Theory explainer
Unit Hydrograph Problems — FE/PE Civil Exam Worked Examples
The unit hydrograph turns a rainfall hyetograph into a flood hydrograph, and the exam tests it in both directions: derive the unit hydrograph from an observed storm, then use it to predict the flood from a new storm. Both directions run on the same two ideas — proportionality and superposition — plus one bookkeeping rule candidates keep forgetting: baseflow is subtracted before you derive, and added back after you predict.
Last reviewed: 2026-11-07. Every worked number below was independently re-solved; steps are shown in full.
What a unit hydrograph is, and what it assumes
A D-hour unit hydrograph is the direct runoff hydrograph a catchment would produce from exactly one unit of rainfall excess — one inch, or one centimetre — falling uniformly over the whole catchment during D hours. It is a property of the catchment, not of any one storm: once derived, the same unit hydrograph predicts the runoff from any storm, provided the storm's excess can be split into D-hour pulses.
The method rests on two assumptions, and the exam tests both. Proportionality (linearity): doubling the rainfall excess doubles every ordinate of the runoff hydrograph, with no change in shape or timing. Superposition (time invariance): the runoff from a multi-pulse storm is the sum of the runoffs each pulse would produce alone, each lagged to its own start time. Together they give the discrete convolution equation in the box above. The method also assumes the rainfall is uniform over the catchment and uniform in intensity during each D-hour pulse — which is why it is a catchment-scale approximation, generally applied to catchments up to about 5,000 km², and why it breaks down for very large or patchy storms.
Deriving a unit hydrograph from an observed storm
Derivation runs the prediction routine backwards. Start with the observed flood hydrograph and a constant (or separated) baseflow, and subtract: observed minus baseflow is the direct runoff hydrograph (DRH). Next, find the rainfall excess depth for the storm — total rainfall minus losses, for example by applying a φ-index over the storm duration. For a storm whose excess falls in a single D-hour pulse, the unit hydrograph is simply every DRH ordinate divided by the excess depth. For a multi-pulse storm, work forward from the first ordinate: the first DRH ordinate depends only on the first pulse, so it gives the first unit ordinate directly; substitute that into the second ordinate to solve for the second unit ordinate, and so on. That back-substitution is called deconvolution, and it is sensitive to measurement error — a small error in an early ordinate propagates into all later ones, which is why derived unit hydrographs are usually smoothed before use.
Predicting with a unit hydrograph: discrete convolution
Prediction is the forward direction, and it is pure bookkeeping. Split the design storm's rainfall excess into D-hour pulses P1, P2, … . For each pulse, take the whole unit hydrograph, multiply every ordinate by that pulse's depth, and shift the copy so it starts when its pulse starts. Add all the copies ordinate by ordinate — that sum is the direct runoff hydrograph. Add baseflow, and the result is the predicted total streamflow hydrograph. The peak of the sum is rarely at the peak of any single copy: it falls where the overlapping copies reinforce each other, which is why the full table matters and a peak-only shortcut fails.
Changing duration: the S-curve (PE depth)
A unit hydrograph is tied to its duration D, but the S-curve lets you change it. Sum lagged copies of the D-hour unit hydrograph, each starting D hours after the last, and the sum tends to a constant discharge — the equilibrium flow for a continuous unit-rate excess. That S-curve is duration-free. To get a D′-hour unit hydrograph, take the S-curve, subtract a second copy lagged by D′ hours, and scale the difference so its volume equals one unit of depth. The exam rarely asks for a full S-curve construction, but it does test whether you know that a 2-hour unit hydrograph cannot be applied directly to a 3-hour storm without conversion.
The volume check: the one-line sanity test
Every unit hydrograph must integrate to exactly one unit of depth over the catchment: the area under the unit hydrograph (in discharge × time) must equal the catchment area times one inch, or times one centimetre. It is the fastest check in the topic — one line, no new data — and it catches scaling errors, unit slips (inches vs feet, hours vs seconds), and hydrographs that were never divided by the excess depth at all. Worked Problem 4 below runs it in SI units; run it mentally on every unit-hydrograph question you meet.
Worked problem 1 Unit hydrograph peak from an observed storm (FE pace, US units)
Given: a storm produces an observed flood-hydrograph peak of 360 cfs. Baseflow is constant at 60 cfs, and the rainfall excess for the storm is 2.0 in over the unit duration. Find the peak ordinate of the unit hydrograph.
Solution:
- Separate baseflow first: peak of the direct runoff hydrograph = \(360 - 60\) = 300 cfs. Dividing the observed peak directly would give 180 cfs per inch — the single most common error in this topic.
- Divide by the rainfall excess: peak of the unit hydrograph = \(\frac{300}{2.0}\) = 150 cfs per inch.
Answer: 150 cfs per inch of rainfall excess. Trap note: the divisor is the rainfall excess (total rainfall minus losses), not the total rainfall. If the stem gives total rainfall and a loss rate, compute the excess first — see the φ-index step in the derivation section above.
Worked problem 2 Convolution: two-pulse storm, plus baseflow (PE pace, US units)
Given: a catchment's 2-hour unit hydrograph ordinates (cfs per inch) at 2-hour intervals are 0, 100, 250, 400, 300, 180, 90, 40, 0. A storm produces 1.5 in of excess in the first 2 hours and 0.5 in in the next 2 hours. Baseflow is constant at 50 cfs. Find the direct runoff hydrograph and the peak total flow.
Solution:
- Split the excess into 2-hour pulses matching the unit duration: \(P_{1} = 1.5\,\text{in}\) \((0–2\,\text{h})\), \(P_{2} = 0.5\,\text{in}\) \((2–4\,\text{h})\).
