Since the late summer of 2005, the EROSFIRE project is investigating the hydrological and soil er... more Since the late summer of 2005, the EROSFIRE project is investigating the hydrological and soil erosion response in recently burnt eucalypt plantations. The study areas are located in the municipalities of Águeda, Albergaria-a-Velha and Sever de Vouga, at reasonably close distances from the project's home base at the University of Aveiro, which has allowed carrying out intensive field campaigns and, thereby, the collection of a rather extensive data set. The present paper will, however, only present initial results that, first and foremost, serve to illustrate the approach that is being applied and tested by the project. Field rainfall simulation experiments (RSE's) constitute the principal technique that is being employed by the EROSFIRE project for measuring soil erosion. Erosion plots are utilized as well but primarily to validate the results of the RSE's, in the sense of comparing erosion rates under artificial versus natural rainfall events as well as for micro-plots versus slope-scale plots. The various erosion measurements are ultimately intended to evaluate the suitability of a selection of erosion models for post-fire conditions. Results are presented of 8 RSE's that were carried out two to three months after a forest wildfire. The amounts of simulated rain that do not infiltrate into the soil, are substantial. The resulting overland flow, however, does not produce as high sediment yields as to be expected on the basis the RSE results of earlier studies in recently burnt pine stands in the same region.
The EROSFIRE project selected field rainfall simulation experiments (RSE's) as a time-and cost-ef... more The EROSFIRE project selected field rainfall simulation experiments (RSE's) as a time-and cost-effective approach to gather runoff and soil loss data under the rapidly changing conditions typical for commercial eucalypt plantations in Portugal following wildfires. These RSE data were first and foremost envisaged for initial parameterization and calibration of a physically-based model like MEFIDIS. Subsequent model assessment at larger spatial scales would then be based on slope-scale erosion plot data collected at a small number of selected sites. The present work, however, will in principle be confined to an analysis of the measurement results of the RSE's in two recently burnt eucalypt plantations. Repeat RSE's were carried out in two adjacent but contrasting eucalypt stands on steep hillslopes in north-central Portugal. This involved six occasions ranging from 3 to 24 months after a moderate severity fire in July 2005. A paired-plot experimental design was employed that comprised two pairs of RSE's at each site and occasion. From a grand total of 46 RSE's: (i) 24 and 22 RSE's involved application rates of 45-50 and 80-85 mm h-1, respectively; (ii) 22 took place in a stand that had been ploughed in downslope direction several years before the wildfire and 24 in an unploughed stand. The results showed a clear tendency for extreme-intensity RSE's to produce higher runoff amounts and greater soil and organic matter losses than the simultaneous high-intensity RSE's on the neighbouring plots. Nonetheless, there were marked exceptions, both in space (for one of the plot pairs) and time (under intermediate soil water repellency conditions). Also, overland flow generation and erosion varied significantly along the various field campaigns. This temporal pattern noticeably differed from a straightforward decline with time-after-fire, rather suggesting a seasonal component that reflected broad variations in topsoil water repellency. The ploughed site produced less runoff and erosion than the unploughed site, contrary to what would be expected if the downslope ploughing had occurred after the wildfire instead of several years before it. Finally, sediment losses at both study sites were noticeably lower than those reported by the other studies involving repeat RSE's, i.e. in Australia and western Spain. This possibly reflects a history of intensive land use in the study region, including in more recent times after the widespread introduction of eucalypt plantations.
