- Email: [email protected]
Minimising phosphorus losses from the soil matrix
Available online at www.sciencedirect.com
Minimising phosphorus losses from the soil matrix Richard W McDowell Phosphorus loss from land, due to agricultural intensification, can impair water quality. The quantity lost is a function of runoff and availability, which is affected by inputs and the ability of the soil to retain P. Losses are exacerbated if surface runoff or drainage occurs soon after P inputs (e.g. fertiliser and/or manure and dung). Strategies to mitigate P losses depend on the farming system. The first step is to maintain a farm P balance (inputs–outputs) close to zero and the agronomic optimum. The next step is to use mitigation strategies in areas that lose the most P, but occupy little of the farm or catchment’s area. Focusing on these areas, termed critical source areas, is more cost-effective than farm or catchment-wide strategies. However, the worry is that mitigation strategies may not keep pace with losses due to increasing intensification. Therefore, a proactive approach is needed that identifies areas resilient to P inputs and unlikely to lose P if land use is intensified. Address AgResearch, Invermay Agricultural Centre, Private Bag 50034, Mosgiel, New Zealand Corresponding author: McDowell, Richard W ([email protected])
Current Opinion in Biotechnology 2012, 23:860–865 This review comes from a themed issue on Phosphorus biotechnology Edited by Andrew N Shilton and Lars M Blank For a complete overview see the Issue and the Editorial Available online 28th March 2012 0958-1669/$ – see front matter, # 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.copbio.2012.03.006
of P tend to be low. Dodds and Oaks [4] predicted total P concentrations in streams and rivers of the conterminous United States in the absence of human land use were less than 50 mg L 1. This is often an order of magnitude less than concentrations in developed catchments [5]. Research over the past three decades has revealed the edaphic (e.g. soil and climatic) factors and management (e.g. the placement and timing of P inputs) practices that result in P loss. While the setting of limits to prevent or improve surface water quality will often take into account economic and social constraints, it is the role of process-based research that defines how achievable a target is (e.g. without harsh financial penalties), or whether or not that target will result in improved water quality [6]. This review outlines the processes involved in P loss, examines some of the technologies devised to mitigate P loss and provides a commentary on how mitigation may be best achieved.
What are the processes controlling the availability of P for loss? Most P loss originates as diffuse sources from agricultural production systems, partly, owing to our success in identifying and mitigating of point sources [7]. In general, P loss increases as the portion of the catchment under agriculture increases [8]. Losses can occur via surface runoff or subsurface flow (viz. subsurface runoff). Runoff from forests, grasslands, and other non-cultivated soils carries little sediment, so P losses are generally dominated by dissolved P, which is immediately algal-available [9]. The cultivation of agricultural land greatly increases erosion, and with it, the loss of particulate-bound P (60–90% of total P; [10]). Some of the particle-bound P is not readily available, but can be a long-term source of P for aquatic biota [9].
Introduction Much work has shown that the efficiency of phosphorus (P) use needs to increase to sustain or improve crop yields, but also to prevent the deterioration of surface water bodies via eutrophication [1]. Isner and Bennett [2] point that about half of the P applied to land (80% of the P mined) is lost to waterways via soil erosion and leaching. This is an expensive waste of P, especially if another price rise, similar to that in 2007–2008, occurs [3]. However, potentially outweighing the cost of P are the effects of eutrophication both monetary (e.g. fish death, effects on tourism and treatment costs if used as potable water) and social or cultural (e.g. loss of waterways for recreation). The cause of P loss is simple: anthropogenic inputs and management of P once in the farm system. Natural losses Current Opinion in Biotechnology 2012, 23:860–865
In acidic soils, P is largely sorbed to Al-oxides and Feoxides, whereas in neutral to alkaline soils, P occurs primarily as Ca-phosphates and Mg-phosphates often precipitated, or sorbed, onto Ca and Mg carbonates. Organic P can form a significant part of soil P, especially in acidic soils and soils that contain much organic matter and nitrogen. The first factor involved in dissolved P loss is soil P solubility. This is a reflection of how much P is added to the soil and the soils’ ability to retain P. Heckrath et al. [11] was one of the first to show a relationship between P losses in tile drainage and soil test P (STP) concentration (i.e. agronomic tests such as Bray, Mehlich, or Olsen). The potential for dissolved P loss in surface runoff or subsurface flow was later approximated by a water or 0.01 M CaCl2 solution [12]. Generally, the relationship between soil test P and runoff P is curvilinear, with P losses increasing with STP www.sciencedirect.com
Phosphorus loss McDowell
861
Figure 1
16000
0.7
0.6
12000 0.5 10000 0.4 8000 0.3 6000
0.01M CaCl2-P (mg/L)
Pasture production (kg dry matter/ha)
14000
0.2 4000
Control 188 SSP 376 SSP 0.01M CaCl2-P
2000
0.1
0
0.0 0
10
20
30
40
Olsen P (mg/L) Current Opinion in Biotechnology
The gap between the agronomic optimum in Olsen P for pasture production (up arrow) and a potential threshold in Olsen P for P loss in subsurface drainage (as estimated by 0.01 M CaCl2-P; down arrow) shows there is little justification in exceeding the agronomic optimum. Data are from plots receiving different rates of superphosphate (SSP; kg ha 1 yr 1) in a trial in Canterbury, New Zealand [44].
and the capacity of the soil to retain added P via sorption decreases. This relationship can also be split into two straight lines, either side of a STP concentration, touted by Hesketh and Brookes [13] as an environmental threshold. Unfortunately, this threshold differed according to the concentration of sorbing materials such as Al-oxides and Fe-oxides or soil pH [14]. Nevertheless, the recommendation of not exceeding an agronomic optimum in STP concentration, as it may exceed an environmental threshold (Figure 1), appears to have been widely adopted [15]. Recent work by Dodd et al. [16] shows that halting fertiliser applications to soils with STP exceeding the agronomic optimum decreases water extractable P at a rate proportional to the soils original STP and P retention capacity. However, they noted a soil may take decades to achieve a water extractable P concentration acceptable to receiving waters, and in the meantime be less profitable. A quicker approach to decreasing water extractable P, but maintain profitability, may be to have a negative P balance by applying less P and more N.
preventing the interaction of manure with runoff, and invertebrate action buries P into the soil [17]. Overall, the magnitude of loss will depend on the rate of application, but also the form and solubility of P. For instance, a low soluble P fertiliser like reactive phosphate rock has been shown to decrease P losses at a catchment scale by about a third compared to highly water soluble superphosphate [18]. Another example is that due a lack of the phytase enzyme, monogastric animals such as pigs and poultry produce manure that contains more P as phytate than manure-P from ruminants. With six phosphate moieties, phytate binds strongly to soil and is thought to be less plant (and possibly algal) available [19]. The addition of phytase into animal feed makes releases most of the orthophosphate in phytate. However, if the change in manure-P form is not accounted for when applying back to land, the increase in orthophosphate can result in increased P losses [20].
