Modelling the impact of increasing supplementary feed allowance on predicted sheep enterprise production, profit and financial risk across southern Australia
A. L. Bates
A * , S. M. Robertson B C , S. R. McGrath B C , M. B. Allworth B C and G. Refshauge D
A
B
C
D
Abstract
Supplementary feeding may alter sheep enterprise production and profit margin, but use may vary across regions, sheep breeds and mating seasons. Supplementary feeding is a means of ensuring adequate nutrition but increases operating costs. Modelling has previously indicated the most profitable sheep enterprises optimise stocking rate and target lamb production, whereas those that minimise supplementary feeding incur the least financial risk.
To explore the impact of increasing supplementary feed allowance on production, profit and financial risk.
Seventy-two sheep farm enterprises were simulated across eight southern Australian locations, including three breeds and three mating seasons. For each enterprise a low grain allowance (LGA) of 30 kg/head.year (threshold used in previous modelling) was compared to a high grain allowance (HGA) of 35 kg/head.year and 42 kg/head.year for Merino and non-Merino ewes (current industry recommendations), respectively. The financial risk of each enterprise was determined via Conditional Value at Risk of gross margins over 30 years, exploring downside risk in the worst 20% of scenarios.
A HGA increased production and profit in 32% of farm enterprises, but financial risk was often increased. Merino enterprises were generally the most profitable, least risky and consumed the greatest amount of supplementary feed, followed by Composite and then Maternal enterprises. Summer and autumn mating was often most profitable, but high supplement consumption in autumn-mated enterprises increased financial risk.
Increasing supplementary feeding may improve production and profit but may also increase financial risk using the parameters examined.
Producers may be able to improve the production, profit and financial risk of an enterprise through increased supplementary feeding, but this will be dependent on breed, input costs, commodity prices and location.
Keywords: AusFarm, composite, grain, management, maternal, Merino, reproduction, sheep production, supplementation, time of mating.
Introduction
Supplementary feeding sheep can support production when feed resources are limiting. This may occur routinely, or during periods of extreme climatic conditions, such as drought. Although supplementary feeding may increase production, the compromise of investment cost versus return may be a barrier to routine use.
Computer modelling is a useful tool for exploring the economic impact of different management practices, although reliant on the user’s inputs for accuracy (Young et al. 2022). Previous modelling suggests the most profitable sheep production systems in southern Australia either optimised stocking rate based on the efficiency of pasture and supplement use, or targeted lamb production (Warn et al. 2006; Kopke et al. 2008; Casburn 2010). Increasing stocking rate risks low pasture availability during summer and autumn and a reliance on supplement feeding. However, grain feeding ewes to achieve the optimum liveweight profile across gestation maximised whole-farm profit in Merino enterprises (Young et al. 2011).
A limitation of previous modelling work is the equal allocation of feed resources to different sheep breeds (i.e. up to 30 kg/head.year for all ewes (Warn et al. 2006)). However the potential intake of non-Merino ewes is suggested to be 25% greater than that of Merino ewes (Blumer et al. 2019). This is reflected in the New South Wales (NSW) Department of Primary Industries (DPI) livestock gross margin budgets, allowing 35 kg/head.year for Merino and 42 kg/head.year for non-Merino ewes (NSW DPI 2021). The current study endeavoured to explore the impost of greater supplement provision on enterprise production, profit and financial risk using updated commodity prices. This was extended to explore the impact of sheep breed and mating season as there are few studies in this space (Fogarty and Mulholland 2014; Robertson et al. 2014). The interactions of supplement provision, breed and mating season were explored using computer simulation modelling.
Materials and methods
Model design
Seventy-two farm enterprises were simulated for eight locations: Bookham, Bungarby, Condobolin, Glen Innes, Narrandera, and Trangie in NSW, Hamilton in Victoria (Vic) and Keith in South Australia (SA). For each location, three ewe breeds (Composite, i.e. multiple cross ewes; Maternal, i.e. first-cross ewes of Merino origin; and Merino ewes) were compared across three mating seasons (spring, summer and autumn). Two supplement levels were applied to each farm enterprise: a high grain allowance (HGA, derived from NSW DPI (2021)) and a low grain allowance (LGA, derived from Warn et al. (2006)). This resulted in 72 LGA and HGA enterprises for production profit and financial risk comparison (total of 144 simulations).
Modelling simulations were developed using the AusFarm modelling software (version 1.5.3) (Moore et al. 2007). Data from simulations run over a 30-year period were used to explore the average optimum stocking rate and impact of environmental variation (including pasture growth and supplement requirements detailed below as sustainability indices) on the enterprise. Any anomalies related to the model initialising were removed by discarding data from an initial 12-month period.
Farm system characteristics
Each simulation consisted of a 2000-hectare (ha) farm with 16 paddocks allocated to cropping rotations and/or pasture production (Table 1). At each location, soil and pasture types and cropping and pasture rotations were selected based on published data and in consultation with industry experts (Supplementary Table S1). A formal pasture validation was not performed for monthly pasture growth. The simulated pasture growth for each location was determined by the authors to be appropriate.
