University of Kentucky

UKnowledge

International Grassland Congress Proceedings XIX International Grassland Congress

Feeding of Conserved Forage - Implications to Grassland

Management and Production
C. W. Holmes
Massey University, New Zealand

P. N. P. Matthews
Massey University, New Zealand

Follow this and additional works at: https://uknowledge.uky.edu/igc

Part of the Plant Sciences Commons, and the Soil Science Commons
This document is available at https://uknowledge.uky.edu/igc/19/19/24
This collection is currently under construction.
The XIX International Grassland Congress took place in São Pedro, São Paulo, Brazil from
February 11 through February 21, 2001.
Proceedings published by Fundacao de Estudos Agrarios Luiz de Queiroz

This Event is brought to you for free and open access by the Plant and Soil Sciences at UKnowledge. It has been
accepted for inclusion in International Grassland Congress Proceedings by an authorized administrator of
UKnowledge. For more information, please contact UKnowledge@lsv.uky.edu.
 FEEDING OF CONSERVED FORAGE -
IMPLICATIONS TO GRASSLAND MANAGEMENT AND PRODUCTION

C.W. Holmes and P.N.P Matthews

Institute of Veterinary, Animal and Biomedical Sciences, Massey University,
New Zealand

Introduction

This review will focus on the use of conserved forages in pastoral dairy farm systems;
especially on their conservation as an aid to good grazing management during periods of rapid
pasture growth, and on their consumption as an aid to good feeding management and grazing
management during periods of slow pasture growth. Conservation of pasture, in order to
move it from one time to another later time, always involves “costs”, either financial
expenditure or physical losses (of dry matter/quality) or both, which must be evaluated in the
whole system.
Most of the data is drawn from New Zealand and Australia, with some from England
and Ireland. The topic has been reviewed before (e.g. Rogers, 1984 and 1985; Leaver, 1985;
Phillips, 1988; Mayne, 1991; Stockdale et al., 1997), but the present review will put greater
emphasis on the effects on pasture management and on the whole pastoral system, and on
recent results, because the cows’ ability to respond to supplementary feeds is now larger than
it used to be, as a result of steady genetic improvement in milk producing capacities. In
recent years there has been marked increase in supplement use on pasture based dairy farming
systems in New Zealand often driven by perceived limitations to production increases in
existing systems rather than profit (Attrill and Miller 1996). This review attempts to identify,
and discuss, the factors involved in supplementary feeding (i.e. the responses of cows and
farming systems to extra feed) to enable sustainable increases in productivity and profitability
on pastoral based systems.
Grazed pasture, the basal ration, is a living source of feed; its supply varies with
season and climate, and its future growth can be influenced by its present grazing
management. A key feature of the interaction between the cow and the pasture is the fact that
an increase in the level of feeding, either from added pasture or added supplements, will
always cause an increase in the residual herbage mass left after grazing. This, in turn, will
affect the future growth and composition of the pasture. The pastoral dairy systems must
achieve close synchrony and balance between feed demand and pasture supply at all times. In
these intensive pastoral systems, used mainly in the Southern Hemisphere, including New
Zealand, Australia (Victoria and Tasmania) and parts of South America and South Africa,
conserved forages are used as a supplement to the basal ration of grazed pasture to help
achieve the balance with feed demand when pasture growth is limiting. The conserved
forages are almost never used as the basal ration, as is commonly the case in many Northern
Hemisphere regions, at least in winter (Thomas and Thomas, 1989).

Milk Production on NZ Dairy Farms; An Example of Intensive Pastoral Systems

There are 3.3 million cows and 3.8 million people in New Zealand; consequently 90%
of milk is exported, without subsidies or incentives, at the “world market price”. Currently
this is about NZ$3.50/kg milksolids (fat plus protein), or 29c/L, with 4.7% fat and 3.8%
protein.
 The costs (c/kg DM) of various feeds are: - Pasture silage, 10 to 15c; maize silage, 14
to 20c; grains, 25 to 40c; concentrates, 40 to 50c; grazed pasture (variable costs), 3 to 5c.
Consequently grazed pasture is the only high-quality feed that can be used as the basal ration
for cows, although silages can be fed strategically, and grain can be fed occasionally in the
grain-growing districts. This financial need to use grazed pasture as the basal ration is made
physically possible as the temperate climate allowing pastures to grow and to be grazed
throughout the whole year. These pastures contain high concentrations of metabolisable
energy and crude protein (10 to 12 MJME/kg DM; 17 to 27% CP; Moller et al., 1996).
Ninety percent of all the cows calve in early spring, in order to synchronise their
increased feed requirements of early lactation with the increased pasture growth in spring-
time. They are later dried-off after relatively short lactations (220 to 240 days), before the
following winter, in order to balance their reduced feed requirements, when non-lactating,
with the reduced pasture growth in winter (see Figure 1). The remaining 10% of cows are
milked through the winter, usually after calving in autumn, either by feeding a high
proportion of silage in their winter ration, or by operating with a low stocking rate during
winter (Garcia and Holmes, 1999). The small local market for short shelf-live products is a
key feature allowing the milk production pattern to follow the seasonality of pasture growth
rather the seasonality of consumer demand.
For the spring calved herd with a moderate stocking rate, a temporary pasture surplus
(growth faster than consumption) occurs during October to December. This can be
conserved, and fed during later periods of pasture deficit (summer, winter or early spring),
which may in turn allow earlier calving or later drying-off. However for the spring-calved
herd with a higher stocking rate, there may be little or no pasture surplus at anytime of year,
but there will be large pasture deficits in summer, winter and autumn (Figure 2).
Supplementary feeds must be imported on to the higher stocked farm to fill periods of deficit.
Average stocking rates have increased from about 1 cow/ha in 1950 to 2.7 cows/ha in
2000; this combined with increased milk yield/cow has created much higher feed demands
per hectare. Therefore most dairy farms now import feed in one form or another, whereas
most farms were self-contained for feeds prior to 1975. With the increased use of
supplementary feeds in New Zealand dairy farming systems farmers must take into account
the likely responses (productive and financial) to the extra feed as well as the short and long
term consequences of this extra feed to the grazing system (grazing management and sward
conditions).

