The actual evapotranspiration (ET_a) estimated models based on the complementary relationship (CR) theory have been applied in various climatic conditions around the world. However, in cold regions, the evaluation of the adaptability of the CR models was performed through complete freeze-thaw cycles, and the adaptability during various periods of soil freeze-thaw cycles has not been evaluated separately. Daily ET_a was measured by lysimeters on alpine grassland in the Qilian Mountains from 2010 to 2017, and the measurements were used to evaluate five CR models during the thawing, thawed, freezing, and frozen periods, respectively. The five models comprised the advection-aridity (AA) model of Brutsaert and Stricker, the GG model proposed by Granger and Gray, Morton's CR areal evapotranspiration (CRAE) model, the Han model, and Brutsaert model. The results show that all five CR models were only able to estimate daily ET_a during the thawed period. None of the models could estimate the daily ET_a during the thawing, freezing, or frozen periods. The basic assumptions of the CR may not be suitable for non-thawed periods with complex energy processes, and no complementary behavior was shown in the non-thawed periods. The CR models must be applied with caution during freeze-thaw cycles in cold regions.

基于互补关系（CR）理论的实际蒸散量（ET _a）估计模型已在世界各地的各种气候条件下应用。然而，在寒冷地区，CR模型的适应性评估是通过完整的冻融循环进行的，并且尚未单独评估土壤冻融循环各个时期的适应性。每日ET _a于2010年至2017年期间，使用祁连山高寒草原的蒸渗仪进行了量热法测量，分别用于解冻，解冻，冻结和冷冻期间的五个CR模型。这五个模型包括Brutsaert和Stricker的对流干旱（AA）模型，Granger和Gray提出的GG模型，Morton的CR区域蒸散（CRAE）模型，Han模型和Brutsaert模型。结果表明，所有五个CR模型仅能在解冻期间估算每日ET _a。该模型没有一个能够估计每天ET_一在解冻，冻结或冻结期间。CR的基本假设可能不适用于具有复杂能量过程的非解冻期，并且在非解冻期未显示互补行为。在寒冷地区的冻融循环过程中，必须谨慎使用CR模型。