I’m reaching out to get an input on how to approach an analysis that aims to assess the environmental consequences of meeting national’s (Ecuador) food needs from a nutritional-health perspective (dietary requirements), relying solely on domestic food production. This includes both conventional and non-conventional practices, and framed within biophysical boundaries, including ecological restoration.
I’m considering whether consequential LCA might be the right approach to apply. However, I’m unsure for a couple of reasons: there is not a direct substitution, but rather a shift in what and how much would be eaten and produced. Second, the database (e.g., EcoInvent) may present challenges, as it would not be representative. Thirdly, land use is key here. So, the implications/trade-offs are different at the national level (under the state’s control mainly) and on a global scale (if any). Anyhow, I’m aware of the limitations I may encounter.
Moreover, Ecuador primarily produces food for domestic consumption but relies on imports (directly and indirectly) for fertilizers, pesticides, seeds, feeds, machinery, and other technologies. Adjusting national policies to align with nutritional requirements and ecological restoration and conservation and changing agricultural practices could have economic implications for national imports (as well as exports, e.g. bananas, shrimp, cocoa, tuna, coffee). An important detail here is that the state regulates the national market.
I’m also considering coupling attributional LCA with system dynamics, though I’m concerned that this might present similar challenges as with C-LCA.
Consequential LCA is relevant whenever you need to support a decision that involves choosing between two or more options, for example to continue to produce and consume a product at the current level, or to increase or decrease the volume produced and consumed, or changing the technology or composition. Whether the decision is small (just one purchase) or large (a policy change for the entire population), it is the consequences of the decision we need to look at.
If your database has insufficient representativeness to provide support for the decision, you need to consider whether you can live with the implied uncertainty, or whether you need to collect more data. Changing to attributional modelling will not make your answer more representative.
When land use is important, you can unfortunately not rely on the consequential version of the ecoinvent database. But there are other more cause-effect based (consequential) models for indirect land use change that you can use, which can also accommodate a distinction between national and global effects.
To model the changes in the global markets resulting from the national changes, you need a good trade model that focus on changes rather than average historical trade patterns. The most simple trick is to use trend forecasting, although I have recently become aware of even more advanced models made by the International Trade Centre.
If you want your model to be dynamic and forward-looking, you can add a dynamic function to your consequential time series. I have a hard time understanding what result you expect from adding a dynamic function to an attributional LCA that is not even consistent within one time period?