Dataset for VOI-Rice: Value-of-information-driven ground-sensor orchestration for UAV-based rice physiological monitoring under cross-flight domain shift

This dataset supports the manuscript “VOI-Rice: Value-of-information-driven ground-sensor orchestration for UAV-based rice physiological monitoring under cross-flight domain shift.” Dataset overview The dataset contains 1,280 paired UAV–ground observations collected from a 51.2-ha irrigated lowland rice landscape in the Plaosan agricultural area, Tlogoadi Village, Mlati District, Sleman Regency, Special Region of Yogyakarta, Indonesia. Data were collected during three UAV and ground-measurement campaigns conducted on 14 February, 24 February, and 4 March 2026. Data acquisition Multispectral imagery was acquired using a fixed-wing vertical-take-off-and-landing Trinity Pro UAV equipped with a MicaSense RedEdge-P sensor. The sensor recorded blue, green, red, red-edge, and near-infrared reflectance. Ground physiological measurements were collected using a SPAD-502 Plus chlorophyll meter and an LI-600 porometer/fluorometer. Dataset contents The workbook contains one worksheet with 1,280 observation records and 447 variables. The variables comprise: 1. Observation, flight, acquisition-domain, sampling-block, and point identifiers.2. Planting dates, UAV acquisition dates, days after planting, timestamps, and spatial coordinates.3. Rice phase, growth-stage, and BBCH-related phenological information.4. Quality-control and spatial-mapping eligibility flags.5. Statistical descriptors of the blue, green, red, red-edge, and near-infrared bands.6. A total of 378 UAV-derived predictor variables, including statistical descriptors of vegetation indices such as NDVI, GNDVI, NDRE, EVI, EVI2, SAVI, OSAVI, CCCI, MTCI, TCARI, ARVI, and related indices.7. Ground-measured physiological and instrument variables, including SPAD, stomatal conductance (gsw), electron transport rate (ETR), effective quantum yield of photosystem II (PhiPS2), leaf temperature (Tleaf), and leaf vapor-pressure deficit (VPDleaf). Phenological coverage The retained observations cover five rice growth-stage groups: - Tillering: 385 observations- Stem elongation/panicle initiation: 165 observations- Booting: 450 observations- Heading–flowering: 170 observations- Grain filling: 110 observations Recommended analytical use The dataset was prepared for leakage-safe cross-flight modelling, leave-one-flight-out validation, UAV feature-domain analysis, uncertainty calibration, out-of-domain assessment, adaptive sampling, and value-of-information-based ground-sensor orchestration. Flight and sampling-block identifiers should be respected when constructing training and validation subsets. Random row-wise splitting may result in information leakage and overly optimistic model evaluation. Quality control and limitations Target-specific missingness is retained for selected physiological variables. The dataset contains curated tabular UAV–ground observations and does not include raw UAV images or complete multispectral orthomosaics. It represents one rice-growing landscape and one seasonal observation window. External validation is therefore required before transferring models or decision policies to other locations, seasons, cultivars, management conditions, or stress scenarios. Physiological deviation values should be interpreted as indicators for targeted verification and not as definitive evidence of agronomic stress or yield loss. License The dataset is released under the Creative Commons Attribution 4.0 International license. Users may share and adapt the dataset provided that the dataset creators and the corresponding Zenodo DOI are appropriately cited.