&ER&
United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-81-091 July 1981
Project Summary
Wastewater Treatment by
Rooted Aquatic Plants in
Sand and Gravel Trenches
Pamela R. Pope
A patented* process developed by
the Max Planck Institute (MPI) of West
Germany to treat industrial wastes
was evaluated as an energy-efficient
method to treat municipal waste-
water. The major goal was to achieve
effluents meeting the U.S. Federal
Effluent Standards using this novel
biological treatment process that re-
quires a minimal amount of
mechanical equipment and manpower
for normal operation.
The Moulton Niguel Water District
(MNWD) of Laguna, California, con-
structed and operated an earthen
trench system using rooted aquatic
plants for the treatment of waste-
water. Two trenches in series were
planted with the reed Phragmites and
the bulrush Scirpus, respectively.
A 2-month study using convention-
al secondary effluent as the trench
influent showed the system was not
effective for removing nitrogen and
phosphorus components.
An 11 -month study demonstrated
that raw screened wastewater applied
to the trench system at a rate not
exceeding 95 mVd (25,000 gpd)
could be treated to secondary effluent
quality. Spatial requirements were
about the same as for a septic tank
system.
This Project Summary was develop-
ed by EPA's Municipal Environmental
Research Laboratory, Cincinnati, OH,
to announce key findings of the
research project which is fully docu-
•U.S. Patent 3,770,623; November 6, 1973.
mented in a separate report of the
same title ('see Project Report ordering
information at back).
Introduction
The MPI process utilizes higher
aquatic plants, such as reeds and bul-
rushes, for the treatment of waste-
water. The system consists of two
earthen trenches, lined with impervious
membranes, operated in series. The
first, designated as the filter trench,
removes coarse suspended solids from
the wastewater. The second, desig-
nated as the elimination trench,
removes dissolved materials from the
effluent of the first trench.
The MNWD services a residential
area, and the wastewater is domestic in
nature. The MPI system, as it was
installed at the MNWD 3A facility, con-
sists of two filter trenches and two
elimination trenches. Two species of
plants were used—a reed Phragmites
communis in the filter trench and a
bulrush Scirpus lacustris in the elimina-
tion trench. A view of the elimination
trench system during construction is
shown in Figure 1
Filter Trenches
The two filter trenches are each 25 m
(75 ft) long, 4 m (12 ft) wide, and 1.3 m (4
ft) deep. They are filled with three layers
of gravel—150 mm (6 in.) of 50-mm (2-
in.) gravel on the bottom; 225 mm (9 in.)
of 19-mm (%-in.) gravel in the middle,
-------
Figure 1. Construction of elimination trench.
and 75 mm (3 in.) of pea gravel on top. A
75-mm (3-in.) layer of silica sand covers
the top. The raw wastewater enters the
MNWD 3A facility at the north side of
the plant and is passed through a roto-
strainer to remove large particles.
The screened influent flows down an
influent channel and is pumped to the
MPI system at a rate of approximately
8.8 to 9.5 X 10"4mVs(14-15gpm).The
influent then flows south through a
control valve into the east end of the
filter trenches. Every 24 hr the flow is
alternated between the two trenches.
Each filter trench contains a central
150-mm (6-in.) plastic pipe with an
open slit running its entire length;
alternating long and short 100-mm (4-
in.) plastic pipes extend from it for even
dispersion of the influent. The waste-
water flows down the central pipe
through the extending pipes and onto
cement splash pads located directly
below. The wastewater percolates
through the trench in a vertical filtering
action leaving a sludge layer on top. The
sand filters out the suspended solids
while the plant system draws moisture
and nutrients. Slow drying of the
deposited solids occurs, and extensive
growth of the plant rootlets and runners
aid in degrading the sludge layer on top
of the sand. Each trench has a perfor-
ated 100-mm (4-in.) plastic pipe extend-
ing the entire length of the trench from
the surface to the bottom in a "U"-
shaped configuration. Flow from the
underdrain goes into this pipe and
through pipe extensions and a butterfly
valve into a sump. This pipe not only
transports the flow but allows aeration
to the bottom since it has openings to
the surface. The sump is a 1.3-m (4-ft)
concrete pipe 3.3 m (10 ft) in height and
is buried 2.6 m (8 ft). A small 0.25-kW
(0.33-hp) pump, activated by a float,
periodically pumps the flow to the
elimination trenches.
Elimination Trenches
Normally, this process would be a
total gravity flow system, with the elimi-
nation trenches so placed as to f acilirate
this, but because of site conditions at
this location, the elimination trenches
had to be constructed 90 m (100 yd)
north of the filter trenches. The two
elimination trenches are 50 m (150 ft)
long, 4 m (12 ft) wide, and 0.75 m (2.5ft)
deep. They are divided in the center by a
weir designed to allow composite samp-
ling in this area and to aid in aeration.
