Nitrate leaching and pasture production from different nitrogen sources on a shallow stoney soil under flood-irrigated dairy pasture.
The leaching of nitrate (N[O.sub.3.sup.-]) from agricultural land and the subsequent contamination of water resources have been recongised as a major environmental issue that could undermine the long-term sustainability of agricultural production in many countries (Spalding and Exner 1993; Jarvis et al. 1995; Addiscott 1996; Cameron et al. 1997; Di and Cameron 2001). Because high concentrations of nitrate in drinking water are deemed to be detrimental to human health, world and national health organisations have established drinking water guidelines, limiting N[O.sub.3.sup.-]-N concentration to a maximum of 10-11.3 mg N[O.sub.3.sup.-]-N/L (World Health Organisation 1984). If N[O.sub.3.sup.-] leached from agricultural land is drained into surface water bodies, it may cause deterioration in quality (e.g. algal bloom). Nitrogen (N) has been shown to be the element that is most limiting and responsible for eutrophication in many temperate estuaries and coastal ecosystems (Howarth 1988). The critical N concentration above which eutrophication may occur depends on the specific aquatic systems; however, eutrophication may become visibly evident when the concentration of total N reaches about 400-600 [micro]g N/L (AEC 1987). To protect water quality, a number of countries or regional authorities have developed regulations to limit N inputs into farming systems, established nitrate vulnerable zones (e.g. EC 1991), and produced codes for good agricultural practices (e.g. MAFF 1991). In New Zealand, regional authorities have also established, or are in the process of establishing, guidelines for land application of wastes and, in some cases, fertilisers to minimise contamination of water resources by farming activities.
One of the major sources of nitrate leaching is the intensively grazed grassland system, e.g. pastures grazed by dairy cattle, where N leaching losses are increased by applications of N fertilisers, organic waste effluents, and N returns in animal urine and dung (Ball et al. 1979; Ryden et al. 1984; Field et al. 1985; Scholefield et al. 1993; Di et al. 1998a, 1998b, 1999; Ledgard et al. 1999; Silva et al. 1999). In intensive pasture systems, e.g. dairying, N fertilisers are increasingly used by New Zealand farmers to stimulate pasture growth, particularly when there is a shortage of feed supply. Dairy shed effluent (DE, the effluent from the milking shed, comprising urine, dung, and washing water), which is rich in N, is also applied back to the pastures to recycle the nutrients. The DE is usually spread to an area around the milking shed, which is about 10% of the area of a dairy farm. In a grazed pasture, 60-90% of the N ingested by the grazing animal is returned to the pasture in the urine and dung and >70% of the N returned is in the urine (Haynes and Williams 1993; Jarvis et al. 1995). The potential for N[O.sub.3.sup.-] leaching from N applied in N fertiliser and waste effluents and from N returned in animal excreta varies widely, depending on soil and environmental conditions, type of N sources applied, timing of application, and application rates (Cameron et al. 1999; Di et al. 1999; Di and Cameron 2001).
In grassland systems, another important soil management practice that may result in significant N[O.sub.3.sup.-] leaching is the ploughing in of pastures, particularly if this occurs before the leaching season (e.g. winter in New Zealand), as large amounts of N are released by mineralisation after ploughing (Cameron and Wild 1984; Addiscott et al. 1991; Francis et al. 1995; Di and Cameron 2001). On dairy farms in New Zealand, pastures are usually renovated every 10-20 years to improve pasture quality, and this is now usually carried out by direct drilling of pasture seeds into the soil after the old pasture has been killed off by spraying herbicides. There appear to have been few direct measurements of N[O.sub.3.sup.-] leaching following pasture renovation by direct drilling. However, it is possible that this may also result in the release of N from soil organic N and thus increase N[O.sub.3.sup.-] leaching.
Dairy farming in New Zealand has been expanding rapidly, with the dairy cattle population increasing from around 3 million in 1982 to 5 million at present. Much of the expansion has been occurring in the South Island, e.g. in the regions of Canterbury and Southland. It is projected that over the next 5 years, the number of dairy farms will increase by 24%, average dairy farm size will increase by 19%, and average herd size will increase by 27% in the South Island (Gaul 2000). On the flood plains of Canterbury, a number of new dairy farms have been or are being established on shallow stony soils which have previously been used for sheep farming. These soils cover more than 200 000 ha in the region. They only have a thin layer (usually no deeper than 30 cm) of fine material on the top, below which the soil contains large amounts of gravel. Because these soils are free draining and are irrigated, there is a general concern that significantly greater nitrate leaching might occur as these soils are converted from sheep farming to dairy farming. There have been no direct measurements of N[O.sub.3.sup.-] leaching under dairy farming on these stony, free-draining soils.
The impact of N[O.sub.3.sup.-] leached from dairy farms on N[O.sub.3.sup.-] concentration in the underlying aquifer depends partly on the amount of dilution that may occur by recharge water from other sources. When assessing the potential impact of the expanding dairy industry on groundwater quality in the region, the dilution effect needs to be taken into account.
The objective of this research project was to determine the amount and concentration of N[O.sub.3.sup.-] leached from a shallow stony soil following the application of urea, DE, and cow urine, and by pasture renovation. The effect of timing of urine returns (spring v. autumn) on N[O.sub.3.sup.-] leaching losses was also determined. The impact of dairy farming on N[O.sub.3.sup.-]-N concentrations in the groundwater was then modelled, taking into account water recharge from different sources.
Materials and methods
Soil and pasture
The soil used was a Lismore stony silt loam (Pallic orthic brown soil, Hewitt 1998; Udic Haplustept loamy skeletal, Soil Survey Staff 1998) from an existing dairy farm in mid-Canterbury, in the South Island of New Zealand (171 [degrees] 43'E, 43 [degrees] 45'S). The soil had a reasonably fertile surface layer in terms of organic carbon (C), total N and Olsen R but had an increasingly high gravel content with depth, reaching about 50% at 60 cm (Table 1).
The existing perennial ryegrass (Loliumperenne)/white clover (Trifolium repens) pasture was renovated in February 1998 by first spraying the paddock with glyphosate (N-(phosphonomethyl) glycine) herbicide, and after 10 days, a mixture of perennial ryegrass and white clover seeds was sown by direct drilling to 10-15 mm depths following standard procedures (Hampton et al. 1999).
