Food can be available to a giraffe without meeting its needs. The animal is a browser, adapted to eating leaves and shoots, and supplying suitable browse throughout the year is difficult for many zoos. Historical reports of zoo giraffes dying with severely depleted fat reserves raised a troubling question: were animals in managed care carrying fewer reserves than giraffes finding their own food?
Answering that question required a common measure. A giraffe's height and weight say how large it is; neither alone describes how much tissue covers its frame. Routine weighing is also difficult outside facilities with trained animals. Irina Clavadetscher and colleagues therefore looked for visible features that could turn photographs into consistent records of body condition.
Their 2021 study in the Journal of Zoo and Aquarium Research found the most useful clues around the hips. Once the team had arranged those contours into seven ordered scores, the zoo–wild comparison overturned their expectation. Zoo giraffes generally had higher scores, consistent with being less limited by food resources than the wild sample. The same scale also picked up seasonal differences in Tanzania, opening a way to investigate how changing conditions in a landscape are reflected in the animals themselves.
Clavadetscher worked with Monica Bond, Louise Martin, Christian Schiffmann, Jean-Michel Hatt and Marcus Clauss. By combining zoo photographs with images from a long-running Tanzanian field project, they made those two settings comparable through the same anatomical features.
Why look at the hips?
Body condition and body size answer different questions. Size describes how big an animal is. Condition concerns the tissue and reserves it carries on that frame. A growing calf and an adult male can have very different weights without either being in poor condition. Our giraffe weight and size guide explains that distinction in the context of kilograms and height.
A condition score looks for anatomical landmarks whose appearance changes as the tissues around them become more or less prominent. The team examined several candidates: ribs, skin folds, shoulders and hips. Giraffe anatomy made some familiar-looking signs surprisingly unhelpful.
Ribs were clearly visible in only two images from a combined pool exceeding 30,000 photographs. Skin folds occurred across the whole range of hip scores and were more common at higher scores. They could not serve as a straightforward sign of an animal running short of reserves.
The shoulder offered only four distinguishable levels, and its appearance did not always agree with the hips. Shoulder shape includes muscle and skin as well as tissue beneath the skin; in adult males, this region is also involved in the physical demands of fighting. A fuller-looking shoulder therefore need not imply the same condition signal as a fuller-looking hip.
Around the hips, the team could distinguish seven stages. At the lowest end, the backbone and angular hip bones were conspicuous, with hollows beside the spine. At higher scores, hollows became shallower, the outline rounded and the area beside the backbone became convex. A rear view made those contours easier to see against the background.
The result was an ordinal scale: the categories run in a meaningful order, but a one-level step does not correspond to a fixed amount of fat or a set number of kilograms. A score gives researchers a shared description to compare, not a hidden weighing machine.
Turning field photographs into usable observations
The wild photographs came from Masai giraffe surveys in the Tarangire Ecosystem between 2012 and 2019. Those surveys used coat patterns to recognise individuals. An image good enough to establish identity was often unsuitable for reading body contours: the animal might be moving, partly concealed, too distant or poorly lit.
Only 532 wild photographs met the condition-scoring requirements. The zoo comparison included 223 giraffes from 90 institutions in Europe and North America. The contrast therefore brings one wild Masai population alongside animals from many facilities and several giraffe taxonomic groups.
Zoos were asked to supply standardised views with the animal standing squarely, its head raised and the photographer at the same level. Front views proved unhelpful. Rear and rear-oblique views gave the clearest view of the hip region, while side photographs could also be used.
This attention to the image is part of what makes the method useful. A shadow can deepen a hollow, a coat patch can imitate a bone edge, and a change in angle can alter the apparent outline. Comparable photographs make it easier to tell a change in the animal from a change in how it was photographed.
A seasonal signal, especially in females
Wild giraffes photographed after the long wet season generally had higher hip scores than those photographed after the dry season. The pattern was clearer in adult females, while adult males showed little seasonal difference.
That is the kind of biological variation a condition indicator should be able to reveal. Seasonal changes in food availability can alter the balance between energy gained and energy spent. For females, pregnancy and milk production may add further demands. The authors discuss reproduction as a possible explanation for the female pattern; the photographs themselves did not establish each female's energy balance or reproductive state.
Nor was this a demonstrated decline and recovery within every individual. The analysis compared seasonal samples from one regional wild population. It shows a pattern worth following through time, with local rainfall and food conditions attached to the interpretation.
Age mattered too. Juveniles had high average hip scores in both the wild and zoo samples, and zoo scores generally decreased with age. Body proportions, growth and life stage affect what a contour means. An adult cannot simply be judged against a calf's appearance.
That distinction suggests a useful application. Repeated photographs of known females, paired with rainfall and feeding observations, could show whether particular animals lose condition in difficult seasons and regain it when browse improves. The present seasonal comparison identifies the pattern to follow; tracking the same individuals would reveal the trajectories behind it.
