8
9
5
6
7
1
2
3
4
Published as: Healy, M.G., Rodgers, M., Mulqueen, J. 2007. Treatment of dairy wastewater using constructed wetlands and intermittent sand filters. Bioresource Technology 98:
2268-2281.
TREATMENT OF DAIRY WASTEWATER USING CONSTRUCTED WETLANDS
AND INTERMITTENT SAND FILTERS
M.G. Healy
*
, M. Rodgers and J. Mulqueen
10
11 Department of Civil Engineering, National University of Ireland, Galway, Ireland.
12
13 Abstract
14
15 In Ireland, the most common method of disposal of dairy parlour washings is by land
16 spreading. This treatment method has numerous problems, namely high labour
17 requirements and the potential for eutrophication of surface and ground waters.
18 Constructed wetlands are commonly used for treatment of secondary municipal
19 wastewaters and they have been gaining popularity for treatment of agricultural
20 wastewaters in Ireland. Intermittent sand filtration may offer an alternative to traditional
21 treatment methods. As well as providing comparable treatment performance, they also
22 have a smaller footprint, due to the substantially higher organic loading rates that may be
23 applied to their surfaces. This paper discusses the performance and design criteria of
24 constructed wetlands for the treatment of domestic and agricultural wastewater, and sand
* Corresponding author. Tel: +353 91 495364; Fax: +353 91 494507. E-mail: mark.healy@nuigalway.ie
1

25 filters for the treatment of domestic wastewater. It also proposes sand filtration as an
26 alternative treatment mechanism for agricultural wastewater and suggests design
27 guidelines.
28
29 Keywords: dairy parlour washings; domestic wastewater; intermittent sand filtration;
30 constructed wetlands; land spreading; organic loading rate.
31
32 1. Introduction
33
34 Dairy parlour washings, containing urine and manure, as well as detergents, spilled milk,
35 and mucus, can have a significant adverse effect on the environment. In Ireland and
36 elsewhere, dairy parlour washings are treated mainly by land spreading with low rate
37 irrigation systems. This has proven problematic because of the labour requirements for
38 effective management and the potential for eutrophication due to run-off and leaching.
39 Problems of surface run-off are particularly prevalent on slow-draining soils with sloping
40 land at field capacity or wetter. In land spreading on free-draining soils, the main nutrient
41 removal processes are filtration, soil adsorption, microbial decomposition, and plant
42 uptake. The latter two processes are active in reducing nitrate-nitrogen (NO
3
-N)
43 concentrations; however, if NO
3
-N passes beyond the root zone it can be leached to the
44 groundwater. Analyses of five Irish borehole (well) waters underlying light textured soils
45 receiving high nitrogen (N) applications have yielded NO
3
-N concentrations greater than
46 the EU maximum allowable concentration (MAC) of 11.3 mg L -1 for drinking water
47 (Richards et al., 1998). In that study, Richards et al. (1998) found, that in a plot
2

48 comprising a sand loam overlying a sandy silt loam to 106 cm deep receiving a mean
49 wastewater application rate of 677 kg N ha -1 yr -1 , mainly as organic N, soil water NO
3
-N
50 concentration was 23 mg NO
3
-N L
-1
.
51
52 The Nitrate Directive, 91/676/EEC (EEC, 1991) has focused considerable attention on the
53 disposal of agricultural wastewaters in Ireland, where about 25% of the land area is
54 devoted to dairy farming (Carton, 2001). A survey of 1132 rivers and streams from 2001
55 to 2003 (Toner et al., 2005) estimated that the percentage of pollution attributed to
56 agriculture was approximately 32%, 32% and 15% in rivers and streams which were
57 slightly, moderately, and seriously polluted, respectively.
58
59 In farms, volumes of water used for wash-down of yard areas, milk spillage, yard area
60 rainfall, and stormwater run-off lead to a final effluent of variable concentration and
61 volume. The water volumes generated may vary according to the practices applied by the
62 farmers. Factors such as frequency of milking and the number of cows present at the
63 same time affect the volumes generated. Dairy parlour washings have been estimated at
64 50 litres per cow per day (Department of the Environment and Department of
65 Agriculture, Food and Forestry, 1996) but this value can be frequently exceeded
66 especially where there is indifferent management of water usage. Studies have shown that
67 when low-pressure, high-volume spray nozzles are used in yard wash-down, water
68 volumes generated can be significantly reduced (Wrigley, 1994).
69
3

70 Concentrations of contaminants in dairy wastewater are significantly higher than
71 municipal effluent, and tend to vary throughout the year. On a 160 cow dairy farm, Ryan
72 (1990) found that total nitrogen (Tot-N) and biochemical oxygen demand (BOD
5
)
73 concentrations in the washings ranged, respectively, from 88 and 954 mg L
-1
during the
74 winter to 225 and 2974 mg L
-1
, respectively, during the summer. These concentrations
75 also depend on the number of rinses of the milking machine as Ryan (1990) found that
76 the BOD
5
concentration of the washings decreased from 19,200 mg L -1 in the first rinse
77 to 144 mg L
-1
in the final rinse. In the hydraulic design for land spreading, the hydraulic
78 conductivity of the soil, the depth to water table and/or bedrock, and the concentration of
79 the suspended sediment can be defining parameters. Maximum land application rates of 5
80 mm per hour are recommended (Department of the Environment and Department of
81 Agriculture, Food and Forestry, 1996). The quantity of wastewater applied should not
82 exceed 50 m
3
per hectare by single tanker application. In The Netherlands, application
83 rates of up to 15 mm per application are recommended (Hack-ten Broeke, 2001).
84
85 In recent years, the use of constructed wetlands (CWs) for the treatment of dairy parlour
86 washings has been gaining popularity, due to their relatively low capital costs and
87 maintenance requirements. However, intermittent sand filtration (ISF) may have the
88 potential to treat dairy parlour washings effectively and, where denitrification is
89 incorporated, to reduce NO
3
-N to low levels. To date, the use of ISF for the treatment of
90 agricultural wastewater has been limited. They have been used in the dewatering of swine
91 wastewater following addition of organic polymers to increase settlement of suspended
92 solids (SS) and organic compounds (Vanotti et al., 2005; Szögi et al., 2006) and in the
4

93 treatment of detergent and milk fat wastewaters (Liu et al., 1998; Liu et al, 2000; Liu et
94 al., 2003).
95
96 This article aims to review the present state of knowledge regarding CWs and ISF for the
97 treatment of domestic wastewater. The performance of CWs in the treatment of dairy
98 parlour washings is also discussed. As ISF may be a viable alternative to conventional
99 CW treatment, this article suggests design protocols which may be adopted for the ISF of
100 dairy parlour washings, as well as detailing some previous experiences in their use.
101
102 2. Domestic wastewater treatment
103
104 2.1 Constructed wetlands
105
106 There are two types of CW: free water surface constructed wetlands (FWS CWs) and
107 subsurface CWs. In FWS CWs, wastewater flows in a shallow water layer over a soil
108 substrate. Subsurface CWs may be either subsurface horizontal flow CWs (SSHF CWs)
109 or subsurface vertical flow CWs (SSVF CWs). In SSHF CWs, wastewater flows
110 horizontally through the substrate. In SSVF CWs, wastewater is dosed intermittently onto
111 the surface of sand and gravel filters and gradually drains through the filter media before
112 collecting in a drain at the base. CWs may be planted with a mixture of submerged,
113 emergent and, in the case of FWS CWs, floating vegetation.
114
5

115 The large surface area of CWs provides an environment for the physical/physico-
116 chemical retention and biological reduction of organic matter and nutrients (Geary and
117 Moore, 1999; Knight et al., 2000). Depending on the type of CW used, its design, organic
118 loading rate and hydraulic retention time (HRT) (Karpiscak et al., 1999), a CW can have
119 a significant nutrient removal capability. However, due to the effect of changing
120 temperatures, the treatment efficiency of these systems tends to change throughout the
121 year (Bachand and Horne, 2000; Healy and Cawley, 2002).
122
123 Although temperature affects the performance of FWS CWs, SSHF CWs and SSVF
124 CWs, they generally meet relevant effluent discharge criteria in colder climates
125 (Maehlum et al., 1995; Vymazal, 2002; Rousseau et al., 2004). In Europe, effluent BOD
5
126 and SS standards for discharge into surface waters range from 250 mg BOD
5
L -1 in The
127 Netherlands to 25 mg BOD
5
L
-1
in Austria and 70 mg SS L
-1
in The Netherlands to 35 mg
128 SS L
-1
in the Czech Republic, respectively (Rousseau et al., 2004). In Norway, where
129 mean winter temperatures can drop below -10 o
C, Maehlum et al. (1995) used an SSHF
130 CW to treat septic tank wastewater. Under an organic loading rate of approximately 4 g
131 BOD
5
m -2 d -1 , BOD
5
and Tot-N was removed by 93% and 48%, respectively (Table 1).
132
133 N removal in CWs is accomplished primarily by physical settlement, denitrification and
134 plant/microbial uptake. Plant uptake does not represent permanent removal unless plants
135 are routinely harvested.
136
6

137 Phosphorus (P) is removed through short-term or long-term storage. Uptake by bacteria,
138 algae and duckweed (Lemma spp.
), and macrophytes provides an initial removal
139 mechanism (Kadlec, 1997). However, this is only a short-term P storage as 35%-75 % of
140 P stored is eventually released back into the water upon dieback of algae and microbes
141 (Richardson and Craft, 1993; White et al., 2000). Anaerobic conditions which exist at the
142 soil/water interface may also cause the release of P back into the water column (Patrick
143 and Khald, 1974). The only long-term P storage in the wetland is via peat accumulation
144 and substrate fixation. The efficiency of long-term peat storage is a function of the
145 loading rate and also depends on the amount of native iron, calcium, aluminium, and
146 organic matter in the substrate (Shatwell and Cordery, 1999). Lake and reservoir
147 sediments have been shown to act as P sinks (Richardson and Craft, 1993; White et al.,
148 2000). At P loading rates of less than 5g P m -2 yr -1 , wetland sediment can absorb greater
149 than 90% of the total incoming P (Faulkner and Richardson, 1989).
150
151 2.1.1 Pretreatment
152
153 Domestic wastewater must undergo septic tank pre-treatment prior to entering a CW
154 (EPA, 2000). In municipal wastewater treatment, an activated sludge plant provides
155 initial settlement, organic carbon removal and partial nitrification (Healy and Cawley,
156 2002). To protect against groundwater contamination, all FWS CWs, SSHF and SSVF
157 CWs should be lined with an impervious layer e.g., a high density polyethylene liner
158 (HDPL).
159
7

