Research
Resident
Researchers
Visiting
Researchers
Research
Use Policy
Teaching
Teaching Use Policy
Teaching
at Hastings
Affiliated Institutions
Museum
Vertebrate Zoology
UC Natural Reserve System
Berkeley Nat. Hist.
Museums
Organz. of Biol. Field Stations
User
Inquiries
Calendar
- Who is at Hastings?
Pre-Arrival Information-Required
To Use
the Reserve
Internship Opportunities
Housing Descriptions
Photo Gallery
Data/Information
Hastings
Bibliography
Current
Weather
Archived
Weather Data
Vegetation
Data
Data
Catalog
History
of Hastings
Natural
History
Amphibians
- Reptiles
Birds
Geology
Invertebrates
- Insects, Spiders
Mammals
Native Grasslands
Oak Woodlands
Plants of Hastings
Webcams in Wildlands
Newsletter,
K-12
Current
Newsletter
Resources for K-12 Teachers
Contact
Us
Office, Resident Staff
Topographic Map of Hastings
Travel / Driving
Instructions
Sketch Map- Building Names
Hastings Hosts Explorations in Automatic Animal Identification and Localization
Dr. Charles Taylor, of UCLA with funding from the National Science Foundation (Awards Number 0410438, and EF-0410438) is exploring ways to use sensors in the wild to identify individual animals from the calls they make. Not only can animals be identified automatically, Dr. Taylor and his colleagues are working on ways to find the animals making calls in the 3-dimensional world. Dr. Taylor and NSF enable field work at Hastings and significantly upgraded our internet connectivity. We are very grateful for this support as it has improved the research and teaching environment at Hastings for hundreds of others.
One
of Dr. Taylor’s students, Yuan Yao, worked for the last 3 years
at Hastings to collect digital recordings of the acorn woodpecker (Yao,
Y., Ling, Y., Trifa, V., and Taylor, C. E. 2006. Vocal individuality
in acorn woodpeckers
,
Melanerpes formicivorus, recognized by hidden Markov
models.
Animal Behaviour, in review). Yuan used sophisticated software to identify
individuals (known because
they
were banded with unique color bands on their legs by Dr. Walter Koenig
and his colleagues at Hastings). A recent publication includes the work
done by Yuan and Taylor who also extended their work to antbirds in Mexico
(Trifa,
V. M., A. N. G. Kirschel, and C. E. Taylor. 2008 Automated species recognition
of antbirds in a Mexican rainforest using hidden Markov models. Journal
of Acoustical Society of America. in press.) Yuan was able to show
that the individual acorn woodpeckers have unique voices and these voices
remain constant over 2-3 years.
Extending this work on automatic identification, Dr. Taylor’s lab studied both antbirds in Mexico and marmots at Rocky Mountain Biological Lab, near Gothic, Colorado (Ali, A. M. S. Asgari, T. C. Collier, M. Allen, L. Girod, R. E. Hudson, K. Yao, C. E. Taylor and D. T. Blumstein. 2008. An empirical study of collaborative acoustic source localization. Wireless Sensor Networks. in press.). By deploying wireless sensors (microphones) in a network, they have shown that we can now both identify individual animals and a network of microphones can determine where they are located. The papers (links above) explain the research in more detail.
This work has many significant implications for those studying animal behavior. For instance, it might soon be possible to replace the tedious work our interns now do by watching birds from blinds to tell where various acorn woodpeckers are living and with who, by installing a network of wireless sensors that can stream data across the internet to databases far away. Our highly trained observers would then be free to do more intensive studies of behavior.