The Response by Cows to the Feeding of Conserved Supplements

In all feeding systems, the cow’s responses to supplementary feed are determined by:
• The amount by which her intake from her basal ration is decreased (the
substitution effect).
• The net amount by which her total nutrient intake is increased.
• The proportion if this net increase which is partitioned into extra milk synthesis.
These factors also apply to grazing conditions, but in addition, grazing cows always
respond to an increase in feeding level (either from extra pasture or from extra supplement) by
leaving a higher post-grazing residual herbage mass (RHM) (Dalley et al., 1999). As
discussed elsewhere, in this review, this increase in RHM will have important subsequent
effects on sward characteristics through changes in the growth, senescence and composition
of the pasture.
The immediate and long term effects of supplementary feed in a grazing system are
illustrated Figure 3. In theory, if all of the metabolisable energy in the extra feed eaten was
absorbed by the udder and converted into milk then an extra 10 MJME in total intake would
produce an extra 6.5 MJ net energy in milk, equivalent to about 0.15 kg milksolids
(milksolids = fat plus protein).
In fact, consumption of 10 MJME as supplementary feed usually causes some decrease
in pasture consumption, and some increase in liveweight gain. Therefore the actual,
immediate response in extra milk production is always smaller than the theoretically possible
0.15 kg MS given above.
The pasture not eaten, and the extra liveweight, may subsequently contribute to the
production of extra milk in carry over, long-term responses (Figure 3). When the immediate
and the long-term effects of adding supplements to a pastoral system are added together, the
total response to extra feed in a grazing system will almost always be smaller than this
theoretical value. Some of the extra nutrients are “lost” from the system in the form of wasted
pasture, which decays, and/or extra liveweight which is never converted into milk.
The size of the response to extra feed depends mainly on the “need” for extra feed, by
the cow or by the system. Large responses will be achieved only if the current performance
of the cows or the system is being severely limited by the lack of feed. These situations
represent “potential deficits”, where the actual levels of intake and production by the cow or
the system are well below their potential levels.
The immediate effect of supplementary pasture silage was illustrated by Bryant and
Trigg (1982), for cows given a restricted daily allowance of pasture, with or without silage.
Consumption of 4 kg silage DM/day caused a 1.7 kg DM decrease in pasture consumption
(substitution rate 0.43) and consequently an increase in residual herbage mass, and increases
in milk production and in liveweight gain. The immediate milk response was 0.15 kg MS
from 4 kg silage DM, or 38 kg MS/kg DM, while the final, total response would depend on
the final fate of the extra liveweight and the extra residual herbage mass.

Substitution Rate

The substitution rate is affected by many factors including pregrazing herbage mass,
and quality of the herbage and the supplement (Stockdale et al., 1997). However the main
factor is the cow’s overall level of feeding (and production) relative to her potential intake
(and production), described mainly as the quantity of supplement fed and the quantity of
pasture offered (Grainger and Matthews, 1989).
For example, for forages: -

Substitution Rate = 0.01 + 0.16 x unsupplemented pasture intake.

Where substitution rate = Decrease in pasture intake

Supplement eaten

And unsupplemented pasture intake = Kg DM eaten/cow daily x 100

Kg LWT of the cow

(Larcombe, 1989; cited by Stockdale et al., 1997).

Substitution rates usually range from 0.3 to 0.7, with even higher values recorded for
cows given generous pasture allowances and large quantities of supplement (e.g. Phillips,
1988).
The substitution effect has obvious negative effects on the cows’ immediate response
to the extra supplementary feed. It will also have negative effects on the long-term responses
by the whole system, if the pasture which is spared by the supplement (substituted) is wasted
and not eaten later.
“Managed substitution” is widely used in commercial and experimental farming
systems, in order to meet targeted levels of management for the cows and the pastures. Some
of these are illustrated by the management decision rules and targets explained by MacDonald
and Penno (1998); and Phillips and Matthews (1994).
Grazing pasture is a relatively slow method of feed consumption, and it requires long
periods of grazing each day (e.g. 9 to 10 hours grazed per day; Watson, 1999). These long
periods are in marked contrast to the time taken to eat diets of silage and concentrates; 3.5 to
4.0 hours per 24 hours (Jackson et al.,1991). Cows presumably prefer to consume feeds
which can be ingested at faster rates, and thus be able to spend less time per day in the activity
of feed consumption.
Rates of feed consumption are:

Pasture 20 to 40 g DM/min
Hay/Silage 50 to 70 g DM/min
Concentrates 250 g DM/min meal
350 g DM/min pellets

(Phillips, 1988; Mayne, 1991; Bramley et al., 1992)

Consumption of supplementary feeds generally cause decreases in total grazing times,

ranging from 22 to 26 minutes per kg DM eaten as concentrates or maize silage (Sarker and
Holmes, 1974; Watson, 1999); up to 30 to 40 mins/kg DM eaten as hay or silage (Phillips and
Leaver, 1985; Mayne, 1991). The decrease in grazing time is probably linked to the rate at
which the supplement can be eaten. However time is rarely the factor which limits the cows’
response to supplements, because the main factor is usually her “potential deficit”.
It is tempting to suggest that forage supplements will cause larger substitution rates
than concentrates, because the forages are eaten more slowly, and cause larger decreases in
grazing time. However there is no clear evidence that this is true (Stockdale et al., 1997),
probably because so many other factors also influence the substitution rate.

Partitioning

The ability of the cow to utilise a high proportion of the extra nutrients, derived from
supplements, into the synthesis of extra milk is another key determinant of her immediate
response to extra feed (Figure 4). A high proportion will be partitioned into extra milk only if
the cow’s potential milk yield is much higher than her current, actual milk yield because of a
relative feed deficit.
Calculations from the data reported and cited by Grainger (1990) and Stockdale and
Trigg (1989) provide estimates for the percentage of extra metabolisable energy partitioned
towards the synthesis of extra milk: -
• Early lactation, and current milk yield which is much lower than potential milk
yield because of severe feed restriction: 70%
• Early lactation, and only moderate feed restriction: 40%
• Late lactation: 35%
• No feed restriction at all: 0%
These values will be higher for cows with high genetic merit for milk production than
for cows with lower genetic merit (Fulkerson, 1997; Ferris et al., 1999), because these high
merit cows have very high potential milk yields, and therefore usually exist in a state of
energy deficit. Logic indicates that these high merit cows will also show smaller substitution
rates, but there is no evidence to justify this belief. Genetic improvements in milk production
capacity have and will continue to make important contributions to increasing responses to
extra feed.
The energy which is eaten but not used by the udder must be converted into extra
liveweight gain, which can contribute later to improved fertility (Garnsworth and Webb,
1999), or to increased milk production (Grainger and Wilhelms, 1982). Under conditions of
continued feed restriction after the period of supplementary feeding, the carry-over effect is
likely to be about 0.5 times the immediate response to the extra feed (Broster and Broster,
1984; Bryant and Trigg, 1982; Rogers, 1985).