160 2.1.2. Media selection
161
162 For FWS CWs, a substrate rich in iron, calcium and aluminium is recommended. For
163 SSHF CWs, a soil or gravel is recommended (Cooper et al., 1996). In SSVF CWs, an
164 active sand layer with a depth of 1.0 m (effective grain size, d
10
=0.25 – 1.2 mm,
165 coefficient of uniformity, Cu<3.5) is recommended (Brix and Arias, 2005).
166
167 2.1.3 Treatment area and organic loading rates
168
169 FWS CWs and SSHF CWs are normally sized in accordance with (Kadlec and Knight,
170 1996):
171
172 A



0 .
0365 Q k

 ln


C
C e i

C
*

C
*

 (1)
173
174 where A = required wetland area (ha)
175
176
C e
= the outlet concentration (mg L
-1
)
C i
= inlet concentration (mg L
-1
)
177 C* = background concentration (mg L
-1
)
178 k = first-order areal rate constant (m yr -1 )
179 Q = hydraulic loading rate (m d
-1
)
180
181 Depending on the water quality parameter used to size the CW, the constants in the
182 model (C* and k) may be different for FWS CWs and SSHF CWs (Kadlec and Knight,
8

183 1996). For example, if using BOD
5
to size a CW, C*= 3.5 + 0.053C
i
for a FWS CW or a
184 SSHF CW and k would be 34 and 180 m yr -1 for a FWS CW and a SSHF CW,
185 respectively (Kadlec and Knight, 1996).
186
187 In CW design, it is difficult to account for variables such as climate variation, pre-
188 treatment control, and time to maturation. Therefore, design guidelines tend to be
189 conservative. Organic and SS loading rates not exceeding 6 g BOD
5
m -2 d -1 and 5 g SS
190 m
-2
d
-1
, respectively, are recommended for FWS CWs (US EPA, 1992). SSVF CWs may
191 be operated in single-pass mode, intermittently loaded 8-12 times d
-1
, or in recirculation
192 mode, intermittently loaded 16 – 24 times d
-1
(Brix and Arias, 2005). Winter and Goetz
193 (2003) recommend a maximum organic loading rate of 20 g COD m
-2
d
-1
and a maximum
194 SS influent concentration of 100 mg L -1 for SSVF CWs. Even when these loading
195 conditions are satisfied, the performance of CWs may be variable.
196
197 2.1.4 Wetland vegetation
198
199 Ireland has a cool temperate west maritime climate. In these climatic conditions, common
200 reed ( Phragmites australis (Cav.) Trin. ex Steud.) and common cattail ( Typha latifolia
201 L.) are mainly planted in CWs. As the amount of oxygen released by the emergent
202 vegetation into the surrounding soil is small (Armstrong et al., 1990), anaerobic
203 conditions predominate.
204
9

205 Harvesting of the emergent macrophytes has a pronounced effect on the growth and
206 nutrient uptake rates. Although nutrient uptake and growth rates are higher in young
207 vegetation stands (Greenway and Whoolley, 2001), other factors such as nutrient loading
208 and hydraulic retention time (HRT) may significantly affect the uptake rates (Reddy et
209 al., 2001; Hardej and Ozimek, 2002). In cool temperate west maritime climates, shoot re-
210 growth depends on the time of year at which harvesting takes place (M.G. Healy,
211 unpublished data). Harvesting during June/July produces good shoot re-growth, whereas
212 August/September harvesting tends not to produce significant re-growth.
213
214 2.1.5 Treatment efficiency
215
216 SSHF CWs are ideal for cold climates because wastewater treatment occurs below the
217 surface (Werker et al., 2002). They are the most common CW system used in Europe
218 (Vymazal, 2005). SSHF CWs have good organic, SS and faecal coliform removal rates
219 but have poor NH
4
-N removal rates (Neralla et al., 2000; Weaver et al., 2001; Steer et al.,
220 2005, Vymazal, 2005). In Texas, USA, Neralla et al. (2000) monitored 8 SSHF CWs
221 comprising gravel media ranging in size from 0.95 to 1.6 cm and receiving domestic
222 effluent from a septic tank. Under organic loading rates ranging from 2 to 5 g BOD
5
m
-2
223 d
-1
, average effluent BOD
5
and SS removals of 80% and 88%, respectively, were
224 measured (Table 1). An average NH
4
-N removal of 39% was also measured and
225 nitrification did not occur. These results were similar to Steer et al. (2005) who studied
226 the performance of 8 SSHF CWs treating domestic effluent, comprising two 25m 2 SSHF
227 CWs connected in series and preceded by a septic tank. Over a 5 yr. duration, average
10

228 BOD
5
and SS removals of 69-98% and 77-83%, respectively, were measured and NH
4
-N
229 was reduced by approximately 70% (Table 1).
230
231 SSVF CWs are commonly used for domestic wastewater treatment. When the organic
232 loading rate does not exceed a maximum allowable organic loading rate of 20 g COD m
-2
233 d
-1
(Winter and Goetz, 2003), they effectively remove organic matter, SS and nutrients
234 (von Felde and Kunst, 1997). Luederitz et al. (2001) intermittently loaded a SSVF CW,
235 comprising a 0.6 m active sand layer (sand/gravel, 0 – 4 mm) 800 m
2
in area and
236 preceded by an anaerobic digester, at an organic loading rate of 35 g COD m
-2
d
-1
(21 g
237 BOD
5
m
-2
d
-1
) and measured a COD removal of 94% and Tot-N removal of 61% (Table
238 1). With a SSVF CW having the same active sand layer but preceded by two unaerated
239 ponds and receiving an organic loading rate of 20 g COD m -2 d -1 (10 g BOD
5
m -2 d -1 ),
240 COD and Tot-N removals were 99.5% and 93.8%, respectively (Table 1).
241
242 Recently, a modified version of the SSVF CW, the two-stage vertical flow constructed
243 wetland (VFCW), has been gaining popularity in France, where there are currently
244 around 400 VFCWs in operation (Molle et al., 2006). The first stage of this system
245 comprises 3 parallel vertical flow sand filters which are alternately intermittently dosed
246 with raw wastewater at an organic loading rate of 300 g COD m
-2
d
-1
. In this first stage,
247 COD and SS removal takes place. They contain a 30 cm-deep fine gravel layer (2-8 mm
248 in size) which overlies a 10-20 cm-deep transition layer (5 mm in size) and a 10-20 cm-
249 deep drainage layer (20-40 mm in size) (Molle et al., 2005). The second stage comprises
250 two identical vertical flow sand filters which contain a 30 cm-deep fine gravel layer
11

251 (effective grain size, d
10
< 0.40 mm) which overlies a 10-20 cm-deep transition layer (3-
252 10 mm in size) and a 10-20 cm-deep drainage layer (20-40 mm in size) (Molle et al.,
253 2005). Nitrification mainly occurs in the second stage. Results from these systems have
254 been good with COD and SS removals of 90% and 95%, respectively, being measured
255 and nitrification at 85% (Molle et al., 2005).
256
257 FWS CWs are also effective in organic matter, SS, and faecal coliform removal (Ran et
258 al., 2004) but, similar to SSHF CWs, have low N removals (Healy et al., 2002; Ran et al.,
259 2004). Studies have reported settlement as the main N removal pathway (Toet et al.,
260 2005). In a 2-cell FWS CW, planted with duckweed and preceded by a preliminary
261 storage tank and a primary sedimentation tank, Ran et al. (2004) measured average BOD
5
262 and SS removals of 71% and 80%, respectively, when the system was loaded at an
263 organic loading rate of 16 g BOD
5
m
-2
d
-1
. Removal of N within the system seemed to be
264 due mainly to sedimentation and plant uptake, as NH
4
-N removals of 14% were measured
265 (Table 1).
266
267 When FWS CWs are in a marsh - retention pond - marsh formation, low flushing rates
268 and little surface cover means that eutrophication may occur in the retention pond during
269 warmer periods of the year (Healy and Cawley, 2002). In a 2-year study of a 3-cell FWS
270 CW for tertiary treatment of municipal wastewaters in Williamstown, Ireland, Healy and
271 Cawley (2002) measured average BOD
5
and SS removals of 49% and 90%, respectively
272 (Table 1) but noted the occurrence of algal blooms in the retention pond during the
273 summer months.
12

274
275 2.2 Intermittent sand filtration
276
277 ISF has been used for the treatment of domestic wastewater for over a hundred years.
278 Sand filters may be operated either in single-pass or recirculation mode. In single-pass
279 mode, following primary sedimentation, the wastewater is intermittently dosed onto a
280 stratified sand filter (Gross and Mitchell, 1985).
281
282 On a single pass through the system, organic carbon removal, ammonification and
283 nitrification occurs. Factors affecting the retention of bacteria in porous media include
284 straining, the grain size of the filter media, and the hydraulic loading rate (Stevik et al.,
285 2004). Removals of greater than 99.9% have been recorded for faecal coliforms
286 (Vanlandingham and Gross, 1998). A study comparing single-pass sand filters (33.5 m
2
),
287 FWS CWs (53 m
2
), and peat biofiltration systems (28 m
2
) for the treatment of septic tank
288 effluent (Puraflo
TM
, Bord na Mona, Ireland) has shown that single-pass sand filters have
289 the greatest organic and nutrient removal efficiency, although the difference in
290 performance between the sand filtration and peat biofiltration systems is small (White,
291 1995). White (1995) measured organic carbon removals of 92% and nitrification of 91%
292 for sand filters. Organic carbon removals of 87 and 82% were measured for the peat
293 biofiltration and constructed wetland systems, respectively. No nitrification occurred in
294 the constructed wetlands, and the percentage nitrification in the peat biofiltration systems
295 was 90%.
296
13