The filter trench effluent enters two
150-mm (6-in.) plastic pipes 4 m (12ft)
long set perpendicular to the trenches at
the south side. The side facing the
trenches is open, and the liquid is
allowed to flow out into an area 1.3 m (4
ft) by 4.0 m (12 ft) in a waterfall-like
action. This area is filled with 15 mm (%
in.) gravel held in place by 50- X 100-
mm (2- X 4-in.) wood baffles. A later
observation (by BWP of New York Inc.)
showed that eliminating the baffles
reduced the operational problems. The
liquid percolates down in a horizontal
manner at a level approximately up to
50 mm (2 in.) below the surface of the
trenches. The trenches are filled with
15-mm (%-in.) gravel with 75 mm (3 in.)
of pea gravel on top. The flow passes
through the weir and runs into two
standpipes that lead into a sump. The
level of the liquid flow is governed by
raising or lowering the standpipes. A
valve in the bottom of the weir allows
periodic draining of the liquid in the
lower portions of the trenches. The total
retention times for the entire system are
estimated to be 6 hr and 8.5 hr at flows
of 133 mVd and 95 mVd, respectively.
Test of MPI System as a
Tertiary Treatment Process
Secondary effluent from the MNWD
extended aeration plant was introduced
to the MPI system on July 1, 1978, at a
loading rate of 56 mVd (15,000 gpd);
the flow was increased to 95 mVd
(25,000 gpd) in August.
The analytical results for samples
obtained during the time extended
aeration effluent was applied to the
system are shown in Table 1.
Overall removal of BOD5, VSS, and
TSS was about 50 percent each month.
COD reduction was about 40 percent.
Ammonium nitrogen removal of 67 per-
cent during August was superior to the
40 percent removal in July. Considera-
tion of the NH4-N, N03-N, and N02-N
values between the 2 months indicates
-------
Table 1. Tertiary Treatment Results for MPI System, Average Values
Item, mg/L
MNWD Secondary Filter Trench
Effluent Effluent
Elimination Overall
Trench Effluent Removal, ',
Flow of 25,000 gpd (95 mVd)
BODS
TSS
VSS
COD
NHA-N
NOt-N
N03-N
TP
15
15
10
58
12
0.8
2.1
11.5
9
8
6
50
8
2
5.5
10
7
7
4
35
6
1.1
7.2
10
53
53
60
39
50
—
—
13
nitrification and subsequent denitrifica-
tion were more active in the system
during the warmer month of August.
TKN samples were not collected during
this period. The MPI system operated as
a tertiary process did not efficiently
remove TP. The extended aeration efflu-
ent applied to the system was of good
quality. The filter trench achieved the
greater part of the overall pollutant
removals; the elimination trench
showed only marginal incremental
removal.
Test of MPI System as a
Secondary Treatment Process
Screened raw wastewater was used
as the influent tothe MPI system in mid-
September at a rate of 56 mVd (15,000
gpd). The flow was increased to 95 mVd
(25,000 gpd) in October and 133 mVd
(35,000 gpd) in January, remaining at
this rate through July. The growth of the
bulrush Scirpus in the elimination
trench is shown in Figure 2.
A thin sludge layer built up and com-
pletely covered the filter trenches by the
end of October. At this time, the
Phragmites had not spread throughout
the trenches. Algae also started to grow
on the sludge layer and may have con-
tributed to later clogging problems. By
mid-December, algae covered a large
portion of the sludge layer, which was
not drying or breaking up as the flow
was directed into the alternate trench.
The problem occurred equally in both
filter trenches. Operation was sus-
pended at this time to consider this
problem and to harvest the Phragmites
from the filter trenches. The Phragmites
had turned brown because of unusually
cold weather and had begun to layover
because of their mature height and
weight. When the sludge layer was
skimmed off a little at a time, about 25
mm (1 in.) of sludge was found deposi-
ted on the filter media; this sludge layer
was wet and becoming septic. Below
this was a layer of compacted organic
matter and fibers that were black in
appearance and felt greasy; this inter-
mingled with the sand, formed an
almost impermeable layer. If a hole was
poked through this layer, the liquid held
in the sludge layer immediately drained
through the remaining sand. The only
solution at this time was to allow the
filter trenches to dry after the harvest-
ing and then to rake out the semidry top
sludge layer carefully. Figure 3 shows
the plants after harvesting.
Subsequent tests conducted in Long
Island, New York, by BWP of New York,
Inc., showed that using four parallel
filter trenches to allow increased drying
time and that draining the filter
trenches three times a week minimized
this sludge problem.
The major objective of this project
was to evaluate the MPI system as a
low-cost wastewater treatment alterna-
tive that would satisfy federal discharge
requirements. These requirements are
attained if final effluent BODS and SS
concentrations do not exceed 30 mg/L
for 30 days average values, or 85 per-
cent overall removal, whichever is more
stringent. The fate of nitrogen and
phosphorus was also monitored. Table
2 summarizes all the data collected.