Lysimeter collection and installation
Undisturbed soil monolith lysimeters, 50 cm diameter and 70 cm deep, were collected 3 weeks after the pasture had been sown and the seeds had germinated. Details of the lysimeter collection and installation procedures have been described elsewhere (Cameron et al. 1992). In brief, this involved placing a metal cylinder casing on the soil surface, digging around the casing, making sure to minimise disturbance to the soil structure inside, and gradually pushing the casing down by small increments. Where stones were encountered by the metal casing, an angle-grinder was used to cut through the stone before pushing down the casing. Once the casing had reached the desired depth (70 cm), the soil monolith was then cut at the base with a cutting plate, secured on the lysimeter casing, and lifted out of the collection site. The lysimeters were transported to a field trench lysimeter facility near Lincoln University, using a specially designed trailer with air-bag suspension to minimise disturbance. The gap between the soil core and the metal casing was sealed using petroleum jelly to stop edge-flow effects (Cameron et al. 1992). The lysimeters were then installed in the field lysimeter facility with the surface of the lysimeters at the same level as that of the surrounding soil surface, in order to maintain normal plant growing conditions. The lysimeters were collected between February and March 1998, and were installed to the lysimeters in April 1998. The experiment began in April 1998 and this paper reports data obtained in the first 2 years.
Lysimeter treatments and maintenance
The treatments on the lysimeters are summarised in Table 2. Each treatment had 4 replicates. The treatments were allocated to the lysimeters in a randomised design. The application rate of 200 kg N/ha for DE equalled the regulatory limit established by the Canterbury Regional Council. The urea 200 was to provide the same N application rate as the DE 200 for comparison purposes. This amount was at the higher end of fertiliser N application rates for most farmers, although some farmers might use higher amounts. The DE 400 was included to assess the effect of DE application rates on N[O.sub.3.sup.-] leaching and pasture production. The rates of DE and urea were split into 4 applications per annum (Table 3). Two treatments (5, 6) included urine applications in the autumn to simulate grazed conditions where urine is deposited onto the paddock, which also receives DE or urea. A seventh treatment was a urine alone treatment applied in the spring (November) to determine the fate of urine deposited in the spring. The application rate of 1000 kg N/ha of urine N was equivalent to the N loading rate under a cow urine patch (Haynes and Williams 1993; Silva et al. 1999). All these treatments were repeated in the second year. An autumn urine treatment and combinations of DE or urea with spring urine applications were not included because of resource constraints.
Dairy effluent as collected from a nearby dairy farm in the morning after milking. It essentially comprised a mixture of cow urine and dung, deposited during the milking session, and cleaning water. A large proportion of the N was in organic forms from the dung (Table 3). The main form of inorganic N was N[H.sub.4.sup.+]-N, with only traces of N[O.sub.3.sup.-] and N[O.sub.2.sup.-]-N. A small amount of the N was in urea form, from the urine. The DE also contained organic C, and significant amounts of phosphorus (P). After the DE was analysed for total N, the volume of DE to give the desired amount of total N was then poured over the lysimeters (Table 3).
Urea was applied in solid form and was broadcast over the surface of the lysimeters followed by irrigation with the same volume of water as that applied in the DE. Irrigation following urea application has been shown to significantly reduce the ammonia loss by volatilisation (Black et al. 1987).
Fresh urine was collected early in the morning during the milking session from Friesian cows, and was analysed and applied to the lysimeters on the same day. The same volume of water was applied to the other lysimeters that did not receive urine to maintain the same moisture input to all lysimeters.
A `maintenance' P fertiliser, potash-sulfur-super (containing 7% P, 10% K, and 17% S), was applied to each lysimeter at 45 kg P/ha.year (Table 2). The annual rate was evenly split into 2 applications, one in the autumn (April) and one in the spring (November) following typical local farming practice.
From November to April (late spring to mid autumn), flood irrigation, at 100 mm, was applied to all the lysimeters at about 3-weekly intervals. The amount of water applied was to represent the average amount of water applied on commercial farms, but this, in reality, might vary between farms. Irrigation water was applied using an electronically controlled metering system to deliver the required volume of water to the lysimeters. From May to October (late autumn to mid spring), simulated rainfall was applied at the end of each month (if necessary), to supplement the natural rainfall received in order to equal the 75th percentile of local rainfall records for the same period of the year. This was done to create a so-called `worst case scenario' in terms of rainfall inputs.
The herbage was cut periodically to simulate typical grazing practice. All the harvested herbage was removed and dry matter yield recorded. At selected times and for selected treatments, the harvested pasture was dissected by hand to separate grass, clover, and weed components. Herbage nitrogen content was analysed on a LECO CNS-2000 analyser. Following each herbage cut, a specially designed mechanical cow hoof was used to simulate cow treading on the lysimeters (Di et al. 2001). The mechanical hoof is made of stainless steel with identical shape and size as an adult Friesian cow hoof. The hoof is mounted onto a compressed air ram which is driven by an air compressor system to provide a treading pressure of 220 kPa to simulate the treading pressure exerted by a cow hoof during walking. The entire surface of the lysimeters was trodden once following each herbage cut. This was based on our observation of hoof print coverage following each grazing rotation (Di et al. 2001).
Leachate collection and analysis
Leachates from the lysimeters were collected as required (when drainage was above 200 mL) or weekly and were analysed for nitrate, nitrite, and ammonium concentrations by flow injection analysis (Tecator Inc., Sweden), or ion exchange chromatography (IEC) (Waters Inc., USA).
Annual N[O.sub.3.sup.-]-N leaching losses were calculated based on N[O.sub.3.sup.-]-N concentrations in the leachate collected from each lysimster and the volume of drainage. Average annual leaching losses were then calculated based on values from the 4 replicates. Statistical analysis of the data was performed using Minitab (Version 11, Minitab Inc., USA).
Water input and drainage
Total water input, including rainfall, irrigation and that applied in the DE was about 1400 mm in the first year and 1700 mm in the second year, with the higher water input in the second year due to higher rainfall (Fig. 1). Total rainfall varied from about 500 mm in the first year (including 53 mm of simulated rainfall) to about 800 mm in the second year (including 32 mm of simulated rainfall). A total of 800 mm irrigation was applied from late spring to mid autumn each year. The amount of water applied in the DE was equivalent to 98 mm in the first year and 78 mm in the second year. Drainage varied from about 610 mm in the first year to about 880 mm in the second year. Significant drainage occurred following the flood irrigation (Fig. 1). In order to reduce drainage volume following flood irrigation, a new irrigation regime is currently being investigated where the water is applied at smaller amounts more frequently.
