How Plants Can Feed at Fraction of Land

Abstruse

Purpose

The expected increase in demand for food raises concerns about the expansion of agricultural land worldwide. To avoid expansion, nosotros demand to focus on increasing land productivity, reducing waste, and shifting man diets. Studies exploring diet shifts so far have ignored competition for land between humans and animals. Our objective was to study the relation between country use, the share of animal protein in the homo nutrition, population size, and land availability and quality.

Methods

Nosotros used linear programming to determine minimum country required to feed a population a diet with 0–80 % of the protein derived from terrestrial domestic animals. Populations ranged from 15 million to the maximum number of people that could exist supported by the organization. The agronomical organization in kingdom of the netherlands was used equally illustration, assuming no import and export of feed and food. Daily energy and protein requirements of humans were fulfilled by a diet potentially consisting of grain (wheat), root and tuber crops (potato, sugar beet), oil crops (rapeseed), legumes (chocolate-brown bean), and animal protein from ruminants (milk and meat) and monogastrics (pork).

Results and discussion

Land is used about efficiently if people would derive 12 % of dietary protein from animals (% PA), particularly milk. The role of animals in such a nutrition is to catechumen co-products from crop production and the man food industry into poly peptide-rich milk and meat. Below 12 % PA, human being-inedible products were wasted (i.e., not used for food production), whereas above 12 % PA, additional crops had to be cultivated to feed livestock. Large populations (40 million or more) could be sustained but if fauna protein was consumed. This results from the fact that at high population sizes, land unsuitable for crop production was necessary to come across dietary requirements of the population, and contributed to food production by providing creature protein without competing for country with crops.

Conclusions

A country use optimization model including crop and animal product enables identification of the optimal % PA in the diet. Land use per capita was lowest at 12 % PA. At this level, animals optimally consume co-products from food product. Larger populations, furthermore, can be sustained simply with diets relatively high in % PA, as land unsuitable for ingather production is needed to fulfil their nutrient demand. The optimal % PA in the human being diet depended on population size and the relative share of land unsuitable for ingather production.

Introduction

Global nutrient demand is projected to increase past 60 % by 2050 (Alexandratos and Bruinsma 2012), considering of a growing world population and increasing wealth. This increased demand for food has raised concerns near environmental impacts related to expansion of agricultural country worldwide (Foley et al. 2011). Pressure on land increases not simply because of future nutrient demands just also because of state degradation (Stringer 2008) and increasing demands for biofuels (OECD/FAO 2014), biomaterials, housing, and infrastructure.

Currently, agronomics already occupies about 38 % of the terrestrial surface of the World, divided among1.5 billion ha of cropland and three.4 billion ha of pastures (Alexandratos and Bruinsma 2012). Meeting the food demand projected for 2050 may require an boosted 0.ii to 1 billion ha of state under agronomics (Tilman et al. 2011). This boosted state will include state of relatively low fertility and productivity and volition be partly located in currently forested or protected areas (Alexandratos and Bruinsma 2012; Foley et al. 2011; Ramankutty et al. 2002). Converting such forested lands to agricultural country conflicts with the demand for nature preservation (Regal Social club of London 2009; Smith et al. 2010; World Bank 2007) and leads to adverse environmental effects (DeFries et al. 2004; Gerber et al. 2013; Millennium Ecosystem Cess 2005; Pielke et al. 2002).

There is considerable agreement, therefore, that humans should minimize further expansion of agricultural land. Limiting global state expansion for nutrient production, still, requires a combination of interventions on the production and consumption side (Foley et al. 2011). Proposed strategies include increasing yields on underperforming lands (Van Ittersum et al. 2013), reducing waste (Papargyropoulou et al. 2014), and shifting human being diets (Stehfest et al. 2009; Wirsenius et al. 2010).

Studies exploring the potential contribution of dietary shifts generally conclude that (i) a vegan diet requires the to the lowest degree land (Hallström et al. 2015) and (two) that land employ decreases when ruminant meat is replaced by monogastric meat (Stehfest et al. 2009; Wirsenius et al. 2010). These studies, yet, do non consider the contest between humans and animals for land. Animals fed with cereals, for instance, directly compete with humans for land. No affair how efficiently produced, straight consumption of cereals by humans is ecologically more efficient than consumption of animal-source food produced by animals fed with these cereals (Foley et al. 2011; Godfrey et al. 2010). Compared to pigs or poultry, ruminants mostly consume less feed that can be consumed directly by humans (De Vries and De Boer 2010; Vellinga et al. 2009). Ruminants, nonetheless, can notwithstanding compete with humans for land, as some of the world'southward grasslands are also suitable for product of arable crops (Suttie et al. 2005). To limit global country apply for nutrient production, therefore, we should swallow livestock products from systems that utilise land that is unsuitable or less suitable for crop production and/or that utilise co-products from nutrient production (Van Zanten et al. 2015). The objective of this written report, therefore, was to identify which factors influence the relation betwixt land use, the share of animate being poly peptide in the human diet, population size, and land availability and quality. We determined the minimum corporeality of land used to feed a growing population a diet varying in the per centum of the poly peptide derived from terrestrial domestic animals. The agricultural system in the Netherlands was used as case-study, bold no import and consign of feed and nutrient.

Material and methods

This study was based on a state use optimization model created in General Algebraic Modelling System (GAMS) version 24.2.

