Brennan_flume

Brennan_flume
1
2
3
4
5
Published as: Brennan, R.B., Fenton, O., Grant, J., Healy, M.G. 2011. Impact of chemical
amendment of dairy cattle slurry on phosphorus, suspended sediment and metal loss to runoff
from a grassland soil. Science of the Total Environment 409: 5111-5118.
doi:10.1016/j.scitotenv.2011.08.016
6
7
Impact of chemical amendment of dairy cattle slurry on phosphorus, suspended sediment
8
and metal loss to runoff from a grassland soil.
9
10
11
R.B. Brennan1, O. Fenton2, J. Grant3, M.G. Healy 1*
12
13
14
15
16
1
17
18
19
20
2
Civil Engineering, National University of Ireland, Galway, Co. Galway, Rep. of Ireland.
Teagasc, Johnstown Castle, Environmental Research Centre, Co Wexford, Rep. of Ireland
3
Teagasc Research Centre, Kinsealy, Co. Dublin.
21
Corresponding author. Tel: +353 91 495364; fax: +353 91 494507. E-mail address:
22
[email protected]
23
24
25
1
26
Abstract
27
28
Emerging remediation technologies such as chemical amendment of dairy cattle slurry have the
29
potential to reduce phosphorus (P) solubility and consequently reduce P losses arising from land
30
application of dairy cattle slurry. The aim of this study was to determine the effectiveness of
31
chemical amendment of slurry to reduce incidental losses of P and suspended sediment (SS)
32
from grassland following application of dairy cattle slurry and to examine the effect of
33
amendments on metal concentrations in runoff water. Intact grassed-soil samples were placed in
34
two laboratory runoff boxes, each 200-cm-long by 22.5-cm-wide by 5-cm-deep, before being
35
amended with dairy cattle slurry (the study control) and slurry amended with either: (i) alum,
36
comprising 8% aluminium oxide (Al2O3) (1.11:1 aluminum (Al):total phosphorus (TP) of slurry)
37
(ii) poly-aluminum chloride hydroxide (PAC) comprising 10% Al2O3 (0.93:1 Al:TP) (iii)
38
analytical grade ferric chloride (FeCl2) (2:1 Fe:TP), (iv) and lime (Ca(OH)2) (10:1 Ca:TP). When
39
compared with the study control, PAC was the most effective amendment, reducing dissolved
40
reactive phosphorus (DRP) by up to 86% while alum was most effective in reducing SS (88%),
41
TP (94%), particulate phosphorus (PP) (95%), total dissolved phosphorus (TDP) (81%), and
42
dissolved unreactive phosphorus (DUP) (86%). Chemical amendment of slurry did not appear to
43
significantly increase losses of Al and Fe compared to the study control, while all amendments
44
increased Ca loss compared to control and grass-only treatment. While chemical amendments
45
were effective, the reductions in incidental P losses observed in this study were similar to those
46
observed in other studies where the time from slurry application to the first rainfall event was
47
increased. Timing of slurry application may therefore be a much more feasible way to reduce
2
48
incidental P losses. Future work must examine the long-term effects of amendments on P loss to
49
runoff and not only incidental losses.
50
51
Keywords: alum; poly-aluminium chloride; lime; ferric chloride; runoff; dairy; slurry;
52
management; grasslands
53
54
Introduction
55
56
Land application of dairy cattle slurry can result in incidental and chronic phosphorus (P) losses
57
to a waterbody (Buda et al., 2009), which may lead to eutrophication (Carpenter et al., 1998).
58
Incidental P losses take place when a rainfall event occurs shortly after slurry application and
59
before slurry infiltrates the soil, while chronic P losses are a long-term loss of P from soil as a
60
result of a build-up in soil test P (STP) caused by application of inorganic fertilisers and manure
61
(Buda et al., 2009). Incidental P losses arising from rainfall events following land application of
62
dairy cattle slurry are the focus of this study.
63
64
Withers et al. (2003) examined the results of a number of studies examining P losses following
65
land application of dairy cattle slurry at different rates and under different climatic conditions
66
(Smith et al., 2001a; Withers et al., 2001; Withers and Bailey, 2003) and found that incidental P
67
losses can account for between 50 and 90% of P losses from land to water. Suspended sediment
68
(SS) losses contribute to particulate phosphorus (PP) in runoff from tillage soils (Regan et al.,
69
2010); however, in grasslands most P loss is in dissolved form with total dissolved phosphorus
70
(TDP) and dissolved reactive phosphorus (DRP) comprising 69% and 60% of total phosphorus
3
71
(TP) load in surface runoff (Haygarth et al., 1998). Incidental SS losses following slurry
72
application can result in high concentrations of SS in runoff, resulting in increased PP losses
73
(Preedy et al., 2001; Withers et al., 2003). This PP can be mineralised and become available to
74
algae (Sharpley, 1993).
75
76
Mitigation methods to reduce incidental P losses include incorporating slurry into soil
77
immediately after land application (Tabbara, 2003), increasing length of buffer zones between
78
slurry application areas and drains and streams (Mayer et al, 2006), enhanced buffers strips
79
(Uusi-Kämppä et al., 2010), timing of slurry application (Hanrahan et al., 2009) and diet
80
manipulation (O’Rourke et al., 2010). The risk of P loss from slurry is strongly related to the
81
water extractable P (WEP) in the slurry (Dou et al., 2003) and amendments which reduce P
82
solubility should reduce P loss to runoff.
83
84
Chemical amendment of slurry using aluminium (Al), iron (Fe), or calcium (Ca) based
85
compounds reduce P solubility in manure (Dao, 1999; Dou et al., 2003; Kalbasi and
86
Karthikeyan, 2004) and reduce P in runoff from plots receiving alum amended poultry litter
87
(Moore and Edwards, 2005) with negligible effect on metal loss (McFarland et al., 2003).
88
Chemical amendments reduce incidental P losses by a combination of the formation of stable
89
metal-phosphorus precipitates (such as Al-P phosphates in the case of alum) and flocculation of
90
the particles in the slurry to form larger particles, which are less prone to erosion
91
(Tchobanoglous et al., 2003). Previous studies have found that there was no risk of increased
92
metal release posed by chemical amendment of poultry litter (Moore et al., 1998), dirty water
93
(McFarland et al., 2003), or horse manure (Edwards et al., 1999). The present study examines the
4
94
effect of chemical amendment of dairy cattle slurry on both P and metal (namely Al, Fe and Ca)
95
losses to runoff. Previous studies have only examined the effect of amendments on P solubility
96
(Dao, 1999; Dao and Daniel, 2002; Dou et al., 2003).
97
98
Chemical amendments can be incorporated into soil to reduce soluble P in soils with high STP
99
(Novak and Watts, 2005), added directly to the manure before land application (Moore et al.,
100
1998), or applied after manure application to reduce P losses in runoff (Torbert et al., 2005).
101
Chemical amendment of poultry litter has been proven to be effective in reducing P losses from
102
poultry litter in the U.S.A. and has been used there for over 30 years (Moore and Edwards,
103
2005). However, there has been limited work involving chemical amendment of dairy manure
104
(Dao, 1999; Dou et al., 2003; Kalbasi and Karthikeyan, 2004). In an incubation study, Dou et al.