Inputs that determine STP concentration include fertilisers and manure or dung (if grazed). The potential for loss from fertiliser or manure/dung is greatest soon after application, and declines exponentially with time as slurry or manure infiltrates into the soil, manure crusts-over
Although these chemical principles can describe the availability of P, hydrologic processes determine whether or not losses occur. As the driver for runoff, rainfall events can be divided into (A) rainfall of low intensity and high frequency that tends to move P in subsurface flow, and,
www.sciencedirect.com
Transport of P from the soil matrix to waterways
Current Opinion in Biotechnology 2012, 23:860–865
862 Phosphorus biotechnology
Figure 2
Percentage of P lost via Infiltration-excess surface runoff
100 Sheep and red deer grazed catchment Sheep and beef cattle grazed catchment 80
90% of all summer-autumn storms
60
40
20
0 0
200
400
600
800
1000
1200
1400
6000 12000
Total stream flow (m3) Current Opinion in Biotechnology
Graph showing the proportion of P contributed by infiltration-excess surface runoff decreases with increasing streamflow as saturation-excess and storm sizes increase. The grey box represents the area where 90% of the summer and fall storms occurred. Data from McDowell and Srinivasan [25].
(B) rainfall events of high intensity and low frequency that tends to move P in surface runoff from a thin layer of P-rich topsoil. As high intensity storms have more kinetic energy and erosive power, more P, especially in particulate forms is lost during surface runoff than storms of low intensity that result in subsurface flow. For example, Sharpley et al. [21] found that despite only accounting for 32% of total flow, surface runoff derived storms contributed the vast majority of total P (80%), much of it as particulate-P. Similarly, surface runoff can be further divided into Hortonian (limited by infiltration rate) and saturation-excess (limited by soil water storage capacity). Infiltration-rate limited surface runoff will have a greater capacity to detach and move soil particles [22], but is restricted to either large storm events or areas where the infiltration rate has been decreased or is low (e.g. dispersion, cattle treading damage, or tracks and lanes). Both saturation-excess and infiltration-excess conditions often combine in effect to yield a complex pattern of P loss. Saturation-excess surface runoff can be described by variable source area (VSA) hydrology [23]. The size and time that these areas actively produce runoff varies rapidly during a storm as a function of precipitation, soil-type, topography, and moisture status. During a rainfall event, a VSA will expand upslope as the saturated area increases, meaning that surface runoff will be restricted to areas close to the stream in dry summer months compared Current Opinion in Biotechnology 2012, 23:860–865
to wetter winter months when the entire hillslope may contribute runoff. By contrast, where infiltration-excess conditions dominate, parts of the catchment can alternate between sources and sinks of runoff as a function of soil properties and rainfall intensity. Srinivasan and McDowell [24] and McDowell and Srinivasan [25] examined the role of both runoff processes in generating P losses from grazed grassland catchments in New Zealand. They found that the infiltration-excess areas forming part of farm infrastructure (e.g. lanes), or created by animal treading and soil compaction (e.g. stock camps or around watering troughs and gateways), accounted for most of the dissolved P lost in small storms that dominated during summer and autumn (Figure 2). Importantly, summerautumn is also when dissolved P would be most detrimental to stream water quality. This finding would have been missed if attention was only paid to winter when most P was lost from areas that contributed P via saturation-excess surface runoff. Although many studies have found that P losses tend to be dominated by surface runoff, subsurface flow losses may be important under certain soil or hydrologic conditions. The loss of P in subsurface flow generally decreases as the degree of soil–water contact increases, owing to sorption by P-deficient subsoils. Exceptions occur where organic matter may accelerate P loss together with Al and Fe, or where the soil has a small P sorption capacity (e.g. some sandy soils), where subsurface flow www.sciencedirect.com
Phosphorus loss McDowell
travels from P-rich topsoil in via macropores, or is intercepted by artificial drainage [26].
How does management affect P losses? Agricultural systems can be categorised into cropping or grassland systems with or without animals that either graze forage or are confined to a barn and supplied feed. Owing to economics, confined animal feeding operations (CAFOs) tend to be larger, but with less land, than grazed operations [27]. The development of intensive and often specialised farming systems, often as CAFOs, has led to regional surpluses of P imported in fertiliser and animal feed compared with P exported in farm produce [28]. The first step in arresting P losses is therefore to balance P inputs with outputs at farm, catchment or regional scales. However, a farm or catchment can have a near zero P balance, but still lose much P if, for example, STP concentrations are enriched and above plant requirements [6]. Hence, mitigation technologies, in addition to a negative P balance, are required to obtain a level of P loss that is acceptable environmentally and agronomically. For all farm types, the presence of good grassland or crop covers, and crop residues, help minimise P loss by decreasing erosion and surface runoff [29]. Similarly, any practice that intercepts surface runoff before it enters the stream, by infiltration or sediment retention, can be effective [30]. Such measures include riparian zones, buffer strips, terracing, cover crops, contour tillage, wetlands (including constructed wetlands) and impoundments or small reservoirs. However, these practices are better at stopping particulate than dissolved P transport. Other mitigation practices that target dissolved P include manure and soil treatment with amendments (e.g. aluminium sulphate) to decrease P solubility and potential release to runoff. If STP concentrations are enriched, periodic tillage of the soil may decrease P loss by redistributing high-P topsoil throughout the root zone [31]. Aeration, which improves infiltration, can decrease P losses where surface runoff is driven by infiltration-excess conditions [32], but is ineffective under saturation-excess conditions [33]. Animal confinement places emphasis on handling manure well, especially where there is insufficient land available to dispose of manure resulting in soil P enrichment. Practices to decrease P loss include the use of soil testing to guide future manure P applications (and not applying manure based on nitrogen contents) and redistribution of manure within and among farms to avoid soil P build ups. Since the availability of manure-P declines exponentially with time since application, applying manure should be restricted to times of year when runoff is unlikely (including snowmelt). Application technologies that incorporate manure into the soil as a slurry also decrease P loss www.sciencedirect.com
863
by increasing interaction and sorption with the soil matrix [34]. Alternatively, incorporating P within the plough layer immediately after application can decrease P losses if erosion is minimised. Many of the mitigation strategies employed on cropping farms, and farms using animal confinement, are applicable for grazing operations, but differences in how a grazed system operates requires additional or modified strategies to decrease P loss. Perhaps the most profound effect is the potential for decreased soil infiltration due to animal treading, especially when soils are wet or in and around areas like watering troughs that receive frequent animal or vehicle traffic [35]. Positioning obvious infiltrationexcess areas like farm tracks, lanes, and gateways so that flow does not reach waterways, and avoiding treading damage with strategies such as restricted grazing during early spring, can decrease P losses [36]. Failure to fence streams off from grazing animals can lead to considerable direct dung-P inputs (up to 30% of catchment loads; [37]) and bank erosion [38]. Manure on grazed dairies takes the form of a dilute slurry called farm dairy effluent (<1% solids), but poor application strategies (i.e. when soils are wet) can result in significant P loss. The answer is to increase storage capacity and apply the effluent at a low rate (e.g. <4 mm h 1) to dry soil, thereby increasing interaction and uptake by the soil [39]. Individually each of these strategies can mitigate a component of P loss; however, only when used collectively in the right place and at the right time can the greatest effect be achieved. This requires knowledge of the spatial and temporal variation of P losses. A hypothesis proposed by Pionke et al. [40] and others is that the majority of P losses come from a minority of a catchment’s area – termed critical source areas (CSAs). This has been widely adopted and explored further on the basis that the cost-effectiveness of mitigation strategies can be maximised if applied to CSAs and not across an entire catchment or farm. McDowell and Nash [41] argued that for New Zealand and Australian dairy farms, which receive no financial subsidies or incentives to decrease P losses, optimal placement and timing of mitigation strategies is the best option to minimise any impact on profitability.