297 Virus removal has also been estimated to occur in the first 30 cm of a stratified sand filter
298 sand (Gross and Mitchell, 1985; Gross, 1990), although removal is dependent on the
299 hydraulic loading rate and degree of saturation of the filter (Reneau et al., 1989). In a
300 series of columns containing medium concrete sand (d
10
= 0.32 mm) and loaded with
301 dechlorinated tap water containing MS2 bacteriophage at hydraulic loading rates of 51,
302 81, 12.2, and 16.3 L m
-2
d
-1
, Vanlandingham and Gross (1998) found average MS2 phage
303 removal efficiencies of 99.6%, 98.6%, 99.9%, and 97.2%, respectively.
304
305 In a normal single-pass filter operation, denitrification is limited by the absence of
306 reducing conditions following nitrification and the lack of an available carbon source.
307 This leads to poor Tot-N reduction and an effluent that is high in NO
3
-N. In order that
308 ISF offers a viable treatment alternative to conventional treatment methods, NO
3
-N in the
309 effluent needs to be denitrified. Recirculation of a major portion of the nitrified effluent
310 from the sand filter through a denitrifying anoxic zone receiving the influent wastewater,
311 located before the filter, offers a solution to this problem (USEPA, 1980). The
312 recirculation ratio is defined by:
313
314
 
Q r
Q
(2)
315
316 where

is the recirculation ratio, Q r is the return flow from the filter to the anoxic
317 recirculation tank, and Q is the influent wastewater. Alternatively, external carbon
318 sources can also be used in an anoxic reactor positioned after the nitrification reactor.
319 Researchers have utilised methanol, ethanol, glucose, and acetate in denitrification (Lamb
14

320 et al., 1990; Carley and Mavinic, 1991). However, the widespread use of these chemicals
321 has been hampered by the possibility of environmental hazards due to the toxicity of
322 substances such as methanol (Lamb et al., 1990). In attempts to reduce the hazards
323 associated with such systems, researchers began to consider septic tank effluent as a
324 source of carbon. However, the success of such a system depends largely on system
325 design. Short contact times, a low carbon to nitrogen ratio (C:N) in the recirculation tank
326 and high recirculation rates all result in poor denitrification rates in the system (Tebbutt,
327 1998; van Buuren et al., 1999).
328
329 To date, a number of variations on recirculating sand filters have been explored. At a
330 hydraulic loading rate of 155–195 L m
-2
d
-1
, Lamb et al. (1990) found that denitrification
331 was achieved when the sand filter effluent was mixed with three different carbon sources
332 (methanol, ethanol and septic tank effluent) prior to entering a buried rock tank. With
333 septic tank effluent as the carbon source, only 25% denitrification occurred, whereas
334 methanol and ethanol produced a mean denitrification of 99%. This poor performance for
335 the septic tank effluent was due to a C:NO
3
-N ratio of 0.7:1 in the rock tank. Ethanol and
336 methanol resulted in C:NO
3
-N ratios of 2:1 and 4:1, respectively. Gold et al. (1992)
337
338 encountered similar poor denitrification performance of a sand filter when the nitrified effluent was recirculated through a recirculation tank with an

ratio of 4:1 to 5:1. Tot-N
339 removal increased from 8.4% (on a single pass through the sand filter) to 20% for
340 recirculation. Again a low BOD
5
:NO
3
-N of approximately 1:1 in the recirculation tank
341 limited denitrification. Optimal conditions for maximum denitrification occur at a COD:
15

342 Tot-N ratio in the range of 4:1 to 5:1 for organic sludge (Henze et al., 1997, Martinez,
343 1997; van Buuren et al., 1999).
344
345 2.2.1 Biological clogging of filter media
346
347 Biological clogging of the filter media is often problematic (Siegrist, 1987). Clogging of
348 the upper layers of the sand filter increases the average water retention time in the filter
349 and reduces the effective area available for water infiltration to a point where ponding
350 occurs.
351
352 Surface clogging may be due to a number of causes. Accumulation of microorganisms on
353 surfaces as biofilms is believed to be the cause of surface sealing (Siegrist and Boyle,
354 1987; Vandevivere and Baveye, 1992; Schwager and Boller, 1997; Bouwer et al., 2000).
355 In this process, hydrated extracellular polymers (exopolymers) as well as cells
356 accumulate on the upper layers of the sand media and give rise to a reduction in
357 permeability (Schwager and Boller, 1997). Siegrist and Boyle (1987) found an
358 accumulation of organic matter in the upper sand layer, and hypothesised that it may have
359 undergone humification, and gradually filled the pore space, reducing the permeability.
360 The type of filter media (Jowett and McMaster, 1995) and the deposition of organic and
361 inorganic solids on the surface layer (Platzer and Mauch, 1997; Rodgers et al., 2004)
362 have also been considered to cause surface sealing.
363
16

364 Siegrist (1987) used gravimetric water content profiles to measure pore size reduction in
365 an aggregate loaded with domestic septic tank effluent, greywater septic tank effluent,
366 and tapwater. The most significant increases in water content near the infiltrative surface
367 were attributed to the pore size reduction due to biomass build-up and, after 62 months of
368 operation, the water contents in the upper 4 cm layer for treating tapwater and domestic
369 septic tank effluent were 0.26 and 0.36, respectively.
370
371 Conservative tracers such as sodium bromide (NaBr) can also be used to illustrate this
372 effect (Schudel and Boller, 1990). Schwager and Boller (1997) used an F-curve
373 (Levenspiel, 1999) in a ‘clean’ and ‘used’ filter to show the effect of biomass growth in
374 the top filter layer on the prolongation of the retention time of a liquid.
375
376 2.2.2 Effect of dosing frequency on treatment performance.
377
378 An increase in dosing frequency will have a beneficial effect on the treatment
379 performance of a single-pass sand filter (Anderson et al., 1985). In a study of 12 shallow
380 sand filters (0.38 m deep), with four sets having the same d
10
(0.29 mm) and uniformity
381 coefficient (Cu=4.5), Darby et al. (1996) found that for the same daily hydraulic loading
382 rate, increasing the application frequency from 4 to 24 times per day resulted in a slight
383 but statistically significant increase in the removal of turbidity, COD and organic-N
384 (Table 2). In a field-scale experiment, Boller et al. (1993) showed that in a sand filter
385 containing coarse sand (d
50
=1.4; Cu=4) dosed daily with septic tank effluent, the filter
386 performed better when loaded with 4 flushes of 10 L m
-2
than when loaded with 1 flush
17

387 of 40 L m
-2
or 2 flushes of 20 L m
-2
. Average sand filter effluent NH
4
-N concentrations
388 were 21.0, 17.2 and 3.8 mg NH
4
-N L -1 at 1, 2, and 4 daily flushes, respectively. The
389 results from a pilot-scale sand filter unit (Boller et al., 1993), loaded at 6 times per day,
390 are tabulated in Table 2.
391
392 2.2.3 Effect of media size and uniformity coefficient on treatment performance.
393
394 The uniformity coefficient (Cu) is defined as (Craig, 1997):
395
396 Cu
 d
60 (3) d
10
397
398 where d
60
and d
10
denote the largest possible sizes of the 60% and 10% fractions,
399 respectively. The higher the uniformity coefficient, the larger the range of particle sizes in
400 the sand. This may affect filter performance at higher hydraulic loading rates, as a well-
401 graded sand means that small particles may fill interstices between large particles, leading
402 to a reduction in the hydraulic conductivity, and possible blocking of the filter media.
403
404 Design specifications recommend that the sizing of the filter media should be in
405 accordance with its use (USEPA, 1980; Ball and Denn, 1997; Loomis and Dow, 1999).
406 For single-pass operation with a filter depth of 0.61 – 0.91 m, a d
10
of 0.33 mm and Cu<3
407 are recommended, whereas in recirculation mode, a d
10
of 1.5-3.0 mm and Cu of 1.3-2.5
408 are recommended. Although increased nitrification has been attributed to a coarse grain
409 size (Nielsen et al., 1993), performance effects appear to be mostly related to hydraulic
18

410 and organic loading rate and dosing frequency (Darby et al., 1996). Nichols and Abboud
411 (1995) found that complete organic carbon removal and good Tot-N removals (68-74%)
412 were still attained when a chip stone (d
10
=1.85 mm; Cu=1.9) and a pea gravel (d
10
=2.05
413 mm; Cu=2.7) were used in a single-layer recirculating sand filter treating effluent from a
414 restaurant.
415
416 2.2.4 Effect of organic and SS loading rates on performance.
417
418 The organic and SS loading rates have a significant effect on the clogging and
419 performance of a filter. The US EPA (1980) recommend that, in single-pass sand filters,
420 the organic loading rate should not exceed 22 g BOD
5
m
-2
d
-1
at a hydraulic loading rate
421 of 40-80 L m -2 d -1 , whereas in recirculating sand filters the organic loading rate should not
422 exceed 22 g BOD
5
m
-2
d
-1
at a hydraulic loading rate of 120 – 200 L m
-2
d
-1
. Results from
423 single-pass and recirculating sand filters are given in Tables 2 and 3, respectively.
424
425 A number of studies have investigated the filter performance under varying loading rates
426 in an attempt to quantify the maximum organic and SS loading rates that can be safely
427 applied to a filter (Darby et al., 1996; Nichols et al., 1997). Nichols et al. (1997) found
428 that, when the COD and SS loading rates on a 1 m-deep, 3-layer stratified sand filter were
429 increased to 15 and 1.0 g m
-2 d
-1
, respectively, reduced rates of nitrification occurred.
430
431 Effert et al. (1985) applied domestic effluent onto a sand filter 76 cm deep (d
10
= 0.4mm;
432 Cu = 2.5) at an organic loading rate of 20 g BOD
5
m
-2
d
-1
and found that almost complete
19

