Subscriber access provided by UNIV OF DURHAM
Nutritional attributes, substitutability, scalability, and environmental intensity of an illustrative subset of current and future protein sources for aquaculture feeds: Joint consideration of potential synergies and trade-offs Nathan L. Pelletier, Dane H Klinger, Neil A Sims, Janice-Renee Yoshioka, and John N Kittinger Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b05468 • Publication Date (Web): 10 Apr 2018 Downloaded from http://pubs.acs.org on April 11, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 37
Environmental Science & Technology
1
Nutritional attributes, substitutability, scalability, and environmental
2
intensity of an illustrative subset of current and future protein sources for
3
aquaculture feeds: Joint consideration of potential synergies and trade-offs
4 5 6
Nathan Pelletier1*, Dane H. Klinger2, Neil A. Sims3, Janice-Renee Yoshioka4, John N.
7
Kittinger4,5
8 9 10
1
11
Columbia, Kelowna, BC, V1V1V7.
[email protected] 12
2
Center on Food Security and the Environment, Stanford University, Stanford, CA, USA 94305
13
3
Kampachi Farms LLC, Kailua-Kona, Hawaii, USA 96740
14
4
Conservation International, Center for Oceans, Honolulu, Hawaiʻi USA 96825
15
5
Arizona State University, Center for Biodiversity Outcomes, Julie Ann Wrigley Global Institute
16
of Sustainability, Tempe, AZ USA 85281
* 340 Fipke Centre for Innovative Research, 3247 University Way, University of British
17 18
Table of Contents/Abstract art
1 ACS Paragon Plus Environment
Environmental Science & Technology
Page 2 of 37
19 20 21
Abstract:
22
Aquaculture is anticipated to play an increasingly important role in global food security, as it may
23
represent one of the best opportunities to increase the availability of healthy animal protein in the context
24
of resource and environmental constraints. However, the growth and sustainability of the aquaculture
25
industry faces important bottlenecks with respect to feed resources, which may be derived from diverse
26
sources. Here, using a small but representative subset of potential aquafeed inputs (which we selected in
27
order to highlight a range of relevant attributes), we review a core suite of considerations that need to be
28
accommodated in concert in order to overcome key bottlenecks to continued development and expansion
29
of the aquaculture industry. Specifically, we evaluate each input’s nutritional attributes, substitutability,
30
scalability, and resource/environmental intensity. On this basis, we illustrate a range of potential synergies
31
and trade-offs within and across attributes that are characteristic of ingredient types. We posit that
32
recognition and management of such synergies and trade-offs is imperative to satisfying the multi-
33
objective decision-making associated with sustainable increases in future aquaculture production.
34
2 ACS Paragon Plus Environment
Page 3 of 37
35
Environmental Science & Technology
Introduction
36 37
Global food production may need to double in the next 30 years to meet the demands of a
38
growing and increasingly affluent population.1 However, the inputs required to sustain production in
39
traditional agricultural sectors are already limited, and competition for these resources from other sectors
40
is intensifying. For example, agriculture accounts for 70% of freshwater demand, and crop and
41
pastureland occupy 30% of land surface area.2 Moreover, expansion of agricultural activities is a major
42
contributor to a range of critical environmental problems, including climate change, biodiversity loss,
43
water pollution, and perturbation of the global nitrogen cycle.3 This is particularly true of terrestrial
44
livestock production, which plays a dominant role in food-related resource and environmental impacts.4
45
In the ocean, fisheries production has stagnated, and there is similarly limited potential for expansion.5-7
46
Aquaculture potentially presents one of the best opportunities for increasing global animal protein
47
production while minimizing resource demands and environmental impacts. Compared to conventional
48
terrestrial animal production, cold-blooded organisms reared in aquatic environments typically have
49
superior feed conversion efficiencies.8 Global aquaculture production, excluding aquatic plants, increased
50
by 1,567% between 1980 and 2014,9 and the sector is expected to grow by an additional 78% over the
51
next 15 years.10 However, the growth potential of aquaculture production is dependent on numerous
52
factors, including the method of farming, the species farmed and, for fed aquaculture, the specific kinds
53
and amounts of feed inputs required.11 Feed price and availability remains one of the critical bottlenecks
54
to expanding aquaculture globally. Demand for aquaculture feed is expected to rise from 30 million MT
55
in 2008 to around 71 million MT in 2020, as more farms rely on commercial feeds and total aquaculture
56
production increases.12
57
Feeds used in intensive aquaculture are typically formulated from multiple ingredients to meet the
58
nutritional requirements of the culture organism at least cost. While the nutritional profile and suitability
59
of feed inputs are clearly essential considerations for the aquaculture sector, so too is the scalability of
3 ACS Paragon Plus Environment
Environmental Science & Technology
Page 4 of 37
60
existing and emerging sources – which presents a second critical bottleneck for the expansion of
61
aquaculture. Specifically, growth of the aquaculture sector may be limited by: (1) the extent to which the
62
availability of feed materials can expand commensurate with increasing demand; (2) the costs of
63
production inputs, which is often a function of alternative uses for each ingredient; and (3) the capacity of
64
the sector to increase production and sales, while maintaining or increasing profitability and efficiency.