How low can you go? In a programme to prevent eutrophication within a catchment, limits are often set that correspond to an acceptable level of algal growth. These are often set for warmer times of year when algal growth is likely and when freshwater is actively used. The setting of limits in terms of concentrations, not loads, can avoid a situation where loads are small due to little rainfall, but concentrations (and effects) in the stream are high. Kleinman et al. [42] argued that targets should be realistic and cite a case in the Chesapeake Bay where in 1987 a reduction target of 40% in P (and N) was to be achieved by 2000. Catchment Current Opinion in Biotechnology 2012, 23:860–865
864 Phosphorus biotechnology
programmes that focused on the ‘low hanging fruit’ achieved a 25% decrease. This target was perhaps ambitious given the complexity of political and economic restrictions. However, in other cases, focusing on ‘low hanging fruit’ may be able to achieve significant effects. For instance, McDowell et al. [43] showed that focusing on better methods to apply dairy shed effluent could significantly decrease P losses and eutrophication in the Pomahaka River, Otago, New Zealand.
Acknowledgements
The optimisation of mitigation strategies, in combination with furthering our understanding of CSAs, will enable us to achieve better decreases in P loss especially in areas of intensive land use. However, there is probably a limit to what is achievable. This will depend on how ‘polluted’ the waterway is, and what is the natural baseline (viz. reference conditions). It is fortunate that the setting of targets often involves the local community who may recall that a river was naturally polluted, and hence may not expect it to be like rivers in other areas. Nevertheless, the worry is that our ability to optimise systems to decrease P losses may not keep pace with the rate of intensification and P loss. Furthermore, given that policy can be reactionary, and often require science to establish the main leakage points before they are managed, the frequency of water quality impairment may increase in the future. I contest that the best approach to solve the problem of P losses, in light of increasing intensification, is to be proactive and identify freshwater systems that are resilient to P inputs, CSAs for mitigation management, and combine these with profitable farming systems that lose little P.
1. Carptenter SR: Eutrophication control is critical to mitigating eutrophication. Proc Natl Acad Sci USA 2008, 105:11039-11040. Reaffirms the need to focus on the management of P loss if eutrophication is to be stopped, but also highlights that this has been the case globally for centuries.
This review was funded as part of the New Zealand Ministry for Science and Innovation’s Clean Water, Productive Land programme (contract C10X006).
References and recommended reading Papers of particular interest, published within the period of review, have been highlighted as: of special interest of outstanding interest
2.
Isner J, Bennett E: A broken biogeochemical cycle. Nature 2011, 489:29-31.
3. Vaccari DA: Phosphorus famine: the threat to our food supply. Sci Am 2009, June:42-47. One of the first papers to outline the need to conserve finite P resources without emotive language. 4.
Dodds WK, Oaks RM: A technique for establishing reference nutrient concentrations across watersheds affected by humans. Limnol Oceanogr: Methods 2004, 2:333-341. Outlines a method for estimating baseline (or reference) conditions using regression techniques.
5.
Soranno PA, Wagner T, Martin SL, McLean C, Novitski LN, Provence CD, Rober AR: Quantifying regional reference conditions for freshwater ecosystem management: a comparison of approaches and future research needs. Lake Reserv Manage 2011, 27:138-148. Excellent summary of the current state of freshwater P management relative to the setting of different limits or thresholds.
6.
Woolsey S, Capelli F, Gosner T, Hoehn E, Hostmann M, Junker B, Paetzold A, Roulier C, Schweizer S, Tiegs SD et al.: A strategy to assess river restoration success. Freshwater Biol 2007, 52:752769.
7.
Conclusions The availability of P for loss by transport mechanisms such as surface runoff and subsurface flow tends to increase with intensification viz. the frequency and quantity of inputs via fertiliser, the saturation of the soil’s sorptive matrix with increasing STP, or surface applications of dung (via grazing animals) or manure. However, hydrologic transport mechanisms vary greatly in space and time. The coincidence of soil-specific or management-specific factors that influence both availability and transport of loss has been termed a critical source area. Management of these CSAs is seen as the next step beyond balancing farm P balances and simple measures (low hanging fruit) in mitigating P losses. The hypothesis is that employing mitigation strategies to CSAs will greatly decrease P losses without affecting farm profitability and therefore represent a substantial improvement in P use efficiency. However, the worry is that focusing on mitigating P losses from CSAs may not enable a farm to reach a catchment target, especially if land use intensification occurs. A proactive approach is therefore needed that identifies streams that are resilient to P inputs and areas unlikely to lose much P, if land use is intensified. Current Opinion in Biotechnology 2012, 23:860–865
Withers PJA, Jarvie HP, Stoate C: Quantifying the impact of septic tank systems on eutrophication risk in rural headwaters. Environ Int 2011, 37:644-653. Highlights the role of point sources within an agricultural catchment, categorised in the past as diffuse pollution.
8.
Omernik JM: Nonpoint source – stream nutrient level relationships: a nationwide study. EPA-600/3-77-105. Corvallis, OR: USEPA; 1977.
9.
Sharpley AN: Assessing phosphorus bioavailability in agricultural soils and runoff. Fertilizer Res 1993, 36:259-272.
10. Sharpley AN, Hedley MJ, Sibbesen E, Hillbricht-Ilkowska A, House WA, Ryszkowski L: Phosphorus transfer from terrestrial to aquatic ecosystems. In Phosphorus Cycling in Terrestrial and Aquatic Ecosystems. Edited by Tiessen H. London, UK: J Wiley & Sons; 1995:171-199. 11. Heckrath G, Brookes PC, Poulton PR, Goulding KWT: Phosphorus leaching from soils containing different phosphorus concentrations in the Broadbalk experiment. J Environ Qual 1995, 24:904-910. The first paper to highlight the potential for a threshold in soil test P concentration above which P losses in drainage increase. 12. McDowell RW, Sharpley AN: Approximating phosphorus release from soils to surface runoff and subsurface drainage. J Environ Qual 2001, 30:508-520. Paper that provides a method to assess the potential P loss in surface runoff and subsurface drainage from the measurement of P in simple soil extracts. 13. Hesketh N, Brookes PC: Development of an indicator for risk of phosphorus leaching. J Environ Qual 2000, 29:105-110. 14. Koopmans GF, McDowell RW, Chardon WJ, Oenema O, Dolfing J: Soil phosphorus quantity-intensity relationships to predict www.sciencedirect.com
Phosphorus loss McDowell
increased soil phosphorus loss to overland and subsurface flow. Chemosphere 2002, 48:679-687. 15. Sharpley AN, Kleinman PJA, Jordan P, Bergstro¨m L, Allen AL: Evaluating the success of phosphorus management from field to watershed. J Environ Qual 2009, 38:1981-1988. 16. Dodd RJ, McDowell RW, Condron LM: Predicting the changes in environmentally and agronomically significant phosphorus forms following the cessation of phosphorus fertiliser applications to grassland. Soil Use Manage 2012, 28:in press.
865
31. Sharpley AN: Soil mixing to decrease stratification of phosphorus in manured soils. J Environ Qual 2003, 32:1375-1384. 32. Moore PA Jr, Formica SJ, Van Epps M, Delaune PB: Effect of pasture renovation on nutrient runoff from pastures fertilized with manure. In Proceedings of the 7th International Symposium on Livestock Environment, American Society of Agricultural Engineers, vol 701. 2005:301.
17. Vadas PA, Harmel RD, Kleinman PJA: Transformations of soil and manure phosphorus after surface application of manure to field plots. Nutr Cycl Agroecosyst 2007, 77:83-99.
33. Curran Cournane F, McDowell RW, Littlejohn R, Houlbrooke D, Condron LM: Is mechanical soil aeration a strategy to alleviate soil compaction and decrease phosphorus and sediment losses from irrigated and rain-fed cattle-grazed pastures? Soil Use Manage 2011, 27:376-384.
18. McDowell RW, Littlejohn RP, Blennerhassett JD: Phosphorus fertiliser form affects phosphorus loss to waterways: a paired catchment study. Soil Use Manage 2010, 26:365-373.
34. Daverede IC, Kravchenko AN, Hoeft RG, Nafziger ED, Bullock DG, Warren JJ, Gonzini LC: Phosphorus runoff: effect of tillage and soil phosphorus levels. J Environ Qual 2003, 32:1436-1444.
19. Turner BJ, McKelvie ID, Haygarth PM: Characterisation of waterextractable soil organic phosphorus by phosphatase hydrolysis. Soil Biol Biochem 2002, 34:27-35.
35. Bilotta GS, Brazier RW, Haygarth PM: Processes affecting transfer of sediment and colloids, with associated phosphorus, from intensively farmed grasslands: erosion. Hydrol Process 2007, 21:135-139.
20. Maguire RO, Sims JT, Saylor WW, Turner BL, Angel R, Applegate TJ: Influence of phytase addition to poultry diets on phosphorus forms and solubility in litters and amended soils. J Environ Qual 2004, 33:2306-2316.
36. Lucci GM, McDowell RW, Condron LM: Phosphorus and sediment loads/transfer from potential source areas in New Zealand dairy pasture. Soil Use Manage 2010, 26:44-52.
21. Sharpley AN, Kleinman PJA, Heathwaite AL, Gburek WJ, Folmar GJ, Schmidt JP: Phosphorus loss from an agricultural watershed as a function of storm size. J Environ Qual 2008, 37:362-368.
37. James E, Kleinman P, Stedman R, Sharpley A: Phosphorus contributions from pastured dairy cattle to streams of the Cannonsville Watershed, New York. J Soil Water Conserv 2007, 62:40-47.
22. Kleinman PJA, Srinivasan MS, Dell CJ, Schmidt JP, Sharpley AN, Bryant RB: Role of rainfall intensity and hydrology in nutrient transport via surface runoff. J Environ Qual 2006, 35:1248-1259.
38. McDowell RW, Wilcock RJ: Sources of sediment and phosphorus in stream flow of a highly productive dairy farmed catchment. J Environ Qual 2007, 36:540-548.
23. Ward RC: On the response to precipitation of headwater streams in humid areas. J Hydrol 1984, 74:171-189. 24. Srinivasan MS, McDowell RW: Identifying critical source areas for water quality: 1. Mapping and validating transport areas in three headwater catchments in Otago, New Zealand. J Hydrol 2009, 379:54-67. 25. McDowell RW, Srinivasan MS: Identifying critical source areas for water quality: 2. Validating the approach for phosphorus and sediment losses in grazed headwater catchments. J Hydrol 2009, 379:68-80. 26. van Beek CL, van der Salm C, Plette ACC, van de Weerd H: Nutrient loss pathways from grazed grasslands and the effects of decreasing inputs: experimental results for three soil types. Nutr Cycl Agroecosyst 2009, 83:99-110. 27. McDowell RW, Kleinman PJA: Efficiency of phosphorus cycling in different grassland systems. In Grassland Productivity and Ecosystem Services. Edited by Lemaire G, Hodgson J, Chabbi A. Oxon, UK: CAB International; 2011:108-119. 28. Kellogg RL, Lander CH, Moffitt DC, Gollehon N: Manure nutrients relative to the capacity of cropland and pastureland to assimilate nutrients: spatial and temporal trends for the United States. USDA-NRCS, USDA-ERS, nps00-0579, GSA National Forms and Publication Center, Fort Worth, TX; 2000, available online at http://www.nrcs.usda.gov/technical/NRI/pubs/ manntr.html. 29. Withers PJA, Jarvis SC: Mitigation options for diffuse phosphorus loss to water. Soil Use Manage 1998, 14:186-192. 30. Maxted JR, McCready CH, Scarsbrook MR: Effects of small ponds on stream water quality and macroinvertebrate communities. N Z J Mar Freshwater Res 2005, 39:1069-1084.
www.sciencedirect.com
39. Houlbrooke DJ, Horne DJ, Hedley MJ, Hanly JA, Scotter DR, Snow VO: Minimising surface water pollution resulting from farm-dairy effluent application to mole-pipe drained soils. I. An evaluation of the deferred irrigation system for sustainable land treatment in the Manawatu. N Z J Agric Res 2004, 47:405-415. 40. Pionke HB, Gburek WJ, Sharpley AN: Critical source area controls on water quality in an agricultural watershed located in the Chesapeake Basin. Ecol Eng 2000, 14:325-335. 41. McDowell RW, Nash D: A review of the cost-effectiveness and suitability of mitigation strategies to prevent phosphorus loss from dairy farms in New Zealand and Australia. J Environ Qual 2012, 41:in press. Highlights two hypotheses: (1) that without subsidies the best chance of decreasing P loss is to provide a wide range of fully costed mitigation strategies so that the end-user can mix and match the best strategy to suit their farm, and (2) that the best mitigation is only achieved when strategies are used where and when they are needed viz. critical source areas vary in space and time. 42. Kleinman PJA, Sharpley AN, McDowell RW, Flaten DN, Buda AR, Tao L, Bergstrom L, Zhu Q: Managing agricultural phosphorus for water quality protection: principles for progress. Plant Soil 2011, 349:169-182. Paper that outlines those future issues facing the management of P for good water quality. A highlight is the discussion of the consequence of a legacy of high P losses from high P soils despite our best efforts (e.g. no further P additions) to mitigate them. 43. McDowell RW, Snelder T, Littlejohn R, Hickey M, Cox N, Booker DJ: State and potential management to improve water quality in an agricultural catchment relative to a natural baseline. Agric Ecosyst Environ 2011, 144:188-200. 44. McDowell RW, Condron LM: Phosphorus and the Winchmore trials: review and lessons learnt. N Zeal J Agric Res 2012, 55:In press.