105
(2003) found that technical grade alum, added at 0.1 kg/kg (kg alum per kg slurry) and 0.25
106
kg/kg, reduced WEP in swine and dairy slurry by 80% and 99%, respectively. Dao (1999)
107
amended farm yard manure with caliche, alum and flyash in an incubation experiment, and
108
reported WEP reductions in amended manure compared to the control of 21, 60 and 85%,
109
respectively. Kalbasi and Karthikeyan (2004) applied untreated and amended dairy slurry to a
110
soil and incubated it for 2 years; alum and FeCl2 were observed to decrease P solubility, while
111
lime amendments increased WEP.
112
113
The objectives of this study were to investigate (i) the effect of chemical amendments on
114
incidental losses of DRP, TDP, dissolved unreactive P (DUP), PP, TP and SS in runoff from a
115
grassed soil receiving dairy cattle slurry (the study control) or chemically amended dairy cattle
5
116
slurry in a laboratory rainfall simulator and (ii) the effect of amendments on metal concentrations
117
in runoff.
118
119
2. Materials and Methods
120
121
2.1. Soil sample collection and analysis
122
123
Intact grassed-soil samples, 70 cm-long by 30 cm-wide by 10 cm deep, were collected from a
124
dairy farm in Athenry, Co. Galway (53°21’N, 8°34’ W). A second set of soil samples, taken to a
125
depth of 10 cm below the ground surface from the same location, were air dried at 40 °C for 72
126
h, crushed to pass a 2 mm sieve, and analysed for Morgan’s P (the national test used for the
127
determination of plant available P in Ireland) using Morgan’s extracting solution (Morgan,
128
1941). Soil pH (n=3) was determined using a pH probe and a 2:1 ratio of deionised water-to-soil.
129
Particle size distribution was determined using B.S.1377-2:1990 (BSI, 1990a). Organic content
130
of the soil was determined using the loss of ignition test (B.S.1377-3; BSI, 1990b). The soil was
131
a poorly-drained sandy loam (58% sand, 27% silt, 15% clay) with a Morgan’s P of 22±3.9 mg P
132
L-1, a pH of 7.45±0.15 and an organic matter (OM) content of 13±0.1%. The soil had a sandy
133
loam texture, which points to moderate drainage on site. However, medium permeable subsoil
134
limits drainage. Historic applications of organic P from an adjacent commercial-sized piggery
135
have led to high STP in the soil used in this study.
136
137
2.2. Slurry collection and analysis
138
6
139
Cattle slurry from dairy replacement heifers was taken from a farm (53°18’ N, 8°47’ W) in
140
County Galway, Republic of Ireland in Winter (February), 2010. The storage tanks were agitated
141
and slurry samples were transported to the laboratory in 10-L drums. Slurry samples were stored
142
at 4°C. Slurry and amended slurry pH was determined using a pH probe (WTW, Germany) and
143
the WEP of slurry was measured at the time of land application after Kleinman et al. (2007). Dry
144
matter (DM) content was determined by drying at 105 °C for 16 h. The TP of the dairy cattle
145
slurry was determined after Byrne (1979). Total potassium (TK), total nitrogen (TN) and TP
146
were carried out colorimetrically using an automatic flow-through unit (Varian Spectra 400
147
Atomic Absorption instrument). Ammoniacal nitrogen (NH4-N) of slurry and amended slurry
148
was extracted from fresh slurry by shaking 10 g of slurry in 200 ml 0.1 M HCl on a peripheral
149
shaker for 1 h and filtering through No 2 Whatman filter paper.
150
151
2.3. Slurry amendment and runoff set-up
152
153
The results of a laboratory micro-scale study by Brennan et al. (2011) were used to select
154
chemical amendments to be examined in the present study. In addition to a grassed soil-only
155
treatment, five treatments were examined: (i) slurry-only (the study control), (ii) industrial grade
156
liquid alum (Al2(SO4)3.nH2O), comprising 8% aluminium oxide (Al2O3) applied at a rate of
157
1.11:1 (Al:TP) (iii) industrial grade liquid poly-aluminium chloride hydroxide (PAC)
158
(Aln(OH)mCl3n-m) comprising 10% Al2O3 at a rate of 0.93:1 (Al:TP) (iv) analytical grade FeCl2
159
at a rate of 2:1 (Fe:TP), and (v) burnt lime (Ca(OH)2) at a rate of 10:1 (Ca:TP). The rates used
160
were based on the results Brennan et al. (2011).
161
7
162
A batch experiment was also conducted using a range of amendment concentrations to construct
163
a multi-point Langmuir isotherm (McBride, 2000):
164
Ce
x
m
=
1 Ce
+
ab b
(1)
165
where Ce is the concentration of P in solution at equilibrium (mg L-1), x/m is the mass of P
166
adsorbed per unit mass of amendments (g kg-1) at Ce, a is a constant related to the binding
167
strength of molecules onto the amendments, and b is the theoretical amount of P adsorbed to
168
form a complete monolayer on the surface. This provided an estimate of the maximum
169
adsorption capacity of the amendments (g kg-1). The amendments were added at a range of rates
170
to 500 g slurry samples and mixed rapidly for 10 min at 100 rpm using a jar test flocculator. The
171
samples were incubated at 11°C for 24 h. Following incubation, 50 g of slurry/amended slurry
172
was mixed with 250 ml of distilled water. The slurry-water solution was then placed on a
173
reciprocating shaker for 1 h. Samples were centrifuged at 14,000 rpm for 5 min to separate the
174
solids from the solution before being passed through a 0.45 µm filter and the P extract was
175
determined using a Konelab nutrient analyser (Konelab 20, Thermo Clinical Labsystems,
176
Finland).
177
178
The equilibrium P concentration (EPC0) (i.e. the point where no net desorption or sorption
179
occurs) was derived using the following formula (Olsen and Watanabe, 1957):
180
181
S’ = kdC-S0
(2)
182
183
where S’ is the mass of P adsorbed from slurry (mg kg-1), C is the final P concentration of the
184
solution, kd is the slope of the relationship between S’ and C, and S0 is the amount of P originally
8
185
sorbed to the amendment (mg L-1). A slurry sample (from the same storage tank as used in the
186
surface runoff experiments) with a DM of 6%, TP of 550 mg L-1 and WEP of 2.26 g kg-1 was
187
used for the isotherm study. An approximate metal: soluble P ratio for each amendment was
188
calculated using the b term from the Langmuir isotherm and WEP of the slurry. These ratios
189
were equivalent to stoichiometric metal: TP ratios of 0.6:1 compared to 1.1:1 used in the present
190
study for alum and 1.5:1 compared to 0.93:1 for PAC and were generally in agreement with the
191
findings of Brennan et al. (2011), but were not in agreement for FeCl2 (0.4:1 compared to 2:1)
192
and lime (0.9:1 compared to 10:1). The isotherm results indicated that lower application rates
193
should be sufficient to bind P in slurry. However, as the Brennan et al. (2011) study was
194
considered to best replicate surface runoff, it was decided to base the application rates on the
195
results of Brennan et al. (2011) and not the batch test used to develop the Langmuir isotherm. As
196
one of the main aims of the present study was to investigate the effect of amendments on metal
197
release, it was considered to be reasonable and conservative to use results from Brennan et al.
