While this is true, forests do hold a lot of water. One of the problems with dry places is not so much that there is no water, but that the water is inconsiste

While this is true, forests do hold a lot of water. One of the problems with dry places is not so much that there is no water, but that the water is inconsistent and only available for a short period of time. Forests act like a sponge and hold water for long periods of time, which can literally transform ecosystems. That said, I know literally nothing about the Gobi desert ecosystem and whether planting a forest will have the desired effect. Interestingly (to me at least), right at the time that the Edo period in Japan was starting, Japan had cut down most of its trees. One of the effects was that the runoff from mountains was much faster and carried less mineral content. This, in turn, reduced the ability for algae to grow in the shallow regions of the sea around Japan, leading to a very real possibility of mass starvation. Somehow somebody realised this and convinced the Shogun to put a moratorium on logging. Despite the really huge hardship this had on the people, it's likely to have saved Japan. By the 18th century Japan had developed fairly advanced silviculture. Some historians in Japan consider this to be a practical miracle given the very limited knowledge that people had about ecology in the Edo period. I'm hoping that our cultures will enter a new period where we start to understand that this kinds of "ecological engineering" is necessary for the health of our societies. Whether or not the planting of trees in the Gobi desert is effective or not (I have no idea), the idea that we need to do these kinds of things is very promising and I hope that whatever successes and failures there are will be fed back into a system of refinement for the next try.