Current Opinion in Biotechnology 2012, 23:860–865
Minimising phosphorus losses from the soil matrix Richard W McDowell Phosphorus loss from land, due to agricultural intensification, can impair water quality. The quantity lost is a function of runoff and availability, which is affected by inputs and the ability of the soil to retain P. Losses are exacerbated if surface runoff or drainage occurs soon after P inputs (e.g. fertiliser and/or manure and dung). Strategies to mitigate P losses depend on the farming system. The first step is to maintain a farm P balance (inputs–outputs) close to zero and the agronomic optimum. The next step is to use mitigation strategies in areas that lose the most P, but occupy little of the farm or catchment’s area. Focusing on these areas, termed critical source areas, is more cost-effective than farm or catchment-wide strategies. However, the worry is that mitigation strategies may not keep pace with losses due to increasing intensification. Therefore, a proactive approach is needed that identifies areas resilient to P inputs and unlikely to lose P if land use is intensified. Address AgResearch, Invermay Agricultural Centre, Private Bag 50034, Mosgiel, New Zealand Corresponding author: McDowell, Richard W ([email protected])
Current Opinion in Biotechnology 2012, 23:860–865 This review comes from a themed issue on Phosphorus biotechnology Edited by Andrew N Shilton and Lars M Blank For a complete overview see the Issue and the Editorial Available online 28th March 2012 0958-1669/$ – see front matter, # 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.copbio.2012.03.006
of P tend to be low. Dodds and Oaks [4] predicted total P concentrations in streams and rivers of the conterminous United States in the absence of human land use were less than 50 mg L 1. This is often an order of magnitude less than concentrations in developed catchments [5]. Research over the past three decades has revealed the edaphic (e.g. soil and climatic) factors and management (e.g. the placement and timing of P inputs) practices that result in P loss. While the setting of limits to prevent or improve surface water quality will often take into account economic and social constraints, it is the role of process-based research that defines how achievable a target is (e.g. without harsh financial penalties), or whether or not that target will result in improved water quality [6]. This review outlines the processes involved in P loss, examines some of the technologies devised to mitigate P loss and provides a commentary on how mitigation may be best achieved.
What are the processes controlling the availability of P for loss? Most P loss originates as diffuse sources from agricultural production systems, partly, owing to our success in identifying and mitigating of point sources [7]. In general, P loss increases as the portion of the catchment under agriculture increases [8]. Losses can occur via surface runoff or subsurface flow (viz. subsurface runoff). Runoff from forests, grasslands, and other non-cultivated soils carries little sediment, so P losses are generally dominated by dissolved P, which is immediately algal-available [9]. The cultivation of agricultural land greatly increases erosion, and with it, the loss of particulate-bound P (60–90% of total P; [10]). Some of the particle-bound P is not readily available, but can be a long-term source of P for aquatic biota [9].
Introduction Much work has shown that the efficiency of phosphorus (P) use needs to increase to sustain or improve crop yields, but also to prevent the deterioration of surface water bodies via eutrophication [1]. Isner and Bennett [2] point that about half of the P applied to land (80% of the P mined) is lost to waterways via soil erosion and leaching. This is an expensive waste of P, especially if another price rise, similar to that in 2007–2008, occurs [3]. However, potentially outweighing the cost of P are the effects of eutrophication both monetary (e.g. fish death, effects on tourism and treatment costs if used as potable water) and social or cultural (e.g. loss of waterways for recreation). The cause of P loss is simple: anthropogenic inputs and management of P once in the farm system. Natural losses Current Opinion in Biotechnology 2012, 23:860–865
In acidic soils, P is largely sorbed to Al-oxides and Feoxides, whereas in neutral to alkaline soils, P occurs primarily as Ca-phosphates and Mg-phosphates often precipitated, or sorbed, onto Ca and Mg carbonates. Organic P can form a significant part of soil P, especially in acidic soils and soils that contain much organic matter and nitrogen. The first factor involved in dissolved P loss is soil P solubility. This is a reflection of how much P is added to the soil and the soils’ ability to retain P. Heckrath et al. [11] was one of the first to show a relationship between P losses in tile drainage and soil test P (STP) concentration (i.e. agronomic tests such as Bray, Mehlich, or Olsen). The potential for dissolved P loss in surface runoff or subsurface flow was later approximated by a water or 0.01 M CaCl2 solution [12]. Generally, the relationship between soil test P and runoff P is curvilinear, with P losses increasing with STP www.sciencedirect.com
Phosphorus loss McDowell
861
Figure 1
16000
0.7
0.6
12000 0.5 10000 0.4 8000 0.3 6000
0.01M CaCl2-P (mg/L)
Pasture production (kg dry matter/ha)
14000
0.2 4000
Control 188 SSP 376 SSP 0.01M CaCl2-P
2000
0.1
0
0.0 0
10
20
30
40
Olsen P (mg/L) Current Opinion in Biotechnology
The gap between the agronomic optimum in Olsen P for pasture production (up arrow) and a potential threshold in Olsen P for P loss in subsurface drainage (as estimated by 0.01 M CaCl2-P; down arrow) shows there is little justification in exceeding the agronomic optimum. Data are from plots receiving different rates of superphosphate (SSP; kg ha 1 yr 1) in a trial in Canterbury, New Zealand [44].
and the capacity of the soil to retain added P via sorption decreases. This relationship can also be split into two straight lines, either side of a STP concentration, touted by Hesketh and Brookes [13] as an environmental threshold. Unfortunately, this threshold differed according to the concentration of sorbing materials such as Al-oxides and Fe-oxides or soil pH [14]. Nevertheless, the recommendation of not exceeding an agronomic optimum in STP concentration, as it may exceed an environmental threshold (Figure 1), appears to have been widely adopted [15]. Recent work by Dodd et al. [16] shows that halting fertiliser applications to soils with STP exceeding the agronomic optimum decreases water extractable P at a rate proportional to the soils original STP and P retention capacity. However, they noted a soil may take decades to achieve a water extractable P concentration acceptable to receiving waters, and in the meantime be less profitable. A quicker approach to decreasing water extractable P, but maintain profitability, may be to have a negative P balance by applying less P and more N.
preventing the interaction of manure with runoff, and invertebrate action buries P into the soil [17]. Overall, the magnitude of loss will depend on the rate of application, but also the form and solubility of P. For instance, a low soluble P fertiliser like reactive phosphate rock has been shown to decrease P losses at a catchment scale by about a third compared to highly water soluble superphosphate [18]. Another example is that due a lack of the phytase enzyme, monogastric animals such as pigs and poultry produce manure that contains more P as phytate than manure-P from ruminants. With six phosphate moieties, phytate binds strongly to soil and is thought to be less plant (and possibly algal) available [19]. The addition of phytase into animal feed makes releases most of the orthophosphate in phytate. However, if the change in manure-P form is not accounted for when applying back to land, the increase in orthophosphate can result in increased P losses [20].
