|
   
The magnitude of temporal variation in soil P–runoff P relationships 1Organic Waste Recycling Unit, NSW Department of Primary Industries, Locked Bag 4, Richmond, NSW 2753, Australia.
2Current address: School of Earth and Environmental Sciences, University of Adelaide, PMB 1, Glen Osmond, SA 5064, Australia. Email: warwick.dougherty@adelaide.edu.au Abstract
The existence and nature of the relationship between soil phosphorus (P) and runoff P is fundamental to models that predict runoff P concentrations, risk assessment frameworks and farm based decisions relating to environmental impact. Knowledge of the relationship can aid in prioritising areas for remedial action to mitigate runoff P losses. However, a number of researchers have observed that there is significant variation in the nature of these relationships with time and soil type. Such variation may dramatically reduce the accuracy of predictive models. Furthermore, the reliability of risk assessment frameworks that include soil P is reduced. This paper reports preliminary research, the aims of which are to a) assess a range of soil P tests as predictors of runoff P and b), quantify the variation in the soil P–runoff P relationship with time. Soil P-runoff P relationships were determined using small rainfall simulation plots (1.5 m2) and large scale runoff plots (1250 m2) on a number of occasions under various management conditions. Preliminary data suggest that the method of soil analysis is an important factor in accurately determining a reliable soil P–runoff P relationship. The quantification of variation in the soil P–runoff P relationship and elucidation of mechanisms of P mobilisation will help provide more accurate methods for identifying areas of farms prone to runoff P losses. Further research to identify those factors influencing the variation in the soil P-runoff P relationship is required. Introduction
The dairy industry is investigating those factors determining the concentration of P in runoff from dairy pastures. They are developing procedures to allow land managers to identify parts of the landscape that are at particular risk to contributing to farm losses of runoff P (Fleming et al. 2003; Melland et al. 2004). The influence of soil P on runoff P is a key component of these procedures. Various authors have demonstrated a significant relationship between soil test P and runoff P concentration (Pote et al. 1996; Torbert et al. 2002). However, the investigations of the relationship between soil P and runoff P concentration have generally been undertaken at single points in time only and there has been little quantification of the variability of these relationships as a result of a range of factors related to time. Understanding this variability is important in accurately assessing the risk that particular paddocks or farms pose in terms of P export in runoff.
This paper presents preliminary results from research that examines a) the ability of various soil tests to predict runoff P concentrations at different times, and b) the variation in these relationships under a wide range of conditions under grazed pastures. Both small and large plot rainfall simulation are used in this research.
Methods
The study was located at the Elizabeth Macarthur Agricultural Institute, Camden, 60 km south west of Sydney. The soils at the site are Brown Chromosols (Isbell 1997) or Red and Yellow Podzolics (Stace et al. 1968). Topsoil texture is clay loam with a pHCa of approximately 5.5.
Small plot rainfall simulation
Plots were established in a mid-slope position using a split plot design with 3 replicates. Split plots were utilised for rainfall simulation at 3 and 12 months after fertiliser application. The soil P status was modified by the addition of fertiliser P at rates of 0, 70, 140, 210 kg/ha. Rainfall simulation was undertaken using a swinging boom rainfall simulator applying rainfall at 80 mm/hr for 40 mins. This is typical of intensities and durations of rainfall used in rainfall simulation methodology (Pote et al. 1999; Torbert et al. 2002) and is approximately equivalent to a 1 in 10 year storm for this site. Soil was sampled by taking 25 individual cores (0-2 cm depth) and compositing these for each plot. Soil samples were then dried, ground and sieved (<2 mm) and stored at 4oC before analysis. Soil samples were then analysed for a range of soil P measures including Olsen P (1954), water soluble P (WSPCa) and equilibrium phosphorus concentration (EPC). WSPCa was determined using the method of Pote et al. (1996). WSPCa was determined by shaking 8 g of soil with 40 mL of 0.01M CaCl2 on an end–over–end shaker for 30 min. The solution was then centrifuged and filtered through a 0.45 μm filter before analysis for molybdate reactive P (Murphy and Riley 1962). Equilibrium phosphorus concentration was determined using the method of Rayment and Higginson (1992) by equilibrating 4 grams of soil in 40mL 0.01M CaCl2 solution containing known concentrations of P (0, 0.5, 1, 2 and 5 mg/l) for 17 hours and then centrifuging samples prior to determining P concentration in the supernatant. Phosphorus sorbed by the soil samples was determined using the change in soil solution P concentration and the ratio of soil:solution. Phosphorus sorbed (mg/l) was plotted against P concentration (mg/l) and the point on a linear regression fit to the data points where no sorption or desorption was occurring was defined as the equilibrium phosphorus concentration (EPC).