198
(2011). In the case of alum and PAC the rates used were approximately equal to 1:1 metal to TP
199
which was in agreement with Brennan et al (2011) and previous batch studies (Dao and Daniel,
200
2002). In the case of FeCl2 the most efficient rate used in the Brennan et al (2011) study was
201
examined. When lime applied at 1:1 in Brennan et al (2011) study there was no effect; therefore
202
the results of Brennan et al (2011) study were used. As one of the main aims of the present study
203
was to investigate the effect of amendments on metal release, it was considered to be reasonable
204
and conservative to use results from Brennan et al. (2011).
205
206
A laboratory runoff box study was chosen over a field study as it was less expensive and allowed
207
testing under standardized conditions. Such studies are a widely used tool in P transport research
9
208
to compare treatments (Hart et al., 2004). This experiment used two laboratory runoff boxes,
209
200-cm-long by 22.5-cm-wide by 5-cm-deep with side walls 2.5 cm higher than the soil surface,
210
and 0.5-cm-diameter drainage holes located at 30-cm-centres in the base (after Regan et al.,
211
2010). Cheese cloth was placed at the base of each runoff box before placing the sods to prevent
212
soil loss. Intact grassed sods from the study site were transported to the laboratory and stored at
213
11°C in a cold room prior to testing. All experiments were carried out within 14 d of sample
214
collection and tests were conducted in triplicate (n=3). Immediately prior to the start of each
215
runoff box experiment, new sods were trimmed and placed in the runoff box; each slab was
216
butted against its adjacent slab to form a continuous surface. Molten candle wax was used to seal
217
any gaps between the soil and the sides of the runoff box, while the joint between adjacent soil
218
samples did not require molten wax.
219
220
The packed sods were then saturated using a rotating disc, variable-intensity rainfall simulator
221
(after Williams et al., 1997), comprising a single 1/4HH-SS14SQW nozzle (Spraying Systems
222
Co., Wheaton, IL) attached to a 450-cm-high metal frame, and calibrated to achieve an intensity
223
of 1.15±1 cm h-1 and a droplet impact energy of 26 kJ cm-1 ha-1 at 85% uniformity. The sods
224
were then left to drain for 24 h before the experiment commenced; the grassed sods were then
225
assumed to be at an approximate ‘field capacity’ (Regan et al., 2010). Amendments were added
226
to the slurry and mixed rapidly (10 min at 100 rpm) using a jar test flocculator immediately prior
227
to land application. Slurry and amended slurry were applied directly to the surface of the intact
228
grassed soil in runoff boxes at a rate equivalent to 33 m3 slurry ha-1 (26 kg TP ha-1), the rate most
229
commonly used in Ireland (Coulter and Lalor, 2008). During each rainfall simulation event, rain
230
was applied until runoff water flowed continuously and then for 1 h while runoff water samples
10
231
were collected. The drainage holes on the base of the runoff boxes were sealed to better replicate
232
field conditions and to ensure that overland flow occurred. The first rainfall simulation (RS1)
233
commenced 48 h after slurry application, then after a 1 h interval the second rainfall simulation
234
(RS2) commenced. The drainage holes at the bottom of the runoff box were opened for a 24 h
235
interval and then closed when the third rainfall event (RS3) commenced. As the soil samples
236
were taken from the mid-slope of a field with a slope of approximately 5%, it would have been
237
unrealistic to allow the soil to remain water-logged for 24 h between RS2 and RS3. All of the
238
surface runoff was collected at 5-min intervals once runoff began. The source for the water used
239
in the rainfall simulations had a DRP concentration of less than 0.005 mg L-1, a pH of 7.7±0.2
240
and an electrical conductivity (EC) of 0.435 dS m-1. Runoff water pH and EC were measured
241
immediately prior to each event using a pH and EC meter.
242
243
2.4. Sample handling and analysis
244
245
Runoff samples were collected 1 L containers (covered to prevent rain water entering container)
246
at the bottom of the runoff box. Immediately after collection, a subsample of the runoff water
247
was passed through a 0.45µm filter and a sub-sample was analysed colorimetrically for DRP
248
using a nutrient analyser (Konelab 20, Thermo Clinical Labsystems, Finland). A second filtered
249
sub-sample was analysed for TDP using potassium persulfate and sulfuric acid digestion (HACH
250
LANGE, Germany). Unfiltered runoff water samples were also collected and TP was measured
251
using the method used for TDP analysis. Particulate P was calculated by subtracting TDP from
252
TP. The DRP was subtracted from the TDP to give the DUP.
253
11
254
Suspended sediment were determined for all samples by vacuum filtration of well-mixed,
255
unfiltered runoff water through Whatman GF/C (pore size: 1.2 µm) filter paper. All water
256
samples were tested in accordance with standard methods for the examination of water and
257
wastewater (APHA, 2005). In order to address the concern of metal release from amendments,
258
identified by Fenton et al. (2008), it was decided to measure Al, Ca and Fe as these were the
259
active metals in the chemical amendments added to slurry. The metal content was determined
260
using an ICP (inductively coupled plasma) VISTA-MPX (Varian, California). The limit of
261
detection for Al and Fe was 0.01 mg L-1 and 1 mg L-1for Ca.
262
263
2.5. Statistical analysis
264
265
The structure of the experiment was a one-way classification with the rainfall events being
266
repeated measures on each experimental unit. Proc Mixed of SAS (2004) was used to analyse the
267
concentrations of DRP, DUP, PP, TP, SS, Al, Ca and Fe with a covariance structure to account
268
for correlations between the repeated measures. An unstructured covariance model was used for
269
most variables and the outcome was interpreted as a factorial of treatment x event. In all cases,
270
the treatment by event interactions were examined. The data for Al and Fe were censored by a
271
limit of detection and PROC NLMIXED of SAS was used to fit a censored Normal-based model
272
while accounting for the correlations by inducing a compound symmetry structure with a random
273
effect.
274
275
3. Results
276
12
277
3.1. Slurry and amended slurry analysis
278
279
The results of the slurry analysis are shown in Table 1. The slurry sample was typical of slurry
280
found on farms in Ireland (Anon, 2010) with a high DM on the upper limit for land application
281
(Lalor, 2011 per com). The slurry TP and TK remained relatively constant. At the rates used in
282
this study, all of the amendments examined reduced the WEP of dairy cattle slurry by
283
approximately 99% compared to the slurry-control (p<0.001). Alum addition reduced slurry pH
284
from approximately 7.5 (control) to 5.4, PAC reduced pH to 6.4 and FeCl2 to 6.7 (p<0.001),
285
while lime addition increased slurry pH to 12.2 (p<0.001).
286
287
The results of the Langmuir isotherm are shown in Fig. 1. The binding strength of alum and PAC
288
was very high, followed by FeCl2 and lime, which had the lowest binding strength of all
289
amendments examined. The EPC0 was determined graphically for alum and PAC; however, as
290
lime and FeCl2 were not in equilibrium, it was not possible to determine EPC0 (Fig 2).
