Manjot Singh Sidhu, Sourav Singh, Rahul Sharma, Stanzin Khenrab, Harmandeep Kaur, Tsering Lanzes
Whether trees in agroforestry systems improve nearby soil has been assessed along two separate axes distance-from-tree spatial gradients and tree- or system-age chronosequences studied as distinct literatures that rarely engage one another. This review integrates both bodies of evidence, drawing on 95 primary studies identified through a documented landscape-scan process, to ask one question: does the mechanism generating the distance-from-tree soil gradient shift identity as the system ages and can that shift explain why gradient existence, direction and magnitude are reported so inconsistently across the literature? The evidence shows that gradient direction and existence are not universal, that age trajectories are not always monotonic, that different soil-property classes can diverge within the same site and that the two literatures are drawn from geographically decoupled contexts. This review proposes as a hypothesis, not an established finding a three-mechanism handoff in which root-zone competition, litterfall-zone organic input and canopy-microclimate modulation contribute in different proportions as a system ages; the canopy-microclimate component has no direct supporting evidence in this database. Evaluated individually against seven documented contradictions in the primary literature, the hypothesis explains one directly, partially explains three and leaves three unresolved, including an age-climate confound the evidence base cannot separate.