Encadrement : Laura Mannocci (chercheuse IRD), David Mouillot (professeur Université de Montpellier)
Organisme d’accueil : UMR MARBEC
Lieu : Montpellier (campus triolet)
Durée : 6 mois
Période souhaitée : de Janvier à Juin 2024
Profil recherché :
- Master 2 ou 5ieme année ingénieur
- Bagage en modélisation statistique
- Maitrise du logiciel R
- Bon niveau d’anglais écrit (en vue d’un mémoire de stage en anglais sous forme d’article scientifique)
Renseignements et candidatures :
Laura Mannocci: https://www.researchgate.net/profile/Laura-Mannocci
Pour candidater, envoyez votre CV et une lettre de motivation par e-mail (laura.mannocci@ird.fr)
----------------------------------------------------------------------------------------------------------
Global assessment of dugong’s body condition through aerial photogrammetry
Context and objective of the internship
The dugong is listed as vulnerable by the IUCN globally (Marsh & Sobtzick 2019), with New Caledonian and East African subpopulations recently assessed as endangered and critically endangered, respectively. To monitor dugong’s abundance, aerial surveys are conducted in different parts of its Indo-Pacific range (Hines et al. 2005; Garrigue et al. 2008; Appoo et al. 2019; Mannocci et al. 2021; Trotzuk et al. 2022). However, aerial surveys in areas of low dugong densities often lead to few sightings despite high amounts of deployed effort. Abundance estimates derived from aerial surveys are also prone to uncertainty, making it difficult to detect changes in population size (e.g. Trotzuk et al. 2022). Assessments of individual body condition from aerial images collected from drones or small aircraft fitted with cameras (e.g. Durban et al. 2016; Hodgson et al. 2020; Ramos et al. 2022) provide a new way of monitoring dugongs, in complement to traditional abundance estimates. Body condition is useful composite indicator of the overall nutritional, physical and health state of individuals (Wilder et al. 2016) and a surrogate of individual fitness, affecting the overall dynamics of the population (Stevenson & Woods 2006). While abundance estimation requires standardized aerial surveys fulfilling strict sampling assumptions and covering vast areas, opportunistic aerial videos are sufficient to derive a body condition index of individuals. With the advent of inexpensive drones and the development of tourism in remote areas, opportunistic aerial videos of dugongs are increasingly available through social media (Mannocci et al. 2021). Combined, videos from social media and standardized aerial surveys provide an unprecedented opportunity to investigate dugong’s body condition throughout their Indo-Pacific range.
The objective of this internship is to assess dugong’s body condition and investigate its geographical variations globally. Specifically, the intern will test the hypotheses that dugong’s body condition varies as a function of temperature (“stockier” individuals are expected in colder environments due to heat loss reduction needs), seagrass condition (“stockier” individuals in areas of high seagrass availability/productivity), remoteness from humans (“stockier” individuals in areas with low human populations and reduced coastal development) and the presence of enforced marine protected areas (“stockier” individuals in protected areas). The intern will also relate dugong’s body condition to the IUCN status of given sub-populations with the expectation of poorer body condition for individuals belonging to endangered and critically endangered sub-populations (New Caledonia and East Africa) compared to larger less threatened subpopulations (e.g. Australia or Qatar).
Methodology
The intern will follow the different steps outlined below:
(1) Assembling of aerial footage (already completed) –
A large database of aerial videos of dugongs collected in different parts of the Indo-Pacific has been assembled by contacting amateurs (through social media), NGOs and academics that own such footage (including our own aerial footage from New Caledonia and Indonesia). Videos were preferred to images because they can be converted in a series of images, allowing multiple views of the same individuals. For each aerial video, the collected metadata included: date, geographic location of the footage and drone/camera used.
(2) Image pre-processing (to do by the intern) –
All videos will be extracted into frames at a rate of 5 frames per second using FFmpeg software. All images will be quality-graded (using a 1-to-3 qualitative scale) based on camera position (nadir, close and centred on individual), body position (animal at the water surface), body posture (straight and non-arching body), environmental conditions (no glare, no water splash, no ripples) (Christiansen et al. 2019). Duplicate images of the same individual will be flagged but kept to investigate measurement errors (see next step).
