The net contribution of dairy production to human food supply: The case of Austrian dairy farms

Agricultural Systems 137 (2015) 119–125

Contents lists available at ScienceDirect

Agricultural Systems j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a g s y

The net contribution of dairy production to human food supply: The case of Austrian dairy farms Paul Ertl *, Hannes Klocker, Stefan Hörtenhuber, Wilhelm Knaus, Werner Zollitsch Department of Sustainable Agricultural Systems, Division of Livestock Sciences, BOKU–University of Natural Resources and Life Sciences, Gregor Mendel Strasse 33, 1180 Vienna, Austria

A R T I C L E

I N F O

Article history: Received 16 December 2014 Received in revised form 13 April 2015 Accepted 14 April 2015 Available online 15 May 2015 Keywords: Human-edible feed conversion efficiency Dairy production Grass-based Food security Ruminant Feed versus food competition

A B S T R A C T

Due to their ability to convert human-inedible fibrous plant materials into high quality animal products, ruminants have always played an important role as net food producers. However, to meet the animals’ nutritional requirements, today’s rations for high yielding dairy cows also contain substantial amounts of potentially human-edible feeds (e.g. cereals and pulses), which increases competition between animal feed and human food availability. The aim of the present study was therefore to calculate the humanedible feed conversion efficiency (heFCE) for 30 Austrian dairy farms operating under different production systems in order to evaluate their contribution to net food production. The heFCE was calculated at farm gate level on a gross energy and crude protein basis, and was defined as potentially human-edible output in the form of animal products (milk and meat) divided by the input of potentially human-edible feedstuffs. The potentially human-edible fraction of all feedstuffs used on the 30 farms was estimated based on available literature using a “low,” “medium,” and “high” scenario, representing low, average, and above average extraction rates of human-edible nutrients from feedstuffs, respectively. The human-edible fraction ranged from 0% for some fibrous feedstuffs up to 100% for some cereals in the high scenario. For the “medium” scenario, heFCE ranged from 0.50 up to 2.95 for energy and from 0.47 up to 2.15 for protein. About half of the analysed farms showed a heFCE below 1, indicating a net loss in food supply. For both energy and protein, heFCE was negatively correlated with the amount of concentrates per kg milk and the total amount of concentrates per cow and year. In addition, we found a positive correlation between heFCE and the area of grassland utilized per ton of milk, as well as a negative correlation between heFCE and the area of arable land required per ton of milk. Therefore, feeding large amounts of concentrates to dairy cows has to be questioned in terms of the heFCE. The results of this study clearly show that grassbased dairy production highly contributes to net food production, particularly if the amount of concentrates per kg milk is reduced. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction With their ability to convert human-inedible plant material into high quality human food, animals have always played an important role in human nutrition and food security (Bradford, 1999; Leng, 2010). When livestock is produced based on grassland or other human-inedible resources, animal production makes an important contribution to total food supply (FAO, 2011; Foley et al., 2011). However, in intensive livestock systems, animals are often fed substantial quantities of potentially human-edible crops, such as grains and pulses, which is a very inefficient way to provide human food and which represents a net drain in total human food production (Cassidy et al., 2013; CAST – Council for Agricultural Science and

* Corresponding author. Tel.: +43 1 47654 3293; fax: +43 1 47654 3254. E-mail address: [email protected] (P. Ertl). http://dx.doi.org/10.1016/j.agsy.2015.04.004 0308-521X/© 2015 Elsevier Ltd. All rights reserved.

Technology, 2013; FAO, 2011). An increasing world population, together with a higher per capita consumption of animal products, will increase the pressure on livestock systems with regard to food efficiency (Cassidy et al., 2013). In order to obtain more sustainable livestock production systems, it is inevitable that less potential human food is fed to animals (Eisler et al., 2014; Herrero and Thornton, 2013). Among various existing concepts to evaluate competition between animal feed and human food, the most promising one is to relate the human-edible output in the form of the animal products to the potentially human-edible input via feedstuffs (FAO, 2011; Oltjen and Beckett, 1996; Wilkinson, 2011). The relation of human-edible output per human-edible input can be described as human-edible feed conversion efficiency (heFCE). As compared to monogastric animals, dairy and grass-based beef and lamb production systems show generally favourable net food production rates (Wilkinson, 2011). From their nutritional ecology, cattle are specialists in digesting fibrous plant materials (e.g. forages) and do not necessarily rely on feeds that could potentially serve as human food