Organisation definition

The system in our case study consisted of production, processing, and consumption of food in kingdom of the netherlands as a standalone system (Fig. ane). The objective of this system was to produce human-edible energy and protein for domestic use. The model estimated the country area required to feed populations ranging from fifteen million (close to the electric current population size) to the maximum number of people that could exist supported by the arrangement. Inside this range, nosotros increased population size past steps of v million people. Every bit we approached the maximum number of people, we increased population size with steps of 0.1 1000000 people. Daily per capita requirements were defined every bit 2000 kcal and 57 g protein (EFSA 2009; EFSA 2012). Full sugar intake was limited to the maximum recommended intake level of ninety g per capita per mean solar day (EFSA 2009). We estimated land use for homo diets varying in the percentage of the protein derived from terrestrial domestic animals ("protein derived from animals" or PA) betwixt 0 % PA (a vegan diet) to 80 % PA. Within this range, we increased % PA past steps of 5 % (and by steps of i % where relevant). Land utilise was determined for cultivation of crops and forages.

Fig. ane

Diagram of the organisation

Full size image

Ingather production system

The electric current Dutch agricultural area of 1842 × 10³ ha (CBS 2013) represented the maximum available area for production of crops and forages. This area consists of clay soils (839 × 10³ ha), sandy soils (779 × 10³ ha), and peat soils (224 × 10³ ha) (Lesschen et al. 2012). Clay and sandy soils can be used for cultivation of crops and forages, whereas peat soils were assumed to be suitable only for cultivation of grass, since most of these are too wet for competitive crop production. For the fact that they are marginal for crop production, peat soils represent so-called marginal lands in this report.

Crops included in the modeling represented the following major groups: grains, root and tuber crops, oil crops, and legumes (Electronic Supplementary Material I). In each nutrient crop grouping, the abundant crop with the largest cultivated expanse in kingdom of the netherlands was chosen (LEI and CBS 2012). Grains were represented past wheat, root, and tuber crops by potatoes and sugar beets, oil crops by rapeseed, and legumes past brown beans. Seven ingather rotations were adopted from Van Ittersum et al. (1995), the length of rotations varying from one to half-dozen years (Electronic Supplementary Textile I). Crop yields were based on current Dutch averages. To compute the almanac yield of each crop within a rotation, nosotros multiplied the crop's yield with its frequency in the rotation (i.due east., years of cultivation in a rotation divided by total years of rotation). In the harvested ingather, we distinguished the main ingather product and human being-inedible products (i.e., wheat harbinger, sugar beet tops and tails, and rapeseed straw). Wheat and maize stubble, tater haulms, sugar beet leaves, and bean straw were left on the field equally source of soil organic carbon. We lowered actual yield levels past 7 % for potatoes, 5 % for beans, and 2 % for wheat and rapeseed to account for product of seeds and seedlings (PPO 2009; PPO 2012). Hence, seed and seedling production was already deemed for in crop yields in our calculations (Y _i,j,t ). In add-on to crops, we considered production of maize silage and grass as forage for dairy cattle. We assumed no effects on yields of climatic differences across the netherlands (Van Wart et al. 2013).

To determine total dry out matter production (ton DM) of harvested product j (Q _j ), we multiplied the state area allocated to crop rotation i on land type l (Ten _i,l ) by the fresh matter yield of harvested product j from the same crop rotation and country type (Y _i,j,l ), and by the dry affair content of harvested product j (DM _j ), then summed across rotations (i = 1,seven) and land types (l = 1,3; Eq. (1)).

$$ {Q}_j={\displaystyle \sum_{i=1}^7}{\displaystyle \sum_{fifty=i}^3}{X}_{i,50}\times {Y}_{i,j,l}\times D{G}_j $$

(1)

Processing of crops

Harvested products (e.m., rapeseed) were divided into food and feed products (e.g., oil and meal) following ratios of dry affair output and dry out thing input of various processing steps (Electronic Supplementary Fabric Two). These dry affair output/input ratios were calculated from fresh matter output/input ratios (Mattsson et al. 2001; Vellinga et al. 2013) and dry matter contents of nutrient (RIVM 2013) and feed products (PDV 2011). To determine dry thing product of nutrient or feed product thou (Q _k ), we multiplied product quantities of harvested product j assigned to process 1000 (Q _j,m ) with the output/input ratio of product g produced from harvested product j in procedure m (C _j,k,k ) (Eq. (2)). Processes that exercise not divide i product into multiple products (e.thousand., ensiling of grass and heating of beans) were assigned an output/input ratio of one.

$$ {Q}_k={Q}_{j,m}\times {C}_{j,chiliad,one thousand} $$

(two)

We converted production quantities of human-edible products into bachelor energy, protein and sugar, using nutrient contents of products found in the Dutch food database NEVO (RIVM 2013). Similarly, for brute feeds (see Sect. 2.4), we converted production quantities into nutrients using feed tables (PDV 2011).

Animal production system

We included two brute production systems with contrasting abilities to use marginal state. We chose pig product equally representative for monogastrics, a system that derives its feeds from land suitable for cultivation of crops, and dairy product equally representative for ruminants, able to apply marginal land. Nosotros chose these systems as within the group of monogastrics and ruminants, pork and dairy products are the largest contributors to protein in the homo diet (PPE/PVV 2013; RIVM 2011).