The system was evaluated for sec-
ondary treatment effectiveness for 11
months. For 5 of the months, the flow
through the system was 95 mVd
(25,000 gpd) or less; for 6 months, 133
mVd (35,000 gpd). Secondary treat-
ment requirements for BOD5 and SS
were achieved all 5 months at the lower
flow. SS residuals and percent removals
met secondary requirements all 6
months at the higher flow rate; how-
ever, the BODs requirement was not
achieved for 5 of the 6 months. The
effluent violated both the concentration
and percent removal requiremerits
three times (January, April, and July);
the percent removal requirement only
was violated twice (May and June).
There was little difference in overall
COD removal for the two application
rates.
The NH<-N and Org-N concentration
values in the effluent during the periods
of 95 mVd application were represen-
tative of conventional secondary treat-
ment residuals. Variations in percent
removals were because of fluctuations
in influent concentrations. The overall
removal of total nitrogen varied from 61
percent in September to 32 percent in
March.
During application of 133 rhVd, the
Org-N residuals were about twice the
values of the lower flow rate results.
During February, a negative removal of
Org-N was noted. The overall removal of
total nitrogen was much lower than
during the 95 mVd application, 18
percent in January to 36 percent in
June.
Nitrite and nitrate nitrogen concen-
trations for all the sample periods show
that nitrification did not occur to any
significant extent at either of the two
flow rates.
The MPI system during both the 95
mVd and the 133 mVd application
rates was not effective for total
phosphorus removal. During the higher
application rate, 2 months (January and
June) showed negative removals.
The major increment of BODs, SS,
VSS, and COD removal occurs at the
filter trench, and the elimination trench
serves as a polishing process (Table 2).
Both trenches in series are necessary
for satisfactory treatment.
The MPI system operated with raw
screened wastewater at an application
rate of 95 m3/d did achieve secondary
effluent quality. Using the trench meas-
urements, the spatial requirements of
-------
Figure 2. Aquatic plant growth in elimination trench.
the MPI system equate to 0.02 mVm-d
(0.5 gpd/ft2). Assuming a per capita
wastewater discharge of 378 L (100
gal), the area required is 2 mVcapita (21
ftVcapita). These two values are very
similar to spatial requirements of a
septic tank system located in a satisfac-
tory percolating soil.
Several operating problems, expected
with new technology development,
were experienced during this
demonstration study. Many of the same
operational problems were encount-
ered at the Long Island, New York.
Several operating problems, expected
with new technology development
were experienced during this demon-
stration study. Many of the same opera-
tional problems were encountered at
the Long Island, New York, installation.
Remedial measures applied at Long
Island included:
2. Recommend harvesting plants not
more than once a year. Frequent
harvesting of the plants used in
the system promotes extra growth
of the root systems and this con-
tributes to clogging.
3. If plant growth becomes excessive
during the year, individual plants
are culled by pulling to thin the
growth.
This initial assessment of the effi-
ciency and spatial considerations for the
MPI system for secondary treatment
indicates it is worthy of further develop-
ment.
The full report was submitted in ful-
fillment of Grant No. R-805279 by the
Moulton Niguel Water District under the
sponsorship of the U.S. Environmental
Protection Agency.
1. Provision for increased area for
the filter trenches, thereby allow-
ing longer idle times for drying.
-------
Figure 3. Filter trench after harvest of plants.
-------
Table 2. Secondary Treatment Results for MPI System, Average Values in mg/L
Sample
Location
BODs TSS COD NH4-N Org-N NOt-N N03-N TP
Influent
Wastewater
Filter Trench
Effluent
Elimination
Trench Effluent
Overall Removal,
Percent
Influent
Wastewater
Filter Trench
Effluent
Elimination
Trench Effluent
Overall Removal,
Percent
Flow of 25,000 gpd (95 mVd)
210 225 405 24 13 0.0 0.1
77 41 179 19
26 20 86 16
0.4 0.9
0.1 0.4
88 91 79 33 31 — —
Flow of 35,000 gpd (133 mVd)
171 181 405 25 17 0.0 0.3
68 48 157 21 13 O.6 1.2
35 19 93 19 11 0.3 0.6
80 89 77 24 35 — —
13
12
12
8
13
13
13
0
Pamela R. Pope is with the Mouhon Niguel Water District, Laguna Niguel, CA
92677.
Ronald F. Lewis is the EPA Project Officer (see below).
The complete report, entitled "Wastewater Treatment by Rooted Aquatic Plants
in Sand and Gravel Trenches," (Order No. PB 81-213 241; Cost: $6.50, subject
to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-757-012/7203
------- |