Large scale faecal (spraint) counts indicate the population status of endangered Eurasian otters (Lutra lutra)

The validity of using spraint (otter faeces) density for population monitoring has been debated for more than 30 years. In this study, we investigated endangered Eurasian otter (Lutra lutra) spraint occurrence and densities at large scales (over 23,800 km2, a quarter of South Korea) over three years (2014–2016). To clarify the spatial heterogeneity of spraint density and count distributions, we applied the global Morans’ I test and hot spot analysis. We also constructed models with 30 environmental factors (six landscape, eight anthropogenic, 13 aquatic health indices, one prey abundance, and two meteorological factors) using generalized linear mixed models with repeated measurements. Our geographical analysis showed regional clusters of otters extending over distances of more than 80 km. The most parsimonious model, a zero-inflated negative binomial model, indicated that our otter spraint counts were significantly positively related to the benthic macro-invertebrate index and precipitation and negatively related to proportion of home range covered by water. In addition, this model showed that absence probabilities of otter spraint were significantly positively related to human populations and negatively related to the number of fish species and altitude. The best explanatory model suggests that our count data was highly related to otter population status, and also affected by anthropogenic disturbance.