Vocal Habits of the Åga

Andria Kroner, Renee Ha, Anne Clark PhD Student Binghamton University

Background and MotivationPair loss

Connectivity

Communication

Methods• Wildlife Acoustics recorders

• Set up in territories known to be active during the main breeding season

• Recording for one week, sunrise to sunset

• 1300 hrs ( > 400Gb) of recordings!

Analysis“Clustering” detections and Software Training :

1. 1 hr at sunrise for each site

2. Defined two broad clusters: “Mariana Crow” and “Not Mariana Crow”

3. Calculated error

4. Accepted detections ≤0.5 distance (error rate 0.047-0.245)

Results- False positivesError rate increased at further distances from center (0.56 error at 0.99 distance)

Most common false positives:

• *Micronesian Starling*

• Black Drongo

• Feral chicken

• Sea bird

• Mariana Kingfisher

0%

20%

40%

60%

80%

100%

120%

140%

160%

180%

200%

5 6 7 8 9 10 11 12 13 14 15 16 17 18

Cum

mul

ativ

e pe

rcen

t/Hou

r

Hour of day

Percent of detections at each site per hour

AGAT GAYA INAY MDUG OKGO RAND SAGU SAKA SONG SPIT SWIM UAGA UALA UPAL

Detections per day

• Highly variable within and between sites!

• Implications for search effort

05

10152025303540

6/29/16 6/30/16 7/1/16 7/2/16 7/3/16 7/4/16 7/5/16 7/6/16

Det

ectio

ns

Date

MDUG

0200400600800

10001200140016001800

6/17/16 6/18/16 6/19/16 6/20/16 6/21/16 6/22/16D

etec

tions

Date

UALA

Are family groups noisier?

No significant difference (p=0.24)

Jose Antonio Diaz

UALAnestling

Summary• Supports field observations of åga vocal behavior*

• Potential option for reducing human effort

• Can keep fine tuning training for auto ID

• Lots of flexibility in additional questions

Future directions• How does pair density affect vocal

behavior?

• How does captive rearing vs. wild rearing affect vocal development?

Sarah Faegre

Thank you!Wildlife Acoustics’ Scientific Product Grant

Suzanne Medina-Guam DAWR

Renee Ha, Sarah Faegre, Rumaan Malhotra- University of Washington

Questions?

Vocal Habits of the ÅgaBackground and MotivationMethodsAnalysisResults- False positivesSlide Number 6Detections per dayAre family groups noisier?SummaryFuture directionsThank you!