Hydrological and SSY monitoring programme to estimate soil erosion in experimental basins at different scales in the Reno catchment

In the year 1997 a hydrologic, turbidimetric and geochemical monitoring programme was set up concerning the river Reno and some of its main tributaries. Thus according to the interest in sediment dynamics, gauging stations were built at the mouth of the Reno river (4139 Km2), at the outlet of the mountain basin of three of its tributaries, the Sillaro (139 Km2), the Lavino (83 Km2) and the Savena (124 Km2), and at the outlet of the minor Gaiana torrent (8.7 Km2). Each monitoring station was equipped with an ultrasonic flow meter, interfaced to an automatic pumped sampler The analysis performed on river samples are: suspended sediment concentration through gravimetric analysis, particle size distribution by the pipette method, turbidity with a laboratory turbidimeter and free settleable solids in Imhoff cone The turbidimetric method showed some shortcomings like the inaccuracy experienced both with highly turbid samples and with the presence of coarse particles. The free settleable solids in Imhoff cones resulted to be a simple, fast and low-cost method to estimate S.S.C.; moreover it has been possible to use it for the all the samples, with less restriction according to sample characteristics than the turbidimetric technique. The fluctuation of suspended sediment concentration appears to be connected to the water flow measured during the sampling. This hypothesis was confirmed by considering the data file of the suspended sediment concentration and the water flow, simultaneously measured. Thanks to the data and the analysis results presented we might state that • significant storm events are clustered during spring and autumn in all the three basins, as expected by their typical Apennines climatic regime • absolute maximum values in S.S.C. occur during summer season, probably because of heavy convective storms and the low vegetation cover for draught and agricultural practices. S.S.C. in average is higher in autumn, probably because of the high frequency of storm events in this season. • mean S.S.C. and maximum S.S.C. correlated well with “rainfall-based” variables, specifically with the maximum rainfall intensity (Imax), the Erosivity Index (E.I.) and the total rainfall height (Ptot) • mean and maximum solid load best correlated with E.I. and rainfall maximum intensity The results seem to indicate as hillsides as main sources of material to the monitored torrents. Cultivated lands on these basins are quite modest, therefore gully formations, pastures and degraded meadows have to be considered as main sources of sediment. The geologic and geomorphologic characteristics of the basins confirm this hypothesis, along with the numerous hydraulic arrengements confirming the highly unstable character of these lands. Nevertheless, the database produced within the context of the Reno basin monitoring programme is now reaching quite a good amount of data, this making it possible to undertake research activities in the field of hydrological numerical modelling and their capabilities. There is the strong belief that mathematical models along with their strengths and weak points should be widely tested, especially since they might be applied outside than research contexts and situations. For the time being, the event oriented, physically based model KINEROS2 is being used and tested. Rainfall-runoff events recorded at the outlet of the mountain catchment (8.7 Km2) of the Gaiana torrent were used in order to calibrate the model, so that KINEROS2 characteristics and capabilities could be assessed.