The Response by the Farming System to the Feeding of Conserved Supplements

Introduction

Successful grazing systems are planned. From the setting of production and financial
objectives and systems planning through to grazing management, monitoring and evaluation.
Grazing management can be defined as the manipulation, within a grazing system, of the feed
supply (pasture and other supplementary forages) and animal feed demand to achieve planned
pasture and animal production targets. Although farmers are likely to have different grazing
management objectives and each system unique, grazing systems in New Zealand have, in
general, been based on high stocking rates, to achieve high levels of pasture utilisation and
per hectare production. These systems have resulted in low average cow production and short
lactation length to enable the animal feed demand profile to closely match that of pasture
growth with little or no reliance on additional supplementary forages. A grazing system can
be represented as a model of feed flow from pasture production through to animal product
(Figure 4). The average pasture cover and the level of animal performance achieved are used
to balance any variation between pasture production and intake per animal. The system is
controlled through the farmers control over animal intakes through stock policy decisions.
For this model to work successfully farmers must be willing to control the herbage intake of
grazing animals even though at times this will result in reduced intakes and lower levels of
animal production (Matthews 1994).
Currently there is considerable debate between the merits of this traditional mainly
pasture based system and alternative systems based on improved animal performance and the
increased use of supplements (e.g Penno et al., 1996, 1998, 1999, Brander and Matthews
1997, Cassells and Matthews 1995, Matthews 1997). If increased levels of production per
animal are to be achieved this effectively reduces the control of the grazing system through
controlling intake levels within the grazing system, and requires increased use of supplements
and/or reducing stocking rate to balance animal demand with pasture growth (Matthews
1994). The following are examples of the grazing objectives for a number of leading New
Zealand dairy farmers attempting to make effective use of additional forage inputs to achieve
a high per cow and per hectare productivity (Matthews 1995):
• To maximise pasture production.
• To make efficient use of the pasture grown.
• To exploit the production potential of the herd.
• To achieve a high per hectare production through high per cow performance.
To plan, manage and monitor grazing systems involves an understanding of the factors
affecting pasture production, average pasture cover, feed intake (requirements) per animal,
animal production and animal responses to supplementary feeding as well as the ability to
quantify and measure each of these components at the farm level. Grazing management also
requires an understanding of the interactions that take place between pastures and grazing
animals. Increased emphasis on system performance will require a clearer understanding of
the responses of both the animals and the farming system to supplementary forages to enable
economic responses to be captured.

Short Term Experiments

The average results from trials in which grass silage was fed with grazed pasture
(Table 1) show the major effects of the initial feeding level on the responses to supplementary
feeding and that that it can in fact reduce milk production (Phillips, 1988).
Average results (per kg DM fed) from 12 studies in which pasture silage was fed
during periods of pasture deficit (Rogers, 1985) were: -
• + 30 gm MS (+ 0.5 litre milk)
• + 150 gm liveweight
• - 0.4 to 0.6 kg pasture DM eaten per kg DM eaten as silage.
More recently an extra 20 to 40 gm MS/kg silage DM were produced by cows at a
high stocking rate, fed 5 kg silage/day in spring or summer (Clark, 1993) with the long-term,
results almost twice as big (Table 2). An extra 0.6 to 0.9 kg extra milk was produced per kg
maize silage DM by cows eating restricted quantities of moderate quality pastures (Stockdale,
1995). These responses decreased sharply in conditions where the total diet contained less
than about 14% crude protein, either because the diet contained a high proportion of maize
silage (CP% about 7 to 9%) or because the grazed pasture contained a low crude protein
concentration. Milk yield was increased by the inclusion of extra protein (fish, soya bean or
cotton seed meals) with the maize silage when it was fed with summer pasture in New
Zealand (MacDonald et al., 1998) or in Australia (Moran and Stockdale, 1992).
Chopped green maize can also be used to supply about half of the total ration for
grazing cows (Moran, 1992), and whole-crop silages (e.g. from conserved wheat) can also be
used successfully as a supplement for grazing cows (Waugh, 1997).
An extra 50 to 60 g MS were produced per kg DM grazed as turnips or as sorghum (at
3 to 4 kg DM/cow daily), by cows given a daily pasture allowance of 25 kg DM in summer
(Harris et al., 1998).
High quality supplements generally produce larger responses (Stockdale et al., 1997).
For example responses to hay, maize silage or concentrates were about 0.45, 0.8 or 1.3 kg
extra milk per kg extra DM. The benefits of silage with higher digestibility were illustrated
by Rogers (1984) and by MacDonald et al., (2000). The high quality of turnips and sorghum
(12 and 11 MJME/kg DM respectively) probably contributed to the relatively high immediate
responses reported above for these feeds.