Product levels of animals were based on Dutch averages. Nosotros modeled hog and dairy product based on animal production units (PUs) per animal place per year. One pig PU consisted of 3.three fattening pigs, 0.12 sows, and 0.07 gilts (Electronic Supplementary Material Three). One pig PU produced 171 kg pork per yr, which corresponds to 1475 MJ and 55 kg human-edible protein per year. Cyberspace energy requirements per pig PU (equivalent to the weighted sum of net energy requirements (PDV 2012) for fattening pigs, sows and gilts) totaled 9901 MJ per twelvemonth (Electronic Supplementary Textile IV). The nutrition of one hog PU had a minimum of xvi % and a maximum of 18 % rough protein (Bikker 2014; Devendra and Clyde Parris 1970) and a digestibility coefficient of at least 80 % (Bikker 2014). Grass, maize silage, straw, sugar beet tops and tails, and saccharide manufacturing plant lime were excluded from consumption past pigs (Bikker 2014). We applied additional restrictions to create a plausible diet (Electronic Supplementary Textile 5).

One moo-cow PU consisted of a dairy cow and its replacement stock, i.e., 0.31 heifers aged one–two years, and 0.34 calves anile 0–one year (Electronic Supplementary Material Vi). Surplus calves were excluded from our analysis. 1 cow PU produced 8502 kg fat-and-protein-corrected-milk and 74 kg meat per twelvemonth, both derived only from the milking cow, corresponding to 22,775 MJ, 303 kg human-edible protein and 383 kg total carbohydrate per year (Electronic Supplementary Fabric Six). Net free energy requirements for one moo-cow PU (equivalent to the weighted sum of net energy requirements (PDV 2012) for the milking cow, replacement heifer, and calf) totaled 51,977 MJ and 606 kg abdominal digestible protein per yr (Electronic Supplementary Material Seven and 8). Rumen degradable poly peptide balance had a lower limit of 0 and an upper limit of 200 one thousand per cow per day (Dijkstra 2014). To assure sufficient structure in the diet, the construction value of the diet (PDV 2012) was at least 1 per kg DM. Maximum feed intake capacity was limited to 14.9 saturation values per day for dairy cows, 3.2 for replacement heifers, and five.9 for replacement calves (PDV 2012). Saccharide mill lime was excluded from consumption past cows. We practical additional restrictions to create a plausible diet (Electronic Supplementary Material V).

Manure production and awarding

Nutrient (i.east., nitrogen and phosphorus) excretion past animals was computed as the difference between nutrient intake and nutrients retained in animals and their products. Nutrient intake was computed based on information about feed intake and food content of feed ingredients (PDV 2011). Food retention in growing pigs was computed from nutrient concentrations in body tissue (Groenestein et al. 2008) and product data (Electronic Supplementary Fabric Iii), and totaled nine.4 kg Northward and 2.0 kg P per pig PU per year. Food retention in milk and body tissue of culled cows and growing immature stock was computed from nutrient concentrations in torso tissue and milk (RVO 2010) and production data (Electronic Supplementary Fabric VI), and totaled fifty.one kg N and 9.64 kg P per cow PU per year.

In line with European Spousal relationship (EU) legislation, awarding of manure to crop and grassland was limited to 170 kg nitrogen from animal manure per hectare per year (RVO 2014). Additionally, we restricted total nitrogen application from manure and bogus fertilizer to the sum of crop- and soil-type-specific maximum nitrogen awarding rates allowed by Eu legislation. The nitrogen fertilizer replacement value of manure was set at threescore % (RVO 2014). Moreover, we restricted full phosphate application to the sum of soil-type-specific maximum phosphate awarding rates for grass and arable land (RVO 2014). These rates depended on the phosphate levels of the soil, every bit determined past Schoumans (2007).

Losses

We accounted for losses of food ingather products, meat, and milk by applying loss fractions (Gustavsson et al. 2011) during postharvest handling, storage, processing, packing, distribution, and consumption. We assumed that at most 21 % of full food ingather product losses could potentially be used as feed (Soethoudt and Timmermans 2013). We assumed 5 % postharvest handling and storage losses for wheat straw, sugar beet tops and tails, and rapeseed straw. In improver, we deemed for conservation and feeding losses for crop products allocated to animals. We assumed conservation losses of 4 % for moist concentrates, 5–ten % for potato pare, silage maize, and beans, xv–17.five % for grass silage and potatoes, 20 % for straw, and 25 % for sugar beet tops and tails (Remmelink et al. 2012). Feeding losses were 2 % for dried concentrates, 3 % for moist concentrates, and v % for roughages (Remmelink et al. 2012). No losses were assigned to fresh grass, as fresh grass yields represented internet production (i.east., intake) by animals.

Objective function

The linear-programming model allocated crop products to humans or animals based on an objective function to minimize state use for all crop rotations i on all land types l (X _i,j ) (Eq. (3)) while meeting energy and poly peptide requirements of the human population.

$$ Min\ {\displaystyle \sum_{i=1}^seven}{\displaystyle \sum_{l=1}^three}{X}_{i,l} $$

(3)

Sensitivity analysis

We explored the impact of changes in the share of different soil types, in crop and fodder yields and in the share of protein from meat in the creature poly peptide consumed on final results of land use and man dietary composition (Table one)

Table 1 Characteristics of the reference situation and, assessed in sensitivity analyses, alternative situations

Full size table

Changes in share of dissimilar soil types

In the reference state of affairs, land consisted of 46 % clay soils, 42 % sandy soils, and 12 % peat soils. To determine the impact of decreasing the share of marginal land (peat soil), we studied a state of affairs in which state consisted of 50 % clay soils, 45 % sandy soils, and v % peat soils (less peat). To determine the impact of increasing the share of marginal land, we studied a situation in which state consisted of 30 % clay soils, 30 % sandy soils, and forty % peat soils (more than peat).

