Collect ideas on master thesis about mapme.protectedareas HOT 2 CLOSED

I have been thinking about the options again a bit more.

One interesting aspect to extend the work could be to look deeper into ways of how to mass process data and improve performance in the scripts. If we think of the big-data-processing script which chains all processes together, we could Analyse the most heavy/lengthy parts and try different strategies:

One possible strategy could be e.g. parallelization in the cloud. Currently we work sequentially on Protected Areas but we could do a lot of these loops also in parallel on multiple cores. Another strategy could be to use RGRASS for specific tasks instead of Terra and SF. So there could be a comparison between both frameworks for certain topics as well.

All of this could be put also in context of cost/benefit analysis (more cloud resources are expensive but we can put a very concrete price tag on that using different VM specifications). Some general conclusions from this work could be on what are currently the most efficient and the most cost-effective tools and setups for mass processing structured Geodata in R.

I know that people do similar work for unstructured (satellite) images so the uniqueness of this work could in looking at these aspects for structured (preprocessed) data products for an applied use-case (conservation).

from mapme.protectedareas.

Thank you @Jo-Schie for initiating this issue and putting up your ideas here.

I am definitely looking forward to continuing what we are working on and incorporate the idea with the thesis topic as I am learning many new things with this internship and specifically, with your supervision. This is something I would like to discuss with you in more detail and with this issue here, it would be very helpful to reach the concrete thesis topic.
Meanwhile, I would start to explore more on the first idea you mentioned here which seems interesting to work on.

from mapme.protectedareas.