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.
Stream stage is the elevation of the water surface at a gauging section, measured above an arbitrary but fixed local datum. It is the quantity that can be recorded continuously, cheaply and reliably — by a staff gauge read manually, or by a float-and-counterweight recorder, a pressure transducer or a bubbler in a stilling well — and it is what a Water Survey of Canada gauge actually transmits.
A rating curve is the calibrated relationship between stage and discharge at that same section, established by making a set of direct discharge measurements (current meter, acoustic Doppler profiler, or dilution gauging) across the full range of flows and fitting a curve through the stage–discharge pairs. The usual fitted form is a power law,
$$Q = a\,(H - H_0)^{b}$$where H is the recorded stage, H0 is the stage of zero flow (the effective datum of the control), and a and b are fitted coefficients whose values reflect the geometry and roughness of the control section.
The two terms are related as measurement and transfer function: stage is measured, discharge is inferred. Because a continuous stage record can be converted through the rating into a continuous discharge record, the pair is what makes long-term streamflow archives such as HYDAT possible at all. The accuracy of every derived quantity — annual runoff volume, flood frequency curve, low-flow statistic — is limited by the accuracy of the rating, particularly at the extrapolated high-flow end where direct measurements are scarcest and the design flood lies.
A common environmental phenomenon that degrades the rating is the formation of an ice cover in winter, which is the dominant problem on Canadian rivers. An ice cover adds a second boundary with its own roughness, converts a free surface into a partially pressurised conduit, and backs water up, so the same discharge is conveyed at a substantially higher stage; applying the open-water rating then over-estimates the flow, sometimes by a factor of two. Water Survey of Canada therefore flags winter records as ice-affected and estimates them from periodic under-ice measurements rather than from the rating. The same class of error is produced by any process that shifts the control: scour or deposition at the control section after a flood, the growth or cutting of aquatic weed through the season, a downstream backwater from a tributary, a tide or a reservoir, or debris and beaver activity lodged on the control. All of them mean a rating curve is a perishable calibration that must be re-verified, not a permanent property of the site.
First, precipitation is spatially variable and the gauge samples an area of about 0.02 m2. A standard 200 mm-diameter gauge orifice collects from roughly one part in 1011 of a 1000 km2 catchment, so it is an extraordinarily small sample of a field that has real structure. Convective summer storms have cell diameters of a few kilometres and rainfall depths that can vary by a factor of five within that distance; frontal systems are more uniform but still show organised bands. Over a catchment with relief, orographic enhancement makes the windward slopes systematically wetter than the valley floor where a gauge is easiest to service, so the single measurement is not merely noisy but biased. Multiplying one point depth by the catchment area therefore returns a volume that can be wrong by a large factor, and in the wrong direction consistently.
Second, the gauge measurement is itself biased low by exposure and instrument effects, and the bias grows with wind speed and with snow. A gauge orifice is an obstruction that accelerates the airflow over it, deflecting the smaller drops and, especially, the snowflakes past the opening; undercatch of 5–15 % for rain and 20–50 % or more for unshielded snow is routine, which is why Nipher or Alter shields are standard on Canadian gauges. Wetting losses on the funnel walls, evaporation between readings, splash-out, and the systematic under-recording of a tipping bucket at high intensity (water is lost while the bucket is in motion) all act in the same direction. So even where the point value is spatially representative, it under-states the depth that fell, and the error is largest in exactly the high-intensity or heavy-snowfall events that matter for design.
Two further considerations reinforce the point in practice: the gauge network is usually sited for accessibility rather than for hydrological representativeness, and a single gauge gives no information at all about the movement of the storm across the basin, which is what determines whether sub-catchment peaks coincide. Together these are why areal-averaging methods, and increasingly gauge-adjusted radar, are used instead of a single point value.
Method 1 — Thiessen polygons. The Thiessen construction assigns to each gauge the portion of the basin that lies closer to it than to any other gauge, and weights the gauge depths by those areas. Applied to the printed figure: join S1–S2, S2–S3, S3–S4 and S4–S1 with straight lines, construct the perpendicular bisector of each connecting line, and extend the bisectors until they intersect one another and the basin boundary. The result is the polygon network sketched in panel (a) — four cells, one per gauge, whose boundaries are equidistant from the two gauges they separate. Planimeter or digitise the area Aj of the part of each polygon that falls inside the watershed divide, and form
$$\bar{P} = \frac{\sum_{j=1}^{n} A_j\,P_j}{\sum_{j=1}^{n} A_j} = \sum_{j=1}^{n} w_j\,P_j , \qquad w_j = \frac{A_j}{A}$$where Pj is the depth at gauge j and the weights sum to unity. The strengths are that the weights are objective, reproducible by any analyst and computed once for a fixed network, and that gauges lying outside the divide can still be used because their polygons may extend into the basin — something the arithmetic average cannot do. The limitations are that the method is purely geometric: it takes no account of topography or of storm orientation, it assumes the depth is constant over each polygon so it cannot represent a gradient, and the whole network must be recomputed whenever a gauge is added or a record is missing. For the figure shown, S1 and S4 command the western half of the basin and would dominate the estimate regardless of where the storm centre actually was.
Method 2 — the isohyetal method. Here the analyst plots the measured depths at their true locations, then draws isohyets — contours of equal precipitation depth — by interpolating between gauges, using knowledge of the storm type and of the terrain to guide the contouring rather than interpolating blindly. In panel (b) the isohyets have been drawn to reflect a storm centred to the north-east, so the 60 mm contour wraps around S3 and the depth falls toward the south-west. The area Aj between each adjacent pair of isohyets is measured, the depth over that strip is taken as the mean of the two bounding isohyets, and
$$\bar{P} = \frac{\sum_{j} A_j \left(\dfrac{P_j + P_{j+1}}{2}\right)}{A}$$with the outermost strips handled by judgement (commonly the mean of the bounding isohyet and the extreme measured value). The isohyetal method is generally the most accurate of the four, because it is the only one that lets physical understanding — orographic enhancement, the track of a convective cell, the sheltering of a lee valley — enter the estimate, and it represents gradients rather than steps. Its costs are that it is labour-intensive, that it must be redone for every storm rather than once for the network, and that it is subjective: two competent analysts contouring the same four points can produce estimates that differ by several per cent, and with a sparse network the drawn contours are largely the analyst’s hypothesis rather than data.
For completeness, the arithmetic average, P̄ = (1/n)∑Pj, is the fastest method and is defensible only when the gauges are evenly distributed over flat terrain and the individual depths differ little from the mean; applied to the figure it would weight S1 through S4 equally even though S2 and S3 sit in the eastern lobe. The grid method superimposes a regular mesh over the basin, estimates the depth at each node by inverse-distance or nearest-neighbour interpolation from the surrounding gauges, and averages the nodal values; it is the natural choice for automated processing and for merging gauge data with radar or reanalysis fields, and it is the form in which most modern GIS-based hydrologic models actually do the calculation.