| Location | Latitude and longitude | Annual rainfall (model output) (mm) | Majority rainfall months (>40 mm) | Mean annual max temperature | Mean annual min temperature | Mean annual rainfall (mm) | Mean monthly rainfall | BOM weather station | |
|---|---|---|---|---|---|---|---|---|---|
| NSW | |||||||||
| Bookham | −34.86, 148.61 | 743.5 | Jun–Feb | 20.8 | 7.5 | 687.9 | 57.3 | Yass (Linton Hostel) | |
| Bungarby | −36.75, 148.98 | 536.1 | Jun, Oct–Mar | 17.9 | 5.2 | 570.0 | 47.5 | Bombala AWS | |
| Condobolin | −33.09, 147.15 | 446.1 | Jun, Nov–Dec, Feb | 24.9 | 10.6 | 438.7 | 36.6 | Condobolin Ag Research Stn | |
| Glen Innes | −29.74, 151.74 | 807.6 | Oct–Mar A | 20.1 | 7.4 | 813.5 | 67.8 | Glen Innes Ag Research Stn | |
| Narrandera | −34.75, 146.56 | 402.6 | Jun, Dec | 24.5 | 10.2 | 427.1 | 35.6 | Narrandera Airport AWS | |
| Trangie | −32.03, 147.98 | 482.6 | Oct–Mar | 25.3 | 11.4 | 473.0 | 39.4 | Trangie Research Stn AWS | |
| Vic | |||||||||
| Hamilton | −37.74, 142.02 | 615.8 | May–Dec | 19.3 | 7.7 | 604.0 | 50.3 | Hamilton Airport | |
| SA | |||||||||
| Keith | −36.10, 140.36 | 432.3 | May–Sept | 22.5 | 9.3 | 432.3 | 36.0 | Keith | |
Livestock enterprise
The three sheep breeds, Composite, Maternal and Merino, were developed using data collected from commercial field observations and in conjunction with advice from industry experts (Bates et al. 2022). Composite ewes were based on a Dorset ewe with a 70 kg standard reference weight (SRW), potential fleece weight of 2.5 kg, fleece yield of 70% and maximum fibre diameter of 27 microns. Maternal ewes were based on a Border Leicester cross Merino ewe with a 70 kg SRW, potential fleece weight of 4 kg, fleece yield of 70% and maximum fibre diameter of 27 micron. Two Merino ewe phenotypes were used, a small frame-size with a 50 kg SRW, potential fleece weight of 4.5 kg, fleece yield of 70% and maximum fibre diameter of 19 microns. The small Merino was used in Bookham, Glen Innes and Hamilton simulations. The second Merino had a 60 kg SRW, potential fleece weight of 5 kg, fleece yield of 70% and maximum fibre diameter of 20 microns. The larger Merino was used in Bungarby, Condobolin, Keith, Narrandera and Trangie simulations. All ewes and weaned lambs had an annual mortality rate of 4%.
The reproductive information for mature ewes was derived from pregnancy scanning rates at different mating body condition scores (BCS), collected at quarter scores (Bates et al. 2022). Spring-mated Composite flock data was collected from highly fecund ewes (http://www.multimeat.com.au/ (accessed on 1 August 2022)) that also received a melatonin-based hormone product (Regulin®, Ceva Animal Health Pty Ltd., Glenorie, NSW, Australia) (Bates et al. 2022). From this dataset, three unique pregnancy scanning outcomes based on mating BCS were used for each breed and mating season (Table 2). These values were determined by the authors to result in a practically different pregnancy scanning outcome for mature ewes. Simulated reproductive rates of maiden ewes, 7–9 month old non-Merino ewes and 18 month old Merino ewes, were calculated as per Freer et al. (2012).
| Pregnancy rate A (%) | ||||||
|---|---|---|---|---|---|---|
| Spring BCS | Summer BCS | Autumn BCS | ||||
| Composite | ||||||
| ≤2.5 | 89 | ≤2.5 | 90 | ≤2.5 | 92 | |
| >2.5 ≤ 3 | 97 | >2.5 ≤ 4.5 | 94 | >2.5 ≤ 4 | 96 | |
| >3 | 93 | >4.5 | 92 | >4 | 95 | |
| Maternal | ||||||
| ≤2.5 | 71 | ≤2.5 | 85 | ≤3 | 77 | |
| >2.5 ≤ 3 | 88 | >2.5 ≤ 3 | 88 | >3 ≤ 4 | 87 | |
| >3 | 87 | >3 | 92 | >4 | 83 | |
| Merino | ||||||
| ≤2 | 89 | ≤2.5 | 80 | ≤2 | 91 | |
| >2 ≤ 3 | 90 | >2.5 ≤ 4 | 86 | >2 ≤ 3.5 | 96 | |
| >3 | 82 | >4 | 85 | >3.5 | 95 | |
| Scanning rate B | ||||||
|---|---|---|---|---|---|---|
| Spring BCS | Summer BCS | Autumn BCS | ||||
| Composite | ||||||
| ≤2.5 | 1.71 | ≤2.5 | 1.43 | ≤2.5 | 1.63 | |
| >2.5 ≤ 3 | 1.84 | >2.5 ≤ 4.5 | 1.61 | >2.5 ≤ 4 | 1.75 | |
| >3 | 1.67 | >4.5 | 1.63 | >4 | 1.77 | |
| Maternal | ||||||
| ≤2.5 | 1.14 | ≤2.5 | 1.28 | ≤3 | 1.30 | |
| >2.5 ≤ 3 | 1.33 | >2.5 ≤ 3 | 1.44 | >3 ≤ 4 | 1.58 | |
| >3 | 1.46 | >3 | 1.53 | >4 | 1.50 | |
| Merino | ||||||
| ≤2 | 1.25 | ≤2.5 | 1.16 | ≤2 | 1.28 | |
| >2 ≤ 3 | 1.34 | >2.5 ≤ 4 | 1.33 | >2 ≤ 3.5 | 1.39 | |
| >3 | 1.12 | >4 | 1.40 | >3.5 | 1.49 | |
A survey of southern Australian sheep producers informed the livestock management practices (Bates et al. 2023). All breeds were commonly mated for three oestrus cycles and mating was simulated on 7th November in spring, 17th January in summer, and 14th March in autumn (Freer et al. 2012; Bates et al. 2023). Maternal sheep producers primarily purchased replacement ewe lambs each year, whereas Composite and Merino sheep enterprises retained ewe lambs as replacement ewes (Bates et al. 2023). Female lambs’ surplus to replacement requirements were sold. Simulated management practices were developed to reflect this.
Lamb sale was flexible to best reflect seasonal and market conditions (Bates et al. 2023). The lamb sale period commenced after weaning (aged 8–13 weeks) and lambs were sold if they reached the target sale weight (46 kg liveweight), if growth reduced to less than 50 g/day (based on 3-day average), or at the end of the six-month sale period. The sale of cast for age ewes (CFA, at six years of age) occurred at weaning to emulate the common producer practice of retaining mature ewes if not pregnant for a ‘second chance’ in the subsequent year (Bates et al. 2023).
