PS 68-118
How well do climate and ecosystem characteristics predict bird abundance at the landscape spatial scale?

Friday, August 15, 2014
Exhibit Hall, Sacramento Convention Center
Nicholas L. Rodenhouse, Biological Sciences, Wellesley College, Wellesley, MA
Zaixing Zhou, Earth Systems Research Center, University of New Hampshire, Durham, NH
Mary E. Martin, Earth Systems Research Center, University of New Hampshire, Durham, NH
Scott Ollinger, Earth Systems Research Center, University of New Hampshire, Durham, NH
John J. Battles, Environmental Science, Policy, and Management, University of California, Berkeley, Berkeley, CA
Background/Question/Methods

Climate and ecosystem characteristics predict bird species richness at continental spatial scales, but do such relationships occur at the landscape spatial scale? Birds like other mobile animals choose where to live based on multiple features of the habitat and social environment. The extent to which these correspond with ecosystem processes and patterns and the drivers of those has seldom been tested. Yet an understanding of these relationships may be key to predicting how climate-driven changes in ecosystem function might affect bird species. We used long-term data collected at the Hubbard Brook Experimental Forest (climate, elevation, aspect and vegetation data), modeled soil wetness (TWI) and ANPP to predict average abundance of foliage gleaning birds, ground foraging birds and all birds at 364 locations stratified to represent the entire 3160-ha Hubbard Brook valley. Point counts of all bird species were carried out three times in each of 12 years spanning 1999-2013. Because estimating bird abundance from point count data is problematic, we used average bird abundance at a location across the entire sampling period as an index to abundance. Higher index values would be found for locations where more birds were present or birds were present more frequently.

Results/Conclusions

We found a surprisingly low ability of the tested variables to predict average bird abundance within northern hardwoods forest, typically < 20%. When the transition zone to spruce-fir forest was included, up to 42% of the variance in the abundance of common foliage-gleaning birds was explained with the top ranking variables including: ANPP, foliar nitrogen, elevation and the percent of basal area that is coniferous. However, these patterns occurred only for south-facing slopes.  Less than 5% of variance could be explained by using these same variables to predict bird abundance on the north-facing slopes. North and south facing-slopes differed significantly in ANPP and foliar nitrogen with greater values of each occurring on south facing slopes. Neither elevation, total basal area of canopy trees, percent of BA that was coniferous trees, nor any measure of bird abundance differed between north- and south-facing slopes. These patterns are consistent with habitat variables and social context being the primary drivers of bird distribution, and they suggest that the food webs of which birds are a part can be sustained over a broad range of values for ecosystem variables.