Abstract

Key Words

Olsen phosphorus, soil nutrients, piezometers, groundwater, nitrogen, potassium.

Introduction

Agricultural ecosystems are noted for increased nutrient flow due to the transfer of nutrients between and within enterprises as well as the environment (van Noordwijk 1999). Accumulation or depletion of nutrients can occur within a production system depending on whether there is net import or export of nutrients. Nutrient budgets for dairy farmlets indicated little changes in extractable phosphorus (P), even at annual fertiliser rates of up to 140kg P/ha. (Gourley 2001), unlike dairy farms in New Zealand and Europe, where large increases in soil extractable P were observed (Goh and Williams 1999). On the other hand accumulations of nitrogen (N) and potassium (K) have been observed on dairy farms with imported hay contributing to the net accumulation of K observed on farms (Goh and Williams 1999, Gourley 2001).

Nutrient redistribution is often observed between and within paddocks on dairy farms. Dairy cows play an important role in this redistribution of nutrients as they ‘recycle’ in dung and urine depositions over 60% of the N, P and K consumed (Haynes and Williams 1993). Phosphorus and K in dung pads is returned at significantly greater rates (248 and 728 kg/ha, respectively) than average fertiliser application rates (Gourley 2001, Aarons et al. 2004). Practices such as the use of night/day paddocks, where cows graze paddocks close to the dairy between evening and morning milking, result in nutrient accumulation in these areas. Stock camps are also high nutrient zones on farms (Gerrish et al. 1993, Gerrish et al. 1995), while nutrient redistribution to the front of strip-grazed paddocks (where these were not back-fenced) has been observed (Aarons 2001).

The accumulation of nutrients on-farm, as well as within the farm, can have a negative impact on the environment if these areas are located along nutrient loss pathways, as P loss in run-off can be related to soil P levels (Carpenter et al. 1998). While these losses are usually minor in production terms, they can have significant impacts on water quality and biodiversity (Rutherfurd et al. 2000).

The Gippsland Dairy Riparian Project – Environmental Monitoring module is located on commercial dairy farms in West Gippsland, where it aims to monitor farm management impacts on the farm and environment, including stream water quality. Soil and water data collected at the research site is presented and the implications for the environment are discussed.

Methods

Site location and description.

The research site is located along an approximately 1.7 km reach at the base of the Sandy Creek catchment, on 2 adjacent commercial dairy farms in West Gippsland (Figure 1). The monitoring data reported in this paper was collected on the upstream dairy property between April 2003 and March 2004. The farmer and sharefarmers milk approximately 180 Friesian cows on 65 ha, with the cows producing between 5000 and 6000L, with 220 kg butterfat and 200 kg protein. The pastures are predominantly ryegrass (Lolium perenne L.) and white clover (Trifolium repens L.). The nutrient management consists of applications of 2 in 1 superphosphate and potash applied in autumn and spring, with N applied strategically as urea when needed.

Figure 1a. Map of research site (green) located at the base of the catchment (red).	Figure 1b. Map of the Sandy creek through the 2 commercial dairy farms

The creek at the research site is a 3^rd order creek with the headwaters approximately 5.5km upstream.

The site comprises ~4% of the catchment area with the upstream landuse predominantly dairy. The area receives ~1100mm of rainfall annually, with most of the rain occurring in winter and early spring. The topography consists of gently rolling hills. The soils of the catchment and the research site appear to have developed on Devonian sediments, Neogene (Quaternary) sediments, or Mid-Tertiary volcanics. Each of the eight soil types identified at this site (Sargeant pers. comm.) are predicted to contribute to either surface or subsurface flows at different locations in this landscape when the soil has reached field capacity.

Piezometer construction.

Four pairs of nested piezometers were installed in the pastures between the dairy and creek on the upstream property to measure deep and shallow groundwater, to confirm nutrient distribution at the site (Figure 2). The shallow piezometers were augured to a depth of 1m, while the deep piezometers were constructed to the watertable or an impermeable surface, which ever came first. A 100mm diameter extendable hand auger was used to dig the holes for each piezometer. The piezometers were made from slotted PVC pipes that were covered with mesh at the base, then inserted into the augured holes. The space around the tubes was backfilled and capped with bentonite to prevent preferential flow pathways developing around the outsides of the PVC tubes. Piezometer tubes extend above the soil surface by approximately 100cm to allow discharge water levels heights to be measured and to ensure no water sitting on the soil surface enters the piezometers. They were end-capped to prevent rain filling the tubes.