120

P. Ertl et al./Agricultural Systems 137 (2015) 119–125

(Gill, 2013; Hofmann, 1989). However, due to enormous increases in animal performance during the last five decades, the inclusion of grains and pulses in cattle’s diet has become necessary to meet the animals’ nutrient and energy requirements (Knaus, 2013). As a result, in some cases beef and dairy production systems even show a heFCE below 1, indicating a net food drain (CAST – Council for Agricultural Science and Technology, 1999; Oltjen and Beckett, 1996; Wilkinson, 2011). Although heFCE has already been calculated for dairy cows based on data from model calculations (Oltjen and Beckett, 1996; Wilkinson, 2011), whole country data (CAST – Council for Agricultural Science and Technology, 1999), or short term feeding trials (Ertl et al., 2015), an analysis of field data from a range of practical dairy farms regarding their heFCE is still lacking. The aim of the present study was therefore to calculate the heFCE for Austrian dairy farms in order to evaluate the potential range of the contribution of dairy production to the net food supply. A major problem when calculating the heFCE for a single animal or a whole production system is evaluating the potential human-edible input via feedstuffs (Ertl et al., 2015; Le Cotty and Dorin, 2012; Wilkinson, 2011). Therefore, the second aim of this study was to provide literature-based estimates for the human-edible fractions (heF) of feedstuffs used on the selected farms. 2. Materials and methods 2.1. Data source On farm data were taken from a national research project on the integrated assessment of the sustainability of selected Austrian milk production systems (Hörtenhuber et al., 2013). On the basis of IACS data (Integrated Administration and Control System of the European Union), 31 Austrian dairy farms were selected for this project. These farms were distributed over the whole country and consisted of 24 conventionally and 7 organically managed operations, which, according to the national statistical database (BMLFUW – Bundestministerium für Land- und Forst-, Umwelt und Wasserwirtschaft, 2014), roughly reflect the actual distribution of conventional and organic dairy farms in Austria. From the overall data set, data related to milk production (e.g. milk yield and composition, amount and composition of concentrates used, livestock sales and purchases), averaged over two years (2010 and 2011) were used for the current study. Due to unusually high animal sales (more than twice as high as the average farm), one of the organic farms had to be excluded from the calculations. According to their milk quota, regional location, and points in the mountain farm register (which identifies and classifies site-related natural and economic challenges affecting individual farms), the remaining 30 farms were assigned to one of the following six production systems: alpine (AL),

alpine intensive (AI), hilly-pasture (HP), hilly-arable (HA), lowlandsmixed (LM), and lowlands-specialized (LS). Table 1 presents average production data for each production system. 2.2. Calculation of the human-edible feed conversion efficiency The heFCE was defined as human-edible output in the form of animal products divided by potential human-edible input via feedstuffs as MJ gross energy (GE) and kg crude protein (CP), at farm gate per year. To calculate the human-edible input via feeding, the potential heF for the GE and CP content of feedstuffs were estimated, based on available literature on food processing or food usage of these commodities (sources given in Table 2 below). The estimated heF were then multiplied with the amount of GE and CP per kg of the respective feedstuff and with the total quantity of each feedstuff fed (kg dry matter). Since calculations were performed at farm gate level, feedstuffs used for dry cows and young stock were also included in the calculations. The human-edible output comprised the amount of GE and CP represented by the milk sold and the net quantity of beef leaving the farm in the respective year. Milk sold was standardized for 4% fat and 3.4% protein (energy corrected milk, ECM). Therefore, 34 g CP and 3.17 MJ per kg milk were presumed (Buttchereit et al., 2010). The net quantity of beef leaving the farm was calculated as the total live weight of cattle leaving the farm minus the total live weight of cattle entering the farm within this year. The human-edible proportion of the live weight of fattening cattle is about 43% (de Vries and de Boer, 2010). However, the human-edible proportion of cattle’s live weight is different between beef and dairy cattle. In Austria, average carcass yield is 49 and 53% for cows and heifers, respectively (Statistics Austria, 2014). About 74% of the carcass can be considered as saleable meat (Minchin et al., 2009; Vestergaard et al., 2007). Presuming an average carcass yield of 51% for dairy cows and heifers, the human-edible proportion of the animal’s live weight was set at 38%. This humanedible meat was assumed to have an average protein content of 19% (de Vries and de Boer, 2010) and an energy content of 6.48 MJ/kg (Bauer et al., 2007). 2.3. Estimation of human-edible fractions For grasses, dried beet pulp, dried distiller’s grains with solubles, lucerne cobs, brewers’ grains, and maize gluten feed no noteworthy potential food uses were found and their heF were therefore estimated to be zero. The basis for estimating the heF for each of the remaining feedstuffs are shown in Table 2. The heF of feedstuffs cannot be seen as one fixed and generally applicable value because they depend on the technology available and other circumstances, such as the degree of food availability. Therefore, heF

Table 1 Main characteristics (average ± standard deviation) of dairy production systems. Item

Production systema AL

AI

HP

HA

LM

LS

Farms (conventional/organic, n) Herd size (lactating cows, n) ECMb produced (kg/cow/year) ECM sold (kg/cow/year) Concentrates (g DM/kg ECMproduced) Maize silage (g DM/kg ECMproduced) Required arable land (ha/t ECMproduced) Required grassland (ha/t ECMproduced) Required total land (ha/t ECMproduced)

3/1

2/3 27 ± 8 8248 ± 824 7648 ± 962 227 ± 33 51c 0.047 ± 0.006 0.112 ± 0.051 0.159 ± 0.056

4/1 20 ± 5 6355 ± 446 5614 ± 498 229 ± 82 0 0.058 ± 0.010 0.216 ± 0.065 0.274 ± 0.056

4/1 32 ± 5 7533 ± 793 6837 ± 737 292 ± 72 162 ± 99 0.077 ± 0.021 0.056 ± 0.020 0.133 ± 0.016

5/0 27 ± 10 8058 ± 625 7129 ± 448 294 ± 51 204 ± 93 0.085 ± 0.013 0.041 ± 0.014 0.126 ± 0.013

6/0 50 ± 18 8403 ± 1178 7746 ± 1160 338 ± 46 187 ± 49 0.083 ± 0.019 0.056 ± 0.020 0.139 ± 0.014

9±4 6818 ± 848 5880 ± 754 337 ± 72 18c 0.063 ± 0.007 0.225 ± 0.006 0.288 ± 0.010

a Based on milk quota and points in the mountain farm register; AL – alpine, AI – alpine-intensive, HP – hilly-pasture, HA – hilly-arable, LM – lowlands-mixed, LS – lowlandsspecialized. b Energy corrected milk. c In this group maize silage was fed on one farm only, therefore no standard deviation is given.