[FIGURE 1 OMITTED]
Nitrate-N leaching losses measured on the lysimeters
Nitrogen in the leachate was predominantly in the form of N[O.sub.3.sup.-] (accounting for up to 99% of the total mineral N leached) with only traces of N[H.sub.4.sup.+] and N[O.sub.2.sup.-] being detected. Following the application of DE and urea, there was a breakthrough of peak N[O.sub.3.sup.-]-N concentrations in the first year, but the concentrations remained very low throughout the second year (Fig. 2). The N[O.sub.3.sup.-]-N peak concentration in the DE 400 treatment was 34.2 mg N/L, and was higher than in the DE 200 (21.3 mg N/L) and urea 200 (11.5 mg N/L) treatments in the first year (P < 0.01).
[FIGURE 2 OMITTED]
Peak N[O.sub.3.sup.-]-N concentrations in the DE plus urine or urea plus urine treatments were significantly higher than in the DE or urea only treatments (P < 0.001) (cf. Figs 2 and 3). The peak N[O.sub.3.sup.-]-N concentrations reached above 400 mg N/L in the urine-applied treatments compared with <35 mg N/L in the DE or urea treatments without urine (Figs 2 and 3). The peak N[O.sub.3.sup.-]-N concentration in the urea + urine treatment was also higher in the first year (P< 0.01), although the peak was narrower, than in the second year. The difference in peak N[O.sub.3.sup.-]-N concentration in the DE + urine treatment was not significant (P > 0.05) between the 2 years.
[FIGURE 3 OMITTED]
The urine applied in the spring (November) resulted in lower peak N[O.sub.3.sup.-]-N concentrations (57-91 mg N/L) than that applied in the autumn in both years (120-417 mg N/L) (P < 0.05) (cf. Figs 4 and 3). However, there was no significant difference in peak N[O.sub.3.sup.-]-N concentration between the 2 years in the spring urine treatment.
[FIGURE 4 OMITTED]
The total annual N[O.sub.3.sup.-]-N leaching losses in the first year were higher in the DE 400 treatment than in the control and urea 200 treatment, but there was no significant difference between the control, DE 200, DE 400, and urea 200 treatments in the second year (Table 4). The annual N leaching losses were significantly higher (P < 0.01) in the first year than in the second year in the control, DE 200 and DE 400 treatments, but there was no significant difference in the urea 200 treatment between the 2 years (P > 0.05).
The annual leaching losses were much higher where urine was applied either alone or together with DE or urea compared with those non-urine treatments (Table 4). However, there was no significant difference (P > 0.05) in the amount of N leached in these urine treatments between the 2 years, even though there was a trend of higher leaching loss in the second year in the DE 400 + urine 1000A treatment.
The percentage of urine N leached was lower for the spring application (29%) than for the autumn application (38-58%) (Table 4). The percentage of urine N leaching loss for the autumn urine treatments was calculated by difference in the amount of N[O.sub.3.sup.-]-N leached between the DE 400 + urine 1000A or the urea 200 + urine 1000A treatments and the DE 400 or urea 200 treatments, respectively. This was based on the assumption that the combination of urine with DE or urea N did not change the amount of N leached from the DE or urea. This assumption may or may not always be valid.
Weighted average leaching losses from a grazed paddock
The N[O.sub.3.sup.-]-N leaching losses measured on the lysimeters, as reported above, were designed to provide an estimate for those areas on a dairy pasture that may receive DE or urea and urine-N inputs from the grazing animal. However, on a grazed dairy pasture, only a fraction of the paddock will be covered by urine patches in any one year. The estimation of annual average N[O.sub.3.sup.-]-N leaching losses in a paddock requires an integration of the data derived from the lysimeters (Di and Cameron 2000):
(1) [N.sub.L] = [N.sub.L1] x [P.sub.1] + [N.sub.L2] x [P.sub.2]
where [N.sub.L] is the annual average N[O.sub.3.sup.-]-N leaching losses from a grazed paddock, [N.sub.L1] and [N.sub.L2] are the leaching losses from the urine and non-urine patch areas, respectively, as determined on the lysimeters, and [P.sub.1] and [P.sub.2] are the proportion of areas covered by urine and non-urine patch areas, respectively. The values of [P.sub.1] and [P.sub.2] will vary depending on the stocking rate. On a dairy farm with 3 cows/ha, the area covered by the urine patches is around 25% of the grazed paddock area (Haynes and Williams 1993; Silva et al. 1999).
Table 5 shows the estimated N[O.sub.3.sup.-]-N leaching losses for 4 different scenarios. The grazing alone scenario means that no fertiliser or effluent N is applied, and urine patches cover 25% of the grazed paddock area. The other 3 represent situations where N fertiliser or effluent N is applied to the whole grazed paddock, and urine patches cover 25% of the paddock area. The data show that urine deposition alone could cause the N leaching loss to reach 129.6 kg N/ha in the first year and 112.2 kg N/ha in the second year, assuming a stocking rate of 3 cows/ha. This would give an annual average N[O.sub.3.sup.-]-N concentration of 12.8-21.2 mg N/L in the drainage water. The application of DE or urea at 200 or 400 kg N/ha to the grazed paddock only caused a slight increase in the annual N leaching loss, particularly in the second year (Table 5).
Modelled impact of dairying on groundwater N[O.sub.3.sup.-]-N concentration
It needs to be recognised that the impact on groundwater quality by N[O.sub.3.sup.-]-N leached below the rooting zone from dairy systems will partly depend on the amount of dilution that will occur in the aquifer (i.e. by recharge water from other sources). If we assume that all the N[O.sub.3.sup.-]-N leached will reach the groundwater (i.e. no further transformations occur), then the effect on groundwater N[O.sub.3.sup.-]-N concentration by leaching from the dairy pasture systems can be modelled using information about the specific aquifer system (i.e. recharge inputs to the aquifer from different sources). A study of the aquifer in the region where the dairy farm was sited on the Lismore soil showed that overall the aquifer in the region between the Rakaia and the Ashburton Rivers behaved as a semi-confined or unconfined system and was recharged mainly from 3 sources: natural precipitation, irrigation, and river plus stock races (Scott and Thorpe 1986). Based on the recharge data and N[O.sub.3.sup.-]-N leaching losses in dairy systems as measured from this study and the recharge data to the aquifer reported in the literature (Table 6), annual average N[O.sub.3.sup.-]-N concentrations in the recharge water entering the aquifer in the region can be modelled as follows:
(2) [N.sub.G] = ([N.sub.D] * [V.sub.D] + [N.sub.O] * [V.sub.O] + [N.sub.R] * [V.sub.R])/([V.sub.D] + [V.sub.O] + [V.sub.R])
where [N.sub.G] is the annual average N[O.sub.3.sup.-]-N concentration in the mixed recharge water; [N.sub.D], [N.sub.O], and [N.sub.R], are the average N[O.sub.3.sup.-]-N concentrations in the recharge water from dairying systems, other non-dairying systems, and river plus stock races, respectively; and [V.sub.D], [V.sub.O], and [V.sub.R] are volumes of annual recharge from the different sources, respectively. The volume of water present in the aquifer was ignored in this calculation because the aquifer in the region is highly dynamic with continuous water inputs (as described above) and outputs (e.g. submarine leakage across the coastline, water extraction for industrial, municipal and agricultural uses, and spring leakage) (Scott and Thorpe 1986). Therefore, in the long-term, the concentration of N[O.sub.3.sup.-]-N in the aquifer is determined by that in the incoming recharge water.