Soil sampling and analyses.

The area between the dairy and the creek was intensively soil sampled to characterise nutrient distribution in the landscape. This area was sub-divided into 33 sampling areas based on slope, drainage and farm management characteristics (Figure 2), and in each area a minimum of 15 soil cores (2.5 cm diameter x 10 cm) was collected. Each core was sectioned to give a 0-5cm and 5-10 cm sample, and the samples from each area were bulked according to depth. The bulked samples were dried at 40^oC for 72 hours then ground and sieved to less than 2mm prior to chemical analysis.

Soil removed during installation of the piezometers was collected from the 0-5cm, 5-20cm and deeper depths for chemical analysis. The soils were dried, sieved and ground as described above. All soil samples were analysed for pH (in water and CaCl₂), electrical conductivity (EC), Olsen (Olsen et al. 1954) P and Colwell (Colwell 1963) K as given by Rayment and Higginson (1992).

Figure 2. Soil sampling areas located between the dairy and the creek. Area 1 is the stand-off area where cows congregate to and from the dairy. The effluent ponds (dams) are located in the bottom left hand corner. Blue circles and numbers in large type represent the location and numbers of the piezometers.

Groundwater sampling and analysis.

The depth to groundwater was measured weekly and water samples collected for pH, EC, nitrate (NO₃), total nitrogen (TN), TP, and total potassium (K) concentrations. Analyses were done according to standard water analytical procedures (see APHA 1998). The EC, NO₃, TN and K results are reported in this paper.

Results and discussion

Soil P levels in the 0-5 cm layer were very high over the site in general with paddock levels of greater than 60 mg/kg P, increasing to more than 200 mg/kg in and adjacent to the stand off area (Figure 3). Olsen P, Colwell K and EC levels were highest at sites 1, 2, 11 and 12, with accumulations at sites 19, 22, 25, 26 and 28 (see Figure 2 for site number locations, Figures 3 to 5 for soil analyses). Colwell K is a good indicator of the presence of dung and urine as previous research has shown a 100% increase in soil Colwell K under dung pads (Aarons et al. 2004) , with the increase sustained long after the pads had completely disintegrated. The patterns recorded in the 0-5cm layer were also observed in the 5-10cm layer, indicating a movement of these nutrients down through the profile. Poorly managed effluent ponds can result in contamination of the environment especially when ponds overflow frequently. Sites 23 and 24 do not appear to function as routes for frequent loss of effluent from the ponds to the environment, as the soil nutrient levels in these areas were not elevated compared with expected levels for fertilised grazed pastures.

The Olsen P and Colwell K levels of the soil removed from various depths when the piezometers were installed confirm the accumulation of nutrients near the surface at this site (Figure 6). Except for piezometer 7, soil P and K were elevated well above the levels recommended for grazed dairy pasture soils (Gourley 2001), with adequate nutrient levels observed to at least 40cm depth. Olsen P levels to 20 cm depth were greater than 41 mg/kg at piezometers 4 and 6, and were 39 mg/kg in the 20 - 40 cm depth sample at piezometer 5. Potassium levels were greater than 340 mg/kg to 50 cm deep. At the piezometer 7 site Olsen P levels were 4 mg/kg and Colwell K levels 190 mg/kg in the 5 to 20 cm soil layer, and were 34 and 850 mg/kg respectively in the 0 to 5 cm layer.

Figure 3. Soil Olsen P (mg/kg) in the 0-5 and 5-10cm layer in each of the sample areas. Those areas with the highest (H) P levels are indicated.

Figure 4. Soil Colwell K (mg/kg) in the 0-5 and 5-10cm soil layers in each of the sample areas. Those areas with the highest (H) K levels are indicated.

Figure 5. Soil EC (dS/m) in the 0-5 and 5-10cm layer in each of the sample areas. Those areas with the highest (H) EC levels are indicated.