P. Ertl et al./Agricultural Systems 137 (2015) 119–125

121

Table 2 Estimation of human-edible fractions of concentrate feedstuffs used on the 30 project farms. Feedstuff

Basis for estimating the human-edible fraction

References

Milling grade depending on type of flour (fine flour – whole grain flour)a Vetrimani et al., 2005; Zwingelberg, 2004 Milling grade depending on type of flour (pearls – dehulled grain flour)a Bhatty, 1986; Kling and Wöhlbier, 1983; Ullrich, 2011 Possible nutrient extraction ratesb,c or milling gradea Eckhoff et al., 1999; Lee et al., 2007; Mestres et al., 1991; Stolp, 2004 Triticale Similar nutrient composition and qualities as wheat = > calculated Peña, 2004; Peña and Amaya, 1992; Serna-Saldivar et al., 2004 in wheat equivalentsa Rye Milling grade depending on type of flour Glitso and Knudsen, 1999; Zwingelberg, 2004 (fine flour – whole grain flour)a Wheat bran High fibre content of wheat bran increases faecal energy losses, Stevenson et al., 2012 thus amounts for human consumption are limiteda b a Peas Possible protein and starch extraction or dehulled whole peas Boye et al., 2010; Hoover et al., 2010; Pelgrom et al., 2013; Vose, 1980 Soybeans Possible protein and fat extraction (concentrates/isolates)c Fischer et al., 2001; Shallo et al., 2001; Wang et al., 2005 or dehulled whole beansa Soybean/sunflower/rapeseed Possible protein (and fat) extraction ratec Aider and Barbana, 2011; DLG, 1997; Fischer et al., 2001; cake and expeller Salgado et al., 2011; Salgado et al., 2012; Shallo et al., 2001; Tan et al., 2011; Villanueva et al., 1999; Wang et al., 2005 b,c Maize silage Potential starch extraction at different maturity stages Baldwin et al., 1992; DLG, 1997; Stolp, 2004 a or harvested as maize grains Wheat Barley Maize

a

Identical factor for the human-edible fraction for protein and energy due to presumed usage of whole grain. Two different factors for the human-edible fractions for protein and energy: human-edible fractions for protein are derived from the potential extraction rates for protein, whereas the human-edible fractions for energy are the sum of the amount of gross energy in the extractable starch and protein divided by the gross energy content of the feedstuff. c Same as described under b, but energy from protein and fat instead of protein and starch extraction. b

were estimated for energy and protein based on available literature for the following 3 scenarios: – “Low”: Recovery of human-edible energy and protein from feedstuffs is lower than described on average in the literature. These recovery rates can be seen as easily achievable without highend technology and/or representing above-average processing losses. – “Medium”: This scenario describes the most likely achievable heF for CP and GE with current standard technology. – “High”: For this scenario, relatively high extraction rates are achieved due to implementation of some kind of sophisticated technology or a moderate change in eating habits (e.g. increasing the consumption of whole grain foods). In addition to these 3 scenarios, heF were also estimated based on the assumptions by Wilkinson (2011).

3. Results In Table 3, heF for protein and energy are shown for the different estimation scenarios. Human-edible fractions range from 0% for feeds not included in the table (all scenarios) and wheat bran (scenario “low”) up to 100% for some cereals (scenario “high”). The biggest differences in the heF between protein and energy are found for rapeseed expeller (scenario “high”), with the heF for protein being 87% as compared to 39% for energy. The calculated heFCE for each farm are shown in Fig. 1 and Fig. 2 for energy and protein, respectively. Average heFCE for energy was 1.01 and ranged from 0.50 for farm LS-75 up to 2.95 for farm HP19-org in the “medium” scenario. For the “low” and “high” scenarios, average heFCE for energy were higher by 0.47 points and lower by 0.22 points, respectively, as compared to “medium”. With 1.72, 1.17, and 0.88 for scenarios “low”, “medium”, and “high”, respectively, average heFCE for protein were slightly higher than for energy.

Table 3 Crude protein (CP) and gross energy (GE) content of concentrates and their estimated human-edible energy (E) and protein (P) fraction (%) for 3 different estimation scenarios and a scenario based on estimations by Wilkinson (2011). Concentrate feedstuff

Wheat Barley Maize Triticale Rye Wheat bran Peas Soybeans Soybean expeller Soybean cake Sunflower expeller Sunflower cake Rapeseed expeller Rapeseed cake Maize silage Other feedstuffsb a b

CP

GE

Protein, estimation scenario

(g/kg)a

(MJ/kg)a

Plow

Pmedium

Phigh

Energy, estimation scenario Elow

Emedium

Ehigh

E=P

126 118 94 126 103 173 239 396 518 493 324 279 382 341 88

18.2 18.4 18.7 18.2 18 18.9 18.3 23.6 19.7 20.8 19.4 21.8 19.4 21.3 19

60 40 70 60 60 0 70 50 50 50 14 14 30 30 19 0

80 65 80 80 80 10 80 92 71 71 30 30 59 59 29 0

100 80 90 100 100 20 90 93 92 92 46 46 87 87 45 0

60 40 70 60 60 0 37 51 30 42 5 20 14 26 19 0

80 65 80 80 80 10 64 64 43 54 12 25 26 36 29 0

100 80 90 100 100 20 90 93 56 65 18 30 39 47 45 0

80 80 80 80 80 20 80 80 80 80 20 20 20 20 0 0

On a dry matter basis; source: INRA et al. (2014). Including: grasses, dried beet pulp, dried distiller’s grains with solubles, lucerne cobs, brewers’ grains, maize gluten feed.