Inputs that determine STP concentration include fertilisers and manure or dung (if grazed). The potential for loss from fertiliser or manure/dung is greatest soon after application, and declines exponentially with time as slurry or manure infiltrates into the soil, manure crusts-over
Although these chemical principles can describe the availability of P, hydrologic processes determine whether or not losses occur. As the driver for runoff, rainfall events can be divided into (A) rainfall of low intensity and high frequency that tends to move P in subsurface flow, and,
www.sciencedirect.com
Transport of P from the soil matrix to waterways
Current Opinion in Biotechnology 2012, 23:860–865
862 Phosphorus biotechnology
Figure 2
Percentage of P lost via Infiltration-excess surface runoff
100 Sheep and red deer grazed catchment Sheep and beef cattle grazed catchment 80
90% of all summer-autumn storms
60
40
20
0 0
200
400
600
800
1000
1200
1400
6000 12000
Total stream flow (m3) Current Opinion in Biotechnology
Graph showing the proportion of P contributed by infiltration-excess surface runoff decreases with increasing streamflow as saturation-excess and storm sizes increase. The grey box represents the area where 90% of the summer and fall storms occurred. Data from McDowell and Srinivasan [25].
(B) rainfall events of high intensity and low frequency that tends to move P in surface runoff from a thin layer of P-rich topsoil. As high intensity storms have more kinetic energy and erosive power, more P, especially in particulate forms is lost during surface runoff than storms of low intensity that result in subsurface flow. For example, Sharpley et al. [21] found that despite only accounting for 32% of total flow, surface runoff derived storms contributed the vast majority of total P (80%), much of it as particulate-P. Similarly, surface runoff can be further divided into Hortonian (limited by infiltration rate) and saturation-excess (limited by soil water storage capacity). Infiltration-rate limited surface runoff will have a greater capacity to detach and move soil particles [22], but is restricted to either large storm events or areas where the infiltration rate has been decreased or is low (e.g. dispersion, cattle treading damage, or tracks and lanes). Both saturation-excess and infiltration-excess conditions often combine in effect to yield a complex pattern of P loss. Saturation-excess surface runoff can be described by variable source area (VSA) hydrology [23]. The size and time that these areas actively produce runoff varies rapidly during a storm as a function of precipitation, soil-type, topography, and moisture status. During a rainfall event, a VSA will expand upslope as the saturated area increases, meaning that surface runoff will be restricted to areas close to the stream in dry summer months compared Current Opinion in Biotechnology 2012, 23:860–865
to wetter winter months when the entire hillslope may contribute runoff. By contrast, where infiltration-excess conditions dominate, parts of the catchment can alternate between sources and sinks of runoff as a function of soil properties and rainfall intensity. Srinivasan and McDowell [24] and McDowell and Srinivasan [25] examined the role of both runoff processes in generating P losses from grazed grassland catchments in New Zealand. They found that the infiltration-excess areas forming part of farm infrastructure (e.g. lanes), or created by animal treading and soil compaction (e.g. stock camps or around watering troughs and gateways), accounted for most of the dissolved P lost in small storms that dominated during summer and autumn (Figure 2). Importantly, summerautumn is also when dissolved P would be most detrimental to stream water quality. This finding would have been missed if attention was only paid to winter when most P was lost from areas that contributed P via saturation-excess surface runoff. Although many studies have found that P losses tend to be dominated by surface runoff, subsurface flow losses may be important under certain soil or hydrologic conditions. The loss of P in subsurface flow generally decreases as the degree of soil–water contact increases, owing to sorption by P-deficient subsoils. Exceptions occur where organic matter may accelerate P loss together with Al and Fe, or where the soil has a small P sorption capacity (e.g. some sandy soils), where subsurface flow www.sciencedirect.com
Phosphorus loss McDowell
travels from P-rich topsoil in via macropores, or is intercepted by artificial drainage [26].
How does management affect P losses? Agricultural systems can be categorised into cropping or grassland systems with or without animals that either graze forage or are confined to a barn and supplied feed. Owing to economics, confined animal feeding operations (CAFOs) tend to be larger, but with less land, than grazed operations [27]. The development of intensive and often specialised farming systems, often as CAFOs, has led to regional surpluses of P imported in fertiliser and animal feed compared with P exported in farm produce [28]. The first step in arresting P losses is therefore to balance P inputs with outputs at farm, catchment or regional scales. However, a farm or catchment can have a near zero P balance, but still lose much P if, for example, STP concentrations are enriched and above plant requirements [6]. Hence, mitigation technologies, in addition to a negative P balance, are required to obtain a level of P loss that is acceptable environmentally and agronomically. For all farm types, the presence of good grassland or crop covers, and crop residues, help minimise P loss by decreasing erosion and surface runoff [29]. Similarly, any practice that intercepts surface runoff before it enters the stream, by infiltration or sediment retention, can be effective [30]. Such measures include riparian zones, buffer strips, terracing, cover crops, contour tillage, wetlands (including constructed wetlands) and impoundments or small reservoirs. However, these practices are better at stopping particulate than dissolved P transport. Other mitigation practices that target dissolved P include manure and soil treatment with amendments (e.g. aluminium sulphate) to decrease P solubility and potential release to runoff. If STP concentrations are enriched, periodic tillage of the soil may decrease P loss by redistributing high-P topsoil throughout the root zone [31]. Aeration, which improves infiltration, can decrease P losses where surface runoff is driven by infiltration-excess conditions [32], but is ineffective under saturation-excess conditions [33]. Animal confinement places emphasis on handling manure well, especially where there is insufficient land available to dispose of manure resulting in soil P enrichment. Practices to decrease P loss include the use of soil testing to guide future manure P applications (and not applying manure based on nitrogen contents) and redistribution of manure within and among farms to avoid soil P build ups. Since the availability of manure-P declines exponentially with time since application, applying manure should be restricted to times of year when runoff is unlikely (including snowmelt). Application technologies that incorporate manure into the soil as a slurry also decrease P loss www.sciencedirect.com
863
by increasing interaction and sorption with the soil matrix [34]. Alternatively, incorporating P within the plough layer immediately after application can decrease P losses if erosion is minimised. Many of the mitigation strategies employed on cropping farms, and farms using animal confinement, are applicable for grazing operations, but differences in how a grazed system operates requires additional or modified strategies to decrease P loss. Perhaps the most profound effect is the potential for decreased soil infiltration due to animal treading, especially when soils are wet or in and around areas like watering troughs that receive frequent animal or vehicle traffic [35]. Positioning obvious infiltrationexcess areas like farm tracks, lanes, and gateways so that flow does not reach waterways, and avoiding treading damage with strategies such as restricted grazing during early spring, can decrease P losses [36]. Failure to fence streams off from grazing animals can lead to considerable direct dung-P inputs (up to 30% of catchment loads; [37]) and bank erosion [38]. Manure on grazed dairies takes the form of a dilute slurry called farm dairy effluent (<1% solids), but poor application strategies (i.e. when soils are wet) can result in significant P loss. The answer is to increase storage capacity and apply the effluent at a low rate (e.g. <4 mm h 1) to dry soil, thereby increasing interaction and uptake by the soil [39]. Individually each of these strategies can mitigate a component of P loss; however, only when used collectively in the right place and at the right time can the greatest effect be achieved. This requires knowledge of the spatial and temporal variation of P losses. A hypothesis proposed by Pionke et al. [40] and others is that the majority of P losses come from a minority of a catchment’s area – termed critical source areas (CSAs). This has been widely adopted and explored further on the basis that the cost-effectiveness of mitigation strategies can be maximised if applied to CSAs and not across an entire catchment or farm. McDowell and Nash [41] argued that for New Zealand and Australian dairy farms, which receive no financial subsidies or incentives to decrease P losses, optimal placement and timing of mitigation strategies is the best option to minimise any impact on profitability.