65
These issues constitute core challenges for the sector and for global food security.
66
As with terrestrial animal production, the resource use and emissions levels characteristic of fed
67
aquaculture will depend on species type, production technology and, importantly, the environmental
68
performance profile of specific inputs to aquaculture feeds. As of 2014, 69% of all farmed aquatic
69
animals were raised with feeds (i.e. not filter feeders or autotrophs).8 Production of marine and
70
diadromous finfish, which generally require higher levels of proteins and lipids in feeds, increased by
71
169% between 2005 and 2014, to 7.3 million tons. Production of freshwater finfish, which generally
72
require fewer protein and lipids, increased by 179% to 42.4 million tons.9 A variety of studies have
73
described the importance of aquafeed production in the overall life cycle environmental impacts of fed
74
aquaculture.13-20 For example, a study of global salmon aquaculture production found that feed inputs
75
account for, on average, 90% of the cradle-to-farm gate impacts of producing farmed salmon.13 At the
76
same time, the environmental performance profiles of potential aquafeed inputs – whether of fisheries,
77
livestock, or agricultural origin – vary widely.13, 17 In light of the increasing attention being paid to the
78
role of food systems, and of animal production in particular, with respect to resource demands and
79
environmental change, considerations of the embedded environmental costs of aquaculture feeds will
80
likely assume an important role in determining the trajectory of the aquaculture sector globally and may
81
hence be considered a third important bottleneck.
82
The nutrient profiles of feeds are often described in terms of four main components: energy,
83
protein, lipid, and carbohydrate.21 Protein often makes up the largest fraction of aquaculture feeds, and
84
increased usage of existing and novel protein sources is critically important in determining future
85
increases in global aquaculture production. However, expanded use of protein inputs to aquaculture feeds
4 ACS Paragon Plus Environment
Page 5 of 37
Environmental Science & Technology
86
may ultimately be constrained by these bottlenecks, specifically: (1) their substitutability (i.e., how their
87
particular nutritional profiles limit the extent to which they might substitute for competing protein inputs);
88
(2) their scalability (i.e. the extent to which production of protein inputs can feasibly be increased over
89
time to match rising demand; and (3) the resource and environmental implications of their production.
90
Moreover, these factors may potentially interact, creating a range of possible synergies or trade-offs.
91
Understanding such potential synergies and trade-offs is imperative to inform decision-making in support
92
of continued aquaculture development and efficient utilization of feed ingredients across multiple animal
93
production sectors.
94
Like terrestrial livestock production, aquaculture may source feed inputs from a wide range of
95
production systems, including marine fisheries, terrestrial crop and animal agriculture, and others.
96
Moreover, feed inputs may be “virgin materials”, co-products of food for human consumption, or
97
recycled biomass. Looking forward, the latter may include substantial but currently underutilized sources
98
– such as fish processing trimmings that are currently, in some cases, slurried and discharged to local
99
waterways.22.
