POPULATION GROWTH AS AN INDICATOR OF FOREST HABITAT QUALITY FOR
SONGBIRD SPECIES NESTING IN THE DRIFTLESS AREA ECOREGION.

Melinda G. Knutson1, Larkin A. Powell2, Gerald J. Niemi3, Mary Hammer-Friberg4, Randy K.
Hines1, Colin M. Sveum1.  
1U.S. Geological Survey, Upper Midwest Environmental Sciences Center, La Crosse, WI 54603 
2University of Dubuque, Dubuque, IA 52001  3Natural Resources Research Institute and
Department of Biology, University of Minnesota, Duluth, MN 55811  4University of Minnesota,
Graduate Program in Conservation Biology, St. Paul, MN 55108.  
 
Population growth (?) is a fundamental population parameter that can be used to estimate habitat
quality.  Many types of demographic data have been used to measure bird habitat quality,
including bird distribution, abundance estimates (numbers of breeding pairs/area), breeding
season productivity (the per-capita number of offspring surviving to the end of the breeding
season), food availability, and survival.  Direct measures of habitat quality such as reproductive
success are strongly preferred over indirect measures such as abundance.  Models are available to
estimate population growth from variables that can be measured in the field, including nesting
success.  Models can incorporate error estimates that are needed to simulate realistic variability
in population growth over time.  We used field data on songbird nesting success and a computer
model to assess population growth rates in upland forests and three types of floodplain forests of
the Driftless Area Ecoregion for 3 species of neotropical migrant birds.  We tested the null
hypothesis that the modeled population growth rate for each species was = 1 over a simulated
200-year time span (stable population).  Data were collected on bird nests in upland and
floodplain forest plots located in southeastern Minnesota, northeastern Iowa, and western
Wisconsin and in the Mississippi River floodplain from 1992 and 1996-1998.   In the floodplain,
we stratified forest units among three habitat classes: mature silver maple forest, mature,
mast-producing bottomland forest, and early successional forest stands.  We estimated breeding
season productivity by modifying Powell's dynamic, stochastic, individual-based model of
songbird reproduction.  We used our nesting data to parameterize the reproductive portions of the
model and we used estimates from the literature for survival parameters.  We found that
productivity was 1.80 (?(1.41 SD) fledglings per year for wood thrush (Hylocichla mustelina)
females, 0.86 (?(1.35) for prothonotary warblers (Protonotaria citrea), and 1.42 (?(0.69) for
American redstarts (Setophaga ruticilla).  Population growth rate predictions were below
replacement for prothonotary warblers  (? = 0.86, ?(0.04, 95% CI) and American redstarts (? =
0.90, ?(0.02, 95% CI).  Wood thrushes (? = 1.04, ?(0.04, 95% CI) did not reject the null
hypothesis of ? = 1.0.  Wood thrushes and American redstarts had high productivity in upland
habitats.  Of the 3 floodplain habitats, prothonotary warblers and American redstarts had lowest
productivity in the early successional stage habitat.  Growth rates of prothonotary warblers
ranged from a low of ? = 0.78 (?(0.04) in early successional floodplain habitats to ? = 0.90
(?(0.04) in maple-dominated floodplain habitat.  American redstart growth rates ranged from ?
=0.98 in upland habitat to ? = 0.91 in floodplain habitats combined.  Wood thrushes attempted
more nests and had more successful nests during the breeding season in our productivity model
than prothonotary warblers or American redstarts.  Wood thrushes also had more parasitized
nests than the other species, but the effects of parasitism were not as negative (no abandonments
or total clutch removals).  Sensitivity analyses revealed differing effects of parasitism rates on the
population growth rate of the three species.  In general, higher nest parasitism rates resulted in
trends of lower annual population growth.  Both the American redstart and the prothonotary
warbler fared poorest in the early successional floodplain habitats.  This is attributed to higher
cowbird parasitism rates and a lower proportion of successful nests in this habitat type. 
Parasitism and predation patterns for redstarts and prothonotary warblers were reversed among
maple and mast forests, indicating that these patterns can show species- rather than
habitat-specific patterns.  Upland habitats supported stable populations of wood thrushes and the
best productivity among habitat types for redstarts. 
     
Keywords: habitat quality, birds, nesting success, productivity, Driftless Area ecoregion