“…and the forests will echo with laughter!”: Remote sensing and bioacoustic integration for assessing Picus viridis distribution in structurally heterogeneous riparian forests

Riparian forests are biodiversity hotspots, yet their structural complexity has long hindered systematic habitat monitoring. Until recently, the lack of integrated technologies limited the identification of priority conservation areas at fine spatial scales.We investigated the habitat suitability of P. viridis, a bio-indicator of mature and structurally complex woodlands, in three riparian biotopes of South Tyrol (Italy). Bird presence data were obtained from 12 autonomous AudioMoth recorders cyclically deployed across the study areas (Valsura: 10, Sluderno: 8, Aurina Valley: 6). Recordings were processed with the BirdNET model and converted into binary presence/absence data. Habitat predictors included LiDAR-derived canopy metrics and vegetation indices from Sentinel-2 imagery, namely mean and standard deviation of the Normalized Difference Vegetation Index (NDVI).Species distribution models were developed using Partial Least Squares Regression, Random Forest, Extreme Gradient Boosting, and Maximum Entropy. Model performance was assessed with the Area Under the Curve (AUC) for both training and test sets. All models consistently identified mean and standard deviation of NDVI and LiDAR-derived canopy height percentiles as key predictors. Suitability maps highlighted high-probability areas for P. viridis in forests with mature trees, complex canopy architecture, and elevated NDVI.This study demonstrates that combining autonomous acoustic monitoring with high-resolution remote sensing offers a cost-effective and replicable framework for avian biodiversity assessment and conservation planning.