|
Activity Number:
|
476
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 7, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
| Abstract - #302120 |
|
Title:
|
Modeling Population Trends by Combining Information from Multiple Time Series
|
|
Author(s):
|
Philip Dixon*+ and David Otis
|
|
Companies:
|
Iowa State University and Iowa State University
|
|
Address:
|
Snedecor Hall, Ames, IA, 50011,
|
|
Keywords:
|
state space model ; detection probability ; combining indices
|
|
Abstract:
|
In the United States, two long-term, nationwide surveys provide three indices of Mourning Dove abundance. These are the Call Count Survey, which provides an index of birds heard and an index of birds seen, and the Breeding Bird Survey. Since 2003, these have been augmented by a direct estimate of population size from band return and harvest information. All four time series are imprecise; the three long-term population indices are potentially biased. We describe a state-space model to integrate all four sources of information into an estimated population trend. The contribution of a time series is down-weighted if that series is estimated to have a large bias or large variance. An extension to the model allows detection probabilities to vary over time. The performance and characteristics of the model are illustrated using data from the U.S. Western Management Unit.
|