Nutrient accumulation areas near the dairy appear to be due to high animal traffic, with accumulation areas down slope potentially the result of the movement of nutrients from the stand-off area. Historical fertiliser management was uniform across the area below the dairy, and therefore could not explain the uneven distribution observed. High extractable soil P levels can be associated with P losses to the environment (Carpenter et al. 1998). Thus the potential exists for there to be significant losses of P from the stand-off and low-lying areas into the Sandy Creek, when surface runoff or shallow subsurface flow occurs, based on the Olsen P results.

Figure 6. Soil Olsen P (a) and Colwell K (b) to depth at the site of piezometers 4, 5, 6 and 7.

Figure 7 shows that a perched water table occurred at piezometer 4 until November 2003, with the deep groundwater rising to within 30 cm of the soil surface. At piezometer 5 discharge conditions occurred for most of the monitoring period, while the deep groundwater was at or near the soil surface during late winter / early spring at piezometers 5 and 6. No water was measured in either the shallow or deep piezometers at location 7 indicating that this piezometer did not intercept any groundwater (Figure 7). Total K and EC in the groundwater are consistent with the elevated nutrient levels measured by the intensive soil sampling. Thus, EC was elevated in the shallow ground water particularly at piezometer 5 and to some extent at piezometer 6, both of which are located in depressions below the dairy and stand off area. Total K in shallow groundwater was elevated at piezometers 4, 5 and 6, all of which are located downslope of the dairy and stand off area. Potassium is mobile in soil water so movement of K from upslope areas could have contributed to the accumulation of K in the depressions where piezometers 5 and 6 are located. While piezometer 4 is located at a slight elevation compared to piezometers 5 and 6, the shallow perched water table K levels were greater than the shallow groundwater levels for 5 and 6. This was in contrast to the greater K levels in soils from the piezometer 5 site (see figure 6). The higher K levels in the perched water table at piezometer 4 could be due to the frequent perching of water restricting K movement further down the profile. K levels in piezometer 6 fluctuated in the shallow groundwater, which may be due to rainfall and/or shallow subsurface lateral flow diluting the shallow groundwater (Figure 8) during the wetter months of the year. The TN and NO₃ levels in shallow and deep water tables reflect the movement of N through the profile at piezometer 4. As the height of the water table decreased TN and NO₃ concentrations increased in piezometer 6, unlike piezometer 5 where the observed discharge may have restricted this decrease (Figure 9).

Stock management is a major contributor to nutrient accumulation areas on dairy farms. At this site a stand-off area is located outside the dairy where the 2 tracks to the dairy meet. Thus the herd of 180 cows congregate in this area on their way into and out of the dairy shed 4 times/day for the duration of the lactation (~280 days/year). The soil sampling data demonstrate the significant nutrient accumulation in the stand-off area. The location of this area upslope of the riparian zone and creek suggests a potential for the transfer of nutrients via runoff and erosion, especially down the tracks, to the lowlying areas close to the creek and into the waterway as well. Nutrient levels are elevated along the track towards the creek as well as in the lowlying areas, most likely due to movement of nutrients along overland flow pathways. The absence of deep or shallow groundwater in piezometer 7, as well as the soil type at this location indicates a greater likelihood of overland flow rather than infiltration. In addition the lower extractable soil P and K levels in the surface (0 to 5 cm) soil as well as to depth, indicate that there has been very little movement of nutrients down the soil profile. Overland flow around piezometer 7 has been observed, as well as movement of water from near piezometer 6 into the creek. These observations, in conjunction with the elevated soil and groundwater nutrient levels indicate the potential for the transfer of degraded water from this farm to the environment. More detailed hydrological observations will help identify the degree of contribution of surface and subsurface flow to creek flow and quality at the site.

Figure 7. Shallow and deep ground water recorded in piezometers at the site. See Figure 2 for locations.

Figure 8. Total K (mg/L) and EC (µS/cm) levels in the shallow (red triangles) and deep (blue circles) groundwater.

Figure 9. Total N (mg/L) and nitrate (mg/L) levels in shallow (red triangles) and deep (blue circles) groundwater.

