Articles
Modelling the effect of biostimulants on horticultural crop performance of poinsettia: an approach combining non-targeted plant analysis methods and data science
Article number
1417_10
Pages
79 – 92
Language
English
Abstract
Poinsettia (Euphorbia pulcherrima Willd.) is an important ornamental horticultural crop sold worldwide with the majority of sales in Europe in December.
The cultivation cycle of poinsettias starts in July and lasts about 15-20 weeks.
Consequently, the rooting phase takes place during the hot period, and this becomes more evident with climate change.
This study aims to assess whether the application of biostimulants promotes poinsettia growth during the summer season and whether such effects can be predicted early in the cultivation process.
Over a four-year cultivation period (2019-2022) we evaluated 10 different types of biostimulants, and we gathered phenotypic data and near infrared spectroscopy (NIRS) measurements during the growth phase (28 days after potting). Additionally, climate data were recorded.
Crop performance was modelled using machine learning tools (GLMNET) with the NIRS data, comprising over 1000 variables for a sample size of 200 plants.
The study revealed a pronounced year-to-year effect, which hides the effect of biostimulants.
The principal component analysis highlights the role of summer relative humidity on better root development of young plants.
Biostimulants based on amino acids exhibited a slight advantage over other types, such as seaweed extracts and microorganisms.
A generalized linear model with penalization was used to predict plant performance, expressed in commercial rating.
The data set was divided, with 75% used for algorithm training and 25% for testing the resulting model.
This predictive approach showed that plant performance could be predicted from rooting variables.
Some biostimulants consistently promoted plant growth at the end of summer over the four years of experiments but other biostimulants exhibited varying effects each year.
The potential impact of biostimulants on early-stage cultivation could be assessed using NIRS data and machine learning tools.
The cultivation cycle of poinsettias starts in July and lasts about 15-20 weeks.
Consequently, the rooting phase takes place during the hot period, and this becomes more evident with climate change.
This study aims to assess whether the application of biostimulants promotes poinsettia growth during the summer season and whether such effects can be predicted early in the cultivation process.
Over a four-year cultivation period (2019-2022) we evaluated 10 different types of biostimulants, and we gathered phenotypic data and near infrared spectroscopy (NIRS) measurements during the growth phase (28 days after potting). Additionally, climate data were recorded.
Crop performance was modelled using machine learning tools (GLMNET) with the NIRS data, comprising over 1000 variables for a sample size of 200 plants.
The study revealed a pronounced year-to-year effect, which hides the effect of biostimulants.
The principal component analysis highlights the role of summer relative humidity on better root development of young plants.
Biostimulants based on amino acids exhibited a slight advantage over other types, such as seaweed extracts and microorganisms.
A generalized linear model with penalization was used to predict plant performance, expressed in commercial rating.
The data set was divided, with 75% used for algorithm training and 25% for testing the resulting model.
This predictive approach showed that plant performance could be predicted from rooting variables.
Some biostimulants consistently promoted plant growth at the end of summer over the four years of experiments but other biostimulants exhibited varying effects each year.
The potential impact of biostimulants on early-stage cultivation could be assessed using NIRS data and machine learning tools.
Authors
N. Guibert, S. Prigent, P. Pétriacq, J.M. Deogratias, C. Cabasson
Keywords
Euphorbia pulcherrima Willd., machine learning, NIR spectroscopy, plant growth-promoting
Groups involved
- Division Vine and Berry Fruits
- Division Plant-Environment Interactions in Field Systems
- Division Horticulture for Human Health
- Working Group Vaccinium Species and Management
- Working Group Rubus and Ribes Species and Management
- Working Group Kiwifruit Culture and Management
- Working Group Strawberry Culture and Management
Online Articles (35)