Soil-crop simulation models are widely used to assess the impacts of soil management and climate change on soil water balance, solute transport and crop production. In this context, it is important that hydrological processes in the soil-crop system are accurately modelled. We suggest here that empirical treatments of soil water flow, water uptake by plant roots and transpiration limit the applicability of crop models and increase prediction errors. We further argue that this empiricism is to a large extent unnecessary, as parsimonious physics-based descriptions of these water flow processes in the soil-crop system are now available. Recent reviews and opinion articles, whilst strongly advocating the need for improvements to crop models, fail to mention the significant role played by accurate treatments of soil hydrology. It seems to us that empirical models of soil water flow have become the elephant in the room.

土壤作物模拟模型被广泛用于评估土壤管理和气候变化对土壤水分平衡、溶质迁移和作物生产的影响。在这种情况下，准确模拟土壤作物系统中的水文过程非常重要。我们在此建议，对土壤水流、植物根系吸水和蒸腾的经验处理限制了作物模型的适用性并增加了预测误差。我们进一步认为，这种经验主义在很大程度上是不必要的，因为现在可以对土壤作物系统中的这些水流过程进行基于物理学的简明描述。最近的评论和意见文章，虽然强烈主张需要改进作物模型，但没有提到准确处理土壤水文所发挥的重要作用。