Wilkinson

122

P. Ertl et al./Agricultural Systems 137 (2015) 119–125 4.5 4.0

heFCE for energy (MJ/MJ)

3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0

Fig. 1. Human-edible feed conversion efficiency (heFCE) for selected Austrian dairy farms on gross energy basis; grey bars show heFCE for the “medium” scenario with dotted bars illustrating the range of heFCE for the scenarios “low” and “high” (characters on x-axis: first two letters indicate the production system for the respective farm; AL – Alpine, HP – hilly-pasture, HA – hilly-arable, AI – alpine-intensive, LM – lowlands-mixed, LS – lowlands-specialized; the following number states herd size, i.e. number of lactating dairy cows on this farm; appendix “org” indicates that these farms are organically managed; the dotted horizontal line indicates a heFCE of 1).

Human-edible feed conversion efficiencies for different production systems are shown in Table 4. The HP production system showed the highest heFCE, both for energy and protein, while LM and LS had heFCE of below 1 for the “medium” and “high” scenario, both for energy and protein. Calculations based on assumptions for heF according to Wilkinson (2011) showed on average the same values for heFCE for energy, and higher heFCE for protein, when compared to the “medium” scenario. For the most important indicators, we observed the following correlations in the “medium” scenario: Human-edible feed conversion efficiencies were negatively correlated with the amount of concentrates used per kg milk (−0.72 and −0.82 for energy and protein, respectively, P < 0.01, n = 30) and also

with the total amount of concentrates fed per cow and year (−0.73 and −0.80 for energy and protein, respectively, P < 0.01, n = 30). Furthermore, there were negative correlations between heFCE and arable land required per ton of milk (−0.51 and −0.72 for energy and protein, respectively, P < 0.01, n = 30) and a positive correlation between heFCE and the area of grassland required (0.64 and 0.55 for energy and protein, respectively, P < 0.01, n = 30). Only weak correlations were found between milk yield per cow per year and heFCE for energy and protein (−0.39 and −0.34, with P = 0.03 and P = 0.06, respectively, n = 30), but a stronger correlation occurred for milk yield per year and total amount of concentrates used per cow and year (0.60, P < 0.01, n = 30).

3.5

heFCE for protein (kg/kg)

3.0

2.5

2.0

1.5

1.0

0.5

0.0

Fig. 2. Human-edible feed conversion efficiency (heFCE) for selected Austrian dairy farms on crude protein basis; grey bars show heFCE for the “medium” scenario with dotted bars illustrating the range of heFCE for the scenarios “low” and “high” (characters on x-axis: first two letters indicate the production system for the respective farm; AL – Alpine, HP – hilly-pasture, HA – hilly-arable, AI – alpine-intensive, LM – lowlands-mixed, LS – lowlands-specialized; the following number states herd size, i.e. number of lactating dairy cows on this farm; appendix “org” indicates that these farms are organically managed; the dotted horizontal line indicates a heFCE of 1).

P. Ertl et al./Agricultural Systems 137 (2015) 119–125

Table 4 Human-edible feed conversion efficiencies (heFCE) for energy and protein for different scenarios for estimation of the potentially human-edible fraction of concentrates. Scenario

heFCEb Energy

heFCE Protein

Low Medium High Wilkinsonc Low Medium High Wilkinsonc

Production systema

Average

AL

AI

HP

HA

LM

LS

1.30 0.89 0.71 0.79 1.67 1.11 0.86 1.12

1.77 1.23 0.99 1.14 2.35 1.58 1.22 1.67

2.66 1.81 1.44 1.52 2.51 1.63 1.24 1.70

1.08 0.75 0.59 0.88 1.57 1.12 0.87 1.35

1.07 0.69 0.52 0.95 1.17 0.78 0.58 1.11

1.06 0.70 0.54 0.83 1.11 0.75 0.57 0.93

1.48 1.01 0.79 1.01 1.71 1.15 0.88 1.29

a Based on milk quota and points in the mountain farm register; AL – alpine, AI – alpine-intensive, HP – hilly-pasture, HA – hilly-arable, LM – lowlands-mixed, LS – lowlands-specialized. b In MJ/MJ human-edible for energy and kg/kg human-edible for protein. c Based on estimation of human-edible fractions according to Wilkinson (2011).

