Acoustic phenology of tropical resident birds differs between native forest species and parkland colonizer species
Data files
Jun 12, 2024 version files 108.16 MB
-
Black-naped_Oriole.zip
792.81 KB
-
BNO.csv
10.59 MB
-
DCuck.csv
10.59 MB
-
Drongo-cuckoo.zip
15.89 KB
-
Koel.csv
10.59 MB
-
Koel.zip
492.78 KB
-
LinBarb.csv
10.61 MB
-
LineatedBarbet.zip
1.56 MB
-
LittleSpiderhunter_SG.zip
854.45 KB
-
LSpider.csv
10.59 MB
-
Pin-striped_Tit-babbler.zip
446.23 KB
-
PSTB.csv
10.60 MB
-
README.md
3.50 KB
-
RTTail.csv
10.59 MB
-
RufousTailedTailorbird.zip
3.75 MB
-
SeasonalityAnalysisExampleCode.R
8.69 KB
-
SHBul.csv
10.59 MB
-
ShortTailedBabbler_SG.zip
2.71 MB
-
STBab.csv
10.59 MB
-
StrawHeadedBulbul.zip
2.18 MB
Abstract
Most birds are characterized by a seasonal phenology closely adapted to local climatic conditions, even in tropical habitats where climatic seasonality is slight. In order to better understand the phenologies of resident tropical birds, and how phenology may differ among species at the same site, we used ~70,000 hours of audio recordings collected continuously for two years at four recording stations in Singapore and nine custom-made machine learning classifiers to determine the vocal phenology of a panel of nine resident bird species. We detected distinct seasonality in vocal activity in some species but not others. Native forest species sang seasonally. In contrast, species which have had breeding populations in Singapore only for the last few decades exhibited seemingly aseasonal or unpredictable song activity throughout the year. Urbanization and habitat modification over the last 100 years have altered the composition of species in Singapore, which appears to have influenced phenological dynamics in the avian community. It is unclear what is driving the differences in phenology between these two groups of species, but it may be due to either differences in seasonal availability of preferred foods, or newly established populations may require decades to adjust to local environmental conditions. Our results highlight the ways that anthropogenic habitat modification may disrupt phenological cycles in tropical regions in addition to altering the species community.
Raw species detection data, R code for data processing, and Kaleidoscope species classifiers are available.
Species Detection Data
Each .csv file contains the results of one species classifier summarised by input file.
Each row represents one unique audio file with a ~30min (29:55) duration and tells the
number of times the target species can be heard within that file.
Site Names:
CCNR - Central Catchment Nature Reserve, Singapore (1.355488, 103.804549)
DAFA - Dairy Farm Nature Park, Singapore (1.358419, 103.777492)
SBWR - Sungei Buloh Wetland Reserve, Singapore (1.441586, 103.735308)
NUS - National University of Singapore Campus, Singapore (1.295020, 103.779385)
Species Names:
LSpider - Little Spiderhunter - Arachnothera longirostra
DCuck - Drongo-cuckoo - Surniculus lugubris
RTTail - Rufous-tailed Tailorbird - Orthotomus sericeus
PSTB - Pin-striped Tit-babbler - Mixornis gularis
STBab - Short-tailed Babbler - Pellorneum malaccense
BNO - Black-naped Oriole - Oriolus chinensis
LinBarb - Lineated Barbet - Psilopogon lineatus
SHBul - Straw-headed Bulbul - Pycnonotus zeylanicus
Koel - Asian Koel - Eudynamys scolopaceus
Variables:
fileName - input file recording. Naming convention is SITE_yyyymmdd_hhmmss.WAV
FOLDER - folder where the input file is located
DURATION - duration of the input file in seconds (max 1795 seconds, aka 29 minutes and 55 seconds)
DATE - date of the recording
TIME - start time of the recording, using a 24 hour clock. There should be one file every half hour.
HOUR - Hour of the day when audio was recorded. Used to make hourly activity plot.
SERIAL.NO - not used. Serial number of the audio recorder.
autoID_target - number of automatic detections of the species’ vocalization within the audio file
manualID_target - number of manually verified (correct) detections of the species’ vocalization
manualID_otherSounds - not used. number of manually identified sounds that were NOT the species’ vocalization
.csv files:
BNO.csv - Black-naped Oriole detections
DCuck.csv - Drongo-cuckoo detections
Koel.csv - Asian Koel detections
LinBarb.csv - Lineated Barbet detections
LSpider.csv - Little Spiderhunter detections
PSTB.csv - Pin-striped Tit-babbler detections
RTTail.csv - Rufous-tailed Tailorbird detections
SHBul.csv - Straw-headed Bulbul detections
STBab.csv - Short-tailed Babbler detections
R Code
SeasonalityAnalysisExampleCode.R generates a calendar plot, radar plot, and
vector-based seasonality index for Little Spiderhunters. Changing the loaded .csv file
can generate the plots for the other species.
Species Classifiers
Each classifier is zipped into a folder. Classifier folders are named according to the species they detect. To use a classifier, load the ‘settings.ini’ file in Kaleidoscope to set up the proper signal parameters, then use the ‘cluster.kcs’ file to scan and cluster new audio.
Files within each species classifier:
cluster.kcs - the trained clustering algorithm
settings.ini - Kaleidoscope settings used for this classifier
Species Classifiers:
Black-naped Oriole.zip
Drongo-cuckoo.zip
Koel.zip
LineatedBarbet.zip
LittleSpiderhunter_SG.zip
Pin-striped Tit-babbler.zip
RufousTailedTailorbird.zip
StrawHeadedBulbul.zip
ShortTailedBabbler_SG.zip
For questions, comments, or additional details, feel free to contact Laura Berman
lauraberman@u.nus.edu
lberman6@wisc.edu
lauraberman11@gmail.com
This is an acoustic phenology dataset. Soundscape recordings were collected 24/7 in Singapore over the course of 2 years. The machine learning software Kaleidoscope Pro was used to make species classifiers for 9 species of birds. Species classifiers are able to automatically detect all occurrences of the target species' song within the 2-year-long dataset. Automatic outputs were manually verified to ensure accuracy.
In addition to the cleaned species detection dataset, we also provide the Kaleidoscope species classifiers. These classifiers can be used to detect the 9 focal species in your own audio data.
Each .csv file contains the results of one species classifier summarised by input file. Each row represents one unique audio file with a ~30min (29:55) duration and tells the number of times the target species can be heard within that file.
SeasonalityAnalysisExampleCode.R generates a calendar plot, radar plot, and vector-based seasonality index for Little Spiderhunters. Changing the loaded .csv file can generate these plots for the other species.