A georeferenced dataset for grapevine phenology and phytosanitary monitoring across Italy

MIRA (MonItoRaggio fitopAtie) is a georeferenced dataset of field observations on grapevine (Vitis vinifera L.) phenology, diseases, insect pests and plant protection treatments, collected from 2020 to 2025 in 70 vineyard stations (65 farm-managed and 5 sentinel) across 9 Italian regions, within the monitoring network established by CREA with regional plant health services (AgriDigit-Agromodelli and MISFITS projects). Data were recorded by regional technicians with a dedicated mobile application, following a common field protocol. The dataset contains about 42,000 records in six related tables: station metadata, host phenology (BBCH scale), disease intensity scale and McKinney disease index (downy mildew, powdery mildew and grey mould), first symptom onset, insect trap captures and plant protection treatments. Hourly weather data for each station (2019–2025, Open-Meteo) are provided as a complement. 📂 Repository contents 📊 MIRA_data.xlsxThe six tables as Excel sheets, with the same content as the .rds files (weather data not included because of their size). 📦 MIRA_data.zip stations.rds: station metadata (70 records) host.rds: host phenology, BBCH scale (24,037 records) disease.rds: disease intensity scale and McKinney disease index (11,867 records) first_symptoms_onset.rds: date of first symptoms (204 records) insect_traps.rds: insect trap captures (5,391 records) treatments.rds: plant protection treatments (421 records) weather_hourly.rds: hourly weather data for each station, 2019–2025 (Open-Meteo) build_mira.R: generates MIRA_data.xlsx from the .rds files; includes commented examples for filtering, joining tables and exporting subsets (R packages: writexl, dplyr) mira_weather.R: examples of combined analyses of weather and monitoring data (R packages: dplyr, ggplot2) The .rds files are the reference version of the dataset. 📄 MIRA_protocol.pdfEnglish translation of the field monitoring protocol followed by the technicians to collect the data. The protocol is part of an unpublished internal technical document (the user manual of the MIRA mobile application) and is provided here only to support the interpretation and reuse of the dataset. 💻 webApp_MIRA.zipSource code of the Shiny web app (app.R and www/) and a copy of MIRA_data.xlsx, from which the app reads the data. To run it locally, open app.R in RStudio and click "Run App".🌐 Online version: geomodellabcrea.shinyapps.io/webApp_MIRA 🔗 All tables are linked through station_code. See README.txt for the data dictionary (data types, units, permissible values and coding of each variable). 🔄 Changes in version 2.0 Revised version of the dataset. Disease variables renamed: incidence_pct to disease_index, severity_sum to intensity_sum, severity_max to intensity_max, as the four-class scale (named "severity classes" in the field protocol) combines the number of symptomatic organs and the extent of symptoms; descriptions corrected. Data entry anomalies checked: confirmed errors corrected or removed, onset dates recalculated, unverified anomalies flagged (data_quality_flag). Station metadata revised; stations registered in the application but not part of the monitoring network excluded. Hourly weather data (weather_hourly.rds) and the script mira_weather.R added. English translation of the field monitoring protocol added. Data dictionary added to README.txt. 💶 Funding This work was supported by two projects funded by the Italian Ministry of Agricultural, Food and Forestry Policies: AgriDigit-Agromodelli, DM n. 36502, 2018–2023, and AgriDigit-MISFITS (Modellistica fitosanitaria), DM n. 418538, 2024–2025. ⚖️ Licence and attribution The MIRA dataset and code are released under Creative Commons Attribution 4.0 International (CC BY 4.0). 🌦️ Weather data by Open-Meteo.com, from the Open-Meteo Historical Weather API, licensed under CC BY 4.0. Changes made: hourly data were checked for completeness and plausible ranges, variable names were standardised, and leaf wetness was added.