433 organic removal occurred, but in the absence of denitrifying conditions, Tot-N was
434 reduced by only 1-11%, with the effluent Tot-N mainly present as NO
3
-N. In an analysis
435 of 50 sand filters, with an effective size (d
10
) of 0.2-0.3 mm, treating domestic effluent,
436 Nielsen et al. (1993) showed that at an average organic loading rate of approximately 10
437 g BOD
5
m
-2
d
-1
, 90-95% of the BOD
5
and 30-45% of the Tot-N were removed.
438
439 The organic and SS loading rates can have a significant effect on the clogging and
440 performance of a filter. A number of studies have investigated the filter performance
441 under varying loading rates in an attempt to quantify the maximum organic and SS
442 loading rate which can safely be applied to a filter (Nichols and Abboud, 1995; Darby et
443 al., 1996).
444
445 2.3 Alternative single-pass filter designs.
446
447 A number of alternative filter designs have been proposed. Latvala (1993) proposed the
448 use of a multilayer intermittent sand filter, wherein wastewater is loaded through three
449 gravel layers at incremental depths in the layered filter column. Varying the organic
450 loading rate between 83 and 166 g COD m
-2
d
-1
, only 54% of the organic matter was
451 removed in the filter.
452
453 Jowett and McMaster (1995) experimented with a set-up similar to that used later by
454 Leverenz et al. (2000). They used a single-pass unsaturated biofilter, composed of foam,
455 to treat synthetic domestic effluent at a hydraulic loading rate of 800 L m
-2
d
-1
. They
20

456 found that the system only performed well under conditions of forced ventilation, giving
457 complete nitrification and over 98% organic matter removal. Using natural air
458 convection, nitrification was incomplete (effluent NH
4
-N concentration was 10.2 mg L
-1
),
459 and organic matter removal was at 93%.
460
461 The multi-soil-layering (MSL) method has also been used to purify domestic effluents
462 (Wakatsuki et al., 1993; Luanmanee et al., 2001). It consists of 1.2 m-deep soil-mixture
463 blocks (containing metal iron and pelletised jute; the jute is used as a carbon-source for
464 denitrification) in a brick-like pattern with zeolite interlayers. They are capable of
465 operating under an organic and hydraulic loading rate of 4-15 g BOD
5
m
-2
d
-1
and 100-
466 850 L m
-2
d
-1
, respectively, without significant clogging. Over a 1-year study period, the
467 BOD
5
and Tot-N was reduced by 80-95% and 80-91%, respectively (Wakatsuki et al.,
468 1993). The Tot-N reduction may be due to NH
4
-N absorption onto the zeolite. The MSL
469 system is, however, a mechanical aeration system. It is highly sensitive to aeration
470 (Luanmanee et al., 2001) and excessive dosing can lead to reduced denitrification rates.
471
472 In a study using non-woven textile fabrics (NWTF) as filter media, Leverenz et al. (2000)
473 used 3 filter configurations (hanging sheets of textile fabric, single layer textile fabric
474 chips (approximately 30mm x 20mm in size), and 3 independent layers of textile fabric
475 chips (size not quoted)) to treat domestic-strength wastewater over a period of 7 months.
476 Due to the random stacking arrangement of the textile fabrics, the filter media had a high
477 porosity and hydraulic conductivity. The filters were operated in recirculation mode with
478 recirculation ratios of 9:1 and 3:1 for two hydraulic loading rates of 410 and 1220 L m
-2
21

479 d
-1
, respectively, on the plan surface area of the filter. Using mechanical aeration, the
480 filters produced an effluent that was low in nutrients and organic matter under maximum
481 COD and SS loading rates of 141 and 76 g m
-2
d
-1
, respectively.
482
483 2.4 SSVF CWs and ISF.
484
485 SSVF CWs remove organic matter, SS and nitrify N (Brix et al., 2002; Weedon, 2003)
486 but, similar to ISF, have poor long-term P removal rates (Brix and Arias, 2005). Both
487 systems are limited by the maximum organic loading rate that may be applied to their
488 surfaces. Depending on the strength of the influent wastewater, a maximum organic
489 loading rate of approximately 24 g COD m
-2 d
-1
may be applied in ISF (Rodgers et al.,
490 2005), whereas a maximum organic loading rate of 20 g COD m -2 d -1 is recommended for
491 SSVF CWs (Winter and Goetz, 2003). SSVF CWs are differentiated from ISF by their
492 surface covering of emergent vegetation which affects the infiltration of wastewater into
493 the filter media (Molle et al., 2006).
494
495 3. Treatment of dairy parlour wastewater
496
497 3.1 Constructed wetlands
498
499 To date, FWS CWs are the most widely used, low-cost and low-maintenance alternatives
500 to land spreading of dairy wastewaters in Ireland. Presently, over 13 farms in the Anne
22

501 Valley catchment, Ireland, use FWS CWs for the treatment of agricultural wastewater
502 (Dunne et al., 2005).
503
504 American guidelines for the design loading of SSHF CWs treating agricultural
505 wastewater (NRCS, 1991) recommend an areal loading rate of 7.3 g BOD
5
m
-2 d
-1
;
506 similar rates are used in wetland design for cool temperate climates (Cooper et al., 1996;
507 Dunne et al., 2005). New Zealand guidelines for the disposal of farm dairy wastewaters
508 (Tanner and Kloosterman, 1997) recommended that an FWS CW should only succeed
509 two waste stabilization ponds (an anaerobic and an aerobic pond, respectively) before
510 entering the wetland with an organic loading rate not exceeding 3 g BOD
5
m
-2
d
-1
. The
511 anaerobic pond reduces the BOD
5
and SS, and the aerobic pond carries out further
512 biological reductions. In a study of a FWS CW treating milkroom parlour wastewater in
513 Connecticut, USA, Newman et al. (2000) did not utilize a stabilization pond and found
514 that, under an organic loading rate of approximately 52 g BOD
5
m
-2 d
-1
, 76% of BOD
5
515 was removed and no nitrification occurred. One study (Mantovi et al., 2003) suggests that
516 the milking parlour should be designed to allow parlour washings and washings from the
517 holding room to be separated, so as the wetland only treats the effluent of lower organic
518 and nutrient content.
519
520 Results from these systems have been variable (Table 4). In a study of planted and
521 unplanted SSHF CWs, where the unplanted SSHF CWs acted as an experimental control,
522 Tanner (1995a,b) found that under 5-day carbonaceous biochemical oxygen demand
523 (CBOD
5
) loading rates ranging from 0.9 to 3.4 g CBOD
5
m
-2
d
-1
(unplanted) and 0.9 to
23

524 4.1 g CBOD
5
m
-2
d
-1
(planted), maximum CBOD
5
removals of 85% and 92%,
525 respectively, were measured. Ammonification was more pronounced with increasing
526 retention time, and Tot-N removal varied between 48 and 80 % for planted CWs. Similar
527 organic loading rates were used in a study on a 3-cell integrated FWS CW in Co.
528 Wexford, Ireland (Dunne et al., 2005) where, under organic loading rates varying from
529 2.7 to 3.5 g BOD
5
m
-2 d
-1
, good organic removal was measured but nitrification was not
530 complete during winter.
531
532 Cronk et al. (1994) also found that under reduced retention times (with loading rates of
533 60 g BOD
5
m
-2
d
-1
) BOD
5
and SS concentrations were not reduced to acceptable levels
534 after treatment in a 1-cell FWS CW, and that no significant reduction of TKN occurred
535 (Table 4). In a study on a dairy farm in Drointon in the UK (Cooper et al., 1996), a SSHF
536 CW was used to treat influent with an average BOD
5
concentration of 1192 mg L
-1
. The
537 system initially utilized only the wetland alone and performed poorly under an organic
538 loading rate of approximately 26 g BOD
5
m
-2
d
-1
. However, when two SSVF CWs and a
539 lagoon were installed in front of the SSHF CW, the system had an organic loading rate of
540 approximately 4 g BOD
5
m -2 d -1 and had good organic and SS removal rates but had
541 limited nitrification due to large fluctuations in the inlet wastewater strength (Table 4).
542 Utilizing a similar set-up (2-stage SSVF CW and a SSHF CW), Job (1992) obtained
543 BOD
5
and SS removals of approximately 65% and 90%, respectively, but nitrification
544 was limited (Table 4).
545
24

546 Even under significantly reduced organic loadings, SSHF and FWS CWs have under-
547 performed. In Italy, a study on a 2-cell FWS CW operated in series and monitored over a
548 26 month period, treating a mixture of domestic and dairy parlour washings at an average
549 influent organic loading rate under 2 g BOD
5
m
-2
d
-1
, showed that anoxic zones which
550 developed in the wetland inlet meant that nitrification was inhibited, producing an
551 effluent Tot-N which was mainly composed of NH
4
-N (Mantovi et al., 2003, Table 4).
552
553 3.2. Sand filtration
554
555 ISF treatment systems may be an efficient treatment alternative to conventional land-
556 spreading/reed bed treatment currently carried out in Ireland. To date, sand filters have
557 been used as a cost effective alternative to conventional septic tank/soil adsorption
558 systems for domestic wastewater (Rich, 1996; Loomis and Dow, 1999). However, they
559 are largely untested for the treatment of dairy parlour washings with high BOD
5
.
560
561 Soil filters have been used as an alternative treatment method for agricultural wastewaters
562 (Martinez, 1997); however they are limited by their hydraulic loading capacity. The
563
564 deposition of dispersed clay particles in the pore spaces of soils reduces the hydraulic conductivity and causes clogging to occur. Martinez (1997) developed the ‘Solepur’
565 process wherein an isolated field treatment plant was used to treat pig slurry. However,
566 the practical application of the process is limited by the low hydraulic loading rate of
567 approximately 1.3 L m -2 d -1 .
568
25

