Stochastic Modeling and Flood Dynamics

My research focuses on advancing the understanding of hydrological processes through theoretical and stochastic approaches, with particular attention to runoff generation, soil moisture dynamics, and flood frequency analysis.

The work bridges process-based hydrology and probabilistic modeling, providing analytical tools to describe complex basin-scale dynamics and support applications in flood risk management, ecohydrology, and hydraulic design.

1. Runoff Generation Mechanisms and Thresholds

A major research line investigates the physical mechanisms controlling runoff generation, emphasizing the role of threshold processes.

The development of theoretically derived flood frequency distributions has allowed a deeper interpretation of extreme events by explicitly incorporating multiple runoff mechanisms. In particular, a two-component probabilistic framework was introduced to account for different hydrological responses within a basin (Gioia et al., 2008).

This approach highlights that:

1. Ordinary floods are typically generated by infiltration excess in limited contributing areas
2. Extreme floods arise when storage thresholds are exceeded across large portions of the basin, producing highly skewed distributions

These findings provide a process-based explanation of flood statistics and establish a link between runoff dynamics and climate, soil, and geomorphology (Gioia et al., 2008).

2. Stochastic Modeling of Soil Moisture and Water Balance

Another core contribution is the development of analytical stochastic models describing soil moisture dynamics at the basin scale.

The soil water balance is formulated as a stochastic differential equation, where rainfall is modeled as a random forcing. This framework enables the derivation of probability density functions (PDFs) of basin-scale variables such as relative saturation and contributing areas (Manfreda & Fiorentino, 2008).

Key results show that:

1. Soil moisture variability is strongly controlled by maximum storage capacity
2. The spatial distribution of storage capacity governs the extent and variability of saturated areas

This approach provides a rigorous probabilistic description of basin dynamics and highlights the importance of spatial heterogeneity in hydrological processes (Manfreda & Fiorentino, 2008).

3. Variable Source Area Concept and Runoff Dynamics

The research advances the variable source area theory, particularly for humid river basins.

Runoff generation is described through saturation excess mechanisms, where only a fraction of the basin contributes to runoff at any given time. This fraction evolves dynamically depending on soil moisture conditions (Manfreda, 2008). The analytical framework developed allows:

1. Modeling of the temporal evolution of saturated areas
2. Derivation of the probability distribution of runoff production
3. Identification of the role of antecedent moisture and basin heterogeneity

These results provide a quantitative basis for understanding hydrological connectivity and runoff response in natural basins (Manfreda, 2008).

4. Theoretical Probability Distributions in Hydrology

A distinctive feature of this research is the use of theoretically derived probability distributions (TDDs) to describe hydrological and hydraulic processes.

These approaches extend classical statistical methods by incorporating physical process understanding into probabilistic models. Applications include flood frequency analysis, runoff production, and hydraulic processes such as bridge scour. For example, a theoretically derived distribution of scour was developed by linking flood statistics with hydraulic variables, offering a new framework for infrastructure risk assessment (Manfreda et al., 2018).

5. Impact of Human Interventions on Flood Regimes

Recent research has focused on the effects of human interventions, such as detention basins, on flood dynamics.

A theoretical framework was developed to quantify how these structures modify the probability distribution of flood peaks, using analytical relationships between inflow and outflow processes (Manfreda et al., 2021). The model shows that:

1. Flood attenuation depends on storage capacity and event duration
2. Hydraulic structures significantly alter the frequency and magnitude of extreme events

These findings support improved design and management of flood mitigation systems under changing environmental conditions (Manfreda et al., 2021).

6. Regionalization and Hydrological Signatures

The research also addresses the challenge of prediction in ungauged basins through the identification of hydrological signatures.

Model parameters have been linked to physically meaningful basin descriptors, enabling regionalization and transferability of models (Gioia et al., 2014). In particular:

1. Soil water loss is influenced by vegetation cover
2. Storage distribution parameters are controlled by topography

This approach provides a pathway toward physically-based parameter estimation and improved predictive capability in data-scarce regions (Gioia et al., 2014).

Concluding Remarks

This body of work contributes to hydrology by integrating stochastic modeling and physical process understanding, analytical probability distributions and environmental dynamics, and natural processes and human interventions. The resulting framework enhances our ability to interpret and predict hydrological behavior across scales, supporting applications in flood risk assessment, ecohydrology, and water resources management.

References

1. Gioia, A., Iacobellis, V., Manfreda, S., & Fiorentino, M. (2008). Runoff thresholds in derived flood frequency distributions. Hydrology and Earth System Sciences. [pdf]
2. Manfreda, S., & Fiorentino, M. (2008). A stochastic approach for the description of the water balance dynamics in a river basin. Hydrology and Earth System Sciences. [pdf]
3. Manfreda, S. (2008). Runoff generation dynamics within a humid river basin. Natural Hazards and Earth System Sciences. [pdf]
4. Gioia, A., Manfreda, S., Iacobellis, V., & Fiorentino, M. (2014). Performance of a theoretical model for the description of water balance and runoff dynamics. Journal of Hydrologic Engineering. [pdf]
5. Manfreda, S., Link, O., & Pizarro, A. (2018). A theoretically derived probability distribution of scour. Water. [pdf]
6. Manfreda, S., Miglino, D., & Albertini, C. (2021). Impact of detention dams on the probability distribution of floods. Hydrology and Earth System Sciences. [pdf]