GRASS COVER, TREE DENSITY, AND LEAF DEVELOPMENT OF MEDITERRANEAN ORCHARDS FROM HIGH RESOLUTION DATA

The study focused on Mediterranean orchards and aimed to explore different remote sensing data (Sentinel 2 data (2016–2023), 1 Pleiades image (2022) and the extraction of Google-satellite-hybrid images (GSH,2017)) to compute key variables affecting water requirements such as tree age and density per...

Full description

Saved in:
Bibliographic Details
Published inInternational archives of the photogrammetry, remote sensing and spatial information sciences. Vol. XLVIII-1/W2-2023; pp. 1531 - 1536
Main Authors Rouault, P., Courault, D., Pouget, G., Flamain, F., Lopez-Lozano, R., Doussan, C., Debolini, M., McCabe, M.
Format Journal Article Conference Proceeding
LanguageEnglish
Published Gottingen Copernicus GmbH 01.01.2023
Copernicus Publications
Subjects
Agricultural management
Agricultural practices
Agriculture
Archives & records
Classification
Crops
Deep learning
Density
Farmers
Flowering
Fruits
Irrigation
Leaves
Orchards
Phenology
Photogrammetry
Remote sensing
Satellite imagery
Satellites
Spectral bands
Teaching methods
Trees
Vegetation
Water requirements
Watersheds
Online AccessGet full text

Cover

More Information
Summary:The study focused on Mediterranean orchards and aimed to explore different remote sensing data (Sentinel 2 data (2016–2023), 1 Pleiades image (2022) and the extraction of Google-satellite-hybrid images (GSH,2017)) to compute key variables affecting water requirements such as tree age and density per plot, leaf development, the inter-row management. Surveys were conducted on 22 farms where accurate information on agricultural practices was collected. The results have shown that a thresholding on the NDVI Sentinel 2 in the summer period allowed the identification of young orchards with an accuracy of 98%. The analysis of temporal profiles of FAPAR allowed the identification of key phenological stages such as flowering and fruit set. Supervised classification was employed to separate grassed and non-grassed plots using three spectral bands of Sentinel 2. Classifications performed from GSH images gave more accurate results (81% well classified) compared with Sentinel 2 (79%) and Pleiades (57%) when identifying grassed plots. The methods presented in this study propose methods easily accessible based on free-to-download data, making them applicable in diverse orchard contexts.
ISSN:2194-9034
1682-1750
2194-9034
DOI:10.5194/isprs-archives-XLVIII-1-W2-2023-1531-2023