Multiscale monitoring of a multispecies case study: two grass species at Sevilleta

Monitoring multiple species and their interactions at multiple scales is critical for any useful habitat conservation plans and for adaptive ecosystem management programs. Viability of single species can be understood only in the context of ecological interactions with other species at multiple spatial and temporal scales. Here, we present a case study of the spatial and temporal dynamics for two perennial grasses, Bouteloua eriopoda and B. gracilis, by using long-term, high resolution transect data from 1989 to 1998 at two sites in the Sevilleta National Wildlife Refuge (Deep Well and Five Points) in central New Mexico, USA. Information entropy was used to describe the spatial distribution of each species and the fractal dimension of information entropy was used to characterize the complexity of species dynamics across scales. When considered individually, the spatial distribution of each species was scale-invariant. However, species joint occurrences changed as scale of resolution increased at both sites. Interactions changed at scales less than 3.2 m at Deep Well and less than 1.6 m at Five Points. Across years, the interactions of these species at Deep Well were significantly different in 1989 from all other years. Our results argue that monitoring multiple species at multiple spatial and temporal scales is necessary to better understand multispecies interactions and community dynamics both of which have important conservation implications under changing environmental conditions.