. The average annual groundwater recharge under optimal irrigation was in the order of YC (108.4 mm), TZ (63.8 mm) and LC (0.4 mm). Furthermore, the groundwater recharge occurred throughout the double cropping seasons in YC, but it occurred only in summer maize season at the LC and TZ sites. The net irrigation amount of the optimal irrigation scheduling was reduced by 51.2%, 24.9% and 77.2% compared to reference practice at the LC, TZ and YC sites, respectively. The amount of groundwater recharge depended on the local precipitation and irrigation, while water table depth and soil texture mainly influenced the delay time of groundwater recharge relative to the water input events.