Predicting gully width and widening rates from upstream contribution area and rainfall: A case study in SW Spain

Abstract

Gully width (W) is usually characterized by having a power relation with discharge. However, calculating runoff discharge in ungauged basins, where these gullies typically form, is inherently difficult and associated with large uncertainties. This paper examines a simplified approach to relating gully width directly to the runoff contributing area (A) in a cultivated area in SW Spain. The effect of rainfall and land use on this W-A relation was also analysed. Gully width was measured in 46 sections through the analysis of 10 historical orthophotos between 1956 and 2013, in a 21 km2 catchment in S Spain. These were validated by field measurements in 2013.Widening rates varied strongly over time, at between ~0 and 2,19 m year−1. No significant differences in the gully widening rates were found between the two land uses present in the study zone (olive groves and herbaceous crops). The data obtained show a significant power relation of the form W = αAβ for all time periods analysed, except for 1980, when many gullies were filled in artificially due to a change in land use in the study area. The power relation coefficient (β) varied at between 0.2 and 0.3. A good correlation was obtained between the number of days with daily rainfall of above 13 mm and the increase in the α coefficient over time. The results of this study on gully width dynamics provide new insights into improving the estimation of gully volume, for example in combination with gully headcut retreat models, and including the effects of different climate scenarios.