Project: To Examine Marginal Cost Of Water For Irrigation Ruataniwha Water Scheme.
Farm Systems Options And Economic Impacts.
June 2013

Prepared By Barrie Ridler

Grazing Systems Ltd

14 JUNE 2013


  • The cost to irrigate is high (Appendix 1).
  • These costs result in the cost of extra feed produced being more costly than other forms of supplement (Appendix 2).
  • Simple calculations show that such costs are not able to be met from dairy. (Appendix 3).
  • A more analytical approach using a linear programming modelling provides a clearer picture of how such costs impact actual farm systems. (Main report Tables. Appendix 4 farm costs).
  • This approach confirms that irrigated pasture under the conditions that apply to the Ruataniwha area is uneconomic when compared to other dairy farm systems that can be implemented within this region,
  • Such systems can substitute irrigated pasture with bought in supplements or can alter the way resources are used and make more final profit.
  • The analysis indicates that no charge can be made for water given the capital, depreciation and running costs associated with irrigation in this area compared to other forms of additional feed or from allocating resources in a more efficient manner.
  • The analysis has not investigated the cost/benefit of the scheme given that water need not be required each year but must be paid for.
  • This questionable need for water on a take or pay basis plus the costs of conversion to an irrigated system make the fixed costs greater than return even when the water may provide additional feed.

Farm Scenario Report Table:

Farm Scenario Report Table

Run 1.0

Data from several farms in the area (irrigated and unirrigated) was used to establish a Base Farm scenario. The data chosen was for a dryer year (2007/08). This then established a typical dairy farm of 275 ha with 610 cows currently producing at 350kgMS / cow. (Run 1.0).

The Base farm pasture production was 8,360 kgDM /ha. dry year. (See Appendix 4 for detailed figures.)

The GSL LP model was then used to compare this Base Farm scenario with a series of Runs where both improved performance. Optional systems were compared over a range of milk solids prices ($6; $7; $8; $9 /kgMS sold). All costs for FWE and for inputs (supplements, nitrogen, grazing off charges, power charges, and crop expenses) remained the same as milk solids price increased but the marginal analysis of the LP allowed changes to input quantity as the MC vs. MR balance changed with increasing MS price.

It is important to note that production per cow can be improved from this Base level without irrigation. It is therefore required that the production possible without irrigation be clearly established before an accurate comparison can be made between systems. This prevents any benefits that are possible from better management alone being attributed solely to irrigation.

Production per cow was increased to 411 kgMS/ cow by a process that many pasture based managers are currently undertaking (and is illustrated by changes that have occurred at Lincoln University Dairy Farm – LUDF). The Ruataniwha area is not mid Canterbury but similar strategies can be undertaken. Such strategies may not yield the same improvements due to differing environments.

The process is:
Reduce number of replacements entering the herd through more emphasis on rearing young stock to recommended live weights. This will have the dual effect of lowering need for replacements as fertility rates climb and longevity in the herd improves. (Appendix 4 detail). This also provides a higher number of cows at the peak of their lactation capability thus reducing the feed used on maintenance and growth during lactation. These stock will also have a greater potential to produce at a higher level due to better intake capacity.

Each animal then needs to be provided with sufficient high quality feed and the level of 411 kg MS/cow can be achieved from pasture and crop in this region. Realistically it would be very difficult to achieve levels above 425kgMS without resort to concentrates which are either marginally economic or not economic.

Run 2.0

This Run uses the production scenario described with the LP model allowed
to adjust all factors to provide the best economic result. Cow numbers are decreased, less supplement is purchased but more milk solids in total are
produced due to more feed being used for milk compared to growth and maintenance.

This Run 2.0 now provides a realistic Base production scenario to compare with irrigation.

Runs 3.0 – 6.0 then include irrigated pasture as an option to use to improve profit at varying levels of milk solids price. (Cost of irrigation and water required Appendices 1 and 2.) The irrigated pasture is not used until the payout reaches $9. However this is partly due to the opportunity of the model to substitute other inputs which have been constrained by limits on other supplementary feeds. When this constraint is removed, the model rejects irrigation and uses a combination of summer crop and supplements instead.

Although irrigation can be used to provide additional feed and milk solids, the $surplus between Income and costs is far less. This is illustrated in the Table by comparison between the improved Base Farm Run 2.0 ($665,626 at $6MS price and 223,900 kg MS) with Run 9.0 ($345,243 at $6 MS price and 363,443 kg MS) a $320,392 advantage for the simpler pasture, farm silage, some supplement and summer crop with reduced stocking rate system in Run 2.0. This is about $1600 REDUCTION per hectare in profit due to the cost of irrigation.

