| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a School of Plant Biology (M084), The University of Western Australia, Nedlands 6009, Western Australia, Australia
b Landcare Research, Private Bag 3127, Hamilton, New Zealand
c Lincoln Environmental, Private Bag 3062, Hamilton, New Zealand
d ESR, PO Box 50.348, Porirua, New Zealand
* Corresponding author (lbarton{at}agric.uwa.edu.au)
Received for publication March 1, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abbreviations: TOC, total organic carbon
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The effectiveness of soil–plant systems to assimilate wastewater-applied N and P will depend on the biological, chemical, and physical attributes of the soil, as well as plant uptake and irrigation management. Wastewater-applied N can be removed biologically via plant uptake (if the cover crop is removed), denitrification, volatilization, or immobilization into the soil organic matter. Any excess N remaining is likely to be leached. Furthermore, additional inorganic N may become available if irrigation increases net N mineralization rates of soil organic matter (Polglase et al., 1995). Wastewater-applied P can be chemically adsorbed by the soil, taken up by plants, or leached from the soil profile. The extent of all these biological and chemical processes will vary with soil type. For example, denitrification is often greater in loamy-textured than sandy-textured soils (Barton et al., 1999a), while P retention increases with increasing iron oxides, aluminium oxides, and aluminosilicate minerals (Brennan et al., 1994).
The way in which nutrients move through the soil profile also affects the extent to which soil biological and chemical processes can assimilate the wastewater (McLeod et al., 1998). Many soil and plant biological processes (e.g., denitrification, plant uptake) occur at greater rates in the topsoil than the subsoil. Hence, increasing the contact time and surface interaction between applied nutrients and the topsoil should increase nutrient assimilation. Contact with the topsoil will be increased if the wastewater moves through micropores within aggregates (i.e., matrix flow) rather than rapidly around aggregates and down cracks and worm channels (i.e., preferential flow). The way wastewater moves through the soil profile will vary according to soil structure and the rate at which the wastewater is applied (McLeod et al., 1998).
In New Zealand, regional authorities advocate the application of wastewater onto land rather than direct discharge into surface waters. Land application of wastewater conforms with the environmental ethics of the indigenous Maori people and is considered to be an alternative to tertiary treatment. Soils vary in their chemical, biological, and physical properties, and can therefore be expected to vary in their ability to assimilate applied wastewater nutrients. The objectives of our study were to determine if four contrasting soils, from which the pasture is regularly cut and removed, varied in their ability to assimilate nutrients from secondary-treated domestic effluent under high hydraulic loadings, in comparison with unirrigated, fertilized pasture.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
The soil cores were resited to a lysimeter facility adjacent to two domestic-effluent treatment ponds, 15 km from Hamilton, New Zealand (37°49' S, 175°17' E). The existing plant cover of the cores was removed by spraying with herbicide (glyphosate) and a new ryegrass (var. Samson) pasture established. The lysimeters were arranged lengthwise in a trench so their tops were at ground level, and exposed to rainfall. Two parallel walls formed a separate trench between the two rows of lysimeters, allowing leachate collection vessels to be buried in the floor of the trench so the leachate could freely drain from the base of the lysimeter to the collection vessel. Effluent irrigation did not begin for at least 6 mo after lysimeter collection. The Gley and Pumice Soils were irrigated from March 1999 to March 2001, while the Allophanic and Recent Soils were irrigated from January 2000 to January 2002 (Table 2).
|
Effluent samples were collected weekly before irrigation, while leachate samples were collected at least fortnightly. Leachate volumes were recorded, and subsamples collected and frozen before analysis. Leachate and effluent samples were bulked proportionally to the weekly volume to make up a 100-mL sample every 4 wk. Bulked samples were then analyzed for total N, nitrate, ammonium, total P, orthophosphate (sometimes referred to as "reactive P"), total organic carbon (TOC), and pH using the methods described below. The ryegrass pasture was harvested and removed when the average pasture height for each soil type (and each treatment) reached 200 to 250 mm. As effluent-irrigated pastures are rarely grazed by animals, the pasture was cut and removed to replicate a "cut and carry" treatment system.
