Unvalidated automated bat species classifications derived from acoustic recordings made at a sample of three National Forest Inventory squares, surveyed as part of the National Woodland Bat Survey. Data from additional squares are available on request.
The National Woodland Bat Survey is an England-wide passive acoustic bat survey conducted across National Forest Inventory squares. Bat echolocation calls within audio recordings are detected and classified to species using automated machine-learning classifiers. These sample data have not been validated, so will include incorrect species classifications. They are of use in applications where it is preferable for classification uncertainty to be modelled and incorporated into analyses, and inference is made at regional or national scale, thereby reducing the impact of individual misclassifications. An alternative dataset of validated and accepted species records will be made available at a future date.
Data are georeferenced using a delocalised grid. The bounding coordinates for each grid square are included in the dataset. Further information about the delocalisation process is provided in the dataset evidence pack.
The 'National Woodland Bat Survey - Example Dataset’ is a project part-funded by the UK Government through Defra’s Natural Capital and Ecosystem Assessment programme (NCEA). Users should ensure they have read the supporting documentation before viewing or making use of the data. This includes the NCEA Analysis Ready Data Limitations Statement and dataset specific limitations/caveats outlined in the supporting documentation.
Attribution statement: © Forestry Commission & Bat Conservation Trust copyright and/or database right 2025. All rights reserved.