291
292
3.2. Water quality analysis
293
294
The average flow-weighted mean concentrations (FWMC) of DRP, DUP and PP in runoff for the
295
three rainfall events are shown in Fig. 3. Alum (114 μg DRP L-1) and PAC (89 μg DRP L-1) were
296
more effective at reducing DRP concentration than lime (200 μg DRP L-1) and FeCl2 (200 μg
297
DRP L-1). There was no significant difference in DRP concentrations in the runoff from grass-
298
only and amended plots. At the rates used, all of the treatments examined resulted in DRP
299
concentrations in runoff greater than the maximum allowable concentration (MAC) of 30 μg
13
300
DRP L-1 for surface waters. However, the buffering capacity of water means that the
301
concentration of a surface waterbody will not be as high as the concentration of runoff, provided
302
runoff from slurry flows over soil which has not received dairy cattle slurry (McDowell and
303
Sharpley, 2002).
304
305
The average concentrations of P in runoff water for the 3 rainfall simulation events were 171 µg
306
DRP L-1, 91 µg DUP L-1 and 373 µg TP L-1 for grassed soil-only treatment compared to 655 μg
307
DRP L-1, 1,290 μg DUP L-1 and 8,390 μg TP L-1 for the slurry-control. Incidental DRP and TP
308
concentrations in runoff water following land application of dairy cattle slurry were 5 and 14
309
times greater than those from grassed-soil. In the present study, alum (p<0.001), PAC (p<0.001),
310
lime (p<0.05) and FeCl2 (p<0.05) reduced DRP losses significantly compared to the slurry-
311
control with reductions similar to those observed in the Brennan et al. (2011) study. The results
312
of both studies are tabulated in Table 2. The average FWMC of TDP was significantly reduced
313
compared to the slurry-control. The difference between grass-only, alum and PAC treatments
314
was not significant and the difference between lime and FeCl2 was also not significant. The
315
average FWMC of DUP was also significantly reduced for all treatments compared to slurry-
316
control.
317
318
There was no significant difference between TP in runoff water from grass-only (373 μg L-1) and
319
alum treatments (506 μg L-1). However, there was a significant difference between grass-only
320
and PAC (1,150 μg L-1) (p< 0.001), lime (1,270 μg L-1) and FeCl2 (2,400 μg L-1) treatments for
321
TP (p< 0.001), with a less significant difference between grass-only and PAC (790 μg L-1) and
322
Fe (1,730 μg L-1) for PP (p< 0.001). Therefore, alum was the best amendment at reducing TP and
14
323
PP loss to runoff. Table 2 shows the TP lost in the runoff expressed as a percentage of the slurry
324
applied. The TP losses from control were in agreement with Preedy et al. (2001), who reported
325
that between 6 and 8% of TP applied was lost to runoff. The TP in runoff from the grass-only
326
treatment comprised approximately 47% DRP compared to 69% reported by Haygarth et al.
327
(1998). This difference may be a result of scale effects or differences in experiment design.
328
While chemical amendment of dairy slurry significantly reduced DRP, DUP, PP and TP in
329
runoff water, the proportions of each faction in runoff from alum, PAC and FeCl2 treatments
330
were similar to slurry-control (Fig. 4).
331
332
Suspended sediment was 162 mg L-1 for the grass-only treatment compared to 3,030 mg L-1 for
333
the slurry-control (Fig. 5). The average FWMC of SS in runoff for the three rainfall events are
334
shown in Fig. 4. Alum resulted in the greatest reduction in SS (an average of 88% for the three
335
rainfall events compared to the slurry-control) (p<0.001). There was no statistical difference in
336
average FWMC of SS between alum, PAC (83% reduction) and lime (82%). All of the
337
treatments resulted in SS concentrations in the runoff which were significantly greater than the
338
grass-only treatment (p<0.005).
339
340
3.3. Metals in runoff water
341
342
The average FWMC of Al, Ca and Fe for the 3 rainfall simulation events are shown in Figs. 6, 7
343
and 8. The average concentrations of metals tested in runoff water for the 3 rainfall simulation
344
events were greater for the slurry-control than the grass-only treatment. Aluminium
15
345
concentrations increased from 60 to 91 µg Al L-1 (not statistically significant), calcium from 84
346
to 108 mg Ca L-1 (p<0.01), and Fe increased from 71 to 151 µg Fe L-1 (p=0.02, RS2).
347
348
The FWMC of Al decreased for all treatments compared to the slurry-control (Fig. 6). There was
349
a significant treatment x event interaction (p<0.001) and differences between events within
350
treatments and between treatments within events were tested. After multiple comparison
351
adjustments, there were no statistically significant differences between treatments. There were
352
some significant decreases to the RS3 event compared to RS1 and RS2 for the lime and slurry-
353
control treatments (p =0.03 and p =0.006). The FWMC of Ca in runoff from all chemically
354
amended slurry treatments was significantly greater than from the slurry-control and the grass-
355
only treatment (p<0.01) (Fig. 7).
356
357
The treatment x event interaction was significant and while no treatments were statistically
358
different across all events, there were some differences between the grass treatment and both
359
alum (p=0.02, RS1) and the slurry-control (p=0.02, RS2), and also between the FeCl2 and slurry-
360
control (p=0.02, RS2).
361
362
4. Discussion
363
364
4.1. Slurry and amended slurry analysis
365
366
The amendments examined significantly reduced WEP in amended slurry compared to the
367
control. This was in agreement with previous studies (Dao, 1999; Dou et al., 2003). Lefcourt and
16
368
Meisinger (2001) reported a 97% reduction in WEP of dairy cattle slurry when 2.5% by weight
369
of alum was added in a laboratory batch experiment. Dao and Daniel (2002) added alum (810 mg
370
Al L-1) and ferric chloride (810 mg Fe L-1) (compared to 1250 mg Al L-1 and 2280 mg Fe L-1 in
371
this study) to dairy slurry and observed that slurry WEP was reduced by 66 and 18%,
372
respectively. At higher application ratios of metal-to-TP, this study showed that greater
373
reductions in WEP are achievable.
374
375
The amendments also changed the pH of the slurry. Lime addition increased slurry pH
376
significantly, resulting in a 25 and 30% reduction in NH4-N and TN of slurry following
377
amendment and mixing (Table 1). This was similar to findings of a study by Molloy and Tunney
378
(1983), who reported an increase in pH to 7.8 and a 50% increase in ammonia (NH3) loss when
379
CaCl2 was added to dairy slurry. This loss in NH4-N was most likely due to NH3 volatilisation,
380
as depending on the pH of a solution, NH4-N can occur as NH3 gas or the ammonium ion (NH4)
381
(Gay and Knowlton, 2005). This reduces the fertiliser value of the slurry and increases NH3
382
emissions from slurry. Addition of alum, PAC and FeCl2 to dairy cattle slurry significantly
383
reduced pH, as expected. This phenomenon has been reported by a number of studies examining
384
the use of amendments to reduce NH3 losses from dairy cattle slurry (Meisinger et al., 2001; Shi
385
et al., 2001). Meisinger et al. (2001) reported a 60% reduction in NH3 loss from dairy cattle
386
slurry when 2.5% by weight of alum was added in a laboratory batch experiment. In a field
387
study, Shi et al. (2001) reported a 92% reduction in NH3 loss. Moore and Edwards (2005) have
388
shown that chemical amendment improves yields due to increased N efficiency. Future work
389
must examine the impact of amendments on gaseous emissions and the risk of ‘pollution
390
swapping’ (the increase in one pollutant as a result of a measure introduced to reduce a different
17
391
pollutant) (Stevens and Quinton, 2008), which must be considered when evaluating amendments
392
for possible recommendations to legislators.