Leachate, Effluent, and Plant Analyses
Total N and P of effluent and leachate samples were determined after digesting samples using a modification of the method of Ebina et al. (1983). Briefly, 6 mL of sample and 6 mL of digest mix (0.074 mol L–1 K2S2O8, 0.075 mol L–1 NaOH) were autoclaved at 121°C for 1 h. Digestion converts N species to nitrate and P species to orthophosphate. Nitrate was measured using a modified hydrazine reduction colorimetric method (Kamphake et al., 1967). Orthophosphate was measured using a modified molybdate and malachite green colorimetric method (Motomizu et al., 1983). Ammonium in undigested samples was measured by colorimetric methods on a flow injection analyzer (QuikChem 8000; Lachat, Milwaukee, WI) (American Public Health Association, 1995). Nitrate and orthophosphate in undigested samples were determined using the methods described above. Organic carbon was measured using a total carbon analyzer (TOC-5000; Shimadzu, Kyoto, Japan). Effluent and leachate pH was measured using a glass electrode pH meter. To determine N and P uptake by the pasture, the herbage was oven-dried (60°C) until a constant dry weight was achieved, weighed, and then analyzed for total N and P (using the flow injection analyzer described above) following a Kjeldahl digest incorporating sulfosalicylic acid (Blakemore et al., 1987).
Data Analyses
Monthly effluent applications of N, P, and TOC were calculated by multiplying irrigation volume (total after 4 wk) by the four-weekly proportionally bulked concentration of the nutrient in the effluent subsample. Similarly, the amount of N and P leached (and forms of N and P) were calculated on a four-weekly basis and then summed to give a total loss after two years. During the study, 76 to 112 leachate samples were analyzed per soil type depending on irrigation treatment. Organic N leached was calculated by subtracting the amount of NH+4 and NO–3 leached from the total N leached. "Unreactive" P was calculated by subtracting orthophosphate leached from total P leached. The total nutrient content of plant samples was aggregated to calculate the total N and P uptake by the pasture during the trial. The balance of N and P was determined by the difference between the nutrients applied and the nutrient leached plus nutrient taken up by the pasture. For the N balance, this remaining component included immobilized N, soil solution N, gaseous losses from denitrification, nitrification, volatilization, plus experimental error. For P it was assumed to be the estimated soil P storage plus experimental error. Changes in soil N and P storage were not determined by measuring changes in N and P soil content, as differences would be difficult to assess after two years, and calculations would require very accurate measures of soil bulk density. Treatment differences in plant uptake, nutrient leaching, and the remaining N and P balance were analyzed using the general analysis of variance procedure in GenStat Sixth Edition (Version 6.1.0.200; VSN International, 2002). Post-hoc pair-wise comparisons of means were made using Tukey P values (Keppel, 1991).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
The N leached from effluent treatments was mainly in organic forms, and represented 69 to 88% of N leached (Tables 5 and 7). Leaching of organic N occurred throughout the study for all soil types with no distinct seasonal distribution pattern (data not shown). The greatest proportion of organic N to total N leached was for the Recent Soil (P < 0.05), with monthly losses varying from 5 to 11 kg N ha–1. Much of the differences in N leaching between effluent-irrigated soil types can be attributed to differences in the amount of organic N leached. For the effluent-irrigated soils, organic N leached in the order: Recent = Gley > Pumice = Allophanic (i.e., a similar ranking to total N leached) (P < 0.05; Table 7). The amount of mineral N (i.e., NO3– and NH4+) leached also varied between effluent-irrigated soils, decreasing in the order: Gley > Recent = Pumice > Allophanic (P < 0.05). For the unirrigated soils, N also leached mainly in an organic N form for the Allophanic, Gley, and Recent (Tables 5 and 7), with the unirrigated Recent Soil leaching a greater amount of organic N than the other soil types (P < 0.05; Table 7). The amount of mineral N leached from the unirrigated treatments did not vary between soil types and was similar to the effluent-irrigated Allophanic and Pumice Soils (P < 0.05; Table 7). The leachate from the effluent-irrigated Gley soil was cloudy and greenish brown in color, while the leachate from both Recent Soil treatments was orange-brown and clear. The leachate from the remaining soil treatments was clear.