Figure 5 shows that as greater proportions of the land are used for dairy farming, the average N[O.sub.3.sup.-]-N concentration in the recharge water increases. If 50% of the land is used for dairying, N[O.sub.3.sup.-]-N concentration in the aquifer is projected to reach 9.7 mg N/L. If 67% of the land is used for dairying, N[O.sub.3.sup.-]-N concentration in the aquifer will reach the New Zealand drinking water guideline of 11.3 mg N/L. The two dotted lines represent a 50% increase or decrease in the concentration of N[O.sub.3.sup.-]-N leached from non-dairying systems to reflect the effect of possible landuse changes and uncertainties in the amount of N leached from non-dairying land. This effect is significant when the area used for dairying is small, but gradually becomes less significant when greater areas of land are used for dairying (Fig. 5).
[FIGURE 5 OMITTED]
Pasture N offtake and dry matter yield
Pasture N offtake in the control, DE, and urea treatments (without urine) ranged from 240 to 313 kg N/ha in the first year, and 345 to 429 kg N/ha in the second year. There was a trend of higher N offtake in the DE 400 treatment than in the other non-urine treatments, but the difference was not statistically significant (P > 0.05) (Fig. 6). The amounts of N offtake were significantly higher where urine was applied compared with the non-urine treatments, ranging from 486-532 in the first year, and 421-587 in the second year (with the exception of the urea 200 + urine 1000 treatment in the second year which suffered from some pest damage) (Fig. 6). The N offtake in the spring urine treatment was just as high as or higher than those in the autumn urine plus DE or urea treatments. Herbage dry matter yield data generally followed similar trends to those of N offtakes (Fig. 7).
[FIGURES 6-7 OMITTED]
Nitrate-N leaching losses measured on the lysimeters
The high N[O.sub.3.sup.-]-N concentrations in the leachate and high total annual N[O.sub.3.sup.-]-N leaching losses in the urine treatments (Figs 3 and 4, Table 4) clearly demonstrate that on these shallow stony dairy pasture soils, the greatest threat to groundwater quality in terms of N[O.sub.3.sup.-] leaching comes from those `hot' urine spots. The high N loading rates (around 1000 kg N/ha) in these urine spots are well above the amount of N that can be taken up by the pasture, the surplus N is thus prone to leaching when there is high drainage. The N[O.sub.3]-N concentrations and total annual leaching losses from the DE and urea treatments are much lower in comparison (Fig. 2 and Table 4), and therefore these N sources, per se, do not pose a great direct threat to groundwater quality by N[O.sub.3.sup.-] leaching. Their impact on N[O.sub.3.sup.-] leaching mainly occurs when the N, taken up by the pasture and consumed by the grazing animal, is returned to patches of the pasture in the concentrated form of urine N.
The higher leaching losses in the control, DE, and urea treatments in the first year compared with the second year (Fig. 2 and Table 4) are probably attributable, mostly, to the effects of pasture renovation, involving the killing off of the existing pasture and re-sowing of new pasture by direct drilling. This is likely to have caused a release of mineral N through mineralisation of organic N in the decaying pasture plants, thus increasing the amount of mineral N available for leaching. The release and increased leaching of N following cultivation of pasture has been reported previously (e.g. Cameron and Wild 1984; Francis et al. 1995); however, the results reported here indicate that increased leaching losses may also occur following pasture renovation by direct drilling. If the difference in the amount of N leached between the first year and second year in the control is taken as the amount of N contributed by mineralisation following pasture renovation, this would amount to 31.2 kg N/ha (Table 4). This was at the lower end of N leaching losses (14-102 kg N/ha) recorded following the ploughing in of pastures in mixed cropping systems in New Zealand (e.g. Francis et al. 1995). The lower disturbance of the soil by direct drilling as used in this study compared with ploughing in mixed cropping may explain the difference in the amount of N released by minealisation and thus in leaching between the direct drilling and ploughing practices.
During the lysimeter collection and installation, every effort was made to ensure minimum disturbance to the soil column. Therefore, the impact by lysimeter sampling and installation on nitrate leaching is likely to be minimal, and is unlikely to be the main cause for the higher leaching losses in the first year. Our previous studies using the same lysimeter technique have not shown significant increases in N[O.sub.3.sup.-]-N leaching due to possible disturbance associated with lysimeter sampling (Di et al. 1998a, 1998b; Silva et al. 1999).
In the urine treatments, however, because of the large amount of urine-N applied, the amount of N released from pasture renovation probably became insignificant in terms of its impact on N[O.sub.3.sup.-] leaching. This effect was therefore probably overshadowed by the variations in N[O.sub.3.sup.-]-N leaching losses from the urine between the 2 years. Therefore, there was no significant difference in the amount of N[O.sub.3.sup.-]-N leached in the urine treatments between the 2 years. The higher but narrower N[O.sub.3.sup.-]-N concentration peaks in the first year (Fig. 3) might be related to the lower drainage volume (610 mm) compared with the second year (877 mm). When water input increased in the second year, this was more likely to have caused macropore flow, thus resulting in broader peaks.
Although the autumn urine treatments (Fig. 3) also received DE or urea, the very low N[O.sub.3.sup.-]-N concentrations in the DE and urea alone treatments in the second year (Fig. 2) would indicate that most of the N in the DE plus urine or urea plus urine treatments in the second year would have been contributed by the urine N. The lower N[O.sub.3.sup.-]-N concentrations and annual leaching losses in the spring urine treatment compared with the autumn-urine treatment (Figs 3 and 4, Table 4) were probably caused by a combination of greater pasture N uptake (see section on herbage N uptake), greater immobilisation, and greater losses by some other processes (e.g. volatilisation and denitrification) due to higher temperatures in the spring and summer period. The coincidence of higher temperatures and soil moisture (e.g. after irrigation in a warm summer day) has been shown to enhance denitrification losses of N (Ambus and Christensen 1995; Henault et al. 1998) and would thus also have reduced the amount of N available for leaching after the urine application in the spring.
