logo

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

Topics

13 min read

Farm facts What did you get out? Why did you decide to join? Why it's important? Farmers & industry benefit Video Management and experience N leaching and drainage Fertiliser N applied and N uptake Demonstrations on arable farm Further information

The Rathgen family manage a mixed dairy and arable farm in St Andrews. Your farm, if similar, could benefit from their insights after joining the FRNL programme. They believe FRNL bridges the gap between environmental concerns and on-farm activities. This programme offers them updated information, enabling them to reduce their environmental impact while boosting their economic benefits. They've also explored different management practices and modelled their effects. Networking with like-minded individuals is another perk. Joining such initiatives can help you stay ahead of new regulations and farm more sustainably, bringing both environmental and economic benefits to the wider farming community.

Farm facts

Area:

312 ha effective milking platform, 137 ha arable

Dominant soils:

Claremont silty loam and Waitohi silty loam.

Average rainfall:

approx. 580 mm

Trading as:

St. Andrews Dairy Limited and Glen River Farm Limited

What did you get out of the FRNL programme?

  • I think FRNL is relevant as a way to help bridge the gap between the environment and what’s happening on farm. Establishing catch crops in wet winters remains a challenge (especially since it is the wet winters that are in the most need of catch crops). Going forward this is something that needs more consideration.
  • The FRNL programme is giving us up to date information that we can use to try and minimise our environmental footprint, while maximising our economic benefit.
  • It is giving us idea's on different management practises that could be useful in our system and, using the data collected, we can run models on different scenarios (using fodder beet on platform, lower stocking rate) and get an idea of the possible outcomes, environmentally and economically.
  • The programme has also allowed us to network and compare with other producers and agencies who have similar goals.  

Why did you decide to join the programme?

“I thought it would be a good way to get some new ideas and network with some new people in the industry.”

Why do you think it's important?

"To keep ahead of new regulations and to farm as environmentally as possible."

How will farmers and the industry benefit?

“Hopefully environmentally and economically.”

Video

Nigel Rathgen discuss following an autumn grazed fodder beet crop with a catch crop.

The dairy farm

The dairy farm has an effective milking platform area of 312 ha. At the start of the programme this was 94% irrigated and stocked at a stocking rate of 3.8 cows/ha. The farm also has an 80 ha dryland support block providing winter grazing.

Since being involved in the FRNL programme, the Rathgens have:

  • Introduced fodder beet onto the milking platform, followed by an oat catch crop

  • Re-grassed a significant proportion of the farm in mixed sward including plantain

  • Increased home-grown feed and pasture eaten

Nitrogen (N) fertiliser use was stable; supplement use and stock numbers have fluctuated. Differences in N leaching between years were maximal 11%.

Management and experience with implementing FRNL options

2015/16

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.

2016/17

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.

2017/18

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.

2018/19

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:

Season Denitrification (kg N/ha) Run-off (mm)
2015-16 0.6 1.7
2016-17 1.2 0.8
2017-18 5.5 46
2018-19 9.5 33

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.

A full report on the catch crop results of the Canterbury dairy monitor farms.

Monthly total rainfall and mean temperature obtained from the National Institute of Atmospheric Research – NIWA (2019) and used in St Andrews simulations.

N leaching and drainage

The average N leaching values generated at 150 cm depth were marginal, ranging from 0 to 7.4 kg N/ha. These low figures reflect the low estimated drainage of the poorly drained Claremont soils which are dominant on the farm. N leaching estimates generated at a soil depth of 60 cm were greater than those estimated at 150 cm depth, but there was similarity in patterns across seasons. Greater values of N leaching and drainage at 60 cm are expected because most crops grown on the farm have deep roots that allow water and N extraction in layers that are deeper than 60 cm.

N leaching and drainage over four seasons (season = 01 Apr - 31 Mar) as estimated by SCRUM-APSIM at a depth of 150cm

Paddock Leaching (kg N/ha) Drainage (mm)
2015-16 2016-17 2017-18 2018-19 2015-16 2016-17 2017-18 2018-19
R02 0.0 0.0 7.1 9.5 0.0 0.0 105 124
R03 0.0 0.0 0.9 1.2 0.0 0.0 51 51
R04 0.0 0.0 0.9 1.2 0.0 0.0 51 47
R05 0.0 0.0 1.1 20.8 0.0 0.0 23 276
R06 0.0 0.0 5.0 6.6 0.0 0.0 99 107
R07 0.0 0.0 1.6 3.1 0.0 0.0 66 68
R08 0.0 0.0 3.0 4.5 0.0 0.0 93 190
R09 0.0 0.0 2.9 8.3 0.0 0.0 58 113
R10 0.0 0.0 0.3 5.1 0.0 0.4 14 141
R11 0.0 0.0 0.9 4.1 0.0 0.0 32 90
R12 0.0 0.0 0.8 8.0 0.0 0.0 32 173
R13 0.0 0.0 5.7 6.7 0.0 0.0 96 151
R14 0.0 0.0 6.6 12.3 0.0 0.0 113 93
R15 0.0 0.0 2.4 6.1 0.0 0.0 64 74
R16 0.0 0.0 1.3 5.5 0.0 0.0 41 93
R17 0.0 0.0 4.1 5.7 0.0 0.0 78 76
R18 0.0 0.0 3.0 3.8 0.0 0.0 125 76
R19 0.0 0.0 1.9 9.6 0.0 0.0 48 174
R21 0.0 0.1 4.7 24.3 0.0 3.0 81 265
Whole-farm average 0.0 0.0 3.1 7.4 0.0 0.1 70 117

