Previous researchers provided evidence that the absence of native birds at elevations below 900m in Hawaii was due to the presence of the Culex mosquito vector and associated disease transmission.  However, recent observations indicated that there may be new or persistent populations of birds in low elevation habitats on Hawaii.  We formulated a number of hypotheses to explain the presence of these birds, and began a field study to address these hypotheses as part of the larger Biocomplexity project.  To do this, we  established  nine 1km² study  plots in ohia dominated forest along an elevational gradient from 25 – 1800m on Mauna Loa.  Birds were captured in mist-nets monthly over a 3 year period at each site to gather information on community composition, demography, reproductive success, and to obtain blood samples for disease diagnostics and genetic analysis.  We were able to provide evidence for the hypothesis that these birds were resident, healthy, and breeding in low elevation habitats despite extremely high presence of malaria and mosquito vectors. One of the exciting implications of this study is that lowland amakihi may have evolved genetic resistance to introduced disease.

We continue to examine the current distribution and abundance of native and exotic forest birds within low elevation habitat in the Puna District of the Big Island Interestingly, our surveys are demonstrating a significant increase in amakihi abundance over the past decade.  It appears these birds may be re-colonizing vast tracts of their former habitat , probably as a result of increased evolution of resistance to avian malaria. The changing composition of the forest bird community in low-elevation Hawaii has important implications for the dynamics of avian malaria in low-elevation Hawaii, and for conservation of Hawaii's lowland forests.

I am currently collaborating on a number of projects with Greg Asner from the Carnegie Institute at Stanford and Natalie Boelman from Columbia. We are using a combination of direct field surveys, bioacoustics, and airborne remote sensing to assess the effects of plant invasion and ecosystem structural variation on avian abundance and community  structure.  We recently tested whether field-based bioacoustics and airborne remote sensing could be used to measure a range of vegetation and avian community properties across the landscape. We then assessed whether Ohia-dominated ecosystems host different avifauna communities than those ecosystems containing  invasive  trees such as Morella faya. We were particularly  interested in understanding how the structure and abundance of native and exotic plants were related to the relative abundance of native and exotic bird species. 

We are examining food as a potential basis for seasonal fluctuations in community structure, movement, and prevalence of disease across an 1800m elevational gradient. We addressed the long-held set of general assumptions that nectar production peaks predictably at different elevations on Hawaii, that nectarivorous birds track these peaks through seasonal migrations across the landscape, and that these movements, in turn, transport disease across the landscape. Monthly nectar availability was estimated for three years at each of our nine study sites by counting ohia flowers in randomly selected trees along transects, and by measuring sucrose concentration with a refractometer.  Abundance of nectar feeding birds was estimated through monthly mist net captures over three years, and through  Variable Circular Plot counts.  We are currently working to create statistical models to describe the relationship between food availability and other factors on bird movement. 

Forest Birds

Forest structure and composition across a large elevational gradient on Hawaii

Uncovering Multi-trophic Biotic Resistance to Biological Invasion in Hawai'i using Bioacoustics, Field Surveys, and Airborne Remote Sensing

Distribution and abundance of native forest birds in low elevation habitats on Hawaii

Is there a relationship between flowering patterns of ohia and the movement of nectar feeding birds across the landscape?

Current Projects

How do native birds persist at low elevations in Hawaii?

Ten years ago, I measured and tagged 1000’s individual trees within two 1km² study areas at Hakalau Forest NWR to set up a long term ecological study on Hawaiian rain forest dynamics.  I am collaborating with Dr. Leonard Freed at UH Manoa, Dr. Patrick Baker at Monash University in Australia and Marcos Gorresen at USGS-BRD to examine the role of stand structure and species composition on tree growth. We are evaluating the influence of neighborhood competitive effects on individual tree growth rate using several classes of models.  The ultimate goals are to predict how forest structure and composition will change over time at Hakalau, and to assess the way these changes will impact the forest bird community.

The role of stand structure and species composition on tree growth rates in a Hawaiian rain forest

Forest structure

Forest structure and composition is a major determinant of bird community structure and dynamics.  We recently completed a survey of the species composition, size, distribution, and abundance of all plants within each of the Biocomplexity study sites across the 1800m elevational gradient on Mauna Loa.  These data will be incorporated into the overall Biocomplexity modeling effort, but are also being analyzed to provide insight into how vegetation changes at various scales across large gradients in elevation in Hawaii.

 

Using radiocarbon to model the age and growth rates of trees in a Hawaiian rain forest

 

The growth and dynamics of tropical forests are poorly understood compared to their temperate counterparts, largely because tropical trees generally do not produce annual growth rings.  Radiocarbon dating is a new and very promising method for estimating age and growth rates of long-lived (>350 years) tropical trees, including ‘ohia (Metrosideros polymorpha).  We are currently taking wood samples and disks from the center of large, fallen ohia trees . We expect radiocarbon dating of these trees will a) provide us with accurate estimates of both age and growth rate for the largest size classes of ‘ohia, b) allow us to test the accuracy of our growth models for the smaller size classes of ‘ohia, and c) facilitate ongoing efforts to build predictive models of the structure and composition of Hawaiian wet forests far into the future.  In addition, radiocarbon dating will be used to determine if tree rings in ‘ohia and other long-lived Hawaiian tree species can be correlated with annual seasonal changes or even inter-annual climate fluctuations, such as El Nino events.  Thus, this technique will provide us with an historical, current, and future perspective of the tree dynamics within a Hawaiian rain forest.

The locations of the nine Biocomplexity study sites on the windward side of the Big Island

Bio-acoustic spectra of avian vocalizations from recordings acquired directly after the dawn chorus. Solid lines indicate sites dominated by native vegetation, dotted indicate exotic dominated sites. (thanks to Natalie Boelman)

A  low elevation wet forest in the Puna District of the Big Island

Graduate student Sam Brooks with a freshly cut disk from a recently fallen Ohia tree at Hakalau (this is not a chainsaw advertisement)

A UH Hilo Biology faculty member with his head in the trees, as usual

The ‘Apapane, doing what it does best