393
394
4.2. Water quality
395
396
The DRP and TP concentrations in runoff water from grass only treatment was well in excess of
397
the MAC of 30 μg DRP L-1 (Flanagan, 1990) and 25-100 μg TP L-1 (USEPA, 1986) for fresh
398
waterbodies.
399
400
This study validated the results of a micro-scale study (Brennan et al., 2011) at meso-scale and
401
demonstrated that PAC is the most effective chemical amendment to reduce incidental DRP
402
losses, with alum being most effective at reducing DUP, PP, TP and SS losses arising from land
403
application of dairy cattle slurry. A limited number of runoff studies have been carried out with
404
chemical amendment of dairy cattle slurry (Elliot et al, 2005; Torbert et al., 2005) and swine
405
slurry (Smith et al., 2001b). Torbert et al. (2005) amended landspread composted dairy manure
406
with ferrous sulphate, gypsum and lime (each at 3:1 metal-to-TP ratio) immediately prior to a 40-
407
min rainfall event with overland flow equivalent to a rainfall intensity of 12.4 cm h-1. Ferrous
408
sulphate reduced DRP loss by 66.3%, while gypsum and lime amendments increased DRP loss.
409
Lime and gypsum were effective for a short time at the beginning of the event and the authors
410
recommended that lime could be used in areas with infrequent and low volume runoff events. In
411
the Torbert et al. (2005) study, amendments were surface applied to slurry immediately after
412
slurry application and just before the first rainfall simulation event occurred. The differences
413
between the results are likely due to a combination of the shorter contact time with lime before
18
414
the first rainfall event and less mixing due to different amendment application methods used in
415
each study. In a plot study, Smith et al. (2001b) amended swine manure with alum and AlCl3 at
416
two stoichiometric ratios (0.5:1 and 1:1 Al: TP). Dissolved reactive phosphorus reductions for
417
alum and AlCl3 at the lower ratio were 33 and 45%, respectively, with 84% for both amendments
418
at the higher ratio, which was similar to reductions observed in the current study.
419
420
The reductions in P losses in the present study were similar to the percentage reductions obtained
421
in other incidental P loss mitigation studies. Hanrahan et al. (2009) reported that incidental TP
422
and DRP losses were reduced by 89 and 65%, respectively, by delaying rainfall from 2 to 5 days
423
after dairy cattle slurry application. This was in agreement with results of O’Rourke et al. (2010).
424
In a plot study, McDowell and Sharpley (2002) applied dairy cattle slurry at 75 m3 ha-1 to the
425
upper end of plots with lengths varying from 1 to 10 m. Increasing the distance from the location
426
where dairy slurry was applied to the runoff water collection point was shown to reduce
427
incidental P concentrations in overland flow by between 70 and 90% when plots were subjected
428
to simulated rainfall with an intensity of 70 mm h-1. Therefore, as there are less expensive
429
methods which can achieve similar reductions in incidental P losses, in future the focus of
430
chemical amendment studies must be to find amendments to bind P in soil with the aim of
431
reducing chronic P losses.
432
433
In order to minimise the effect of the larger variation in the study control than in runoff from
434
grass-only and amended slurry runoff boxes and to detect differences between treatments, the
435
slurry-control was excluded from the statistical analysis of TP and PP. The reduction in TP and
436
PP losses when alum, PAC and FeCl2 was added to slurry was a result of a combination of
19
437
precipitation and floc formation, which led to a decrease in SS loss in runoff water. In the case of
438
lime addition, the reductions were a result of the formation of Ca-P precipitates. The average
439
FWMC of TP for the slurry-control during the three rainfall simulation events was 8,390 μg L-1.
440
This was similar to 7,000 μg L-1 reported by Preedy et al. (2001) in a rainfall simulation study to
441
examine incidental P loss from dairy slurry.
442
443
Measures such as increasing the time between slurry application and the first rainfall event are as
444
effective as chemical amendment at reducing incidental losses of P. Chemical amendment
445
immobilises soluble P in slurry applied to soil and could therefore be included as a low capital
446
cost management tool to reduce farm P status and chronic P losses. The cost of chemical
447
amendments in comparison to other treatment methods (e.g. transporting to other farms,
448
anaerobic digestion, separation and composting) is likely to be the most significant factor in the
449
future implementation of chemical amendments. Economies of scale were not considered in this
450
study and this could considerably reduce costs. The cost of amendment, calculated after Brennan
451
et al. (2011), based on the estimated cost of chemical, chemical delivered to farm, addition of
452
chemical to slurry, increases in slurry agitation, and slurry spreading costs as a result of
453
increased volume of slurry due to the addition of the amendments to slurry, is shown in Table 2.
454
At the scale of the present study, alum and ferric chloride provide the best value in reducing on
455
TP loss from slurry. These are preliminary estimates and if the cost of using these amendments
456
as a mitigation measure is to be accurately calculated, then the optimum dosage for each
457
amendment at field-scale needs to be determined.
458
459
4.3. Metals in runoff water
20
460
461
Previous studies (Moore et al., 1998; Edwards et al., 1999) have reported that chemical
462
amendment of poultry litter posed no significant risk of increased metal release to runoff water.
463
The findings of the present study also validate this for chemical amendment of dairy cattle slurry.
464
Moore et al. (1998) associated an increase in Ca release from alum treatment to a displacement
465
of Ca in Ca-P bonds by Al. This is also likely to be the cause for PAC and FeCl2 with Ca
466
displaced by Al and Fe. The increase in Ca from the lime treatment was expected as a high rate
467
of lime was applied. The FWMC of Fe (Fig. 8) decreased for all treatments except alum, which
468
increased Fe loss by 30% compared to the slurry-control; this was most likely a result of pH
469
effect of alum, which increased the Fe solubility leading to higher Fe losses. There are acute
470
(acute concentrations being short-term concentration and chronic being a long-term
471
concentration) MAC (750 μg L-1) and chronic MAC (87 μg L-1) for Al in runoff (USEPA, 2009).
472
The Al concentrations observed in the present study were below all MAC with the exception of
473
slurry-control during RS2 and grass-only treatment in RS2, which exceeded chronic MAC. There
474
is no MAC for Ca in water. Iron concentrations in runoff were all below the chronic MAC of
475
1,000 μg L-1 (USEPA, 2009).