The zoo result overturned the starting expectation
Zoo giraffes generally scored higher than the wild sample. The reported means were about 4.5 and 3.7, respectively, with overlapping distributions. The contrast was greatest when zoo animals were compared with wild photographs taken after the dry season. It remained when adults and sexes were examined separately.
The team had expected the difficulties of feeding browsers in zoos to produce the opposite result. Instead, the observed difference was consistent with zoo animals being less constrained by seasonal food resources. The authors also considered improvements in giraffe nutrition and husbandry as a possible part of the explanation.
The result challenges a general expectation of low condition in zoo giraffes, but it cannot identify which husbandry practices produced the difference. The two samples were photographed across different years and conditions, and the animals were not assigned experimentally to their settings. Nor did the study compare illness or survival across score levels, so a fuller outline is not automatically evidence of better health.
Even the highest score, 7, occurred in both settings. The scale describes appearance; its upper end was not established as an obesity threshold. A low score likewise calls for context: limited food, reproduction and illness could produce different reasons to investigate a similar outline.
How much confidence belongs in a score?
When the same observer scored 141 images again after two weeks, 87.2% received exactly the same hip score; the rest differed by one level. That shows useful consistency for the person who developed the scale. Agreement between different observers was not tested.
The image itself introduced more variation. Rear and side views of 73 giraffes received identical scores in just over 64% of comparisons. A small difference between two photographs may therefore reflect viewing conditions rather than a biological change.
There was also an encouraging link with a separate measure of condition. In small zoo subsets, hip scores correlated with body mass divided by shoulder height, including a group of 14 adult females. But that index had not itself been validated for giraffes, measurements and photographs could be almost 100 days apart, and the study did not directly measure body fat.
Those findings support the score as a promising monitoring tool while explaining why repeated, comparable observations are more informative than a confident verdict on one photograph.
Connecting appearance with what an animal eats
A 2025 follow-up by Louise Martin, Clavadetscher and Clauss returned to the same zoo photographic scores and linked them with a husbandry and feeding survey. Its analysis of 211 giraffes from 89 institutions found lower scores among lactating females than females that had never given birth. Estimated consumption of high-fibre pellets was positively associated with scores, while cereals and low-fibre pellets showed the opposite association.
This extended the original dataset from describing condition to asking what might influence it. Because the survey estimated consumption and tested many associations without adjusting for multiple comparisons, the links are exploratory. They are also snapshots: a diet might influence condition, while an animal losing condition might be offered a different diet. Following actual intake and condition through time would help distinguish those directions.
A complementary 2021 study by Keisuke Kawase and colleagues followed one zoo giraffe using husbandry training to monitor weight, food intake, condition scores and blood measurements. The animal’s intake, weight and scores changed with the seasons, alongside changes in several blood measures. This illustrates what repeated observations can add: a changing outline can be interpreted against an animal’s own feeding and weight history. The study used an earlier eight-level condition scale and one animal, so it cannot validate Clavadetscher’s seven-level method across giraffes.
Questions a useful photograph opens up
The photographic method opens up several questions about monitoring giraffes over time.
Can different observers read the same change? Having several scorers independently assess comparable photographs would show how well the categories transfer beyond the person who developed them, and where uncertainty is greatest.
Which tissues change when the outline changes? Pairing repeat photographs with safely obtained weight and non-invasive tissue measurements in trained giraffes could establish whether changes in a score track fat, muscle or both, while accounting for growth and pregnancy.
Can a condition trajectory reveal difficulty early? Following known wild females through seasons could test whether a sustained score change adds information about later survival or reproduction beyond age, rainfall and habitat. The useful outcome would be an earlier signal to investigate, with its meaning grounded in the animal's wider circumstances.
Sources
- Clavadetscher, I., Bond, M. L., Martin, L. F., Schiffmann, C., Hatt, J.-M. & Clauss, M. (2021). Development of an image-based body condition score for giraffes Giraffa camelopardalis and a comparison of zoo-housed and free-ranging individuals. Journal of Zoo and Aquarium Research, 9(3), 170–185. DOI: 10.19227/jzar.v9i3.615. Published open-access article.
- Martin, L. F., Clavadetscher, I. & Clauss, M. (2025). Survey of husbandry and feeding management and potential influencing factors on the body condition score of zoo-housed giraffes Giraffa camelopardalis spp. in Europe and Northern America. Journal of Zoo and Aquarium Research, 13(2), 86–95. DOI: 10.19227/jzar.v13i2.867. Published open-access article.
- Kawase, K., Hirayama, K., Kawano, M., Yayota, M., Ito, S. & Shiihara, S. (2021). Long-term monitoring for assessing the nutritional status of a captive giraffe using body weight and blood compositions. Animal Behaviour and Management, 57(2), 39–45. DOI: 10.20652/jabm.57.2_39. Publisher’s full paper.
