[MARMAM] New paper on estimating body mass of free-living whales using drone photogrammetry and 3D modelling
f.christiansen at live.se
Wed Oct 2 05:07:47 PDT 2019
My co-authors and I are happy to announce the publication of the following paper in Methods in Ecology and Evolution:
Christiansen, F., Sironi, M., Moore, M.J., Di Martino, M., Ricciardi, M., Warick, H.A. & Uhart, M.M. 2019. Estimating body mass of free-living whales using aerial photogrammetry and 3D volumetrics. Methods in Ecology and Evolution: 1-11. DOI: /10.1111/2041-210X.13298
In this study we used unmanned aerial vehicles, or drones, to measure the body shape (length, width and height) of southern right whales in Pen¨ªnsula Vald¨¦s, Argentina, from which we estimated body volume. We also calculated the body girth of the free-living whales, and developed a model to predict body volume from body length and girths. Using this model we then predicted the volume of North Pacific right whales caught in whaling operations, for which body length, girth and mass was known. From this we could calculate a volume-to-mass conversion factor, or body density, of the dead whales, which we then applied to the free-living southern right whales to calculate their body mass.
Please find the abstract and a link to the paper below:
1. Body mass is a key life©\history trait in animals. Despite being the largest animals on the planet, no method currently exists to estimate body mass of free©\living whales.
2. We combined aerial photographs and historical catch records to estimate the body mass of free©\living right whales (Eubalaena sp.). First, aerial photogrammetry from unmanned aerial vehicles was used to measure the body length, width (lateral distance) and height (dorso©\ventral distance) of free©\living southern right whales (Eubalaena australis; 48 calves, seven juveniles and 31 lactating females). From these data, body volume was estimated by modelling the whales as a series of infinitely small ellipses. The body girth of the whales was next calculated at three measurement sites (across the pectoral fin, the umbilicus and the anus) and a linear model was developed to predict body volume from the body girth and length data. To obtain a volume©\to©\mass conversion factor, this model was then used to estimate the body volume of eight lethally caught North Pacific right whales (Eubalaena japonica), for which body mass was measured. This conversion factor was consequently used to predict the body mass of the free©\living whales.
3. The cross©\sectional body shape (height¨Cwidth ratio) of the whales was slightly flattened dorso©\ventrally at the anterior end of the body, almost circular in the mid region, and significantly flattened in the lateral plane across the posterior half of the body. Compared to a circular cross©\sectional model, our body mass model incorporating body length, width and height improved mass estimates by up to 23.6% (mean = 6.1%, SD = 5.27). Our model had a mean error of only 1.6% (SD = 0.012), compared to 9.5% (SD = 7.68) for a simpler body length©\to©\mass model. The volume©\to©\mass conversion factor was estimated at 754.63 kg/m3 (SD = 50.03). Predicted body mass estimates were within a close range of existing body mass measurements.
4. We provide a non©\invasive method to accurately estimate body mass of free©\living whales while accounting for both their structural size (body length) and relative body condition (body width). Our approach can be directly applied to other marine mammals by adjusting the model parameters (body mass model script provided).
The paper can be accessed from the following link:
This study was done in collaboration with the Digital Life Project at the University of Massachusetts at Amherst and CG artist Robert Gutierrez, which created a 3D model of a southern right whale. To access the 3D mesh model of a right whale you can follow this link:
To access a full colour 3D model of a southern right whale you can follow this link:
Assistant Professor/Postdoctoral Research Fellow
Aarhus Institute of Advanced Studies
Aarhus University, Denmark
f.christiansen at aias.au.dk
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the MARMAM