Trinity Canyon Vineyards - satellite monitoring

Trinity Canyon Vineyards - Satellite Remote Sensing Dataset =========================================================== Associated publication: Khlghatyan, A.; Bergamaschi, A.; Medvedev, A.; Muradyan, V.; Asmaryan, S.; Dell'Acqua, F. "Stable Seasonal Trends in Satellite-Derived Vegetation Indices Over Vineyards: Preliminary Results from Trinity Canyon, Armenia." Submitted to Applied Sciences, 2026. Pre-print available at: https://www.preprints.org/manuscript/202606.1319with DOI: 10.20944/preprints202606.1319.v1 =========================================================== STUDY AREA =========================================================== Trinity Canyon Vineyards, Vayots Dzor region, Armenia (approx. 4.4 ha, altitude 1290-1320 m). The vineyard contains six grape varieties planted in two row orientations: N-S and E-W =========================================================== FOLDER STRUCTURE =========================================================== S1_DPRVI/ S1_Incidence_Angle_Rasters/ S2_LAI/ Trinity_shp/ NOTE - METEOROLOGICAL DATA NOT INCLUDED: The daily air temperature records from the Areni meteorological station and the in-situ Thermochron TR-2L logger used to compute the Winkler Index (Growing Degree Days) are NOT distributed here because they are commercial/proprietary data. =========================================================== S1_DPRVI =========================================================== Dual-polarimetric Radar Vegetation Index (DpRVI) products derived from Sentinel-1 SLC data. Contents: - GeoTIFF raster images - one per acquisition date, one per relative orbit (72, 79, 152, 174). - Excel table with plot-level mean DpRVI values extracted for each grape variety, orbit, and date. Source data: 77 Sentinel-1 IW SLC scenes (VV+VH), downloaded from the Copernicus Browser. Acquisitions span four orbit configurations: Preprocessing (ESA SNAP v13.0): S-1 TOPS Split -> Apply Orbit File -> Calibration -> S-1 TOPS Deburst -> Polarimetric Matrix Generation (C2) -> Multilooking -> Polarimetric Speckle Filtering -> Range-Doppler Terrain Correction Index formula: DpRVI = q(q + 3) / (q + 1)^2, where q = lambda2 / lambda1 with lambda1, lambda2 the eigenvalues of the 2x2 covariance matrix C2. Values range from 0 (deterministic/bare surface) to 1 (dense, random vegetation canopy). Spatial extraction: All pixels within each plot boundary were spatially averaged to produce plot-level mean time series used for parabolic fitting analysis. =========================================================== S1_INCIDENCE_ANGLE_RASTERS =========================================================== Per-scene local incidence angle rasters co-registered with the DpRVI products. Contents: - GeoTIFF rasters of local incidence angle (degrees) - one per acquisition date and orbit. - Excel table summarising mean incidence angle per orbit and scene. =========================================================== S2_LAI =========================================================== Leaf Area Index (LAI) products derived from Sentinel-2 Level-2A imagery. Contents: - GeoTIFF raster images - one per acquisition date (31 scenes total, 18 were used from April to August 2024). Source data: Sentinel-2 L2A (atmospherically corrected surface reflectance, BOA) downloaded from the Copernicus Browser. Retrieval method: Custom evalscript deployed in Sentinel Hub applying a neural network algorithm using bands B03, B04, B05, B06, B07, B8A, B11, B12 alongside viewing and solar geometry angles to produce pixel-wise LAI estimates (m2/m2). Use in combined analysis: LAI acquisition dates were matched with the nearest DpRVI acquisition within a +/-2-day window to enable joint optical-radar analysis. Only temporally coincident pairs were retained. =========================================================== TRINITY_SHP =========================================================== Shapefiles defining the spatial framework of the study. Contents: - Parts.shp: Plot boundary polygons: delineate the six grape variety plots within the vineyard. Used for plot-level pixel extraction of DpRVI mean values. - Points_LAI: Point files for LAI-DpRVI comparison: pre-defined sampling points positioned within vine rows to minimise inter-row soil contamination. Used for pixel-wise extraction of co-located DpRVI and LAI values for the combined optical-radar scattergram analysis. =========================================================== LICENSE AND ATTRIBUTION =========================================================== The satellite data used to generate the products in this repository are derived from the Copernicus Programme of the European Union. S1_DPRVI (DpRVI derived from Sentinel-1): Contains modified Copernicus Sentinel data 2024. S1_Incidence_Angle_Rasters (incidence angle rasters from Sentinel-1): Contains modified Copernicus Sentinel data 2024. S2_LAI (LAI derived from Sentinel-2): Contains modified Copernicus Sentinel data 2024.