4. Discussion To the best of our knowledge, this has been the first study investigating heFCE in practice on farms and estimating heF for various feedstuffs based on food processing literature. Compared to Wilkinson (2011), our estimations of heF are more differentiated and therefore allow us to depict a range of the potential heF of feedstuffs. Although heFCE for energy on average did not differ between the “medium” and “Wilkinson” estimation scenarios, great differences were observed between production systems, suggesting that the generalized heF of 0, 20, and 80% stated by Wilkinson (2011) for different feedstuffs (see Table 3) may result in an overestimation of heFCE for more intensive production systems. In our calculations for heFCE we presumed that from a food safety perspective, all animal feeds were potentially edible for humans. One of the major food safety-related aspects of the competition between animal feed and human food are mycotoxin contaminations, which are potentially threatening human and animal health (Friedman, 1996). Maximum or guidance levels for mycotoxins in animal feeds are mostly higher than for contamination of cereals for human consumption (Cheli et al., 2014). Thus, one can assume that not all feeds would be suitable for human consumption due to contaminations, with mycotoxins serving as an example herein. However, feeds for dairy cattle have generally stricter regulations on maximum contamination levels as compared to e.g. beef cattle (Murphy et al., 2006) and no valid data are available on the percentage of cereals used as feed for dairy cattle that would not meet the respective requirements for human food. Therefore, we did not include this specific aspect in our calculations, but we are aware that this may be a matter of concern in some cases. The heFCE reported herein for the individual farms of between 0.38 and 4.31 for energy and 0.36 and 3.25 for protein (depending on the estimation scenario for the heF) are slightly lower than values reported in earlier studies (Wilkinson, 2011; CAST – Council for Agricultural Science and Technology, 1999; Oltjen and Beckett, 1996), but within the same order of magnitude. CAST – Council for Agricultural Science and Technology (1999) reported heFCE for different countries ranging from 0.79 to 4.61 for energy, and from 1.06 to 14.30 for protein, whereas Wilkinson (2011) calculated a heFCE for dairy in the UK of 2.13 and 1.41 for energy and protein, respectively. In contrast to our calculations, Wilkinson (2011) calculated the human-edible input per human-edible output. Therefore, we used the reciprocal value of his results for comparisons with ours. Comparing two different Californian dairy rations, Oltjen and Beckett

123

(1996) estimated heFCE of 0.57 and 0.96 for energy and protein, respectively, for a ration containing large proportions of maize and soy and 1.28 and 2.76 for a least cost ration in which industrial by-products were used as concentrate supplements. In a shortterm feeding trial with mid lactating dairy cows, the substitution of commonly used concentrates with by-products from the food processing industry increased the heFCE from 1.39 to 5.55 and from 1.60 to 4.27 on an energy and protein basis, respectively (Ertl et al., 2015). The results of earlier studies and the results of the present work show that heFCE for dairy production systems show a large variation, resulting in a great potential for increasing heFCE. As indicated by the calculated correlations, one of the main measures to enhance heFCE is to reduce the amount of concentrates per kg milk and increase the milk production from grassland, from which human food cannot be directly derived. The outstanding role of grassland in terms of human food security is indisputable (FAO, 2011; Leng, 2010; O’Mara, 2012) and explains why farms of the HP production system, achieving only a moderate average milk yield of 6355 kg per cow and year, showed the highest heFCE for energy and protein. In addition to reducing concentrate supplementation, the feeding of human-inedible concentrates such as by-products from the food processing industry is another way to achieve a favourable heFCE (FAO, 2011). Thus, the composition of the diet rather than milk yield determines food efficiency of a dairy production system. However, at present, food efficiency in terms of heFCE is not rewarded by the markets and therefore is not given high priority (Lapierre et al., 2005). This is particularly the case in the affluent regions of the world where inexpensive concentrates are readily available. It is, however, likely that the issue of contributions of livestock production systems to the actual (net) food supply and hence the efficiency of utilization of potentially human-edible feed inputs in livestock production will gain importance if resources will become even more limited in the near future (Leng, 2010). In the present study, heFCE was slightly higher for protein than for energy, which is in agreement with earlier publications (Ertl et al., 2015; CAST – Council for Agricultural Science and Technology, 1999; Oltjen and Beckett, 1996). Supplementation of easily fermentable carbohydrates helps to increase protein efficiency and is an effective way to provide additional protein to the animal by increasing microbial protein synthesis (Van Soest, 1994). Among other reasons, such as a quantitatively better balance of energy and protein supply, the synchronization of high-quality carbohydrate supply to the rumen microbiota with the release of ammonia allows the animal to recapture the latter and thereby increases efficiency of protein utilization (Nocek and Russell, 1988; Poppi and McLennan, 1995). Therefore, particularly in grassland-based production systems, energy supplements need to be provided at lower milk yield levels than protein supplements. This might partly account for the lower heFCE for energy. Another likely explanation is the efficient recycling of nitrogen in ruminants, while about 6% of GE is already lost as methane during ruminal fermentation (Flachowsky, 2003; Johnson and Johnson, 1995; Van Soest, 1994). When discussing the conversion efficiency of human-edible protein, the amino acid patterns and the biological value of the proteins also have to be taken into consideration (Smith et al., 2013). Generally, plant proteins have a relatively unfavourable amino acid composition and a lower biological value, when compared to animal proteins (Friedman, 1996; Oltjen and Beckett, 1996; Smith et al., 2013). Therefore, from a nutrition physiology point of view, even heFCE values below 1 for protein do not necessarily reflect a loss, as the protein output from a livestock production system is probably of higher quality for humans as compared to the (plant) feed protein input. However, profound data on the consequences of the protein quality change during this transformation process are lacking and require further research.

