In the abstract they argue that this is because the woodland improvement program located woods in socioeconomically deprived areas. But in general they shouldn't be making any causal claims with observational data like this. It's a frequent problem that popular press takes a study and reframes an association as causation for their story.