😅Hydrological Modeling Unit 15 – Flood Forecasting and Extreme Events

Key Concepts and Definitions

• Flood defined as an overflow of water that submerges land that is usually dry
• Extreme events characterized by their rarity, intensity, and potential for significant impacts
• Return period represents the average time between events of a certain magnitude or greater
• Hydrograph depicts the rate of flow versus time past a specific point in a river, or other channel or conduit carrying flow
  • Consists of a rising limb, peak, and falling limb
• Hydrological modeling involves mathematical representations of the rainfall-runoff process and river routing
• Uncertainty in flood forecasting arises from input data, model structure, and parameter estimation
• Risk assessment evaluates the likelihood and consequences of flood events
• Flood frequency analysis uses historical data to estimate the probability of future flood events

Flood Types and Characteristics

• Riverine floods occur when water levels in rivers, lakes, or streams overflow their banks
  • Often caused by heavy rainfall, snowmelt, or a combination of both
• Flash floods characterized by rapid onset and high velocity water flow
  • Typically result from intense rainfall over a short period (a few hours or less)
• Coastal floods happen when storm surges, high tides, or tsunamis cause seawater to inundate coastal areas
• Urban floods result from inadequate drainage infrastructure in cities during heavy rainfall events
• Dam and levee failure floods occur when these structures fail, releasing large volumes of water
• Floods can cause extensive damage to infrastructure, property, and pose risks to human life
• Flood characteristics include depth, velocity, duration, and extent of inundation

Extreme Event Analysis

• Focuses on understanding and quantifying the probability and magnitude of rare, high-impact events
• Flood frequency analysis uses historical flood data to estimate the likelihood of future events
  • Involves fitting a probability distribution (e.g., Gumbel, Log-Pearson Type III) to the data
• Return period (or recurrence interval) is a key concept in extreme event analysis
  • Represents the average time between events of a certain magnitude or greater (100-year flood)
• Probable Maximum Flood (PMF) is the largest flood that can reasonably be expected in a given area
  • Estimated using Probable Maximum Precipitation (PMP) and hydrological modeling
• Climate change is expected to alter the frequency and intensity of extreme events in many regions

Hydrological Modeling Techniques

• Lumped models treat the entire watershed as a single unit with averaged properties
  • Examples include the Rational Method and the SCS Curve Number method
• Distributed models divide the watershed into smaller units (grid cells or sub-basins) to capture spatial variability
  • Examples include SWAT, HEC-HMS, and VIC models
• Physically-based models incorporate the underlying physical processes of the hydrological cycle
  • Solve equations for infiltration, evapotranspiration, and surface and subsurface flow
• Conceptual models use simplified representations of hydrological processes
  • Often based on the concept of interconnected storage reservoirs
• Data-driven models, such as artificial neural networks, use machine learning to identify patterns in data

Data Sources and Collection Methods

• Precipitation data obtained from rain gauges, weather radar, and satellite observations
  • Rain gauge networks provide point measurements of rainfall depth
  • Weather radar provides high-resolution estimates of rainfall intensity and spatial distribution
• Streamflow data collected using stream gauges that measure water level (stage) and convert it to discharge
• Topographic data, such as digital elevation models (DEMs), used to delineate watersheds and derive terrain attributes
• Land use and land cover data obtained from satellite imagery and field surveys
  • Influences infiltration, runoff, and evapotranspiration processes
• Soil data, including soil type and hydraulic properties, obtained from soil surveys and field measurements
• Remote sensing techniques, such as LiDAR and InSAR, provide high-resolution elevation and surface deformation data

Flood Forecasting Methods

• Deterministic forecasting uses a single set of input data and model parameters to generate a single forecast
  • Examples include the Sacramento Soil Moisture Accounting (SAC-SMA) model and the HEC-RAS model
• Probabilistic forecasting incorporates uncertainty by using an ensemble of input data and model parameters
  • Ensemble forecasting involves running multiple model simulations with slightly different initial conditions
• Short-range forecasting (up to 48 hours) relies heavily on real-time observations and weather forecasts
  • Nowcasting techniques use radar and satellite data to provide short-term (0-6 hours) forecasts
• Medium-range forecasting (2-15 days) uses a combination of observations and numerical weather prediction models
• Long-range forecasting (months to seasons) incorporates climate models and teleconnections (El Niño/La Niña)
• Operational flood forecasting systems integrate various data sources, models, and decision support tools

Uncertainty and Risk Assessment

• Uncertainty in flood forecasting arises from multiple sources
  • Input data uncertainty due to measurement errors, sampling bias, and spatial variability
  • Model structure uncertainty due to simplifications and assumptions in the model representation
  • Parameter uncertainty due to the difficulty in estimating model parameters from limited data
• Uncertainty quantification techniques include sensitivity analysis, Monte Carlo simulation, and Bayesian methods
• Risk assessment combines the likelihood and consequences of flood events
  • Consequences can be measured in terms of economic losses, casualties, and environmental impacts
• Flood risk maps delineate areas with different levels of flood hazard and vulnerability
  • Used for land use planning, insurance pricing, and emergency response planning
• Risk communication involves effectively conveying flood risk information to stakeholders and the public

Real-World Applications and Case Studies

• The European Flood Awareness System (EFAS) provides early warning for floods across Europe
  • Uses a combination of weather forecasts, hydrological models, and real-time observations
• The U.S. National Water Model (NWM) is a continental-scale hydrological modeling system
  • Simulates streamflow and other hydrological variables for 2.7 million river reaches
• The Yangtze River flood of 1998 was one of the most severe floods in China's history
  • Caused by a combination of heavy rainfall, deforestation, and inadequate flood control measures
• The 2011 Brisbane flood in Australia resulted from a combination of heavy rainfall and dam release
  • Highlighted the challenges of balancing flood control and water supply objectives
• The 2021 European floods, particularly affecting Germany and Belgium, caused widespread damage and loss of life
  • Demonstrated the need for improved flood risk management and climate change adaptation
• Case studies provide valuable lessons for improving flood forecasting and risk management practices