Most of the studies include impact assessments in their results. Only Yang et al. (2011) limit their publication to the inventory step, and Jorquera et al. (2010) only assess the energy bal­ance. All the other publications assess the potential reduction of greenhouse gas emissions in addition to the energy balance. However, only three studies estimate other environmental impacts, as defined by the LCA ISO norm: abiotic depletion, potential acidification, eutrophi­cation, ozone depletion, human toxicity, marine toxicity, photochemical oxidation, ionizing radiation, land use, freshwater toxicity, and terrestrial toxicity. In most of the studies, climate change is assessed with the characterization factors given by the IPCC (IPCC, 2006) for a tem­poral horizon of 100 years. Brentner et al. (2011) and Campbell et al. (2011) use different char­acterization factors, and Khoo et al. (2011) do not present the used methodology to assess climate change. Table 13.9 illustrates the divergence of characterization factors among the dif­ferent methods.

As explained in the previous sections, perimeters, modeling assumptions, and impact assessment methods can differ significantly among the publications. This results in a large

variability of the results related to the global-warming potential (GWP) and the energy return on investment (EROI) and hampers the capacity to compare results. However, we gathered results for these two indicators within the selected studies. Table 13.10 pre­sents the GWP of the various publications, and Figure 13.5 illustrates the relationship between EROI and GWP. Coproduct management has an important influence on the climate-change results. In some studies, climate change impact is negative, which means that the considered system fixes more greenhouse gases than it emits. In Batan et al. (2010), the negative score is due to the substitution of algal oilcakes to soybean oilcakes used to feed livestock. In Sander and Murthy (2010), it corresponds to the substitution of algal oilcakes to maize for the production of bioethanol. Finally, in Campbell et al. (2011), it corresponds to the electricity production from biogas produced by anaerobic digestion of the algal oilcakes.

When only the NER is considered to determine the EROI, favorable values are determined by most of the studies. However, when CER is taken into account, the EROI is limited (1.8 for the best case, 0.96 for the less favorable). It can also be observed that poor EROI (between 0 and 1) corresponds to high GWP.