The multi-scale effects of neighborhood composition on patterns of associational resistance and susceptibility

Background/Question/Methods

Associational resistance, AR (or susceptibility, AS) occurs when neighboring plants increase (or decrease) the amount of damage to a focal plant. A number of studies have tested for the existence of AR/AS, examined mechanisms that may lead to AR/AS patterns, and discussed AR/AS applications in agro-forestry.  However, many studies have traditionally used substitutive and additive experimental designs which confound the effects of total plant density, heterospecific plant density, and conspecific plant density on plant damage.  Furthermore, past studies have ignored the effects of neighborhood size on patterns of AS/AR which could obscure the mechanisms by which neighboring plant influence damage patterns.  Using a response surface experimental design, we examined how total plant density and the density of heterospecific neighbors (hereafter referred to as “neighborhood composition”) influenced insect herbivore damage.  We examined how plant damage patterns varied with neighborhood composition at two spatial scales (neighbors immediately surrounding a focal plant and neighbors within a 1m² quadat).  The focal plant species was Solanum carolinense (Carolina horsenettle) and the heterospecific neighbor species was Solidago altissima (Tall goldenrod).

Results/Conclusions

The effect of neighborhood composition on damage varied according to spatial scale.  As the proportion of heterospecific neighbors immediately surrounding a focal Solanum plant increased, there was a linear increase in the amount of leaf damage (percent leaf area removed by insects), suggesting AS with increasing Solidago neighbors.  In addition, there was a negative linear relationship between neighbor distance and leaf damage (irrespective of neighbor identity), suggesting AS when neighbors are in close proximity.  At larger spatial scales (neighbors within a 1m² quadrat),  we discovered a unimodal relationship between leaf damage on Solanum and the proportion of Solidago in the neighborhood, suggesting AS when Solidago neighbors are at low to intermediate densities, but AR when Solidago neighbors are at intermediate to high densities.  We also found an increase in leaf damage on Solanum in neighborhoods with increasing total plant density (suggesting AS), indicating that Solanum in close proximity with neighbors experienced more damage, irrespective of neighbor identity.  Since the relationships between the proportion of Solidago neighbors and damage on Solanum varied according to spatial scale, this suggests that multiple mechanisms are operating to produce AR/AS patterns.