GB2214502A - Sewage treatment system - Google Patents

Abstract

A reed bed sewage treatment system comprises a series of generally parallel spaced liquid-conducting channels (1,2), reed bed support medium (soil) (4) extending between neighbouring channels, means (6) for delivering sewage to be treated to alternate ones (1) of said channels wherefrom it can flow outwardly through said medium to connect as treated liquid in other channels (2) between said alternate channels, and means (7) for conducting said treated liquid from the system. The channels may be defined between upstanding stone-filled gabions (3). Impervious plates (11) reduce short circuit surface flows between channels (1) and (2). <IMAGE>

Description

SEWAGE TREATMENT SYSTEM This invention relates to a sewage treatment system, more particularly to a reed bed sewage treatment system.

In recent years, interest has grown in the use of artificially constructed reed beds for the treatment of sewage. As is known, reeds take atmospheric oxygen down to their roots, and the sewage treatment technique essentially involves contacting the sewage with the root-zone (rhizosphere) of the reeds. For this purpose, the sewage to be treated must be flowed through the root-zone, and this requires tho root zone to have an adequately high hydraulic conductivity (kf as defined by d'Arcy's Law). Typically, values in excess of 0.001 (10E 3) m/s are required by most currently designed reed beds with realistic geometries for the flow to pass through the root-zone. However the hydraulic conductivities of saturated soils seem rarely to rise much above .00001 (10E-5) m/s, two orders of magnitude too low.

It has been claimed (Prof* R. Kikuth, State University of Hessen) that the rhizomes of phragmites will increase the hydraulic conductivity of the growing medium.

This may be true for some media but seems not to be so for many . In very small (0.5m) beds, being monitored by yus, those containing soil are showing hydraulic conductivities of 3 x 10E-5 mis. Two natural reed beds, one on peat and one on clay, showed hydraulic conductivities of 1OE-6 m/s or lower. In other trial reed beds, the hydraulic conductivity had not improved over a four year period.

The problem of hydraulic conductivity has had two profound effects to date on the design of reed beds: 1. Beds have been constructed with relatively steep gradients in order to allow them to maintain a greater hydraulic gradient, thus forcing greater flows through the root-zone. However this causes problems with the preferred method of weed-control (flooding), and frequently still fails to force enough flow through the root-zone. If it fails, because of the gradient (up to 5% gradients have been tried) the resulting surface flow erodes the surface of the bed.

2. Beds have been constructed using gravel to ensure that at least the initial hydraulic conductivities are high enough to allow the flow to pass through the root-zone.

This has disadvantages in terms of cost, in terms of loss of ion-exchange properties, and possibly in terms of treatment rates (certainly in the pilot scale beds, those with soil have given higher rates of treatment than those with gravel). Also, it is to be expected that these initial high hydraulic conductivities will fall off with time.

It is possible that, in reed beds of low conductivity in the original growing medium, significant rates of treatment might be achieved in the F#horizon, the layer of composting reed debris that accumulates on the surface of a reed bed. If this is eventually proven, it will be an additional reason to stop constructing beds with significant surface gradients We have now found a way in which the problem of hydraulic conductivity can be overcome without the necessity of resorting to the designs (1) and (2) described above with their attendant disadvantages. As we have said above, a soil reed bed gives generally better sewage treatment than a gravel reed bed.We have found that with very small soil reed beds (O.5m x Im), it is possible to get all the sewage flow through the root-zone even though the hydraulic conductivity is about 3 x 1OE-5. However, the problem is that the only legitimate way to scale up these beds is to make them wider. (For the avoidance of doubt, the width of a bed is its extent measured in a horizontal plane in a direction perpendicular to the general direction of sewage flow therethrough.) Thus, soil reed beds of a size such as 1200m wide x 6m long are indicated, and such geometries especially the enormous width, are virtually impractical.

According to the present invention, we have found a system design which is more practical and in which all the dryweather flow can still pass through the root-zone using a soil bed medium.

According to one aspect of the present invention, there is provided a reed bed sewage treatment system which comprises a series of generally parallel spaced liquidconducting channels, reed bed support medium extending between neighbouring channels, means for delivering sewage to be treated to alternate ones of said channels wherefrom it can flow outwardly through said medium to collect as treated liquid in the channels between said alternate channels, and means for conducting said treated liquid from the system.

The arrangement of channels in accordance with the invention effectively provides a reed bed of very substantial width and yet relatively compact.

In order that the invention may be more fully understood, one embodiment thereof is shown by way of example only in Figures 1 and 2 of the accompanying drawings, Figure 1 being a plan view and Figure 2 being a cross-sectional view.

Referring to Figures 1 and 2, the reed bed system comprises a series (six;are shown) of parallel channels 1,2, each (except for the end channels) formed between pairs of O.5m wide stone-filled gabions 3. Each channel is 0.5m wide and the distance between neighbouring pairs of gabions is 3m The regions between pairs of gabions are filled with soil 4 and planted with reeds. The depth of soil is 0.6m after settlement under saturated conditions. At each end of the overall bed system, the channel 1,2 is defined between a single gabion 3 and a raised bund 5 or the like. As illustrated, an impermeable membrane 9 can be provided at the bottom of the bed system to prevent fluid loss to surrounding ground.

A supply pipe or open channel 6 is connected to one end of the alternate channels 1, to feed influent sewage (for treatment) thereto. The other channels, 2, are connected to an effluent pipe or open channel 7 to conduct the treated liquid away to, for example, an effluent chamber 8. This is preferably a variable head chamber as more fully described hereinafter. The connections should all be made at a low level such that the water levels in all of the inlet channels is the same, and likewise the water level in all of the effluent channels will be that of the variable head outlet.