The Run 2.0 system is capable of producing through a dry year (as modelled). Both the non irrigated and irrigated systems suffer some reductions in drought (such as 2013 summer) as 400mm is not sufficient to grow the extra pasture and supplement purchase becomes prohibitive with 906 cows compared to the 545 cows in the pasture/crop/farm silage. The Run 2.0 cows can milk longer into the drought but when the turnips and farm silage run out (late March) the herd is dry. The model dries some cows off in early March as well.

By comparison all cows must be dry by early March for the irrigated 906 cow herd due to much higher demand. Increased levels of supplements required to maintain the higher herd numbers adds much greater costs. Further runs verified the preference for supplements to be purchased, fed out and utilised at costs stated in Appendix 4 rather than use irrigation.

The difference between profits varied from $1120 / ha to $1600 /ha in favour of non irrigated pasture systems.

Although not a part of the economic analysis, N leach figures were also produced and show large increases in N leach with higher stocking rates, production and irrigation. This will likely be a problem area in the future for any farm that does decide to use irrigation in this area.


The ability of LP to integrate factors of production with the marginal economics that apply to each possible substitution provides a very clear insight into the economic difference and systems required under optional management scenarios. However the data produced can be daunting as multiple factors are altering between each run. In this case, per cow production, base pasture growth, quality, costs and choice of substitutes has remained constant.

Herd number and use of additional feed changes as milk solids price changes to reflect the marginal cost and return changes that are occurring. It is important to take all these factors into account as each scenario is compared, remembering that milk solids price change is also occurring between Runs 3 – 6.

This analysis indicates that if pivot irrigation costs are as stated in Appendix 1, the water distribution fee cannot be charged without causing a loss. The actual cost of extra pasture from irrigation (despite being of higher quality and no cash cost to feed out) is still marginally more expensive than from supplementary feed even with the associated costs of feeding and utilisation of supplements included.

Although no analysis of the actual requirement for water has been undertaken, any cost of water on the “take or pay” basis proposed will incur greater average losses in comparison to an optimised “dry” pasture system.

It should be noted that irrigation (if it covers a water deficit period and produces an additional 4000kgDM/ha consumed per year) occurs in a period between November and March. Water does not affect pasture supply outside these times (although some better growth rate has been allowed within the model). This is shown by the 2 weekly modelling of feed demand and supply within the GSL model which produces a more “unbalanced” feed supply profile with irrigation.

Higher stocking rates to take advantage of any extra feed in summer months requires large inputs of supplementary feeds between July and late October (at a time when pasture should be supplying the high quality feed required for production above 400 kgMS/cow) and again from early March until cows are sent grazing. This adds more costs to a system which is not matched to the long cold winters and frosts which provide little good pasture growth from May until late September or October.

LUDF has measured pasture growth as high as 70 kgDM/ha/day in August. This region struggles to grow at 20 kgDM/ha/day at this time.

Anecdotal comment from farmers currently using water in the region is that they cannot pay the indicated for water due to the actual costs of running their farms and the irrigation system.

It seems they are correct.

Basis for Analysis.

A linear programming (LP) resource allocation model (GSL LP model) was used to provide a number of options for a dairy farm.

  • Data from several farms in the area were used to establish pasture growth rates and pasture quality throughout each 2 weekly period for each year. These data included a typical “dry” year for this area.
  • Data was extended to include factors of quality, timing, cost and quantity of the actual resources and inputs of the farm. Although this was not always precise, discussions with those concerned and some advisors in the area allowed adjustments of inputs until the model provided answers that represented a dairy farm system for the area.
  • The costs included in these runs are those directly associated with the FWE.
  • All costs associated with the land must be added to the figures presented. These should include such costs as rates, depreciation, and wages of management, capital or financing costs and depend upon individual farm circumstances.
  • Any crops or supplementary feed costs are calculated as needed within the analysis.
  • The additional costs associated with added inputs (such as crops, supplements, nitrogen) and all costs associated with irrigation (development, infrastructure, maintenance, depreciation, pumping, power and labour) are included when such inputs are added included during the Linear Programming (LP) process.
  • The cost of water is added as a separate input from the other irrigation costs as a means to investigate break even prices for either water or milk solids.
  • The dairy herd consisted of predominantly FxJ cows.
  • These cows were estimated at a mature LW at calving of 500 kg. CS 5.0
  • The maximum milk solids production per cow was based upon the DairyNZ recommendation of a herd average 0.83 MS / kgLW (about 415 kgMS/cow).
  • Cull cows were at a schedule price of $3.00/kg CW and calves at $45/hd.
  • The figures therefore are a comparison between varying systems- with the main emphasis on comparing alternative systems with or without irrigation and some change to summer rainfall.
  • The additional costs (Marginal Cost) and additional income (Marginal Return) for the different systems are then compared between systems as calculated by the GSL linear programming model which may substitute between alternatives or be constrained to specific options.
  • The runs produced are examples of the relative resources used for each option and the likely comparison of final profit. Data were gathered from various sources (dairy farms in the area).
  • These data were used to establish an initial “Base” system using a linear programming (LP) resource allocation model (GSL LP model).
  • This system was either constrained or optimised for individual inputs depending on the systems being investigated.
  • Data on feed use, timing, productions, pasture and supplement flows from all sources were also reported but not all included in the summary Tables.
  • The inputs required for an N leach figure from the Overseer ® program were produced but were calculated separately from the GSL program.
  • These data are summarised in the attached spreadsheet.
  • Use of the LP methodology allowed a series of such runs to be completed with some inputs remaining constant. The LP optimisation allowed substitution of similar resources or different combinations of inputs and output to create optional systems.
  • This methodology allowed a number of farm systems to be investigated and provided systems data and comparative economic data between the differing combinations of the farm’s specific resources.

