Bird associations with floristics and physiognomy differ across five biogeographic subregions of the Great Basin, USA

ABSTRACT The majority of management plans for birds on public lands across the western United States do not recognize the geographic variation in a given species' habitat. We examined associations of plant species and functional groups with occupancy of 19 bird species across 5 biogeographic subregions of the Great Basin (central, western, Sierra Nevada, northern, and eastern), USA. We hypothesized that occupancy was associated with floristics (individual plant species) within subregions, and with physiognomy (characterized by functional groups) across the Great Basin. We used two methods to evaluate bird-vegetation associations within and across subregions. First, we examined which covariates of floristics and physiognomy were significantly associated with occupancy in each subregion. Second, for each bird species, we compared covariate estimates between each of the 10 pairs of subregions. We classified the effects of covariates on occupancy in 2 subregions as significantly different if <5% of their posterior distributions overlapped. The plant species and functional groups that were associated significantly with occupancy varied considerably among subregions. Twenty-four percent of bird-plant associations that were significant at the Great Basin level were not significant in any subregion. Associations between occupancy and floristics differed the most between the Sierra Nevada and central or western subregions, and the least between the eastern and western subregions. Associations between occupancy and physiognomy differed the most between the Sierra Nevada and western and central subregions, and the least between the northern and western subregions. These differences and similarities may reflect variations in climate or bird communities or differences in sampling effort. In addition, the number and strength of associations between occupancy and floristic or physiognomic covariates varied substantially among bird species and subregions. We recommend that the management of birds across the Great Basin or other large ecoregions evaluate and account for geographic variation in environmental attributes associated with occupancy, and not assume bird-plant relations are consistent across the Great Basin. LAY SUMMARY The majority of management plans for birds on extensive public lands across the western United States do not recognize geographic variation in a given species' habitat. Local identities of plant species (floristics) and vegetation structure (physiognomy) can affect availability of nesting sites, food, shelter, and protection from predators. We examined associations of floristics and physiognomy with occupancy of 19 species of birds across 5 biologically distinct subregions of the Great Basin (central, western, Sierra Nevada, northern, and eastern), USA. We hypothesized that associations between birds and floristics varied among subregions, whereas associations between birds and physiognomy were consistent. Associations between bird occupancy and floristics and physiognomy were inconsistent across the Great Basin. Associations in the Sierra Nevada subregion were the most distinct. A single, regional management approach may not be effective for some Great Basin bird species. We recommend that management plans for birds evaluate local associations with vegetation, and not assume bird-plant relations are consistent across the Great Basin. RESUMEN La mayoría de los planes de manejo de aves en tierras públicas a lo largo del oeste de Estados Unidos no reconocen la variación geográfica en el hábitat de una especie determinada. Examinamos asociaciones de especies de plantas y grupos funcionales con la ocupación de 19 especies de aves en cinco subregiones biogeográficas de la Gran Cuenca (central, oeste, Sierra Nevada, norte y este), EEUU. Hipotetizamos que la ocupación estaba asociada con la florística (especies de plantas individuales) dentro de las subregiones y con la fisonomía (caracterizada por grupos funcionales) en toda la Gran Cuenca. Usamos dos métodos para evaluar las asociaciones de aves y vegetación dentro y entre subregiones. Primero, examinamos qué covariables florísticas y fisonomías estaban significativamente asociadas con la ocupación en cada subregión. Segundo, para cada especie de ave, comparamos estimaciones de covariables entre cada uno de los 10 pares de subregiones. Clasificamos los efectos de las covariables sobre la ocupación en dos subregiones como significativamente diferentes si <5% de sus distribuciones posteriores se superponen. Las especies de plantas y los grupos funcionales que se asociaron significativamente con la ocupación variaron considerablemente entre las subregiones. El 24% de las asociaciones de aves y plantas que fueron significativas a nivel de la Gran Cuenca no lo fueron en ninguna subregión. Las asociaciones entre ocupación y florística difirieron más entre Sierra Nevada y las subregiones central u oeste, y menos entre las subregiones este y oeste. Las asociaciones entre ocupación y fisonomía difirieron más entre Sierra Nevada y las subregiones oeste y central, y menos entre las subregiones norte y oeste. Estas diferencias y similitudes pueden reflejar variaciones en el clima o en las comunidades de aves, o diferencias en el esfuerzo de muestreo. Además, el número y la fuerza de las asociaciones entre la ocupación y las covariables florísticas o fisionómicas variaron sustancialmente entre las especies de aves y las subregiones. Recomendamos que el manejo de las aves en la Gran Cuenca u otras grandes ecorregiones evalúe y tome en cuenta la variación geográfica en los atributos ambientales asociados con la ocupación, y no asuma que las relaciones entre aves y plantas son consistentes en toda la Gran Cuenca.