This arrangement effectively produces a reed bed that is 100m wide and 3m long (or 4m if the gabions are included). In particular it provides an inlet tone of the ditch and gabion type that is loom long.

At a conductivity of 4 x 1OE-5 m/s, this bed accepts the flow from 100 people through the root-zone.

(This assumes that the bed can maintain a hydraulic gradient of 10% As there is a possible fall of nearly 0.6m in every 3m section it is possible that even higher hydraulic gradients could be maintained.) There will be a plan area of 3 sq.m per person of reeds in soil, though the reeds (or at least the rhizomes) will invade the gabions, and it may be reasonable to consider this as 4 sq.m per person.

The width of the bed can be any value. The number of channels 1,2 illustrated can be varied as desired to alter the bed width at will. Also, the length of each soil segment can be varied, eg. to lie typically between Im and 12m. At 12m, a hydraulic conductivity in the soil of 2 x 1OE-4 would be required, for the same flow as assumed for the 3m width presented in this design. However it would be pointless going to 12m unless the flow could be increased pro-rata otherwise the land area per person would go up pro rats. At four times the flow rate (to keep the land area at 3 sq.m per person) a hydraulic conductivity of 8 x 1OE-4 would be required. This latter value is greater than would be expected for soil.

Figures 1 and 2 show the inlet (1) and outlet (2) ditches each connected in parallel. They could equally be connected in series to give a serpentine flow pattern.

It is to be understood that the particular design shown in Figures 1 and 2 is merely illustrative of the invention and not limiting. As will be clear to those skilled in the art, various modifications are possible especially in the construction of the channels or ditches, the gabions and generally.

Among the advantages achievable by the present invention are: i) The system is simple to construct. In particular, no complex distribution systems are needed. The bed system can be about 75% soil and if local soil is used, the costs of construction should be modest, ii) The system is completely horizontal, which makes for ease of construction and allows good water level control for limiting weed growth and promoting rhizome penetration.

iii) It is possible to construct movable walkways 10 (see Figures 1 and 2) supported by the gabions such that the beds can be filled with soil and planted with reeds without the need for anyone to walk on the soil and compact it.

iv) Should the system fail as a sub-surface-flow reed bed, it would be possible to raise the outlet level and operate as a shallow reed-filled, lagoon.

In use of the beds of the invention, during times of high inflow or if hydraulic conductivities remain too low, some surface flow may occur, and this is liable to short circuit. Two solutions to this problem in accordance with further features of the present invention, are as follows. One is to provide a gentle reverse slope into the soil surface, such that it is say 50mm higher against the outlet gabions than the inlet gabions. The other solution is to place impervious plates (PVC) against the outlet gabions as the soil is being placed, such that they extend about 100 mm above the soil surface and 1SO mm below the soil surface. These would have the effect of holding back surface flow and preventing short circuiting occurring via channelling on the soil surface. Such plates 11 are shown in Figure 2.

A further aspect of the present invention concerns the effluent chamber 8. In very wet weather, reed bed treatment systems are particularly susceptible to deterioration of effluent quality because the higher flows tend to flow quickly across the surface of the bed and receive little or no treatment. We have now devised a variable head outlet for connection to the reed bed, whereby back-up of the reed bed can be made to occur to accommodate at least some of the extra flow. In particular, the first peak of storm flow can be so accommodated, and this is usually the most polluted. Figure 3 illustrates schematically one form of effluent chamber of the invention.

As shown, the chamber is in two halves A,B Region A contains a variable head outlet 20 connected to the reed bed. Separating regions A and B is a weir 21 permanently set at a given height, eg. 0.7m, above the reed bed base.

Through the bottom of the weir is a pipe 22 controlled by a valve 23. The valve transmits flows equal to 1.5 DWF. At flows higher than this, chamber A starts to back up. Once it has backed up above the level of outlet 20, it will start to back up the whole reed bed. In conditions of very wet weather it could back up all the way to the top of the weir, converting the reed bed to a 100mm deep lagoon.

However before it gets this far it will have already backed up several days' DWF within the reed bed. Also by the time it gets this far, the extra head will be pushing significantly more than 1.5 DWF through the pipe through the weir. The outlet system of the present invention can be used not only with the reed bed designs of the present invention, but with any other reed bed with a horizontal surface.

Whilst the various aspects of the invention have been described with reference to the treatment of sewage, it has to be understood that the invention applies also to the treatment of other similarly biodegradable materials in reed beds, eg. certain industrial wastes and the like.

Claims (9)

CLAIMS:

1. A reed bed sewage treatment system which comprises a series of generally parallel spaced liquid-conducting channels, reed bed support medium extending between neighbouring channels, means for delivering sewage to be treated to alternate ones of said channels wherefrom it can flow outwardly through said medium to collect as treated liquid in the other channels between said alternate channels, and means for conducting said treated liquid from the system.

2. A system according to claim 1 wherein the reed bed support medium comprises soil.

3. A system according to claim 1 or 2 wherein the parallel channels are defined between upstanding stonefilled gabions.

4. A system according to claim 1, 2 or 3 which also includes means to reduce short circuit surface flow from said alternate channels to said other channels.

5. A system according to claims 3 and 4 wherein said short circuit flow reduction means comprises impervious plates mounted against the gabions defining said other channels.

6. A system according to claims 3 and 4, wherein said short circuit flow reduction means comprises providing the soil surface at a higher level against the gabions defining the other channels than against the gabions defining said alternate channels.

7. A system according to any of claims 1 to 6, which includes a variable head outlet connected to the reed bed to provide back-up during periods of high flow.

8. A reed bed sewage treatment system substantially as herein described with reference to Figures 1 and 2, or Figures 1, 2 and 3, of the accompanying drawings.

9. A method of treating sewage which comprises passing it through a reed bed system as claimed in any of claims 1 to 8.