Here, we review a representative subset of important current and emerging protein
100
sources for aquaculture feeds in order to illustrate such potential synergies and trade-offs. First, we
101
consider soy (a relatively low-input leguminous crop) produced in two major production regions (the US
102
and Brazil) that are characterized by differing production conditions and environmental implications. Soy
103
is processed into either soybean meal (a low-energy process producing a feed input whose use may be
104
constrained by anti-nutritional factors) or soy protein concentrate (an energy-intensive process resulting in
105
an output of superior nutritional value). Second, we consider three types of marine protein: (1) Peruvian
106
anchoveta meal, which is derived from a large-scale, energy efficient reduction fishery; (2) herring by-
107
product meal, derived from a smaller-scale, less energy efficient fishery for human consumption (in this
108
case, herring roe); and (3) Antarctic krill meal, which has unique nutritional properties but is derived from
109
a comparatively energy-intensive fishery and low-yielding rendering process. We also consider two
110
potential protein feed inputs derived from livestock processing materials. These are poultry by-product
111
meal, which has high nutritional value, and feather meal, which has comparatively lower nutritional
5 ACS Paragon Plus Environment
Environmental Science & Technology
Page 6 of 37
112
value. For both, raw material provision is characterized by non-trivial resource demands and emissions
113
due to the inefficiencies inherent to biological feed conversion. The former is also produced via an
114
energy-intensive rendering process relative to the latter. Finally, we also consider insect meal – an
115
emerging protein source that is oft-touted as a sustainable alternative to animal proteins. For each, we
116
evaluate aspects of their nutritional attributes, substitutability and scalability. We also employ Life Cycle
117
Assessment (LCA) models (that we created on a methodologically consistent basis from previously
118
reported, peer-reviewed studies in order to ensure comparability) to quantitatively assess and describe
119
some of the resource utilization and environmental impacts characteristic of the life cycle of each feed
120
input considered. On this basis, we discuss how these attributes of each protein source may both interact
121
and influence their immediate and future utility as a protein source for aquaculture feeds. We recognize
122
that protein provides a somewhat crude (albeit preferable to mass) basis for making comparisons among
123
feed inputs, since proteins can differ in nutritional value and be of varying nutritional utility depending on
124
what species and at what life stage they are fed. However, in the absence of more refined metrics such as
125
Protein Digestibility Corrected Amino Acid Scores or Dietary Indispensable Amino Acid Scores that are
126
generalizable across species, we nonetheless believe this provides a reasonable first-order basis for our
127
analysis.
128
The balance of the manuscript is hence organized into three sections. The first compares and
129
contrasts each feed input in terms of its nutritional profile, potential substitutability, and scalability. The
130
second describes the varied resource and environmental considerations that may be attributable to the
131
different ingredient types, highlighting generalizable insights and underscoring where context-specific
132
factors may be determining. The third section, which synthesizes and concludes, provides summary
133
recommendations for decision support for feeding future aquaculture production.
134 135
6 ACS Paragon Plus Environment
Page 7 of 37
Environmental Science & Technology
136 137 138 139
Nutritional Profiles, Substitutability and Scalability
140 141
Protein is required in fish and crustacean diets as a source of amino acids for protein synthesis
142
and for gluconeogenesis and metabolic energy.23,
24
143
crustaceans is often described as a single value, but the value represents the variety and ratio of up to 22
144
different amino acids, ten of which cannot be synthesized by fish and crustaceans and must be acquired
145
through dietary sources (for reviews of amino acids and nutrition, see
146
protein in an organism’s diet can change as a result of the size of the organism, the profile of amino acids
147
in the protein, the bioavailability of amino acids in the protein, and the level of dietary energy in the
148
feed.29, 30 There has been an extensive body of research focused on determining the optimal protein and
149
amino acid levels for fish growth under common culture conditions.31 Generally, an ideal feed will
150
contain similar amino acids as the proteins in the cultured animal.32
The specific requirement for protein in fish and
25-28
). The total requirement for
151
Protein sources vary widely in both the amount of crude protein and the type and ratio of amino
152
acids they provide.33, 34 Amino acid deficiencies can result in reduced growth and feed utilization.24 Diets
153
deficient in specific amino acids have been associated with reduced immune function and increased
154
overall susceptibility to disease.29, 30 Deficiencies in individual amino acids can also lead to erosion of the
155
dorsal fin (lysine deficiency), spinal deformities (tryptophan, leucine, lysine, arginine or histidine
156
deficiencies), or cataracts (methionine, tryptophan, or histidine deficiencies) (as reviewed in
157
addition, many plant-based protein sources contain anti-nutritional factors which can interfere with
158
utilization of nutrients in a feed and negatively impact the health and growth of an organism.36, 37. In some
159
cases, processing of protein ingredients can help increase digestibility, but at added cost.38
160 161
24, 35
). In
Protein ingredients are often among the most expensive components in aquaculture feeds, and farmers have a strong incentive to source the cheapest proteins that satisfy their animal’s nutritional
7 ACS Paragon Plus Environment
Environmental Science & Technology
162
needs.35, 39 Traditionally, fishmeal has provided an inexpensive, reliable, and nutritious source of protein
163
for aquaculture feeds, but growing demand from aquaculture, other sectors, and human consumption,
164
coupled with stagnating or decreasing supply from capture fisheries, has increased fishmeal prices to
165
historically high levels.40-42 Many sectors in the aquaculture industry now diversify feed formulations to
166
include large amounts of non-fishmeal protein ingredients.33, 43
Page 8 of 37
167
For a protein source to be viable for feed use in a range of aquaculture species, it must have
168
adequate protein content (48-80%), a suitable amino acid profile, high digestibility, and acceptable
169
palatability.44 Additionally, it must have low levels of fiber, starch, non-soluble carbohydrates, and anti-
170
nutritional factors that might impair performance.44 A protein source should ideally also be scalable, to
171
meet the growing demand from the aquaculture sector.
