I'd suggest you sart by going to the Ecopath website ( https://ecopath.org ) and click 'Ecobase' at the top of the page. The link takes you to a database that has all the ecopath models available--if you scroll down to the bottom you will note that there is a fairly large selection for Africa. You may be able to download some of the models from there directly.
Currently I am working on a project regarding Fukushima (Japan) and the Gulf of Lion (Mediterranean). These use EwE and the Ecotracer routine for contaminant tracing; and I am based in France.
Ewe has Ecotracer built into the software. Once you have a balanced Ecopath model you can use the Ecotracer routine. It's located on the side bar menu at the bottom under tools; see Walters and Christensen 2018 (Ecotracer: analyzing concentration of contaminants and radioisotopes in an aquatic spatial-dynamic food web model) for a full description.
best to look at some papers, and it depends on what the model is being used for/or what the question is. You will note that the previous Walters and Christensen paper uses Ecosim to look at changes going forward in time to biomass levels.
I completely agree with everything Shawn Booth has said above. I'll add that if you are looking to use Ecotracer with Ecosim, then when moving from Ecopath to Ecosim you only need to add any data that will help to answer your question. If you are less interested in changes to biomass but more interested in how a contaminant moves through an ecosystem then there is less pressure on having more robust biomass data through time.
We (Tierney et al. 2018 Modelling marine trophic transfer of radiocarbon (14C) from a nuclear facility) used a mass balanced Ecopath model and ran that forward through time in Ecotracer. We did not address the changes in biomass over this time as this wasn't a priority in that study.
Our study was also spatially resolved (in Ecospace) and used the spatial-temporal framework (Steenbeek et al. 2013 Bridging the gap between ecosystem modeling tools and geographic information systems: Driving a food web model with external spatial–temporal data). This was essential for our study but again may not be necessary in others.
You should get in touch with the working group of Prof. Matthias Wolff at ZMT. He and his group are working for many years with / on Ecopath models, also creating R-Routines for it.