This manuscript introduces an exciting approach to simulation and counterfactual work, in the shape of a genetic algorithm optimising land-use selection, applied here to the conservation of pollinators in a farming landscape.
The approach is extremely promising, and the paper is well written and clear, with great attention to possible improvements or expansion of questions.
The main shortcoming is, in my opinion, the fact that the results are entirely driven by the underlying population model, and that one is referred but not really explained. Considering the volume of the paper that is dedicated to visualising and interpreting the results, I would want to understand a bit more how the modelling assumptions impacted the optimization process. Ideally, some kind of sensitivity analysis (e.g. is the optimal land use different for different spatial or temporal decisions in the model?) could be run, but at least a slightly fuller description of the model would help. For instance, foraging distances are maxima, how does individual foraging affects the landscape composition? A full analysis is beyond the scope of this paper, but a "feel" for it would make the optimisation method more convincing.

I thank the authors for providing clear responses and edits on the manuscript.