LCA Food 2012 in Saint Malo, France! - Manifestations et colloques ...

PARALLEL SESSION 4B: DIET 8 th Int. Conference on LCA in the
Agri-Food Sector, 1-4 Oct 2012
Comparing the environmental impact of human diets varying in
amount of animal-source food – the impact of accounting for
nutritional quality
Heleen R.J. van Kernebeek * , Simon J. Oosting, Imke J.M. de Boer
Animal Production Systems Group, Wageningen University, The Netherlands
Corresponding author. E-mail: Heleen.vankernebeek@wur.nl
ABSTRACT
Studies comparing the environmental impact of human diets that vary in amount of animal-source food generally not account for the
nutritional quality of these diets. Because diets may differ in amount of energy and nutrients, the impacts of these diets are not comparable.
To make impacts comparable, we quantified the amount of 12 nutrients in diets described in 13 peer reviewed studies. We
also computed their composite nutritional quality, expressed as the Nutrient Rich Diet9.3 score (NRD9.3). We expressed the GWP
per unit protein and per unit NRD9.3. Diets that had higher levels of animal protein had higher (excessive) levels of total protein and
were generally lower in composite nutritional quality. Diets that had lower levels of animal protein had lower GWPs per gram protein
and per unit NRD9.3. Accounting for composite nutritional quality gave stronger contrasts in GWP between human diets than
original comparisons.
Keywords: GWP, human diets, protein, nutritional quality, NRD9.3
1. Introduction
The production and consumption of animal-source food products (ASFP) has a high environmental impact
and leads to high resource use (Cordell et al., 2009; Steinfeld, 2006). Choosing a diet low in ASFP
could mitigate environmental impacts and improve resource use efficiency. Several studies assessed the environmental
impact of human diets that varied in their ASFP concentration (Carlsson-Kanyama and Gonzalez,
2009; Davis et al., 2010; Saxe et al., 2012). They generally compared impacts of two or more diet
scenarios using a life cycle perspective. Life cycle assessment (LCA) is an holistic approach to assess the
environmental impact during the entire production chain. These studies support the general conclusion that
plant-based diets have a lower environmental impact than animal-based diets.
To compare environmental impacts of different diet scenarios, the environmental impact should be expressed
on the basis of a so-called functional unit (FU). A FU depends on the main function of the system,
and the primary function of food production is to satisfy the human body’s need for nutrients such as kcal,
protein, fibre, vitamins and minerals. Studies generally use daily intake or energy, protein or fat content of
the diet as functional units. These elements, however, only partly define the nutritional quality of a diet.
Smedman et al., (2010), therefore, accounted for the content of 21 nutrients, when comparing emissions of
greenhouse gases along the life cycle of various beverages. The FU that Smedman et al., (2010) used, was
the nutrient density (NDS). The NDS of a food product is a nutrient over energy ratio representing the composite
nutrient score of a product (Hansen, 1973). In the present paper we developed another composite nutrient
score at the diet level, and used this score as a FU in the comparison of the environmental impact of
human diets.
The main aim of this paper was to evaluate whether the comparison of environmental impacts among diets
that varied in their ASFP-concentration, as reported in literature, would be different when nutritional
quality was accounted for. We reviewed 13 published studies that assessed the environmental impact of human
diets that varied in their ASFP-concentration.
2. Methods
2.1. Selection of published studies
We found 13 studies that met the following five selection criteria: the study contained more than one national
diet scenario; diets within studies varied in ASFP-concentration; the weight of each food product included
in the diets was given; diets were not designed for specific groups (e.g. infants, people with health
problems or a specific gender); the article is published in a peer-reviewed scientific journal.
2.2. Calculation of individual nutrient scores of each diet
395