Changes in crop and grass yields

To determine the touch on of differences in relative productivity of clay, sand, and peat soils, we decreased yields on sandy soils by twenty % and on peat soils by 50 %.

Changes in meat content of the diet

In the reference situation, we did not ready requirements for meat consumption. I possible outcome, therefore, was that PA could come mainly from milk. To determine the impact of meat consumption, we forced meat (pork and/or beef) to constitute at least fifty % of PA, as this is the current situation in kingdom of the netherlands (RIVM 2009).

Results

Land use

Reference situation

The relation between the minimum amount of country needed to feed a specific population and the percentage of the protein derived from animals (% PA) in the nutrition was nonlinear (Fig. 2). As % PA increased, land employ initially decreased upward to about 12 % PA, and later on increased. Diets with near 12 % PA, therefore, systematically had the lowest land use. Furthermore, as population size increased, the possible range of % PA in the diet became more narrow. This implies that larger populations could not exist supported by a vegan diet or a diet containing a high % PA.

Fig. 2

Minimum land (10³ ha) needed for feeding the total population with diets varying in percentage of dietary protein from animals (% PA) in the reference situation. mln 1000000

Full size prototype

The corporeality of country needed per capita increased equally population size increased (Table 2). Per capita land use increased with population size considering high yielding soils, i.e., clay soils, were cultivated kickoff, followed by sandy soils (Fig. 3). This order follows from the more often than not higher yields at rotation level on dirt soils than on sandy soils. As population size increased, therefore, the relative contribution of lower yielding soils increased, explaining the increase in per capita land apply (Table ii and Fig. 3).

Table 2 Per capita land use index for diets varying in percentage of dietary protein from animals (% PA) and various populations in the reference situation

Total size tabular array

Fig. 3

Country use (m²/capita) per soil type for diets varying in pct of dietary protein from animals (% PA) and populations of 15 million (left) and 35 million (right) people in the reference situation. For a population of 35 million people, 39 % PA was the last feasible choice. mln 1000000

Full size image

Food product on clay soils was sufficient to feed a population of fifteen meg people a diet with 0–25 % PA. In the range from 30 to threescore % PA, in addition to dirt soils, sandy soils were used for the production of silage maize, every bit silage maize had college yields on sandy soils than on clay soils. If PA exceeded 60 %, sandy soils were predominantly used for the production of crops, equally all clay soils were fully used. Feeding a population of 35 million, notwithstanding, required all dirt soils and most of the sandy soils, fifty-fifty at low % PA. From xv % PA upward, in addition to dirt and sandy soils, peat soils were used to produce grass (see Electronic Supplementary Material X for diet composition per moo-cow PU). Diets with more than 39 % PA were not feasible.

Impact of changes in share of different soil types

Decreasing the share of peat soils (i.east., from 12 % in the reference situation to five %) increased the maximum number of people that could be fed from the state, whereas increasing the share of peat soils (i.e., from 12 % in the reference situation to twoscore %) decreased the maximum number of people that could be fed from the land (Fig. 4). This difference in the number of people that can be fed tin can be explained past the higher productivity of clay and sandy soils than peat soils. Furthermore, in the state of affairs with a smaller share of peat soils, the maximum number of people (i.eastward., 43.six million) consumed diets with about 15 % PA, whereas in the situation with a larger share of peat soils, the maximum number of people (i.e., 31.5 million) consumed diets with nearly 44 % PA. In other words, when population size increases in a region with a larger share of marginal land, this population can be supported merely if a relatively loftier percentage of its protein comes from fauna sources. Moreover, a vegan diet is just viable for smaller populations in such a situation, i.east., larger populations can only be sustained when animal poly peptide is consumed. The feasible share of animal protein in the man nutrition, therefore, depends on the population size in combination with the share of marginal land.

Fig. 4

Minimum land (10³ ha) needed for feeding the total population with diets varying in percentage of dietary poly peptide from animals (% PA) in alternative situations with v % (left) and 40 % (right) of total land area underlain by peat soils. mln one thousand thousand

Full size prototype

Impact of changes in ingather and grass yields

Decreasing crop yields on sandy and peat soils did not increment per capita land use compared to the reference situation as long as % PA was less than thirty %, because only clay soils were used in that range to feed a population of 15 million in both the reference and alternative situation (Fig. 5). When % PA was thirty % or more than, per capita land use increased relatively quickly compared to the reference state of affairs (Fig. 3), because of the lower availability of highly productive land.