The annual percentage leaching losses of urine N applied in the autumn (38-58%) were much higher than the 12% leaching loss recorded in the companion study on the deep stone-free Templeton soil (Silva et al. 1999). The leaching losses on this shallow Lismore soil are also higher than those reported on a deep silt loam soil by Field et al. (1985), who found a 20% leaching loss in the first leaching season following urine application, and a 48% leaching loss after 2 leaching seasons.
Weighted average leaching losses from a grazed paddock
The estimated high N[O.sub.3.sup.-]-N leaching losses from a grazed paddock in the grazing only scenario demonstrate, again, the importance of `hot' urine spots in causing N[O.sub.3.sup.-] leaching in grazed pastures (Table 5). The calculated paddock leaching losses are higher than those recorded in the companion lysimeter using the deep stone-free Templeton soil (Silva et al. 1999), which showed N leaching losses of up to 60 kg N/ha per annum. The drainage volumes of 610-877 mm per year recorded in this study were also higher than that in the earlier study (410 mm). Ledgard et al. (1999) also reported high N[O.sub.3.sup.-]-N leaching losses in a 3-year study on dairy farms in the Waikato in the North Island of New Zealand. Nitrate N leaching losses ranged from 20 to 74 kg N/ha per year when no N fertiliser was applied, 59 to 101 kg N/ha per year when urea N was applied at 200 kg N/ha per year, and 100 to 204 kg N/ha per year when urea was applied at 400 kg N/ha (3.24 cows/ha). Cow urine N was also recognised as the main source of N leached in their studies based on the observation of skewed distributions of N[O.sub.3.sup.-] concentrations between individual leachate collectors and low leaching losses in ungrazed lysimeters. Studies in the UK on beef cattle grassland have recorded N[O.sub.3.sup.-]-N leaching losses ranging from 39 to 162 kg N/ha per annum (Ryden et al. 1984; Schlefield et al. 1993). These results show that the amount of N that may be leached can vary significantly on different soils and under different environmental and management conditions.
The high N leaching losses as estimated in Table 5 for the present study were mainly attributable to the shallow, stony, and free-draining nature of the Lismore soil. The Lismore soil only has about 20-30 cm of fine soil at the top, below which the soil stone content increases. The soil therefore has a low capacity to retain water and nutrients. When N in the form of urine or fertilisers is applied to the soil, the surplus N that is not taken up by the pasture is thus prone to leaching. In addition, the leaching losses reported here were measured under flood irrigation. Some dairy farms in the region are spray-irrigated, and the amount of water applied is usually less (about 50 mm/application) than that applied in flood (100 mm/application). The amount of water input can affect drainage volume, denitrification potential, and N[O.sub.3.sup.-]-N concentrations in the drainage (DJ et al. 1998a, 1998b). Therefore, the results reported here may or may not be applicable to spray-irrigated conditions. The development of regulatory guidelines and management practices designed to reduce N[O.sub.3.sup.-] leaching should take into consideration these variations in N[O.sub.3.sup.-] leaching potential in different soils and in environmental and management conditions. Computer models have a role to play to incorporate soil and environmental conditions when estimating N dynamics and N[O.sub.3.sup.-] leaching losses (Scholefield et al. 1991; Di and Cameron 2000).
The calculations for data in Table 5 assumed discrete urine patches without interference with one another. In reality, this assumption may or may not always hold as some urine patches may completely or partially overlap, and this uncertainty is a subject that is currently under investigation.
Modelled impact of dairying on groundwater N[O.sub.3.sup.-]-N concentration
Figure 5 represents a simple way of assessing the potential impact that the expanding dairy industry may have on groundwater N[O.sub.3.sup.-]-N concentration in the region. At present, about 5% of the land in the region is used for dairying, this would result in a modelled N[O.sub.3.sup.-]-N concentration of 3.9 mg N/L after dilution. A survey of groundwater quality in wells in the region showed that N[O.sub.3.sup.-]-N concentrations in deep wells (>40 m) were all below 4-6 mg N/L and decreased with depth (Smith 1993). In shallow wells (<40 m) N[O.sub.3.sup.-]-N concentrations ranged between 0.1 and 15 mg N/L. It is clear that because of the dilution effect, the N[O.sub.3.sup.-]-N concentration in the mixed aquifer can be significantly lower than that measured directly below the rooting zone of a grazed dairy farm (Table 5). However, as dairy farming expands, the N[O.sub.3.sup.-]-N concentration in the aquifer is likely to increase (Fig. 5). What is considered to be an acceptable increase of N[O.sub.3.sup.-]-N in the groundwater below the drinking water guideline is a subject of intense public debate in the region, and the regional regulatory authority may establish an upper limit above which any further increase will become unacceptable.
It should be noted that several simplified assumptions were made in modelling the average N[O.sub.3.sup.-]-N concentration in the mixed aquifer. The modelled average N[O.sub.3.sup.-]-N concentrations in the mixed recharge water (Fig. 5) probably represent a worst case scenario, as leaching losses and drainage volume from a shallow, stony soil (Lismore soil) were used as representing those of dairy systems in the region. Although Lismore-like soils are widespread in the Canterbury region, covering more than 200 000 ha, other soil types, which have greater depths to gravels and are less free-draining, are also used for dairying. The projected average N[O.sub.3.sup.-]-N concentrations in the recharge (Fig. 5) therefore probably represent the upper boundary of N[O.sub.3.sup.-]-N concentrations if the aquifer is completely mixed.
However, the recharge water from different sources may or may not be completely mixed, depending on their geographical locations and hydrogeological conditions. A less than uniform mixing in the short-term will result in spatial and vertical fluctuations in N[O.sub.3.sup.-]-N concentration in the aquifer. Recharge from rivers and water races is likely to have the greatest dilution effect on adjacent areas. There are also likely to be seasonal fluctuations in aquifer N[O.sub.3.sup.-]-N concentrations due to changes in recharge from dairying and non-dairying sources in different seasons. However, the modelled results still provide a useful long-term perspective with regard to the potential impact that the expanding dairy industry may have on groundwater quality in terms of N[O.sub.3.sup.-]-N concentration in the region. Clearly, further research is required to improve our knowledge of the hydrogeological and hydrochemical conditions of the aquifer in the region in order to improve our ability to carry out such calculations.
