Hyperspectral Reflectance Dataset (350–2500 nm) of Wheat Flag Leaves under Four Levels of Cereal Leaf Beetle Damage

This dataset contains hyperspectral reflectance measurements (350–2500 nm) of winter wheat (Triticum aestivum L., cv. Bc Anica) flag leaves affected by varying levels of damage caused by the cereal leaf beetle (Oulema melanopus L., Coleoptera: Chrysomelidae). The data were collected at the experimental fields of the Bc Institute for Breeding and Production of Field Crops in Zagreb, Croatia (45°44'49.1"N, 15°56'13.4"E) during the 2021/2022 growing season.

Spectral measurements were obtained using a Spectral Evolution® SR-2500 full-range spectroradiometer (350–2500 nm) equipped with a contact probe and internal halogen light source. Each spectrum represents an average of three readings taken on the adaxial surface of individual flag leaves, with calibration performed every ten scans using a BaSO₄ white reference panel.

Leaves were visually classified into four categories of cereal leaf beetle (CLB) damage:

1. Healthy (0% tissue loss)

2. Slightly damaged (10–15% tissue loss)

3. Moderately damaged (15–30% tissue loss)

4. Severely damaged (30–60% tissue loss)

The final dataset comprises 210 averaged reflectance spectra, distributed across the four damage categories (52, 52, 46, and 60 samples, respectively). Each record includes reflectance values for 2,151 spectral bands (1 nm interval) and a categorical damage label (1–4).

This dataset supports research on the early detection and classification of pest-induced stress in cereal crops using hyperspectral sensing and machine learning, contributing to precision pest management and sustainable agricultural practices.