Male glow-worms combine flying with walking and impede one another in their scramble to reach females
Data files
Sep 24, 2025 version files 23.12 GB
-
Arena_analysis.zip
636.30 KB
-
Field_landing_analysis.zip
832.97 KB
-
Field_sensitivity_analysis.zip
6.36 MB
-
RawData_arena.zip
13.02 GB
-
RawData_Field_videography.zip
10.09 GB
-
README.md
10.16 KB
Abstract
This dataset consists of 3 sets of infrared videos and the associated scripts for analysis. Infrared videography data recorded in a laboratory arena (videos of glow-worm males' movements towards a dummy female). Infrared videography data recorded in the field (videos of glow-worm males landing near a dummy female placed on a trap). Infrared videography data recorded in the field (videos of glow-worm males landing near a dummy female placed on a wire, i.e., without a trap).
Dataset DOI: 10.5061/dryad.83bk3jb4k
Description of the data and file structure
To determine whether features of the male glow-worm approach to glowing females conform to those of scramble competition, we adopted an observational approach using infrared videography of a dummy female (green LED) at night or in a behavioural arena in the lab.
1- Arena videography:
First, we explored the movements of individual males during their approach to a dummy female over short distances in a laboratory arena. The infrared camera was placed above an arena containing a dummy female (green LED). A single male was placed on a starting platform and its behaviour was quantified from the moment it left the platform to the moment it reached the green LED. We collected the type of behaviour (Walk, Flight or Leaps) as well as the distance covered by each behaviour and the time it took each male to reach the LED. This data was analysed using the R scripts located in the corresponding arena analysis folder.
2- Field videography:
Data were obtained using an infrared camera placed in the field above a green LED (dummy female). Males can be seen landing around the traps.
The data folder contains 2 subfolders, one for the videos recorded above a green LED mounted above a custom made trap, and one for the videos without a trap.
In the video with the LED mounted on a trap, we quantified several aspects of the final stage of males’ approach: (i) the mode of movement (walking vs flying); (ii) the time of their landing; (iii) the distance from the dummy female at which they landed; and (iv) the number of other males present when they landed. We investigated how these parameters influenced the time males took to reach the dummy female. Those variables are extracted and analysed using the R scripts located in the corresponding field videography analysis folder: "Field_landing_analysis".
In the video with the LED mounted on a wire (without a trap), we quantified the distance between males and the number and duration of male-male interactions on the ground. We also observed the behaviour of males directly on the LED, extracting the time each male spent on the LED alone or with other males, and the mode of exit from the LED. These data were analysed using the R scripts located in the corresponding field videography analysis folder: "Field_sensitivity_analysis"
Files and variables
File: RawData_arena.zip
Description: This folder contains the videos recordings for each individual males tested in the arena
File: RawData_Field_videography.zip
Description: This folder contains the videos recordings obtained in the field, in two subfolders separating data with and without a trap. Each day/video folder contains the original long video (.h264) and/or a .avi or mp4 conversion of the same video. Additionally, the "Without_trap" subfolder contains a .tif file of 16-20 frames extracted from each video every 2-minutes interval.
File: Arena_analysis.zip
Description: This folder contains the code used to analyse the data collected manually from the arena videos and produce the corresponding stats and graphs.
In the data folder the file "Arena_analysis_20210803_clean.csv" contains all data extracted from the videos and used in the analysis.
Columns names are as follows:
- Date : Date of the video
- gw_nbr : ID of the male in the video, number starting with "x" indicate males that did not leave the platform and where excluded from the analysis.
- Final position : Last place males were found in (choices are in order of appearance : 1-On the LED, 2-Still in Flight, 3-On the platform, 4-On the Base of the LED)
- Time to reach LED base (in seconds)
- Time to climb (to the LED), also in seconds
- 1st action : the first thing a male did to exit the platform (note that jump is the equivalent of "leap", in the manuscript, the other categories retained the same name i.e. flight, fall and climb down)
- Distance of 1st action (in centimetre) : how far the male went with the first action described in the previous column.
- Time of 1st action (seconds): how long the first action lasted
- 2/3/4th action : when applicable, the following actions made by males on their way to the LED
- Distance of 2/3/4th action (in centimetre)
- Time of 2/3/4th action (seconds)
- behaviour from base to LED : observations made on how males tackled the final climb to the LED after reaching it's base.