Pasture N offtake and dry matter yield
The treatment effect on pasture N offtake and dry matter yield was not only affected by the amount of N applied, but also by changes in clover growth induced by the N applications. In the control the N offtake was mainly derived from the soil N and biological N fixation by white clover. Where N was applied, the clover growth was suppressed. From February to April 1999, the clover component of the pasture varied from 52 to 63% in the control, 30 to 40% in the DE 400 treatment, 12 to 17% in the urea 200 treatment, and 3 to 5% in the urea 200 + urine 1000 treatment (data not presented). Therefore, where no N was applied in the control, a greater amount of N was supplied by biological N-fixation compared with other N applied treatments. The dramatic suppressing effect on clover growth by the urine-N should be taken into account when developing N budgets for grazed pastures.
Although the Lismore soil used in this study is a shallow stony soil, the soil fertility in the surface layer had been built up for dairy farming in the preceding decade (Table 1). The higher amounts of N fixation, high initial soil fertility and the maintenance P, K, and S fertilisers applied were probably responsible for the good performance in terms of pasture N offtake and dry matter yield in the control compared with the DE and urea treatments (without urine).
Although only a fraction (12-51%) of N in the DE applied was in mineral N forms (mainly N[H.sub.4.sup.+]) at the time of application (Table 3), the DE contained other nutrients (e.g. P, S, and K) which would be available for plant uptake. In addition, the application of DE has previously been shown to stimulate microbial activities and increase the mineralisation rate of organic N in the soil and organic N applied in the DE (Di et al. 1999; Zaman et al. 1999). Therefore, DE can be a useful nutrient source for pasture production in soils where these nutrients are deficient. Previous studies (e.g. Di et al. 1998b) have shown that DE can be as effective as N fertilisers in stimulating pasture production when applied at the same annual N rate.
The competitive performance of the spring urine treatment in N offtake and dry matter yield compared with the autumn-urine treatments which received additional DE or urea inputs (Figs 6 and 7) partly reflects the better timing of N input in relation to pasture N uptake and growth. The urine N applied in the spring, right at the start of a warm season with fast pasture growth, had a greater chance of being taken up by the pasture and thus a better agronomic value than the urine N applied in the autumn. This was probably one of the reasons for the lower N[O.sub.3.sup.-]-N leaching losses from the spring urine than from the autumn urine (Table 4).
The generally lower herbage N offtake and dry matter yield in most of the treatments in the first year than in the second year were probably because the pasture was newly sown in the first year, and the lysimeters received lower water input in the first year due to lower rainfall (Fig. 1).
This study showed that large amounts of N[O.sub.3.sup.-]-N, ranging from 112 to 162 kg N/ha per year, could be lost by leaching from this free-draining shallow and stony soil, depending on the amount and forms of N applied and pasture conditions. The largest contribution to N[O.sub.3] leaching loss came from the cow urine returns. Nitrate leaching losses from the urea or dairy effluent applied per se were much lower. The amount of N leached from the urine also varied, depending on the time of application, with a lower leaching loss (29% of the urine N applied) for the urine applied in spring than the urine N applied in the autumn (38-58%). Pasture renovation in the form of direct-drilling may also have caused an increase in N[O.sub.3] leaching in the first year, probably because of increased mineralisation of soil organic N. At this site the N[O.sub.3.sup.-] leached from the dairy farm will be diluted within the underlying aquifer by recharge water from other sources. A simple model was developed to assess the effect of this dilution and to calculate the effect of increasing the proportion of land covered for dairy farming in this region. The modelled average N[O.sub.3.sup.-]-N concentrations in the mixed recharge water below the dairy farm were significantly lower than those directly measured under the rooting zone (e.g. 3.9 mg N/L v. 13-27 mg N/L).
Table 1. Properties of the Lismore soil used Depth Stone content Fine earth Whole soil (cm) (% v/v) bulk density bulk density (g/[cm.sup.3]) (g/[cm.sup.3]) 0-20 10.0 0.98 1.15 20-40 14.5 1.07 1.31 40-60 45.9 1.29 1.95 pH (A) Organic C Total N Olsen P CEC Base (mg C/g) (mg N/g) ([micro] ([cmol. sat. g/g) sub.c]/kg) (%) 0-7.5 5.9 36.5 3.5 53 17.5 66 (A) Measured at 1:2.5 air-dried soil: water ratio. Table 2. Description of the treatments DE, dairy effluent from the milking shed; A, autumn applied; S, spring applied. DE and urea rates were split into 4 annual applications, in April/May, August, November, and February, each year. Urine was applied in April (autumn) for treatments 5 and 6, and in November (spring) for treatment 7, in a single application Treatment Description N applied (kg N/ha) no. DE Urea Urine 1 Control 0 0 0 2 DE 200 200 0 0 3 DE 400 400 0 0 4 Urea 200 0 200 0 5 DE 400 + 400 0 1000 urine 1000A (autumn) 6 Urea 200 + 0 200 1000 urine 1000A (autumn) 7 Urine 1000S 0 0 1000 (spring) Table 3. Chemical properties of the dairy effluent used The effluent contained less than 1% of N[O.sub.3.sup.-] and N[O.sub.2.sup.-] N. The % solids refer to the dry weight of the solid over the wet weight of the effluent. The volumes applied are for the DE 200 treatment. The volumes for the DE 400 double these Application pH N[H.sub.4. Urea-N Total N date sup.