- Scale the unit hydrograph by \(P_{1}\) with no lag: 0, 150, 375, 600, 450, 270, 135, 60, 0.
- Scale the unit hydrograph by \(P_{2}\) and lag it by one interval \((2\,\text{h})\): 0 \((at 0\,\text{h})\), 0, 50, 125, 200, 150, 90, 45, 20, 0 — the copy starts at 2 h, so its first non-zero ordinate lands at 4 h.
- Add the copies ordinate by ordinate to get the direct runoff hydrograph:
| Time (h) | 0 | 2 | 4 | 6 | 8 | 10 | 12 | 14 | 16 | 18 |
|---|---|---|---|---|---|---|---|---|---|---|
| Pulse \(1 \times 1.5\) (cfs) | 0 | 150 | 375 | 600 | 450 | 270 | 135 | 60 | 0 | 0 |
| Pulse \(2 \times 0.5\), lagged (cfs) | 0 | 0 | 50 | 125 | 200 | 150 | 90 | 45 | 20 | 0 |
| DRH (cfs) | 0 | 150 | 425 | 725 | 650 | 420 | 225 | 105 | 20 | 0 |
Answer: DRH peak 725 cfs at 6 h; peak total flow = \(725 + 50 = 775\) cfs. Trap note: the peak falls at 6 h, not at either pulse's unit-hydrograph peak alone — it is where the two scaled copies overlap most strongly. A peak-only calculation \((1.5 \times 400 = 600, or 2.0 \times 400 = 800)\) misses the lag entirely.
Worked problem 3 Deriving unit ordinates from a single-pulse storm (FE pace, US units)
Given: after baseflow separation, a storm's direct runoff ordinates at 2-hour intervals are 0, 120, 360, 480, 240, 60, 0 cfs. The rainfall excess for the storm, falling in a single 2-hour pulse, is 1.2 in. Derive the 2-hour unit hydrograph.
Solution:
- The storm is a single pulse, so every DRH ordinate is the unit ordinate times 1.2 in. Divide each ordinate by 1.2.
- Unit ordinates: \(\frac{0}{1.2} = 0\); \(\frac{120}{1.2} = 100\); \(\frac{360}{1.2} = 300\); \(\frac{480}{1.2} = 400\); \(\frac{240}{1.2} = 200\); \(\frac{60}{1.2} = 50\); \(\frac{0}{1.2} = 0\) — in cfs per inch.
Answer: 2-hour unit hydrograph ordinates 0, 100, 300, 400, 200, 50, 0 cfs per inch. Trap note: these ordinates are the input to Worked Problem 2's style of prediction only if the durations match — this is a 2-hour unit hydrograph, so it takes 2-hour pulses. Feeding it 1-hour or 3-hour pulses without an S-curve conversion is a duration error, not a rounding error.
Worked problem 4 Triangular unit hydrograph peak from the volume check (PE pace, SI units)
Given: a catchment of \(250\,\text{km}^{2}\) has a triangular 6-hour unit hydrograph with a time base of 72 h. Find its peak discharge.
Solution:
- The volume under the unit hydrograph must equal one unit of depth over the catchment: \(V = 250 \times 10^{6}\,\text{m}^{2}\) × 0.01 m = \(2500000\,\text{m}^{3}\).
- For a triangle, \(V = \frac{1}{2}\) × time base × peak, so peak = 2V / time base = \(2 \times \frac{2500000}{(72 \times 3600)} = \frac{5000000}{259200}\) = \(19.3\,\text{m}^{3}/\text{s}\).
Answer: peak of the 6-hour unit hydrograph ≈ \(19.3\,\text{m}^{3}/\text{s}\) per cm of rainfall excess. Trap note: the two unit conversions carry this problem — \(\text{km}^{2}\) to \(m^{2} (×10^{6})\) and hours to seconds \((×3600)\). Forgetting either inflates or deflates the peak by orders of magnitude, which is exactly what the volume check exists to catch.
Frequently asked questions
What is a unit hydrograph?
The direct runoff hydrograph from one unit of rainfall excess — one inch (US) or one centimetre (SI) — falling uniformly over the catchment in a specified duration D. A D-hour unit hydrograph applies only to rainfall excess split into D-hour pulses.
How is a direct runoff hydrograph built from a unit hydrograph?
By discrete convolution: split the excess into D-hour pulses, multiply the whole unit hydrograph by each pulse depth, lag each scaled copy by its pulse start time, add the copies ordinate by ordinate, then add baseflow for the total hydrograph.
How is a unit hydrograph derived from an observed storm?
Subtract baseflow from the observed hydrograph to get the direct runoff hydrograph, find the rainfall excess depth, and divide every ordinate by that depth (single-pulse storm) or deconvolve pulse by pulse (multi-pulse storm).
What must the volume under a unit hydrograph equal?
Exactly one unit of depth over the catchment area — area × one inch (US) or area × one centimetre (SI). If it does not, the hydrograph is not a unit hydrograph.
What are the assumptions behind the unit hydrograph method?
Linearity and time invariance (proportionality and superposition), uniform rainfall over the catchment, and uniform intensity during duration D. The method is a catchment-scale approximation, generally used up to about 5,000 km².
What is the difference between a unit hydrograph and an S-curve?
The hydrograph from a continuous unit-rate rainfall excess, built by summing lagged copies of the unit hydrograph. It is the tool for converting a unit hydrograph from one duration to another: subtract two S-curves lagged by the new duration and scale to one unit of depth.
Free FE/PE flashcards + study tips, by email
Join the Water Resources Hub email list and get the free 40 water-resources flashcards (PDF + Anki) straight away, then study tips and practice problems by email. Email address only — no spam, unsubscribe any time.