How low can you go? In a programme to prevent eutrophication within a catchment, limits are often set that correspond to an acceptable level of algal growth. These are often set for warmer times of year when algal growth is likely and when freshwater is actively used. The setting of limits in terms of concentrations, not loads, can avoid a situation where loads are small due to little rainfall, but concentrations (and effects) in the stream are high. Kleinman et al. [42] argued that targets should be realistic and cite a case in the Chesapeake Bay where in 1987 a reduction target of 40% in P (and N) was to be achieved by 2000. Catchment Current Opinion in Biotechnology 2012, 23:860–865
864 Phosphorus biotechnology
programmes that focused on the ‘low hanging fruit’ achieved a 25% decrease. This target was perhaps ambitious given the complexity of political and economic restrictions. However, in other cases, focusing on ‘low hanging fruit’ may be able to achieve significant effects. For instance, McDowell et al. [43] showed that focusing on better methods to apply dairy shed effluent could significantly decrease P losses and eutrophication in the Pomahaka River, Otago, New Zealand.
Acknowledgements
The optimisation of mitigation strategies, in combination with furthering our understanding of CSAs, will enable us to achieve better decreases in P loss especially in areas of intensive land use. However, there is probably a limit to what is achievable. This will depend on how ‘polluted’ the waterway is, and what is the natural baseline (viz. reference conditions). It is fortunate that the setting of targets often involves the local community who may recall that a river was naturally polluted, and hence may not expect it to be like rivers in other areas. Nevertheless, the worry is that our ability to optimise systems to decrease P losses may not keep pace with the rate of intensification and P loss. Furthermore, given that policy can be reactionary, and often require science to establish the main leakage points before they are managed, the frequency of water quality impairment may increase in the future. I contest that the best approach to solve the problem of P losses, in light of increasing intensification, is to be proactive and identify freshwater systems that are resilient to P inputs, CSAs for mitigation management, and combine these with profitable farming systems that lose little P.
1. Carptenter SR: Eutrophication control is critical to mitigating eutrophication. Proc Natl Acad Sci USA 2008, 105:11039-11040. Reaffirms the need to focus on the management of P loss if eutrophication is to be stopped, but also highlights that this has been the case globally for centuries.
This review was funded as part of the New Zealand Ministry for Science and Innovation’s Clean Water, Productive Land programme (contract C10X006).
References and recommended reading Papers of particular interest, published within the period of review, have been highlighted as: of special interest of outstanding interest
2.
Isner J, Bennett E: A broken biogeochemical cycle. Nature 2011, 489:29-31.
3. Vaccari DA: Phosphorus famine: the threat to our food supply. Sci Am 2009, June:42-47. One of the first papers to outline the need to conserve finite P resources without emotive language. 4.
Dodds WK, Oaks RM: A technique for establishing reference nutrient concentrations across watersheds affected by humans. Limnol Oceanogr: Methods 2004, 2:333-341. Outlines a method for estimating baseline (or reference) conditions using regression techniques.
5.
Soranno PA, Wagner T, Martin SL, McLean C, Novitski LN, Provence CD, Rober AR: Quantifying regional reference conditions for freshwater ecosystem management: a comparison of approaches and future research needs. Lake Reserv Manage 2011, 27:138-148. Excellent summary of the current state of freshwater P management relative to the setting of different limits or thresholds.
6.
Woolsey S, Capelli F, Gosner T, Hoehn E, Hostmann M, Junker B, Paetzold A, Roulier C, Schweizer S, Tiegs SD et al.: A strategy to assess river restoration success. Freshwater Biol 2007, 52:752769.
7.
Conclusions The availability of P for loss by transport mechanisms such as surface runoff and subsurface flow tends to increase with intensification viz. the frequency and quantity of inputs via fertiliser, the saturation of the soil’s sorptive matrix with increasing STP, or surface applications of dung (via grazing animals) or manure. However, hydrologic transport mechanisms vary greatly in space and time. The coincidence of soil-specific or management-specific factors that influence both availability and transport of loss has been termed a critical source area. Management of these CSAs is seen as the next step beyond balancing farm P balances and simple measures (low hanging fruit) in mitigating P losses. The hypothesis is that employing mitigation strategies to CSAs will greatly decrease P losses without affecting farm profitability and therefore represent a substantial improvement in P use efficiency. However, the worry is that focusing on mitigating P losses from CSAs may not enable a farm to reach a catchment target, especially if land use intensification occurs. A proactive approach is therefore needed that identifies streams that are resilient to P inputs and areas unlikely to lose much P, if land use is intensified. Current Opinion in Biotechnology 2012, 23:860–865
Withers PJA, Jarvie HP, Stoate C: Quantifying the impact of septic tank systems on eutrophication risk in rural headwaters. Environ Int 2011, 37:644-653. Highlights the role of point sources within an agricultural catchment, categorised in the past as diffuse pollution.
8.
Omernik JM: Nonpoint source – stream nutrient level relationships: a nationwide study. EPA-600/3-77-105. Corvallis, OR: USEPA; 1977.
9.
Sharpley AN: Assessing phosphorus bioavailability in agricultural soils and runoff. Fertilizer Res 1993, 36:259-272.
10. Sharpley AN, Hedley MJ, Sibbesen E, Hillbricht-Ilkowska A, House WA, Ryszkowski L: Phosphorus transfer from terrestrial to aquatic ecosystems. In Phosphorus Cycling in Terrestrial and Aquatic Ecosystems. Edited by Tiessen H. London, UK: J Wiley & Sons; 1995:171-199. 11. Heckrath G, Brookes PC, Poulton PR, Goulding KWT: Phosphorus leaching from soils containing different phosphorus concentrations in the Broadbalk experiment. J Environ Qual 1995, 24:904-910. The first paper to highlight the potential for a threshold in soil test P concentration above which P losses in drainage increase. 12. McDowell RW, Sharpley AN: Approximating phosphorus release from soils to surface runoff and subsurface drainage. J Environ Qual 2001, 30:508-520. Paper that provides a method to assess the potential P loss in surface runoff and subsurface drainage from the measurement of P in simple soil extracts. 13. Hesketh N, Brookes PC: Development of an indicator for risk of phosphorus leaching. J Environ Qual 2000, 29:105-110. 14. Koopmans GF, McDowell RW, Chardon WJ, Oenema O, Dolfing J: Soil phosphorus quantity-intensity relationships to predict www.sciencedirect.com
Phosphorus loss McDowell
increased soil phosphorus loss to overland and subsurface flow. Chemosphere 2002, 48:679-687. 15. Sharpley AN, Kleinman PJA, Jordan P, Bergstro¨m L, Allen AL: Evaluating the success of phosphorus management from field to watershed. J Environ Qual 2009, 38:1981-1988. 16. Dodd RJ, McDowell RW, Condron LM: Predicting the changes in environmentally and agronomically significant phosphorus forms following the cessation of phosphorus fertiliser applications to grassland. Soil Use Manage 2012, 28:in press.