476
477
From previous studies, adverse effects are not expected due to alum amendment to manure. In a
478
plot study, Moore el al. (1998) amended poultry litter with alum to examine the effect of alum
479
amendment on runoff concentrations of metals. Alum treatment significantly reduced Fe in
480
runoff. Runoff Al concentrations were not affected by treatment and Ca concentrations increased
481
after treatment. Moore et al. (2000) also found Al loss from a small-scale catchment was
482
unaffected by alum treatment. In order to determine the effect of long-term additions of alum to
21
483
poultry litter, Moore and Edwards (2005) began a 20-yr study in 1995. The most significant
484
findings of this study were that long-term land application of alum-amended poultry litter did not
485
acidify soil in the same way as NH4-N fertilisers and that Al availability was lower from plots
486
receiving alum-treated poultry manure than NH4-N fertiliser. McFarland et al. (2003)
487
incorporated alum into soil prior to application of dairy dirty water and reported no difference in
488
Al concentrations in runoff between control and alum amended plots.
489
490
5. Conclusion
491
492
The results of this study demonstrate that chemical amendment was very successful in reducing
493
incidental losses of DRP, TP, PP, TDP, DUP and SS from land-applied slurry. The results of the
494
study demonstrate that PAC was the most effective amendment for decreasing DRP losses in
495
runoff following slurry application, while alum was the most effective for TP and PP reduction.
496
Incidental loss of metals (Al, Ca and Fe) from chemically amended dairy cattle slurry was below
497
the MAC for receiving waters. Future research must examine the long-term effect of
498
amendments on P loss to runoff, gaseous emissions, plant availability of P and metal build-up in
499
the soil. If amendments to slurry are to be recommended and adopted as a method to prevent P
500
losses in runoff, the impact of such applications on slurry-borne pathogens, as well as pathogen
501
translocation to the soil and release in surface runoff, needs to be addressed. The long-term
502
effects on microbial communities in soil must also be examined. The results of this study show
503
that even with chemical amendment, P concentration in runoff was above the MAC. Therefore,
504
amendments may not be the best option for minimising incidental P losses, as timing of
505
applications may be just as effective at controlling incidental P losses, and may be much more
22
506
cost effective. However, chemical amendment immobilises soluble P in slurry and has the
507
potential to reduce chronic P losses. The use of chemical amendments in combination with other
508
mitigation methods such as grass buffer strips would likely increase the effectiveness of the
509
measures. Future work should focus on using amendments to reduce P solubility in slurry to
510
decrease P loss from high P soils by binding P in slurry once it is incorporated into the soil,
511
thereby allowing farmers to apply slurry to soil without further increasing the potential for P loss.
512
513
Acknowledgements
514
515
The first author gratefully acknowledges the award of a Walsh Fellowship by Teagasc to support
516
this study. The authors are also grateful for assistance provided by Teagasc and NUI Galway
517
staff and colleagues with special mention to Peter Fahy, Ana Serrenho, Aoife Keady, John
518
Regan, Sean Murphy, Tony Murphy, Theresa Cowman, Denis Brennan, Linda Moloney-Finn
519
and Maria Radford.
520
521
522
523
524
525
526
527
528
23
529
References
530
531
Anon, 2010. S.I.610. 2010. European Communities (Good Agricultural Practice for Protection of
532
Waters) Regulations 2010. Available on 25th January 2011 at:
533
http://www.environ.ie/en/Legislation/Environment/Water/FileDownLoad,25133,en.pdf.
534
535
APHA. Standard methods for the examination of water and wastewater. American Public Health
536
Association (APHA). 1995; APHA, Washington.
537
538
Brennan RB, Fenton O, Rodgers M, Healy MG.. Evaluation of chemical amendments to control
539
phosphorus losses from dairy slurry. Soil Use Man. 2011; 2011; 27: 238-246.
540
541
British Standards Institution. 1990a. British standard methods of test for soils for civil
542
engineering purposes. Determination of particle size distribution. BS 1377:1990:2. BSI, London.
543
544
British Standards Institution. 1990b. Determination by mass-loss on ignition. British standard
545
methods of test for soils for civil engineering purposes. Chemical and electro-chemical tests. BS
546
1377:1990:3. BSI, London.
547
548
Buda AR, Kleinman PJ, Bryant RB, Feyereisen GW. Effects of hydrology and field management
549
on phosphorus transport in surface runoff. J Environ Qual 2009;38:2273-2284.
550
24
551
Byrne E. Chemical analysis of agricultural materials – methods used at Johnstown Castle
552
Research Centre, Wexford. Published by An Foras Taluntais.1979; 194 pages.
553
554
Carpenter SR, Caraco NF, Correll DL, Howarth RW, Sharpley AN, Smith VH. Nonpoint
555
pollution of surface waters with phosphorus and nitrogen. Ecol. Appl. 1998; 8:559-568.
556
557
Coulter BS, Lalor S. Major and micro nutrient advice for productive agricultural crops. 3ed
558
Teagasc Johnstown Castle Wexford. 2008. P. 116.
559
560
Dao TH. Co-amendments to modify phosphorus extractability and nitrogen/phosphorus ration in
561
feedlot manure and composted manure. J. Environ. Qual. 1999;28:1114-1121.
562
563
Dao TH, Daniel TC. Particulate and dissolved phosphorus chemical separation and phosphorus
564
release from treated dairy manure. J. Environ. Qual. 2002;31:1388-1398.
565
566
Dou Z, Zhang GY, Stout WL, Toth JD, Ferguson JD. Efficacy of Alum and Coal Combustion
567
By-Products in Stabilizing Manure Phosphorus. J. Environ. Qual. 2003;32:1490-1497.
568
569
Edwards DR, Moore PA, Workman SR, Bushee EL. Runoff of metals from alum-treated horse
570
manure and municipal sludge. J. Am. Water Res. Assoc. 1999;35:155-165.
571
572
Elliott H, Brandt R, O'Connor GA. Phosphorus losses from surface-applied biosolids. J Environ
573
Qual. 2005;34:1632-1639.
25
574
575
Fenton O, Healy M. Schulte R. A review of remediation and control systems for the treatment of
576
agricultural waste water in Ireland to satisft the requirements of the Water Framework Directive.
577
Biology & Environment: Proceedings of the Royal Irish Academy 2008;108(2), 69-79.
578
579
Flanagan PJ. Water quality regulations are legal guidelines used to safeguard public health.
580
‘Parameters of Water Quality’, Environmental Research Unit, Dublin 4. 1990.
581
582
Gay SW, Knowlton KF. Ammonia Emissions and Animal Agriculture. Virginia Cooperative
583
Extension. Biological Systems Engineering Publication No. 2005;442-110.
584
585
Hart MR, Quin BF, Nguyen ML. Phosphorus Runoff from agricultural land and direct fertilizer
586
effects. J. Environ. Qual. 2004;33:1954-1972.
587
588
Hanrahan LP, Jokela WE, Knapp JR. Dairy Diet Phosphorus and Rainfall Timing Effects on
589
Runoff Phosphorus from Land-Applied Manure. J Environ Qual. 2009;38(1), 212-217.
590
591
Haygarth PM, Hepworth L, Jarvis SC. Forms of phosphorus transfer in hydrological pathways
592
from soil under grazed grassland. Eur. J. Soil Sci. 1998;49:65-72.
593
594
Kalbasi M, Karthikeyan KG. Phosphorus dynamics in soils receiving chemically treated dairy
595
manure J. Environ. Qual. 2004;33:2296-2305.
596
26
597
Kleinman PJA, Sullivan D, Wolf A, Brandt R, Dou Z, Elliott H, Kovar J, Leytem A, Maguire R.