Appendix 1. (Per hectare costs and assuming existing “dry” farm) Yearly per ha cost irrigation.

DairyNZ irrigation conversion figures/ha.
Water, re-fencing, tree removal works including earthworks, races and
culverts $2050
Pivot system (includes some charge for pipeline to farm and allowance for irregular shapes and contour) $5000
$550 cost of finance (interest only)
$675 depreciation and insurance of irrigation equipment
$150 depreciation other assets

Total $1375/ha/year.

Additional costs with increased production:

Additional cow and part replacement $1800 interest only $150
Additional shares (400 x $7.50 but assume dividend covers this)
Additional accommodation (depn and interest on capital only) $100.
Cost of capital and depreciation to change to irrigation/ha/yr $1625.
Add: cost of power ($78) and pumping costs ($150) = $228/ha/yr

So total to costs (less water charge) to apply 400mm/ha/year = $1853.

Appendix 2.

Simple marginal cost of additional dry matter produced from irrigation.
(Refer also to simple calculator spreadsheet.)
Assume extra 4000kg DM is consumed from 400mm extra water over summer months.
Yearly cost from Appendix 1 is about $1850/ha.
Add cost of water at 25 cents per cubic metre (1mm/ha =10 cum) 400mm = 4000 cum x $0.25 = $1000.
So additional 4000kgDM costs ($1850+$1000) $2850.

This costs each additional kgDM at ($2850/4000kgDM) $0.70 /kgDM.

Appendix 3.
Simplistic breakeven price for milksolids if using irrigated pasture (assume one additional 400kgMS cow/ha is now milked on this extra feed).
Simplistically at an efficiency ratio of 12kgDM/kgMS the feed cost alone will require a MS payment of $8.40/kgMS.

This however ignores the additional costs of running the cow for a year
(animal health, proportion of labour costs and other FWE) of between $400 – $600/cow per year.
A better answer for this simple calculation is that these additional milksolids
from irrigated pasture would require a Milksolids price of about $9.50 -$ 10.00 /kgMS payment.
But this assumes ALL feed for the animal comes from irrigation and this is not so as winter and spring/autumn feed may be another supplement or pasture.

NOTE: These assumptions are also based on the assumption that water will be required every summer to the 400mm/ha application rate that (take or pay arrangement) is purchased.
If this is not required, the only costs that reduce are part of the power costs (some of this cost is fixed charge) and pumping maintenance costs.
The fixed costs involved with this type of irrigation greatly escalate the risks involved to the farmer.
There are other more profitable options to choose from in terms of farm systems.
These are investigated in more depth with the GSL LP resource allocation model.

Appendix 4.

Costs and productions.
Base Farm dry year growth 8,360 kgDM/ha/year. Low autumn, winter (<10kgDM/ha/day) and spring PGR.
Normal year 11,000kgDM 82% utilisation.
Pasture quality dry in late spring 11.7MJME/kgDM.
Supplementary feed costs vary from 30-32 cents /kg DM and 10.5 – 11.0 MJME/kgDM.
Feed out costs 4-5 cents /kgDM.
Utilisation of supplements 85-90%
Farm Working expenses about $900/cow.
All additional input costs added within GSL model as used. Turnip crop 11.5 tonne on effluent $850 to new grass. Utilisation 80% fed Feb/March.
Farm silage 15 cents make, 5 cents feed out 85% utilisation 10.5MJME. 10-15% wastage in making paddock to pit out. Grazing off charges $8; $12; $19 per week.

Nitrogen 1.70/kgN $20/ha application.
Average 10:1 response rate.

Irrigated pasture. 400mm applied 375mm effective /ha November to early March if water allocation lasts.
Produces about 4- 5000kg extra DM in dry year (14,050 kgDM/ha/year) with 85% utilised. 11.8MJME average.

Herd replacement 23% to 18%
4 to 3% losses
Longevity from 5 to 7 years in herd.
2yr olds 75% improved to 85% of mature cow performance
3 yr olds 85% improved to 95% of mature cow performance. 500 kgLW CS 5.0 at calving.
6 km allowed per day for walking.