Résumé
The runoff infiltration partitioning has direct consequences on preservation of water resources in rural territories, both in agricultural plots and uncultivated areas (e.g. ditches, channels, grass strips), and requires a better understanding of the variability of soil infiltration capacity by disentangling the complex links between soil, vegetation and management. The general objective of the study was to investigate the temporal variation in quasi-steady ponded infiltration rates of a fluvisol soil under bare and different cover crop type (a Malvaceae with a tap-root system and a Poaceae with a fibrous root system) and management conditions (burning, mowing, and chemical weeding) that are commonly found in the Mediterranean vineyards. A modified double-ring infiltration method was used to repeat measurement of the quasi-steady ponded infiltration, fcp, on the same location over time. Placed on a 64 m2 plot area with minimal distances between individual measurements of 30 cm, the setup allowed evaluation of variability among measurements that were conducted within the plot. The results showed a significantly lower fcp for bare soil than covered soil, and a two-fold higher fcp for soil covered by Malvaceae than Poaceae. A seasonal effect in fcp was observed, with the highest fcp in summer and the lowest in winter. The study revealed a strong spatial variability in fcp along a transect of a few tens of centimeter, and showed no significant effect of management strategies compared to the vegetated control. The results revealed the importance of considering both plant traits and season rather than vegetation management strategies to explain quasi-steady ponded infiltration rates.