598
Moore, P, Saporito L, Sharpley AN, Shober A, Sims T, Toth J, Toor G, Zhang H, Zhang T.
599
Selection of a water extractable phosphorus test for manures and biosolids as an indicator of
600
runoff loss potential. J. Environ. Qual. 2007;36:1357-1367.
601
602
Lefcourt AM, Meisinger JJ. 2001. Effect of adding alum or zeolite to dairy slurry on ammonia
603
volatilisation and chemical composition. J. Dairy Sci. 2001;84:1814-1821.
604
605
Mayer PM, Reynolds SK, McCutchen MD, Canfield TJ. Riparian Buffer Width, Vegetative
606
Cover, and Nitrogen Removal Effectiveness: A Review of Current Science and Regulations.
607
EPA/600/R-05/118. Cincinnati, OH, U.S. Environmental Protection Agency. Available at:
608
<http://www.epa.gov/nrmrl/pubs/600R05118/600R05118.pdf> verified 12 July 2011. , 2006.
609
610
McBride, M.B. 2000. Chemisorption and precipitation reactions. p. B-265 – B-302. In: M.E.
611
Sumner (ed). Handbook of Soil Science. CRC Press. Boca Raton, Fl.
612
613
McDowell, R.W. and Sharpley, A.N. (2002) Effect of plot scale and an upslope phosphorus
614
source on phosphorus loss in overland flow, pp. 112-119, Blackwell Publishing Ltd.
615
616
McFarland, A.M.S., L.M. Hauck and A.P. Kruzic. Phosphorus reductions in runoff and soils
617
from land-applied dairy effluent using chemical amendments: An observation. The Texas Journal
618
of Agriculture and Natural Resources. 16:47-59 available at
27
619
http://www.tarleton.edu/Departments/txjanr/Volumes/Vol%2016%20-
620
%202003/V16_03Art07.pdf. 2003.
621
622
Meisinger JJ, Lefcourt AM, Thompson RB. Construction and validation of small mobile wind
623
tunnels for studying ammonia volatilisation Appl. Eng. Agric. 2001;17:375-381.
624
625
Moore PA, Daniel TC, Gilmour JT, Shreve BR, Edwards DR,Wood BH. Decreasing metal
626
runoff from poultry litter with aluminum sulphate. J. Environ. Qual. 1998;27: 92-99.
627
628
Moore PA, Edwards DR. Long-term effects of poultry litter, alum-treated litter and ammonium
629
nitrate on aluminium availability in soils. J. Environ. Qual. 2005;34:2104-2111.
630
631
Moore PA, Daniel TC, Edwards DR. Reducing phosphorus runoff and inhibiting ammonia loss
632
from poultry manure with aluminium sulphate. J. Environ. Qual. 2000;29:37-49.
633
634
Morgan MF. Chemical soil diagnosis by the Universal Soil Testing System. Connecticut.
635
Connecticut agricultural Experimental Station Bulletin 450 Connecticut. New Haven. 1941.
636
637
Molloy SP, Tunney H. A Laboratory Study of Ammonia Volatilization from Cattle and Pig
638
Slurry. Irish Journal of Agricultural Research 1983;22(1), 37-45.
639
28
640
Novak JM,Watts DW. An alum-based water treatment residual can reduce extractable
641
phosphorus concentrations in three phosphorus-enriched coastal plain soils. J. Environ. Qual.
642
2005;34:1820-1827.
643
644
Olsen SR, Watanabe FS. A method to determine a phosphorus absorption maximum of soils as
645
measured by the Langmuir isotherm. Soil Science Society Proceedings 1957; 31: 144-149.
646
647
O'Rourke SM, Foy RH, Watson CJ, Ferris CP, Gordon A. Effect of varying the phosphorus
648
content of dairy cow diets on losses of phosphorus in overland flow following surface
649
applications of manure. J Environ Qual 2010;39(6), 2138-2146.
650
651
Preedy N, McTiernan K, Matthews R, Heathwaite L, Haygarth P. Rapid incidental phosphorus
652
transfers from grassland. J. Environ. Qual. 2001;30:2105-12.
653
654
USEPA. United States environmental protection agency. Office of water regulations and
655
standards, DC 20460 1986. Verified 12 April 2011. Available at:
656
http://water.epa.gov/scitech/swguidance/standards/criteria/aqlife/upload/2009_01_13_criteria_go
657
ldbook.pdf
658
659
USEPA. United States Environmental Protection Agency. Office of water, science and
660
technology. National recommended water quality criteria for priority pollutants. 2009. Verified
661
12 April 2011. Available at:http://water.epa.gov/scitech/swguidance/standards/current/index.cfm
662
29
663
Uusi-Kämppä, J., Turtola, E., Närvänen, A., Jauhiainen, L. and Uusitalo, R. (2010) A
664
preliminary study on buffer zones amended with P-binding compounds. Turtola, E., Ekholm, P.
665
and Chardon, W. (eds).
666
667
Regan JT, Rodgers M, Healy MG, Kirwan L, Fenton O. 2010. Determining phosphorus and
668
sediment release rates from five Irish tillage soils. J. Environ. Qual. 2010;39:1-8.
669
670
SAS Institute. SASV9.1. SAS/STAT® User’s Guide. Cary, NC. , SAS Institute Inc. 2004.
671
672
Sharpley AN. Assessing phosphorus bioavailability in agricultural soils and runoff. Nutr. Cycl.
673
Agroecos. 1993; 36:3:259-272.
674
675
Shi Y, Parker DB, Cole NA, Auvermann BW, Mehlhorn JE. Surface amendments to minimise
676
ammonia emissions from beef cattle feedlots. American Society of Agricultural Engineers
677
2001;44(3), 677-682.
678
679
Smith KA, Jackson DR, Withers PJA. Nutrient losses by surface run-off following the
680
application of organic manures to arable land. 2. Phosphorus. Environmental Pollution
681
2001a;112(1), 53-60.
682
683
Smith DR, Moore PA Jr, Griffis CL, Daniel TC, Edwards DR, and Boothe DL. Effects of alum
684
and aluminium chloride on phosphorus runoff from swine manure. J. Environ. Qual. 2001b;
685
30:992-998.
30
686
687
Stevens CJ, Quinton JN. Policy implications of pollution swapping. Physics and Chemistry of
688
the Earth, Parts A/B/C. 2009;34(8-9), 589-594.
689
690
Tabbara H. Phosphorus Loss to Runoff Water Twenty-Four Hours after Application of Liquid
691
Swine Manure or Fertilizer. J. Environ. Qual. 2003; 32: 1044-1052.
692
693
Tchobanoglous, G., Burton, F.L., Stensel, H.D. Wastewater engineering treatment and reuse, 4th
694
Ed. McGraw-Hill: Boston, 2003; 1848 pp.
695
696
Torbert HA, King KW, Harmel RD. Impact of soil amendments on reducing phosphorus losses
697
from runoff in sod. J. Environ. Qual. 2005;34:1415-1421.
698
699
Tunney H. A Note on a Balance Sheet Approach to Estimating the Phosphorus Fertiliser Needs
700
of Agriculture. Irish Journal of Agricultural Research. 1990;29:2:149-154.
