The metric used to measure the performance of a food service activity needs to:
Given these requirements, the absolute amount of CO2e emitted by a food service activity is not a good metric, as it is dependent on its scale. Rather, the emissions need to be normalized by a factor that enables their fair comparison regardless of differences in size and type.
In the field of Life Cycle Assessment of food and diets, the choice of this normalizing factor, called Functional Unit, is a discussed topic that is still unstandardized (Green et al., 2023, McAuliffe et al., 2020). In the FAO report “Integration of environment and nutrition in life cycle assessment of food items: opportunities and challenges” a pool of experts in the nutritional and LCA fields evaluated the currently used functional units and provided best practice recommendations (McLaren et al., 2021). Simple Functional Units based on individual factors (mass, protein content, caloric intake) are easy to use but only address specific issues (e.g. overconsumption, protein deficiency), and are therefore not suitable as a general unit to use across different food categories. Therefore, the combination of multiple nutrients is recommended. Existing nutrient indices combine multiple nutrients to encourage (e.g. vitamins, iron) and sometimes nutrients to limit (e.g. saturated fat, sodium), weighted by their Recommended Dietary Intake (RDI). By using these indices, the multifunctional aspect of nutrition is better accounted for when comparing different food items.
The well-established Nutrient Rich Food (NRF9.3) score considers 9 nutrients to encourage (protein, fiber, vitamins A, C, and D, calcium, iron, potassium, and magnesium) and 3 nutrients to limit (saturated fat, added sugar, sodium) (Fulgoni et al.,2009), and is a validated metric for nutritional purposes in the USA (McLaren et al., 2021). It is calculated by normalizing each nutrient by its RDI or its Maximum Daily Intake (MDI), summing the encouraged nutrients and subtracting the restricted nutrients:
$$ NRF9.3 = \sum_i \frac{Nutrient_i}{RDI_i} - \sum_j \frac{Nutrient_j}{MDI_j} $$
where:
i : the nutrients to encourage
j: the nutrients to limit
RDI: the Recommended Dietary Intake of the nutrients to encourage
MDI: the Maximum Dietary Intake of the nutrients to limit.
The NRF9.3 score can be reported per kg or per calorific content of each food.
This metric has a moderate complexity, but the inclusion of added sugars can be challenging since they cannot be easily distinguished from other sugars; total sugars are sometimes used instead. An additional unsolved problem is that the NRF9.3 score can be negative, but a negative functional unit cannot be used in LCA. To avoid this, additional rules need to be set, or the nutrients to limit need to be reported as a separate index. This further complicates an already hard to understand metric.
Overall, the FAO report concluded that no standard unique functional unit exists, and that open questions remain on, among many, the number of nutrients to consider, the treatment of nutrients that should be limited, and the use of nutrient indices. They recommend to consider as well some non-nutrient components (e.g. fibers), to adapt the metric used to the target population (e.g. population-specific RDI), and to report the results both per functional unit and per mass/volume to help interpretation and subsequent dietary studies (McLaren et al., 2021).
Eaternity developed its own functional unit to include multiple nutrients and at the same time keep the data requirements at a level that allows a streamlined calculation of carbon footprints. The Daily Food Unit (DFU) considers at the same time proteins, fats, energy, mass, and water content. Similarly to the NRF9.3, the nutrient contents are normalized by their Recommended Dietary Intake, according to the World Health Organization (WHO) recommendations. The DFU is then calculated as the averaged sum of the nutrients contributions to the RDI:
$$ DFU_x = \frac{\frac{protein_x}{50}+ \frac{fats_x}{66}+\frac{rest kJ_x}{6000}+\frac{water_x}{2500}+\frac{DW_x}{600}}{5} $$
Where: