A misleading tail: a long-term study of reptile responses to multiple disturbances undermined by a change in surveying techniques

Long-term ecological monitoring is crucial to understanding the complex dynamics of ecosystems, communities, and populations. Despite this, monitoring data are lacking or rare for the vast majority of biodiversity. Here we report the results of 19 years (2003-2022) of continuous annual monitoring of reptile species at Booderee National Park (BNP) on the east coast of south-eastern Australia. We tested the effects of time, habitat type, fire, and climate on detections of five reptile species. Our study revealed declines in detections of two skink species over time (<i>Lampropholis delicata </i>and<i> </i><i>Ctenotus </i><i>taeniolatus</i>), which we suspect was partly driven by weather conditions influencing activity of these species. We also identified broad vegetation type associations for two congeneric species with <i>L. delicata </i>being associated with forested sites<i>, </i>and <i>Lampropholis guichenoti</i> associated with more shrubby sites. Our results also demonstrated a clear association between <i>Cryptophis nigrescens </i>and <i>L. delicata</i> and fire, with the probabilities of detection of both species decreasing with time since fire in the short term. At about the midway point of our study (in 2011), we were forced to make a change in the way our data were collected. The change heavily influenced our findings, and so breached the integrity of the time series in our dataset. We acknowledge that a simple but crucial step to mitigate this breach would have been to conduct calibration that allowed subsequent analysis to control for a change in field survey methodology. Whilst improvements in the effectiveness of field survey methods might be possible through new technologies, it is crucial to maintain the integrity of long-term datasets as data collection continues.