Self Contained Systems; Forage Conserved on the Farm

The interaction between stocking rate and date of silage conservation was studied in
Taranaki, New Zealand, (Thomson et al., 1984). Early conservation had an adverse effect at
the higher stocking rate, reducing milk produced per cow prior to 1st December because
conservation induced a feed deficit in early lactation. However, at the lower stocking rate,
early conservation had a beneficial effect, increasing milk produced per cow after 1st
December, because of decreased dead matter concentration in the pasture during summer.
Another Taranaki experiment measured the productivity of farmlets on which surplus
pasture in spring was either harvested and conserved as silage, or carried through on the
paddock as “deferred grazing” to be grazed in late December and January (Thomson et al.,
1989). There were only small differences in milk yield between the two systems at both
stocking rates, except in one year when, at the higher stocking rate, the silage farmlet
produced 165kg milk solids per hectare more than the deferred grazing farmlet, due entirely to
extra milk produced after December 10, from 50 extra days in milk. The results suggest that
conservation as silage will reduce the variability of milk production between years at higher
stocking rates, but it was not “profitable” every year even at the higher stocking rate.
An experiment in Victoria, Australia, compared two conservation systems, one with
hay and the other with silage, both at 2 cows/hectare (Thomas & Matthews, 1991). The silage
was conserved at earlier dates and was of higher digestibility, (68% v 61%) than the hay. The
herd on the silage farmlet produced more milk per cow (4380 v 4049 litres) than the herd on
the hay farmlet, mainly due to differences while the supplement was fed during the summer.
However there was no difference between the two farmlets in economic surplus because the
silage farmlet incurred greater costs, because it was forced to purchase extra silage, whereas
the hay farmlet sold surplus hay. The loss of DM in the silage stack was 25%, which was one
reason for the need to purchase extra silage.

a - Dry Matter losses

Mowing, wilting and chopping resulted in slightly higher quality silage than silage
direct cut with a forage harvester (+ 3% milk per cow). However direct cutting caused
smaller losses of DM in the paddock and in the stack, and thereby produced 20% more milk
per hectare of silage (Small & Gordon, 1988). This experiment, and that by Thomas and
Matthews (1991) above, emphasises the important consequences of the losses of DM which
can occur during the conservation process, on the productivity of the whole farm. These
losses extend into the management of the silage face, especially for maize silage, and to the
feeding methods, which can be over 20% if silage is fed in the field (Phillips, 1988).

b - Growing Forage Crops

The total supply of feed was increased by growing 18 t DM/ha as maize on 20% or
50% of the farm, harvesting it as silage and feeding it to the cows as required (Campbell et
al., 1978). Milk production per hectare was increased, but, with the yields costs and prices in
the 1970’s, the gross margin per hectare was reduced. A more recent experiment showed
benefits from growing maize on 12% of the farm’s area in only 1 year out of two (Thomson et
al., 1998). A theoretical study of the incorporation of maize silage on dairy farms in Southern
England assumed yields of 12 to 18 t DM/hectare from maize and 10 t DM from pasture
(Doyle & Phipps, 1987). The model predicted increases in profit due to inclusion of maize on
up to 100% of the farm area, although the biggest increase was from 0% to 25% of the area as
maize. Obviously the results of such experiments will be very dependent on the yields of
DM from the maize, and the costs and prices involved.
Turnips can provide a large supply of “grazeable”, high quality feed (12 MJME/kg
DM) in summer when pasture growth and quality can both be low, and the relatively high
immediate responses (50 to 60g extra milksolids per kg extra DM) were described above.
However in the whole system, growing the crop reduces the area of pasture which causes a
temporary increase in stocking rate. This resulted in the conservation of less silage, and in
lower average body condition and daily milk yield of the cows by the time when consumption
of the turnips could begin (Thompson et al., 1997). The overall net effects of the turnip crop
on the whole farm’s productivity will depend on the size of the adverse effects while the crop
is growing and the size of the benefits while the crop is eaten. There is unlikely to be much
net benefit unless the turnip yield exceeds about 12 t DM/ha and the crop allows the dry-off
date to be delayed, giving extra days in milk.

c - The use Of Nitrogen fertiliser to Conserve extra silage

Extra silage can be conserved if extra nitrogen is applied, and this can result in extra
milk provided that the extra feed is needed by the cows and by the system. However, with
present cost and prices (1 kg N costs $1; 1 kg MS worth $3.5), this strategy will result in only
small increases in profit.
The application of 428 kg N/ha produced an extra 3.3 t pasture DM/ha, which allowed
the conservation of an extra 1.6 t silage DM with 3.3 cows/ha (McGrath et al, 1998b). An
extra 69kg MS were produced per cow, with an extra 27 days of lactation (mainly due to
feeding of silage to allow drying-off to be delayed). However profit was not increased. A
combination of 100kg N/ha plus conservation of extra silage plus earlier calving, produced
small increases in milk per cow and per hectare (Thomson et al., 1991).

d - In-situ Conservation of Forage

Pasture can be stored in situ, without being harvested, by simply shutting the gate to
exclude grazing cattle, and allowing the herbage to accumulate until needed for grazing at a
later date. Autumn-saved pasture and foggage are examples of this (Wallace 1958, Holmes,
1989), and the practice is common in tropical systems (Butterworth, 1985). Prolonged in situ
storage is associated with increased senescence and other changes in composition and quality,
and these are larger if storage occurs in spring/summer when the grass is going through its
reproductive phase and temperatures are high, than if storage occurs during winter with lower
temperatures and light intensity (L’Huillier, 1988; Holmes, et al., 1992).
Pasture can be stored from late spring until summer by “deferred grazing” (McCallum
et al., 1991; L’Huillier, 1988), a low-cost technique which has some application in intensive
pastoral systems. The stored herbage is of relatively low quality, but it can contribute to
reseeding and to a fallowing effect on the soils.
Pasture can be stored until early spring by storage from autumn (Wallace, 1958) or by
gradual accumulation over the winter period (Bryant and MacDonald, 1983). In the latter
case, an extra 200 kg DM accumulated/ha in July to September resulted in the production of
an extra 3 to 8 kg milksolids/cow, or 12 to 24 kg milksolids/ha, in early lactation by highly-
stocked farmlets which experienced a pasture deficit in early lactation. These responses are
equivalent to very efficient conversion of the extra feed into milk in these intensive systems.
However the changes in herbage composition and in pasture structure which are
associated with winter storage of pasture (Holmes et al., 1992), can also cause subsequent
problems in pasture production and quality, and ultimately in milk production (Clark et al.,
1994). Accumulation of excessive herbage masses in early spring must be avoided.
Growth of grazeable crops (e.g. turnips; green-feed maize) on part of the farm
represent a form of in situ conservation of feed. These incur costs of growing the crop, but do
not suffer from deterioration in quality, and may also result in improved pastures
subsequently after reseeding back into grass.
Shorter term conservation is achieved through the use of rotational grazing with forage
being accumulated between grazings. This is one reason for the common practice of
lengthening rotations as the system is moving from a period of feed surplus to deficit, or fast
to slower pasture growth rates (Matthews 1994). In most instances this represents a feed
transfer and intake rationing strategy to reduce the impact of lower pasture growth rates rather
than the addition of extra feed into the system.
Long-term System Studies, With Imported Supplements