124

P. Ertl et al./Agricultural Systems 137 (2015) 119–125

5. Conclusions Potential heF of feedstuffs vary between 0 and 100%. The method presented herein allows for a sound reflection of this variability in heF, which needs to be considered in the debate on the role of animal production in relation to food security and the competition between animal feed and human food. From their natural feed spectrum, ruminants and particularly dairy cows do not necessarily compete with humans for food. However, about half of the dairy farms analysed in this study showed a heFCE below 1, indicating a net reduction of potential food during the conversion from feed into the animal products. The amount of concentrates used per kg milk has been shown to be a factor that strongly influences heFCE and, if minimized, will contribute to the fulfilment of the unique role of dairy cattle as efficient converters of human-inedible feedstuffs into high quality food. In terms of heFCE, grass-based dairy production systems are more favourable than systems with high concentrate inputs, because despite lower milk yield per cow, grass-based dairy production provides a net gain in human food supply. The beneficial effect of energy supplementation of grasslandbased diets on heFCE for protein will be even more enhanced if concentrates with a low heF are utilized, thereby creating a winwin-situation from a food security perspective. Acknowledgments The authors thank the Austrian Federal Ministry of Agriculture, Forestry, Environment and Water Management (grant number 100783/3) for funding the project which the field data for this study were derived from. We particularly acknowledge the valuable comments received from two anonymous reviewers on an earlier version of this paper. References Aider, M., Barbana, C., 2011. Canola proteins: composition, extraction, functional properties, bioactivity, applications as a food ingredient and allergenicity – a practical and critical review. Trends Food Sci. Technol. 22, 21–39. http://dx .doi.org/10.1016/j.tifs.2010.11.002. Baldwin, R.L., Donovan, K.C., Becket, J.L. 1992. An update on returns on human-edible input in animal agriculture. In: Proceedings California Animal Nutrition Conference, May 1992, University of California, Davis, CA, USA. Bauer, A., Bauer, F., Hiesberger, J., Hilbert, F., Hofbauer, P., Paulsen, P., et al., 2007. Tierproduktion und veterinärmedizinische Lebensmittelhygiene: ein synoptisches Lehrbuch. Academic Publishers, Wageningen, The Netherlands. Bhatty, R.S., 1986. Physiochemical and functional (breadmaking) properties of hull-less barley fractions. Cereal Chem. 63, 31–35. BMLFUW – Bundestministerium für Land- und Forst-, Umwelt und Wasserwirtschaft. 2014. Grüner Bericht – Bericht über die Situation der österreichischen Land- und Forstwirtschaft. BMLFUW, Vienna, Austria. Boye, J., Zare, F., Pletch, A., 2010. Pulse proteins: processing, characterization, functional properties and applications in food and feed. Food Res. Int. 43, 414–431. http://dx.doi.org/10.1016/j.foodres.2009.09.003. Bradford, G.E., 1999. Contributions of animal agriculture to meeting global human food demand. Livest. Prod. Sci. 59, 95–112. http://dx.doi.org/10.1016/S0301 -6226(99)00019-6. Buttchereit, N., Stamer, E., Junge, W., Thaller, G., 2010. Evaluation of five models fitted for fat:protein ratio of milk and daily energy balance. J. Dairy Sci. 93, 1702–1712. http://dx.doi.org/10.3168/jds.2009-2198. Cassidy, E.S., West, P.C., Gerber, J.S., Foley, J.A., 2013. Redefining agricultural yields: from tonnes to people nourished per hectare. Environ. Res. Lett. 8, 034015. http://dx.doi.org/10.1088/1748-9326/8/3/034015. CAST – Council for Agricultural Science and Technology. 1999. Animal Agriculture and Global Food Supply. Task Force Report No. 135. Page 105. Ames, Iowa, USA. CAST – Council for Agricultural Science and Technology. 2013. Animal Feed vs. Human food: challenges and Opportunities in Sustaining Animal Agriculture Toward 2050. Issue Paper 53. Vol. 53, Ames, Iowa, USA. Cheli, F., Battaglia, D., Gallo, R., Dell’Orto, V., 2014. EU legislation on cereal safety: an update with a focus on mycotoxins. Food Control 37, 315–325. http://dx .doi.org/10.1016/j.foodcont.2013.09.059. de Vries, M., de Boer, I.J.M., 2010. Comparing environmental impacts for livestock products: a review of life cycle assessments. Livest. Sci. 128, 1–11. http:// dx.doi.org/10.1016/j.livsci.2009.11.007.

