UPDATES: December 2010

UPDATES is a monthly email newsletter on stormwater management, assessment (including monitoring), and maintenance research at St. Anthony Falls Laboratory and the University of Minnesota.

Click HERE to subscribe to UPDATES Newsletter

Or Text SWUPDATES to 22828 to Join. Message and data rates may apply.

For Email Marketing you can trust


December 2010 (vol. 5, issue 7)

Measuring Effects of Trash and Vegetation on the SAFL Baffle

Contributed by Kurtis D. McIntire, Graduate Research Assistant, St. Anthony Falls Laboratory, University of Minnesota (Advisors: Omid Mohseni and John S. Gulliver)

Funded by the Minnesota Department of Transportation

Can a standard sump operate as a pollution prevention device? Previous St. Anthony Falls Laboratory (SAFL) research showed that standard sumps (manholes) can retain suspended solids. Unfortunately, these retained solids wash out at high flow rates, making them self-cleaning with no significant pollutant removal. However, the November 2009 UPDATES reported on research that documented a standard sump retrofitted with a SAFL Baffle will capture slightly more solids and reduce sediment washout to near-zero at high flow rates, making the standard sump an effective stormwater treatment best management practice (BMP).

This being said, the operation of sumps fitted with a SAFL Baffle when clogged with debris (trash and vegetation) is unknown. The Minnesota Department of Transportation funded a new study at SAFL to assess the effects of trash and vegetation on the performance of standard sumps retrofitted with a SAFL Baffle at low flow rates and sediment washout at high flow rates. Three laboratory test series provide the backbone for several SAFL Baffle design guidelines in this edition of UPDATES.

First off, what makes up stormwater debris? Studies of highways in Southern California (CALTRANS, 2000 and Kim et al, 2004) and with a proprietary trash collector in a Texas neighborhood (Weir et al, 2010) showed debris larger than 0.5 cm in stormwater is roughly composed of 81-90% vegetation, 5-19% trash, and 0-13% sediment. Potential cloggers of the SAFL Baffle are limited to floatable and neutrally buoyant debris, meaning they float or remain suspended in the water column. This means that debris which quickly sinks (e.g. large sediment, metal and glass trash) was not used in the following test series.

Table 1: Stormwater Debris Composition from Three Studies.

Source Trash Vegetation Sediment
CALTRANS, 2000 19% 81% 0%
Weir et al, 2010 5% 82% 13%
Kim et al, 2004 - 90% -


Figure 1: A schematic of two standard sumps retrofitted with a SAFL Baffle. Left - A 6-ft diameter and 3-ft deep "shallow" sump; Right - A 6-ft diameter and 6-ft deep "deep" sump. 

Figure 2: Basic dimensions of the 6-ft diameter and 3-ft deep sump.

The first series of laboratory washout tests was conducted on a 1:6 scale model of a 6–ft diameter and 6-ft deep sump retrofitted with a SAFL Baffle. Prior to each test, the sump was filled with a specified amount of sediment. The scale model was then fed with trash and vegetation at low flow rates until the sump appeared inundated. Vegetation was simulated at the model scale with star shaped plastic confetti, and trash was simulated with cut wood dowels. Despite clogging from debris, little washout of sediment was recorded at the end of each test. This indicates that deep sumps retrofitted with the SAFL Baffle should experience little washout of sediment at high flow rates.

Figure 3: A 1:6 scale model of a deep sump retrofitted with a SAFL Baffle fed with simulated trash and vegetation. The inlet pipe is shown on the right.

Surprised by this result, we decided to test the extreme limits of the baffle: its ability to operate with plastic bags in the flow in a shallow sump. For this second testing series, washout tests were conducted on a full scale (6-ft diameter, 3-ft deep) shallow sump equipped with a 3 inch hole diameter SAFL Baffle. The full scale testing was conducted following the methodology developed for the 1:6 scale model. Vegetation and trash were simulated using debris combinations of leaves, small plastic grocery bags, and/or plastic bottles.

Two of the three tests in this series included an extreme number of plastic bags in the debris combination. During these tests, plastic bags quickly clogged the SAFL Baffle (see Figure 3). As a result, the sump's water flow patterns changed dramatically - from traveling through the baffle to traveling under the baffle. This water flow pattern traveled closely to the sediment bed at the bottom of the sump, and washout was visible. It is possible to clog the SAFL Baffle with an extreme number of plastic bags.

Figure 4: A 6x3 sump and 3” hole diameter SAFL Baffle inundated with plastic grocery bags and plastic bottles. The view is from the upstream side of baffle after the draining the sump.

Test number 3 of this series utilized leaves as the only form of debris entering the shallow sump. Leaves clogged the baffle to a lesser degree than plastic bags, so flow patterns were split between traveling through the baffle and traveling under the baffle (see Figure 5). Preliminary results in Table 2 show that stormwater debris, especially an extreme number of plastic bags, can clog a SAFL Baffle and cause significant washout in shallow sumps experiencing high flow rates.

Figure 5: A 6x3 sump and 3” hole diameter SAFL Baffle inundated with leaves (view from the downstream side of baffle).

Table 2: 6-ft x 3ft, 3" SAFL Baffle Hole - Shallow Sump Washout Test Series. An effluent concentration of 100 mg/L is the median background level of urban runoff.

Test # Bags Added Bottles Added Leaves Added Flow Rate (cfs) Effluent Concentration (mg/L)
1 - 7.0 1,305
2 6.9 978
3 - - 6.7 434


In the third test series, a washout test series was conducted on a 1:6 scale model of a 6×3 ft sump to find ways to reduce washout in shallow sumps. These tests were conducted similarly to the previous 1-ft by 1-ft (i.e. a 1:6 scale model of a 6×6 ft sump) scale model tests. However, debris loading amounts were varied, flow rates were approximately the same between tests, and an alternative SAFL Baffles design was tested. Table 3 shows that sediment washout is dependent on the amount of debris loaded and the size of the SAFL Baffle's holes.

Table 3: 1:6 scale model of a 6x3 ft sump - Shallow Sump Washout Test Series. Baffle hole dimensions are given for full scale.

Test # Baffle Holes (inches) Debris Loading (%) Gross Pollutants Load (g) Effluent Concentration (mg/L)
1 3 100 319 157
2 3 0 0 6
3 3 50 163 27
4 4 78 100 314 12


Despite the washout caused by debris in shallow sumps, they can still act as effective stormwater BMPs. By implementing one or more of the following solutions, washout caused from debris will decrease:

  1. Using a SAFL Baffle with larger hole openings (5” instead of 3" or 1")
  2. Installing another device at an upstream manhole or catch basin to collect trash
  3. Scheduling maintenance during the fall season

Further testing will be conducted to assess the effects of trash and vegetation on the baffle’s ability to capture suspended sediments at low flow rates.

Supplementary Materials

  1. November 2009 UPDATES Newsletter
  2. Videos: www.youtube.com/kurtmcintire


  • Weir, Bob, Steve Esmond, and Tom Harris. "Separation and Handling of Gross Solids and other Pollutants with Minimal Maintenance." 2010.
  • California Department of Transportation, “CALTRANS District 7 Litter Management Pilot Study.” 26, Jun. 2000.
  • Kim, Lee-Hyung, Masoud Kayhanian, and Michael K. Stenstrom. "Event Mean Concentration and Loading of Litter from Highways During Storms." Science of the Total Environment 330 (2004): 101-113.


We want to hear from you!!!

Let us know your thoughts, experiences, and questions by posting a comment. To get you thinking, here are a few questions: 

  • Do you think this type of testing could be performed on other types of stormwater treatment devices?