Abstract.Fire can considerably change hydrological processes, increasing the risk of extreme flooding and erosion events. Although hydrological processes are largely affected by scale, catchment-scale studies on the hydrological impact of fire in Europe are scarce, and nested approaches are rarely used. We performed a catchment-scale experimental fire to improve insight into the drivers of fire impact on hydrology. In north-central Portugal, rainfall, canopy interception, streamflow and soil moisture were monitored in small shrub-covered paired catchments pre-and post-fire. The shrub cover was medium dense to dense (44 to 84 %) and pre-fire canopy interception was on average 48.7 % of total rainfall. Fire increased streamflow volumes 1.6 times more than predicted, resulting in increased runoff coefficients and changed rainfall-streamflow relationships -although the increase in streamflow per unit rainfall was only significant at the subcatchment-scale. Fire also fastened the response of topsoil moisture to rainfall from 2.7 to 2.1 h (p = 0.058), and caused more rapid drying of topsoils after rain events. Since soil physical changes due to fire were not apparent, we suggest that changes resulting from vegetation removal played an important role in increasing streamflow after fire. Results stress that fire impact on hydrology is largely affected by scale, highlight the hydrological impact of fire on small scales, and emphasize the risk of overestimating fire impact when upscaling plot-scale studies to the catchmentscale. Finally, they increase understanding of the processes contributing to post-fire flooding and erosion events.

No abstract

No abstract

The present paper reports the analysis of a diary study of a German-speaking child and reveals interrelations between semantic and phonological development: some of the subject's overextensions were explained by a deliberate avoidance of a form and the use of an easily pronounecable substitute rather than by erroneous attempts to map linguistic and non-linguistic knowledge. These results suggest that there may be phonological as well as lexical or conceptual reasons for children to (apparently) overextend the meaning of a word. The findings emphasize the danger of a modular analysis. Different linguistic levels should not be seen individually and independent of one another.

No abstract