|
Nitrogen and Phosphorus Balance
For the Allophanic and Recent Soils, the amount of recovered N (leached plus N taken up by the pasture) exceeded the amount of effluent N applied by up to 180 kg N ha–1 after two years (Table 5). Native soil N present in the soil before irrigation is the likely source of this unaccounted N. For the Gley and Pumice Soils, we did not recover all the N applied (Tables 5 and 6), and it is likely the unrecovered N has either been denitrified or stored in the soil. Similarly, P leaching and plant uptake do not account for all the applied effluent P and it appears the soils have stored P by varying amounts (Table 6). Greatest estimated P storage after two years occurred for the Pumice Soil (228 kg P ha–1), followed by Gley (171 kg P ha–1), Allophanic (146 kg P ha–1), and Recent Soils (90 kg P ha–1) (P < 0.05; Table 6). Apart from the Recent Soil, applying effluent increased estimated P storage for all irrigated soil types (Table 6).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In our study, N leached mainly in the organic form from all effluent-irrigated soil types. Other studies reporting N leaching from wastewater-irrigated soils often measure only inorganic forms of N in the leachate (Hook and Burton, 1979; Sopper and Kerr, 1979; Kim and Burger, 1997). Had we measured only inorganic N leaching in our study, N leaching losses would have been underestimated by up to 90%. Furthermore, we would have concluded that the Recent Soil leached similar amounts of N to the Pumice Soil. Similarly, P leaching losses would have been 40 to 70% lower if we had measured only orthophosphate and not total P. We strongly recommend measuring total N and total P, in addition to inorganic forms, when measuring the leaching of these elements from effluent-irrigated soils. This is of particular importance to regulatory bodies interested in limiting nutrient losses from land treatment systems to ground water. For example, in New Zealand, application of effluent to land requires a consent that meets the Resource Management Act (New Zealand Government, 1991), which is administered by Regional Authorities. In practice, consents generally require monitoring of inorganic N in leachate, rather than leaching of total N (including organic N).
Effluent irrigation only increased P leaching from the Gley Soil, with losses less than 25 kg P ha–1 yr–1. Phosphorus leaching is generally not considered to be a concern for land treatment systems as, with the exception of some sandy soils (Iskandar and Syers, 1980; Latterell et al., 1982), most soils are thought to have sufficient capacity for retaining the applied P (e.g., Cameron et al., 1997; McLaren and Smith, 1996). For example, P leaching has been shown to be low (<2 kg P ha–1) from effluent-irrigated forest (e.g., Burton and Hook, 1979; Tomer et al., 2000) and pasture (Kardos and Hook, 1976) soils. Despite this, significant P leaching was observed in this study from the Gley Soil (clay loam), which has a moderate P retention index (Table 1). Furthermore, the P retention index measured before effluent irrigation ranked soil P storage as: Allophanic > Pumice 
Gley >> Recent (Table 1). Our estimated P storage was greater in the Pumice Soil than the Allophanic and Gley Soils (Table 6).
For the Allophanic (irrigated and unirrigated) and Recent Soils (irrigated only), the amount of N recovered in leachate and by plant uptake was greater than the amount of applied N. For the Allophanic Soil treatments, we suggest that the additional N is a result of net soil N mineralization, and has subsequently been taken up the plant. For the irrigated Recent Soil, the additional N is also soil derived, but has been leached as organic N. The leaching of organic N from the irrigated Recent Soil is discussed in more detail below.
