Opening a larger window onto forest elephant ecology

African forest elephants (Loxodonta cyclotis) can be observed reliably in forest clearings but are difficult to observe directly in their otherwise forested habitat. To obtain data on population demography and activity cycles, conservation programmes commit considerable human and financial resources to observations at clearings. However, individual elephants spend only a tiny proportion of their annual activity budget in a particular clearing, and there may be demographic differences between daytime and nighttime populations. Using acoustic monitoring tools to assess elephant numbers continuously for long time periods at multiple clearings, we show that daytime‐only observations are likely to be biased. In this study, (i) 79% of all elephant visitation occurred at night, (ii) sometimes large changes in nocturnal elephant numbers occurred without proportional changes in daytime elephant numbers, and (iii) there were indications that a different demographic of the population visits clearings at night compared with the day. These results suggest that acoustic monitoring should be incorporated into forest elephant monitoring programmes to augment direct observation and that more study is needed to explicitly identify any biases inherent in daytime‐only observation.