Large plot rainfall simulation
Large-scale rainfall simulation was also undertaken on 6 plots (50 × 25 m) using bike-shift sprinkler irrigation to apply simulated rainfall for 10 hours at 8 mm/hr. These plots were located in the paddock adjacent to that used for the rainfall simulations. The forms and concentrations of P in runoff from these large-scale simulations were previously compared with those in runoff occurring as a result of natural rainfall on two occasions when the two event types occurred within several days of each other. There was no significant difference (P<0.05) in runoff P concentrations and forms of P between event types. Therefore, artificially induced runoff events were used in this experiment to represent ‘real’ runoff events on these large plots.
Soils were sampled by taking 30 cores (0-2 cm) along a permanent sampling transect in each plot. Soil samples were analysed for WSPCa using the method described above. Runoff flow and volumes were measured using Greenspan depth sensors and logger using a DataTaker data-logger. Runoff samples were collected automatically using a peristaltic water sampler triggered at pre-determined runoff volumes by the data-logger.
Water sample handling and analysis
All runoff samples were filtered (< 0.45 μm) immediately upon collection, chilled to 4oC, and subsequently analysed for dissolved reactive P (DRP) within 24 hours of collection. Phosphorus was determined using colorimetric analysis (Murphy and Riley 1962). All runoff P concentrations are reported as flow-weighted averages.
The DRP concentration of the water used in all rainfall simulations was <0.05 mg/L. The concentration of phosphorus in all runoff samples was at least an order of magnitude higher than that used in the rainfall simulations; consequently no adjustment was made to the runoff concentrations.
Statistical methods
A preliminary examination of the relationships between soil P and runoff P concentrations were undertaken using linear regression. Data were checked for normal distribution of variances and if required log transformed. In all cases, data from runoff events occurring less than 40 days after fertiliser application were excluded from analysis as fertiliser is known to directly influence runoff P concentrations immediately following its application (Nash et al. 2000).
Results
The P status of the soil varied widely across the sites examined. The Olsen P ranged from 13-162 mg/kg (0-2 cm) for the small rainfall simulation plots and from 17-109 mg/kg for the large plots. For the large runoff plots, soil P was measured immediately prior to 9 separate runoff events.
Small plot rainfall simulation
The regressions between each of the soil tests and runoff P concentrations are shown in Figure 1. On the small scale simulation plots there was a highly significant (P<0.01) relationship between the three soil P tests tested and runoff P on each occasion under which simulation was undertaken. However, there were significant differences (P<0.05) in the slopes of the regressions at T3 and T12 for Olsen P. However, there were no significant differences (P<0.05) in the slopes of the regressions between T3 and T12 for the WSPCa and EPC regressions.
Figure1. Regressions between soil P and rainfall simulation runoff DRP concentrations for three different soil P tests at two different times (3 and 12 months) after fertiliser application.
Large plot rainfall simulation
For DRP, the slopes of the relation with WSPCa ranged from 0.277 ± 0.029 (P <0.001) to 0.078 ± 0.0.026 (P <0.05). Regressions for the minimum and maximum slopes are shown in Figure 2. The differences among the events in both the intercepts and slopes of their relations with WSP were significant (P <0.05).
Figure 2. Regressions of WSPCa against DRP in runoff from large grazed plots showing the minimum and maximum regressions for individual events.
Discussion
Preliminary analysis of the data show that runoff phosphorus concentrations increase as soil test P increases. However, the relationship between soil test P and runoff P was observed to vary significantly with time, depending on the soil test used. In the comparison of soil test P to predict runoff P under simulated rainfall/runoff at different times, the differences in the relationship between Olsen P and runoff P were large. Olsen P provides a measure of the labile P pool using 0.5M NaHCO3 adjusted to a pH of 8.5. In contrast, the WSPCa and EPC methods provided much smaller differences in regressions at each of the two times. Both of these soil tests were undertaken in weak electrolyte solutions with no pH adjustment. The WSPCa has been proposed as a better indicator of potential P loss as it attempts to more closely mimic the extraction of P from soil in runoff by using a short extraction time, relatively close soil:solution ratio and uses a weak electrolyte as the extractant (Pote et al. 1996). However, EPC provided regressions that were very similar at both simulation times. Agronomic measures of soil P such as Olsen P are likely to provide a poorer indication of runoff P indication. This may be exacerbated when standard sampling depths such as 0-10 cm are used.
