The activities carried out made it possible to substantially achieve the objectives set out in the funding application, which aimed to assess the extent to which changes in the agricultural landscape, agricultural drainage systems, and soil characteristics may have altered the water retention capacity of the territory and the generation of flood runoff in the agricultural lowlands of Romagna.
With reference to the objective of identifying representative areas of the Romagna territory and building an integrated knowledge base, the study focused on the Canale di Via Cupa catchment and, in particular, on the portion located upstream of the town of Russi. This area covers approximately 57.6 km² and was identified as the area most affected by the flooding that occurred in May 2015. An integrated database was developed, including historical and recent aerial and satellite imagery, soil data, geometric information on the land reclamation drainage network, topographic surveys of the flooded areas, rainfall data relating to the 2015 and 2023 events, and information on agricultural drainage systems. This result made it possible to integrate the agronomic, territorial, and hydrological components envisaged in the project proposal within a single case study.
With regard to the objective of reconstructing the changes that have occurred in farm structure and the rural landscape since the first half of the twentieth century, a multitemporal analysis of imagery from the periods 1931–1937, 1976–1978, and 2023–2024 was carried out across ten agricultural sample areas. The analysis documented a profound simplification of the agricultural landscape and a progressive consolidation of fields. The average field size increased from approximately 0.62 ha in 1931–1937 to 1.19 ha in 1976–1978, reaching approximately 8.47 ha in 2023–2024. This increase was observed in all the areas examined and confirms the transition from a mosaic of small fields to much larger cropping units, which are better suited to mechanisation and modern intensive farming systems.
In relation to the objective of assessing the effects of changes in agricultural drainage systems on water retention capacity, the variation in storage volumes associated with the field drainage ditch network was quantified. The estimated average storage volume decreased from approximately 334.6 m³/ha in 1931–1937 to 162.4 m³/ha in 1976–1978 and to 58.7 m³/ha in 2023–2024. The comparison between the earliest and most recent periods therefore indicates an average reduction of approximately 82%, with decreases ranging from about 71% to 94% across the individual areas. This result demonstrates a substantial loss in the capacity of agricultural land to temporarily store rainfall and slow its transfer towards the main land reclamation drainage network.
With reference to the objective of assessing the ecological simplification of the agricultural landscape, the presence of trees along field boundaries, drainage ditches, and farm roads was also quantified. In 1931–1937, an average of approximately 61.7 trees per hectare was recorded. By 1976–1978, this value had already fallen to approximately 2.1 trees per hectare, while no trees were identified in the ten areas examined in the 2023–2024 imagery. The analysis therefore documented the almost complete disappearance of the linear tree features that characterised the historical agricultural landscape and that could contribute to rainfall interception, increased surface roughness, and improved soil porosity and infiltration through their root systems.
To address the objective of characterising soil quality and drainage capacity at both farm and territorial scales, twelve soil samples were collected and analysed across four representative areas. The soils showed an alkaline reaction, with pH values ranging from 8.5 to 9.1. Electrical conductivity ranged from 86.5 to 345 µS/cm and was therefore below the regional reference value of 2,000 µS/cm in all samples. Soil organic matter content ranged from 1.30% to 2.89%, with average values across the four areas ranging from 1.68% to 2.04%.
The samples were assigned to hydrological soil groups B, C, and D. Six of the twelve samples belonged to group D, which is associated with lower infiltration capacity and a greater tendency to generate surface runoff. Experimental measurements showed infiltration rates ranging from 0.009 to 0.450 cm/h, with mean values by area ranging from approximately 0.03 to 0.21 cm/h. Nine of the twelve samples had values below 0.1 cm/h, indicating frequently limited infiltration capacity despite marked variability among sampling points. Statistical analysis showed no significant differences among the four areas in pH, electrical conductivity, soil organic matter content, measured infiltration rate, infiltration rate estimated from soil texture, or the results of the dispersion test. The findings therefore indicate substantial variability within the areas, but no systematic and statistically demonstrable differences among the areas considered. Under particularly intense or prolonged rainfall, the low infiltration rates observed in several samples may contribute to the generation of surface runoff.
In relation to the objective of assessing the hydrological impact of agricultural transformations through modelling, a linear reservoir hydrological model was developed to reconstruct the catchment response to rainfall events and estimate flood discharges. The maximum conveyance capacity of the Canale di Via Cupa, estimated using the Chézy hydraulic formula, was approximately 20.89 m³/s. Simulations based on observed rainfall produced peak discharges of approximately 25.58 m³/s for the May 2015 event and 21.10 m³/s for the May 2023 event. These results are consistent with the impacts observed in the area: in 2015, the simulated discharge exceeded the estimated capacity of the canal and extensive flooding occurred, whereas in 2023 the system reached conditions close to its maximum capacity.
With regard to the objective of comparing current conditions with historical configurations characterised by more extensive agricultural drainage systems, an initial modelling scenario representative of farming conditions in the first half of the twentieth century was developed. This scenario was characterised by greater water storage capacity and a denser network of field drainage ditches and agricultural drainage systems. Under the preliminary scenario considered, the increase in catchment retention capacity reduced the simulated peak discharge to approximately 17.07 m³/s, corresponding to an indicative reduction of 33% compared with the peak reconstructed for the 2015 event. This result highlights the potential contribution of traditional agricultural drainage systems, the conservation of the minor drainage network, and the restoration of vegetated features in slowing runoff generation and reducing the risk of canal overflow.
Within the overall framework of the activities carried out, the results confirm the hypothesis underlying the funding application: the hydraulic vulnerability of lowland agricultural areas does not depend exclusively on the intensity of meteorological events, but is also influenced by the way in which the territory is managed. The increase in field size, the substantial reduction in drainage ditches and their associated storage volumes, the disappearance of trees, and the limited infiltration capacity of some soils may contribute to accelerating the hydrological response of the catchment and increasing the discharges conveyed through the land reclamation drainage network.
The results obtained also provide the knowledge base required to identify the adaptation strategies envisaged by the project. The analyses indicate that potentially effective measures include the conservation and restoration of the minor drainage network, an increase in water storage capacity at farm and territorial scales, the restoration of trees and other vegetated features, and the adoption of agronomic practices aimed at protecting soil structure, organic matter content, and infiltration capacity.