Although concern for ephemeral gully (EG) erosion has been growing within the research community, knowledge gaps still need to be identified for a state-of-the-art review. This review is based on an extensive search of published literature in the Web of Science database that specifically addressed EG erosion. This review included 173 papers published in 56 scientific journals over the last 30 years. The results indicated that the main EG erosion topics were related to gully erosion processes, factors, methodology, models, and the detrimental effects of EGs. The definition of EG was identical worldwide. There is an urgency to define critical morphological values under different conditions for identifying EG channel locations in fields in order to map and model their development quickly and accurately. Topographic factors have been much more studied than other contributing factors, including climate, land management, soil, and hydraulics. Fifteen different methodological approaches or technologies, which were categorized into four kinds of methods, including field monitoring, modelling, simulation experiments, and review, have been employed to assess and quantify EG erosion. The average soil erosion rate of EGs in 36 publications increased and decreased with increase of slope and catchment areas, respectively, and the number of EGs linearly increased with catchment area. The complexity of the development of EG channels and the driving force mechanisms should also be studied intentionally, e.g., the quantification of soil detachment and sediment transport mechanisms for both surface and subsurface flow during different sub-processes, including gully head advance, sidewall collapse and bed incision. It was concluded that the interactive effects of different contributing factors on EG erosion across various spatio-temporal scales should be further researched. The comprehensive utilization or application of new methods and technologies, e.g., automatic monitoring technologies, artificial intelligence, machine learning algorithms, would be helpful to improve study efficiency and accuracy. The scientific basis of prediction models still needs further development and refinement. This review provides recommendations and suggestions to improve knowledge, prediction, and control of EG erosion.

尽管研究界对临时性沟壑（EG）侵蚀的关注日益增长，但仍需找出知识差距以进行最新的综述。这篇评论是基于对Web of Science数据库中已发表文献的广泛搜索，这些文献专门针对EG侵蚀。这篇综述包括了过去30年在56种科学期刊上发表的173篇论文。结果表明，主要的EG侵蚀主题与沟蚀过程，因素，方法，模型以及EG的有害影响有关。EG的定义在世界范围内是相同的。迫切需要在不同条件下定义关键的形态学值，以识别油田中的EG通道位置，以便快速准确地绘制地图并对其发展进行建模。与其他因素（包括气候，土地管理，土壤和水力）相比，对地形因素的研究更多。十五种不同的方法学方法或技术已被分类为四种方法，包括现场监测，建模，模拟实验和回顾，用于评估和量化EG侵蚀。36种出版物中的EG的平均土壤侵蚀率分别随坡度和集水面积的增加而增加和减少，并且EG的数量随集水面积的增加而线性增加。还应有目的地研究EG通道开发的复杂性和驱动力机制，例如，量化不同子过程中地表和地下流的土壤离析和沉积物迁移机制的量化，包括沟壑前移，侧壁塌陷和床切口。得出的结论是，应进一步研究不同时空尺度上不同促成因素对EG侵蚀的交互作用。自动监控技术，人工智能，机器学习算法等新方法和新技术的综合利用或应用，将有助于提高学习效率和准确性。预测模型的科学基础仍然需要进一步发展和完善。这篇综述提供了建议和建议，以提高对EG侵蚀的认识，预测和控制。得出的结论是，应进一步研究不同时空尺度上不同促成因素对EG侵蚀的交互作用。自动监控技术，人工智能，机器学习算法等新方法和新技术的综合利用或应用，将有助于提高学习效率和准确性。预测模型的科学基础仍然需要进一步发展和完善。这篇综述提供了建议和建议，以提高对EG侵蚀的认识，预测和控制。得出的结论是，应进一步研究不同时空尺度上不同促成因素对EG侵蚀的交互作用。自动监控技术，人工智能，机器学习算法等新方法和新技术的综合利用或应用，将有助于提高学习效率和准确性。预测模型的科学基础仍然需要进一步发展和完善。这篇综述提供了建议和建议，以提高对EG侵蚀的认识，预测和控制。预测模型的科学基础仍然需要进一步发展和完善。这篇综述提供了建议和建议，以提高对EG侵蚀的认识，预测和控制。预测模型的科学基础仍然需要进一步发展和完善。这篇综述提供了建议和建议，以提高对EG侵蚀的认识，预测和控制。