569 In ISF of dairy parlour wastewaters, best performance is achieved when the filter is
570 operated in recirculation mode. In a laboratory study, Healy et al. (2004) used a 0.9 m
571 deep sand filter with a recirculation ratio of 3:1; to treat high strength synthetic
572 wastewater (Table 3), the filter comprised stratified layers of coarse (d
10
= 0.45 mm) and
573 fine sand (d
10
= 0.11 mm). They found that it was able to treat synthetic agricultural
574 wastewater with an organic loading rate of 23.4 g COD m
-2
d
-1
and a system hydraulic
575 loading rate of 6.7 L m -2 d -1 . A maximum system hydraulic and organic loading rate of 10
576 L m
-2
d
-1
and 30 g COD m
-2
d
-1
, respectively, produced organic and Tot-N removals of
577 99% and 86%, respectively. Using the same filter operated in single-pass mode, Rodgers
578 et al. (2005) found that at a BOD
5
loading rate of 22 g m
-2
d
-1
, 37% of Tot-N occurred
579 (Table 2).
580
581 In this section, based on laboratory studies using intermittent sand filters for the treatment
582 of synthetic dairy parlour washings (Healy et al., 2004; Rodgers et al., 2004; Rodgers et
583 al., 2005; Healy et al., 2006), the design rationale for a pilot-scale sand filter is discussed
584 (Figure 1 and Table 5). Possible differences that may exist between laboratory synthetic
585 dairy parlour washings and farm dairy parlour washings are acknowledged and
586 conservative design is recommended.
587
588 3.2.1 Farmyard management: house-keeping and water usage
589
590 In farms, volumes of water used for wash-down of yard areas, pipeline cleaning and
591 rinsing and stormwater run-off from soiled yards lead to a final effluent of variable
26

592 volume and concentration. The water volumes generated may vary according to the
593 practices applied on the farms. Factors such as hosing down practices and the number of
594 cows milked affect the volumes generated. The design specification detailed below
595 applies to farms with a water usage not greater than 50 L cow
-1
day
-1
(Department of the
596 Environment and Department of Agriculture, Food and Forestry, 1996). Prior to the
597 installation of a recirculating sand filtration system, water usage should be monitored by
598 the installation of a water meter on the rising main and cleaning practices revised to
599 minimise volumes.
600
601 3.2.2 Organic and SS loading rates
602
603 The operation and performance of sand filters are determined by the organic and
604 suspended solids loading applied to them. A design specification similar to that illustrated
605 in Figure 1 is proposed. This design proposes a holding tank comprising two
606 compartments, a settlement chamber with a retention time of 1 day, into which the dairy
607 parlour and yard washings and three-quarters of the volume of effluent from the sand
608 filter flow and a pumping chamber with a pump assembly and screen. Settled wastewater
609 flows from the settlement chamber through a baffled pipe into the pumping chamber. The
610 contents of the pumping chamber are pumped a minimum of 4 times a day onto the sand
611 filter to give a hydraulic loading of 10 L m
-2
dose
-1
at each pumping (40 L m
-2
d
-1
).
612
613 3.2.3 Sand filter media, type and depth
614
27

615 Design specifications recommend that the sizing of the filter media should be in
616 accordance with its use (USEPA, 1980; Ball and Denn, 1997; Loomis and Dow, 1999).
617 For a recirculating sand filter, sands with a d
10
of 1.5-3.0 mm and a coefficient of
618 uniformity of 1.3-2.5 are recommended. A sand media depth of not less than 0.9 m is
619 recommended. The sand filter must be regularly checked for ponding and ideally the
620 column could be instrumented with tubes for time domain reflectrometry (TDR) probes
621 to monitor the build-up of biomass within the column (Healy et al., 2004; Rodgers et al.,
622 2005).
623
624 3.2.4
Wastewater application to the sand filter, including the loading rate and
625 frequency
626
627 The wastewater is applied to the sand filter via a distribution manifold (Figure 1) at 4
628 times the influent hydraulic flow rate. The distribution manifold should be designed so
629 that the water flow from each orifice should be approximately the same (US EPA, 1982).
630
631 A maximum filter loading rate of 40 L m -2 d -1 of a 3000 mg L -1 COD and 300 mg L -1 Tot-
632 N wastewater was successfully treated in the laboratory (Healy et al., 2006). It is
633 proposed that this design loading be used in the full-scale plant. In laboratory studies,
634 good results were achieved with the application of wastewater in four daily doses to a
635 sand filter (Healy et al., 2004; Rodgers et al., 2005). The pumps were sized to operate for
636 a period of 5 minutes per dose.
637
28

638 3.2.5
Wastewater treatment-post sand filtration
639
640 After passing through the sand filter, the remaining quarter of the effluent from the sand
641 filter flows by gravity onto a woodchippings/sand tank where further denitrification
642 occurs. Healy et al. (2005) found that when a loading rate of 2.9 mg NO
3
-N kg
-1
d
-1
was
643 applied to a woodchippings/sand horizontal flow filter, a 97% reduction in NO
3
-N
644 occurred.
645
646 The outflow from the woodchippings/sand horizontal flow filter can lead to a percolation
647 area in permeable soil or to a suitable watercourse in slow-draining soils. The settlement
648 and pumping chambers must be regularly checked and desludged.
649
650 3.2.6
Actions to be taken in the event of clogging of the sand filter
651
652 If excess organic and suspended solids loading rates are applied to the sand filter,
653 clogging of the surface layer will occur. In a study of a laboratory intermittently loaded
654 stratified sand filter treating synthetic agricultural wastewater, Rodgers et al. (2004)
655 investigated the characteristics of the clogging layer, following the occurrence of surface
656 ponding. Measurements of field-saturated hydraulic conductivity within the upper layer
657 of the sand filter indicated that the effects of the build-up of organic and inert matter may
658 have extended to a depth of 0.165 m into the layer. If clogging occurs, it would be
659 necessary to replace the surface layer of sand.
660
29

661 4. Conclusions and recommendations
662
663 This article reviewed the currently used practices of constructed wetlands, single-pass
664 and recirculating sand filters for the treatment of dairy parlour wastewaters. To date, ISF
665 of high strength wastewaters is an emerging technology; therefore, there is not strong
666 research antecedence in their use in an agricultural setting. While FWS CWs, SSHF and
667 SSVF CWs are regularly used for the treatment of agricultural wastewaters, they appear
668 to be limited by the organic loading rate that may be applied to them. In comparison, ISF,
669 though possibly requiring greater monitoring and maintenance can accept a higher
670 organic loading rate and has a much smaller footprint. Research should be prioritised in
671 the following areas:
672 (1) Pilot-scale studies examining ISF for the treatment of dairy parlour wastewaters should be completed. 673
674 (2) Work needs to be done to optimise the design of CWs – particularly under Irish climatic conditions – to reduce the areal requirement and/or increase the organic 675
676 loading application rate.
677
678 Acknowledgements
679
680 The authors are grateful to Teagasc for the award of a Walsh fellowship to the first author
681 and for financial support for the work. The authors acknowledge the technical help of M.
682 Rathaille, M. O’Brien and G. Hynes (NUI, Galway).
683
30

684 REFERENCES
685
686 Anderson, D.L., Siegrist, R.L., Otis, R.J., 1985. Technology assessment of intermittent
687 sand filters. Report No. PB85-232585. Municipal Facilities Division, Office of Municipal
688 Pollution Control, Washington, D.C. USA.
689
690 Armstrong, W., Armstrong, J. and Beckett, P.M., 1990. Measurement and modelling of
691 oxygen release from roots of Phragmites australis . In: P.F. Cooper, B.C. Findlater (eds),
692 Constructed Wetlands in Water Pollution and Control, Oxford, pp. 41-51. UK.
693
694 Bachand, P.A.M., Horne, A.J., 2000. Denitrification in constructed wetland free-water
695 surface wetlands:II. Effects of vegetation and temperature. Ecol. Eng. 14, 17-32.
696
697 Ball, J.L., Denn, G.D., 1997. Design of recirculating sand filters using a standardized
698 methodology. In: M.S. Bedinger, A.I. Johnson, and J.S. Fleming (eds). Site
699 characterisation and design of on-site septic systems, ASTM STP 898. American Society
700 for Testing and Materials. West Conshohocken, PA, USA, pp. 191-206.
701
702 Boller, M., Schwager, A., Eugster, J., Mottier, V., 1993. Dynamic behaviour of
703 intermittent buried filters. Wat. Sci. and Tech. 28 (10), 99-107.
704
31

705 Bouwer, E.J., Rijnaarts, H.H.M., Cunningham, A.B., Gerlach, R., 2000. Biofilms in
706 porous media. In: J.D. Bryers (ed.), Biofilms II: Process analysis and applications. Wiley,
707 NY, USA, pp. 123 – 158.
708
709 Brix, H., Arias, C.A., Johansen, N.H., 2002. BOD and nitrogen removal from municipal
710 wastewater in an experimental two-stage vertical flow constructed wetland system with
711 recycling. In: Proceedings of the eight international conference on wetland systems for
712 water pollution control. Arusha, Tanzania, 16 – 19 September, pp. 400 – 410.
713
714 Brix, H., Arias, C.A. 2005. The use of vertical flow constructed wetlands for on-site
715 treatment of domestic wastewater: new Danish guidelines. Ecol. Eng. 25, 491-500.
716
717 Carley, B.N., Mavinic, D.S., 1991. The effects of external carbon loading on nitrification
718 and denitrification of a high ammonia landfill leachate. J. Wat. Poll. Cont. Fed. 63, 51–
719 59.
720
721 Carton, O.T. 2001. Environmental pressures and national environmental legisation with
722 respect to nutrient management. In: P De Clercq et al. (ed.), Nutrient Management
723 Legislation in European Countries. Department of Soil Management and Soil Care,
724 Faculty of Agricultural and Applied Biological Sciences. Wageningen, The Netherlands,
725 pp. 89-204.
726
32

