16-Civ-B4 Engineering Hydrology · May 2013
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Paper format. National Examinations, May 2013 — 98-Civ-B4 Engineering Hydrology. Three hours, closed book, one candidate-prepared two-sided 8½″ × 11″ aid sheet, and one approved Casio or Sharp calculator whose model must be declared. Seven problems are printed; page-1 Note 4 states that any five (5) questions constitute a complete paper and that only the first five answers appearing in the workbook are marked. Each problem carries twenty (20) points, so the examinable total is 5 × 20 = 100 points. The page-7 marking scheme breaks each problem into its sub-parts. All seven problems are solved here, because this set is a study resource rather than a timed sitting; the sub-part mark values shown below are the printed ones.
Reference texts. V. T. Chow, D. R. Maidment and L. W. Mays, Applied Hydrology (hydrologic cycle, unit hydrographs, routing, frequency analysis, infiltration); L. W. Mays, Water Resources Engineering, 3rd ed. (design application, rainfall–runoff, groundwater); W. Viessman and G. L. Lewis, Introduction to Hydrology, 5th ed. (measurement, areal precipitation, snowmelt energy budget); V. T. Chow, Open-Channel Hydraulics (1959) (flood-wave propagation, gradually varied unsteady flow); C. W. Fetter, Applied Hydrogeology, 4th ed. (Darcy’s law, hydraulic conductivity, transmissivity). Canadian practice references: Environment and Climate Change Canada / Water Survey of Canada HYDAT archive and the ECCC Engineering Climate Datasets (IDF curves and the IDF_CC climate-adjustment tool), and the Transportation Association of Canada Guide to Bridge Hydraulics, 2nd ed.
Check — conventions used throughout this paper. A hydrologic year is taken as 365 days = 31 536 000 s unless a question says otherwise. Water density is 1000 kg/m3 and the latent heat of fusion of ice is 334 kJ/kg. Where the printed data are internally inconsistent — and Question 7(i) is such a case — the inconsistency is demonstrated arithmetically, the governing conservation requirement is stated, and the corrected reading actually used is declared at the point of use, as page-1 Note 1 invites (“the candidate is urged to submit… a clear statement of any assumptions made”).
Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.
A conceptual model represents the watershed as a small set of idealised storage elements and transfer functions whose parameters are lumped over the whole basin and calibrated against observed flow. The unit hydrograph, the linear-reservoir cascade, the Muskingum routing equation and the rational method are all conceptual models: they encode a physically motivated idea — that a basin behaves like a series of leaky reservoirs, or that runoff response is linear and time-invariant — but they do not solve the equations of motion. Their parameters (K, X, the time of concentration, the runoff coefficient) are effective quantities that absorb everything the model does not represent explicitly.
A numerical model solves the governing partial differential equations — continuity and momentum for surface flow, Richards’ equation for unsaturated flow, the groundwater flow equation for the subsurface — on a discretised grid or network, using finite-difference, finite-element or finite-volume schemes. Its parameters are, at least in principle, measurable physical properties: bed slope, Manning roughness, cross-section geometry, saturated conductivity.
The contrasts that matter to an engineer are these. Conceptual models are cheap, transparent, robust with sparse data, and defensible in front of an approving authority, but they are only valid inside the range of conditions over which they were calibrated, they return a single lumped hydrograph rather than a spatial field, and they cannot represent backwater, reverse flow or a structure failure. Numerical models are spatially distributed, can represent structures and boundary conditions explicitly, and can be extrapolated beyond calibration on physical grounds, but they demand detailed topography and roughness data, are expensive to build and to run, and can be numerically unstable if the space and time steps violate the Courant condition. In practice the two are complementary: a conceptual model is used to generate the inflow hydrograph, and a numerical model routes it through the reach of interest.
A concrete case where the numerical model is required rather than preferred: a stormwater outfall discharging into a tidal river through a highway embankment. The peak of the catchment hydrograph may coincide with high tide, so the outfall runs submerged and the water surface in the trunk sewer is controlled from downstream. A conceptual model that pushes a hydrograph through a fixed rating curve has no mechanism to represent this — it assumes the outflow depends only on the upstream state. A one-dimensional unsteady solution of the Saint-Venant equations, with the tidal stage as the downstream boundary condition, reproduces the backwater, the reduced conveyance and the resulting surcharge, and shows whether the system floods. The same argument applies to any dam-break, levee-overtopping or flood-plain-storage problem, and to the routing of a hydrograph through a looped or pressurised pipe network.
