18-Env-A2 Hydrology and Municipal Hydraulics Engineering · December 2013
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
National Exams — December 2013 — 04-Env-A2 / Hydrology and Municipal Hydraulics Engineering. 3 hours duration; closed book with a candidate-prepared 8½×11 in double-sided aid sheet; Casio or Sharp approved calculator only. Any five questions constitute a complete paper (first five answers marked, 20 marks each, 100 marks total); all seven are solved below for completeness.
Reference texts. Davis & Cornwell, Introduction to Environmental Engineering (6th ed.) — hydrology, stormwater management and water-demand chapters; Metcalf & Eddy, Wastewater Engineering: Treatment and Resource Recovery (5th ed.) — sanitary sewer hydraulics, Manning/Harmon design formulas; MWH’s Water Treatment: Principles and Design (3rd ed.) — distribution systems, pipe-network analysis and pump selection; Chow, Open-Channel Hydraulics — Manning's n tables and specific-energy theory; Chow, Maidment & Mays, Applied Hydrology — flood-frequency analysis; CCME water quality guidelines — cold-water fishery thermal protection.
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.
Three reasons: (1) peak attenuation — storing runoff temporarily and releasing it slowly through a metered outlet directly reduces the peak discharge reaching the receiving channel, which is the single largest driver of downstream flooding and channel erosion; (2) gravity settling time — holding water (even briefly in a dry pond, longer in a wet pond's permanent pool) gives suspended sediment and particulate-bound pollutants time to settle out under gravity before the water is released, directly improving effluent quality; and (3) thermal and biological buffering — a held volume has time to equilibrate somewhat toward ambient temperature and allows some biological uptake of nutrients (in a vegetated wet pond), both of which are lost if runoff is conveyed directly to the receiving watercourse with no retention time at all.
On-site system — permeable pavement / bioretention cell. Two routine maintenance measures: (1) periodic vacuum sweeping or surface cleaning to prevent fine sediment and debris from clogging the infiltration surface, which is the single most common cause of permeable-system failure; and (2) inspection of the underdrain/overflow structure after major storms to confirm it is not blocked, since a clogged underdrain can cause the cell to surcharge and fail exactly when it is needed most. Off-site system — regional wet detention pond. Two routine maintenance measures: (1) periodic sediment/bathymetric survey and dredging of the permanent pool (typically every 10–20 years) to restore the design treatment volume as accumulated sediment reduces it; and (2) outlet structure (riser/orifice, trash rack) inspection and debris clearing, since a blocked low-flow orifice defeats the pond's peak-attenuation function even though the pond itself appears to be working.
Given/Find. How a flood-frequency curve is generated, an example of its use, and two assumptions underlying its application.
Generation. The annual-maximum instantaneous (or daily) streamflow is extracted from the gauge record for each year on file, giving an annual-maximum series. This series is fit to a probability distribution commonly used for hydrologic extremes — typically Log-Pearson Type III (Environment Canada/USGS standard practice) or the Gumbel distribution — using the method of moments (mean, standard deviation and, for LP3, skew of the log-transformed data). The fitted curve is plotted against return period $T$ (or exceedance probability, via a plotting-position formula such as Weibull's $P=m/(N+1)$) on log-probability paper, and a confidence band (e.g., 90%) is added to show the sampling uncertainty around the fitted line at each return period, which visibly widens at longer return periods where extrapolation beyond the gauged record is greatest.
Example use. A bridge or culvert crossing a stream is designed to safely pass the flow associated with a specified return period (e.g., the 100-year flood for the structure's hydraulic opening, checked against the 500-year event for overtopping/scour risk); the engineer reads the design streamflow directly off the fitted curve at the required $T$, rather than needing a flow of that magnitude to have actually been observed in the (much shorter) gauge record.
Two assumptions to consider. (1) Stationarity — the method assumes the statistical properties of the annual-maximum series (mean, variance) do not change over time, an assumption increasingly violated by upstream land-use change (urbanization) and a shifting climate, which can make historical curves under-predict future flood magnitudes. (2) Record length vs. extrapolation target — a curve fit to, say, 30 years of record is being used to estimate a 100- or 500-year event well beyond the observed range; the confidence band widens accordingly, and the designer must recognize the quoted design flow carries substantially more uncertainty than the same-looking number for a return period close to the record length. In practice a designer facing both limitations at once — a short, possibly non-stationary record — should treat the upper end of the confidence band, not just the fitted line, as the governing design value for a critical structure, and should periodically refit the curve as additional years of gauge record become available rather than relying indefinitely on an analysis performed once at the start of a project.
