Longitudinal Multimodal Big Data Infrastructure for Precision Poultry Monitoring (Sample Dataset Release)

This record provides a structured sample release from a longitudinal multimodal agricultural sensing dataset collected over 22 consecutive weeks across five commercial-style poultry barns. The dataset integrates heterogeneous sensor streams including RGB video (1080p, 30 fps), audio (48 kHz WAV), radiometric thermal imaging (rJPEG), and twice-daily environmental measurements (temperature and relative humidity). The full dataset comprises approximately 10.7 terabytes of synchronized multimodal data. To ensure farm confidentiality and responsible data governance, this Zenodo record contains representative sample files illustrating the data structure, file naming conventions, metadata schema, and preprocessing workflows described in the associated Frontiers in Big Data manuscript. The sample release includes: Representative video segments in MP4 format Representative audio clips in WAV format Sample radiometric thermal image files Structured environmental CSV files Metadata documentation Folder hierarchy description Data dictionary Preprocessing pipeline documentation The objective of this release is to demonstrate the multimodal acquisition architecture, synchronization strategy, storage framework, and preprocessing design supporting scalable agricultural data systems. Access to the complete dataset is available upon request under a Data Access Agreement to protect farm confidentiality and animal facility privacy. This dataset supports research in: Multimodal data integration Temporal synchronization of heterogeneous streams Agricultural Big Data infrastructure Sensor fusion research Scalable AI deployment in precision agriculture Associated manuscript:A Longitudinal Multimodal Big Data Infrastructure for Precision Poultry Monitoring