Evaluating the consistency of leaf reflectance-based varietal discrimination in Brazilian vineyards

Abstract

The coexistence of Vitis vinifera and Vitis labrusca within the same regions makes Brazilian viticulture a unique system for varietal identification and origin certification. Non-destructive spectral methods are essential to support traceability within Denomination of Origin (DO) and Indication of Provenance (IP) frameworks. Contact spectroradiometry enables the detection of subtle reflectance differences linked to biochemical and structural leaf traits.

This study evaluated the consistency of spectral discrimination across datasets collected in December 2024 and February 2025 in vineyards of Mariana Pimentel, Rio Grande do Sul, Brazil. Reflectance spectra (350–2500 nm) were obtained using a spectroradiometer from adaxial and abaxial leaf surfaces of V. vinifera, V. labrusca, hybrid, and rootstock varieties. Spectral preprocessing, implemented in Python, included detector alignment, smoothing, and normalization.

Spectral ratios, first derivatives, and standard deviations were applied to identify discriminative wavelengths, refined through Forward Stepwise Linear Discriminant Analysis (LDA). Using the 2024 dataset, ten wavelengths achieved 93% classification accuracy. With 2025 dataset, the adaxial dataset yielded 94% accuracy with wavelengths selected by model and the abaxial dataset, 97% accuracy.

The most relevant bands were concentrated in the VNIR (400–700 nm) and SWIR1 (1350–1550 nm) regions. These results confirm the consistency of selected wavelengths across years and leaf surfaces, demonstrating the potential of contact spectroradiometry for varietal identification, vineyard monitoring, and certification of geographic origin in Brazilian viticulture.

References

Blackburn, G. A., 2006. Hyperspectral remote sensing of plant pigments. Journal of Experimental Botany, 58(4), 855-867. https://doi.org/10.1093/jxb/erl123.

Camargo, U. A., Tonietto, J., Hoffmann, A., 2011. Progressos na viticultura brasileira. Revista Brasileira de Fruticultura, 33(spe1).

Lacar, F., Lewis, M., Grierson, I., 2001. Use of hyperspectral reflectance for discrimination between grape varieties. IGARSS 2001. Scanning the Present and Resolving the Future. Proceedings. IEEE 2001 International Geoscience and Remote Sensing Symposium (Cat. No.01CH37217), 6, 2878–2880 vol.6.

Mazzia, V., Comba, L., Khaliq, A., Chiaberge, M., Gay, P., 2020. UAV and Machine Learning Based Refinement of a Satellite-Driven Vegetation Index for Precision Agriculture. Sensors, 20(9). https://www.mdpi.com/1424-8220/20/9/2530.

Pithan, P. A., Ducati, J. R., Garrido, L. R., Arruda, D. C., Thum, A. B., Hoff, R., 2021. Spectral characterization of fungal diseases downy mildew, powdery mildew, black-foot and Petri disease on Vitis vinifera leaves. International Journal of Remote Sensing, 42(15), 5680–5697.

Protas, J. F. d. S., Camargo, U. A., Mello, L. M. R. d., 2002. A viticultura brasileira: realidade e perspectivas. Simpósio Mineiro de Viticultura e Enologia, 1., 2002, Andradas, MG. Anais. Viticultura e Enologia: atualizando conceitos, Caldas: EPAMIG, 17–32. Acesso em: 16 set. 2025.

Ray, S., Jain, N., Miglani, A., Singh, J., Singh, A. K., Pan- igrahy, S., Parihar, J., 2010. Defining optimum spectral narrow bands and bandwidths for agricultural applications. Current science, 93, 1365-1369.

Schroeder, D. J., 2000. Astronomical Optics. 2nd edn, Academic Press, San Diego, CA, Beloit College.

Tsai, F., Philpot, W., 1998. Derivative Analysis of Hyperspectral Data. Remote Sensing of Environment, 66(1), 41-51. VIVC, 2025. Isabel — vitis international variety catalogue. https://www.vivc.de/index.php?r=passport/viewid=5560. Accessed: 2025-09-21.

Acknowledgments

This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001.