Ewe BCS (1–5) management was derived from the average BCS reported as optimum at mating and lambing (Bates et al. 2023), which coincides with previously modelled profitability and welfare ranges (Kingwell 2002). This BCS management was simulated via minimum BCS targets across the reproductive cycle of the ewe through access to feed resources (pasture, crop stubbles, or supplement).
Composite ewes did not receive short-term supplementary feeding prior to mating in autumn and spring and had a BCS 2.5 target between weaning and mating. Composite ewes carrying a single fetus had a BCS 2.5 target between mating and late pregnancy (30 days before lambing), after which the target increased in BCS 3.5. Composite ewes carrying multiple fetuses had a BCS 3.0 target between mating and late pregnancy, after which the target increased to BCS 3.5. When mated in summer, the short-term supplementary feeding period was 28 days pre-mating with a BCS 3.0 target.
Maternal ewes received short-term supplementary feeding for 56 days prior to mating and had a target of BCS 3.0. Maternal ewes carrying a single fetus had a BCS 2.5 target between mating and late pregnancy, after which the target increased to BCS 3.0. Maternal ewes carrying multiple fetuses had a BCS 3.0 target between mating and lambing.
Merino ewes received short-term supplementary feeding for 28 days prior to mating and had a target of BCS 3.5. Merino ewes carrying a single fetus had a BCS 2.5 target between mating and late pregnancy, after which the target increased in BCS 3.5. Merino ewes carrying multiple fetuses had a BCS 3.0 target between mating and late pregnancy, after which the target increased to BCS 3.5. The target between lambing and weaning and for all non-pregnant ewes of all breeds was BCS 2.5. The target for lambs of all breeds post weaning was BCS 2.5.
Grazing management
Pasture management and livestock rotations were based on a series of rules to ensure optimum pasture and livestock growth was achieved. If necessary, animals were rotated throughout the farming system every 7 days. Ewes with multiple fetuses and growing lambs had priority and were moved to the best available pasture. Pregnant ewes had the next highest priority (i.e. allocated the next best pasture) followed by non-pregnant ewes. If crops were present, sheep were able to graze crop stubbles if cover was above 50% for a maximum of 135 consecutive days. Supplementary feed was supplied in the paddock if there was adequate groundcover (≥70%) and sheep were below a minimum BCS target. If groundcover was inadequate (<70%), sheep were moved into one of three containment feedlots. Sheep were assigned to feedlots based on class to mitigate overgrazing (DEDJTR 2018): mature ewes, maiden ewes and lambs. Barley grain (13.7 MJ/kg DM, 12% CP) was utilised as supplementary feed in all instances. Sheep were removed from the feedlot when groundcover increased 2% above the groundcover threshold (i.e. 72%).
Two sustainability indices, one for supplementary feed and one for groundcover, were implemented to ensure supplement consumption and groundcover remained within realistic bounds. As the models were unique, the interaction between the environment, including weather and pasture growth, and animal pressure at each location varied with stocking rate. Models were run at varying stocking rates for a 30-year period and the attainment of the described sustainability indices determined after the fact. The 30-year trial period was blocked into three 10-year blocks. In each block, the LGA and HGA levels could be surpassed the maximum number of times (described below), with no upper/lower limit. If these levels were surpassed, the next highest stocking rate was investigated, and so on. This allowed the optimum average stocking rate, in conjunction with resource use, to be determined. The provision to exceed these thresholds was implemented to mimic the feeding required to overcome an extreme weather event, such as drought.
For supplement consumption, the LGA simulations allowed all ewes to consume up to a threshold 30 kg/head.year in all years, and in four out of 10 years to consume more than this threshold (derived from Warn et al. (2006)). The HGA simulations allowed Merino ewes to consume up to a threshold 35 kg/head.year in all years and non-Merino ewes up to 42 kg/head.year in all years (derived from NSW DPI (2021)), and in four out of 10 years for all ewes to consume more than these threshold levels. As it was not possible to separate unweaned lambs and ewes, the supplement level applied to all ewes (with or without lambs). The supplement threshold did not apply to weaned lambs (until the time of sale) or ewe lambs (if retained for replacement). The threshold level for ewes may also not be reached each year (i.e. less than, or up to, the threshold level was not consumed). The sustainability indices for groundcover allowance applied to both LGA and HGA simulations and allowed the available pasture from 1st January to 30th April each year to fall below 800 kg DM/ha in two out of 10 years (derived from Warn et al. (2006)).
Financial information
Income and expenses pertaining solely to the sheep enterprise were used to determine the gross margins (i.e. overall profitability) and were based primarily on the NSW DPI (2021) sheep gross margins (April–September 2020) (File S1 and S2). Annual pasture maintenance was A$38/pasture ha. A value of A$315/t for supplementary grain was used for all locations. Rams were mated at 1.5% of the breeding ewe flock (Making More From Sheep 2008) and each year 20% was replaced and sold for A$160.80/head. Replacement Composite and Maternal rams cost A$1100.00/head and Merino rams A$1800.00/head. A skin price of A$4.52 for lambs and A$5.62 for mature sheep was used (Meat and Livestock Australia Limited 2022a, 2022b). An additional expense of A$7/ewe was included in spring-mated Composite simulations to account for hormone use pre-mating. Composite and Maternal CFA ewes were sold for AUD $6.26/kg carcass weight (cwt) and Merino CFA ewes for A$6.33/kg cwt. Composite and Maternal lamb price for <21 kg carcass was A$7.86/kg cwt and >21 kg carcass was A$8.20/kg cwt. Merino lamb price for <20 kg carcass was A$7.50/kg cwt and >20 kg carcass was A$7.55/kg cwt. Composite and Maternal wool price for <28 micron was A$4.88/kg clean and >28 micron was A$4.20/kg clean. Merino wool price for <19 micron was A$14.32/kg clean and >19 micron was A$12.39/kg clean.
