Local ecological knowledge and line-transect surveys for estimation of species occupancy data in Central Amazon

The Amazon, renowned for its exceptional biodiversity, faces significant threats from deforestation. Effective biological monitoring, essential for conservation strategies, remains challenging in these remote areas. This study assesses the accuracy of Local Ecological Knowledge (LEK) in estimating habitat occupancy in the Central Amazon, comparing traditional line-transect surveys with structured interviews of local people. Sampling targeted 13 fluvial islands and 5 continuous forest sites, focusing on 10 taxa, including birds, mammals, and reptiles. The interviews provided higher detection (p=0.73) and occupancy (ψ=0.70) probabilities compared to line-transect (p=0.42, ψ=0.52). The positive correlation between interviews and line-transect occupancy estimates (0.69; p<0.03), suggests the potential of LEK-based methods as a surrogate for traditional surveys. The interviews provided insights into iconic species occupancy, including brown-throated sloth (Bradypus variegatus) and jaguar (Panthera onca). The jaguar, brown-throated sloth and red howler monkey (Alouatta juara) demonstrated highest occupancy probabilities in line-transects. The jaguar, brown-throated sloth, and red howler monkey showed the highest detection probabilities in interviews, while red howler monkey exhibited the highest in line-transects. We highlight the potential of integrating LEK-based approaches with traditional surveys to attempt more comprehensive, rapid and cost-effective monitoring of biodiversity, crucial for conservation efforts in the Amazon and similar remote ecosystems globally.