Fig. 5

State employ (m^two/capita) per soil type for diets varying in pct of dietary protein from animals (% PA) and populations of xv million (left) and 35 meg (right) people in the alternative situation with 20 % lower yields on sandy soils and 50 % lower yields on peat soils compared to the reference situation. For a population of 35 1000000 people, 23 % PA was the last feasible option. mln one thousand thousand

Full size epitome

We expected decreasing crop yields on sandy soils to upshot in higher land use on these soils compared to the reference state of affairs, under high population pressure or loftier % PA. For a population of 15 1000000, this was indeed the example for diets with 50 % PA and more (Fig. 5). For diets containing less than 50 % PA, however, sandy soils were not used in the alternative situation. This resulted from the relatively small departure in yield of maize silage between clay and sandy soils in the reference situation. After reducing yields by twenty % on sandy soils, yield of maize silage was higher on clay soils, which postponed the utilise of sandy soils to a higher % PA.

To feed 35 million people, more sandy soils were used then in the reference situation, due to their lower yields. From 20 % PA up, sandy soils were fully used and peat soils were needed to produce animate being poly peptide. The maximum feasible % PA for this population was lower than that in the reference situation.

Consumption of animal protein

Reference situation

When % PA was less than 10 %, daily poly peptide intake per capita equaled the recommended intake level of 57 g, simply this recommended level was often exceeded when % PA exceeded 10 % (Fig. 6) (run across Electronic Supplementary Material Ix for human diet composition). Our simulations also testify that animal protein was mainly provided by milk (fixed ratio of poly peptide from milk and beef of 14:ane) (Fig. 6), which is due to higher protein productivity of dairy cows than of pigs (De Vries and de Boer 2010).

Fig. half dozen

Per capita protein intake (g/24-hour interval) from crops, milk, beef, and pork for diets varying in percentage of dietary protein from animals (% PA) and populations of xv million (left) and 35 1000000 (right) people in the reference situation. For a population of 35 million people, 39 % AP was the last viable option. The horizontal line indicates the daily poly peptide requirement of 57 g/cap. mln million

Full size image

Impact of changes in meat content of the nutrition

When requiring that at least 50 % of the dietary poly peptide of a population of xv million came from meat (in our model, beef or pork), the percentage of dietary protein from pork gradually increased from about 2 % (i.e., PA = 5 %) to about 37 % (PA = 80 %) (Fig. 7), at the expense of dietary protein from milk in particular. Land use in this alternative scenario, therefore, is slightly college than that in the reference scenario. Hence, replacing dietary protein from milk with that from meat implies that we can eat less protein derived from animals. For a population of 35 1000000 people, maximum % PA in the nutrition decreased to 35 % (Fig. 7), compared to 39 % in the reference situation (Fig. 6) (run into Electronic Supplementary Fabric 11 for diet composition per pig PU).

Fig. 7

Per capita protein intake (1000/day) from crops, milk, beef, and pork for diets varying in percentage of dietary protein from animals (% PA) and populations of 15 one thousand thousand (left) and 35 million (right) people in the alternative situation in which meat contributed at least 50 % of PA. For a population of 35 million people, 35 % PA was the last feasible option. The horizontal line indicates the daily protein requirement of 57 one thousand/cap. mln one thousand thousand

Full size image

Discussion and conclusions

Our model provides insights into relations between land use, the share of animal poly peptide in the human nutrition, population size, and land availability and quality. We demonstrated that at lower population sizes, consumption of near 12 % of dietary protein from animal source foods resulted in the most efficient utilise of agricultural country. In the range from 0 to 20 % PA, land use remains more than or less stable, whereas beyond 20 % PA state use increases more strongly. At the highest population size that could exist supported by the state, however, the optimal percent of dietary protein derived from beast source foods ranged from 15 to 45 %.

Minimizing land employ resulted in per capita land utilize of 400–800 g²/year, values lower than actual land use values reported by Meier and Christen (2013), Terluin et al. (2013), and Van Oorschot et al. (2012). This implies that humans can use country more efficiently if they would accept ascetic diets. Diets resulting from our analysis consisted of a limited multifariousness of products because nosotros used proxies for the 5 major groups of crop production, and for monogastric and ruminant production. We retrieve, however, that including a wider variety of found-based or terrestrial domestic animal-based products would not take affected our conclusions. Products from fisheries were not considered, every bit these systems do not use land.

In our results, animal protein in the human diet consisted mainly of milk, and beefiness was consumed as a co-product of milk production (milk/beef ratio of fourteen:ane). When requiring that at least 50 % of the brute protein consumed should come from meat, pork was added to the human diet. Permit us consider, nonetheless, what would have happened had we included beefiness production from suckler beef systems. Suckler cows tin exploit marginal lands by producing beef via grazing. Beef from suckler cows, however, would have been included in the human diet simply if nosotros had defined minimum requirements for beef consumption, or if marginal lands had been suitable only for grazing of suckler cows. This is because from a land use perspective grazing of dairy cows is preferred to grazing of suckler cows considering dairy cows produce animal poly peptide more than efficiently (De Vries and De Boer 2010). Furthermore, feed produced on dirt or sandy soil is converted more efficiently to creature protein by pigs than by beef cattle. We realize, however, that producing beefiness or mutton on marginal lands unsuitable for grazing of dairy cattle can be of utmost importance in other areas, and this will result in an increase of per capita country use.

Another important finding of our written report is that a vegan nutrition ever required more land than a nutrition with small-scale amounts of fauna protein. In other words, country is used most efficiently if people consume pocket-sized amounts of animal protein, which is too referred to as the "default livestock diet" (Fairlie 2010). The function of animals in a default livestock diet is to convert co-products from arable production (e.grand., straw) and the human food industry (e.g., beet pulp) that cannot be consumed straight by humans into protein-rich milk and meat. When no animal protein is produced, every bit suits a vegan diet, these human being-inedible products are wasted (i.e., not used for nutrient product) or used as a bio-energy source, and additional crops will have to be cultivated to run into nutritional free energy and protein requirements of the population. Consequently, larger populations could not be supported by a vegan nutrition and a population cannot exceed a certain size unless animal protein is consumed.

Larger populations as well could not be supported by a diet with a high percentage of protein derived from animal source foods. A population of 35 million people, for example, could not be supported from a diet containing twoscore % PA or more. When need for beast poly peptide exceeds the default livestock diet, additional crops will take to be cultivated, resulting in higher land use. At higher population sizes this country is not available, which limits consumption of high amounts of animal protein, and, thus, the possible percentages of dietary protein derived from animal-source foods decreased. Moreover, at the highest population size that could be supported by the country, the optimal pct of dietary poly peptide derived from animal source food ranged from 15 to 45 %, the verbal value depending on assumed ingather yields and the share of marginal land. Peat soils, although relatively productive in the netherlands, were considered marginal land equally they are suitable only for grass production. Also, peat soils are somewhat less productive than dirt soils in the Netherlands. Thus, increasing the share of peat soils increased the optimal percentage of dietary protein from fauna source foods at the highest population size, but at the aforementioned time decreased the maximum number of people that could be fed. In contrast, decreasing the relative share of peat soils, and hence increasing the relative share of arable soils, would increment the number of people that could be fed and lower the optimal percent of dietary poly peptide from animal source foods at higher population sizes. The optimal percentage of dietary protein from animals in future diets, therefore, depends on the share of marginal land in the world, together with the productivity of these marginal lands (which is atypically high in the Netherlands). Moreover, the optimal per centum of dietary protein from animals besides depends on the type of crops and the extent to which co-products are harvested. A higher availability of co-products for feed would shift optimum country use to higher % PA only also reduce carbon inputs to the soil.

A final important conclusion is that our results contradict results of life cycle assessment (LCA) studies that explored land employ of diets differing in the per centum of protein derived from animals. These LCA studies advise that vegan diets require the least amount of land, followed past vegetarian diets (Hallström et al. 2015; Meier and Christen 2013). Optimization, as employed in our study, accounts for the unsuitability of marginal lands to grow crops, the suitability of animals to apply human-inedible products, and the co-production of meat and milk. These aspects are non included in LCA studies, and explain the different results. Our country-utilize optimization model could be extended to the employ of other limited resources such every bit fossil free energy and phosphorous, and the emission of, for example, greenhouse gases.

References

• Alexandratos Northward, Bruinsma J (2012) World agriculture towards 2030/2050: the 2012 revision. ESA Working paper No. 12–03. FAO, Rome

• Bikker P (2014) Wageningen UR Livestock Enquiry. Personal communication, Wageningen

  Google Scholar

• CBS (2013) Statistics Netherlands (Centraal Agency voor de Statistiek). http://statline.cbs.nl/statweb/

• De Vries Chiliad, De Boer IJM (2010) Comparing environmental impacts for livestock products: a review of life bike assessments. Livest Sci 128:1–xi

  Article  Google Scholar

• DeFries RS, Foley JA, Asner GP (2004) Land-use choices: balancing man needs and ecosystem function. Forepart Ecol Environ ii:249–257

  Article  Google Scholar

• Devendra C, Clyde Parris EC (1970) Optimum crude protein levels for growing and finishing pigs in a tropical surroundings. Trop Anim Wellness Pro ii:162–166

  Article  Google Scholar

• Dijkstra J (2014) Personal advice. Creature Sciences. Wageningen University, Wageningen

  Google Scholar

• EFSA (2009) Scientific Opinion of the Console on Dietetic products, Nutrition and Allergies on a request from European Committee on the review of labelling reference intake values for selected nutritional elements. EFSA Journal 1008:i–fourteen

  Google Scholar

• EFSA (2012) EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA); Scientific Opinion on Dietary. Reference Values for poly peptide. EFSA Journal 10:2557

  Google Scholar

• Fairlie S (2010) Meat: a beneficial extravagance. Hyden Business firm, Petersfield

  Google Scholar

• Foley JA et al (2011) Solutions for a cultivated planet. Nature 478:337–342

  CAS  Article  Google Scholar

• Gerber PJ, Steinfeld H, Henderson B, Mottet A, Opio C, Dijkman J, Falcucci A, Tempio G (2013) Tackling climatic change through livestock—a global assessment of emissions and mitigation opportunities. Food and Agriculture System of the United Nations, Rome

• Godfrey HCJ et al (2010) Food security: the claiming of feeding nine billion people. Scientific discipline 327:812–818

  Article  Google Scholar

• Groenestein CM, Van Bruggen C, Hoeksma P, Jongbloed AW, Velthof GL (2008) Nadere beschouwing van stalbalansen en gasvormige stikstofverliezen uit de intensieve veehouderij. Wageningen University and Research Center, Wageningen

  Google Scholar

• Gustavsson J, Cederberg C, Sonesson U, van Otterdijk R, Meybeck A (2011) Global food losses and food waste. Extend, causes and prevention. Food and Agriculture Arrangement of the United nations, Rome

  Google Scholar

• Hallström E, Carlsson-Kanyama A, Börjesson P (2015) Ecology impact of dietary change: a systematic review. J Cleaner Prod 91:ane–11

  Article  Google Scholar

• LEI and CBS (2012) Land- en tuinbouwcijfers 2012 [Agronomical and horticultural figures]. Landbouw Economisch Instituut (LEI), The Hague

  Google Scholar

• Lesschen JP, Heesmans H, Mol-Dijkstra J, van Doorn A, Verkaik Due east, van den Wyngaert I, Kuikman P (2012) Mogelijkheden voor koolstofvastlegging in de Nederlandse landbouw en natuur. Alterra Wageningen UR, Wageningen

  Google Scholar

• Mattsson B, Wallén Eastward, Blom A, Stadig Thou (2001) Livscykelanalys av matpotatis (Life cycle assessement of potatoes). SIK - The Swedish Establishment for Food and Biotechnology. Every bit cited by Gustavsson J, Cederberg C, Sonesson U, van Otterdijk R, Meybeck A 2011: Global food losses and food waste matter. Extend, causes and prevention, Food and Agronomics Organization of the United Nations, Rome, Italy

• Meier T, Christen O (2013) Ecology impacts of dietary recommendations and dietary styles: Frg as an case. Environ Sci Technol 47:877–888

  CAS  Article  Google Scholar

• Millennium Ecosystem Assessment (2005) Ecosystems and Human Well-existence: Synthesis

• OECD/FAO (2014) Biofuels, in OECD-FAO Agronomical Outlook 2014. OECD Publishing

• Papargyropoulou Due east, Lozano RK, Steinberger J, Wright N, Ujang ZB (2014) The food waste hierarchy as a framework for the management of food surplus and food waste. J Cleaner Prod 76:106–115

  Article  Google Scholar

• PDV (2011) CVB Veevoedertabel 2011. Chemische samenstellingen en nutritionele waarden van voedermiddelen. Production board for beast feed, The Hague

  Google Scholar

• PDV (2012) Tabellenboek Veevoeding 2012. Production Lath Animal Feed, The Hague

  Google Scholar

• Pielke RA Sr et al (2002) The influence of land-use change and landscape dynamics on the climate system: relevance to climate-change policy beyond the radiative issue of greenhouse gases. Philos T Roy Soc A 360:1705–1719

  CAS  Article  Google Scholar

• PPE/PVV (2013) Cattle, meat and eggs in the Netherlands. Indicators 2012 (Vee, Vlees en Eieren in Nederland Kengetallen 2012). Zoetermeer, kingdom of the netherlands (in Dutch)

• PPO (2009) Kwantitatieve Informatie Akkerbouw en Vollegrondsgroenteteelt 2009. Praktijkonderzoek Plant & Omgeving B.V. Businessunit Akkerbouw, Groene ruimte en Vollegrondsgroenten. Wageniningen UR, The Netherlands

• PPO (2012) Kwantitatieve Informatie Akkerbouw en Vollegrondsgroenteteelt 2012. Praktijkonderzoek Constitute en Omgeving. Stichting Dienst Landbouwkundig Onderzoek, Wageningen UR

  Google Scholar

• Ramankutty Northward, Foley JA, Norman J, McSweeney G (2002) The global distribution of cultivable lands: current patterns and sensitivity to possible climate change. Global Ecol Biogeogr xi:377–392

  Article  Google Scholar

• Remmelink GJ, Blanken K, Middelkoop JCV, Ouweltjes W, Wemmenhove H (2012) Handboek Melkveehouderij 2012. Wageningen UR Livestock Research, Lelystad

  Google Scholar

• RIVM (2009) Consumptiepeiling, Jongvolwassenen, consumptie productgroepen, bronnen van macronutriënten. Rijkinstituut voor Volksgezondheid en Milieu. Ministerie van Volksgezondheid Welzijn en Sport

• RIVM (2011) Contribution (weight %) of food products to the intake of energy and macronutrients for people in the historic period of 19–30 years. Young adults, consumption of product groups and sources of macronutrients. (Bijdrage (gewogen %) van voedingsmiddelen aan de inneming van energie en macronutriënten voor 19-30-jarigen. Jongvolwassenen, consumptie productgroepen, bronnen van macronutriënten). In Dutch. . National Institute for Public Health and the Surroundings, the Netherlands, www.voedselconsumptiepeiling.nl (Version 11–2009)

• RIVM (2013) NEVO Table 2013. National Institute for Public Health and the Surroundings, The Netherlands

  Google Scholar

• Royal Society of London (2009) Reaping the benefits: science and the sustainable intensification of global agriculture. London

  Google Scholar

• RVO (2010) Guidance subcontract-specific excretion of dairy (Handreiking bedrijfsspecifieke excretie melkvee. Versie per 1 januari 2010 van kracht)

• RVO (2014) Tabellen Mestbeleid 2014–2017. Land Bureau for enterprising Netherlands (Rijksdienst voor Ondernemend Nederland)

  Google Scholar

• Schoumans OF (2007) Trends in de fosfaattoestand van landbouwgronden in Nederland in de periode 1998–2003. Alterra, function of Wageningen University and Research Middle, Wageningen

