Worksheet Grade 12 Science

Flood Frequency and Culvert Sizing on Tullimore Creek

A grade 12 science worksheet using 20 years of peak-flow records to rank floods, find return periods and exceedance risk, and size a replacement culvert.

Objectives

  1. Students will rank an annual peak-flow record and compute return periods using a plotting-position formula.
  2. Students will convert an annual exceedance probability into the risk of at least one exceedance over ten years.
  3. Students will size a replacement culvert from a design flow and a maximum allowable velocity.

Vocabulary

Key words: annual peak flow, return period, exceedance probability, plotting position, design flow, continuity equation

Teacher overview

Winford is a fictional town whose main road crosses Tullimore Creek on a culvert installed decades ago. The road floods every few winters, and the town has twenty years of annual peak-flow records for the creek. This worksheet has students rank that record, convert ranks into return periods and exceedance probabilities, and then size a replacement culvert from a stated design flow and a velocity limit.

The recurring misreading here is treating a return period as a schedule. A twenty-one-year flow is not a flow that arrives once every twenty-one years in an orderly rotation; it is a flow with about a 4.8 percent chance of being exceeded in any single year, which is why item 5 asks students to convert the return period into a probability over the life of a structure.

Answer Key

  1. Answer: Rank 3.
    Sorted from largest to smallest, the record runs 255, 220, 204, 185, 176, 159, 154, 140, 131, 118, 113, 103, 99, 89, 83, 72, 67, 61, 51, 48. The 204 cfs peak from 2008 sits third.

  2. Answer: 21 years.
    The largest peak is 255 cfs at rank m = 1. T = (20 + 1) / 1 = 21 years.

  3. Answer: 4.8 percent. p = 1 / 21 = 0.0476, which is 4.8 percent when rounded to one decimal place.

  4. Answer: 2.63 years.
    The 140 cfs peak sits at rank m = 8. T = 21 / 8 = 2.625, which rounds to 2.63 years.

  5. Answer: 38.6 percent.
    R = 1 - (1 - 0.0476) raised to the tenth power. (0.9524) squared is 0.90707, raised to the fourth is 0.82278, raised to the eighth is 0.67697, and multiplying the eighth power by the second gives the tenth power, 0.61405. R = 1 - 0.61405 = 0.38595, or 38.6 percent.

  6. Answer: 24.17 square feet.
    A = Q / V = 290 cfs / 12 ft/s = 24.1667 square feet, which rounds to 24.17 square feet.

  7. Answer: 5.55 feet.
    D squared = 24.1667 / 0.785 = 30.786. The square root of 30.786 is 5.548, which rounds to 5.55 feet.

  8. Answer: 6 feet, giving 10.26 ft/s.
    The next whole-foot size above 5.55 feet is 6 feet. Its area is 0.785 x 6 x 6 = 28.26 square feet, so V = 290 / 28.26 = 10.262 ft/s, which rounds to 10.26 ft/s. That is under the 12 ft/s cap.

  9. Answer: 187.5 MGD.
    290 cfs / 1.547 cfs per MGD = 187.46 MGD, which rounds to 187.5 MGD. The figure is large because a flood peak is many times a town's daily water use.

  10. Answer: 150.7 cfs.
    A = 0.785 x 4.0 x 4.0 = 12.56 square feet. Q = 12.56 x 12 = 150.72 cfs, which rounds to 150.7 cfs.

  11. Answer: 7 peaks.
    The peaks above 150.7 cfs are 255, 220, 204, 185, 176, 159, and 154, which is seven of the twenty years.

  12. Answer: 2.9 years.
    20 years of record / 7 exceedances = 2.857 years per exceedance, which rounds to 2.9 years. On that record the crossing is overtopped roughly every three years.

  13. Answer: Twenty years of record supports return periods no longer than 21 years, so a 1 percent flow has to be extrapolated.
    The plotting-position formula assigns the longest return period to the largest observed peak, and with n = 20 that ceiling is (20 + 1) / 1, or 21 years. A flow with a 1 percent annual chance has a 100-year return period, far outside the record. The formula can rank and space the observed floods and give a defensible estimate of frequent events, yet estimating a 100-year flow requires fitting a statistical distribution to the record or borrowing data from similar gauged watersheds, and the result carries wide uncertainty.

  14. Answer: The existing crossing floods about every three years, while the replacement handles a flow the record has never reached.
    Item 12 shows the current culvert is overtopped on average every 2.9 years, so the road closures and repair costs recur every few years. Item 5 shows that even the largest peak in the record has a 38.6 percent chance of being matched or beaten within ten years, and the 6-foot culvert passes the 290 cfs design flow with velocity to spare. Over the life of the crossing, one larger investment can cost less than repeated repairs and road closures.

Differentiation

Support

Give students the fictional Tullimore Creek peaks already sorted from largest to smallest with a blank rank column, so ranking is not the barrier to the return-period work. Work item 2 together, then have students complete items 3 and 4 before continuing. For items 6 through 8, supply a worked example using a different design flow so the sequence of area, then diameter, then rounded size, then recomputed velocity is modeled before students try it.

Extension

Have students plot the twenty peaks against their return periods on semi-log paper, fit a straight line by eye, and read off an estimate for the 50-year flow in the fictional Winford record. Ask them to compare that estimate with the 290 cfs design flow and to write two sentences on how much confidence the extrapolation deserves.

ELL supports

Provide a term card that separates the three ideas students most often merge: rank, return period, and probability, each with one example drawn from the table. Give a sentence frame for items 13 and 14, such as "The record supports return periods up to ___ years because ___." Allow the item 14 response as a short list of reasons with numbers attached rather than a paragraph.

Extensions

Students can pull the annual peak-flow record for a real gauged stream near the school from the United States Geological Survey and repeat items 1 through 5 on that record, then compare the length of the real record with the twenty years available in the fictional Winford case. A second option is to have students calculate the risk of at least one exceedance across a 50-year culvert design life rather than ten years, and explain why designers of long-lived structures use the longer window.