Abstract

At the first grazing, treatments E and F resulted in higher (P<0.05) crude protein (CP) content compared to treatments A and B. The metabolisable energy (ME) content of treatment A was higher (P<0.05) than D, E and F. By the second grazing, the ME content of A, B and C was higher than E and F. The concentration of phosphorus, sulphur and magnesium at the first grazing was higher (P<0.05) with treatments D, E and F compared to A. This study indicates the potential to use dairy effluent to increase millet DM yield and improve it’s nutritive value in dryland areas of southern Victoria.

Media summary

Key Words

Crude protein, metabolisable energy, nitrogen, potassium, sodium

Introduction

Dairy effluent is recognised as a significant point source in the pollution of waterways. It contains organic matter, microbial contaminants, nutrients and suspended solids that can all impinge upon water health. In Victoria, the State Environment Protection Policy (SEPP) states that within 5 years (2005) all dairy waste will be retained within the boundary of the property. To achieve this, dairy farmers will require clear guidelines on appropriate and sustainable systems to effectively use dairy effluent.

It is estimated that only 50% of dairy farms in the dryland regions of Victoria have suitable dairy effluent systems and of these only 25% are managed effectively (IRIS Research 2000). Furthermore, most farmers apply effluent to less than 10% of their available land, often to the same area each year. Despite being seen by many farmers as an undesirable waste, dairy effluent contains relatively large amounts of agronomically useful nutrients especially nitrogen (N) and potassium (K). Studies in New Zealand (Goold 1980, Roach et al. 2000) indicate substantial DM yield increases are possible when effluent is applied to perennial pasture throughout the year. It should however be noted that the climatic conditions under which these trials were conducted are different to those in southern Victoria where effluent application in winter is likely to lead to nutrient run off or leaching. Given this limitation, there is potential to investigate the use of effluent on summer active forages. Previous work (Jacobs and Ward 2003) highlighted DM yield increase when effluent was applied to turnips.

This paper reports results generated from the first year of a three year study that compares millet DM responses and changes in nutritive characteristics and mineral content for a range of dairy effluent application rates.

Methods

This study was conducted on a commercial dairy farm near Terang (38°14’S, 142°55’E) in western Victoria on a basalt derived fine sandy clay loam soil. A paddock was removed from the grazing rotation on 1 September 2002 and N was applied (50 kg N/ha) one week later. On 16 October, pasture was cut for silage and bales removed from the paddock within three days. Immediately following removal of silage bales, the area was grazed to remove remaining residual pasture and then mouldboard ploughed. Eight days later the area was power harrowed and on 11 November sown with millet (Echinochloa utilis cv. Shirohie) at a rate of 20 kg/ha with 200 kg/ha single superphosphate (17.6 kg P, 22 kg S).

From 27 December to 4 January, six levels of second pond dairy effluent, 0 (A), 15 (B), 30 (C), 45 (D), 60 (E) and 75 (F) mm/ha were applied to plots at the site. Treatments were randomly allocated to plots (12 m x 12 m) within each block, and replicated six times in a randomised block design. Effluent was applied via a pressurised spray system (Irrifrance, Bosch Engineering) with sprinklers located on a 12 m x 12 m grid system (corner of each plot) with each sprinkler covering a 90^o arc ensuring a uniform distribution. For each 12 m x 12 m plot a buffer zone of 2 m was established to minimise impacts from adjacent plots. Application rates did not exceed 15mm/ha/d to avoid run off of effluent. Irrigation was only undertaken when wind conditions were such that drift did not occur.

Dry matter yield measurements were taken 11 weeks after sowing and again on regrowth seven weeks later. At each harvest, 6 quadrats (1.0 m²) were collected per plot, weighed individually and subsampled on a plot basis. Samples were collected approximately three hours after sunrise, stored in airtight bags and packed with ice in insulated containers. The samples were further divided with one portion being used to determine DM yield by drying at 100°C for 24 h. The remaining sample was dried at 60°C for 72 h, ground through a 1mm screen (Tecator Cyclotec 1093 sample mill) and used to determine nutritive characteristics and mineral content.

Analysis of samples for nutritive characteristics was undertaken at FEEDTEST, Pastoral and Veterinary Institute, Hamilton using near infrared spectroscopy. Metabolisable energy (ME) (MJ/kg DM) values were calculated from predicted DM digestibility values (SCA 1990). Mineral analysis of leaf and root was by a microwave digestion (Lautenenschlaeger 1989, Nackashima et al 1988)) followed by Inductively Coupled Plasma - Optical Emission Spectroscopy (SCL 1987). Statistical analysis was undertaken using analysis of variance (ANOVA) (GenStat Committee 2000) with significance declared if P<0.05.

Results

Prior to effluent application soil test results (0-10 cm) were: pH _(H2O) 5.3, Olsen P 41.5 mg/kg and Skene K 230 mg/kg. Effluent composition (Table 1) indicates a high K and sodium (Na) content.

Table 1. pH, electrical conductivity (EC) (dS/m), sodium adsorption ratio (SAR), phosphorus (P), potassium (K), sulphur (S), nitrogen (N), calcium (Ca), magnesium (Mg), sodium (Na) (mg/L) of effluent

Treatments C, D, E and F resulted in an increase (P<0.05) in DM yield at the first grazing compared to the control treatment (A) (Table 2). Treatments D, E and F increased (P<0.05) DM yield compared to the lowest level of effluent application (B). This trend continued through to the second grazing where the DM yield of A was lower (P<0.05) than treatments D, E and F. Total DM yields over the two grazings showed that, apart from B, all other levels led to an increase (P<0.05) in DM yield compared to the A. Furthermore, treatments D, E and F resulted in higher (P<0.05) total DM yields than B and C.

