I am doing some environmental assessment studies using GIS and Remote sensing applied to mining areas. Which of the previous index would you recommend to do that.
I think there is no general answer on this. Both have pros and cons regarding different aspects, for example the density of vegetation, the impact of soil and its moisture, the type of vegetation and its leaves...
Look what other studies used in your study area (or comparable ecosystems) and think of what kind of change you want to assess.
You can also combine indices. I used CARTs to classify sampling points into four classes for impact assessment (urban, vegetation, open and water) based on NDVI, EVI, IBI, NMDWI, NDBI, and BUI.
EVI was better to separate 'vegetation' and 'open' while NDVI was more often used to separate 'vegetation' and 'water'/'urban'.
If you are more interested to vegetation degradation, I suggest NDVI in case of low biomass. In case of high biomass, EVI is recommended due to the NDVI saturation problem.
EVI actually complements NDVI especially if the image being used has some cloud covering. However in a mining area, the vegetation is often not dense so maybe an NDVI should suffice. And it also depends on what is being monitored, for example if it is to monitor the percentage vegataion cover of a mine tailings dump, then an NDVI should be okay provided there is no cloud cover on the image being used.