This is based on some thoughts that I had recently. It isn't meant to be a comprehensive guide to reforestation or to anything really.
As forest defenders and reforesters, it's important for us to choose our battles. We cannot be everywhere at once doing everything that needs to be done, so until we can destroy the systems of oppression responsible for deforestation, we would do well to select high-priority areas to protect or to reforest. I thought of four main criteria for making such a selection.
Conservation Value
Conservation value is basically an assessment of how much would be lost if the forest were destroyed, or how much could still be saved/regained if degraded forest were protected and restored.
How important is the area to the integrity/contiguity of the forest?
Does it contain key bodies of water that need to be protected from contamination? How many animals live there or depend on using the area as a safe passage as they travel from place to place?
Is it adjacent to any areas that are already protected (such that protecting/reforesting this area would effectively expand the area under protection)?
Does the area have a notable impact on the local/regional climate (e.g. does it transpire enough moisture to affect rainfall downwind)?
Are there any rare or endangered plant species in the area, or would reforestation of the area restore the native habitat of these species?
Taken together, factors like these give an idea of the conservation value of the forest. Pristine rainforest, or even an area with pristine forest fragments and deforested patches between them, could be considered as having greater conservation value than an island of dry forest that was logged and cleared of all understory growth for use as goat grazing land.
Return on Investment
We have limited time and energy, so the more results that we can get for the effort that we put in, the more effective our actions will be.
Protecting or reforesting a fairly remote area bordered by intact forest on three sides, in a rainforest climate, where the land is flat or only gently sloped, will be much easier with a higher chance of success than trying to do the same in an area where urban sprawl is rapidly approaching on three sides, with high traffic through the area, in a savanna climate that requires carrying buckets of water up a steep slope in the dry season.
All else equal, it makes sense to focus on the easy areas first, as this will result in a greater land area under forest cover than if we were to start with the difficult areas which require more time and effort (if they can be saved at all).
It's Dangerous To Go Alone
You can plant more trees if you're alive with all of your arms and legs intact.
Forest defenders are vulnerable to being attacked by cattle ranchers, gold miners, loggers, and other people who want to destroy the forest, especially if we sabotage these destructive activities. With the environmental crisis quickly approaching multiple tipping points, the years ahead will likely be more turbulent than any time before as capitalists fight for the diminishing natural resources. Desperation will increase the tendency toward violence.
One way to reduce the risk associated with forest protection and reforestation is to find strength in numbers. Get some "ecoterrorists" together, get organised, and coordinate your efforts in particular areas. Having a decentralised local support network not only keeps more watchful eyes on the forest and makes reforestation projects easier ("many hands make light work"), it also makes it more difficult for the destroyers of the forest to get away with violence. As they become outnumbered in an area, it will be less and less practical for them to take the forest by force.
Natural Regeneration
I think of this from two different angles.
Areas with a high capacity for natural regeneration need the least help. If simply left alone, these areas can reforest themselves. If there is no clear and present danger to the local forest, it might make more sense to focus our efforts elsewhere. There are other areas (e.g. the tropical dry evergreen forest of India or parts of the Tumbes-Piura dry forest) that have been extremely deforested with a lower capacity for regeneration due to small remaining forest size (fewer seed trees and seed-dispersing animals), seasonal precipitation, and severe pressure from grazing operations, and these areas really do need long-term assistance in order to keep them from being destroyed entirely. The need is more urgent in areas with a lower capacity for natural regeneration.
Areas with a high potential for natural regeneration will be easier to reforest, meaning more area reforested per unit effort. Protect the area from grazing cows/goats, broadcast an abundance of seeds, and then nature will do the work to reforest large amounts of land while we focus on more intensive reforestation in the parts that need extra attention (such as pasture lands surrounded by other deforested lands, with no adjacent forest to supply seeds). Even if we get injured or otherwise incapacitated, most of the forest will grow back on its own, and it only needs someone to be living there to protect it. The chance of success is higher in areas with a higher capacity for natural regeneration.
No warranty either expressed or implied, not even warranty of fitness for a particular purpose, don't do anything illegal, etc. It would be cool if this gave anyone any new ideas about how and where to reforest or if it started a discussion in the comments, but I don't have any specific intention in posting this. I might expand on these ideas later on or provide more specific examples if anyone is interested.