Assessing the role of pre-fire ground preparation operations and soil water repellency in post-fire runoff and inter-rill erosion by repeated rainfall simulation experiments in Portuguese eucalypt plantations

This study addressed the impacts of contrasting pre-fire ground preparation operations on post-fire runoff and inter-rill erosion in six eucalypt plantations in north-central Portugal, with a special emphasis on the role of soil water repellency in the seasonal patterns of overland flow generation. To this end, a down slope ploughed, a contour ploughed and a terraced site were compared with three unploughed sites. Runoff and erosion data were collected in the field by carrying out rainfall simulation experiments (RSEs) with two intensities (45–50 and 80–85 mm h− 1) at six occasions during the first one to two years following wildfires in 2005 and 2006.Overall runoff coefficients varied markedly amongst the six study sites and between the two intensities (7 to 55%). While runoff figures were comparable to those of prior RSE studies in recently burnt areas, overall sediment losses were comparatively low (7–155 g m− 2) but contained a substantial organic matter fraction (29–74%). Apparently, the inter-rill erosion rates were essentially sediment-limited, fitting in with the long history of intensive land use that is typical in the Mediterranean Basin. The hydrological and erosion impacts of the three pre-fire ground preparation operations were minor, probably because these operations took place several years before the latest wildfire. Overall, the two rainfall intensities produced the expected differences but this effect was only statistically significant for simultaneous RSEs. Furthermore, the effect of rainfall intensity varied markedly between the study sites and, occasionally, between the two plot pairs at the same site. This impeded an erosion risk ranking of the six study sites that was consistent for both rainfall intensities. Runoff and erosion rates did not decrease in a simple or pronounced manner with time-since-fire. These temporal patterns could in part be attributed to changes in soil water repellency but other factors were involved as well. Removal of the protective soil cover by litter in particular appeared to play a key role in the increase in sediment losses following logging and wood extraction.

► We used rainfall simulations to asses post-fire runoff and inter-rill erosion.
► We compared different pre-fire ploughing techniques and two rainfall intensities.
► Runoff amounts were similar as reported by others but sediment losses were low.
► An intensive land use and moderated fire severity could be originated in that result.
► A pattern between different pre-fire practices was not found.