Wildfires can have important impacts on hydrological and soil erosion processes, due to the destr... more Wildfires can have important impacts on hydrological and soil erosion processes, due to the destruction of vegetation cover and changes to soil properties. According to Shakesby and Doerr (2006), these wildfire effects are: i) much better known at small spatial scales (especially erosion plots) than at the scale of catchments; ii) much better studied with respect to overland flow and streamflow (and, then, especially peak discharges) than to soil erosion. Following up on a precursor project studying runoff generation and the associated soil losses from micro-plot to slope-scale in Portuguese eucalypt forests, the EROSFIRE-II project addresses the connectivity of these processes across hillslopes as well as within the channel network. This is done in the Colmeal study area in central Portugal, where the outlet of an entirely burnt catchment of roughly 10 ha was instrumented with a gauging station continuously recording water level and tubidity, and five slopes were each equipped with 4 runoff plots of < 0,5 m2 ("micro-plot") and 4 slope-scale plots as well as 1 slope-scale sediment fence. Starting one month after the August 2008 wildfire, the plots were monitored at 1- to 2-weekly intervals, depending on the occurrence of rainfall. The gauging station became operational at the end of November 2008, since the in-situ construction of an H-flume required several weeks. A preliminary analysis of the data collected till the end of 2008, focusing on two slopes with contrasting slope lengths as well as the gauging station: revealed clear differences in runoff and erosion between: (i) the micro-plot and slope-scale plots on the same hillslope; (ii) the two slopes; (iii) an initial dry period and a subsequent much wetter period; (iv) the slopes and the catchment-scale, also depending on the sampling period. These results suggest that the different processes govern the hydrological and erosion response at different spatial scales as well as for different periods, with soil water repellency playing a role during the initial post-fire period. The current presentation will review these preliminary results based on the data collected during the first year after the wildfire.
Current procedures of water management often use simple models and parameterisation to predict de... more Current procedures of water management often use simple models and parameterisation to predict demand that may fail in years with extreme climatic and hydrological conditions. The simulation platform ADEAUMIS, presented in a companion paper , was developed to improve decision-making in strategic water management. In order to estimate irrigation demand over large areas, it includes a spatial data base and a bio-decisional model. The main variables of the data base are irrigated area, weather data and agricultural practices (crops, sowing dates, varieties and irrigation decision rules). ADEAUMIS was applied in the Baïse sector, in south-western France, in 1998 and 2000.
Current procedures of water management often use simple models and parameterisation to predict de... more Current procedures of water management often use simple models and parameterisation to predict demand that may fail in years with extreme climatic and hydrological conditions. The simulation platform ADEAUMIS, presented in a companion paper , was developed to improve decision-making in strategic water management. In order to estimate irrigation demand over large areas, it includes a spatial data base and a bio-decisional model. The main variables of the data base are irrigated area, weather data and agricultural practices (crops, sowing dates, varieties and irrigation decision rules). ADEAUMIS was applied in the Baïse sector, in south-western France, in 1998 and 2000.
Since the late summer of 2005, the EROSFIRE project is investigating the hydrological and soil er... more Since the late summer of 2005, the EROSFIRE project is investigating the hydrological and soil erosion response in recently burnt eucalypt plantations. The study areas are located in the municipalities of Águeda, Albergaria-a-Velha and Sever de Vouga, at reasonably close distances from the project's home base at the University of Aveiro, which has allowed carrying out intensive field campaigns and, thereby, the collection of a rather extensive data set. The present paper will, however, only present initial results that, first and foremost, serve to illustrate the approach that is being applied and tested by the project. Field rainfall simulation experiments (RSE's) constitute the principal technique that is being employed by the EROSFIRE project for measuring soil erosion. Erosion plots are utilized as well but primarily to validate the results of the RSE's, in the sense of comparing erosion rates under artificial versus natural rainfall events as well as for micro-plots versus slope-scale plots. The various erosion measurements are ultimately intended to evaluate the suitability of a selection of erosion models for post-fire conditions. Results are presented of 8 RSE's that were carried out two to three months after a forest wildfire. The amounts of simulated rain that do not infiltrate into the soil, are substantial. The resulting overland flow, however, does not produce as high sediment yields as to be expected on the basis the RSE results of earlier studies in recently burnt pine stands in the same region.