A unit hydrograph of duration D for a given watershed is the direct-runoff hydrograph that results from one unit depth of effective rainfall (1 mm, or 1 cm, or 1 inch, depending on the units declared) generated uniformly over the whole watershed at a uniform rate over the duration D. It is therefore a normalised response function of the basin: the direct-runoff hydrograph produced by any other storm of the same duration is obtained by scaling the ordinates in proportion to the actual excess depth, and by a storm of several successive blocks of excess by superposing the scaled and lagged responses. The convolution statement is
$$Q_n = \sum_{m=1}^{n \le M} P_m\, U_{n-m+1}$$where Pm is the excess-rainfall depth in the mth block of duration D, U are the unit-hydrograph ordinates, and Qn is the direct-runoff ordinate at time n. Note that the unit hydrograph describes direct runoff only — baseflow must be separated out before it is derived and added back afterwards.
Three rules must be respected when deriving one for a particular watershed.
Rule 1 — the storms used must be simple, isolated and uniform. The derivation should use single-peaked events, produced by a storm whose rainfall excess is reasonably uniform over the watershed area and reasonably uniform in time over the chosen duration, with a well-defined start and a clean recession that is not contaminated by a second burst. Duration should be roughly one-quarter to one-third of the basin lag; a storm much longer than that violates the uniformity assumption, and one much shorter yields ordinates dominated by measurement noise. Several such events should be analysed and their unit hydrographs averaged — by averaging the peak and the time to peak first, then sketching a curve of correct volume through them, never by averaging ordinates directly, which flattens the peak.
Rule 2 — the volume under the unit hydrograph must equal one unit of runoff depth over the watershed area. This is the normalising condition and it is also the arithmetic check: the area under the derived curve, divided by the watershed area, must return exactly the unit depth. After baseflow separation and after any smoothing of the recession, the ordinates must be rescaled so the condition is met, otherwise every subsequent design flow inherits a volume error.
Rule 3 — the linearity and time-invariance assumptions must actually hold, and the duration must be stated. A unit hydrograph presumes that response is proportional to input (double the excess, double every ordinate) and that the base length is independent of the excess intensity. That assumption degrades badly if the watershed is large — above roughly 5000 km2 a single storm cannot be uniform, and the basin should be subdivided — or if the events span a wide range of magnitudes, since large events run off faster than small ones. It also fails if the land use, channel geometry or storage has changed between the calibration period and the design application, so a unit hydrograph derived before urbanisation cannot be applied afterwards. Finally, a unit hydrograph is meaningless without its duration attached: a 1-hour and a 6-hour unit hydrograph for the same basin are different curves, and converting between them requires the S-curve method or Snyder’s synthetic relations.
[Figure not reproduced: Figure 2.1 — Redrawn from the diagram printed with the question: the hyetograph is split into losses and rainfall excess, and the resulting hydrograph shows the rising limb, crest segment and recession curve above the separated baseflow. See the official exam paper.]
The peak discharge and the time to peak. The peak ordinate Qp is the flow the conveyance must pass, and the time to peak Tp — measured from the centroid of the rainfall excess to the peak — is what allows the designer to decide whether two sub-catchments will peak together. For a stormwater collection system the peak sets every pipe diameter, inlet capacity and culvert size through the continuity and normal-depth calculation; for a flow-control structure it sets the maximum stage and therefore the freeboard. The pairing of the two is what makes the diagram useful: a catchment with a short time to peak produces a tall, narrow hydrograph, which is severe for conveyance but easy to control with a small amount of storage, whereas a long time to peak gives a low, broad hydrograph that is easy to convey but expensive to detain. The steepness of the rising limb also fixes the closure rate a gate or valve must achieve if the structure is actively operated.
The volume under the hydrograph — the area between the curve and the separated baseflow. Detention and retention facilities are sized by volume, not by peak. Once the outflow rating of the control structure is fixed (by the orifice and weir geometry), the required storage is the maximum accumulated difference between the inflow hydrograph and the outflow hydrograph, so the whole area under the curve matters, including the long tail of the recession. This is why a design that limits the post-development peak to the pre-development peak may still be inadequate: urbanisation typically increases volume as well as peak, and the extended release of that extra volume can cause downstream channel erosion even though the peak criterion is satisfied. In Canadian practice this is the reason quantity control is now routinely paired with an extended-detention or erosion-control volume criterion.
The recession curve and the base length. The recession represents the drainage of storage — first channel storage, then interflow, then groundwater — and it decays approximately exponentially, Q(t) = Q0kt, with a recession constant that is a stable property of the basin. Its practical significance is threefold: it fixes how long a detention pond stays full and therefore whether it will have recovered its storage before the next storm of a multi-day event arrives; it sets the duration of the erosive flow that the outlet channel and its bank protection must survive; and it defines the base length that must be respected when hydrographs from several sub-catchments are combined, since two peaks that arrive at different times may still overlap on their recessions and produce a downstream flow larger than either alone. The point at which the recession rejoins the baseflow line is the end of direct runoff and is where baseflow separation is performed before any unit-hydrograph work.