- comments : any additional observations
- Protocol : the specific setting of the arena in the video
File: Field_landing_analysis.zip
Description: This folder contains the code used to analyse the data collected manually from the field videos with a trap. Analysis is split in 2 scripts looking at the i) landing distances and approach strategy, and ii) the time to reach the LED depending on landing time, distance and the number of neighbours. Within the "Data" folder, structure is as follows:
Field_landing_analysis/Data/
├── Individual_counts
│ └── Count
│ └── XY
- "Count" contains a single "_count.csv" file for each video with the following columns:
- ROI point nbr = the reference of the Region of Interest marked on the video, allow for cross referencing with the ROI in the corresponding XY coordinates files.
- Description of Behaviour: brief description of males behaviour on their way to the LED. Behaviour are described using the following abbreviations :
- WF = walk and fly
- CB = climb down
- GONE = left the field of view
- WI = Walked into the field of view (rather than landed)
- W = Walk
- F = Flight
- FI = Flew in the trap
- TT = landed on Top of Trap
- Male ID = ID of the individual glow-worm observed at the ROI location, males can appear multiple times if they performed multiple actions before entering the trap or exiting the field of view.
- Landing location: within each of the marked circle (C1 = <10cm, C2=10-20cm, C3=>20cm) or on top of the trap (TT) or directly in the trap (IN)
- Initial landing location: Same as previous but limited to the first male landing (and not any subsequent landing).
- "XY" contains two .csv files for each video :
- "_XY_coord.csv" files contain the X and Y coordinates in pixels for each male action described in the previously described "count.csv" files. Additional columns are :
- Slice = the frame number at which the ROI is located
- Behaviour to trap = Overall strategy to the trap (no = did not reach the trap, WO = walking only, WF = combining Walk and flight, F = Flight only)
- "_LED.csv" contains the location of the LED in pixels
- "_XY_coord.csv" files contain the X and Y coordinates in pixels for each male action described in the previously described "count.csv" files. Additional columns are :
Additionally, a summary file called "2024_GW_landing_videos_data_summary.csv" summarises the counts of males observed in and around the trap, their landing location and their behaviour to the LED. Column names follow the previously described abbreviations.
File: Field_sensitivity_analysis.zip
Description: This folder contains the code used to analyse the data collected manually from the field videos without trap. Analysis is split in 3 scripts looking at the i) males on the ground, ii) males reaching the LED, ii) males on the LED.
Within the "Data" folder, we have a subfolder and 3 .csv files :
Field_sensitivity_analysis/Data/
├── Sampled_coordinates
├── 2025_field_videography_direct_contacts.csv
├── 2025_field_videography_males_on_LED.csv
├── 2025_field_videography_sampled_male_counts.csv
- "Sampled_coordinates" contains a .csv file for each video describing males location in the 16 sample frames extracted for analysis. Columns are as follows:
- Type = Descriptor of male location, 1 = On the LED, 2 = On the wire, 3 = On the cloth
- Slice = Frame number
- X, Y = coordinates in pixels
- X(cm), Y(cm) = coordinates in centimetre
- "2025_field_videography_direct_contacts.csv" summarises the direct interactions between males in each video. Columns are as follows:
- Video_date, video_ID = date and ID of the analysed video
- interaction_start, interaction_end = frame number at the start and end of the interaction
- nbr of males involved = count of males participating in the interaction
- interaction_length_sec = duration of the interaction in seconds
- N_total_landing = number of males observed in total
- "2025_field_videography_males_on_LED.csv" describes all males behaviour on the LED. Columns are as follows:
- Video_date, video_ID = date and ID of the analysed video
- video_duration = total length in seconds
- male_count = number of males observed in total
- male_at_LED = ID of male on the LED in that time frame
- on_LED_start = frame number when the male touched the LED
- on_LED_end = frame number when the male exited the LED
- time_on_LED = in seconds, total time spent by the male on the LED
- males_on_LED_ID = ID of all males present on the LED during this time frame
- exit_type = mode of exit from the LED, when several males were present, includes male ID
- exit_code = standardised mode of exit
- exit_group_id = ID given to the male or group of males who exited the LED simultaneously
- size_of_exit_group = number of males who exited simultaneously
- "2025_field_videography_sampled_male_counts.csv". For each video, once 16-19 random frames are extracted, the number of males present on the LED, on the Wire, or on the cloth were summarised here. Columns are as follows:
- Video_date, video_ID = date and ID of the analysed video
- video_duration = total length in seconds
- whole_video_male_count = number of males observed in total
- Slice = random frame number
- On_LED, On_wire, On_cloth = count of males at each location
- Actual_slice = Frame number to cross-reference with original video frames.
Code/software
All statistical analyses were conducted in R 4.3.2 (R Core Team, 2023). The packages used are listed at the top of each script.
