Associational effects are widespread indirect interactions between neighboring plants, mediated by herbivores. Previous studies have shown that the identity of a neighboring plant species can strongly increase (associational susceptibility) or decrease (associational resistance) damage to a focal species. However, predicting the direction and magnitude of associational effects has been challenging for several reasons. First, herbivory often varies across environmental gradients, with herbivory tending to be strongest in warmer climates. Thus, the strength of associational effects might also vary across gradients. Second, the degree of difference in traits between neighboring plants may also influence the outcome of their interactions with herbivores. We conducted a meta-analysis using 250 estimates of insect herbivore abundance on agricultural crops neighbored by a conspecific or heterospecific from 40 studies undertaken on five continents. Associational effects were estimated as the Cohen’s d effect size of herbivore abundance in monocultures relative to polycultures. We extracted mean annual temperature (MAT) and calculated the phylogenetic distance between neighboring plant species for each study. MAT and phylogenetic distance interacted to predict the magnitude of associational effects. For focal crops with closely related neighbors (congeners), associational effects became slightly more negative (i.e., benefited from associational resistance) at warmer sites, but this effect was much stronger for focal crops planted with distant neighbors. This work highlights that variability in associational effects are driven by interactions between climate and phylogenetic relatedness of the neighboring plants.