Pasture silage (5 kg DM/cow daily, with 10.8 MJ ME/kg DM) was fed for 30 days in
either spring, summer or autumn or at all 3 times (90 days) and these four systems were
compared with a fifth system which received no silage (Clark, 1993). All five systems carried
a relatively high stocking, 3.8 cows/Ha. Some results are shown in Table 2, relative to the
system that fed no silage.
Feeding silage in autumn also allowed the cows to be milked for an extra week, which
contributed to their bigger response. The farmlet on which silage was fed at all three times
also had an additional 450kg DM/Ha average herbage mass at the end of the lactation. The
high stocking rate used in this experiment ensured that pasture deficits did occur, but a rigid
design of this sort does not allow for the exact matching, in time and quantity, between
pasture deficits and silage fed which can be achieved on the best farms.
Maize silage was fed to cows, grazed at two stocking rates (MacDonald, 1999) and
some results for 3 years are shown in Table 3. An increase in feed demand, by increasing
stocking rate, with no increase in feed supply caused decreases in feed available per cow, and
consequent decreases in milk yield per cow and even per hectare, plus reduced fertility.
However, when the increase in feed demand was associated with an increase in feed supply
(as maize silage), the adverse effects on performance per cow were overcome, and milk
production per hectare was increased considerably.
In two other treatments, with 4.42 cows/Ha, the extra feed was supplied either as
maize grain or as a balanced ration including maize silage and concentrates. The responses
per cow and per hectare were greater than from maize silage, mainly due to the higher
concentrations of ME/kg DM in the other two feeds, but the responses were similar if
expressed per MJME.(Penno et al, 1998).
Of the three systems shown above, only that at 3.34 cows/Ha without maize silage and
that at 4.42 cows/Ha with maize silage are sustainable, and sensible. The response by the
whole system can be expressed as an extra 398 kg MS/ha, from a combination of an extra 5.8t
DM/ha plus an extra 1.08 cow/ha plus an extra 11 days in milk per cow.
In an earlier stage of the experiment, maize silage was also fed at the lower stocking
rate (3.34 cows/Ha), but the productivity of this system was limited by its low feed demand
(Penno et al, 1996).
Best responses from whole systems are usually obtained from several integrated inputs
that affect both feed demand and feed supply.
In another study, imported maize silage was fed in either Spring, Summer or Autumn
(Thomson et al, 1998). In both years the responses were lowest in summer, possibly due to its
relatively low protein concentration (see Stockdale 1995). When fed in Spring or in Autumn,
the response to maize silage was about 70g and 165g extra milksolids per kg maize silage DM
in Year 1 and Year 2 respectively, with much lower milk yields achieved in Year 2. There
were no obvious reasons for the differences between years; but they illustrate the risk attached
to the achievement of high responses (Thomson et al., 1998).
The latter study is being continued, with the feeding of maize silage targeted at
specific periods of pasture deficit, including deficits caused by early calving, late drying off at
a higher stocking rate.
A similar experiment to that described by Macdonald (1999), was carried out in
Victoria, Australia, but concentrates were fed instead of maize silage (Grainger, 1995; cited
by Stockdale et al., 1997). A combination of extra feed demand (higher stocking rate) and
extra feed supply, produced extra milk in all years, up to 1200 to 1800 kg milksolids/Ha with
3.5 to 4.7 cows/ha and 0.8 to 1.7 t concentrates fed/cow. However the extra concentrates
were profitable only in the 2 years when the concentrates cost 13c/kg, and not in the 3rd year
when they cost 24c/kg, with $5.4/kg received per kg milkfat.

Results from Commercial Farms

Data for the top quartile and the bottom quartile of farms (classified on the basis of
profit per hectare) are presented in Table 4 (Howse and Leslie, 1997). The top farms
successfully combined extra feed demand/ha (extra cows producing higher yields per cow)
with extra feed supplied in order to produce more milk and profit. Attrill and Miller (1996)
concluded, however, that the increased use of supplementary feeds on dairy farms in New
Zealand had, on average, increased production but not profitability.
In the traditional highly stocked grazing system in New Zealand supplementary
forages have been used either to enable the system to a) survive crisis situations (e.g.
droughts) and b) maintain the high stock numbers over the non-productive winter period
when pasture growth rates are low. For most seasonal supply dairyfarming systems in New
Zealand the summer is the most climatically variable time of the year. Research carried out in
the area of summer management (MacDonald 1987, Gray et al, 1992) has mostly focused on
survival tactics, with decision rules over the summer autumn period aimed at protecting the
new season often resulting in the farmer relenting control and short term production losses.
Increased profitability to added supplement is most likely to occur when it is used in a
more productive capacity during the lactation period (Brander and Matthews 1997, Matthews
1997) through either increasing the number of lactation days (Penno et al., 1999, 2000) or the
average production per lactation day (Cassells and Matthews 1995). This change is illustrated
by the change in total supplement use (forage supplements and off-farm grazing) in a high
producing dairy farm (Table 5) as the emphasis changed from high production per hectare
based on a high stocking rate to one based achieving a high production per hectare through
increased per cow performance (Cassells and Matthews 1995).
Brief details are given in Table 6 of two more extreme examples of farms using
supplementary feeds in very intensive pastoral systems in New Zealand. Again both systems
have combined extra feed demand plus extra feed supply to produce large quantities of milk
per cow and per hectare, profitably.