+]-N (mg/L) Apr. 1998 7.6 76.3 n.d. 150 Aug. 1998 7.5 37.8 6.1 300 Nov. 1998 7.8 58.5 1.5 280 Feb. 1999 7.9 43.8 1.0 165 May 1999 7.9 84.0 3.9 340 Aug. 1999 7.3 15.0 9.8 200 Nov. 1999 7.3 65.2 3.7 290 Feb. 2000 7.3 84.4 n.d. 240 Application Total P Total C % solids Vol. appl. date (L) Apr. 1998 32.5 882 0.3 6.5 Aug. 1998 n.d. 2503 0.8 3.3 Nov. 1998 22.5 794 0.3 3.5 Feb. 1999 64.3 1588 0.5 5.9 May 1999 123.2 5149 2.2 2.9 Aug. 1999 46.3 2597 1.0 4.9 Nov. 1999 62.4 3226 1.3 3.4 Feb. 2000 33.4 1235 0.5 4.1 n.d., not determined. Table 4. Total annual N[O.sub.3.sup.-]-N leaching losses as measured on the lysimeters Within columns, values followed by the same lower case letters are not significantly different (P > 0.05) Treatment Description no. 1998-1999 N leached (kg N/ha) ([+ or -] s.e.) 1 Control 36.0 ([+ or -] 3.5)a 2 DE 200 55.0 ([+ or -] 3.4)ab 3 DE 400 78.3 ([+ or -] 9.5)b 4 Urea 200 30.7 ([+ or -] 3.3)a 5 DE 400 + 456.7 ([+ or -] 23.0)d urine 1000A 6 Urea 200 + 569.1 ([+ or -] 65.7)d urine 1000A 7 Urine 1000S 327.1 ([+ or -] 44.8)c Treatment Description Annual no. N[O.sub.3. sup.-]-N leaching losses N% leached (A) 1 Control n.a. 2 DE 200 9.5 3 DE 400 10.5 4 Urea 200 n.a. 5 DE 400 + 30.1 urine 1000A 37.8 (urine) (B) 6 Urea 200 + 44.4 urine 1000A 53.8(urine) (B) 7 Urine 1000S 29.1 Treatment Description Annual N[O.sub.3.sup.-]-N leaching losses no. 1999-2000 N% N leached leached (A) (kg N/ha) ([+ or -] s.e.) 1 Control 4.8 ([+ or -] 1.2)a n.a. 2 DE 200 7.6 ([+ or -] 0.6)a 1.4 3 DE 400 18.7 ([+ or -] 3.4)a 3.5 4 Urea 200 17.1 ([+ or -] 7.6)a 6.2 5 DE 400 + 596.0 ([+ or -] 45.7)c 42.2 urine 1000A 57.7 (urine) (B) 6 Urea 200 + 569.2 ([+ or -] 14.7)c 47.0 urine 1000A 55.2 (urine) (B) 7 Urine 1000S 299.2 ([+ or -] 28.9)b 29.4 n.a., not applicable. (A) N leaching losses above those in the control as a percentage of total N applied. (B) N leaching losses above those in the corresponding DE or urea treatments as a percentage of urine N applied. Table 5. Weighted annual average N[O.sub.3.sup.-]-N leaching losses and concentrations in the drainage water from grazed fields using Eqn 1, assuming 25% of the field is covered with urine patches annually (equivalent to 3 cows/ha) DE or urea was assumed to have been applied throughout the paddock, where appropriate. N[O.sub.3.sup.-]-N leaching loss from the urine was estimated on the basis of the average percentage of urine N leached when applied in the spring and autumn (Table 4) Treatments N leaching losses Average concentration (kg N/ha) (mg N/L) 1998-1999 1999-2000 1998-1999 1999-2000 Urine input only 129.6 112.2 21.2 12.8 (no DE or urea applied) DE 200 + urine 148.6 115.0 24.3 13.1 DE 400 + urine 161.9 127.6 26.5 14.6 Urea 200 + urine 134.3 122.9 22.0 14.0 Table 6. Data used for modelling average N[O.sub.3.sup.-]-N concentrations in the mixed recharge water between the Rakaia and Ashburton rivers on the Canterbury plains as affected by the expansion of dairying The total area between the two rivers is 1350 [km.sup.2] Source Recharge rate N[O.sub.3.sup.-]-N concentration (mg N/L) Average annual recharge 370 mm/year (A) 4.8 (B) from non-dairying farmland Recharge from rivers and 2.96 x [10.sup.8] 0.1 (C) stock races [m.sup.3]/year (A) Recharge from irrigated 744 mm/year (D) 17.7 (D) dairy land (A) Scott and Thorpe (1986). (A) Based on leaching losses and landuse areas reported in Burden (1984). Effects by landuse changes and by uncertainty in this value is shown in Fig. 5. (C) Ministry for the Environment (1997). (D) Measured on the Lismore soil lysimeters as reported in this paper. The N[O.sub.3.sup.-]-N concentration was the average of all the treatments in Table 5 in both years. It was calculated using the average annual N[O.sub.3.sup.-]-N leaching loss (131.5 kg N/ha) and the average annual drainage (743.5 mm) of the two years.
We thank the New Zealand Foundation for Research, Science and Technology for funding the research, and Trevor Hendry, Steve Moore, and Neil Smith for technical support.
Addiscott TM (1996) Fertilizers and nitrate leaching. In `Agricultural chemicals and the environment'. (Eds RE Hester, RM Harrison) Issues in Envronmental Science and Technology 5, 1-26.
Addiscott TM, Whitmore HP, Powlson DS (1991) `Farming, fertilizers and the nitrate problem.' (CAB International: Wallingford, UK)
AEC (1987) Nutrients in Australian waters. Australian Environmental Council Report 19, Department of Arts, Heritage and Environment, AGPS, Canberra, Australia.
Ambus P, Christensen S (1995) Spatial and seasonable nitrous oxide and methane fluxes in Danish forest, grassland, and agroecosystems. Journal of Environmental Quality 24, 993-1001.
Ball PR, Keeney DR, Theobald PW, Nes P (1979) Nitrogen balance in urine-affected areas of a New Zealand pasture. Agronomy Journal 71, 309-314.
Black AS, Sherlock RR, Smith NP (1987) Effect of timing of simulated rainfall on ammonia volatilisation from urea applied to soil of varying moisture content. Journal of Soil Science 38, 679-687.
Burden RJ (1984) Chemical zonation in groundwater of the central plains, Canterbury. Journal of Hydrology (NZ) 23, 100-119.
Cameron KC, Di HJ, McLaren RG (1997) Is soil an appropriate dumping ground for our wastes? Australian Journal of Soil Research 35, 995-1035.
Cameron KC, Di HJ, McLaren RG (1999) Comparison of nitrogen leaching losses from different forms and rates of organic wastes, fertilsers and animal urine applied to Templeton soil lysimeters. In `Best soil management practices for production'. (Eds LD Currie, MJ Hedley, DJ Horne, P Loganathan) pp. 241-249. (Fertiliser and Lime Research Centre: Massey University, Palmerston North, NZ)
Cameron KC, Wild A (1984) Potential aquifer pollution from nitrate leaching following the plowing of temporary grasslands. Journal of Environmental Quality 13, 274-278.