727 Cooper, P., Job, G.D., Green, M.B., Shutes, R.B.E., 1996. Reed beds and constructed
728 wetlands for wastewater treatment. WRc, Swindon, UK, 184 pp.
729
730 Craig, R.F. 1997. Soil Mechanics. 7 th
edition. E & FN Spon, London. E&FN Spon,
731 London., UK, 447 pp.
732
733 Crites, R., Lekven, C., Wert, S., Tchobanoglous, G., 1997. A decentralized wastewater
734 system for a small residential development in California. The Small Flows J. 3, 3-10.
735
736 Cronk, J.K, Kodmur, V., Shirmohammadi, A., 1994. An evaluation of wetlands for the
737 treatment of dairy effluent: results from the first year of operations. Presented at the
738 International winter meeting of the American Society of Agricultural Engineers. Pap. No.
739 94-2600, ASAE, St. Joseph, MI, USA, pp. 11-19.
740
741 Darby, J., Tchobanoglous, G., Nor, M.A., 1996. Shallow intermittent sand filtration:
742 performance evaluation. The Small Flows J. 2, 3-14.
743
744 Department of the Environment & Department of Agriculture, Food and Forestry. 1996.
745 Code of good agricultural practice to protect waters from pollution by nitrates.
746 Department of the Environment and Department of Agriculture, Food and Forestry,
747 Dublin, Ireland.
748
33

749
Dunne, E.J., Culleton, N., O’Donovan, G., Harrington, R., Olsen, A.E., 2005. An
750 integrated constructed wetland to treat contaminants and nutrients from dairy farmyard
751 dirty water. Ecol. Eng. 24, 221-234.
752
753 EEC. 1991. Council Directive concerning the protection of waters against pollution
754 caused by nitrates from agricultural sources. Council of European Communities,
755 91/676/01-07.
756
757 Effert, D., Morand, J., Cashell, M., 1985. Field performance of three onsite effluent
758 polishing units. Proceedings of the fourth national symposium on individual and small
759 community sewage systems. New Orleans, LA, USA. December 10-11, 1984. ASAE
760 Publication 07-85, pp. 1-11.
761
762 EPA, 2000. Wastewater treatment manuals: Treatment systems for single houses.
763 Environmental Protection Agency, Wexford, Ireland. 58 pp + appendices.
764
765 Faulkner, S.P., Richardson, C.J., 1989. Physical and chemical characteristics of
766 freshwater wetland soils. In: D.A. Hammer (ed.), Constructed wetlands for wastewater
767 treatment. Lewis Publishers, Chelsea, MI, pp. 71-72.
768
769 Geary, P.M., Moore, J.A., 1999. Suitability of a treatment wetland for dairy wastewaters.
770 Wat. Sci. and Tech. 40 (3), 179-185.
771
34

772 Gold, A.J., B.E. Lamb, G.W. Loomis, J.R. Boyd, V.J. Cabelli, and C.G. McKiel. 1992.
773 Wastewater renovation in buried and recirculating sand filters. J. Env. Qual. 21, 720-725.
774
775 Greenway, M., Woolley, A., 2001. Changes in plant biomass and nutrient removal over 3
776 years in a constructed wetland in Cairns, Australia. Wat. Sci. and Tech. 44 (11-12), 303-
777 310.
778
779 Gross, M., Mitchell, D., 1985. Biological virus removal from household septic tank
780 effluent. Proceedings of the fourth national symposium on individual and small
781 community sewage systems. New Orleans, LA, USA. December 10-11, 1984. ASAE
782 Publication 07-85, pp. 295-304.
783
784 Gross, M. 1990. Virus removal by sand filtration of septic tank effluent. J. Env. Eng. 116,
785 711-720.
786
787 Hack-ten Broeke, M.J.D., 2001. Irrigation management for optimizing crop production
788 and nitrate leaching on grassland. Agric. Wat. Manage. 49, 97-114.
789
790 Hardej, M., Ozimek, T. 2002. The effect of sewage sludge flooding on growth and
791 morphometric parameters of Phragmites australis (Cav.) Trin. Ex Steudel. Ecol. Eng. 18,
792 343-350.
793
35

794 Healy, M.G., Cawley, A.M., 2002. The nutrient processing capacity of a constructed
795 wetland in western Ireland. J. Env. Qual. 31, 1739-1747.
796
797 Healy, M.G., Rodgers, M., Mulqueen, J. 2004. Recirculating sand filters for treatment of
798 synthetic dairy parlor washings. J. Env. Qual. 33, 713-718.
799
800 Healy, M.G., Rodgers, M., Mulqueen, J. 2005. Denitrification of a nitrate-rich synthetic
801 wastewater using various wood-based media materials. J. Env. Sci. & Health – Part A 41,
802 1-7.
803
804 Healy, M.G., Rodgers, M., Mulqueen, J. 2006. Performance of a stratified sand filter in
805 removal of chemical oxygen demand, total suspended solids and ammonia nitrogen from
806 high strength wastewaters. J. Env. Management. In press.
807
808 Henze, M., Harremoes, P., Jansen, J.L.C., Arvin, E., 1997. Wastewater treatment:
809 biological and chemical processes. 2 nd
edition. Springer Publication, Berlin, Germany.
810
811 Job, G.D., 1992. Treatment of medium strength industrial and agricultural effluents using
812 reed bed treatment systems. Ph.D. thesis, University of Birmingham, UK.
813
814 Jowett, E.C., McMaster, M.L., 1995. On-site wastewater treatment using unsaturated
815 adsorbent biofilters. J. Env. Qual. 24, 86-95.
816
36

817 Kadlec, R.H. 1997. An autobiotic wetland phosphorus model. Ecol. Eng. 8, 145-172.
818
819 Kadlec, R.H., Knight, R.L. 1996. Treatment wetlands. CRC Press, Baco Raton, FL, USA.
820 893 pp.
821
822 Karpiscak, M.M., Freitas, R.J., Gerba, C.P., Sanchez, L.R., Shamir, E., 1999.
823 Management of dairy waste in the Sonoran Desert using constructed wetland technology.
824 Wat. Sci. and Tech. 40 (3), 57-65.
825
826 Knight, R.L., Payne Jr., V.W.E., Borer, R.E.,Clarke Jr., R.A., Pries, J.H., 2000.
827 Constructed wetlands for livestock management. Ecol. Eng. 15, 41-55.
828
829 Lamb, B.E., Gold, A.J., Loomis, G.W., McKiel, C.G., 1990. Nitrogen removal for on-
830 site sewage disposal: a recirculating sand filter / rock tank system. Trans. ASAE 33, 525-
831 531.
832
833 Latvala, A. 1993. Multilayer intermittent sand filter. Wat. Sci. and Tech. 28 (3), 125-132.
834
835 Levenspiel, O. 1999. Chemical reaction engineering. 3 rd
edition. John Wiley & sons. New
836 York, USA. 668 pp.
837
37

838 Leverenz, H., Darby, J., Tchobanoglous, G., 2000. Evaluation of textile filters for the
839 treatment of septic tank effluent. Report No. 2000-1. Centre for Environmental and Water
840 Resources Engineering, University of California, Davis. CA, USA.
841
842 Liu, Q., Mancl, K.M., Tuovinen, O.H. 1998. Effect of inoculation on the biodegradation
843 of butterfat-detergent mixtures in fixed-film sand columns, Biores. Tech. 64, 27-32.
844
845 Liu, Q., Mancl, K., Tuovinen, O.H. 2000. High fat wastewater remediation using layered
846 sand filter biofilm systems, In: Proceedings of the international symposium on animal,
847 agricultural and food processing wastes, ASAE, St. Joseph, Michigan, pp. 242-248.
848
849 Liu, Q., Mancl, K., Tuovinen, O.H. 2003. Biomass accumulation and carbon utilization in
850 layered sand filter biofilm systems receiving milk fat and detergent mixtures, Biores.
851 Tech. 89, 275-279.
852
853 Loomis, G., Dow, D., 1999. Guidelines for the design and use of sand filters in critical
854 resource areas. Rhode Island Department of Environmental Management, NY, USA.
855
856 Loudon, T.L., Thompson, D.B., Fay, L., Reese, L.E., 1985. Cold climate performance of
857 recirculating sand filters. Proceedings of the fourth national symposium on individual
858 and small community sewage systems. New Orleans, LA. December 10-11, 1984. ASAE
859 Publication 07-85, pp. 333-342.
860
38

861 Luanmanee, S., Attanandana, T., Masunaga, T., Wakatsuki T., 2001. The efficiency of a
862 multi-soil-layering system on domestic wastewater treatment during the ninth and tenth
863 years of operation. Ecol. Eng. 18, 185-199.
864
865 Luederitz, V., Eckert, E., Lange-Weber, M., Lange, A., Gersberg, R.M., 2001. Nutrient
866 removal efficiency and resource economics of vertical flow and horizontal flow
867 constructed wetlands. Ecol. Eng. 18, 157 – 171.
868
869 Maehlum, T., Jenssen, P.D., Warner, W.S., 1995. Cold-climate constructed wetlands.
870 Wat. Sci. and Tech. 32 (3), 95-101.
871
872 Mantovi, P., Marmiroli, M., Maestri, E., Tagliavini, S., Piccinini, S., Marmiroli, N., 2003.
873 Application of a horizontal subsurface flow constructed wetland on treatment of dairy
874 parlour wastewater. Bioresource Tech. 88, 85-94.
875
876 Martinez, J. 1997. Solepur: a soil treatment process for pig slurry with subsequent
877 denitrification of drainage water. J. Eng. Res. 66, 51-62.
878
879
Molle, P., Liénard, A., Boutin, C., Merlin, G., Iwema, A., 2005. How to treat raw sewage
880 with constructed wetlands: an overview of the French systems. Wat. Sci. and Tech. 35(6),
881 11-21.
882
39

883
Molle, P., Liénard, A., Grasmick, A., Iwema, A., 2006. Effect of reeds and feeding
884 operations on hydraulic behaviour of vertical flow constructed wetlands under hydraulic
885 overloads. Wat. Res. 40, 606 – 612.
886
887 Neralla, S., Weaver, R.W., Lesikar, B.J., Persyn, R.A., 2000. Improvement of domestic
888 wastewater quality by subsurface flow constructed wetlands. Biores. Tech. 75, 19 – 25.
889
890 Newman, J.M., Clausen, J.C., Neafsey, J.A., 2000. Seasonal performance of a wetland
891 constructed to process dairy milkhouse wastewater in Connecticut. Ecol. Eng. 14, 181-
892 198.
893
894 Nichols, J.R., Abboud, N.M., 1995. Recirculating sand filters in Connecticut. Fluid /
895 Particle Separation J. 8, 175-189.
896
897 Nichols, D.J., Wolf, D.C., Gross, M.A., Rutledge, E.M.,1997. Renovation of septic
898 effluent in a stratified filter. In: M.S. Bedinger, J.S. Fleming, and A.I. Johnson (eds), Site
899 characterisation and design of on-site septic systems. ASTM STP 1324. American
900 Society for Testing and Materials. West Conshohocken, PA, USA, pp. 235-247.
901
902 Nielsen, J., Lynggaard-Jensen, A., Hasling, A., 1993. Purification efficiency of Danish
903 biological sand filter systems. Wat. Sci. and Tech. 28 (10), 89-97.
904
40

