Go to: OVERVIEW (SaveOurCounty)     DETAILS (listener)     PLANNING     SCHOOLS     ENVIRONMENT     EROSION     Report corrections & broken links to Webmaster     Get updates on local issues

 

Rain, Rain, Go Away: Rain Leaks into Sanitary Sewers

 

Rain and ground water can leak into sanitary sewers through cracks, breaks, and illegal connections from gutters. Treating large flows of rain water raises the cost of operations and the stress on treatment plants and operators. Rain leaking in also means there is real risk that raw sewage can leak back out into the ground water through the same cracks in pipes and fittings, whenever a local water table drops below the sewer pipes.

 

There is a low cost way to estimate the total rain leaking into a system. It uses existing daily flow records, daily rain data already collected by others, and a spreadsheet. The results show if there are major leakage problems to find and fix, and may motivate people to solve the problems.

 

Table 1 was calculated by putting into a spreadsheet the daily flows of each sewage plant and daily rain nearby. Shepherdstown has a rain gauge at the sewer plant, on one edge of the sewage collection area. For Charles Town, the nearest gauge is at an elementary school three miles outside the edge of the sewage collection area. Most parts of the country have a variety of rain gauges on the web, so one is usually not far away.

 

Table 1. Effect of Rain on Sewage Flows

 

Shepherdstown, West Virginia, May-November 2004

 

Daily Flow = 266,000             average gallons per day, not including rain

               + 88,000 gallons per inch of rain falling in the same day (standard error = + 11,000 gallons)

               + 37,000 gallons from each inch of rain which fell in the previous day ( + 11,000 gallons)

               + 23,000 gallons from each inch of rain which fell two days before ( + 11,000 gallons)

 

Total effect in Shepherdstown is 148,000 gallons of extra flow per inch of rain, 56% above normal flow.

 

Charles Town, West Virginia, July-October 2004

 

Daily Flow = 1,005,000          average gallons per day, not including rain

             + 152,000 gallons per inch of rain falling in the same day (standard error = + 16,000 gallons)

               + 36,000 gallons from each inch of rain which fell in the previous day ( + 16,000 gallons)

               + 35,000 gallons from each inch of rain which fell two days before ( + 16,000 gallons)

 

Total effect in Charles Town is 223,000 gallons of extra flow per inch of rain, 22% above normal flow.

 

A @regression@ is a tool, common in spreadsheet and statistical packages, which looks at the data and finds the best possible formula to relate rainfall (independent variable) and daily wastewater flow (dependent variable). These examples look at rainfall from the same day and the two previous days, since it takes time for water to flow into the system.

 

The calculation is not exact of course, since flow depends on many factors besides rain: commuting patterns, school and business schedules, etc. The goal here is to see how much of the flow does depend on rain, and it is substantial. A big rainfall clearly adds a lot to sewage flows, both on the same day and on later days, as the water leaks into the sewer pipes.

 

Any wastewater system can do this analysis, since daily flow is already recorded, and there are widespread rain gauges on the web. You can get rainfall for the last year from a growing free network of gauges at http://www.wunderground.com/weatherstation/ListStations.asp. A bigger free network, in nearly every zip code, is at http://www.instaweather.com/aws, though it only gives 30 days at a time, so you have to download data monthly. The government network at http://www.noaa.gov/climate.html is smaller and costs a little; it is sometimes more reliable, and you can get data as far back as you want.

 

A short-cut approach is to compare rainfall and daily flows for just a few days before and after a big rainstorm. Table 2 shows this approach. This is faster than entering long periods of daily  flows and rainfall.

 

Table 2. Shepherdstown before and after a Big Rainfall

 

Daily Rain (Inches)

Daily Sewage (Gallons)

Days with Little Effect from Rain (Average =

 251,600 gal.)

Above Average Flow from Rain

Extra Flow per Inch of Rain (based on 2.39" rain)

 

 

 

 

 

 

Tue 6/1/04

0.02

270,200

U

 

 

Wed 6/2/04

0.03

174,500

U

 

 

Thu 6/3/04

0

363,700

U

 

 

Fri 6/4/04

0.03

289,600

U

 

 

Sat 6/5/04

2.39

608,000

 

356,400

149,000

Sun 6/6/04

0

414,000

 

162,400

68,000

Mon 6/7/04

0.01

328,000

 

76,400

32,000

Tue 6/8/04

0

216,000

U

 

 

Wed 6/9/04

0

195,600

U

 

 

 

Total effect is 249,000 gallons per inch of rain, 99% above normal flow.

 

Shepherdstown had 2.39 inches of rain on June 5, and sewage flows doubled that day. Average flow on 6 days which were not much affected by rain was 251,600 gallons per day. Compared to this average, the total extra flow on June 5-7 was 595,200 gallons, or 249,000 gallons per inch of rain. Table 2 shows the same kind of situation as Table 1, and the numbers differ only because Table 1 is an average based on seven months of data, and Table 2 is based on nine days.

 

After estimating how much of your flow comes from rainfall, you will want to find and fix the spots where rain enters the collection system. This search can involve smoke tests, remote cameras, and flow meters around the system. If your system has flow meters recording flows in different parts of the system, you can compare those daily flows to the same rainfall data, to estimate how much infiltration is in each part of the system.

 

The method in this paper estimates extra inflows which are related to rain. It may underestimate if there are steady leaks where a cracked pipe is below the water table, and it may overestimate if there are major customers who send more sewage in rainy weather. The method is more accurate with rain than snow, since snow melts unpredictably, and may reach sewers many days after it falls. It is also more accurate if it includes a day or two with rain over an inch, to see a big effect on the sewage flow.

 

The method in this paper does not show where rain comes in, but it may motivate people to find and fix those leaks. Every gallon costs money to process. Large rain loads also force a plant to expand before it would otherwise need to, and rainfall does not pay sewer bills.