Cameron KC, Smith NP, McLay CDA, Fraser PM, McPherson RJ, Harrison DF, Harbottle P (1992) Lysimeters without edge-flow: an improved design and sampling procedure. Soil Science Society of America Journal 56, 1625-1628.
Di HJ, Cameron KC (2000) Calculating nitrogen leaching losses and critical nitrogen application rates in dairy pasture systems using a semi-empirical model. New Zealand Journal of Agricultural Research 43, 139-147.
Di HJ, Cameron KC (2001) Nitrate leaching in temperate agroecosystems: sources, factors and mitigating strategies. Nutrient Cycling in Agroecosystems (in press).
Di HJ, Cameron KC, Moore S, Smith NP (1998a) Nitrate leaching from dairy shed effluent and ammonium fertiliser applied to a free-draining soil under spray or flood irrigation. New Zealand Journal Agricultural Research 41, 263-270.
Di HJ, Cameron KC, Moore S, Smith NP (1998b) Nitrate leaching and pasture yields following the application of dairy shed effluent or ammonium fertilizer under spray or flood irrigation: results of a lysimeter study. Soil Use and Management 14, 209-214.
Di HJ, Cameron KC, Moore S, Smith NP (1999) Contributions to nitrogen leaching and pasture uptake by autumn-applied dairy effluent and ammonium fertilizer labeled with [sup.15]N isotope. Plant and Soil 210, 189-198.
Di HJ, Cameron KC, Milne J, Drewry JJ, Smith NP, Hendry T, Moore S, Reijnen B (2001) A mechanical hoof for simulating animal treading under controlled conditions. New Zealand Journal of Agricultural Research 44, 111-116.
EC (1991) Council directive of 12 December 1991 concerning the protection of waters against pollution caused by nitrates from agricultural sources. 91/676/ECC, Legislation 1375/1-1375/8, European Community.
Field TRO, Theobald PW, Ball PR, Clothier BE (1985) Leaching losses of nitrate from cattle urine applied to a lysimeter. Proceedings of Agronomy Society of New Zealand 15, 137-141.
Francis GS, Haynes RJ, Williams PH (1995) Effects of the timing of ploughing-in temporary leguminous pastures and two winter cover crops on nitrogen mineralization, nitrate leaching and spring wheat growth. Journal of Agricultural Science 124, 1-9.
Gaul P (2000) South Island dairying-2005. In `Proceedings of the South Island Dairy Event'. Invercargill. pp. 177-189. (The Professional Development Group, Lincoln University: Canterbury, NZ)
Hampton JG, Kemp PD, White JGH (1999) Pasture establishment. In `New Zealand pasture and crop science'. (Eds J White, J Hodgson) pp. 101-115. (Oxford University Press: Auckland, NZ)
Haynes RJ, Williams PH (1993) Nutrient cycling and soil fertility in the grazed pasture ecosystem. Advances in Agronomy 46, 119-199.
Henault C, Devis X, Page S, Justes E, Reau R (1998) Nitrous oxide emission under different soil and land management conditions. Biology and Fertility of Soils 26, 199-207.
Hewitt AE (1998) `New Zealand soil classification.' 2nd edn (Manaaki Whenua Press: Lincoln, Canterbury, NZ)
Howarth RW (1988) Nutrient limitation of net primary production in marine ecosystems. Annual Review of Ecological Systems 19, 898-1110.
Jarvis SC, Scholefield D, Pain B (1995) Nitrogen cycling in grazing systems. In `Nitrogen fertilization in the environment'. (Ed. PE Bacon) pp. 381-419. (Marcel Dekker: New York)
Ledgard SF, Penno JW, Sprosen MS (1999) Nitrogen inputs and losses from clover/grass pastures grazed by dairy cows, as affected by nitrogen fertilizer application. Journal of Agricultural Science, Cambridge 132, 215-225.
Ministry for the Environment (1997) The state of the environment report. Ministry for the Environment, Wellington, NZ.
MAFF (1991) `Code for good agricultural practice for the protection of water.' Ministry of Agriculture, Fisheries and Food. (MAFF Publications: London)
Ryden JC, Ball PR, Garwood EA (1984) Nitrate leaching from grassland. Nature 311, 50-53.
Scholefield D, Lockyer DR, Whitehead DC, Tyson KC (1991) A model to predict transformation and losses of nitrogen in UK pastures grazed by beef cattle. Plant and Soil 132, 165-77
Scholefield D, Tyson KC, Garwood EA, Armstrong AC, Hawkins J, Stone AC (1993) Nitrate leaching from grazed grassland lysimeters: effects of fertilizer input, field drainage, age of sward, and patterns of weather. Journal of Soil Science 44, 601-613.
Scott DM, Thorpe HR (1986) Ground water resources between the Rakaia and Ashburton rivers. Publication No. 6 of the Hydrology Centre, Ministry of Works and Development, Christchurch.
Silva RG, Cameron KC, Di HJ, Hendry T (1999) A lysimeter study of the impact of cow urine, dairy shed effluent and nitrogen fertilizer on drainage water quality. Australian Journal of Soil Research 37, 357-69.
Smith VR (1993) Groundwater quality in Canterbury, results of the summer 1992/1993 survey. Canterbury Regional Council, Christchurch, NZ.
Soil Survey Staff (1998) `Keys to soil taxonomy.' 8th edn (United States Department of Agriculture: Washington, DC)
Spalding RF, Exner ME (1993) Occurrence of nitrate in groundwater--a review. Journal of Environmental Qualility 22, 392-402.
World Health Organization (1984) `Guidelines for drinking water quality. Vol 1. Recommendations.' (WHO: Geneva)
Zaman M, Di HJ, Cameron KC (1999) A field study of gross rates of N mineralisation and nitrification and their relationships to microbial biomass and enzyme activities in soils treated with dairy effluent and ammonium fertilisers. Soil Use and Management 15, 188-194.
Manuscript received 1 February 2001, accepted 25 June 2001
H. J. Di and K. C. Cameron
Centre for Soil and Environmental Quality, PO Box 84, Lincoln University, Canterbury, New Zealand.
|Printer friendly Cite/link Email Feedback|
|Author:||Di, H.J.; Cameron, K.C.|
|Publication:||Australian Journal of Soil Research|
|Article Type:||Statistical Data Included|
|Date:||Mar 1, 2002|
|Previous Article:||Mineralisation of nitrogen contained in mature subterranean clover, capeweed and annual ryegrass, and subsequent nitrogen use by wheat in dryland...|
|Next Article:||Phosphorus sorption and nitrogen transformation in two soils treated with piggery wastewater.|