(3) Body measurements (to do by the intern) –
Photogrammetry techniques will be used to derive a simple, relative body condition index based on the maximum body width to body length ratio for each individual, as developed by Ramos et al. (2022) for manatees. For each individual, body length and body width will be measured in existing free and open-source software (imageJ or morphometriX). In the absence of altitude information to derive ground sampling distance, all measurements will be done in pixels. When possible, several measurements (body width and body length) of the same individual will be made to quantify measurement errors. Mean and standard deviation of body condition will be derived for each geographic location (at country or regional scale) from all unique individuals measured in that location.
(4) Statistical modelling (to do by the intern) –
Statistical models will be developed in R to relate body condition to potential predictors reflecting socio-environmental conditions (such as temperature, seagrass availability, remoteness from humans, presence of an enforced marine protected area), as well as the IUCN status of the sub-population (when known).
References
Appoo, J., van de Crommenacker, J., Sanchez, C., Currie (SANBI), J., Burt, A.J., Fleischer-Dogley, F., et al. (2019). The Use of Aldabra and its Protected Waters by Marine Mammals.
Christiansen, F., Sironi, M., Moore, M.J., Di Martino, M., Ricciardi, M., Warick, H.A., et al. (2019). Estimating body mass of free-living whales using aerial photogrammetry and 3D volumetrics. Methods in Ecology and Evolution, 10, 2034–2044.
Durban, J.W., Moore, M.J., Chiang, G., Hickmott, L.S., Bocconcelli, A., Howes, G., et al. (2016). Photogrammetry of blue whales with an unmanned hexacopter. Marine Mammal Science, 32, 1510–1515.
Garrigue, C., Patenaude, N. & Marsh, H. (2008). Distribution and abundance of the dugong in New Caledonia, southwest Pacific. Marine Mammal Science, 24, 81–90.
Hines, E.M., Adulyanukosol, K. & Duffus, D.A. (2005). Dugong (dugong Dugon) Abundance Along the Andaman Coast of Thailand. Marine Mammal Science, 21, 536–549.
Hodgson, J.C., Holman, D., Terauds, A., Koh, L.P. & Goldsworthy, S.D. (2020). Rapid condition monitoring of an endangered marine vertebrate using precise, non-invasive morphometrics. Biological Conservation, 242, 108402.
Mannocci, L., Villon, S., Chaumont, M., Guellati, N., Mouquet, N., Iovan, C., et al. (2021). Leveraging social media and deep learning to detect rare megafauna in video surveys. Conservation Biology, 1.
Marsh, H. & Sobtzick, S. (2019). Dugong dugon (amended version of 2015 assessment). The IUCN Red List of Threatened Species 2019: e.T6909A160756767. IUCN Red List of Threatened Species. Available at: https://www.iucnredlist.org/en. Last accessed 3 September 2020.
Ramos, E.A., Landeo-Yauri, S., Castelblanco-Martínez, N., Arreola, M.R., Quade, A.H. & Rieucau, G. (2022). Drone-based photogrammetry assessments of body size and body condition of Antillean manatees. Mamm Biol, 102, 765–779.
Stevenson, R. & Woods, W. (2006). Condition indices for conservation: New uses for evolving tools. Integrative and comparative biology, 46, 1169–90.
Trotzuk, E., Findlay, K., Taju, A., Cockcroft, V., Guissamulo, A., Araman, A., et al. (2022). Focused and inclusive actions could ensure the persistence of East Africa’s last known viable dugong subpopulation. Conservation Science and Practice, 4, e12702.
Wilder, S.M., Raubenheimer, D. & Simpson, S.J. (2016). Moving beyond body condition indices as an estimate of fitness in ecological and evolutionary studies. Functional Ecology, 30, 108–115.