Treatments E and F resulted in higher (P<0.05) crude protein (CP) content compared to both the control (A) and the lowest rate of effluent application (B) (Table 3). Treatment F also gave rise to a higher (P<0.05) CP content than either C or D. The ME content of treatment A was higher (P<0.05) than D, E and F. By the second grazing, there was no difference in CP content, whilst the neutral detergent fibre (NDF) content of A was lower than for D, E and F. The ME content of A, B and C was higher than E and F at the second grazing.

Table 2. The effect of different effluent application rates (A 0; B 15; C 30; D 45; E 60; F 75 mm/ha) on millet dry matter yield (t DM/ha) over subsequent grazing periods and growth rates from sowing to Grazing 1 (S-G1) and grazing 1 to Grazing 2 (G1-G2) (kg DM/ha/d)

Table 3. The effect of different effluent application rates (A 0; B 15; C 30; D 45; E 60; F 75 mm/ha) on millet metabolisable energy (ME) (MJ/kg DM), crude protein (CP), neutral detergent fibre (NDF), and water soluble carbohydrate (WSC) (%DM) content

At the first grazing treatments D, E and F had higher (P<0.05) P, S and Mg content than A (Table 4). Potassium content of A was lower (P<0.05) than treatments C, D and E, whilst A also had a lower (P<0.05) Na content than all other treatments. By the second grazing there was no effect of effluent application on mineral content apart from the S content of A being (P<0.05) higher than treatments C, D, E and F.

Discussion and Conclusions

The concentration of nutrients within the effluent used for this study falls well within the ranges quoted by Kane (pers comm.) from a study of 158 effluent ponds in south west Victoria. Furthermore, effluent composition was similar to that of Jacobs and Ward (2003) who found P, K and N levels of 35, 427 and 122 kg/ML respectively.

The DM yield responses of millet to applied effluent ranged from a 13 to 38% compared to the control treatment. The responses at the lower rates (15-30 mm/ha) are similar to those observed for turnips (Jacobs and Ward 2003) in an earlier study. Comparative studies to this work (Jacobs et al. 2004) where effluent was applied to perennial pasture showed responses ranging from 19 to 44%. Whilst there are no comparative data for millet or other C4 species, these results would indicate that similar responses are likely irrespective of species under comparative conditions. Initial data from year 2 of this study (unpublished) indicates similar millet DM yield responses to effluent. One of the challenges with multi nutrient solutions such as dairy effluent is determining the key factors responsible for such DM responses. Given the high soil levels of P, K and S at the site, it is postulated that the DM responses were largely a result of N and water.

To the authors knowledge there are no other studies that have been conducted to measure the effect of effluent application on the nutritive characteristics of millet. The data collected in this study do show positive CP responses to applied effluent, however there was also a negative effect on the ME content. This drop in ME is likely to be a reflection of increased DM yield and an additional rise in the NDF content.

Table 4. The effect of different effluent application rates (A 0; B 15; C 30; D 45; E 60; F 75 mm/ha) on millet phosphorus (P), potassium (K), sulphur (S), calcium (Ca), magnesium (Mg) and sodium (Na) (%DM) content

In conclusion, dairy effluent has the potential to increase DM yields of millet during summer, a period when feed is often limiting on dryland farms in southern Victoria. This study will continue for a further two years and assist in determining long term sustainable practices for the use of effluent in terms of achieving a balance between production and environmental implications.

References

IRIS Research (2000). A survey of Natural Resource Management on Australian Dairy Farms. Technical report.

Genstat 5 Committee (1997) 'Genstat 5.41 Reference Manual'. Oxford Science Publications, Oxford, UK.

Goold GJ (1980). Rates of farm dairy effluent applied to pastures on clay soils in Northland. New Zealand Journal of Experimental Agriculture 8, 93-99.

Jacobs JL and Ward GN (2003). Effect of different rates of dairy effluent on turnip DM yields and nutritive characteristics. Proceedings of the 11^th Australian Agronomy Conference, Geelong, Victoria.

Jacobs JL Ward GN and McKenzie FR (2004) Effect of dairy effluent on DM yields and nutritive characteristics of perennial pasture in late spring and summer. Proceedings of the 11^th Australian Agronomy Conference, Brisbane, Queensland.

Lautenenschlaeger W (1989). Atomic Spectroscopy Advances. Microwave Digestion in a Closed-Vessel, High Pressure System. Spectroscopy 4(9), 16-21.

Nackashima S, Sturgeon E, Willie SN and Berman SS (1988). Acid Digestion of Marine Samples for Trace Element Analysis Using Microwave Heating. Analyst. 113 159-163.

Roach CG, Stevens G, Clark DA and Nicholas P. (2000). Effects of effluent and urea application on groundwater, soil and pasture at WTARS. Proceedings of the New Zealand Grassland Association. 62, 173-178.

SCA (1990) Feeding standards for Australian Livestock. Ruminants. CSIRO Publications, Melbourne, Australia.

State Chemistry Laboratory (1987). ‘Method 013. Determination of calcium, magnesium, sodium, potassium, sulphur, phosphorus, iron, copper, zinc and manganese in plant material by inductively coupled plasma- Optical emission spectroscopy’. Department of Natural resources and Environment, Werribee, Vic.