Sunken lanes or ‘road gullies’ are a common geomorphic feature in the European loess belt. These landforms are usually created by various processes over time (water erosion, mass movements and traffic). They are important for ecology and biodiversity, for recreation and transport, as well as from a hydrological and geomorphological perspective (e.g. by enhancing the runoff and sediment connectivity of the landscape). Nonetheless, little is known about their origin and evolution. The formation of sunken lanes is typically only understood in qualitative terms, while no studies so far have provided a quantitative analysis of their initiation conditions. To address this research gap and to improve our geomorphic understanding of this landform, we investigated to what extent sunken lanes can be treated and described as a gully erosion process. More specifically, we explored to what extent their position in the landscape and their characteristics can be explained by the topographic slope-area (SA) threshold that typically controls gully initiation: $S=k{A}^b$, with k and b empirical constants. Based on field surveys and LIDAR data analyses, we determined the slope gradient of the soil surface (S) and contributing area (A) for 132 representative sunken lane heads in the Belgian loess belt and collected data on several other characteristics (position of initiation and sedimentation points, width, depth and length of each sunken lane). Our results show a very large scatter on the S and A values. Moreover, the morphological characteristics of sunken lanes (e.g. width and length) showed no clear relation with these values. However, a very weak SA-threshold relation with a low b-exponent (−0.016) and a low k-coefficient (0.022) could be identified. Overall, our results show that the formation of sunken lanes is not in line with the topographic threshold conditions that govern the initiation of gullies across different environments. The low k-coefficient and the lack of a strong correlation between initiation slope and contributing area point towards an overall higher susceptibility to the formation of sunken lanes than the formation of gullies. This supports the conclusion that not only concentrated flow detachment and transport, but especially human activities (mainly traffic erosion) play a major role in the initiation and development of sunken lanes.

下陷的车道或“沟壑”是欧洲黄土带的常见地貌特征。这些地貌通常是随着时间的流逝（水蚀，群众运动和交通）的各种过程而形成的。它们对于生态和生物多样性，娱乐和运输以及从水文和地貌学的角度（例如，通过增强景观的径流和泥沙连通性）而言都是重要的。但是，关于它们的起源和进化知之甚少。下沉的车道的形成通常只能从定性的角度进行理解，而到目前为止，尚无研究对其起始条件进行定量分析。为了解决这一研究空白并提高我们对这种地形的地貌认识，我们调查了在何种程度上可以将下陷的车道视为沟蚀过程。$\mathrm{小号}=ķ{\mathrm{一种}}^b$，具有k和b的经验常数。基于实地调查和LIDAR数据分析，我们确定了比利时黄土带中132个典型下沉车道头的土壤表面（S）和贡献面积（A）的坡度，并收集了其他几个特征数据（起爆位置和起伏位置）每个沉没泳道的沉降点，宽度，深度和长度）。我们的结果表明，S和A值存在很大的分散。此外，下沉的车道的形态特征（例如宽度和长度）与这些值没有明确的关系。但是，可以识别出非常弱的SA阈值关系和低b指数（-0.016）和低k系数（0.022）。总体，我们的结果表明，下沉车道的形成与在不同环境中控制沟渠形成的地形阈值条件不一致。较低的k系数以及起始坡度与贡献面积之间缺乏强相关性，这意味着对下陷车道形成的总体敏感性要比对沟渠的形成更高。这支持这样的结论：不仅集中的流动分离和运输，而且特别是人类活动（主要是交通侵蚀）在下沉车道的形成和发展中起着重要作用。