Artificial inoculation with fungal species involved in the Petri disease was performed on in vitr... more Artificial inoculation with fungal species involved in the Petri disease was performed on in vitro grapevine shoots. After 2 months, plantlets were observed and the presence of any leaf chlorosis or necrosis, considered as symptoms of the disease, were recorded. The fungal inoculation caused an increase in the number of symptomatic in vitro plants and in the incidence of external
ABSTRACT Since 1740 until now, the Book of Vine Branches encompasses beautiful pictorial represen... more ABSTRACT Since 1740 until now, the Book of Vine Branches encompasses beautiful pictorial representations of shoots and sprouts, collected every year on April 24th in the vineyards of the Hungarian town of K ő szeg. A long time series (1740 – 2009) of sprouts phenological devel- opment coded with the BBCH scale is obtained through paintings visual inspection. Consequently, a model relating sprouts development with thermal resources accumulation is developed on the base of Wien daily air temperature series for the 1857 – 2009 period. Thermal resources is obtained with the Normal Heat Hours approach, a response curve method that takes into account the nonlinear response of phenological development to temperature. The model, calibrated and validated on the 1930 – 2009 period, shows an overestimation of thermal resources when applied to the 1857 – 1929, due to the more advanced phenology of that phase. Since this advance can be ascribed to genetics and management instead of more favorable thermal conditions, model outputs previous to 1930 are decreased by a constant factor. Finally, the 1740 – 2009 time series of grapevine spring thermal resources is obtained applying the model to the complete phenological time series. The analysis of the series highlights a first phase (1740 – 1780) characterized by high and stable availability of thermal resources during the 1740 – 1780, a second phase (1781 – 1820) with the lowest level of resources, a phase of intermediate availability (1821 – 1929) and a final present phase (1930 – 2009) with renewed high availability. The adopted method overcomes the lack of correlation for the post 1960 period shown by previous works.
Flavonoids are a class of bioactive compounds largely represented in grapevine and wine.
They als... more Flavonoids are a class of bioactive compounds largely represented in grapevine and wine. They also affect the sensory quality of fruits and vegetables, and derived products. Methods available for flavonoid measurement are time-consuming, thus a rapid and cost-effective determination of these compounds is an important research objective. This work tests if applying machine learning techniques to texture analysis data allows to reach good performances for flavonoid estimation in grape berries. Whole berry and skin texture analysis was applied to berries from 22 red wine grape cultivars and linked to the total flavonoid content. Three machine-learning techniques (regression tree, random forest and gradient boosting machine) were then applied. Models reached a high accuracy both in the external and internal validation (R² ranging from 0.75 to 0.85 for the external validation and from 0.65 to 0.75 for the internal validation, while RMSE went from 0.95 mg g-1 to 0.7 mg g-1 in the external validation and from 1.3 mg g-1 to 1.1 mg g-1 in the internal validation).
Artificial inoculation with fungal species involved in the Petri disease was performed on in vitr... more Artificial inoculation with fungal species involved in the Petri disease was performed on in vitro grapevine shoots. After 2 months, plantlets were observed and the presence of any leaf chlorosis or necrosis, considered as symptoms of the disease, were recorded. The fungal inoculation caused an increase in the number of symptomatic in vitro plants and in the incidence of external
ABSTRACT Since 1740 until now, the Book of Vine Branches encompasses beautiful pictorial represen... more ABSTRACT Since 1740 until now, the Book of Vine Branches encompasses beautiful pictorial representations of shoots and sprouts, collected every year on April 24th in the vineyards of the Hungarian town of K ő szeg. A long time series (1740 – 2009) of sprouts phenological devel- opment coded with the BBCH scale is obtained through paintings visual inspection. Consequently, a model relating sprouts development with thermal resources accumulation is developed on the base of Wien daily air temperature series for the 1857 – 2009 period. Thermal resources is obtained with the Normal Heat Hours approach, a response curve method that takes into account the nonlinear response of phenological development to temperature. The model, calibrated and validated on the 1930 – 2009 period, shows an overestimation of thermal resources when applied to the 1857 – 1929, due to the more advanced phenology of that phase. Since this advance can be ascribed to genetics and management instead of more favorable thermal conditions, model outputs previous to 1930 are decreased by a constant factor. Finally, the 1740 – 2009 time series of grapevine spring thermal resources is obtained applying the model to the complete phenological time series. The analysis of the series highlights a first phase (1740 – 1780) characterized by high and stable availability of thermal resources during the 1740 – 1780, a second phase (1781 – 1820) with the lowest level of resources, a phase of intermediate availability (1821 – 1929) and a final present phase (1930 – 2009) with renewed high availability. The adopted method overcomes the lack of correlation for the post 1960 period shown by previous works.
Flavonoids are a class of bioactive compounds largely represented in grapevine and wine.
They als... more Flavonoids are a class of bioactive compounds largely represented in grapevine and wine. They also affect the sensory quality of fruits and vegetables, and derived products. Methods available for flavonoid measurement are time-consuming, thus a rapid and cost-effective determination of these compounds is an important research objective. This work tests if applying machine learning techniques to texture analysis data allows to reach good performances for flavonoid estimation in grape berries. Whole berry and skin texture analysis was applied to berries from 22 red wine grape cultivars and linked to the total flavonoid content. Three machine-learning techniques (regression tree, random forest and gradient boosting machine) were then applied. Models reached a high accuracy both in the external and internal validation (R² ranging from 0.75 to 0.85 for the external validation and from 0.65 to 0.75 for the internal validation, while RMSE went from 0.95 mg g-1 to 0.7 mg g-1 in the external validation and from 1.3 mg g-1 to 1.1 mg g-1 in the internal validation).
Uploads
Papers by Lorenzo Lovat
They also affect the sensory quality of fruits and vegetables, and derived products. Methods available
for flavonoid measurement are time-consuming, thus a rapid and cost-effective determination of these
compounds is an important research objective. This work tests if applying machine learning
techniques to texture analysis data allows to reach good performances for flavonoid estimation in
grape berries.
Whole berry and skin texture analysis was applied to berries from 22 red wine grape cultivars and
linked to the total flavonoid content. Three machine-learning techniques (regression tree, random
forest and gradient boosting machine) were then applied. Models reached a high accuracy both in the
external and internal validation (R² ranging from 0.75 to 0.85 for the external validation and from 0.65
to 0.75 for the internal validation, while RMSE went from 0.95 mg g-1 to 0.7 mg g-1 in the external
validation and from 1.3 mg g-1 to 1.1 mg g-1 in the internal validation).
They also affect the sensory quality of fruits and vegetables, and derived products. Methods available
for flavonoid measurement are time-consuming, thus a rapid and cost-effective determination of these
compounds is an important research objective. This work tests if applying machine learning
techniques to texture analysis data allows to reach good performances for flavonoid estimation in
grape berries.
Whole berry and skin texture analysis was applied to berries from 22 red wine grape cultivars and
linked to the total flavonoid content. Three machine-learning techniques (regression tree, random
forest and gradient boosting machine) were then applied. Models reached a high accuracy both in the
external and internal validation (R² ranging from 0.75 to 0.85 for the external validation and from 0.65
to 0.75 for the internal validation, while RMSE went from 0.95 mg g-1 to 0.7 mg g-1 in the external
validation and from 1.3 mg g-1 to 1.1 mg g-1 in the internal validation).