905 NRCS, 1991. Technical requirements of agricultural wastewater treatment. National
906 Bulletin No. 210-1-17. National Resource Conservation Service, United States
907 Department of Agriculture, Washington, DC, USA.
908
909 Patrick, W.H., Khald, R.A., 1974. Phosphate release and sorption by soils and sediments:
910 effect of aerobic and anaerobic conditions. Science 186, 53-55.
911
912 Platzer, C., Mauch, K., 1997. Soil clogging in vertical flow reed beds – mechanisms,
913 parameters, consequences and… solutions? Wat. Sci. and Tech. 35 (5), 175-181.
914
915 Ran, N., Agami, M., Oron, G., 2004. A pilot study of constructed wetlands using
916 duckweed ( Lemma gibba L.) for treatment of domestic primary effluent in Israel. Wat.
917 Res. 38, 2241 – 2248.
918
919 Reddy, G.B., Hunt, P.G., Phillips, R., Stone, K., Grubbs, A. 2001. Treatment of swine
920 wastewater in marsh-pond-marsh constructed wetlands. 2001. Wat. Sci. and Tech. 44 (11
921 – 12), 545-550.
922
923 Reneau, R.B., Hagedorn, C., Degen, M.J., 1989. Fate and transport of biological and
924 inorganic contaminants from on-site disposal of domestic wastewater. J. of Env. Qual. 18,
925 135-144.
926
41

927 Rich, L.G. 1996. Low-tech systems for high levels of BOD5 and ammonia removal.
928 Public Works-New Jersey 127, 41-42.
929
930 Richards, K., Ryan, M., Coxon, C., 1998. Leaching of N compounds from swards used
931 for dairying that are N-based and irrigated with dirty water/slurry. End of project report.
932 ARMIS 3635. Teagasc. Johnstown Castle Research Centre, Wexford, Ireland.
933
934 Richardson, C.J., and C.B. Craft. 1993. Effective phosphorus retention in wetlands: Fact
935 or fiction. In: G.A. Morshiri (ed.), Constructed wetlands for water quality improvement.
936 Lewis Publishers, London, pp. 271-282.
937
938 Rodgers, M., Mulqueen, J., Healy, M.G., 2004. Surface clogging in an intermittent
939 stratified sand filter. Soil Sci. Soc. Am. J., 68, 1827-1832.
940
941 Rodgers, M., Healy, M.G., Mulqueen, J., 2005. Organic carbon removal and nitrification
942 of high strength wastewaters using stratified sand filters. Wat. Res. 39, 3279-3286.
943
944 Rousseau, D.P.L., Vanrolleghem, P.A., De Pauw, N., 2004. Constructed wetlands in
945 Flanders: a performance analysis. Ecol. Eng. 23, 151-163.
946
947 Ryan, M. 1990. Properties of different grades of soiled water and strategies for safe
948 disposal. Environmental Impact of Landspreading of Wastes. In: Proceedings of the
42

949 symposium held at Johnstown Castle Centre for Soils and Environmental Research and
950 Development, Co. Wexford, Ireland. 30-31 May, 1990. pp. 43-58.
951
952 Schudel, P., Boller, M., 1990. Onsite wastewater treatment with intermittent buried
953 filters. Wat. Sci. and Tech. 22 (3/4), 93-100.
954
955 Schwager, A., Boller, M., 1997. Transport phenomena in intermittent filters. Wat. Sci.
956 and Tech. 35 (6), 13-20.
957
958 Shatwell, T., Cordery, I., 1999. Nutrient storage in urban wetlands. In: B. Ellis (ed.)
959 Impacts of urban growth on surface water and groundwater quality. Publ. 259. IAHS,
960 Wallingford, UK, pp. 339-347.
961
962 Siegrist, R.L. 1987. Soil clogging during subsurface wastewater infiltration as affected by
963 effluent composition and loading rate. J. of Env. Qual. 16, 181-187.
964
965 Siegrist, R.L., Boyle, W.C., 1987. Wastewater-induced soil clogging development. ASCE
966 J.l of Env. Eng. ASCE 113, 550-566.
967
968 Steer, D.N., Fraser, L.H., Seibert, B.A., 2005. Cell-to-cell pollution reduction
969 effectiveness of subsurface domestic treatment wetlands. Biores. Tech. 96, 969 -976.
970
43

971 Stevik, T.K., Aa, K., Ausland, G., Hanssen, J.F., 2004. Retention and removal of
972 pathogenic bacteria in wastewater percolating through porous media: a review. Wat. Res.
973 38, 1355-1367.
974
975 Sz Ö gi, A.A., Vanotti, M.B., Hunt, P.G., 2006. Dewatering of phosphorus extracted from
976 liquid swine waste. Biores. Tech. 97, 183-190.
977
978 Tanner, C.C., Clayton, J.S., Upsdell, M.P., 1995a. Effect of loading rate and planting on
979 treatment of dairy farm wastewaters in Constructed Wetlands-I. removal of oxygen
980 demand, suspended solids and faecal coliforms. Wat. Res. 29, 17-26.
981
982 Tanner, C.C., Clayton, J.S., Upsdell, M.P., 1995b. Effect of loading rate and planting on
983 treatment of dairy farm wastewaters in constructed wetlands-II. removal of nitrogen and
984 phosphorus. Wat. Res. 29, 27-34.
985
986 Tanner, C.C., Kloosterman, V.C., 1997. Guidelines for constructed wetland treatment of
987 farm dairy wastewaters in New Zealand. NIWA Science and Technology Series No. 48.
988 Hamilton, New Zealand.
989
990 Tebbutt, T.H.Y. 1998. Principles of water quality control. 5 th
edition. Butterworth-
991 Heinemann, Oxford, UK.
992
44

993 Toet, S., Van Logtestijn, R.S.P., Schreijer, M., Kampf, R., Verhoeven, J.T.A., 2005. The
994 functioning of a wetland system used for polishing effluent from a sewage treatment
995 plant. Ecol. Eng. 25, 101-124.
996
997
998
Toner, P., Bowman, J., Clabby, K., Lucey, J., McGarrigle, M., Concannon, C.,
Clenaghan, C., Cunningham, P., Delaney, J., O’Boyle, S., MacGarthaigh, M., Craig, M.,
999 Quinn, R. 2005. Water quality in Ireland. Environemental Protection Agency, Co.
1000 Wexford, Ireland.
1001
1002 U.S. Environmental Protection Agency. 1980. Design manual: onsite wastewater
1003 treatment and disposal systems. EPA Report no. 625/1-80-012. Cincinnati, OH, USA.
1004
1005 U.S. Environmental Protection Agency, 1992. Onsite wastewater treatment systems
1006 manual. EPA Report no. 625/R-00/008. Cincinnati, OH, USA.
1007
1008 Van Buuren, J.C.L., Abusam, A., Zeeman, G., Lettinga, G., 1999. Primary effluent
1009 filtration in small-scale installations. Wat. Sci. and Tech. 39 (5), 195-202.
1010
1011 Vandevivere, P., Baveye, P., 1992. Effect of bacterial extracellular polymers on the
1012 saturated hydraulic conductivity of sand columns. Appl. Environ. Microbiol., May 1992,
1013 1690-1698.
1014
45

1015 Vanlandingham, D.S., Gross, M.A., 1998. Contaminant distribution in intermittent sand
1016 filters. ASAE Publication 03/98, 380-387.
1017
1018 Vanotti, M.B., Rice, J.M., Ellison, A.Q., Hunt, P.G., Humenik, F.J., Baird, C.L., 2005.
1019 Solid-liquid separation of swine manure with polymer treatment and sand filtration. Am.
1020 Soc. Agric. Eng. 48, 1567-1574.
1021
1022 Von Felde, K., Kunst, S., 1997. N- and COD-removal in vertical-flow systems. Wat. Sci.
1023 and Tech. 35(5), 79-85.
1024
1025 Vymazal, J., 2002. The use of sub-surface constructed wetlands for wastewater treatment
1026 in the Czech Republic: 10 years experience. Ecol. Eng.18, 633-646.
1027
1028 Vymazal, J., 2005. Horizontal sub-surface flow and hybrid constructed wetlands systems
1029 for wastewater treatment. Ecol. Eng. 25, 478 – 490.
1030
1031 Wakatsuki, T., Esumi, H., Omura, S., 1993. High performance and N & P-removable on-
1032 site domestic waste water treatment system by multi-soil-layering method. Wat. Sci. and
1033 Tech. 27 (1), 31-40.
1034
1035 Weaver, R.W., Lane, J.J., Johns, M.J., Lesikar, B.J., 2001. Uptake of 15N by
1036 macrophytes in subsurface-flow wetlands treating domestic wastewater. Env. Tech. 22,
1037 837 – 843.
46