Calculations
Stocking rate was determined on a per pasture hectare basis as sheep had access to pasture paddocks year-round in all enterprises. A set stocking rate was implemented across the 30-year trial (i.e. stocking rate did not vary each year). The average optimum stocking rate at mating (average number of ewes mated per pasture hectare per year) over the 30-year trial was determined for LGA and HGA by considering supplementary feed allowance, pasture availability and/or if maximum average gross margin was achieved (i.e. the maximum gross margin was able to be captured even if stocking rate could be increased without violating the sustainability indices). To be included in analysis, an enterprise must have met both sustainability indices, which may be met simultaneously. The number of ewes mated (head) was converted to DSE in the month of mating (DSE being a 50 kg dry sheep equivalent) to standardise reporting across the three breeds (Table 3) and was reported on a per pasture hectare basis.
| Mating season | Grain allowanceA | Composite | Maternal | Merino | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Spring | Summer | Autumn | Spring | Summer | Autumn | Spring | Summer | Autumn | ||||||||||||
| per ha | DSE | ewes | DSE | ewes | DSE | ewes | DSE | ewes | DSE | ewes | DSE | ewes | DSE | ewes | DSE | ewes | DSE | ewes | ||
| NSW | ||||||||||||||||||||
| Bookham | LGA | 8.2 | 5.4 | 10.0 | 6.9 | 10.8 | 8.9 | 9.1 | 5.9 | 10.6 | 7.0 | 10.1 | 7.0 | 8.8 | 7.5 | 10.9 | 6.9 | 10.0 | 8.9 | |
| HGA | 8.8 | 5.9 | 11.5 | 8.4 | 10.8 | 8.9 | 9.9 | 6.4 | 11.8 | 8.2 | 10.1 | 7.0 | 9.4 | 8.0 | 11.3 | 7.3 | 10.8 | 9.7 | ||
| Bungarby | LGA | – | – | – | – | – | – | – | – | – | – | – | – | – | – | 3.2 | 2.4 | 1.6 | 1.2 | |
| HGA | – | – | – | – | – | – | – | – | – | – | – | – | – | – | 3.6 | 2.8 | 3.6 | 2.8 | ||
| Condobolin | LGA | 1.7 | 1.0 | 1.6 | 1.0 | 1.5 | 1.0 | 1.9 | 1.2 | 2.8 | 1.8 | 1.8 | 1.2 | 1.2 | 0.8 | 1.8 | 1.2 | 1.2 | 0.8 | |
| HGA | 1.7 | 1.0 | 2.3 | 1.5 | 3.9 | 2.5 | 1.9 | 1.2 | 3.7 | 2.3 | 3.6 | 2.3 | 1.2 | 0.8 | 1.8 | 1.2 | 2.9 | 2.0 | ||
| Glen Innes | LGA | 11.8 | 7.4 | 11.4 | 6.9 | 12.9 | 8.4 | 11.0 | 7.0 | 11.0 | 7.0 | 12.3 | 8.2 | 10.8 | 9.3 | 10.8 | 9.3 | 12.9 | 11.3 | |
| HGA | 11.8 | 7.4 | 11.4 | 6.9 | 12.9 | 8.4 | 11.0 | 7.0 | 11.0 | 7.0 | 12.3 | 8.2 | 10.8 | 9.3 | 10.8 | 9.3 | 12.9 | 11.3 | ||
| Narrandera | LGA | 1.7 | 1.0 | 2.4 | 1.5 | 2.3 | 1.5 | 1.8 | 1.2 | 1.8 | 1.2 | 1.8 | 1.2 | 2.3 | 1.6 | 2.3 | 1.6 | 1.7 | 1.2 | |
| HGA | 2.5 | 1.5 | 2.4 | 1.5 | 2.3 | 1.5 | 1.8 | 1.2 | 1.8 | 1.2 | 1.8 | 1.2 | 2.3 | 1.6 | 2.3 | 1.6 | 1.7 | 1.2 | ||
| Trangie | LGA | – | – | 2.5 | 1.5 | 2.4 | 1.5 | – | – | 2.7 | 1.8 | 2.6 | 1.8 | 1.8 | 1.2 | 2.3 | 1.6 | 2.8 | 2.0 | |
| HGA | – | – | 2.5 | 1.5 | 2.4 | 1.5 | – | – | 2.7 | 1. 8 | 2.6 | 1.8 | 1.8 | 1.2 | 2.3 | 1.6 | 2.8 | 2.0 | ||
| Vic | ||||||||||||||||||||
| Hamilton | LGA | 1.6 | 1.0 | 3.1 | 2.0 | 4.8 | 3.0 | 3.7 | 2.3 | 3.7 | 2.3 | 5.4 | 3.5 | 2.9 | 2.4 | 4.3 | 3.6 | 5.8 | 5.0 | |
| HGA | 3.3 | 2.0 | 4.2 | 2.5 | 4.8 | 3.0 | 3.7 | 2.3 | 4.6 | 2.9 | 6.2 | 4.1 | 4.3 | 3.6 | 4.3 | 3.6 | 5.8 | 5.0 | ||
| SA | ||||||||||||||||||||
| Keith | LGA | 4.3 | 2.5 | 3.4 | 2.0 | 3.3 | 2.0 | 3.8 | 2.3 | 3.8 | 2.3 | 3.7 | 2.3 | 3.6 | 2.4 | 3.0 | 2.0 | 2.9 | 2.0 | |
| HGA | 4.3 | 2.5 | 3.4 | 2.0 | 3.3 | 2.0 | 3.8 | 2.3 | 3.8 | 2.3 | 3.7 | 2.3 | 3.6 | 2.4 | 3.0 | 2.0 | 2.9 | 2.0 | ||
All simulations were profitable under the assumptions used, but 13 of the 144 simulations (9%) violated at least one of the sustainability indices, even at a stocking rate of one ewe per hectare; and these simulations were removed from further analysis. The respective HGA enterprises were also removed as both are required for profit and risk comparison, which resulted in 63 enterprises for comparison (total 126 simulations). As each simulated location is unique in its composition (e.g. pasture/cropping sequences, weather and soil data), comparisons between locations are not appropriate.
