Ecological Constraints on Sexual Selection in a Human-Modified Landscape

Abstract

Anthropogenic activities are changing the sensory landscape, interfering with transmission and reception of sexual signals. These changes are leading to alterations in mating behaviour with consequences to fitness. In systems where mate-finding involves long-distance signalling by one sex and approach by the other sex, the spatial distribution of signallers can have implications for male and female fitness. Spatial distribution of signallers is typically determined by an interplay of multiple factors, both ecological and evolutionary, including male competition, female choice and resources, such as calling and oviposition sites. We investigated the possible influence of resource distribution (signalling sites) on the strength and direction of sexual selection acting on false-leaf katydid Onomarchus uninotatus males, signalling in a human-modified landscape in the Western Ghats, India, a biodiversity hotspot. The landscape has changed from evergreen forests to plantations owing to human settlements. We first determined the spatial distribution of calling males and of available calling sites, which are trees of the genus Artocarpus, in the landscape. Using the information on male spacing, call transmission and hearing thresholds, the perceptual spaces of male signals were computed to understand the acoustic environment of calling males and females. It was found that both calling males and females could hear calls of males from neighbouring trees with a probability of 0.76 and 0.59, respectively. Although calling males were found to be spaced apart more than predicted by chance, significant overlap was seen in their acoustic ranges. Clustering of males enables females to easily sample multiple males, facilitating mate choice, but is detrimental to males as it increases competition for females. Using simulations, we determined the optimal spatial distributions of O. uninotatus males for female choice, and for reduction of male competition, given the signalling site distribution. The observed distribution of signallers was then compared with the hypothetical optimal distributions to examine the drivers of signaller spacing. Spacing of calling males in the field was found to be not optimal for either males or females. Resource distribution was found to limit the effectiveness of sexual selection drivers in pushing male spacing toward fitness optima of males or females.