ASSESSMENT OF AFRICAN RHINOCEROS TEMPERAMENT AND ITS IMPLICATIONS FOR MANAGEMENT

nsight into an animal’s mental state, i.e., how it responds to the environment, conspecifics and humans, is important for interpreting its wellbeing and can be monitored via temperament assessments. The temperaments of 49 black (Diceros bicornis, BR) and 125 white (Ceratotherium simum, WR) rhinos in North American zoos were assessed using temperament surveys completed by rhino keepers. Median survey ratings by the keepers on 12 temperament traits were clustered by principal components analysis into temperament profiles related to bold/fearful, active/inactive and anthrophilic/-phobic traits for both species, and reactive/stoic traits for white rhinos. Additionally, novel object tests with a rope basket were conducted on 32 black and 66 white rhinos to measure exploration/avoidance, active/inactive, and bold/fearful temperaments. Proportion of time active and alert, frequency of interactions, and latency to approach the basket were analyzed. Temperament profiles and novel object test behaviors grouped by species were analyzed using linear mixed-effects models and negative binomial models, respectively, for their relationship to age, sex, sociality (individual/social housed), and institution. Additionally, the influence of test parameters (e.g., location of test, ambient temperature), were included for the novel object test behaviors. The survey temperament profiles indicated that males of both species were more active than females, and activity declined with age in WR (p<0.05). The institution at which WR were housed was associated with anthrophilic/-phobic and reactive/stoic traits (p<0.05). Age, temperature and enclosure size were associated with behaviors in both species during the novel object test, whereas sex, calf presence and social housing additionally had associations with WR behaviors (p<0.05). Results suggest younger BR and WR and individually housed WR might be more likely (p<0.05) to engage in novel situations. Multiple behavioral indicators of active/inactive, anthrophilic/-phobic, bold/fearful, exploratory/avoidant, and sociable/aggressive with conspecifics were correlated in both assessments (p<0.05) reinforcing the utility of both tools for assessing the temperament of rhinos. Chemosensory ratings from the survey in BR were negatively related (p<0.05) to retreat and stimulated during the novel object test, indicating more exploratory and less fearful rhinos. Numerous other behavioral responses of BR to the basket were related to survey traits, and BR engaged more frequently with it than WR; together these suggest the novel object test was a good tool for assessing BR temperament. Pant calls are a signal to conspecifics and possibly humans; during the novel object test they were negatively related (p<0.05) to survey traits of positive interactions with humans, and thus were a behavioral indicator for more anthrophobic WR. The gate wait behavior during the novel object test also correlated with multiple WR survey traits. Thus, social WR appeared to react more to the novel separation from conspecifics rather than the novel basket. Location of the novel object test (e.g., enclosure size, indoors or out, and ambient temperature) also was associated with behavioral responses of the rhinos. Therefore, potential influence of study design should be considered when interpreting temperament data gathered via novel object tests. Surveys provided a complex temperament profile, however also were potentially influenced by institution, particularly in WR. Both temperament tools illustrate rhinos’ tendency towards fearfulness and exploration, which allows modification of training, the environment, or other aspects of care to minimize potentially aversive experiences. Consideration of a rhino’s natural activity and comfort with humans enables tailoring expectations for daily behaviors, interactions with enrichment, conspecifics, and people. Improving individual care in these ways could promote positive mental states and overall wellbeing.