Conclusion

The intensive soil sampling to depth and the piezometer data demonstrated elevated nutrient levels within this landscape. Stock management (high stock densities and frequency of use) near the dairy is postulated to contribute to the accumulation of nutrients observed at the stand-off area, with the potential for transport from this area to downslope depressions. The distribution of nutrients at this site and the predicted hydrological pathways at this location indicate the potential for stock management around the dairy to contribute to degraded water quality in the Sandy Creek.

Acknowledgments

The authors thank the research site farmers (Peter and Helen Snape, Charlie and Jan Pinch) and sharefarmers (Gary Pocklington and Kate Vansippart) for accommodating the research activities on their property. This research is supported by the Department of Primary Industries, Dairy Australia, GippsDairy, Land and Water Australia, and the West Gippsland CMA, with additional collaboration with The University of Melbourne, The University of Western Australia and Waterwatch.

References

Aarons SR (2001). Cycling of organic phosphorus in grazed dairy systems - Final Report to the Dairy Research and Development Corporation. Department of Natural Resources and Environment, Agriculture Victoria, Ellinbank, Victoria.

Aarons SR, O’Connor CR, Gourley CJP (2004). Dung decomposition in temperate dairy pastures: I Changes in soil chemical properties Australian Journal of Soil Research. 42, 107-114.

APHA. (1998). Standard Methods for Examination of Water and Wastewater. 20^th edition. In L. S> Clesceri, A. E. Greenberg, & A. D. Eaton (Eds). Baltimore: American Public Health Association – American Water Works Association – Water Environmental Federation.

Carpenter SR, Caraco NF, Correll DL, Howarth RW, Sharpley AN, Smith VH (1998). Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological Applications 8, 559-568.

Colwell JD (1963). The estimation of the phosphorus fertiliser requirements of wheat in southern New South Wales by soil analysis. Australian Journal of Experimental Agriculture and Animal Husbandry. 3,190-198.

Gerrish JR, Brown JR, Peterson PR (1993). Impact of grazing cattle on distribution of soil minerals. In ‘Proceedings of the American Forage and Grassland Council’. pp. 66-70. Des Moines, IA, USA.

Gerrish JR, Peterson PR, Brown JR (1995). Grazing management affects soil phosphorus and potassium levels. In ‘Proceedings of the American Forage and Grassland Council’. pp. 175-179. Lexington, KY, USA.

Goh KM, Williams P (1999). Comparative nutrient budgets of temperate grazed pastoral systems. In ‘Nutrient disequilibria in agroecosystems: concepts and case studies’. (Eds E Smaling, O Oenema, L Fresco) CABI Publishing, Cambridge, UK.

Gourley CJP (2001). Phosphorus for dairy farms 1995-2001. Final report to the Dairy Research and Development Corporation. Department of Natural Resources and Environment, Ellinbank, Victoria.

Gourley C, Ridley A (2004). Controlling non-point pollution in Australian agricultural systems. Pedosphere (In press).

Haygarth PM, Sharpley AN (2000). Terminology for phosphorus transfer. Journal of Environmental Quality 29, 10-15.

Haynes R J, Williams PH (1993). Nutrient cycling and soil fertility in the grazed pasture ecosystem. Advances in Agronomy 49,119-199.

Olsen SR, Cole CV, Watanabe FS, Dean LA (1954). Estimation of available phosphorus in soils by extraction with sodium bicarbonate. USDA Circular No. 939.

Rayment GE, Higginson FR (1992). Australian Laboratory Handbook of soil and water chemical methods. Australian soil and land survey handbooks series, v. 3. Inkata Press

Rutherfurd ID, Jerie K, Marsh N (2000). A rehabilitation manual for Australian streams: Volume 2. Land and Water Resources Research and Development Corporation (LWRRDC) and the Cooperative Research Centre for Catchment Hydrology.

van Noordwijk M (1999) Nutrient cycling in ecosystems versus nutrient budgets of agricultural systems. In 'Nutrient disequilibria in agroecosystems : concepts and case studies.' (E Smaling, O Oenema,L Fresco). 1-26 (CABI Publishing: Cambridge, UK).