PAM + No Treatment PAM
Gypsum
Erosion plots after very wet rainfall simulator runs in summer 1998 study,
showing differences in surface condition and rilling between the treatments. Rills
were painted for greater clarity.
Erosion Control Study Using Polyacrylamide (PAM)
and PAM+Gypsum on a Steep Slope
Dennis Flanagan, Kiran Chaudhari, Darrell Norton
Objectives:
PAM and PAM+Gypsum have been shown in previous studies (mostly on lower
agricultural slopes) to improve infiltration, decrease runoff, and decrease soil loss. The
objectives of this study were to determine the effectiveness of PAM (P) and
PAM+Gypsum (PG) in reducing runoff and soil loss on very steep slopes typical of
construction sites, landfills, mines, etc. Overall objectives of this study (in combination
with 2 natural rainfall sites) were to examine the use of PAM and PAM+Gypsum as
erosion control methods during critical period of vegetation establishment on plots, and
determine their effect on both erosion control and grass establishment.
Treatments:
- Control
- PAM applied at 80 kg/ha in a liquid solution Estimated cost $529/ha
- PAM applied at 80 kg/ha in a liquid solution
plus Gypsum applied broadcast dry at 5 Mg/ha Estimated cost $579/ha
Site Characteristics:
Slope: 32%
Soil: Silt loam topsoil over gravel subsoil
Plots: 3 m x 9 m, 3 replications
Rainfall Simulator Study in 1998:
Dry run: 64 mm/h for 1 hour
Wet run: 64 mm/h for 1 hour
Very wet run:
64 mm/hr for 15 minutes
28 mm/hr for 15 minutes
100 mm/hr for 15 minutes
Results:
PAM and PAM+Gypsum treatments significantly reduced sealing, runoff, and sediment
loss. Rill formation on the treated plots was virtually eliminated.
Treatment Total % Total Sediment %
Runoff reduction Loss (Mg/ha) reduction
(mm)
Dry Run Control 41.5 a 76.3 a
PAM 5.9 b 85.7 1.0 b 98.7
PAM+Gypsum 4.7 b 88.6 0.7 b 99.1
All Runs Control 148.4 a 210.4 a
Combined
PAM 89.7 b 39.6 35.0 b 83.3
PAM+Gypsum 71.0 b 52.1 19.3 b 90.8
* Runoff rate and sediment loss mean values followed by same lower case letter are not significantly different at
P<0.05 using t test for multiple means comparison.