1038
1039 Weedon, C.M., 2003. Compact vertical flow constructed wetland systems – first two
1040 years’ performance. Wat. Sci. and Tech. 48(5), 15 -23.
1041
1042 Werker, A.G., Dougherty, J.M., McHenry, J.L., Van Loon, W.A., 2002. Treatment
1043 variability for wetland wastewater treatment design in cold climates. Ecol. Eng. 19, 1-11.
1044
1045 White, K.D. 1995. Performance and economic feasibility of alternative on-site
1046 wastewater treatment and disposal options: peat biofilters, constructed wetlands, and
1047 intermittent sand filters. In: Proceedings of the 68 th
Annual Conference and Exposition.
1048 Water Environment Federation, Miami, FL, USA. pp. 129-138.
1049
1050 White, J.S., Bayley, S.E., Curtis, P.J., 2000. Sediment storage of phosphorus in a northern
1051 prairie wetland receiving municipal and agro-industrial wastewater. Ecol. Eng. 14, 127-
1052 138.
1053
1054 Winter, K.J., Goetz, D., 2003. The impact of sewage composition on the soil clogging
1055 phenomena of vertical flow constructed wetlands. Wat. Sci. and Tech. 48(5), 9 – 14.
1056
1057 Wrigley, R. 1994. Managing dairy-shed wastes. Volume 2. Dairy Research and
1058 Development Corporation, Victoria, Australia.
1059
1060
47

1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1061
1062
List of figures.
1063 Figure 1. Illustrated layout of a proposed recirculation sand filter system for treating dairy
1064 wastewaters.
1065
1066
1067
1068
1080
1081
1082
1083
48

Table 1. Data from wetlands treating domestic wastewaters.
____________________________________________________________________________________________________________________________________________________________________________
Reference Location Wetland type
†
Organic loading Pre-treatment Influent Quality Effluent Quality rate (mg L -1 ) (mg L -1 )
______________________________________________________________________________________
BOD
5
COD SS Tot-N NH
4
-N BOD
5
COD SS Tot-N NH
4
-N
____________________________________________________________________________________________________________________________________________________________________________
Healy and Cawley, 2002
Luederitz et al., 2001
Maehlum et al., 1995.
Neralla et al., 2000
Ran et al., 2004
Steer et al., 2005
‡
Ireland FWS
Germany SSVF
Germany SSVF
Norway
USA
Israel
USA
SSHF
SSHF
SSHF
FWS
SSHF
1 g BOD
5
m -2 d -1 Aerator/clarifier 18
21 g BOD
5
m -2 d -1 Anaerobic digester 490
10 g BOD
5
m -2 d -1 2 unaerated ponds 335
4 g BOD
5
m -2 d -1 Septic tank
2 g BOD
5
m -2 d -1 Septic tank
5 g BOD
5
m -2 d -1 Septic tank
16 g BOD
5
m -2 d -1 Storage tank and
200
69
177
142
- sedimentation tank
Septic tank 131
-
816
691
420
-
-
298
-
52
-
-
-
39
26
65
25
29
126
100
91
-
-
-
-
2.2
-
-
-
49
24
51
31
9
24
3
10
10
17
40
23
-
44
12
195
-
-
93
-
5
-
-
-
16
7
13
3
15
66
6
48
-
-
-
-
0.8
-
-
-
32
18
44
2
SSHF - Septic tank 173 - 45 - 41 53 - 8 - 23
____________________________________________________________________________________________________________________________________________________________________________
†
SSHF = subsurface horizontal flow constructed wetland; SSVF = subsurface vertical flow constructed wetland; FWS = free-water surface constructed wetland.
‡
The first wetland has a population equivalent (P.E.) of 1. The second wetland has a P.E. of 6.
49

Table 2.
Filter design and loading rates from previous studies on single – pass sand filters.
__________________________________________________________________________________________________________________________________________________________________
Reference Hydraulic loading rate Wastewater type Frequency
(L m
-2
d
-1
) (times d
-1
)
Filter loading rates Filter depth Filter media
COD SS
__ g m
-2
d
-1
__ (m)
COD
% removals
Tot-N NH
4
-N PO
4
-P SS
__________________________________________________________________________________________________________________________________________________________________
Darby et al. (1996) 81 Domestic 24 11.5 2.3 0.38 Single-layer filter 94.9 - 99.5 - 98.1
326
41
163
24
4
4
47
5.8
23
9.4
1.2
4.7 d
10
= 0.29 mm
Cu= 4.5
71.6
96.3
92.1
-
-
-
95.7
99.7
-
-
90.7
98.2
99.2 - 94.7
98.6 26.1 93.8 Boller et al. (1993) 30
Nichols et al. (1997) 12.5
50
Rodgers et al. (2005) 20
6.7
EPA guidelines
†
<40
Domestic
Domestic
Agricultural
Agricultural
Domestic
6
-
-
4
4
>2
8.0 1.7
3.8 0.2
15 1.0
26
24
9.3
5.3
5.0
3.9
1.0
1.0
0.9 m
0.9 m
0.61-0.91
Single-layer filter 90.5 -
Cu = 4.4
Stratified sand - filter -
-
-
Stratifed sand filter 96
Single-layer filter d
10
= 0.4 - 1 mm
Cu< 4
99
27
37
99
85
37
36
89
-
88 94 100
100 83 100
__________________________________________________________________________________________________________________________________________________________________
†
US EPA (1980), Calculations based on a typical flow of 24 L m
-2
d
-1
with a septic tank effluent COD and SS concentration of 389 and 163 mg L
-1
, respectively (EPA, 1999)
50

Table 3. Filter design and loading rates from previous studies on recirculating sand filters (adapted from Healy et al., 2006).
________________________________________________________________________________________________________________________________________
Reference Forward flow
†
Recirculation Filter loading rates Filter media Filter depth Comments ratio COD SS
L m -2 d -1 g m -2 d -1 m
________________________________________________________________________________________________________________________________________
Crites et al. (1997) 9.6 5:1 ~2 <1 Single-layer filter 1.0 Domestic wastewater. Effluent d
10
= 3 mm; C u
<2 treated to discharge standards.
Loudon et al. (1985) 42 4:1 ~7 2 Single-layer filter 1.0 Domestic effluent. Nitrification d
10
= 0.3mm; C u
=4 incomplete in filter.
Healy et al. (2004) 6.67 4:1 23.4 3.2 Stratified sand filter 1.0 Synthetic agricultural-strength
Top layer: d
10
= 0.45mm; effluent.
Cu: 3
EPA guidelines
‡
120-200 3:1-5:1 34 5.8 Single-layer filter 0.61-0.91 d
10
= 1.0-1.5 mm; C u
<4
________________________________________________________________________________________________________________________________________
†
The forward flow is the hydraulic loading rate on the sand filtration system.
‡
US EPA (1980). Calculations based on a forward flow of 160 L m -2 d -1 at a recirculation ratio of 5:1, and with COD, SS, and TKN concentrations of 35, 6 and 8 mg L -1 in the recirculation tank (using data from Leverenz et al., 2000)
51

Table 4. Data from wetlands treating agricultural wastewaters.
____________________________________________________________________________________________________________________________________________________________________________
Reference Location Wetland type
†
Organic loading Pre-treatment Influent Quality Effluent Quality
rate (mg L -1 ) (mg L -1 )
________________________________________________________________________
BOD
5
SS Tot-N NH
4
-N BOD SS Tot-N NH
4
-N
____________________________________________________________________________________________________________________________________________________________________________
Cooper et al., 1996 Drointon, UK SSHF 4 g BOD
5
m -2 d -1 Anaerobic lagoon + 2 vertical
350 400 - 200 61 100 - 180
Cronk et al., 1994 Maryland, US FWS
Geary and
Moore, 1999
Job, 1992 Rugeley, UK
Karpiscak et al., 1999 Arazona, US
Knight et al., 2000 US
Mantovi et al., 2003 Casina, Italy
Australia FWS
SSHF
FWS
FWS
~60 g BOD
5
m -2 d -1
5.6 g BOD
5
m -2 d -1
- flow reed beds
Settling tank
+ anaerobic pond
Anaerobic lagoon + 2 vertical
Solids separator
7130
220
400
5540 -
- -
150
126
400 - 350 flow reed beds
~12 g BOD
5
m -2 d -1
22 g BOD
5
m -2 d -1
4 anaerobic lagoons
Settling basin
~1.9 g BOD
5
m -2 d -1 Septic tank +
Solids separator
242 911 - 188
442 1111 103 105
451 680 64 22
2730 990
90
150 50
246
28
-
641
141 592
60
-
-
-
- 112
51 42
33
120
105
300
25
___________________________________________________________________________________________________________________________________________________________________________
†
SSHF = subsurface horizontal flow constructed wetland; FWS = free-water surface constructed wetland.
52

Table 5. Sizing a full-scale recirculation sand filter system for treating dairy parlour wastewater based on a laboratory study.
________________________________________________________________________________________________________________________________________________
Herd size Daily wastewater g COD to Min. holding tank Recirculation Sand filter Woodchipping/ Volume flow
† be treated
‡ volume flow
¶ area
¥ sand mass reqd.
§ reqd.
# m 3 d -1 g COD d -1 m 3 m 3 m 2 kg m 3
Q 3Q
________________________________________________________________________________________________________________________________________________
50 2.5 7500 10 7.5 247 5 x 10 4 64
100 5.0 15000 20 15.0 493 1 x 10 5 128
150 7.5 22500 30 22.5 740 1.6 x 10 5 205
200 10 30000 40 30.0 987 2 x 10 5 256
¶
§
#
________________________________________________________________________________________________________________________________________________
†
With each cow producing 50 L d -1 (0.05 m 3 d -1 ) (Department of the Environment and Department of Agriculture, Food and Forestry, 1996)
‡
Assuming a yearly average COD concentration of 3000 mg L -1 .
¥
In laboratory conditions, optimum COD loading rate applied to recirculating sand filter was 30.4 g COD m -2 d -1 under a system hydraulic loading rate of
10 L m -2 d -1 and an average organic carbon concentration of 3045 mg COD L -1 .
Assuming a final effluent from the recirculating sand filter has a NO
3
-N concentration of 60 mg L -1 with each cow producing 50 L d
In the laboratory experiment, a NO
3
-N loading rate of 2.9 x 10 -6 g NO
3
-N g -1 woodchippings/sand mixture produced 98% NO
In the laboratory study the woodchippings/sand mixture had a bulk density of 0.78 g cm -3 , which is applied here.
-1 .
3
-N reduction .
53

Figure 1.
4Q
From dairy farm
Q 3Q
Settlement chamber
Screened pump
Intermittent sand filter
Q Q
Woodchippings/ sand tank
To percolation
54