Data from: Deciduous forests hold conservation value for birds within South Andaman Island, India

Authors: Arpitha Jayanth, Zankhna Patel, Mohammed Mubeen, Karthikayan M, Rohit Naniwadekar
Affiliations - 
Arpitha Jayanth, Zankhna Patel, Mohammed Mubeen and Rohit Naniwadekar - Nature Conservation Foundation, Mysore, India
Karthikayan M - Pondicherry University, Pondicherry, India

We assessed the importance of deciduous and evergreen habitats for forest birds (including several endemic and threatened species) within South Andaman Island.
To this end, we compared the composition and diversity (taxonomic, functional and phylogenetic) of forest birds across the two habitat types, and evaluted species-specific responses to habitat variables within the island.
The dataset collected for this study is published here. Bird species traits were sourced from Tobias et al. (2022) and Wilman et al. (2014). 
Phylogenetic trees were sourced from birdtree.org (Jetz et al. 2012).

Geographic Coverage: South Andaman Island, Andaman and Nicobar Islands, India
Taxonomic Coverage: Birds, 54 species
Temporal Coverage: March-May 2022 and December 2022-April 2023


Brief summary of field methods: 
We conducted line-transect surveys for forest birds across evergreen and deciduous forests in South Andaman Island, India. 
Twenty-seven transects were sampled either in the morning (between 0515 – 0930 hrs) or in the afternoon (between 1515 – 1730 hrs) from March-May 2022 and December 2022-April 2023.
For every bird detected, we noted species identity, group size (if visible), time of detection, and whether the bird(s) was seen or heard.
Detections of flying birds and nocturnal species (owls and nightjars) were excluded.
For each transect, we measured - 
a. tree density and basal area (using the point-centered quarter method, or PCQ); 
b. canopy cover, proportion of deciduous trees, and presence of cane, bamboo runners and clumps of standing bamboo (averaged across PCQ points);
c. number of large trees and cut logs (counted within a 20m-wide belt along each transect); and 
d. distance from nearest settlement/village (using Google Earth Pro).

Description of data files:

1. species_list.csv - This dataset contains all 54 bird species codes and scientific names used in the analysis
sp_id: four-letter bird species code, used consistently throughout the analysis
sp_latest: latest scientific name of bird species
sp_tree: scientific name used for extracting phylogenetic tree from birdtree.org
common_name: common name of bird species

2. transects_veg.csv - This dataset contains transect-level habitat variables.
transect: transect ID
site: site within the island
density_ha: tree density per hectare
BA_per_ha: basal area per hectare (in m2)
large_trees: number of trees with DBH >=150cm in a 20m belt along the transect
cut_logs: number of cut logs in a 20m belt along the transect
prop_cane: proportion of PCQ points with cane in a 10m radius
prop_RB: proportion of PCQ points with bamboo runners in a 10m radius
prop_SB: proportion of PCQ points with erect bamboo stands in a 10m radius
avg_canopy: canopy cover averged across all PCQ points for a transect
cut_poles: number of cut logs (10cm <= DBH > 30cm) in a 20m belt along the transect
cut_adults: number of cut logs (DBH >= 30cm) in a 20m belt along the transect
prop_D: proportion of trees measured for PCQ that are deciduous
cluster: habitat cluster assigned from the clustering analysis
habitat: same as cluster, but as evergreen and deciduous
dist_village_m: distance of transect to nearest human settlement/village (in m)

3. Andaman_birds.csv - This dataset contains information on bird detections during line transect surveys.
site: site within the island
transect: transect ID
date: date of sampling (dd-mmm-yy)
year: project year of sampling - Year1 (Y1) or Year 2 (Y2)
replicate: sampling temporal replicate within that year
start_time: start time of transect (hh:mm, 24h format)
end_time: end time of transect (hh:mm, 24h format)
observers: observer code
weather: weather condition for >50% of sampling time
length: length of transect (in m)
sp_time: time of detecting bird species
sp_id: bird species code (refer to species_list.csv)
cluster_size: group size i.e., number of individuals detected (if visible)
seen: 1 if detection was visual
heard: 1 if detection was only auditory
session: if transect was walked in the morning (AM) or afternoon (PM)
effort_rep: number of times the transect was walked over the entire study period

