1. Recent findings suggest that the colonization of habitat patches may be affected by the quality of surrounding patches. For instance, patches that lack predators may be avoided when located near others with predators, a pattern known as risk contagion. Alternatively, predator avoidance might also redirect dispersal towards nearby predator-free patches resulting in so-called habitat compression. However, it is largely unknown how predators continue to influence these habitat selection behaviors at increasing distances from outside of their own habitat patch. In addition, current information is derived from artificial mesocosm experiments, while support from natural ecosystems is lacking. 2. This study used bromeliad landscapes as a natural model system to study how oviposition habitat selection of Diptera responds to the cues of a distant predator, the carnivorous elephant mosquito larva. 3. We established landscapes containing predator-free bromeliad habitat patches placed at increasing distances from a predator-containing patch, along with replicate control landscapes. These patches were then left to be colonized by ovipositing bromeliad insects. 4. We found that distance to predators modulates habitat selection decisions. Moreover, different dipteran families had different responses suggesting different habitat selection strategies. In some families, predator-free patches at certain distances from the predator patch were avoided, confirming risk contagion. In other families, these patches received higher number of colonists providing evidence of predator induced habitat compression. 5. We confirm that effects of predators in a natural ecosystem can extend beyond the patch in which the predator is present and that the presence or absence of remote predator effects on habitat selection depends on the distance to predators. The notion that perceived habitat quality can depend on conditions in neighboring patches forces habitat selection studies to adopt a landscape perspective and account for the effects of both present and remote predators when explaining community assembly in metacommunities.

This dataset was collected during a field experiment. It is the raw dataset and has not been processed in any way. It contains abundances of dipteran larva found in bromeliads that were subjected to different treatments as well as morphological measurements of each bromeliad 

Each row represents an individual bromeliad. Bromeliads were arranged into clusters of four, each placed in one of 4 positions within the cluster, indicated by the POSITION column. Bromeliads in the same cluster therefore have the same CLUSTER id. In the TREATMENT column, for bromeliads in the center position, 'present' indicates that a predator was present in that bromeliad. For bromeliads in positions d1, d2, and d3, 'present' indicates that they were found in a cluster that had a predator in the center position (but they themselves were predator-free). Therefore clusters P1 to P5 all have predators in the center position. C1 to C3 were control clusters, 'absent' indicating that the bromeliad was found in a cluster that did not have a predator in the bromeliad in the center position. Also note, in the main text, the term 'landscape' is used instead of cluster to indicate a group of four bromeliads.