865
31. Sharpley AN: Soil mixing to decrease stratification of phosphorus in manured soils. J Environ Qual 2003, 32:1375-1384. 32. Moore PA Jr, Formica SJ, Van Epps M, Delaune PB: Effect of pasture renovation on nutrient runoff from pastures fertilized with manure. In Proceedings of the 7th International Symposium on Livestock Environment, American Society of Agricultural Engineers, vol 701. 2005:301.
17. Vadas PA, Harmel RD, Kleinman PJA: Transformations of soil and manure phosphorus after surface application of manure to field plots. Nutr Cycl Agroecosyst 2007, 77:83-99.
33. Curran Cournane F, McDowell RW, Littlejohn R, Houlbrooke D, Condron LM: Is mechanical soil aeration a strategy to alleviate soil compaction and decrease phosphorus and sediment losses from irrigated and rain-fed cattle-grazed pastures? Soil Use Manage 2011, 27:376-384.
18. McDowell RW, Littlejohn RP, Blennerhassett JD: Phosphorus fertiliser form affects phosphorus loss to waterways: a paired catchment study. Soil Use Manage 2010, 26:365-373.
34. Daverede IC, Kravchenko AN, Hoeft RG, Nafziger ED, Bullock DG, Warren JJ, Gonzini LC: Phosphorus runoff: effect of tillage and soil phosphorus levels. J Environ Qual 2003, 32:1436-1444.
19. Turner BJ, McKelvie ID, Haygarth PM: Characterisation of waterextractable soil organic phosphorus by phosphatase hydrolysis. Soil Biol Biochem 2002, 34:27-35.
35. Bilotta GS, Brazier RW, Haygarth PM: Processes affecting transfer of sediment and colloids, with associated phosphorus, from intensively farmed grasslands: erosion. Hydrol Process 2007, 21:135-139.
20. Maguire RO, Sims JT, Saylor WW, Turner BL, Angel R, Applegate TJ: Influence of phytase addition to poultry diets on phosphorus forms and solubility in litters and amended soils. J Environ Qual 2004, 33:2306-2316.
36. Lucci GM, McDowell RW, Condron LM: Phosphorus and sediment loads/transfer from potential source areas in New Zealand dairy pasture. Soil Use Manage 2010, 26:44-52.
21. Sharpley AN, Kleinman PJA, Heathwaite AL, Gburek WJ, Folmar GJ, Schmidt JP: Phosphorus loss from an agricultural watershed as a function of storm size. J Environ Qual 2008, 37:362-368.
37. James E, Kleinman P, Stedman R, Sharpley A: Phosphorus contributions from pastured dairy cattle to streams of the Cannonsville Watershed, New York. J Soil Water Conserv 2007, 62:40-47.
22. Kleinman PJA, Srinivasan MS, Dell CJ, Schmidt JP, Sharpley AN, Bryant RB: Role of rainfall intensity and hydrology in nutrient transport via surface runoff. J Environ Qual 2006, 35:1248-1259.
38. McDowell RW, Wilcock RJ: Sources of sediment and phosphorus in stream flow of a highly productive dairy farmed catchment. J Environ Qual 2007, 36:540-548.
23. Ward RC: On the response to precipitation of headwater streams in humid areas. J Hydrol 1984, 74:171-189. 24. Srinivasan MS, McDowell RW: Identifying critical source areas for water quality: 1. Mapping and validating transport areas in three headwater catchments in Otago, New Zealand. J Hydrol 2009, 379:54-67. 25. McDowell RW, Srinivasan MS: Identifying critical source areas for water quality: 2. Validating the approach for phosphorus and sediment losses in grazed headwater catchments. J Hydrol 2009, 379:68-80. 26. van Beek CL, van der Salm C, Plette ACC, van de Weerd H: Nutrient loss pathways from grazed grasslands and the effects of decreasing inputs: experimental results for three soil types. Nutr Cycl Agroecosyst 2009, 83:99-110. 27. McDowell RW, Kleinman PJA: Efficiency of phosphorus cycling in different grassland systems. In Grassland Productivity and Ecosystem Services. Edited by Lemaire G, Hodgson J, Chabbi A. Oxon, UK: CAB International; 2011:108-119. 28. Kellogg RL, Lander CH, Moffitt DC, Gollehon N: Manure nutrients relative to the capacity of cropland and pastureland to assimilate nutrients: spatial and temporal trends for the United States. USDA-NRCS, USDA-ERS, nps00-0579, GSA National Forms and Publication Center, Fort Worth, TX; 2000, available online at http://www.nrcs.usda.gov/technical/NRI/pubs/ manntr.html. 29. Withers PJA, Jarvis SC: Mitigation options for diffuse phosphorus loss to water. Soil Use Manage 1998, 14:186-192. 30. Maxted JR, McCready CH, Scarsbrook MR: Effects of small ponds on stream water quality and macroinvertebrate communities. N Z J Mar Freshwater Res 2005, 39:1069-1084.
www.sciencedirect.com
39. Houlbrooke DJ, Horne DJ, Hedley MJ, Hanly JA, Scotter DR, Snow VO: Minimising surface water pollution resulting from farm-dairy effluent application to mole-pipe drained soils. I. An evaluation of the deferred irrigation system for sustainable land treatment in the Manawatu. N Z J Agric Res 2004, 47:405-415. 40. Pionke HB, Gburek WJ, Sharpley AN: Critical source area controls on water quality in an agricultural watershed located in the Chesapeake Basin. Ecol Eng 2000, 14:325-335. 41. McDowell RW, Nash D: A review of the cost-effectiveness and suitability of mitigation strategies to prevent phosphorus loss from dairy farms in New Zealand and Australia. J Environ Qual 2012, 41:in press. Highlights two hypotheses: (1) that without subsidies the best chance of decreasing P loss is to provide a wide range of fully costed mitigation strategies so that the end-user can mix and match the best strategy to suit their farm, and (2) that the best mitigation is only achieved when strategies are used where and when they are needed viz. critical source areas vary in space and time. 42. Kleinman PJA, Sharpley AN, McDowell RW, Flaten DN, Buda AR, Tao L, Bergstrom L, Zhu Q: Managing agricultural phosphorus for water quality protection: principles for progress. Plant Soil 2011, 349:169-182. Paper that outlines those future issues facing the management of P for good water quality. A highlight is the discussion of the consequence of a legacy of high P losses from high P soils despite our best efforts (e.g. no further P additions) to mitigate them. 43. McDowell RW, Snelder T, Littlejohn R, Hickey M, Cox N, Booker DJ: State and potential management to improve water quality in an agricultural catchment relative to a natural baseline. Agric Ecosyst Environ 2011, 144:188-200. 44. McDowell RW, Condron LM: Phosphorus and the Winchmore trials: review and lessons learnt. N Zeal J Agric Res 2012, 55:In press.
Current Opinion in Biotechnology 2012, 23:860–865