N leaching and drainage over four seasons (season = 01 Apr - 31 Mar) as estimated by SCRUM-APSIM at a depth of 60cm.

Paddock Leaching (kg N/ha) Drainage (mm)
2015-16 2016-17 2017-18 2018-19 2015-16 2016-17 2017-18 2018-19
R02 1.3 10.6 24.9 7.5 0 8 152 138
R03 0.0 0.5 3.7 19.0 0 17 79 90
R04 0.0 1.1 3.7 18.7 0 17 79 87
R05 0.2 11.6 26.2 4.7 0 24 64 280
R06 0.0 8.7 28.4 9.4 0 29 172 144
R07 2.8 1.3 16.8 37.3 0 27 96 95
R08 0.0 1.6 12.9 2.4 0 14 168 191
R09 0.0 1.5 38.8 33.4 0 24 76 138
R10 0.0 2.3 7.5 12.1 0 24 63 153
R11 0.1 1.7 30.7 36.2 1.5 17 87 147
R12 0.0 3.9 15.3 9.9 0 11 64 177
R13 0.0 2.5 16.6 39.2 0 15 127 179
R14 1.4 11.5 33.4 51.0 0 19 150 130
R15 1.4 3.9 19.4 7.6 2.9 25 85 103
R16 0.0 1.6 13.1 9.1 0 16 76 123
R17 0.0 3.0 31.7 10.2 0 14 112 91
R18 0.1 0.7 4.3 25.7 0.3 18 158 102
R19 0.0 12.3 14.0 3.6 0 21 94 178
R21 1.9 18.5 38.2 30.0 11.6 29 109 287
Whole-farm average 0.5 6.3 20.0 21.8 0.7 19 109 143

Fertiliser N applied and N uptake

Farm records show that most paddocks were sown in crops soon after completion of grazing and harvesting events, allowing the utilisation of residual soil N and N returned in manure and urine. Farm records showed a reduction in the average N fertiliser applied over the studied seasons.

Total N application (fertiliser and estimates of manure and urine return) and N uptake (as estimated by SCRUM-APSIM) over four seasons (season = 01 Apr - 31 Mar)

Paddock Applied fertiliser + excreta (kg N/ha) N uptake (kg N/ha)
2015-16 2016-17 2017-18 2018-19 2015-16 2016-17 2017-18 2018-19
R02 269 249 198 158 295 341 226 216
R03 54 42 0 221 288 149 114 276
R04 54 44 0 221 97 147 112 276
R05 198 0 39 35 217 124 119 31
R06 27 0 189 83 141 78 242 159
R07 240 152 192 263 229 160 255 276
R08 238 89 184 35 314 156 230 81
R09 54 163 177 166 70 153 169 232
R10 191 46 52 86 182 117 96 158
R11 159 240 231 204 154 248 297 310
R12 104 46 58 136 112 107 100 137
R13 201 95 182 212 195 159 231 227
R14 209 256 240 221 225 344 286 296
R15 17 81 205 243 140 187 287 281
R16 228 185 0 35 217 181 142 214
R17 253 143 187 152 244 139 328 288
R18 195 155 37 221 116 287 78 276
R19 100 197 189 43 94 207 257 72
R21 118 46 228 102 78 226 228 169
Whole-farm average 166 130 139 161 186 195 205 223

Demonstrations on arable farm

Estimating N fertiliser rates

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.

Paddock Crop Pre-sowing soil mineral N (kg N/ha) N rate estimated by Applied N (kg N/ha) Yield (t DM/ha) N use efficiency* Leaching (kg N/ha) Residual N (kg N/ha)
R12 Oats 17.1 Farmer Model 140 120 9.92 10.3 1.35 1.63 - - 14.6 14.6

Further information

An article relating to the SCRUM-APSIM modelling work was published in Agronomy NZ.

A full report on the arable monitor farms, including the detailed results of the catch crop demonstrations on this farm and at Chertsey, and an N fertiliser rate demonstration on another arable monitor farm.

Last updated: Aug 2023
Share:
Tags related to “Forages for Reduced Nitrate Leaching, Chris Nigel and Ross Rathgen (Canterbury)”