• Smith P et al (2010) Contest for state. Philos T Roy Soc B 365:2941–2957

  Article  Google Scholar

• Soethoudt H, Timmermans T (2013) Monitor food waste (Monitor voedselverspililng). Mid-term report. . Wageningen UR Nutrient & Biobased Enquiry, Wageningen, kingdom of the netherlands

• Stehfest E, Bouwman L, van Vuuren DP, den Elzen MGJ, Eickhout B, Kabat P (2009) Climate benefits of changing diet. Climate Modify 95:83–102

  CAS  Article  Google Scholar

• Stringer L (2008) Can the Un Convention to Gainsay Desertification guide sustainable use of the world's soils? Front Ecol Environ 6:138–144

  Commodity  Google Scholar

• Suttie JM, Reynolds SG, Battello C (2005) Grasslands of the world. FAO, Rome

  Google Scholar

• Terluin IJ, Kamphuis BM, Oudendag DA, Van Leeuwen MGA (2013) Voedselvoorziening in Nederland onder buitengewone crisisomstandigheden. LEI, Social Economics Research Institute, Wageningen

  Google Scholar

• Tilman D, Balzer C, Colina J, Befort BL (2011) Global food demand and the sustainable intensification of agriculture. Proc Natl Acad Sci 108:20260–20264

  CAS  Article  Google Scholar

• Van Ittersum MK, Hijmans RJ, Scheele D (1995) escription and user guide of GOAL-QUASI: an IMGLP model for the exploration of future land utilize vol 2. Research Institute for Agrobiology and Soil Fertility. The C.T. de Wit Graduate Schoolhouse for Production Ecology, Wageningen

  Google Scholar

• Van Ittersum MK, Cassman KG, Grassini P, Wolf J, Tittonell P, Hochman Z (2013) Yield gap assay with local to global relevance—a review. Field Crop Res 143:iv–17

  Article  Google Scholar

• Van Oorschot M, Rood T, Vixseboxse East, Wilting H, Van der Esch S (2012) De Nederlandse voetafdruk op de wereld: hoe groot en hoe diep? Planbureau voor de Leefomgeving (PBL), Den Haag

  Google Scholar

• Van Wart J et al (2013) Employ of agro-climatic zones to upscale simulated crop yield potential. Field Crop Res 143:44–55

  Article  Google Scholar

• Van Zanten HHE, Mollenhorst H, Klootwijk CW, Van Middelaar CE, De Boer IJM (2015) Global food security: country use efficiency of livestock systems. Int J of Life Cycle Assess (in press)

• Vellinga Boob tube, Van Laar H, Thomasse MA, De Boer IJM, Berkhout P, Aiking H (2009) Environmental bear upon of animal feed (Milieueffecten van diervoeders). Report 205. Animate being Sciences Grouping, Wageningen University and research center, Lelystad, Holland

• Vellinga Goggle box, Blonk H, Marinussen G, Van Zeist WJ, De Boer IJM, Starmans D (2013) Methodology used in FeedPrint: a tool quantifying greenhouse gas emissions of feed production and utilization. Wageningen UR Livestock Research, Lelystad

  Google Scholar

• Wirsenius S, Azar C, Berndes G (2010) How much country is needed for global food production under scenarios of dietary changes and livestock productivity increases in 2030? Agr Syst 103:621–638

  Commodity  Google Scholar

• World Depository financial institution (2007) Earth Development Report 2008: Agriculture for evolution. Washington DC

  Volume  Google Scholar

Download references

Acknowledgments

We thank Marcel Lubbers, Bert Rijk (Plant Production Systems Grouping, Wageningen University) and Jantine van Middelkoop (Wageningen UR Livestock Research) for their contributions to the country use model, crop product data, and livestock data, respectively, and Mike Grossman for language editing.

Author data

Affiliations

1. Animal Production Systems Group, Wageningen University, P.O. Box 338, 6700 AH, Wageningen, Holland

   Heleen R. J. Van Kernebeek, Simon J. Oosting & Imke J. Yard. De Boer

2. Institute Production Systems group, Wageningen University, Wageningen, The Netherlands

   Martin K. Van Ittersum

3. Livestock Research, Wageningen University and Inquiry Centre, Wageningen, The Netherlands

   Paul Bikker

Corresponding author

Correspondence to Heleen R. J. Van Kernebeek.

Boosted information

Responsible editor: Thomas Jan Nemecek

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Rights and permissions

Open up Admission This article is distributed under the terms of the Creative Eatables Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted apply, distribution, and reproduction in any medium, provided yous give advisable credit to the original author(s) and the source, provide a link to the Creative Eatables license, and indicate if changes were made.

Reprints and Permissions

Nearly this article

Cite this article

Van Kernebeek, H.R.J., Oosting, Southward.J., Van Ittersum, K.M. et al. Saving land to feed a growing population: consequences for consumption of crop and livestock products. Int J Life Cycle Assess 21, 677–687 (2016). https://doi.org/ten.1007/s11367-015-0923-6

Download citation

• Received: xvi Dec 2014

• Accustomed: 16 June 2015

• Published: 15 July 2015

• Issue Date: May 2016

• DOI : https://doi.org/10.1007/s11367-015-0923-6

Keywords

• Animal product
• Contest for resource
• Crop production
• Homo diets
• Land employ
• Optimization

bardenalievespecon67.blogspot.com

Source: https://link.springer.com/article/10.1007/s11367-015-0923-6

Share this post