701
702
Tunney H. Phosphorus needs of grassland soils and loss to water. In Steenvoorden J, Claesses F,
703
Willems J. (Eds.), International conference on agricultural effects on ground and surface waters.
704
International association of hyrdogical sciences, Wageningen, Netherlands. 2000. p. 63-69.
705
706
Williams JD, Wilkins DE, McCool DK, Baarstad LL, Klepper BL, Papendick RI. A new rainfall
707
simulator for use in low-energy rainfall areas. Appl. Eng. Agric. 1998;14:243-247.
708
31
709
Withers PJA and Bailey, G.A. (2003) Sediment and phosphorus transfer in overland flow from a
710
maize field receiving manure. Soil Use and Management 19(1), 28-35.
711
712
Withers, P.J.A., Clay, S.D. and Breeze, V.G. (2001) Phosphorus Transfer in Runoff Following
713
Application of Fertilizer, Manure, and Sewage Sludge. J. Environ. Qual. 30(1), 180-188.
714
715
Withers, P.J.A., Ulén, B., Stamm, C. and Bechmann, M. (2003) Incidental phosphorus losses –
716
are they significant and can they be predicted? Journal of Plant Nutrition and Soil Science
717
166(4), 459-468.
718
719
720
721
722
723
724
725
726
727
728
729
730
731
32
732
List of figures
733
734
Fig. 1 Langmuir isotherm fitted to phosphorus in amended slurry data.
735
736
Fig. 2 Phosphorus sorption isotherm for amended slurry data.
737
738
Fig. 3 The average flow weighted mean concentration of dissolved reactive phosphorus (DRP),
739
dissolved un-reactive phosphorus (DUP) and particulate phosphorus (PP), which comprise total
740
phosphorus (TP) in runoff from each rainfall simulation event.
741
742
Fig. 4 Average flow weighted mean concentrations of suspended sediment in runoff.
743
744
Fig. 5 The average % of dissolved reactive phosphorus (DRP), dissolved un-reactive phosphorus
745
(DUP) and particulate phosphorus (PP), which comprise total phosphorus (TP) in runoff after
746
three rainfall simulation events.
747
748
Fig. 6 Average flow weighted mean concentrations of Al in runoff and rain water.
749
750
Fig. 7 Average flow weighted mean concentrations of Ca in runoff and rain.
751
752
Fig. 8 Average flow weighted mean concentrations of Fe in runoff and rain.
753
754
33
Ce/(x/m) (mg L-1)
Ce (mg L-1)
755
Fig. 1 Langmuir isotherm fitted to phosphorus in amended slurry data
756
757
758
759
760
761
762
763
764
765
766
767
34
S’ (mg kg-1)
Ce (mg L-1)
768
769
Fig 2 Phosphorus sorption isotherm for amended slurry data.
770
771
772
773
774
775
776
777
778
779
780
35
781
782
783
784
785
Fig 3 The average flow weighted mean concentration of dissolved reactive phosphorus (DRP), dissolved unreactive
phosphorus (DUP) and particulate phosphorus (PP), which comprise total phosphorus (TP) in runoff from three
rainfall simulation events.
786
787
788
789
790
791
792
793
794
795
796
36
797
798
799
800
Fig 4 The average % of dissolved reactive phosphorus (DRP), dissolved un reactive phosphorus (DUP) and
particulate phosphorus (PP), which comprise total phosphorus (TP) in runoff after three rainfall simulation events.
801
802
803
804
805
806
807
808
809
810
811
812
813
814
37
815
816
817
Fig 5 Average flow weighted mean concentrations of suspended sediment in runoff.
818
819
820
821
822
823
824
825
826
827
828
829
38
830
831
832
Fig 6 Average flow weighted mean concentrations of Al in runoff and rain water.
833
834
835
836
837
838
839
840
841
842
843
844
845
39
846
847
848
Fig 7 Average flow weighted mean concentrations of Ca in runoff and rain water.
849
850
851
852
853
854
855
856
857
858
859
860
861
40
862
863
864
Fig 8 Average flow weighted mean concentrations of Fe in runoff and rain water.
865
866
867
868
869
870
871
872
873
874
875
876
877
41
878
879
880
881
Table 1
Stoichiometric ratio at which the amendments were applied and slurry dry matter (DM), pH and average
concentrations of NH4- N, water extractable phosphorus (WEP), total nitrogen (TN), total phosphorus (TP) and total
potassium (TK) (n=3).
Rate
DM
pH
-1
%
Slurry
WEP
-1
TN
TP
-1
TK
-1
mg L
g kg DM
mg L
mg L
mg L-1
10.5 (0.04)
7.47 (0.05)
1760 (123)
2.22 (0.34)
4430 (271)
1140 (76)
4480 (218)
1.1:1 [Al:TP]
9.4 (0.16)
5.40 (0.12)
1770 (21)
0.002 (0.0004)
4570 (176)
1140 (69)
4360 (84)
PAC
0.93 [Al:TP]
9.6 (0.28)
6.37 (0.05)
1760 (143)
0.0013 (0.0003)
4750 (448)
1180 (165)
4680 (448)
Lime
10:1 [Ca:TP]
8.2 (0.29)
12.2 (0.12)
1320 (141)
0.0056 (0.0003)
3190 (263)
1140 (96)
4810 (227)
FeCl2
2:1 [Fe:TP]
() standard deviation
10.1 (0.22)
6.7 (0.06)
1700 (11)
0.0022 (0.0006)
4340 (372)
1120 (51)
4720 (386)
Alum
882
883
884
885
886
NH4-N
887
42
888
889
Table 2
From preliminary study and current study, showing cost of treatments and total phosphorus (TP) lost from runoff box.
a
Preliminary agitator test
stoichiometric DRP
ratio
reduction
Treatment
metal: TP
DRP
reduction
Cost per m3
treated slurryb
metal: TP
%
€ m-3
TP loss as %
of TP applied
Cost per kg
P reduction
P lost per
hectare
€ kg P-1
kg P ha-1
Slurry
-
-
-
-
1.90
7.70
-
2.90
Alum
0.98:1
87
1.11:1
83
7.40
0.46
66.70
0.17
0.98:1
88
0.93:1
86
8.80
1.05
91.10
0.40
5:1
74
10:1
69
10.20
1.16
111.00
0.44
c
PAC (AlCl3)
Lime
890
891
892
893
894
895
%
Runoff box
stoichiometric
ratio
FeCl2
2:1
88
2:1
67
7.00
2.20
61.00
0.19
a
Taken from Brennan et al. (2011).
b
The cost m-3 and cost effectiveness have been updated from Brennan et al. (2011) to reflect the slight change in ratio of metal:TP in the present runoff box study.
c
Laboratory grade aluminium chloride (Al2(SO4)3.nH2O) was used in Brennan et al. (2011). Commercially available commercial grade liquid poly-aluminium chloride was used in the present study.
Note: All treatments were found to be significantly different to the control (p<0.001) in the Brennan et al. (2011) study. However, these were not significantly different to each other. In this study, all
treatments were significantly different to the slurry-control. Alum and AlCl2 were significantly different to lime and FeCl2, but not to each other. (€1.00 is approximately equal to $1.37or £1.59)
896
43
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement