Til innholdet




Prosjektnummer: 900568
Status: Avsluttet
Startdato: 10.12.2010
Sluttdato: 01.11.2011

Resource budgets and analysis of feed ingredients in salmon diets: Main project / Ressursregnskap og analyse av fôrråvarer i lakseoppdrett: Hovedprosjekt

Results achieved
Executive summary

Atlantic salmon (Salmo salar) is the dominating species in Norwegian aquaculture, and in 2010, 927 876 tons were slaughtered. The sustainability of the salmon industry has been questioned, and the salmon industry has been criticized for the use of fish meal and oil in the production of salmon feed. At present, 27 per cent of the global fish meal production and 68 per cent of the fish oil production is used in feed for salmonids worldwide. Two decades ago, the main ingredients for Norwegian salmon feed were fish meal and fish oil. However, in 2010 only 52 per cent of the ingredients were of marine origin, and the remaining 47 per cent of plant origin (on dry matter basis). The fish-in-fish-out (FIFO) ratio is the amount of forage fish used to produce the amount of fish oil and meal required to produce 1 kg of salmon. The FIFO ratio for fish oil and fish meal in Norwegian fish farming has decreased from 7.2 and 4.4 in 1990 to 2.3 and 1.4, respectively, in 2010. When correcting for use of by-products from capture fisheries, the 2010 values were 1.8 and 1.1, respectively. The limited supply of fish meal and fish oil makes this shift from marine towards plant ingredients necessary, but introduces other challenges from a sustainability perspective.

The Food and Agricultural Organization (FAO) projects that 70 per cent more food need to be produced globally within 2050 to feed a population of 9 billion people and calls for urgent action in developing food systems that uses less energy and emits less greenhouse gases (FAO 2011a). The global food sector is responsible for around 30 per cent of the world’s energy consumption and contributes to more than 20 per cent of the global greenhouse gas (GHG) emissions (FAO 2011b). In addition, land use changes (mainly through deforestation) contribute to another 15 per cent of GHG emissions.

This increase in food production will have to come through improvements in efficiency of livestock systems because most of the land area suitable for agriculture is already utilized. 30 per cent of the worlds cereal production is currently used to feed livestock, and livestock productions also consume large amounts of freshwater, both for irrigation of feed crops and for drinking. Freshwater is becoming increasingly scarce and the livestock sector is probably the largest source of water pollution (FAO 2011b). The expansion and intensification of the livestock production sector the last decades has led to degradation of 20 per cent of the world’s pastures due to overgrazing. Deforestation to grow animal feed crops has led to extinction of many plants and animals and released large amounts of carbon dioxide into the atmosphere. The global food production is also heavily dependent on the use of phosphorus fertilizer. However, the current use of phosphorus is not sustainable due to losses at all stages from mining to crop field to human consumption. Phosphorus is not cycled at present, but moves through an open one way system where the final losses end up in the ocean.

Several indicators and methods for measuring sustainability and eco-efficiency of aquaculture productions have been developed, such as the simple fish-in-fish-out-ratio, forage fish dependency ratio, marine nutrient dependency ratio and various nutrient retention ratios. More extensive methods such as the ecological footprint model and life cycle analysis (LCA) are also applied for assessing the sustainability of aquaculture and other food production system. These methods have their strengths and weaknesses, and the outcome of an analysis will depend on which impacts are included in the analysis and how the impacts are allocated between co-products in production processes that generate several products.

Evaluation of sustainability of aquaculture is complicated, and different aspects have to be addressed in order to evaluate the sustainability of Norwegian salmon production. There is currently no single method that is robust enough to cover all environmental impacts related to food production and several methods must be used in combination to evaluate the ecoefficiency of food production.

A Life cycle analysis (LCA) was performed for production of salmon with the impact factors
i) occupation of agricultural land,
ii) energy use,
iii) carbon footprint and
iv) ocean primary production using five differently formulated feeds:
Diet 1: The average commercial feed in 2010,
Diet 2: High content of marine ingredients (88 per cent of the diet),
Diet 3: 2010 diet with marine ingredients only from the North Atlantic,
Diet 4: Containing poultry by-products, and
Diet 5: High content of plant ingredients (85 per cent of the diet).

In conclusion, considerable changes in the salmon diet formulation did only cause minor changes in the carbon footprint except for the diet containing a high amount of poultry by-products (2020 LAP) which had a higher carbon footprint (3.4 CO2e/kg, similar to Swedish chicken). This is a consequence of allocating the carbon footprint from poultry production to the poultry by-products according to their mass. Changing the diet composition from 85 per cent plant ingredients to 88 per cent marine ingredients resulted in almost the same carbon footprint (2.47 and 2.40 CO2e/kg respectively). Excluding marine ingredients from South America and the Mexican Gulf from the 2010 diet increased the carbon footprint with 7 per cent to 2.75 CO2e/kg.

The Norwegian farmed salmon has a lower climate impact than the Swedish pig and chicken. The CO2 footprint for the farmed salmon was 2.6 kg CO2 equivalents/kg edible product in 2010, whereas the CO2 footprint for chicken and pig production was 3.4 and 3.9 kg CO2 equivalents/kg edible product respectively. The land occupation per kg edible product of Norwegian salmon was 3.32 m2/kg which is lower than that of both Swedish pig (8.35 m2/kg) and chicken (6.95 m2/kg). Increasing the content of plant ingredients to 85 per cent of the salmon diet will require 5.55 m2 agricultural land to produce one kg of edible product. Production of 1 kg of edible chicken and pig require 2-3 times more phosphorus fertilizer compared to salmon production. In addition, salmon retain roughly twice as much of the phosphorus in the diet compared to chicken and pig.

The total agricultural area used for production of Atlantic salmon in 2010 was 5440 km2 which is equivalent to half of the total cropland area in Norway. The total industrial energy input for the 2010 production of salmon in Norway was 41 400 TJ. 95 per cent of the industrial energy input was used for harvest, production and transport of feed ingredients and feed. The ratio industrial energy input/energy output in the salmon product was 3.6 per kg live weight and 6.2 per kg edible product respectively. For tracing of nutrient flows and estimating the nutrient retention efficiency mass balance models are more suited than LCA models. Access to representative data on nutrient composition of the feed, final product and, particularly in the parts of the salmon that are not consumed by humans, was vital for tracking the nutrient flows when making a resource budget for the Norwegian salmon production in 2010. The Norwegian aquaculture industry has an accurate system for reporting detailed aquaculture production data, and information of ingredients used for feed production in 2010 was provided by BioMar, Ewos and Skretting. Marine Harvest provided data on nutrient content in salmon. Data on fish composition was also obtained from official databases (Nifes sjømatdata, Matvaretabellen). With this information, the total nutrient flow in Norwegian salmon farming in 2010 could be estimated.

In 2010, Norwegian salmon farming consumed 1 236 000 tons of feed, with an energy content close to 31 000 MJ, and 460 853 tons of protein. In total, 612 097 tons of salmon fillet, containing 6 646 390 GJ, and 121 807 tons of protein was produced. Salmon is an important source of the nutritionally important fatty acids EPA and DHA, and of the 49 373 tons of EPA+DHA in the feed, 12 909 tons were retained in the edible part of salmon. The retention of EPA and DHA was 58 per cent in the whole body and 26 per cent in the fillet. The retention of protein and energy was 26 and 21 per cent in the edible part, respectively. These retention data can however not be compared to single productions or controlled studies, since all losses during the production of feed and salmon are included in the data used in the present study.

The conclusion from this study is that salmon farming is a more efficient way of producing nutrients for human consumption compared to chicken and pork production. Salmon farming occupies less agricultural land, uses less of the non-renewable phosphorus resources and has lower climate impact per kg product produced for human consumption. Salmon also retain the nutrients in the feed more efficiently than chicken and pig and is thus a more efficient converter of feed nutrients to nutrients for human consumption compared to land animal productions such as chicken and pig. Theoretical calculations indicate that using fish meal and oil from capture fisheries for salmon production may in fact provide more marine protein, energy and EPA and DHA for human consumption compared to utilizing the marine fishery resources directly as a human food source.
Part 2 of the project is published as a report evaluating strenghts, weaknesses, opportunities and threats related to different feed raw materials used for salmon feeds. The report looks into marine ingredients, plant ingredients, microbial ingredients, terrestrial animal by-products (protein and oils), blue mussels and insect meal, and also GMO ingredients.
A sustainable production of salmon
During the last decade, the production of Atlantic salmon has increased by almost 70 per cent from around 900 000 tonnes worldwide in 2000 to more than 1 500 000 tonnes today, and farmed salmon is the most widely consumed sea product in the industrialised world (Naylor and Burke 2005). The growth in the salmon industry has raised concerns about the environmental impacts of salmon farming. Increasing consumer awareness on sustainability and food safety puts pressure on the aquaculture industry to document that the production of salmon is safe and environmentally sustainable. What characterizes a sustainable production is however not clearly defined, but it is generally accepted that it includes a responsibility for present as well as future generations. According to the of United Nations Brundtland commission (1987), a sustainable development meets the needs of the current population without compromising the ability of future generations to meet their own needs.

The feed is the main input factor in salmon production, so an understanding of how different feed formulations affect environmental impacts and resource utilisation is vital. The dependence of the aquaculture feed industry on fish meal and fish oil and the effect this may have on wild fish stocks is the main argument against sustainability.

Of the worlds total fishery production in 2007, 81 per cent was used for human consumption and 14.5 per cent was reduced to fish meal and oil (FAO 2008). Of the total production of fish meal and oil in 2006, between 56–68 per cent of the fish meal and 83–89 per cent of the fish oil produced were consumed by the aquaculture industry (Jackson 2006, 2007, Tacon 2008, FAO 2008) and 55 per cent were consumed by the salmon industry (FAO 2008). The rest of the fish meal is used in terrestrial animal feed production (Pigs 9 per cent, poultry 31 per cent) (IFFO). Some fish oil is also used for human consumption and industrial purposes. The landings of industrial feed fish and production of fish meal have been fairly stable the last 20 years (FAO 2008). In 2007, 52 per cent of the world’s fish stocks were fully exploited, 19 per cent were overexploited, 8 per cent depleted and 1 per cent were recovering from depletion and the remaining 20 per cent were underexploited or moderately exploited (FAO 2008).

Thus, a further growth in the production of salmon and aquaculture can not depend on an increase in the catch volume of wild fish beyond sustainable limits, but must rather rely on increased use of alternative sources of lipid and protein.

However, sustainability not only applies to the use of marine ingredients, but must also include agriculture crops used in production of salmon feed. Deforestation, soil erosion, use of pesticides and fertilisers, gene modification of plants and microorganisms, depletion of water resources and pollution of rivers and lakes are some effects that may be considered.

It is therefore necessary to do an analysis of strengths, weaknesses, opportunities and threats (SWOT) for the major feed ingredients used in the Norwegian salmon industry today and for ingredients that potentially may be used in the future. This concerns availability, economic viability, fish performance, health and welfare, food safety, feed quality, feed technology, sustainability and other environmental impacts.

The main project is based on FHF's project “Resource budgets and analysis of feed ingredients in salmon diets: A preproject” (FHF-900522).
1. to create a resource budget for ingredients used in salmon diets in comparison with feed for other meat productions;
2. to review the methods available for calculating resource use and eco-efficiency;
3. to make a SWOT analysis (strengths, weaknesses, opportunities and threats) of the most important ingredients used in salmon diets today and ingredients potentially used in the future.
Expected project impact
Although some recent reports exists (Ellingsen and Aanondsen 2006, Naylor et al., 2009), the efficiency and sustainability in terms of use of energy and nutrients in different food production systems are seldom compared and some of the existing data almost 20 years old (Austreng 1994). Thus, there is a need to update the existing data and develop more accurate methods to calculate the efficiency in resource use in salmon farming in comparison with other important feed production systems as pigs and poultry.

The SWOT analysis should focus on marine feed ingredients because the use of these is most controversial at the moment. The SWOT analysis does not need to be comprehensive, but rather look at the main groups of ingredients and overall trends to identify threats before they become large problems and identify the limiting resources in terms of sustainability.
Project design and management
The project will be lead by Nofima Marin, and involve international scientists when needed. An industry group with participation from the large feed producers and some salmon producing companies will be involved in and follow the project closely.
The project is divided between the Resource budget and the SWOT analysis.
a) The resource budget
The main focus of this part of the project will be to calculate the efficiency of resource use in Norwegian salmon production and compare it with the efficiency of other food productions in Norway such as pig and poultry production. However, the salmon industry competes for resources in a global market, so it is necessary to view salmon production in a wider perspective and look at alternative ways to utilise a limited resource pool.

Only a small part of the earths land area is suitable for agriculture. What is grown on this area is controlled by market forces. Large areas and water resources are used to produce cotton to make clothes for the fashion industry in the developed countries. Other non-food crops that are grown in large scale are coffee, tea, tobacco and linen. In addition, an increasing amount of farm land is used to produce crops used for production of biofuels, and crops that could be used for human consumption such as soy beans, corn and barley are fed to animals to produce meat for a market with willingness to pay. Thus, the production of crops for human consumption may be increased significantly if there is willingness to change the allocation of purchasing power in the world. However, as long as the world’s economy is based on a capitalistic system the market forces decide what will be produced.

It is important to view salmon production in this perspective, if the production of salmon had been stopped tomorrow, what would be the alternative? The resources in salmon production would not be directed towards people in developing countries with low buying power, but would rather be used for production of other high priced food items for the consumers in the industrial parts of the world. When discussing whether a food production is ethical or sustainable it must be evaluated in relation to how the world actually is, and not in relation to an ideal world.

The major challenge in the next 30–40 years is to produce enough food to sustain a population of 9 billion people on earth. The available resource pool must be utilised as effective as possible to yield as much energy and protein as possible, and with less space available on land, growing food in the ocean is an attractive possibility. This project will resolve how effective and sustainable salmon production is compared to other food productions and what methods that may be used for comparison.

The concept sustainable production of food has to be discussed and preferably defined. It is also of interest to look at the efficiency of fisheries for human consumption in terms of total catch, how much that reaches the market and the edible yield of the total catch.

There are several available methods to express the efficiency in utilisation of resources and energy for production of salmon and other food production systems (described below). There are also several ways to express the environmental impact of a production (Carbon footprints, ecological footprints, integrated eco-efficiency models).

This project will review the different methods available and evaluate their strengths and weaknesses and discuss what methods are useful for comparing different food productions. It is also important to assess what methods/models that evaluate the efficiency of resource utilisation and sustainability.

i) Conversion efficiency
The retention of nutrients and energy in the edible part of the animal and in the whole animal can be calculated based on the content in the feed, feed utilisation and energy and nutrient content. The intake of nutrients (g) may also be related to the gain of the animal (kg) and expressed as g protein eaten/kg gain (Protein efficiency ratio, PER), g fat eaten/kg gain (Lipid efficiency ratio, LER), and energy/kg gain (Energy efficiency ratio, EER).

Another approach is to calculate nutrient-to-nutrient ratios for protein, fat, energy and polyunsaturated fatty acids (EPA and DHA). Irrespective of the method used, it is very important that the data used in the calculations are clearly defined and that the calculations are reproducible. Experimental data must be clearly separated from data obtained under commercial farming conditions.

It is difficult to study the effect of single ingredients in the diet on resource utilisation in salmon production. Alternatively, one can measure the effect of different feed formulations, for example the standard salmon feed from 20 and 10 years ago, the standard today and what is believed to be the situation in 10 years from now. Data on feed formulations and FCR (Feed Conversion Ratio) from commercial farming must be made available for the project to make such an analysis possible. The trends in other food productions such as pig and poultry will also be evaluated in a similar way.

The fish in/fish out ratio (FIFO) transforms the amount of fish meal and oil that is used to produce one kg of farmed fish back to wild fish weight equivalents. It is often used, both in scientific publications (Tacon and Metian, 2008, Naylor et al., 2009) and in the public debate. The published values for this ratio in salmon production vary considerably, from 5 (Naylor et al. 2009) to less than 1 (Torstensen et al. 2008) depending on the inclusion levels of fish meal and oil, the FCR and whether the fat content in the fish meal is considered in the calculation.

However, this weight-to-weight approach does not take into account the nutrient and energy content of the feed fish and the salmon product.

ii) Eco-efficiency methods
An alternative approach to express the efficiency in resource utilisation is by using a model that analyses the eco-efficiency of different productions. Life Cycle Assessment (LCA) is an ISO-standardized (ISO 14040-14043) analytical framework for evaluating the environmental impacts of products or processes (eco-efficiency). The LCA framework is used to quantify the energy and material inputs and environmental impacts associated with each stage of a product’s life cycle, from resource extraction and processing, consumption, disposal and recycling. LCA is increasingly being used to evaluate food production systems, including aquaculture (Papatryphon et al., 2004, Ellingsen and Aanondsen 2006, Pelletier and Tyedmers, 2007, Ellingsen et al., 2009, Kruse et al., 2009), both to compare environmental performance of competing products and to identify areas in a production process that has the largest negative environmental impacts (environmental “hot spots”).

One example of an eco-efficiency model is BASF’s model developed in collaboration with Roland Berger Consulting. This model is used by the Eco-Institute in Freiburg, Germany and is accepted by the Wuppertal Institute. The ecological calculations in the model are done according to the ISO-rules 14040 ff. This model calculates the production costs as well the ecological consequences. Such a model can be used as a fundament for strategic decisions and evaluation of different productions in terms of sustainability. A model can also be used for case-studies and to study trends over time and to compare different productions. However, it is important to critically evaluate the data that goes into the model and how each parameter is weighted because this is of vital importance for the outcome of the analysis.

Energy analysis is another method for measuring eco-efficiency. It calculates the total energy requirement of a certain production and provides a measure of the energy cost of the production. Because a lot of the energy consumed in industrial production comes from fossil energy sources resulting in environmental problems (climate change, acid precipitation etc.) energy analysis is also used as a measure of sustainability. The calculation of carbon footprints is one example.

Ecological footprint analysis is another measure of eco-efficiency that has been used to evaluate aquaculture production and fisheries (Folke et al., 1998). The resource and energy requirement to support an activity or production is expressed in the form of an ecosystem support area and is thus a measure of relative ecological efficiency.

From a resource efficiency perspective it is important how much of the animal that is utilised, either as human food or as ingredients in animal food or for other purposes. Parts of the product not suited for human consumption may be used as a feed resource in other production systems (such as blood and bone meal, feather meal) or have other valuable applications (cosmetics, health products, pet food, salmon oil for omega 3 capsules, etc). Resources such as fish meal and oil are spared by using by-products from salmon industry for such productions, but currently there are no good ways of expressing the ecological value of “recycled” resources in animal production systems.

Attempts have been made to adjust the FIFO ratio according to the share of the spent fish meal and oil that comes from by-products, but this project will address these issues further and suggest possible ways to express both use of by-products in salmon production and also the value of by-products from salmon production and other animal food production systems.

b) The SWOT analysis
The project will evaluate the strengths, weaknesses, opportunities and threats associated with the use of different ingredients in salmon feed. Both feed ingredients commonly used today and ingredients that may potentially be used in the future will be evaluated in terms of fish health and welfare, food safety, feed quality, feed technology, discharge and other environmental impacts. Opportunities in the form of competitive advantages, (availability, price, volume, consumer preferences and reputation) also need to be evaluated.

The SWOT analysis will not be comprehensive, but rather look at the main groups of ingredients and overall trends. To identify threats before they become large problems and identify the limiting resources in terms of sustainability will be important goals for the SWOT analysis. The SWOT analysis shall evaluate the basis for the legislation and retailer demands on use of several potential ingredients such as LAPs and GMO and challenge the scientific basis for these regulations and demands.

The project shall contribute to establish a knowledge based platform for making new ingredients available for the salmon industry and identify gaps of knowledge and point out future areas for research and development. A special emphasis should be put on marine feed resources, because there is a need for unbiased environmental standards/documentation on fish stocks that are used in production of fish meal and oil today.

It is also important to identify the opportunities that exist in utilisation of new marine feed ingredients such as krill and fish silage.

The main focus of the SWOT analysis is:
1) to dentify major constraints in available resources for salmon production;
2) to valuate the situation for marine ingredients
a. assess the status of the fish stocks used in production of fish meal and oil;
b. opportunities and consequences of harvesting ingredients further down in the marine food web;
c. by-products from fisheries and aquaculture;
3) to dentify the opportunities in utilising new ingredients in salmon production from land-based productions (vegetabilia, animal by-products, GMO, single cell protein and oil);
4) to challenge the scientific basis for the legislation that restricts the use of certain feed ingredients in Europe (LAPs, GMOs) and identify gaps of knowledge.
Dissemination of project results
Nofima will arrange a workshop early in the project period where the participants in the project meet to discuss the content and methods to be used in the project. The project will result in one or several publications with contributions from internationally acknowledged researchers and at least 3 reports.

Suggested themes for the publications/reports from the project:
1. Review of methods for estimating resource utilization and eco-efficiency
2. Efficiency in resource utilization in salmon production in comparison with other meat productions
3. SWOT analysis of feed ingredients in salmon production
a. Environmental assessment of marine resources in salmon feed production
b. Environmental assessment of terrestrial resources in salmon feed production (vegetabilia, GMOs, LAPs, single cell organisms)

The results from the project will also be presented in the form of publications in popular journals such as Norsk fiskeoppdrett and as fact sheets and presentations at conferences and meetings (national and international).