Conditional value at risk (CVaR) was used to quantify the expected financial outcome that would occur in a worst-case scenario and was used to explore the downside risk of the annual gross margins (A$/ha) of each simulation. As per Moore (2014), a 20% level was implemented, measuring the average gross margins in the worst 20% of cases over the 30-year trial period. A high CVaR value indicates lower financial risk if all else is equal. The benefits of CVaR as a measure of financial risk include readily understandable terms, meaningful units and mathematical coherence (Rockafellar and Uryasev 2002). The CVaR of average gross margin for HGA and LGA were calculated. The change in CVaR was then determined by subtracting the CVaR of LGA from the CVaR of HGA. The change in CVaR indicates the direction of risk, i.e. a negative CVaR change equates greater risk, no CVaR change is equal risk, and a positive CVaR change indicates reduced risk.
Results
Stocking rate
A HGA increased the mean stocking rate in 20 out of 63 farm enterprises (32%) (Table 3). Supplement allowance did not alter stocking rates at any of the Glen Innes, Trangie, or Keith enterprises (25/63, 40%). In Glen Innes and Keith simulations this was due to little reliance on supplementary feed and being limited by pasture production. Trangie was reliant on supplementary feed allowance, which was not eased when supplementary feed allowance was increased from LGA to HGA. Merino enterprises (44%, 8/18) and an autumn mating (50%, 9/18) had the greatest mean stocking rates. Maternal and spring-mated enterprises often had the lowest mean stocking rates. Generally, stocking rate was similar below an average annual rainfall of 500 mm, however, there was no influence of rainfall pattern on stocking rate.
Supplementary feed
Greater stocking rates resulted in greater total consumption of supplementary feed. To reduce confounding by stocking rate, supplementary feed was explored as mean supplement consumption per ewe mated per year (Fig. 1). Breed and mating season impacted supplement consumption within a location, however, the impact of location itself on supplement consumption per ewe mated was greater.
Mean total supplement consumption (kg) per DSE (50 kg dry sheep equivalent) for autumn-mated ewes at each location (in order of ascending average rainfall) when supplied the high grain allowance.
Merino enterprises generally had the highest supplement consumption per ewe mated, followed by Composite and then Maternal enterprises (Fig. 1). Self-replacement in Merino and Composite enterprises meant a greater number of sheep were carried across the year compared with Maternal systems where all female lambs were sold and replacement ewes purchased prior to short-term supplementary feeding. Autumn mating was most often associated with the highest consumption of supplement in Composite and Merino enterprises; followed by summer then spring. Summer- and autumn-mated Maternal enterprises consumed the most supplement. However, Keith and Narrandera simulations had greater supplement consumption during summer and spring.
Profit
Increasing the supplement allowance from LGA to HGA increased profit when stocking rate was increased (32%) (Table 4). As average rainfall increased so too did average stocking rate and profitability. Within each location, the most profitable enterprise by breed was generally Merino followed by Composite and then Maternal. This was generally driven by the additional income from wool production and sale of CFA ewes (Tables S2–S4). Maternal enterprises were often the least profitable overall; although they generally generated the greatest income from lamb production as all lambs (i.e. male and female) were sold, however this was counteracted by purchasing replacement ewes. The most profitable mating season was summer and autumn as spring-mated enterprises often had the lowest income from lamb, wool and ewe sales (Tables S2–S4).
| Mating season | Grain allowance A | Composite | Maternal | Merino | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Spring | Summer | Autumn | Spring | Summer | Autumn | Spring | Summer | Autumn | |||
| NSW | |||||||||||
| Bookham | LGA | 786 | 873 | 789 | 678 | 733 | 520 | 895 | 829 | 783 | |
| HGA | 811 | 921 | 789 | 690 | 760 | 520 | 919 | 852 | 814 | ||
| Bungarby | LGA | – | – | – | – | – | – | – | 209 | 117 | |
| HGA | – | – | – | – | – | – | – | 232 | 274 | ||
| Condobolin | LGA | 106 | 112 | 114 | 98 | 161 | 86 | 78 | 139 | 86 | |
| HGA | 106 | 232 | 281 | 98 | 213 | 179 | 78 | 139 | 236 | ||
| Glen Innes | LGA | 1003 | 912 | 1116 | 893 | 856 | 955 | 1202 | 1144 | 1286 | |
| HGA | 1003 | 912 | 1116 | 893 | 856 | 955 | 1202 | 1144 | 1286 | ||
| Narrandera | LGA | 119 | 170 | 140 | 113 | 100 | 63 | 197 | 174 | 112 | |
| HGA | 189 | 170 | 140 | 113 | 100 | 63 | 197 | 174 | 112 | ||
| Trangie | LGA | – | 160 | 155 | – | 145 | 118 | 124 | 173 | 220 | |
| HGA | – | 160 | 155 | – | 145 | 118 | 124 | 173 | 220 | ||
| Vic | |||||||||||
| Hamilton | LGA | 119 | 283 | 427 | 226 | 280 | 371 | 282 | 419 | 508 | |
| HGA | 247 | 357 | 427 | 226 | 349 | 412 | 396 | 419 | 508 | ||
| SA | |||||||||||
| Keith | LGA | 268 | 302 | 291 | 214 | 299 | 249 | 291 | 303 | 280 | |
| HGA | 268 | 302 | 291 | 214 | 299 | 249 | 291 | 303 | 280 | ||
Financial risk
The financial risk (CVaR value) of an enterprise only changed if stocking rate, and therefore profit, changed. Hence, Glen Innes, Trangie and Keith enterprises experienced no change in financial risk. Increased profit was met with increased risk in the majority (60%, 12/20) of circumstances (Fig. 2). Increased profit and reduced risk did occur and were generally concentrated in Hamilton spring- and summer- (n = 4) mated enterprises and autumn-mated Merino enterprises (n = 3; Bookham, Bungarby, Condobolin).
Change in average gross margins (A$/ha) and respective change in conditional value at risk (CvaR, A$/ha) as a result of changing from low to high grain allowance for Composite (a), Maternal (b) and Merino (c) ewes when there was a change in average gross margin. Change in CVaR was calculated by subtracting the CVaR for low grain allowance from the CVaR for high grain allowance.
Across the study, Merino enterprises generally had the least financial risk, followed by Composite and then Maternal enterprises (Table 5). Summer mating generally offered the least financial risk and autumn mating the most. There were also incidences where the CVaR for LGA and/or HGA was negative, indicating high financial risk as the average gross margin was a loss in the lowest 20% of cases. This was apparent in 20 out of 126 (16%) simulations and was most frequent in Maternal enterprises, during an autumn mating or in Condobolin or Trangie simulations.
| Mating season | Grain allowance A | Composite | Maternal | Merino | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Spring | Summer | Autumn | Spring | Summer | Autumn | Spring | Summer | Autumn | |||
| NSW | |||||||||||
| Bookham | LGA | 439 | 497 | 183 | 279 | 333 | −286 | 404 | 493 | 317 | |
| HGA | 339 | 416 | 183 | 166 | 267 | −286 | 378 | 491 | 359 | ||
| Bungarby | LGA | – | – | – | – | – | – | – | 35 | 47 | |
| HGA | – | – | – | – | – | – | – | 26 | 47 | ||
| Condobolin | LGA | 1 | 38 | 19 | −25 | −15 | −22 | 12 | 49 | −7 | |
| HGA | 1 | 38 | −17 | −25 | −49 | −92 | 12 | 49 | 30 | ||
| Glen Innes | LGA | 706 | 542 | 455 | 597 | 485 | 657 | 867 | 602 | 817 | |
| HGA | 706 | 542 | 455 | 597 | 485 | 657 | 867 | 602 | 817 | ||
| Narrandera | LGA | 70 | 80 | 59 | 43 | 39 | 9 | 117 | 97 | 64 | |
| HGA | 95 | 80 | 59 | 43 | 39 | 9 | 117 | 97 | 64 | ||
| Trangie | LGA | – | 11 | −7 | – | −33 | −74 | 7 | 29 | 1 | |
| HGA | – | 11 | −7 | – | −33 | −74 | 7 | 29 | 1 | ||
| Vic | |||||||||||
| Hamilton | LGA | 78 | 261 | −6 | 66 | 249 | −142 | 160 | 112 | 158 | |
| HGA | 118 | 316 | −6 | 66 | 283 | −159 | 174 | 112 | 158 | ||
| SA | |||||||||||
| Keith | LGA | 92 | 256 | 200 | 58 | 225 | 116 | 157 | 254 | 136 | |
| HGA | 92 | 256 | 200 | 58 | 225 | 116 | 157 | 254 | 136 | ||
Discussion
Increasing the supplementary feed allowance to align with current supplement budgets recommended by NSW DPI (2021) resulted in greater production through higher stocking rates. In these instances, profit increased due to greater income from the sale of lamb, wool and CFA ewes. This occurred in 32% of enterprises that met the requisite sustainability indices and is consistent with other studies (Amidy et al. 2017; Robertson and Friend 2020). Increased production and profit were met with increased risk in the majority of cases (60%) and was dependent on location, breed and mating season. This was expected given that the thresholds implemented by Warn et al. (2006), and adapted in the current study, were developed to reduce the financial risk associated with supplementary feeding and for pasture conservation. Increased financial risk associated with production and higher stocking rate has also been reported in previous modelling work (Salmon et al. 2004; Amidy et al. 2017).
Decreases in financial risk when stocking rate, supplement consumption and average gross margin increased were most common at Hamilton. The optimum stocking rate at Hamilton was limited by the supplementary feed threshold, as were most other simulations where financial risk reduced when profit increased. This contradicts previous reports of elevated supplementary feed leading to greater financial risk (Warn et al. 2006; Casburn 2010; Robertson et al. 2014). The pasture species used at each location will have influenced enterprise production and therefore risk (Byrne et al. 2010; Moore 2014; Thomas et al. 2018). The Hamilton location utilised lucerne, which had the greatest growth during spring and summer and may account for the improvement in financial risk (Moore 2014), but most remaining incidences of reduced financial risk occurred in enterprises that did not utilise lucerne.
Further, the rainfall and temperature patterns at each location determines pasture growth, quality and supply. A commonality between locations with reduced risk at HGA was a uniform rainfall between 1991 and 2020 (Bureau of Meteorology 2022). Direct comparison between locations cannot be made, however those with a greater proportion of cropping (Condobolin and Trangie) generally had increased risk. Further, Bookham and Glen Innes had similar average annual rainfall, but the reliance and cost of supplementary feed was much greater at Bookham. Weather was not explored further in the current study but may be a potential driver of financial risk. These differences are likely reflective of complex interactions of enterprise diversity, including sheep breed, mating season, climate, soil, pasture species and growth. As such, it is important that the results are interpreted within the assumptions of the current study.
Merino simulations often had the least financial risk, differing from some previous reports (Warn et al. 2006; Kopke et al. 2008). This is likely due to higher wool income and relatively high reproductive rates compared to previous reports (Kleemann and Walker 2005). Stocking rates in Merino enterprises were often greatest of the three sheep breeds and accounted for greater supplement consumption, whereas the high supplement consumption of Composite enterprises is reflective of high reproductive rates. The sale of all female lambs resulted in lower supplement consumption in Maternal simulations. When comparing breeds, the results of this study indicate financial risk was not dependent on supplement feed consumption and may be linked to other mating management practices such as mating season and ewe replacement (i.e. retained or purchased). The ability for a self-replacing enterprise to focus on wool and/or meat production depending on season and market value is beneficial (Kopke et al. 2008; Robertson and Friend 2020), and may lend itself to reduced financial risk. The purchase of replacement ewes in Maternal systems counteracted the sale of additional lambs and increased the financial risk of these enterprises. This relationship, however, will alter as commodity prices, especially meat, wool and supplementary feed, and reproductive rate vary relative to each other.
An autumn mating generally had the greatest reliance on supplementary feed, indicating pasture supply did not align well with feed demand in contrast to other studies (White et al. 1983; Robertson et al. 2014). There is evidence that mating one month later than the current study (April rather than March) may better align the pasture supply curve with feed requirements (Robertson and Friend 2020), but this would be dependent on location and finishing system (Moore 2009). The high consumption of grain may have been exacerbated by high pregnancy scanning rate and high lamb survival of autumn-mated ewes, leading to greater financial risk. Simulated lamb survival to marking may be greater than expected in practice and must be considered when interpreting results (Robertson and Friend 2020). Producers must consider the pasture growth curve, reproductive and production benefits and lamb survival challenges associated with mating season.
As is common in modelling studies, static prices and price grids were used across the simulations (Moore 2014; Robertson et al. 2014). Given the complexities in responses to breed and season at each location, price variation was not considered. Natural fluctuations in commodity prices, therefore, were not captured across the 30-year period modelled, and as such, the profitability of each simulated enterprise may be under or over-estimated. The commodity prices used in the current study were generally greater than that used by Warn et al. (2006), and the relationship between these varied. For instance, grain price has increased by approximately 52%, lamb price by 61–64% (dependent on breed), broad fibre wool by 7%, but fine wool has declined in value by approximately 19% (Warn et al. 2006; NSW DPI 2021). Further, as producers have pursued greater fleece weights and lamb growth, ewe liveweights have also increased (Walkom and Brown 2017; Masters and Ferguson 2019). The ewe SRW reported by the NSW DPI (2021) and used in the current study were also greater than those used by Warn et al. (2006). Given the increase and altered relationship of supplement price, value of lamb meat, wool, and larger-framed ewes, increasing feed to Merino ewes by 5 kg/head, or approximately 14%, and Maternal and Composite ewes by 12 kg/head, or approximately 29%, to algin with NSW DPI (2021) budget recommendations is a justifiable progression and aligns with recommendations from the Lifetime Maternals project (Blumer et al. 2019).
A proportion of the simulated enterprises did not meet the sustainability indices. The models were developed to represent an ‘average’ enterprise and were informed by a producer survey, previous research and gross margins budgets available as a producer resource (Warn et al. 2006; NSW DPI 2021; Bates et al. 2023). It may be common to use higher/lower feeding allowances at different locations dependent on pasture and feed resources but published literature to support such an assertion was not found. As this was unknown, the threshold levels used in the current study were determined to be a suitable benchmark to apply across locations. Greater understanding of supplement usage by producers would improve the application of modelled outcomes.
There were limitations in the development of these models. Pasture productivity was validated against available resources, however, may differ in practice. For example, previous work has demonstrated a stocking rate of 13–25 DSE/ha at Hamilton (EverGraze 2012), however, this was not achieved using the assumptions applied in this study. The distribution of conception dates across the mating period was not collected from the commercial farms and, therefore, was not compared to that generated by the model. Labour costs were not included in the financial assessment and any increased labour cost associated with increased stocking rate were not reflected in average gross margins. Transporting wool and supplementary feed and storing supplementary feed were not accounted for, which will have regional variation, and may also impact overall enterprise profitability. Highly precise results or recommendations are not possible from this modelling study; however, it is possible to identify opportunities and challenges in the management practices explored.
Conclusions
Under the assumptions used in the current study, increasing the supplementary feed allowance from the LGA (30 kg/head.year for all ewes) to HGA (35 kg/head.year for Merino and 42 kg/head.year for non-Merino ewes) increased profitability in 32% of the simulated farm enterprises through increased stocking rate leading to greater production outputs; lamb sales, wool and CFA ewes. This predominantly increased financial risk. In most modelled enterprises, increasing the supplementary feed allowance did not influence production, profit or financial risk because the grain allowance rule did not limit the optimum enterprise in the first instance (i.e. limited by groundcover threshold, or maximum average gross margin was achieved). Restricting the amount of supplementary feed as a means of reducing costs may hinder production and profit when comparing breeds. Financial risk was associated with supplement consumption when comparing mating seasons and so reducing consumption may increase profitability in this instance. Producers may be able to improve profit and reduce financial risk by altering breed, stocking rate and mating season, by retaining ewes for replacement or modifying the amount of supplement fed relative to their location. It is noted that alterations to these management practices vary in difficulty and may not appeal to or be practical for all producers. Input costs and commodity prices will also affect profit and financial risk and therefore need to be considered before making strategic changes. The current study delivers value in identifying the potential opportunities and challenges of different management practices.
Data availability
The data that support this study cannot be publicly shared due to ethical or privacy reasons and may be shared upon reasonable request to the corresponding author if appropriate.
Declaration of funding
This research was funded by Meat and Livestock Australia (Project No. L.LSM.0020), the New South Wales Department of Primary Industries and an Australian Government Research and Training Scholarship awarded to A.L.B.
Acknowledgements
The authors thank Neville Herrmann and Andrew Moore for their assistance in AusFarm model development and understanding. The expert consultants engaged at NSW DPI, SARDI and privately are thanked for their assistance in developing pasture and cropping rotations. The anonymous reviewers are thanked for their assistance in improving the clarity of this paper.
References
Amidy MR, Behrendt K, Badgery WB (2017) Assessing the profitability of native pasture grazing systems: a stochastic whole-farm modelling approach. Animal Production Science 57(9), 1859-1868.
| Crossref | Google Scholar |
Bates AL, McGrath SR, Robertson SM, Refshauge G (2022) Mating conditions and management practices influence pregnancy scanning outcomes differently between ewe breeds. Animals 12(21), 2908.
| Crossref | Google Scholar | PubMed |
Bates AL, McGrath SR, Allworth MB, Robertson SM, Refshauge G (2023) A cross-sectional study of commercial ewe management practices for different sheep breeds across southern Australia. Animals 13(3), 388.
