Accessibility drives research efforts on Amazonian sarcosaprophagous flies

The tropics hold most of the planet's biodiversity but face significant knowledge gaps. This is particularly concerning in the Brazilian Amazon, where anthropogenic disturbances are driving species loss. Our study focused on sarcosaprophagous flies, a group with key roles in public health and ecosystem functioning. Using 8244 occurrence records of flies and machine learning, we mapped knowledge distribution at three levels: families, the best-sampled species and a null model simulating chance knowledge probability. Analyses revealed substantial biases. Sampling was uneven, with approximately 40% of forested areas showing <10% probability of occurrence for families and species, while 80% of the region aligned with null expectations. Knowledge probability increased with accessibility, and species were better documented in degraded areas, exceeding chance expectations, whereas remote areas of high conservation value, including Quilombola territories, were neglected. These patterns were consistent across taxonomic levels, highlighting that addressing knowledge biases requires more than increasing research effort. Our findings underscore the importance of studying less-charismatic groups, such as sarcosaprophagous flies. We bring new insights into the value of targeted surveys in remote areas and collaborative engagement with local traditional communities, essential for building a comprehensive understanding of biodiversity and promoting effective conservation in the Amazon.