terclim by ICS banner
IVES 9 IVES Conference Series 9 GiESCO 9 Using the fraction of transpirable soil water to estimate grapevine leaf water potential: comparing the classical statistical regression approach to machine learning algorithms

Using the fraction of transpirable soil water to estimate grapevine leaf water potential: comparing the classical statistical regression approach to machine learning algorithms


Context and purpose of the study – Weather uncertainty is forcing Mediterranean winegrowers to adopt new irrigation strategies to cope with water scarcity while ensuring a sustainable yield and improved berry and wine quality standards. Therefore, more accurateand high-resolution monitoring of soil water content and vine water status is a major concern. Leaf water potential measured at pre-dawn (YPD) is considered to be in equilibrium with soil water potential and is highly correlated with soil water content at the soil depth where roots extract water.
The aim of this study is to evaluate a dataset of eco-physiological data collected in a 3-year vineyard irrigation trial to assess the explanatory power of the fraction of transpirable soil water (FTSW) to predict YPD by comparing the classical statistical regression approach with a machine learning algorithm (MLA).

Material and methods – Deficit irrigation trials were conducted from 2013 to 2015 in a commercial vineyard in the Alentejo (southern Portugal). Trial plot was planted with Vitis vinifera (L.) cv. Aragonez (ARA)(syn. Tempranillo), grafted onto 1103 Paulsen rootstock and spaced 1.5 m within and 3.0 m between N-S oriented rows. The experimental layout was a randomized complete block design with two treatments: sustained deficit irrigation (SDI – control; ~30% Etc) and regulated deficit irrigation (RDI; ~15% Etc) and 4 replicates per treatment. The YPD and soil water content were measured the day before and the day after each irrigation event by using a capacitance probe down to a soil depth of 1 m and a Scholander pressure chamber. Models predicting YPD from FTSW were trained on 600 data cases and validated on an independent dataset (10% of all available data) using MATLAB R2022b (Mathworks, USA) and STATISTICA 13 (Tibco, USA). 

Results – Our results show that 87.6% of the observed YPD variability is explained by the FTSW using a linear regression model (LRM) with a linear-logarithmic transformation of the independent variables. The accuracy of the prediction model, as measured by root mean squared error (RMSE), in the independent validation dataset, was 0.08 MPa. These results were compared to the estimation accuracy of a set of MLAs. Two support vector machine (SVM) algorithms with a quadratic and a medium Gaussian kernel function, and three Gaussian process regression (GPR) algorithms with an exponential, a squared exponential and a rational quadratic kernel functions were tested. All trained MLAs showed an accuracy in explaining the variability of the YPD (86-87%) similar to the LRM. An increase in the model explained variability of the independent dataset from 89 to 91% was observed in all MLAs, with an accuracy of 0.087 to 0.096MPa as measured by the RMSE.
Both statistical methods indicate that YPD can be estimated with good accuracy using FTSW as an explanatory variable. Regarding the comparative performance of the two types of statistical models no differences were found in the ability of the tested models to estimate YPD.


Publication date: June 29, 2023

Issue: GiESCO 2023

Type: Poster


Ricardo EGIPTO1*, J. Miguel COSTA2, José SILVESTRE1, Carlos M. LOPES2

1INIAV IP – Polo de Inovação de Dois Portos, 2565-191 Dois Portos, Portugal
2LEAF – Linking Landscape, Environment, Agriculture and Food, Instituto Superior de Agronomia, Univ. Lisboa, Portugal

Contact the author*


deficit irrigation, soil water content, machine learning algorithms


GiESCO | GIESCO 2023 | IVES Conference Series


Related articles…


Climate change has brought several impacts that are becoming increasingly intense during the last few years and put at risk the quality of the berries or even the plant’s sustainability. Such extreme climatic events impact the composition of the wine while modulating its quality and the consumer preferences (Tempère et al., 2019). The three most important changes that take place in the must are: 1) decrease acidity, 2) increase of the concentration of sugar, hence increase of alcohol in the wine, and 3) modification
of the sensory balance and the development for example of cooked fruit aromas.


In Italy, in the past two decades, the rate of temperature increases (0.0369 °C per year) was slightly higher compared to the world average (0.0313 °C per year). It has also been indicated that the number and intensity of heat waves have increased considerably in the last decades. (IEA, 2022). Viticultural zones can be classified with climatic indexes. Huglin’s index (HI) considers the temperature in a definite area and has been considered as reliable to evaluate the thermal suitability for winegrape production (Zhang et al., 2023).

A synthesis approach on the impact of elevated CO2 on berry physiology and yield of Vitis vinifera

Besides the increase in global mean temperature the second main challenge of a changing climate is the increase in atmospheric carbon dioxide (CO2) in relation to physiology and yield performance of grapevines. The benefits of increasing CO2 levels under greenhouse environment or open field studies have been well investigated for various annual crops. Research under free carbon dioxide enrichment on field-grown perennial plants such as grapevines is limited to a few studies. Further, chamber and greenhouse experiments have been conducted mostly on potted vines under eCO2 conditions.

Microbial ecosystems in wineries – molecular interactions between species and modelling of population dynamics

Microbial ecosystems are primary drivers of viticultural, oenological and other cellar-related processes
such as wastewater treatment. Metagenomic datasets have broadly mapped the vast microbial species
diversity of many of the relevant ecological niches within the broader wine environment, from vineyard
soils to plants and grapes to fermentation. The data highlight that species identities and diversity
significantly impact agronomic performance of vineyards as well as wine quality, but the complexity
of these systems and of microbial growth dynamics has defeated attempts to offer actionable
tools to guide or predict specific outcomes of ecosystem-based interventions.

Searching for the sweet spot: a focus on wine dealcoholization

It is well known that the vinification of grapes at full maturation can produce rich, full-bodied wines,
with intense and complex flavour profiles. However, the juice obtained from such grapes may have very
high sugar concentration, resulting in wines with an excessive concentration of ethanol. In addition, the decoupling between technological maturity and phenolic/aromatic one due to global warming, exacerbates this problem in some wine-growing regions. In parallel with the increase of the mean alcohol content of wines on the market, also the demand for reduced alcohol beverages has increased in recent years, mainly as a result of health and social concerns about the risks related to the consumption of alcohol.