| Crossref | Google Scholar | PubMed |
Blumer S, Behrendt R, Hocking Edwards J, Young J, Thompson A (2019) Lifetime maternals - Phase II: feeding standards for maternal ewes (Project code: L.LSM.0008) [Report]. Meat and Livestock Australia Limited. Available at https://www.mla.com.au/research-and-development/reports/2019/feeding-standards-for-modern-maternal-ewes/
Bureau of Meteorology (2022) Average annual, seasonal and monthly rainfall Vol. 2022. Commonwealth of Australia. Available at http://www.bom.gov.au/jsp/ncc/climate_averages/rainfall/index.jsp
Byrne F, Robertson MJ, Bathgate A, Hoque Z (2010) Factors influencing potential scale of adoption of a perennial pasture in a mixed crop-livestock farming system. Agricultural Systems 103(7), 453-462.
| Crossref | Google Scholar |
Casburn G (2010) The profitability of a lamb meat system is greatly influenced by lambing time. Australian Farm Business Management Journal 7(2), 37-43.
| Google Scholar |
EverGraze (2012) Hamilton proof site: Key messages Vol. 2024. (EverGraze) Available at https://www.evergraze.com.au/research-and-demonstration/hamilton-proof-site/index.html
Fogarty NM, Mulholland JG (2014) Seasonal reproductive performance of crossbred ewes in intensive lamb-production systems. Animal Production Science 54(6), 791-801.
| Crossref | Google Scholar |
Freer M, Moore AD, Donnelly JR (2012) The GRAZPLAN animal biology model for sheep and cattle and the GrazFeed decision support tool [Reviewed 2012]. Available at https://grazplan.csiro.au/wp-content/uploads/2007/08/TechPaperMay12.pdf
Kingwell R (2002) Sheep animal welfare in a low rainfall Mediterranean environment: a profitable investment? Agricultural Systems 74(2), 221-240.
| Crossref | Google Scholar |
Kleemann DO, Walker SK (2005) Fertility in South Australian commercial Merino flocks: sources of reproductive wastage. Theriogenology 63(8), 2075-2088.
| Crossref | Google Scholar | PubMed |
Kopke E, Young J, Kingwell R (2008) The relative profitability and environmental impacts of different sheep systems in a Mediterranean environment. Agricultural Systems 96(1–3), 85-94.
| Crossref | Google Scholar |
Making More From Sheep (2008) Making More From Sheep. Available at https://www.makingmorefromsheep.com.au/
Masters DG, Ferguson MB (2019) A review of the physiological changes associated with genetic improvement in clean fleece production. Small Ruminant Research 170, 62-73.
| Crossref | Google Scholar |
Meat and Livestock Australia Limited (2022a) Australia - Sheepskin prices - Lambs - Calendar year. Available at http://statistics.mla.com.au/Report/List
Meat and Livestock Australia Limited (2022b) Australia - Sheepskin prices - Merino sheep - Calendar year. Available at http://statistics.mla.com.au/Report/List
Moore AD (2009) Opportunities and trade-offs in dual-purpose cereals across the southern Australian mixed-farming zone: a modelling study. Animal Production Science 49(10), 759-768.
| Crossref | Google Scholar |
Moore AD (2014) The case for and against perennial forages in the Australian sheep–wheat zone: modelling livestock production, business risk and environmental interactions. Animal Production Science 54(12), 2029-2041.
| Crossref | Google Scholar |
Moore AD, Holzworth DP, Herrmann NI, Huth NI, Robertson MJ (2007) The common modelling protocol: a hierarchical framework for simulation of agricultural and environmental systems. Agricultural Systems 95(1–3), 37-48.
| Crossref | Google Scholar |
NSW DPI (2021) Livestock gross margin budgets Vol. 2021. Available at https://www.dpi.nsw.gov.au/agriculture/budgets/livestock
Robertson SM, Friend MA (2020) Performance of sheep systems grazing perennial pastures. 4. Simulated seasonal variation and long-term production. Animal Production Science 60(3), 423-435.
| Crossref | Google Scholar |
Robertson SM, Southwell AF, Friend MA (2014) Modelling the risk of different joining times and lamb sale policies. Animal Production Science 54(10), 1694-1698.
| Crossref | Google Scholar |
Rockafellar RT, Uryasev S (2002) Conditional value-at-risk for general loss distributions. Journal of Banking and Finance 26(7), 1443-1471.
| Crossref | Google Scholar |
Salmon L, Donnelly JR, Moore AD, Freer M, Simpson RJ (2004) Evaluation of options for production of large lean lambs in south–eastern Australia. Animal Feed Science and Technology 112(1), 195-209.
| Crossref | Google Scholar |
Stone G, Pozza RD, Carter J, McKeon G (2019) Long Paddock: climate risk and grazing information for Australian rangelands and grazing communities. The Rangeland Journal 41(3), 225-232.
| Crossref | Google Scholar |
Thomas DT, Moore AD, Bell LW, Webb NP (2018) Ground cover, erosion risk and production implications of targeted management practices in Australian mixed farming systems: lessons from the Grain and Graze program. Agricultural Systems 162, 123-135.
| Crossref | Google Scholar |
Walkom SF, Brown DJ (2017) Genetic evaluation of adult ewe bodyweight and condition: relationship with lamb growth, reproduction, carcass and wool production. Animal Production Science 57(1), 20-32.
| Crossref | Google Scholar |
Warn L, Webb Ware J, Salmon L, Donnelly J, Alcock D (2006) Analysis of the profitability of sheep wool and meat enterprises in southern Australia (Project 1.2.6) [Report: Australian Sheep Industry CRC]. Sheep CRC, Armidale, Australia. Available at https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.472.8604&rep=rep1&type=pdf
White DH, Bowman PJ, Morley FHW, McManus WR, Filan SJ (1983) A simulation model of a breeding ewe flock. Agricultural Systems 10(3), 149-189.
| Crossref | Google Scholar |
Young JM, Thompson AN, Curnow M, Oldham CM (2011) Whole-farm profit and the optimum maternal liveweight profile of Merino ewe flocks lambing in winter and spring are influenced by the effects of ewe nutrition on the progeny’s survival and lifetime wool production. Animal Production Science 51(9), 821-833.
| Crossref | Google Scholar |
Young M, Vercoe PE, Kingwell RS (2022) Optimal sheep stocking rates for broad-acre farm businesses in Western Australia: a review. Animal Production Science 62(9), 803-817.
| Crossref | Google Scholar |