DLG. 1997. DLG-Futterwerttabelle für Wiederkäuer. 7. ed. DLG-Verl., Frankfurt am Main. Eckhoff, S.R., Du, L., Yang, P., Rausch, K.D., Wang, D.L., Li, B.H., et al., 1999. Comparison between alkali and conventional corn wet-milling: 100-g procedures. Cereal Chem. 76, 96–99. http://dx.doi.org/10.1094/Cchem.1999.76.1.96. Eisler, M.C., Lee, M.R.F., Tarlton, J.F., Martin, G.B., Beddington, J., Dungait, J.A.J., et al., 2014. Steps to sustainable livestock. Nature 507, 32–34. Ertl, P., Zebeli, Q., Zollitsch, W., Knaus, W., 2015. Feeding of by-products completely replaced cereals and pulses in dairy cows, and enhanced edible feed conversion ratio. J. Dairy Sci. 98, 1225–1233. http://dx.doi.org/10.3168/jds.2014-8810. FAO. 2011. World Livestock 2011 – Livestock in Food Security. FAO, Rome, Italy. Fischer, M., Kofod, L.V., Schols, H.A., Piersma, S.R., Gruppen, H., Voragen, A.G.J., 2001. Enzymatic extractability of soybean meal proteins and carbohydrates: heat and humidity effects. J. Agric. Food Chem. 49, 4463–4469. http://dx.doi.org/ 10.1021/Jf010061w. Flachowsky, G., 2003. Contributions of agriculture to improve food security and food safety. Fresenius Environ. Bull. 12, 467–489. Foley, J.A., Ramankutty, N., Brauman, K.A., Cassidy, E.S., Gerber, J.S., Johnston, M., et al., 2011. Solutions for a cultivated planet. Nature 478, 337–342. http://dx.doi .org/10.1038/Nature10452. Friedman, M., 1996. Nutritional value of proteins from different food sources. A review. J. Agric. Food Chem. 44, 6–29. http://dx.doi.org/10.1021/Jf9400167. Gill, M. 2013. Converting feed into human food: the multiple dimensions of efficiency. In: H. P. S. Makkar and D. Beever (Eds.), Proceedings of the FAO Symposium. FAO and Asian-Australasian Association of Animal Production Societies, Bangkok, Thailand. Glitso, L.V., Knudsen, K.E.B., 1999. Milling of whole grain rye to obtain fractions with different dietary fibre characteristics. J. Cereal Sci. 29, 89–97. http://dx.doi .org/10.1006/jcrs.1998.0214. Herrero, M., Thornton, P.K., 2013. Livestock and global change: emerging issues for sustainable food systems. Proc. Natl. Acad. Sci. U.S.A. 110, 20878–20881. http://dx.doi.org/10.1073/pnas.1321844111. Hofmann, R.R., 1989. Evolutionary steps of ecophysiological adaptation and diversification of ruminants: a comparative view of their digestive system. Oecologia 78, 15. Hoover, R., Hughes, T., Chung, H.J., Liu, Q., 2010. Composition, molecular structure, properties, and modification of pulse starches: a review. Food Res. Int. 43, 399–413. http://dx.doi.org/10.1016/j.foodres.2009.09.001. Hörtenhuber, S., Kirner, L., Neumayr, C., Quendler, E., Strauss, A., Drapela, T., et al. 2013. Integrative Bewertung von Merkmalen der ökologischen, ökonomischen und sozial-ethischen Nachhaltigkeit landwirtschaftlicher Produktionssysteme am Beispiel von Milchproduktionssystemen. (accessed 12.11.14). INRA, CIRAD, AFZ, FAO. 2014. Feedipedia – Animal Feed Resources Information System. (accessed 25.11.14). Johnson, K.A., Johnson, D.E., 1995. Methane emissions from cattle. J. Anim. Sci. 73, 2483–2492. Kling, M., Wöhlbier, W. 1983. Handels-Futtermittel II Teil B. Ulmer, Stuttgart, Germany. Knaus, W., 2013. Re-thinking dairy cow feeding in light of food security. AgroLife Sci. J. 2, 36–40. Lapierre, H., Berthiaume, R., Raggio, G., Thivierge, M.C., Doepel, L., Pacheco, D., et al., 2005. The route of absorbed nitrogen into milk protein. Anim. Sci. 80, 11– 22. Le Cotty, T., Dorin, B., 2012. A global foresight on food crop needs for livestock. Animal 6, 1528–1536. http://dx.doi.org/10.1017/S1751731112000377. Lee, K.M., Herrman, T.J., Rooney, L., Jackson, D.S., Lingenfelser, J., Rausch, K.D., et al., 2007. Corroborative study on maize quality, dry-milling and wet-milling properties of selected maize hybrids. J. Agric. Food Chem. 55, 10751–10763. http://dx.doi.org/10.1021/Jf071863f. Leng, R.A., 2010. Decline in available world resources: implications for livestock production systems. In: Odongo, N.E., Garcia, M., Viljoen, G.J. (Eds.), Sustainable Improvement of Animal Production and Health. FAO, Rome, Italy. Mestres, C., Louisalexandre, A., Matencio, F., Lahlou, A., 1991. Dry-milling properties of maize. Cereal Chem. 68, 51–56. Minchin, W., Buckley, F., Kenny, D.A., Monahan, F.J., Shalloo, L., O’Donovan, M., 2009. Effect of grass silage and concentrate based finishing strategies on cull dairy cow performance, carcass and meat quality characteristics. Meat Sci. 81, 93–101. http://dx.doi.org/10.1016/j.meatsci.2008.07.001. Murphy, P.A., Hendrich, S., Landgren, C., Bryant, C.M., 2006. Food mycotoxins: an update. J. Food Sci. 71, R51–R65. http://dx.doi.org/10.1111/j.1750-3841 .2006.00052.x. Nocek, J.E., Russell, J.B., 1988. Protein and energy as an integrated system – relationship of ruminal protein and carbohydrate availability to microbial synthesis and milk-production. J. Dairy Sci. 71, 2070–2107. http://dx.doi.org/ 10.3168/jds.S0022-0302(88)79782-9. Oltjen, J.W., Beckett, J.L., 1996. Role of ruminant livestock in sustainable agricultural systems. J. Anim. Sci. 74, 1406–1409. O’Mara, F.P., 2012. The role of grasslands in food security and climate change. Ann. Bot. 110, 1263–1270. http://dx.doi.org/10.1093/aob/mcs209. Pelgrom, P.J.M., Vissers, A.M., Boom, R.M., Schutyser, M.A.I., 2013. Dry fractionation for production of functional pea protein concentrates. Food Res. Int. 53, 232–239. http://dx.doi.org/10.1016/j.foodres.2013.05.004. Peña, R.J., 2004. Food uses of triticale. In: Mergoum, M., Gómez-Macpherson, H. (Eds.), Triticale Improvement and Production. FAO, Rome, Italy.

