Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Recurso digital |
| Sprache: | Englisch |
| Veröffentlicht: |
Zenodo
2025
|
| Schlagworte: | |
| Online-Zugang: | https://doi.org/10.5281/zenodo.19455964 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Inhaltsangabe:
- <p>This thesis aims to advance vineyard precision irrigation by improving methods for evaluating grapevine water status, which is critical for optimizing grapevine water stress and ensuring sustainability in viticulture. Accurate monitoring of vineyard water use, with high spatial and temporal resolution, is essential for enhancing grape productivity and quality. A significant challenge in this field is the absence of universally applicable methods to assess vine water status across diverse conditions. The first key objective of this research was to develop machine learning algorithms (MLAs) for accurately estimating actual crop evapotranspiration (ETc), a key indicator of vine water status. The study compared eight state-of-the-art supervised MLAs using atmospheric variables such as net radiation (Rn), wind speed (u), and vapor pressure deficit (VPD), along with plant variables like stomatal conductance to water vapor (gsw). These algorithms were evaluated against the FAO-56 Kc-ET0 method and actual ETc measurements from an eddy covariance system. Results showed that the top five algorithms explained over 89% of the variability in measured ETc, achieving an average accuracy 2.5 times higher than the FAO-56 Kc-ET0. The second major milestone was the development of a biophysical model to predict grapevine canopy conductance (gc) to water vapor, which is highly sensitive to water stress. The model combined grapevine gsw and atmospheric demand (Rn and VPD) to estimate gc, providing a reliable method for assessing grapevine transpiration under arid conditions. This model explained over 91% of the variability in gc as measured by the Penman-Monteith method, demonstrating its effectiveness in simulating the grapevine’s response to water scarcity and environmental stress. A third milestone involved a comprehensive review of the effects of soil temperature (ST) on vineyard performance in Mediterranean climates, examining its impact on ecophysiological processes and agronomic outcomes. The review also highlighted the potential of imaging techniques to assess ST and vertical canopy temperature profiles, providing valuable insights for precision irrigation practices. Building on this review, the fourth milestone developed a modeling framework to estimate energy flux dynamics at the soil and plant levels under arid conditions. This framework focused on evaluating sensible and latent energy fluxes within the vineyard during grapevine berry maturation. The approach, tested against measurements from an eddy covariance system, showed that aggregated fluxes estimated using a three-source clumped model explained 98% of the variability in turbulent fluxes, with a consistent error margin of less than 3 W.m-2 for sensible heat fluxes. Although latent heat fluxes were slightly underestimated, this framework offers an improved understanding of energy dynamics in vineyards under regulated deficit irrigation. Overall, the results of this research demonstrate the effectiveness of non-destructive remote sensing techniques for assessing grapevine water status. The developed models and methods—ranging from estimating actual crop evapotranspiration to evaluating canopy conductance and energy flux balances—provide valuable tools for vineyard management. These tools can facilitate the adoption of deficit irrigation strategies, improving water use efficiency and supporting sustainable viticulture practices.</p>