Factors Influencing Nitrogen and Phosphorus Leaching
Apart from the Recent Soil, N leaching from effluent-irrigated soils appears to be linked either to plant N uptake or to the soil's tendency for preferential flow. For the Allophanic and Pumice Soils, high plant N uptake decreased the availability of N for leaching. While it is not entirely clear why the Allophanic and Pumice Soils had greater plant N uptake than the Gley Soil, it is suspected to be due to restricted soil aeration affecting plant growth. The lower N uptake may also have been partly due to the way the effluent moved through the soil profile. In a companion study, McLeod et al. (2001) showed that the Allophanic and Pumice Soils had a lower tendency for preferential flow than the Gley Soil due to their uniformly porous soil structure leading to predominately matrix flow. Decreasing the incidence of preferential flow and keeping the wastewater in the active root zone increases the opportunity for plant uptake. In the Gley Soil, low plant N uptake and a tendency for preferential flow contributed to high N leaching. Preferential flow in the Gley Soil was attributed to large structural cracks, as well as worm holes and root channels present in the soil (McLeod et al., 2001).
Nitrogen leaching from the effluent-irrigated Recent Soil was high, and a similar quantity to the effluent-irrigated Gley Soil, despite having similar plant N uptake to the Allophanic Soil. Furthermore, the amount of recovered N (leached plus N taken up by the pasture) for the effluent-irrigated Recent Soil exceeded the amount of effluent N applied. As previously discussed, this additional N appears to be soil derived and has leached as organic N. For the irrigated Recent Soil we consider that the increased hydraulic load appears to be responsible for the amount of N leached rather than the N contained in the effluent. This is supported by two observations. First, the TOC to organic N ratio of the leachate from the irrigated Recent Soil (14) and unirrigated Recent Soil (19) were similar and much greater than the TOC to organic N ratio of the effluent (3.8), indicating the organic N was not directly derived from the effluent. Second, the organic N concentration of the leachate from the irrigated Recent Soil (2.9) is very similar to that from the unirrigated Recent Soil (2.6; Table 7). If the N leached from the irrigated Recent Soil was derived directly from the applied effluent (as a result of preferential flow) we would have expected the leachate from the effluent-irrigated soil to have a higher organic N concentration than that from the unirrigated soil, and a TOC to organic N ratio similar to that of the effluent. Indeed, this is the case for the irrigated Gley Soil, where preferential flow was responsible for increased N and P leaching.
The amount of P leached from the effluent-irrigated soils was related to plant P uptake and soil P storage. For example, moderate P leaching losses were recorded from the Recent Soil, due to high pasture P uptake, while low P leaching losses were recorded from the Allophanic and Pumice Soils, due to a combination of high pasture P uptake and soil P retention. Phosphorus leaching was greater from the Gley Soil than the other soil types due to low pasture P uptake, which was consequence of poor growth (Table 4) related to anaerobic soil conditions. In our study, the estimated P that has gone into storage ranged from 45 to 114 kg P ha–1 yr–1. These values were similar to that reported for an effluent-irrigated forest soils (78–109 kg P ha–1 yr–1; Burton and Hook, 1979; Falkiner and Polglase, 1997; McLay et al., 2000); but not as high as that reported for a pasture grown on a sandy loam soil (44–332 kg P ha–1 yr–1 depending on annual application rate; Kardos and Hook, 1976).
Implications for Design
Designing land treatment systems so that preferential flow is limited and the cover crop can fully utilize the applied N (and any additional N mineralized from the soil) is a practical approach to limiting N leaching. Limiting preferential flow can be achieved by choosing soil types that do not have a tendency for preferential flow, and using irrigation rates that minimize preferential flow (McLeod et al., 1998). Various management strategies will optimize plant uptake of wastewater-applied nutrients in a land treatment system. Ideally, nutrient loading rates should match plant requirements, and be adjusted to suit seasonal differences and differences due to the age of the crop. For example, pasture species grown outside the tropics often take up more nutrients during the warmer months, while tree plantations have higher nutrient demands before canopy closure (Stewart et al., 1990; Cromer et al., 1983). In some instances, uptake can be further enhanced if the biomass is regularly harvested and removed from the site (Hook and Burton, 1979). Nitrogen uptake should also be optimized by ensuring other elements essential for plant growth, and not supplied by the wastewater irrigation, are applied. In grazed systems, matching N inputs to plant requirements will be more difficult due to the uneven distribution of urine N. Although denitrification, volatilization, and soil storage are alternative sinks for wastewater-applied N, these processes are often variable, more difficult to predict or manage than plant uptake, and can occur at low rates in soils suited for land treatment systems (Smith et al., 1996; Barton et al., 1999b; Smith and Bond, 1999; Mending et al., 2001; Master et al., 2003).
