Lecture 3 Part 1
Infiltration-Excess vs. Saturation-Excess Runoff
Runoff generation in this lecture is driven by two main processes. Infiltration-excess runoff, also known as Hortonian overland flow, occurs when the rainfall rate exceeds the soil’s infiltration capacity. In other words, if the rainfall rate R(t) is larger than the infiltration rate f(t) at the surface, water cannot infiltrate fast enough and runs off the surface. Saturation-excess runoff occurs when the soil pore space becomes saturated, either because rainfall continues for a long duration or because the soil has a low infiltration capacity that becomes saturated; in this case, infiltration is constrained by the available pore space and rainfall that cannot infiltrate becomes runoff. Formally, if infiltration capacity is exceeded, surface runoff forms; if infiltration capacity can accommodate the rainfall, surface runoff may be minimal or absent. A compact representation is:
while if $R(t)\le f(t)$, $Q{surf}(t)=0$ (ignoring other pathways like interflow for the moment).
Infiltration capacity is not constant. It tends to vary with soil type, moisture, history, and time. Even when rainfall intensity is not extreme, f(t) slowly declines over time as the soil becomes wetter and pores fill. A general way to express this is that f(t) is a decreasing function of time (i.e., $f'(t)<0$) after rainfall begins, so infiltration rates fall even if rainfall is modest. A simple expression to reflect the temporal decrease is:
where $f0$ is the initial infiltration capacity and $\phi(t)$ captures the decline with time.
Two distinct rainfall-driving regimes emerge:
Infiltration-excess (Hortonian) runoff tends to occur with short, intense rainfall and/or in areas with low permeability (e.g., clay-rich or compacted soils). Cracking clays can create transient infiltration pathways when soils are dry, but when soils are wet they may behave as low-permeability zones with limited infiltration.
Saturation-excess runoff tends to occur with long-duration, gentle-to-moderate rainfall when the pore space becomes saturated, causing the infiltration rate to fall and rainfall to contribute to surface runoff.
This framework helps explain why rainfall events with high intensity often produce substantial runoff, especially if the catchment is wet (soil moisture near saturation) or has low infiltration capacity. Conversely, gentler, prolonged rains in a catchment with high infiltration capacity may infiltrate more readily and produce less immediate runoff.
Temporal and spatial variability on hillslopes
Hillslopes are not uniform. Infiltration-excess runoff can occur on portions of a slope with relatively low infiltration capacity, while further downslope, infiltration may be higher and runoff minimal. The result is a mosaic of processes along the slope: some areas contribute surface runoff, some re-infiltrate, and some contribute to interflow or groundwater processes. The lecture emphasizes that the simple picture of a uniform infiltration-capacity is inappropriate for real landscapes.
Two common patterns are highlighted:
On the upslope, infiltration-excess flow may dominate where soils are less permeable or where infiltration is temporarily hindered (e.g., by crusts, compaction, or perched groundwater).
Further downslope, infiltration can resume (or baseflow contribute) as infiltration capacity recovers due to drainage or less resistance, leading to a patchwork of runoff and infiltration zones. This is why rainstorms do not produce uniform runoff across a catchment.
Groundwater interactions: exfiltration, base flow, and saturated areas
When soils and underlying materials become wetted, groundwater processes begin to influence surface hydrology. Saturation-excess runoff can be accompanied by exfiltration of groundwater to the surface, i.e., base flow contributing to surface water and, in saturated zones, water exfiltrates rather than infiltrates. The transcript notes several related ideas:
Base flow may emerge at the surface when the groundwater table is high and water exfiltrates.
Saturation-excess runoff can occur in areas where the water table is near or at the surface, leading to ponding and surface flow, even if rainfall alone would not have produced significant infiltration-excess runoff.
Infiltration-excess and saturation-excess processes are not mutually exclusive in a real catchment; different parts of the landscape can experience different dominant processes depending on soil, moisture, and rainfall regime.
A practical takeaway is that rainfall-runoff behavior depends not just on rainfall intensity but also on catchment moisture status and the spatial distribution of soil permeability and groundwater conditions.
Rainbow Beach: a groundwater-pulse mechanism distinct from infiltration/saturation excess
Rainbow Beach is used as an illustrative aside to show that not all ponding events fit neatly into infiltration-excess or saturation-excess frameworks. In March of a given year, the area (sand-dominated) experienced heavy rainfall and ponding due to a groundwater mound rising to the surface under the sand. In this case:
The sand has high infiltration capacity, so rainfall does not produce traditional infiltration-excess runoff on the beach itself.
The key mechanism was groundwater mound development that lifted the water table to the surface, creating ponding on the surface regardless of infiltration rate.
Once rainfall ceased, ponding dissipated quickly as the water table subsided.