Summary of the Factors Affecting Short and Long-term Responses to Supplementary

Forages

Two simple lists (shown in Table 7) illustrate the size of probable responses to
supplementary forages and the factors that will contribute to them. The largest responses will
be achieved by cows and systems with the highest potential feed demands and milk
production, from high quality supplements given at the right times and in the right quantities.
Appropriate use of supplements in efficiently planned and integrated systems can
probably achieve the highest responses because of the “long-term” benefits of good feeding.
e.g. through effects on body condition as shown by Stockdale (1999) and sward conditions
(Benson and Matthews 2001).
This will involve the positive attributes of feed substitution or managed substitution.
In the efficient intensive grazing system, the substitution effect is used deliberately to “save”
pasture during a pasture deficit, while simultaneously maintaining the cow’s level of feeding.
In this case the substitution is done by the farm manager and not by the cow. For example,
the cow’s total required feeding level is 15 kg DM/day; with pasture only this would require a
daily allowance of about 35 kg DM/cow daily. During a period of slow pasture growth (e.g.
dry summer), the cows pasture intake can be reduced to 10 kg DM/day by reducing the
pasture allowance (by reducing the area grazed each day) to about 20 kg DM / cow daily
while simultaneously feeding 5 kg DM/cow daily as silage. This combination of reduced
pasture intake and consumption of supplementary feed can ensure that the cows’ total feeding
level, and the postgrazing residual herbage mass are both maintained.
The overall objectives of this managed substitution are to: -
• Maintain the cow’s total level of feeding.
• Decrease the daily rate at which pasture is consumed, so that it is equal to the daily rate
of pasture growth.
• Maintained the pre and post grazing herbage masses at the required, target, levels.
“Managed substitution” is illustrated by the experimental data for cows given either a
pasture allowance of 45-kg DM/cow daily with no supplements, or a pasture allowance of 30
kg DM/cow daily plus 8 kg DM/day as maize silage and cotton seed meal (Moran and
Stockdale, 1992). Both feeding combinations resulted in similar levels of intake, milk yield
and postgrazing residual herbage mass, and pasture consumption was reduced by the addition
of maize silage without increased wastage of pasture.
The relations between pasture allowance, pasture intake and residual herbage mass at
different levels of supplement intake, and the ability to manage substitution have not been
studied in detail. However, there will probably be a close relationship between total
allowance of feed (pasture plus supplement) and total feed intake (pasture plus supplement).
The increasing emphasis being given to supplement use in grazing systems will
require the establishment and management of sward conditions to enable pasture growth to be
maximised and used efficiently (Matthews 1995). An example of appropriate sward targets is
given in Table 7. There needs to be increased emphasis on pre-grazing and post-grazing
sward conditions rather than average pasture cover as these control both pasture quality and
animal intake. Phillips and Matthews (1994) suggested that in a dairy farm system there were
few occasions when average cover should be outside the range of 1900 to 2100 kg DM/ha
(Table 7). The net effect is to limit the ability to buffer differences in pasture growth and
animal demand through building up and running down cover. This means that average
pasture cover cannot act as a large feed reserve in the grazing system. To do this would then
force pre-grazing levels outside the desirable range as average cover increased and also post-
grazing residuals outside this desirable range as pasture cover was decreased.
While manipulating grazing rotation lengths has been one of the key grazing
management decisions made by New Zealand farmers (Figure 4) Matthews (1997) suggests
that as farmers make increased use of supplements to buffer the variability in pasture growth
rate rotation length will become less important for transferring feed but geared more to the
replacement rate of pasture in the sward. It is also suggested that stock numbers rather than
intake (production) per animal will in future be used to buffer the system along with an
increased use of supplementary forages.
To gain maximum benefit (both production and financial) supplementary feed will be
used not only to meet animal requirements on the short term but also for the maintenance of
appropriate sward conditions to ensure long term production responses are achieved.

References

Attrill, B. and Miller G. (1996). Ruakura Farmers Conference.

