Ajla Mulaomerović-Šeta

Rapid urban development and increasing hydro-meteorological variability are creating new challenges for flood risk management in urban areas. In this context, the concept of a smart city offers opportunities to improve the understanding, prediction, prevention, and management of flood events through the integration of environmental monitoring, digital technologies, data, modelling, and decision-support systems. However, becoming a "smart city" in terms of flood management requires more than the introduction of individual technological solutions. This paper explores what Sarajevo would need to develop an effective smart flood management system. The analysis considers the key components required for such a system, including reliable and spatially distributed meteorological and hydrological monitoring, real-time data acquisition and transmission, integration of heterogeneous data sources, high-resolution geospatial databases, hydraulic and hydrological modelling, flood forecasting and early warning, and decision-support tools for emergency response. Particular attention is given to the integration of existing monitoring and information systems and to the interoperability of data generated by different institutions and infrastructure systems. The paper further considers the importance of defining clear institutional responsibilities, data-sharing mechanisms, communication channels, and procedures for translating technical information into timely decisions. Rather than proposing a single technological solution, the paper identifies the essential components and their interconnections that should form the basis for Sarajevo's transition towards smart flood risk management. The proposed framework can serve as a basis for assessing the current level of preparedness and identifying priorities for future investments in monitoring, modelling, digital infrastructure, institutional capacity, data exchange, and warning communication.

Regional analysis is often used for flood quantile estimation in ungauged catchments. The regionalization procedure has two phases: the formation of homogeneous regions and flood quantile estimation. The presented research results consider the first phase of the regional analysis for 41 catchments in Serbia. The catchment similarity attributes are catchment area and catchment mean elevation. The number of formed regions and the number of stations within the regions are determined by maximising the mean silhouette width of the region. Regions were first obtained by cluster analysis and then adjusted to comprise catchments with a positive silhouette width. For the three formed regions, homogeneity was checked by the Gini index - GI.

Flood quantile estimation in ungauged basins is often performed using regional analysis. A regionalization procedure consists of two phases: the definition of homogeneous regions among gauged basins, i.e., clusters of stations, and information transfer to the ungauged sites. Due to its simplicity and widespread use, a combination of hierarchical clustering by Ward’s algorithm and the index-flood method is applied in this research. While hierarchical clustering is very efficient, its shortcomings are the lack of flexibility in the definition of clusters/regions and the inability to transfer objects/stations from one cluster center to another. To overcome this, using silhouette width for induced clustering of stations in flood studies is proposed in this paper. A regionalization procedure is conducted on 53 gauging stations under a continental climate in the West Balkans. In the induced clustering, a negative silhouette width is used as an indicator for the relocation of station(s) to another cluster. The estimates of mean annual flood and 100-year flood quantiles assessed by the original and induced clustering are compared. A jackknife procedure is applied for mean annual flood estimation and 100-year flood quantiles. Both the Hosking–Wallis and Anderson–Darling bootstrap tests provide better results regarding the homogeneity of the defined regions for the induced clustering compared to the original one. The goodness-of-fit measures indicate improved clustering results by the proposed intervention, reflecting flood quantile estimation at the stations with significant overestimation by the original clustering.