When rain falls on a basin, not all of it becomes a flood. Part of it is caught on leaves and evaporates, part fills small depressions, part infiltrates into the soil, and what remains begins its journey over the surface toward the channels. Surface runoff modeling is an engineering attempt to trace and calculate this journey: how much will reach the basin outlet, when, and at what peak. These figures determine the size of a culvert, a small dam, or a flood map.
The journey begins with the concept of excess rainfall, which is the part of the rain that is not absorbed by abstractions and infiltration and therefore becomes runoff. This part is not constant. It varies with the intensity and duration of the storm, the soil type, the land cover, and the soil moisture before the storm. Light rain may be completely absorbed by dry soil, while the same intense storm falling on saturated soil or an impervious surface turns into direct flood flow. The excess rainfall then travels across the slopes to the valleys and collects at the outlet in a time curve called the hydrograph, from which the peak, the runoff volume, and the arrival time are read.
Models vary according to their level of detail. The simple rational method estimates the peak alone for small basins. Lumped models treat the basin as a single unit. Semi-distributed models, including HEC-HMS, divide the basin into sub-basins and connect them with channel reaches. Physically based distributed models solve the equations of water movement on a grid of cells and require far more data. A more complex model is not always a more accurate one; the appropriate model is the one whose data are available and which suits the purpose of the study.
A semi-distributed model consists of components, each performing a specific function. The loss component separates excess rainfall from total rainfall, and one of its best-known methods is the Curve Number (SCS-CN), which links soil type and land use to a single number; the higher the number, the greater the runoff. The transform component converts excess rainfall into a hydrograph using the unit hydrograph, and its shape depends on the lag time or time of concentration, that is, the time water needs to travel across the basin from the farthest point to the outlet. Next comes the routing component, which simulates wave movement in the channels, and finally the baseflow component, which is usually marginal in arid regions because their rivers are seasonal.
These concepts appear in the study we discussed for the Hilla and Diwaniyah basins. The curve numbers ranged from about 88 to 91, which are high values indicating a nearly impervious surface that generates substantial runoff from little rain, and the lag times ranged from about two to seven hours depending on the area and slope of the sub-basin. When a dry year was compared with a wet one, the simulated peak flow at the outlet rose from 131 to 194 m³/s in Hilla, and from 135 to 224 m³/s in Diwaniyah. This comparison is a reminder of a basic fact: the variation of rainfall from one year to another is what controls runoff most, not the change in the basin surface alone.
Arid basins have particular characteristics that make them harder to model. Their storms are short, intense, and spatially irregular; their soils may develop a crust that reduces infiltration; and their channels are seasonal wadis that flow only after rain, so part of the water is lost through seepage across the wadi bed. The problem is compounded by the scarcity of rainfall and discharge stations, so satellite data such as CHIRPS are used. These are suitable but need local validation, as we mentioned in the remote sensing article.
This is where calibration and validation come in, which turn the model from a theoretical calculation into a reliable tool. The model parameters are adjusted until its results approach measured discharges in a given period, and then tested in another period that was not used in calibration. The quality of the match is measured by indicators such as the Nash–Sutcliffe Efficiency (NSE) and percent bias. In regions with scarce measurements, adequate discharge records are often absent, so the results are read as indicators of the trend and magnitude of the phenomenon rather than precise predictions, and comparison with documented flood events or remote sensing data is used to support them.
The practical workflow can be summarized in steps: prepare the digital elevation model, soil and land cover maps, and rainfall data; then delineate and subdivide the basin; then estimate parameters such as the curve number and lag time; then simulate a design storm or historical storms; then calibrate and validate if data are available; then test scenarios such as climate change or land use change; and finally use the results in structure design and in producing hazard maps and early warning systems.
Surface runoff modeling thus shows that a flood is not merely a quantity of rain but the result of an interaction among a storm, soil, surface, terrain, and time. The real value of a model is not in the number it gives, but in understanding what drives that number, which allows the engineer to ask: what if urbanization increases, the storm changes, or the climate shifts?
Presented by: Asst. Lect. Zahraa Hussein Mohammed.
References
USDA Natural Resources Conservation Service (2004). National Engineering Handbook, Part 630 Hydrology, Chapter 10: Estimation of Direct Runoff from Storm Rainfall.
USDA Natural Resources Conservation Service (2010). National Engineering Handbook, Part 630 Hydrology, Chapter 15: Time of Concentration.
Chow, V.T., Maidment, D.R., Mays, L.W. (1988). Applied Hydrology. McGraw-Hill, New York.
U.S. Army Corps of Engineers, Hydrologic Engineering Center. HEC-HMS Technical Reference Manual.
Beven, K.J. (2012). Rainfall-Runoff Modelling: The Primer (2nd ed.). Wiley-Blackwell.
Hawkins, R.H., Ward, T.J., Woodward, D.E., Van Mullem, J.A. (2009). Curve Number Hydrology: State of the Practice. American Society of Civil Engineers.
Moriasi, D.N., et al. (2007). Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Transactions of the ASABE, 50(3), 885–900.
Nash, J.E., Sutcliffe, J.V. (1970). River flow forecasting through conceptual models, Part I. Journal of Hydrology, 10(3), 282–290.
Zahraa Hussein Mohammed Aziz (2024). Projections of Climate Changes in Hydrological Regime of the middle Mesopotamian plain. Master’s thesis in Civil Engineering (Water Resources), University of Babylon, Iraq.