Plot-level VNIR–SWIR hyperspectral reflectance dataset for potato under nitrogen fertilization and field-to-laboratory measurement conditions

This dataset contains plot-level visible–near-infrared to short-wave infrared (VNIR–SWIR) hyperspectral reflectance data collected from Russet Burbank potato under nitrogen fertilization treatments near Carberry, Manitoba, Canada, during the 2024 growing season. The dataset includes field canopy spectra, laboratory fresh-leaf spectra, laboratory dried-powder spectra, laboratory-measured petiole nitrate-N (PNN) reference values, corrected randomized treatment metadata, quality-control summaries, cleaned reflectance files with low-quality wavelength regions removed, and preprocessed spectra.

The experiment followed a randomized complete block design with 15 nitrogen treatments and four replicates, resulting in 60 plots. Spectra were collected using an ASD FieldSpec Pro spectroradiometer over the 350–2500 nm range at 1 nm intervals. Cleaned reflectance files exclude wavelength regions affected by sensor-edge noise and strong water-absorption features: 350–400, 1290–1500, 1790–2050, and 2350–2500 nm. The preprocessed files were generated from the cleaned reflectance data using Savitzky–Golay smoothing, standard normal variate correction, and first-derivative transformation.

This dataset is associated with the related research article, “VNIR–SWIR hyperspectral spectroscopy for nitrogen assessment in potato crops: Deep learning and regression models across field and laboratory conditions,” published in Smart Agricultural Technology.