The EROSFIRE project selected field rainfall simulation experiments (RSE's) as a time-and cost-ef... more The EROSFIRE project selected field rainfall simulation experiments (RSE's) as a time-and cost-effective approach to gather runoff and soil loss data under the rapidly changing conditions typical for commercial eucalypt plantations in Portugal following wildfires. These RSE data were first and foremost envisaged for initial parameterization and calibration of a physically-based model like MEFIDIS. Subsequent model assessment at larger spatial scales would then be based on slope-scale erosion plot data collected at a small number of selected sites. The present work, however, will in principle be confined to an analysis of the measurement results of the RSE's in two recently burnt eucalypt plantations. Repeat RSE's were carried out in two adjacent but contrasting eucalypt stands on steep hillslopes in north-central Portugal. This involved six occasions ranging from 3 to 24 months after a moderate severity fire in July 2005. A paired-plot experimental design was employed that comprised two pairs of RSE's at each site and occasion. From a grand total of 46 RSE's: (i) 24 and 22 RSE's involved application rates of 45-50 and 80-85 mm h-1, respectively; (ii) 22 took place in a stand that had been ploughed in downslope direction several years before the wildfire and 24 in an unploughed stand. The results showed a clear tendency for extreme-intensity RSE's to produce higher runoff amounts and greater soil and organic matter losses than the simultaneous high-intensity RSE's on the neighbouring plots. Nonetheless, there were marked exceptions, both in space (for one of the plot pairs) and time (under intermediate soil water repellency conditions). Also, overland flow generation and erosion varied significantly along the various field campaigns. This temporal pattern noticeably differed from a straightforward decline with time-after-fire, rather suggesting a seasonal component that reflected broad variations in topsoil water repellency. The ploughed site produced less runoff and erosion than the unploughed site, contrary to what would be expected if the downslope ploughing had occurred after the wildfire instead of several years before it. Finally, sediment losses at both study sites were noticeably lower than those reported by the other studies involving repeat RSE's, i.e. in Australia and western Spain. This possibly reflects a history of intensive land use in the study region, including in more recent times after the widespread introduction of eucalypt plantations.
Wildfires can have important impacts on hydrological and soil erosion processes, due to the destr... more Wildfires can have important impacts on hydrological and soil erosion processes, due to the destruction of vegetation cover and changes to soil properties. According to Shakesby and Doerr (2006), these wildfire effects are: i) much better known at small spatial scales (especially erosion plots) than at the scale of catchments; ii) much better studied with respect to overland flow and streamflow (and, then, especially peak discharges) than to soil erosion. Following up on a precursor project studying runoff generation and the associated soil losses from micro-plot to slope-scale in Portuguese eucalypt forests, the EROSFIRE-II project addresses the connectivity of these processes across hillslopes as well as within the channel network. This is done in the Colmeal study area in central Portugal, where the outlet of an entirely burnt catchment of roughly 10 ha was instrumented with a gauging station continuously recording water level and tubidity, and five slopes were each equipped with 4 runoff plots of < 0,5 m2 ("micro-plot") and 4 slope-scale plots as well as 1 slope-scale sediment fence. Starting one month after the August 2008 wildfire, the plots were monitored at 1- to 2-weekly intervals, depending on the occurrence of rainfall. The gauging station became operational at the end of November 2008, since the in-situ construction of an H-flume required several weeks. A preliminary analysis of the data collected till the end of 2008, focusing on two slopes with contrasting slope lengths as well as the gauging station: revealed clear differences in runoff and erosion between: (i) the micro-plot and slope-scale plots on the same hillslope; (ii) the two slopes; (iii) an initial dry period and a subsequent much wetter period; (iv) the slopes and the catchment-scale, also depending on the sampling period. These results suggest that the different processes govern the hydrological and erosion response at different spatial scales as well as for different periods, with soil water repellency playing a role during the initial post-fire period. The current presentation will review these preliminary results based on the data collected during the first year after the wildfire.
Current procedures of water management often use simple models and parameterisation to predict de... more Current procedures of water management often use simple models and parameterisation to predict demand that may fail in years with extreme climatic and hydrological conditions. The simulation platform ADEAUMIS, presented in a companion paper , was developed to improve decision-making in strategic water management. In order to estimate irrigation demand over large areas, it includes a spatial data base and a bio-decisional model. The main variables of the data base are irrigated area, weather data and agricultural practices (crops, sowing dates, varieties and irrigation decision rules). ADEAUMIS was applied in the Baïse sector, in south-western France, in 1998 and 2000.
Current procedures of water management often use simple models and parameterisation to predict de... more Current procedures of water management often use simple models and parameterisation to predict demand that may fail in years with extreme climatic and hydrological conditions. The simulation platform ADEAUMIS, presented in a companion paper , was developed to improve decision-making in strategic water management. In order to estimate irrigation demand over large areas, it includes a spatial data base and a bio-decisional model. The main variables of the data base are irrigated area, weather data and agricultural practices (crops, sowing dates, varieties and irrigation decision rules). ADEAUMIS was applied in the Baïse sector, in south-western France, in 1998 and 2000.
Uploads
Papers by Sergio Prats