Validation of the Elephant Welfare Assessment Tool (EWAT) using Behavioural and Fecal Glucocorticoid Metabolites measures in captive Asian elephants (Elephas maximus) at tourist camps in Chiang Mai, Thailand

Abstract

Elephant welfare in captive settings is highly controversial, especially in tourist camps in Thailand, where many venues are now boycotted by Western tourists. This shift is largely driven by public perceptions of what constitutes “ethical” interactions with elephants. However, to move beyond beliefs and emotions, structured and scientifically grounded tools are needed to assess elephant welfare. The Elephant Welfare Assessment Tool (EWAT), developed by the Center of Elephant and Wildlife Health at Chiang Mai University, was designed to address this need. It comprises 42 indicators categorized into four domains: Environment, Nutrition, Health, and Behaviour and Mental State. The primary objective of this study was to validate the EWAT by comparing it with two complementary welfare indicators: a Behaviour tool (BV) specifically developed for this research (based on focal sampling and Qualitative Behaviour Assessment), and fecal glucocorticoid metabolite (FGM) concentrations as a physiological stress marker. Data were collected from 31 elephants housed at three tourist camps in Thailand. We found no relationship between the EWAT and either of the two other indicators. However, elephants with higher BV scores had lower FGM concentrations, indicating better overall welfare. Moreover, we found that FGM concentrations were lower in Riding elephants compared to No riding elephants, contradicting the public belief that “Riding” is systematically negative for elephant welfare. We also examined the prevalence and determinants of stereotypic behaviours (SB) and found that 93.5% of elephants displayed SB at least once. Old elephants and those in Camp B were less likely to exhibit SB. We observed that FGM concentrations increased with total SB duration, refuting the coping hypothesis (i.e. SB help reduce stress), and supporting the idea that SB in this context may reflect ongoing stress or frustration. Finally, we assessed the effect of chaining on behaviour. Elephants expressed a higher proportion of negative behaviours – mostly stereotypies – and a lower proportion of positive behaviours when chained. These results indicate that the EWAT requires adjustments to improve its sensitivity by: (1) incorporating more detailed behavioural assessments, particularly of SB; (2) evaluating behaviours under both chained and unchained conditions; and (3) integrating FGM analysis whenever feasible. With these refinements, the EWAT would integrate a broader range of indicators, allowing a more accurate interpretation of welfare outcomes within a single comprehensive tool. We hope this study will contribute to the development of evidence-based guidelines for elephant management in tourist camps and that future research will include longer sampling periods to better capture welfare dynamics over time.