Bramley, A.J., Dodd F.H., Mein G.A. and Bramley J.A. (1992). Machine Milking and
Lactation. Chapter 8, Insight Books, Berkshire, England.
Brander, J.S. and Matthews P.N.P. (1997). Dairyfarming Annual, Massey University, 49:
77-84.
Bryant, A.M. and Trigg T.E. (1982). In: Dairy Production from Pasture. Edited by K.L.
MacMillan and V.K. Taufa; New Zealand Society of Animal Production.
Bryant, A.M. and MacDonald. (1983). Proceedings of New Zealand Society of Animal
Production, 43: 93-95.
Brookes, I.M. (1996). Dairyfarming Annual, Massey University, 48: 77-84.
Broster, W.H. and Broster V.J. (1984). Journal of Dairy Research, 51: 149-196.
Burnham, D., Daysh C., Purchas G., Holmes C.W. and Brookes I.M. (1995).
Dairyfarming Annual, Massey University, 28-31.
Butterworth, M.H. (1985). Beef cattle nutrition and tropical pastures. Longmans, London.
Campbell, A.G., Clayton D.G. and MacDonald K.A. (1978). AgLink No. 166; Ministry of
Agriculture, New Zealand.
Clark, D.A. (1993). Ruakura Farmers Conference, 41-46.
Clark, D.A., Carter W., Walsh B., Clarkson F.H. and Waugh C.D. (1994). Proceedings of
New Zealand Grassland Association, 56: 55-59.
Cassells J.B. and Matthews P.N.P. (1995). Dairyfarming Annual, Massey University, 47:
40-45.
Dalley, D.E., Roche J.R., Grainger C. and Moate P.J. (1999). Australian Journal of
Experimental Agriculture, 39: 923-931.
Doyle, C.J. and Phillips R.H. (1987). Grass and Forage Science, 42: 411-428.
Ferris, C.P., Gordon F.G., Patterson D.C., Mayne C.S. and Kilpatrick D.J. (1999).
Journal of Agricultural Science, Cambridge, 132: 467-481.
Fulkerson, W.J. (1997). Report of Dairy Research Group, New South Wales Agriculture,
29-36.
Garcia, S.C. and Holmes C.W. (1999). New Zealand Journal of Agricultural Research, 42:
347-362.
Garcia, S.C., Cayzer F.J., Holmes C.W. and MacDonald A. (1998). Dairyfarming Annual,
Massey University, 50: 207-218.
Garnsworth, P.C. and Webb R. (1999). In: Recent Advances in Animal Nutrition,
Nottingham University. Pp 39-52.
Grainger, C. (1990). Australian Journal of Experimental Agriculture, 30: 495-501.
Grainger, C. and Matthews, G.L. (1989). Australian Journal of Experimental Agriculture,
29: 355-360.
Grainger, C. and Wilhelms. (1982). Australian Journal of Experimental Agriculture, 19: 395
Gray D.I., Lynch G.A., Lochhart J.C., Parker W.J., Todd E.G. and Brookes I.M. (1992)
Proceedings of New Zealand Animal Production Society, 52: 11-12.
Harris, S.L., Clark D.A., Waugh C.D., Copeman P.J. and Napper A.R. (1998).
Proceedings of New Zealand Animal Production Society, 58: 121-124.
Holmes, W. (1989). Grass; its production and utilisation. Blackwell Publications.
Holmes, C.W., Hoogendoorn C.J., Ryan M.P. and Chu A.C.P. (1992). Grass and Forage
Science, 47: 309-315.
Howse, S.W., Isherwood P., Miller D.B., Wells J.L., Riddick C.M., Thomson N.A. and
Clark D.A. (1996). Proceedings of New Zealand Grassland Association, 57: 157-160.
Howse, S. and Leslie M. (1997). Ruakura Farmers Conference, 10-19.
Jacobs, J.L., Rigby S.E., McKenzie Ward G.N. and Kearney G. (1998). Australian
Journal of experimental agriculture, 38: 131-138.
Kolver, E.S., Penno J.W., MacDonald K.A., McGrath J.M. and Carter W.A. (1999).
Proceedings of New Zealand Grassland Association, 61: 139-145.
Leaver, J.D. (1985). In: Grazing, Edited by J. Frame; British Grassland Society; Occasional
Symposium, No. 19, 79-88.
L’Huillier, P. (1988). Ruakura Farmers Conference, 19-26.
Matthews, P.N.P. (1994). Dairyfarming Annual, Massey University, 46: 143-151.
Matthews, P.N.P. (1995). Dairyfarming Annual, Massey University, 47: 171-174.
Matthews, P.N.P. (1997). Dairyfarming Annual, Massey University, 49: 67-71.
Mayne, C.S. (1991). In: Management Issues for the Grassland Farmer in the 1990’s. Ed.
C.S. Mayne, British Grassland Society Occasional Symposium, No. 25, 55-74.
McCallum, D.A., Thomson N.A. and Judd T.G. (1991). Proceedings of New Zealand
Grassland Association, 33: 79-83.
McGrath, J.M., Penno J.W., Davis K.L. and Wrenn R. (1998a). Proceedings of New
Zealand Grassland Association, 60: 259-264.
McGrath, J.M., Penno J.W., MacDonald K.A. and Carter W.A. (1998b). Proceedings of
New Zealand Society of Animal Production, 58: 117-120.
MacDonald, K.A. (1999). Ruakura Farmers Conference, 78-87.
MacDonald, K.A. and Penno J.W. (1998). Proceedings of New Zealand Society of Animal
Production, 58: 132-135.
MacDonald, K.A., Penno J.W., Kolver E.S., Carter W.A. and Lancaster J.A. (1998).
Proceedings of New Zealand Society of Animal Production, 58: 102-105.
MacDonald, K.A., Nicolas P., Kidd J., Penno J. and Napper A. (2000). Ruakura Farmers
Conference, 88-89.
Moller, S., Parker W.J. and Edwards N.J. (1996). Proceedings New Zealand Grassland
Association, 57: 173-177.
Moran, J.B. and Stockdale C.R. (1992). Australian Journal of Experimental Agriculture,
52: 279-285.
Moran, J.B. (1992). Australian Journal of Experimental Agriculture, 32: 293.
Penno, J.W., Holmes C.W., MacDonald K.A. and Walsh B.J. (1998). Proceedings of New
Zealand Society of Animal Production, 58: 113-116.
Penno, J.W., MacDonald K.A. and Bryant A.M. (1996). Ruakura Farmers Conference, 11-
19.
Phillips, C.J.C. (1988). Grass and Forage Science, 43: 215-230.
Phillips, C.J.C. and Leaver J.D. (1985). Grass and forage Science, 40: 193-199.
Phillips, R.M. and Matthews P.N.P. (1994). Dairyfarming Annual, Massey University,
46:153-157.
Rogers, G.L. (1984). In: “Silage in the 80’s”. Armidale, N.S.W. Ed. By T.J. Kempton, A.G.
Kaiser and T.E. Trigg. 331-351.
Rogers, G.L. (1985). In: The Challenge: Efficient Dairy Production. Ed. By T.I Phillips,
Australian and New Zealand Societies of Animal Production, 85-108.
Sarker, A.B. and Holmes W. (1974). Journal of the British Grassland Society, 29: 141-143.
Small, J.C. and Gordon F.G. (1988). Journal of Agricultural Science (Cambridge), 111:
369.
Stockdale, C.R. and Trigg T.E. (1989). Australian Journal of Experimental Agriculture, 29:
601-611.
Stockdale, C.R., Dellow D.W., Grainger C., Dalley D. and Moate P.J. Supplements for
Dairy Production in Victoria. Dairy Research and Development Corporation. Victoria.
Stockdale, C.R. (1995). Australian Journal of Experimental Agriculture, 35: 19-26.
Stockdale, C.R. (1999). Australian Journal of Experimental Agriculture, 39: 803-809.
Thomas, C. and Thomas P.C. (1989). In: “Silage for Milk Production”; Ed. C.S. Mayne.
British Grassland Society Occasional Symposium, No. 23, 49-60.
Thomas, G.W. and Matthews G.L. (1991). Australian Journal of Experimental Agriculture,
31: 195-203.
Thonson, N.A. (1997). Ruakura Farmers Conference, 106-113.
Thomson, N.A., Logan J.F. and McCallum D.A. (1984). Proceedings of New Society of
Animal Production, 44: 67-70.
Thomson, N.A., McCallum D.A. and Prestidge R.W. (1989). Ruakura Farmers
Conference, 50-56.
Thomson, N.A., Roberts A.H.C., Judd T.G. and Clough J.C. (1991). Proceedings of New
Zealand Grassland Association, 53: 85-90.
Thomson, N.A., Davis K.L., McGrath J.M., Hainsworth R.J. and Clough C.J. (1998).
Proceedings of New Zealand Grassland Association, 60: 51-55.
Thomson, N.A., Exton P.R., McLean N.R. and Dawson J.E. (1997). Proceedings of New
Zealand Society of Animal Production, 57: 165-168.
Thompson, N., Keiller P.R. and Yates C.W. (1989). Grass and Forage Science, 44: 195-
203.
Waugh, T. (1997). Dairyfarming Annual, Massey University, 49: 62-66.
Wallace, L.R. (1958). Dairyfarming Annual, Massey University, 30: 9-20.
Watson, L.A.( 1999). Thesis for Master of Applied Science, Massey University.