There were large differences in plant uptake of N and P in the irrigated soils, which have implications for designing land treatment systems reliant on biomass harvesting. The reason for these differences is not entirely clear, but for the Gley Soil low plant uptake was probably due to poor soil aeration. Interestingly, plant uptake in the irrigated soils was strongly correlated to applied inorganic N. If this relationship was causal it implies plants are predominantly accessing applied inorganic N in the Gley and Pumice soils. However, for the Allophanic and Recent Soils, mineralization of organic N (from either soil organic matter or effluent) must have contributed to plant biomass production. Plant biomass production from the effluent-irrigated Gley Soil was lower than other soils, probably due to fact that the soil was often saturated as a result of the soil's low hydraulic conductivity. Whatever the reason for the differences in plant uptake of N, the large variation of plant removal of nutrients between soil types needs to be accounted for in land treatment design. We advocate biomass trials to determine the amount of nutrients plant can remove.
Phosphorus leaching losses can be minimized by limiting preferential flow, ensuring wastewater-applied P matches plant uptake and the capacity of soil to retain P. Phosphorus sorption isotherms are usually used to predicting potential soil P storage in a land treatment system (e.g., Ryden and Pratt, 1980; Feigin et al., 1991). However, this approach can lead to an incorrect estimation of the soil's ability to retain P (Falkiner and Polglase, 1997; Menzies et al., 1999). Phosphorus sorption isotherms have been shown to underestimate the capacity of a sandy loam in a Monterey pine (Pinus radiata D. Don) plantation to retain effluent applied P. The authors attributed the discrepancy to the short (17 h) equilibrium time and wide solution to soil ratio used to produce the sorption isotherm (Falkiner and Polglase, 1997). By contrast, Menzies et al. (1999) found that sorption isotherms overestimated the ability of an effluent-irrigated soil to retain P, as the interaction between the wastewater P and the soil was limited by the low hydraulic conductivity of the soil. Similarly, in our study, P leaching from the Gley Soil may have been high due to preferential flow limiting the contact between wastewater-applied P and soil P adsorption sites. Consequently, the effectiveness of a soil as a phosphate sink should not be based on phosphate adsorption capacity alone and needs to take into account the hydraulic conductivity of soil and the tendency for preferential flow. For soil types where contact between the wastewater and the soil may be hindered, we recommend choosing irrigation rates that will limit preferential flow and/or not relying on the soil to retain P when designing the land treatment system.
In conclusion, soil systems vary in their capacity to assimilate wastewater-applied N and P. Choosing soil types and irrigation management practices that minimize preferential flow and maximize the time wastewater spends in the active root zone will increase the opportunity for plant uptake and soil storage of applied nutrients. If preferential flow is minimized, then matching wastewater nutrient loadings to plant N uptake, and to plant P uptake plus sustainable soil storage, will limit leaching of these nutrients. Native soil N may be an additional source of N and needs to be considered when designing land-treatment systems that increase hydraulic loading and leaching of this organic N.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Related articles in JEQ:
This article has been cited by other articles:
|
|
 |
|
  M. McLeod, J. Aislabie, J. Ryburn, and A. McGill Regionalizing Potential for Microbial Bypass Flow through New Zealand Soils J. Environ. Qual., August 8, 2008; 37(5): 1959 - 1967. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. J. Livesley, M. A. Adams, and P. F. Grierson Soil Water Nitrate and Ammonium Dynamics under a Sewage Effluent Irrigated Eucalypt Plantation J. Environ. Qual., October 24, 2007; 36(6): 1883 - 1894. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Mertens, R. Stenger, and G. F. Barkle Multiobjective Inverse Modeling for Soil Parameter Estimation and Model Verification Vadose Zone J., August 24, 2006; 5(3): 917 - 933. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Adeli, D. E. Rowe, and J. J. Read Effects of Soil Type on Bermudagrass Response to Broiler Litter Application Agron. J., January 5, 2006; 98(1): 148 - 155. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
