Data validation of botanical field surveys: an R Shiny tool for the New Italian National Forest Inventory (IFNI 2025)

The digitalization of data collected during botanical field surveys represents a critical step in the scientific workflow: it is in this transition step — from the field to the database — that observer errors occur most frequently. Physical and mental fatigue, large data volumes, and demanding working conditions make small transcription errors inevitable; if not detected early, these errors propagate through the entire analytical chain and may become impossible to correct after their storage and publication. The field surveyor, who made field observation, is the most likely person to check and correct those errors, if feedback is received while the context of the survey is still fresh in memory. In this contribution we present a self-validation approach developed within the framework of the New Italian National Forest Inventory 2025 (IFNI 2025), a nationwide survey campaign involving expert botanists working independently across hundreds of plots throughout Italian forest ecosystems. The tool, implemented as an R Shiny application, enables surveyors to validate their own data autonomously before submission to the data manager, intercepting errors at the source rather than downstream. The adopted approach — defined as a zero-error gate — ensures that no data reaches the coordinator until all formal checks have been passed, relieving the coordinator from the mechanical review of files and allowing field surveyors to focus on verifying the completeness of the campaign. The system accommodates any number of observers without additional configuration and applies the same validation rules consistently to all users regardless of their experience. Preliminary results from the IFNI 2025 campaign confirm the effectiveness of the approach in eliminating transcription errors prior to data submission.