4. traits_all.csv - This dataset contains trait information for all 54 bird species, sourced from Tobias et al. (2022) and Wilman et al. (2014).
sp_id: bird species code (refer to species_list.csv)
sp_latest: latest scientific name of bird species
sp_tree: scientific name used for extracting phylogenetic tree from birdtree.org
beak_length_culmen: from Tobias et al. 2022
beak_length_nares: from Tobias et al. 2022
beak_width: from Tobias et al. 2022
beak_depth: from Tobias et al. 2022
tarsus_length: from Tobias et al. 2022
wing_length: from Tobias et al. 2022
kipps_distance: from Tobias et al. 2022
secondary: from Tobias et al. 2022
hand_wing_index: from Tobias et al. 2022
tail_length: from Tobias et al. 2022
mass: from Tobias et al. 2022
sp_morphoTraits: species name in Tobias et al. 2022. If blank, 'sp_latest' was used.
diet_inv: from Wilman et al. 2014
diet_Vend: from Wilman et al. 2014
diet_Vect: from Wilman et al. 2014
diet_Vfish: from Wilman et al. 2014
diet_Vunk: from Wilman et al. 2014
diet_scav: from Wilman et al. 2014
diet_fruit: from Wilman et al. 2014
diet_nect: from Wilman et al. 2014
diet_seed: from Wilman et al. 2014
diet_plantO: from Wilman et al. 2014
diet_cat: from Wilman et al. 2014
sp_EltonTraits: species name in Wilman et al. 2014. If blank, 'sp_latest' was used.

5. all_birds.nex - Phylogenetic tree used in the phylogenetic diversity analysis, sourced from birdtree.org (Jetz et al. 2012).

The following files were used in the HMSC (Hierarchical Modelling of Species Communities) analysis:

6. bird_P.nex - Phylogenetic tree used in the HMSC analysis, sourced from birdtree.org (Jetz et al. 2012).

7. birdtraits_hmsc.csv - This dataset contains traits for 33 species used in the HMSC analysis
sp_id: bird species code (refer to species_list.csv)
sp_tree: scientific name used for extracting phylogenetic tree from birdtree.org
beak_width: from Tobias et al. 2022
tarsus_length: from Tobias et al. 2022
kipps_distance: from Tobias et al. 2022
tail_length: from Tobias et al. 2022
diet_fruit: from Wilman et al. 2014
diet_nect: from Wilman et al. 2014
diet_seed: from Wilman et al. 2014
diet_animal: from Wilman et al. 2014

8. data_hmsc.csv - This dataset contains transect-level variables and species-wise detections for each sampling replicate (N = 269).
transect: transect ID
rep: temporal replicate of sampling (from rep1 to rep15)
[columns 4-36]: Each of these columns contain total number of detections (both visual and auditory) of bird species across each sampling replicate of a transect. Columns names are same as 'sp_tree' names for each bird species (refer to species_list.csv)
density_ha: tree density per hectare for each transect
BA_per_ha: basal area per hectare (in m2) for each transect
prop_D: proportion of trees measured for PCQ that are deciduous
cut_adults: number of cut logs (DBH >= 30cm) in a 20m belt along the transect
dist_village_m: distance of transect to nearest human settlement/village (in m)
long: longitude of transect centroid, in UTM format
lat: latitude of transect centroid, in UTM format


REFERENCES:
1. Jetz, W., G. H. Thomas, J. B. Joy, K. Hartmann, and A. O. Mooers. 2012. The global diversity of birds in space and time. Nature 491:444–448.
2. Tobias, J. A., C. Sheard, A. L. Pigot, A. J. M. Devenish, J. Yang, et al. 2022. AVONET: morphological, ecological and geographical data for all birds. Ecology Letters 25:581–597.
3. Wilman, H., J. Belmaker, J. Simpson, C. de la Rosa, M. M. Rivadeneira, and W. Jetz. 2014. EltonTraits 1.0: Species-level foraging attributes of the world’s birds and mammals. Ecology 95:2027–2027.

