Climate change is expected to significantly increase the frequency and severity of climate-related extreme events, including floods (Calvin et al., 2023). These events are likely to impose increasing costs on the economy, individuals, and companies, affecting both the productive capacity and the overall economic stability (Fatica et al., 2024; Cl`o et al., 2024; Benincasa et al., 2024). 

The rising frequency of hydrologic risks underscores the need for effective mitigation and adaptation measures aimed at minimizing the adverse consequences of climate change. Among these measures, state-level interventions, such as investments in infrastructure designed to reduce hydrologic risks, can play a crucial role in enhancing economic resilience and safeguarding productive assets. 

Italy represents a particularly relevant case for studying the impacts of hydrologic risk mitigation investments. The country is highly exposed to the effects of climate change, with its diverse topography and extensive river network, making it especially prone to flood-related hazards (Trigila et al., 2021). 

Moreover, the Italian productive system is primarily composed of micro, small, and medium enterprises, which tend to be more vulnerable to extreme weather events and often lack adequate insurance coverage against such risks (ECB and EIOPA, 2024). These factors make it crucial to assess the effectiveness of public investments in flood risk mitigation in Italy, as such initiatives can provide valuable protection for businesses and local economies. 

Evaluating the effectiveness of public investments in mitigating hydrologic risk requires information on public expenditures, flood risk levels, and recorded flood events. Focusing on public investments, a key initiative is the Repertorio Nazionale degli interventi per la Difesa del Suolo (the National Repository of Soil Defense Interventions, hereinafter ReNDiS) (ISPRA, n.d.). 

This is a database compiled by the Istituto Superiore per la Protezione e la Ricerca Ambientale (Italian Institute for Environmental Protection and Research, hereinafter ISPRA) that tracks interventions aimed at mitigating hydrologic risk across Italy. While ReNDiS is well aligned with the scope of this study, it has limitations, as it mainly includes projects funded by the Ministero dell’Ambiente e della Sicurezza Energetica (Ministry of Environment and Energy Security). 

On the contrary, OpenCUP (DiPE and MEF, 2025), an open-access portal provided by the Dipartimento per la programmazione e il coordinamento della politica economica (Department for the Planning and Coordination of Economic Policy, hereinafter DiPE) of the Presidenza del Consiglio dei Ministri (Presidency of the Council of Ministers), accounts for public interventions financed through local, national, European, and even private funds. Nevertheless, it includes investments in a wide variety of topics, hence the necessity of a specific filtering before using it for the purposes of this study.

In terms of hydrologic risk mapping, a valuable source of information is the Mosaicatura ISPRA delle aree a pericolosit`a idraulica (ISPRA mosaic of hydraulic risk areas) (ISPRA, 2020b), mapped by the Autorit`a di Bacino (River Basin Authorities). 

Last updated in 2020, the ISPRA Mosaic identifies high, medium, and low flood risk zones in Italy. The classification is based on three scenario analyses that simulate extreme events of increasing severity, identifying areas that could be inundated under each scenario. As regards the exposure to hydrologic risk, it is essential to consider the entire network of watercourses flowing within the Italian territory. 

To this end, multiple data sources exist. The dataset Reticolo Idrografico Nazionale (National Hydrographic Network) (ISPRA, 2020a), compiled by ISPRA, provides a representation of Italian hydrography limited to natural watercourses, available in shapefile format at a 1:250,000 scale. 

In contrast, two additional sources also include artificial canals. Firstly, Humanitarian OpenStreetMap Team (2025), within the Humanitarian Data Exchange project, offers the monthly updated Italy Waterways map. The second source is instead a map developed within the framework of the Strati Prioritari di Interesse Nazionale1 (Priority layers of National Interest, hereinafter DBPrior10k), as detailed in Section 2.1. 

The choice of a hydrographic network which includes artificial canals, as well as natural rivers, is crucial for the purposes of this study, since both types of watercourse are affected by flood events. Finally, regarding flood occurrences in Italy, there are various potential sources. ISPRA’s annual reports on climate events (Berti and Lucarini, 2024) - which we detail in Section 2.3 - provide information on actual flood events, reporting the affected river basin, but not the impacted municipality. 

Differently, the dataset by Gatto et al. (2023) collects all hydrogeological disasters (both landslides and floods) that occurred in Italy between 2013 and 2022, which caused such severe impacts as to require the declaration of national-level emergencies. Although the dataset provides detailed data at the municipal level, it is not possible to distinguish between landslides and floods, nor there is any information about the rivers affected by such events. 

In addition, Copernicus Emergency Management Service (n.d.) offers mapping services in cases of natural hazards. The platform documents the extent of flooded areas for selected major flood events worldwide. However, a key limitation is that it does not provide a comprehensive inventory of all flood events, instead focusing only on the most significant ones. 

Despite this, it can be valuable for case studies, as it offers precise spatial data on affected areas. Similarly, the institute Popolazione a Rischio da Frana e da Inondazione in Italia (Population at risk of landslides and floods in Italy, hereinafter Polaris) (CNR, 2024), part of the Consiglio Nazionale delle Ricerche (National Council of Research, CNR), publishes detailed descriptions of selected flood events, providing information on the most affected municipalities and rivers, although without offering precise data on the extent of flooded areas. 

Despite the availability of valuable data sources, analyzing the distribution and effectiveness of flood risk mitigation and adaptation measures in Italy remains challenging. This is primarily due to the absence of a comprehensive database that integrates all the necessary information for such assessments. 

To address this limitation, our work makes a key contribution by developing a novel dataset that integrates multiple information sources essential for evaluating the effectiveness of public investments in mitigating flood risks. Specifically, we combine data on the Italian hydrography, records of public investments targeted at mitigating hydrologic risks, and historical data on flood events across Italy. 

Among the available possibilities in terms of hydrography, we opted for DBPrior10k, because it reports both natural and artificial watercourses. As for public interventions, we selected OpenCUP: despite encompassing a much broader scope than the one required for this study, it offers the flexibility to be filtered according to our specific needs. 

Finally, as regards flood events, we think that ISPRA’s annual reports on climate events are the most suitable option for our case: while municipal-level information is not readily available, we can reconstruct it using hydrography data. The resulting dataset provides a unique spatial representation of flood mitigation investments at the river-municipality level together with their evolution over time. 

The construction of this dataset represents an initial step in a broader project aimed at informing policy discussions on the benefits and cost-effectiveness of public investments in hydrologic risk management, as well as identifying potential investment gaps that require further attention. We intend to use this dataset to conduct additional analyses in future work. 

One possible extension involves integrating data on extreme rainfall events in Italy to assess whether public investments in hydrologic risk mitigation effectively reduce the likelihood of flooding. Another promising direction would be to combine this foundational dataset with firm-level data to evaluate whether, in the event of a flood, firms located in areas with greater public investment experience less severe consequences. 

The remainder of the paper is structured as follows. Section 2 describes the data sources, while Section 3 details the methodology used to construct the dataset. Section 4 presents the final dataset, while Section 5 presents the next steps and possible use cases of the dataset.