172
The concept of scalability has multiple definitions across sectors and disciplines.45 A quantitative
173
analysis of scalability to meet demand from aquaculture feeds would require extensive economic
174
modeling, and is outside the scope of this review. Instead, we review and compare key factors that will
175
likely influence scalability, including the availability of inputs to production, and the market price of
176
protein meals.46 Specifically, we evaluate major inputs to production that could increase production costs
177
through either supply or demand dynamics, such as supply constraints, or use in other sectors. The market
178
price of each protein source is also considered, including discussion of historical price volatility,
179
innovations in production technology, and demand for the protein from other sectors. Given the inherent
180
difficult of forecasting future global production frontiers, factors are qualitatively determined to be
181
favorable, uncertain, or unfavorable for scalability.
182 183
Fishmeal
184 185
Historically, fishmeal has been the primary protein source in aquaculture feeds. It has a high
186
protein content (e.g. 65–72% crude protein), suitable amino acid profile, and numerous trace elements
8 ACS Paragon Plus Environment
Page 9 of 37
Environmental Science & Technology
187
that are beneficial for the health of fish and crustaceans. The specific protein content of fishmeal varies
188
depending on the species of fish used, and the season and location in which they were caught.47 Fishmeal
189
is also highly digestible and palatable, and lacks anti-nutrients that can be toxic or impede growth.44, 48
190
Fishmeal is produced by cooking, pressing, drying and milling raw fish or fish parts to produce a
191
brown powder.49 Global fishmeal production increased from 1961 to 1985 and has since leveled out, with
192
substantial annual and multi-year fluctuations. The primary input to fishmeal is whole fish or fish parts.
193
On average, producing one kilogram of fishmeal requires 4.4 – 4.6 kilograms of whole fish.49 There is
194
limited capacity to increase fishmeal production, due to limited access to additional fish as raw
195
ingredients. Most fisheries for the small, pelagic fish used in fishmeal production (e.g. anchoveta and
196
sardines) are either at maximum capacity, or are currently overharvesting.50 Additionally, many small
197
pelagic fisheries are prone to dramatic fluctuations in abundance due to cyclical climatic variations every
198
2 to 7 years, complicating management.51 Expanding fishmeal production by using additional fish species
199
or populations would result in a net loss of available seafood for human consumption52 and would also
200
provide only limited additional production, as most fisheries are near, at, or beyond long-term production
201
capacity.5 There is some capacity to increase fishmeal production by utilizing fish that are currently
202
discarded as bycatch, although doing so may not be economically attractive and can lead to additional
203
incentives to overfish.53,
204
processing by-products will likely continue to increase in the near-term,55 although the total production
205
level is constrained by the total amount of fishery landings and aquaculture production.12
54
Additionally, the amount of fishmeal production derived from seafood
206
Fishmeal prices increased substantially in the mid-2000’s, and have remained higher than
207
terrestrial protein meals.42, 56 Aquaculture used 73% of global fishmeal production in 2010 (competing
208
uses included swine, 20%; poultry, 5%; and other sectors, 2%), and use in aquaculture is expected to
209
remain constant or increase.49, 57 Due to the limited supply of fish and the increasing demand, fishmeal
210
prices are expected to increase through 2025 and beyond, and high prices will continue to incentivize
211
many aquaculture enterprises to substitute other types of protein for fishmeal.8 The inputs to production
212
and the market price of fishmeal are therefore unfavorable for scalability of fishmeal use in aquaculture.
9 ACS Paragon Plus Environment
Environmental Science & Technology
Page 10 of 37
213
We summarize below key attributes of each alternative protein source considered in terms of nutritional
214
value, potential substitutability for fishmeal, and scalability.
215 216
Soybean Meal
217 218
Soybean meal, the co-product of oil extraction from soya beans, is the most frequently used non-
219
fishmeal protein source in aquaculture, due to its high digestibility, suitable amino acid profile, and low
220
price.47 Its performance with carnivorous aquatic species is diminished, relative to fishmeal, due to its
221
lower protein content (47% crude protein), and low concentrations of two essential amino acids,
222
methionine and lysine.34, 58, 59 Some species of fish also find soybean meal unpalatable without addition of
223
other protein meals or attractants.60 In addition, soybean meal contains anti-nutritional factors (e.g.
224
protease inhibitors, lectins, phytic acid, saponins, phytoestrogens, antivitamins, and allergens) that can
225
lead to reduced gut health and may ultimately be toxic.36, 48, 61 While extensive (>30%) use of soybean
226
meal is common in omnivorous species, use in carnivorous species is currently limited (