Table 1 - The effect of pasture allowance on the response to additional supplement (Phillips
1988)

Silage Fed Substitution Rate Extra Milksolids

(kg DM/cow/day) (gm/cow/day)

Fed with generous

pasture 5.0 1.1 - 31
Fed with restricted
pasture 4.9 0.3 + 105

Table 2 - Production responses (kg milksolids/cow) to pasture silage fed at different stages
of lactation Clark, 1993)

Spring Summer Autumn Spring, Summer &

Autumn

During the feeding period 5.7 3.6 12.0

During the whole lactation 10.1 10.0 28

Responses to the extra feed

(gm extra MS/kg extra DM) 69 67 79 62
Table 3 - Systems response to increased stocking rate with or without the addition of maize
silage (MacDonald, 1999)

Stocking Rate (Cows/Ha)

3.34 4.42 4.42

Maize silage fed (t DM/Ha) 0 0 5.8

Days in milk: 266 217 277

Milk produced (kg MS)

Per cow:: 362 269 364
Per hectare: 1208 1190 1606

Percentage of cows anoestrous at start of mating: 11 35 13

Table 4 - The use of supplementary feeds on commercial dairy farms in New Zealand
(Howse and Leslie, 1997)

Quartile
Top Bottom

Profit ($/ha) 2280 1130

Cows/ha 3.3 2.8
Milksolids produced
kg/cow 334 267
kg/ha 1100 742
Feed purchased (t DM/ha): 1.1 0.3
Table 5 - The effect of changes in stocking rate and supplement use on per cow and per
hectare production (adapted from Cassells and Matthews 1995)

1980-1988 1989-1993 1994-1995

Stocking rate (cows/ha) 3.93 3.47 3.03

Production
265 318 409
(kg milksolids/cow/year)
(kg milksolids/ha/year) 1035 1107 1239

Feed Inputs
Supplement (kg DM/ha/year) 1433 551 2455
Grazing (kg DM/ha/year) 1060 605 924
Total (kg DM/ha/year) 2493 1156 3379

Table 6 - An example of two New Zealand dairy farms using high levels of supplements

J and S Van G and F

Der Poel Edgecombe

Cows/ha 4.15 4.1

Milksolids produced
kg/cow 430 540 (in 315 days)
kg/ha 1790 2200
Supplements fed t DM/ha 4.9 6.0
Main Supplement used Maize silage Cannery byproducts
Table 7 - Factors affecting the likely level of response to additional feeds
(a) Immediate, short-term responses:

Low Response High Response Very High Response

30 g MS/kg DM 60 g MS/kg DM 100 g MS/kg DM

Moderate initial level of Low initial level of feeding Very low initial level
feeding of feeding

Cows with low potential Cows with high potential feed ONLY by high
feed demand and milk yield. demand and milk yield. potential cows in early
lactation, and high
e.g. mid to late lactation e.g. early lactation quality supplement
e.g. moderate genetic merit e.g. high genetic merit
e.g. some disease problems e.g. healthy cows
e.g. thin cows e.g. adequate body condition

Low quality supplement High quality supplement

(<9 MJME/kg DM) (>11 MJME/kg DM)

Grazing on high quality Grazing on low quality

pastures. pastures.

(b) Total, long-term response (including carry-over effect)

Low Response 50 g MS/kg DM High Response 100 g MS/kg DM

Low immediate response (30 g MS/kg DM) High immediate response (60 g MS/kg DM)

Moderate feed demand /ha High feed demand/ha

No extra days in milk achieved from extra Extra days in milk achieved from the extra
feed feed (from earlier calving and/or later dry-off)
Table 8 - Seasonal pasture targets for a seasonal supply dairy farm (adapted from Phillips and
Matthews 1994)

Season Average Pasture Pre-grazing herbage Post-grazing herbage

Cover mass (kg DM/ha) mass (kg DM/ha)
(kg DM/ha)

Late-autumn 1750-2000 2400-2700 1200-1400

Winter 1900-2100 2500-2700 800-1000
Early-spring 1900-2000 2500-2700 1300-1400
Late-spring 2000-2200 2500-2700 1500-1600
Rates of pasture growth or consumption (kg DM/ha daily)

60
Silage Conserved

50
!! Pasture growth
"- - -" Feed required
40

30

240 days in milk

20
300 kg MS/cow
750 kg MS/ha

10

Calved Conceived Dried-off

% cows in the herd

Aug Oct Dec Feb Apr Jun

Figure 1 - Moist Summer/Cool Winter: 2.5 cows/ha : No Supplements

Rates of pasture growth or consumption (kg DM/ha daily)

60

Supplements fed !! Pasture growth

50 "- - -" Feed required

Supplements fed
40

30

240 days in milk

20 300 kg MS/cow
900 kg MS/ha

10

Calved Conceived Dried-off

% cows in the herd

Aug Oct Dec Feb Apr Jun

Figure 2 - Moist Summer/Cool Winter: 3 cows/ha: Supplements Fed Spring and Autumn
 Supplementary Feed
Wasted
Reduced pasture intake
IMMEDIATE
EFFECTS Pasture Spared Increased Total Intake

Partitioned
Extra liveweight Extra Milk
Certain Fatter Condition

Certai
Higher Herbage Mass

LONG-TERM
EFFECTS More Waste Certai

Extra Pasture

Conserved Increased Improved

Intake Fertility

Higher Yields per day Longer Lactation’s Possible

Extra Milk

Possible

TOTAL EFFECTS

Extra Improved
Milk Fertility?

Figure 3 - Immediate and long-term, or carry-over, responses of cows to supplementary feeding

(Modified from Brookes 1996)
 Conservation

Increase

Increase

Pasture Animal Animal

Average Pasture
Production
Growth Cover
Intake

Decrease
Decrease
Decay

Supplements

Figure 4 - Feed Flow Model of a Grazing System (adapted from Matthews 1994)