P. Ertl et al./Agricultural Systems 137 (2015) 119–125

Peña, R.J., Amaya, A., 1992. Milling and breadmaking properties of wheat triticale grain blends. J. Sci. Food Agric. 60, 483–487. http://dx.doi.org/10.1002/jsfa .2740600413. Poppi, D.P., McLennan, S.R., 1995. Protein and energy-utilization by ruminants at pasture. J. Anim. Sci. 73, 278–290. Salgado, P.R., Ortiz, S.E.M., Petruccelli, S., Mauri, A.N., 2011. Sunflower protein concentrates and isolates prepared from oil cakes have high water solubility and antioxidant capacity. J. Am. Oil Chem. Soc. 88, 351–360. http://dx.doi.org/ 10.1007/s11746-010-1673-z. Salgado, P.R., Drago, S.R., Ortiz, S.E.M., Petruccelli, S., Andrich, O., Gonzalez, R.J., et al., 2012. Production and characterization of sunflower (Helianthus annuus L.) protein-enriched products obtained at pilot plant scale. LWT-Food Sci. Technol. 45, 65–72. http://dx.doi.org/10.1016/j.lwt.2011.07.021. Serna-Saldivar, S.O., Guajardo-Flores, S., Viesca-Rios, R., 2004. Potential of triticale as a substitute for wheat in flour tortilla production. Cereal Chem. 81, 220–225. http://dx.doi.org/10.1094/Cchem.2004.81.2.220. Shallo, H.E., Rao, A., Ericson, A.P., Thomas, R.L., 2001. Preparation of soy protein concentrate by ultrafiltration. J. Food Sci. 66, 242–246. Smith, J., Sones, K., Grace, D., MacMillan, S., Tarawali, S., Herrero, M., 2013. Beyond milk, meat, and eggs: role of livestock in food and nutrition security. Anim. Front. 3, 6–13. http://dx.doi.org/10.2527/af.2013-0002. Statistics Austria, 2014. Lebend & Schlachtgewichte. Endergebnis 2013. Vol. 2014. Statistics Austria. Stevenson, L., Phillips, F., O’Sullivan, K., Walton, J., 2012. Wheat bran: its composition and benefits to health, a European perspective. Int. J. Food Sci. Nutr. 63, 1001– 1013. http://dx.doi.org/10.3109/09637486.2012.687366. Stolp, K.-D., 2004. Maisstärke. In: Heiss, R. (Ed.), Lebensmitteltechnologie: Biotechnologische, chemische, mechanische und thermische Verfahren der Lebensmittelverarbeitung, sixth ed. Springer, Berlin, Germany.

125

Tan, S.H., Mailer, R.J., Blanchard, C.L., Agboola, S.O., 2011. Canola proteins for human consumption: extraction, profile, and functional properties. J. Food Sci. 76, R16–R28. http://dx.doi.org/10.1111/j.1750-3841.2010.01930.x. Ullrich, S.E., 2011. Barley, Production, Improvement, and Uses. Wiley-Blackwell, Chichester, West Sussex, UK; Ames, Iowa, USA. Van Soest, P.J., 1994. Nutritional Ecology of the Ruminant, second ed. Comstock Publ., Ithaca, NY, USA. Vestergaard, M., Madsen, N.T., Bligaard, H.B., Bredahl, L., Rasmussen, P.T., Andersen, H.R., 2007. Consequences of two or four months of finishing feeding of culled dry dairy cows on carcass characteristics and technological and sensory meat quality. Meat Sci. 76, 635–643. http://dx.doi.org/10.1016/j.meatsci.2007.02 .001. Vetrimani, R., Sudha, M.L., Rao, P.H., 2005. Effect of extraction rate of wheat flour on the quality of vermicelli. Food Res. Int. 38, 411–416. http://dx.doi.org/ 10.1016/j.foodres.2004.10.013. Villanueva, A., Vioque, J., Sanchez-Vioque, R., Clemente, A., Bautista, J., Millan, F., 1999. Production of an extensive sunflower protein hydrolysate by sequential hydrolysis with endo- and exo-proteases. Grasas Aceites 50, 472–476. Vose, J.R., 1980. Production and functionality of starches and protein isolates from legume seeds (Field Peas and Horsebeans). Cereal Chem. 57, 406–410. Wang, H., Wang, T., Johnson, L.A., 2005. Effect of alkali on the refunctionalization of soy protein by hydrothermal cooking. J. Am. Oil Chem. Soc. 82, 451–456. http://dx.doi.org/10.1007/s11746-005-1092-3. Wilkinson, J.M., 2011. Re-defining efficiency of feed use by livestock. Animal 5, 1014–1022. http://dx.doi.org/10.1017/S175173111100005x. Zwingelberg, H., 2004. Mahlerzeugnisse aus Weizen und Roggen. In: Heiss, R. (Ed.), Lebensmitteltechnologie: Biotechnologische, chemische, mechanische und thermische Verfahren der Lebensmittelverarbeitung, sixth ed. Springer, Berlin, Germany.