Yield Loss Functions of Winter Wheat Under Single and Combined Drought–Flood Stress During the Booting–Flowering Stage in the Huaibei Plain

ABSTRACT To establish yield loss functions for winter wheat under different water stress conditions (single drought, single waterlogging, and drought‐flood abrupt alternation) in the Huaibei Plain, controlled cylinder experiments with gradient water treatments were conducted over two consecutive growing seasons (2023–2024 and 2024–2025) during the critical booting‐flowering growth stage across two independent growing seasons, representing replicated experiments conducted at the same phenological phase. The cumulative drought stress degree (SD) was calculated using the average soil moisture content, field capacity (0.28 g·g −1 ), and wilting coefficient (0.093 g·g −1 ) of the test cylinders, with the lower limit of suitable soil moisture content (θ s ) determined as 65% of the field capacity based on long‐term irrigation experiment data in the study area. The cumulative waterlogging stress degree (SFW, cm·day), an intuitive indicator representing the total surface flooding load on crops, was defined as the cumulative value of daily flooding depth during the waterlogging period. S ‐shaped logistic curve functions for wheat yield loss under single drought and single waterlogging stress were constructed, with the determination coefficients ( R 2 ) of 0.90 and 0.87, respectively, showing excellent fitting performance. A multiple linear regression function was further developed to characterize wheat yield loss under drought‐flood abrupt alternation, which incorporated the interaction term of SD and SFW (SD×SFW) to reflect the combined effect of dual stresses. The disaster‐causing point, bursting point (inflection point of the logistic curve), and attenuation point of water stress for maintaining high and stable wheat yield were quantitatively identified under single stress conditions: for drought stress, the three points corresponded to SD values of 3.03, 5.89, and 12.70 (drought durations of 4, 8, and 13 days, respectively); for waterlogging stress (10 cm flooding depth), they were 20.55, 33.76, and 46.97 cm·day (waterlogging durations of 2, 4, and 6 days, respectively). Overall analysis revealed that drought‐flood abrupt alternation during the booting‐flowering stage exerted a compensation effect on wheat yield (reflected by the negative interaction coefficient of −0.0059 in the regression model), meaning the actual yield loss under combined stress was lower than the additive loss from single drought and single waterlogging stress. This study advances the quantitative assessment of compound drought‐flood disaster losses for dryland crops, and provides important technical support for precise regional flood and drought risk evaluation, the formulation of farmland irrigation and drainage project construction standards, and the development of integrated drought‐flood disaster mitigation measures in the Huaibei Plain, thus safeguarding regional food security and promoting sustainable agricultural development.