Forages for Reduced Nitrate Leaching, Chris Nigel and Ross Rathgen (Canterbury)

Reducing stocking rate, imported supplement and establishing plantain as part of a mixed sward

This season some large farm system changes occurred. These included reducing the stocking rate by 0.5 cows/ha, reducing the total imported supplement from 918 to 551 kg DM/cow, and establishing plantain as part of a mixed sward. Similar rates of N fertiliser were used to earlier years (297 kg N/ha).

Plantain

• In late spring and early summer new pasture containing plantain was direct drilled. The Rathgens believe in spring-summer sowing for herbs, not going later than December-January.
• Diploid ryegrasses are used in the mix because they find they are more robust in their heavier soils.
• A typical seeding rate will total about 28-30kg/ha, made up of a mix of:
  • Plantain – 1 kg/ha
  • Chicory – 1 kg/ha
  • Red clover – 2 kg/ha
  • White clover – 4 kg/ha
  • Timothy – 1 kg/ha
  • Prairie grass – 5 kg/ha
  • Italian ryegrass – 4 kg/ha
  • Ryegrass – 10 kg/ha
• Chicory and plantain seeds are much smaller than ryegrass seeds so there are more seeds per kilogram and sowing rates aren’t indicative of the resulting proportion of each pasture species to the others in the sward.

Growing fodder beet on the platform for autumn feed, reducing imported supplement and continuing to establish plantain as part of a mixed sward

This season was largely about refining systems after the large change that occurred in the 2015-16 season. The main system change that was implemented was the inclusion of 6 ha (2% of milking platform) of fodder beet on the milking platform for use as autumn feed and to begin transitioning for winter fodder beet.

The amount of imported supplement was further reduced from 551 to 391 kg DM/cow. Pasture harvested increased slightly, replacements ran on the milking platform for parts of October, December, January and March to control pasture surpluses, and milk production remained steady.

Fodder beet

• Fodder beet was grazed in April and May. Crop yield was about 25 t DM/ha.
• Initially the fodder beet was lifted, with the intention of 1 herd getting the lifted beet on the feed pad and one herd grazing in situ. If the weather was looking like deteriorating, then the intention was to lift a stockpile and stop grazing. Lifting was taking about 25 t wet weight per hour, depending on how wet the soil was.
• However to simplify the system and save on labour and costs, both herds were moved to grazing in situ, with a herd grazing each end of the paddock.
• One herd was being milked 3 times in 2 days and the other OAD due to the wet weather, lameness and lower BCS
• At dry off, cows were being offered up to 8 kg DM/cow of beet.

Introducing catch crop, increasing area of fodder beet, increasing imported supplement and continuing to establish plantain as part of a mixed sward

A catch crop of oats and Italian ryegrass were grown following last autumn’s fodder beet and this was harvested in November. The area of fodder beet established was increased to 12.4 ha. However, having this area out of rotation made logistics challenging. Slightly less N fertiliser was used, 280 kg N/ha (compared with 300 kg N/ha in 2016-17). However, this was offset by an increase in imported supplement from 391 to 657 kg DM/cow.

Catch crops

The Rathgens took part in an oats/Italian ryegrass catch crop trial, with oats and Italian being sown immediately after the fodder beet was finished being grazed (mid-June).

• Fallow areas were left to allow a soil mineral N comparison with areas where catch crop had been sown.
• At harvest (mid Oct) the oat crop yielded 1.7 tonnes DM/Ha at 12.6 MJM/kg DM.
• Soil mineral N values at harvest show that soil beneath the oat crop had 18% less soil mineral N than soil left fallow.

Increasing imported supplement use due to dry summer and autumn, continuing to establish plantain as part of a mixed sward

The oats-Italian catch crop established after last autumn’s fodder beet was harvested in December – a month later than planned due to weather. This achieved more yield but poorer quality. Fodder beet for autumn feed was established in two smaller paddocks (9 ha) to reduce area out while maintaining the logistical advantages of having two paddocks. N fertiliser increased slightly from 280 to 287 kg N/ha and a greater proportion was applied in autumn due to poor growing conditions. Feed deficit also resulted in an increase in imported supplement from 657 to 908 kg DM/cow. This increased use of inputs in autumn increased the purchased N surplus and estimated N leaching. The results of this year illustrate how a few poor autumn months can determine the overall environmental performance, with Overseer’s N leaching based on long-term average weather conditions.

Arable farm – Annual summary of modelling study

The Simple Crop Resource Uptake Model operating within the Agricultural Production Systems sIMulator (SCRUM-APSIM) was used to simulate N balances on the arable part of the farming enterprise.

Model estimates indicated minimal drainage and N leaching, due to the farm being characterised by Claremont soils, which are heavy and poorly drained.

Modelling outcomes identified the following factors that increased the amount of leachable soil N:

• application of N fertiliser in excess of crop N requirements,
• mineralisation of N-rich crop residues,
• paddocks remaining fallow during the high N leaching risk period (autumn-winter) and
• “leaky” soils e.g. the free-draining Eyre and Lismore soils.

Evaluation of N remaining in the soil at the end of the season

Strategies that can reduce the amount of soil N available for leaching include:

• sowing catch crops immediately after the summer harvest to mop up residual soil mineral N or N mineralised from soil organic matter and crop residues, and
• reduced N fertiliser use by calculating requirements with a recommendation system that accounts for soil mineralisation.

Evaluation of N remaining in the soil at the end of the season indicated high N values for some paddocks, but prompt sowing after the harvest of previous crops allowed utilisation of residual soil N. The farm also implements under-sowing which provides continuous crop cover, but the benefits cannot be estimated because this management practice is not yet implemented in SCRUM-APSIM.

Leaching was greater when at least two of these factors were in play. For example, excessive N application to spring-sown crops resulted in high residual soil N at risk of leaching if no crops were sown after the summer harvest to take up the N. Similarly, mineralised N from residues retained in paddocks was available for leaching if there was no vegetation to take up water and N and reduce drainage/leaching.

Denitrification and run-off were assessed as additional pathways of N loss due to the poorly drained Claremont soils on farm. The results indicated an association with poorly drained soils with increases in denitrification and run-off.

Indicative model estimates of whole farm average denitrification (kg N/ha) and run-off (mm) for each of the four seasons:

Overall results have shown that leaching is mainly influenced by rainfall through its impact on drainage, but farm management practices determine the amount of soil N at risk of leaching.

The industry-agreed Good Management Practice for nutrient management is to match the nutrient supply from the soil and fertiliser to the demand from the crop to reach its expected yield. To do this with confidence, farmers require reliable information and methods for working out how much fertiliser to apply to their crops.

Comparisons were carried out in crops sown in spring 2017

Comparisons were carried out in crops sown in spring 2017, comparing farmers’ current N application rates with APSIM forecasts based on deep mineral N sampling. Models like APSIM use a mass balance approach to determine how much fertiliser should be applied to the crop to achieve its potential yield. Paddocks were selected and divided into two sections to demonstrate crop performance using farmer- and model-estimated fertiliser N application rates.

Paddock sections that received model-estimated N rates yielded similar dry matter (DM), slightly greater N use efficiency and the same soil residual N at harvest compared with the paddock sections which used farmer-estimated N rates. Modelled N leaching was negligible on this farm.

Applied N fertiliser, crop yield, N use efficiency and model-predicted N leaching and residual soil N at harvest for demonstration paddock in 2017-18 season. Soil sampling date 10 Aug 2017 for pre-sowing soil mineral N.