EventsThe 6th International Electronic Conference on Remote Sensing
Published
This submission belongs to the session S3. Remote Sensing in Semi-Arid and Arid Areas of the event The 6th International Electronic Conference on Remote Sensing
Published date
14 Oct, 2026
Academic Editor
author-avatarTimo Balz
Citation
Eleftheria Alexoudaki, Georgios K. Vasios, A smart farming multi-scale spatial framework for monitoring viticultural resilience under climate change in Lemnos Island, Greece, in Proceedings of The 6th International Electronic Conference on Remote Sensing, 19 October–21 October 2026, MDPI: Basel, Switzerland
Share
Email
Facebook
Twitter
LinkedIn

A smart farming multi-scale spatial framework for monitoring viticultural resilience under climate change in Lemnos Island, Greece

image
1. Laboratory of Sustainable Agrifood and Smart Farming, Department of Food Science and Nutrition, School of Environment, University of the Aegean, 81400, Myrina, Lemnos, Greece.
Abstract

Traditional Mediterranean viticultural systems are increasingly vulnerable to climate-change-induced aridification. To empower local stakeholders, this study introduces an integrated Smart Farming framework deployed on Lemnos Island, Greece, combining cloud-based Google Earth Engine (GEE) dashboards with statistical scripting in R. To resolve canopy-scale omission errors (53.6%) caused by traditional low-density vineyards, regional Random Forest classifications were spatially overlaid with official 2025 administrative parcel registries. The multi-scale architecture operates via a unified, single-click parcel query executing across three interconnected modules: a Scientific module extracting 40-year climate-vegetation datasets (1986–2025) to track biomass versus surface temperature (NDVI vs. LST) and ecosystem deficits versus leaf moisture (dEVI vs. NDWI); a Farmer module integrating on-demand Sentinel-2 soil-adjusted indices (SAVI, NDWI) with live "Hephestos" weather station telemetry; and an Academic Validation module cross-referencing satellite precipitation against National Observatory of Athens database. Results demonstrated robust annual precipitation alignment between CHIRPS and ground station records (R² = 0.51; annual variance 2.6%; absolute deviation 11.2 mm). Long-term time-series analysis via Bai–Perron structural change tests confirmed a significant shift breakpoint (in 2000) accompanied by decadal vegetative curve advances across cultivars. At the parcel scale, coupling satellite surface thermal thresholds (LST > 32°C) and canopy moisture deficits (NDWI < 0.02) with live weather station data (precipitation = 0 mm) reliably triggered automated, cultivar-specific water and heat stress diagnostics on the user console. This hybrid framework proves that integrating administrative boundaries with multi-temporal remote sensing, time-series statistics, and live ground telemetry delivers actionable climate resilience metrics for arid viticulture.

Keywords
Smart Farming
Google Earth Engine
Sentinel
Decision-Support Dashboards
Random Forest Classification
Vineyards
Spatiotemporal Maize Growth Monitoring and Yield Estimation Using Remote Sensing: A Case Study of Nyagatare District, Rwanda
Remote Sensing-Driven Interpretable Random Forest to Predict Maize End-of-Season WRSI Drought in Kirehe District, Rwanda (2001–2023)