There were also significant differences in the slope and intercept of relationship between WSPCa and DRP for runoff events from the large plots at different times. Management conditions on these plots varied greatly between events. Various management factors were identified as having significant effects on runoff P concentrations (data not presented). These factors included DSF, biomass, days since grazing and interaction terms between P rate and application number. However, these effects were identified only in relation to P fertiliser rate and could not be tested with WSPCa data due to the low number of runoff events for which WSPCa and management data were available. If soil test P is to be used more reliably, the effect of these parameters in conjunction with various soil P tests is required.
Conclusions
Soil P provides a relative indicator of likely runoff P concentrations in runoff. However, the relationship between soil P and runoff P varies significantly with time. Preliminary data suggest that agronomic measures of soil P such as Olsen P are likely to provide a poorer indication of runoff P concentration. This may be exacerbated when standard agronomic sampling depths such as 0-10 cm are used. Further investigation of the relationship between soil and runoff P and the factors affecting this relationship are needed if soil P status is to be reliably used as part of tools used for guiding P management. These future investigations should include the effect of time since fertiliser application, pasture management, soil type, sampling depth and soil test type.
Acknowledgements
This research was funded by Dairy Australia and Incitec-Pivot. Biometrical advice and assistance by Paul Nicholls and Damian Collins is gratefully acknowledged.
References
Fleming N, Gepp M, Cox J, Davies P, Valentine S, Gregory N, Dougherty W (2003) 'Flaxley Farmlets - Nutrient loads in run-off water at high stocking densities. DRDC project DAS 10815. Milestone Report No. 3.' South Australian Research and Development Institute, Milestone Report No. 3, Adelaide.
Isbell RF (1997) 'The Australian Soil Classification.' (CSIRO Publishing, Melbourne, Victoria.)
Melland AR, Gourley CJP, Peverill KI, Strickland P, Awty I, Scott JM (2004) Making better fertiliser decisions, a project for better informing production and environmental decisions relating to pasture nutrient management. In 'Nutrient Management Forum'. Albany, Western Australia. (Centre of Excellence in Natural Resource Management and Department of Agriculture)
Murphy J, Riley JP (1962) A modified single solution method for the determination of phosphate in natural waters. Analytica Chimica Acta 27, 31-36.
Nash D, Hannah M, Halliwell D, Murdoch C (2000) Factors affecting phosphorus export from a pasture-based grazing system. Journal of Environmental Quality 29, 1160-1166.
Olsen SR, Cole CV, Watanabe FS, Dean LA (1954) 'Estimation of available phosphorus in soils by extraction with sodium bicarbonate extraction.' U.S. Department of Agriculture, Circular No. 939.
Pote DH, Daniel TC, Nichols DJ, Sharpley AN, Moore PA, Miller DM, Edwards DR (1999) Relationship between phosphorus levels in three Ultisols and phosphorus concentrations in runoff. Journal of Environmental Quality 28, 170-175.
Pote DH, Daniel TC, Sharpley AN, Moore PA, Edwards AC, Nichols DJ (1996) Relating extractable soil phosphorus to phosphorus losses in runoff. Soil Science Society of America Journal 60, 855-859.
Rayment GE, Higginson FR (1992) Phosphorus. In 'Australian Handbook of Soil and Water Chemical Methods' pp. 59-94. (Inkata Press: Melbourne)
Stace HCT, Hubble GD, Brewer R, Northcote KH, Sleeman JR, Mulcahy MJ, Hallsworth EG (1968) 'Handbook of Australian Soil.' (Rellim Publishers, Australia)
Torbert HA, Daniel TC, Lemunyon JL, Jones RM (2002) Relationship of soil test phosphorus and sampling depth to runoff phosphorus in calcareous and non-calcareous soils. Journal of Environmental Quality 31, 1380-1387. 
|