This scenario is a reminder that perched groundwater and perched water tables can drive surface ponding independently of the surface infiltration capacity and can coexist with or override infiltration/excess processes in other parts of a catchment.
Large rainfall totals and flood potential
The lecture emphasizes that large rainfall totals often occur together with other large totals and are associated with higher runoff. Specifically:
Higher rainfall totals tend to produce more runoff because the infiltration capacity is exceeded for longer durations and/or rainfall occurs when the catchment is already moist.
The timing and clustering of heavy rainfall events matter: a rainstorm that arrives on a wet catchment with high rainfall intensity will produce much more runoff than the same storm on a dry catchment.
The discussion notes real-world examples (e.g., urban Brisbane after heavy rainfall) and connects soil moisture status (e.g., root-zone soil moisture) to flood risk and runoff generation.
Quantitatively, one can describe the relationship between rainfall and runoff as a growing mismatch between rainfall input and the soil’s ability to infiltrate, leading to elevated surface runoff as rainfall totals rise and soil moisture remains high.
Modeling in hydrology: purpose, types, and calibration
The lecture introduces modeling as a central tool in hydrology for extrapolating processes in space and time. Models are used to interpret observations, test hypotheses, and project scenarios (e.g., floods, droughts, or climate-change impacts). A few key points:
Models are simplifications of reality. They require assumptions, parameterizations, and often calibration to data. They are not perfect representations but are useful tools when applied with care.
The course distinguishes between physical models (e.g., scale experiments, physical representations of flow) and mathematical models (e.g., differential equations, rainfall-runoff relationships, and empirical relations).
Calibration is the process of adjusting model parameters to reproduce observed data. The Brisbane River catchment study is used as an example: observed flows in the city, impacted by dams, are matched by a model (the Erbs model) with dam operations included. Calibration targets the red curves (fitted values).
A scenario often used is to remove dam constraints in the model to estimate natural flows. This yields a dataset that is uniform with respect to dam presence/operation, enabling fair comparisons and the derivation of flood statistics such as the 1% annual exceedance probability flood (the 1% flood). This illustrates how modeling supports extrapolation in time and space and helps derive robust flood quantiles.
Two important caveats are highlighted:
Models are simplifications of reality; the accuracy depends on assumptions, data quality, and understanding of processes.
When data and processes are complex (e.g., urban catchments with dams, land-use change, and groundwater interactions), multiple models or model components may be used to capture essential dynamics, and results should be interpreted with uncertainty in mind.
The two sorts of models: physical and mathematical
The lecture contrasts two broad classes of models:
Physical models: These are tangible representations or experiments that simulate flow and transport in a scaled or analog form. They help visualize processes, test hypotheses about mechanisms, and validate mathematical formulations under controlled conditions.
Mathematical models: These are sets of equations and algorithms that describe the system using variables and parameters. They include physically based models (e.g., Richards’ equation for infiltration, groundwater flow equations) and empirical or conceptual models (e.g., simple rainfall-runoff relationships, unit hydrographs).
The goal is to use models to understand, quantify, and predict hydrological behavior. In practice, we often use both types in a complementary fashion: physical models to gain mechanistic insight and mathematical models for prediction and scenario analysis. The progression of scientific understanding—from Newtonian physics to relativity as described in the lecture—illustrates how models evolve with better data and theory. Similarly, hydrology relies on evolving models that become progressively more capable as knowledge improves, even while acknowledging that no model captures every detail of a real catchment.
Practical implications for exams and study
Be able to explain the difference between infiltration-excess and saturation-excess runoff, including the key drivers (rainfall intensity vs. infiltration capacity, soil moisture state, pore-space saturation).
Understand that infiltration capacity declines with time during rainfall and that rainfall on a saturated catchment tends to produce more runoff;
Recognize that hillslopes show spatial variability in infiltration and runoff generation; runoff can originate in patches rather than uniformly downslope.
Distinguish groundwater-related ponding (exfiltration, perched water tables) from infiltration-excess and saturation-excess mechanisms; Rainbow Beach is an example of groundwater-driven ponding independent of the two main runoff processes.
Appreciate how rainfall totals interact with catchment moisture to influence flood risk; large totals often coincide with higher runoff, especially on previously wetted catchments.
Grasp the role of hydrological modeling: why we calibrate to observed data, how dams affect observed flows, and how removing dams in a model can yield a uniform baseline for flood analysis (e.g., the 1% flood).
Differentiate between physical and mathematical models, and recognize that they are both used to interpret and predict hydrological behavior, with ongoing refinement as knowledge advances.
If you want, I can convert these notes into a condensed study guide with a quick glossary of terms and a few practice questions focusing on likely exam prompts (e.g., deriving runoff type given R and f, explaining Rainbow Beach phenomena, or outlining the Brisbane River modeling workflow).