Garlic (Allium sativum) feature-specific nutrient dosage based on using machine learning models
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , , , , , , , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1371/journal.pone.0268516 http://hdl.handle.net/11449/240992 |
Resumo: | Brazil presents large yield gaps in garlic crops partly due to nutrient mismanagement at local scale. Machine learning (ML) provides powerful tools to handle numerous combinations of yield-impacting factors that help reducing the number of assumptions about nutrient management. The aim of the current study is to customize fertilizer recommendations to reach high garlic marketable yield at local scale in a pilot study. Thus, collected 15 nitrogen (N), 24 phosphorus (P), and 27 potassium (K) field experiments conducted during the 2015 to 2017 period in Santa Catarina state, Brazil. In addition, 61 growers’ observational data were collected in the same region in 2018 and 2019. The data set was split into 979 experimental and observational data for model calibration and into 45 experimental data (2016) to test ML models and compare the results to state recommendations. Random Forest (RF) was the most accurate ML to predict marketable yield after cropping system (cultivar, preceding crops), climatic indices, soil test and fertilization were included features as predictor (R2 = 0.886). Random Forest remained the most accurate ML model (R2 = 0.882) after excluding cultivar and climatic features from the prediction-making process. The model suggested the application of 200 kg N ha-1 to reach maximum marketable yield in a test site in comparison to the 300 kg N ha-1 set as state recommendation. P and K fertilization also seemed to be excessive, and it highlights the great potential to reduce production costs and environmental footprint without agronomic loss. Garlic root colonization by arbuscular mycorrhizal fungi likely contributed to P and K uptake. Well-documented data sets and machine learning models could support technology transfer, reduce costs with fertilizers and yield gaps, and sustain the Brazilian garlic production. |
id |
UNSP_dcdd3be9b12a72c36563cfdff6bba9e6 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/240992 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Garlic (Allium sativum) feature-specific nutrient dosage based on using machine learning modelsBrazil presents large yield gaps in garlic crops partly due to nutrient mismanagement at local scale. Machine learning (ML) provides powerful tools to handle numerous combinations of yield-impacting factors that help reducing the number of assumptions about nutrient management. The aim of the current study is to customize fertilizer recommendations to reach high garlic marketable yield at local scale in a pilot study. Thus, collected 15 nitrogen (N), 24 phosphorus (P), and 27 potassium (K) field experiments conducted during the 2015 to 2017 period in Santa Catarina state, Brazil. In addition, 61 growers’ observational data were collected in the same region in 2018 and 2019. The data set was split into 979 experimental and observational data for model calibration and into 45 experimental data (2016) to test ML models and compare the results to state recommendations. Random Forest (RF) was the most accurate ML to predict marketable yield after cropping system (cultivar, preceding crops), climatic indices, soil test and fertilization were included features as predictor (R2 = 0.886). Random Forest remained the most accurate ML model (R2 = 0.882) after excluding cultivar and climatic features from the prediction-making process. The model suggested the application of 200 kg N ha-1 to reach maximum marketable yield in a test site in comparison to the 300 kg N ha-1 set as state recommendation. P and K fertilization also seemed to be excessive, and it highlights the great potential to reduce production costs and environmental footprint without agronomic loss. Garlic root colonization by arbuscular mycorrhizal fungi likely contributed to P and K uptake. Well-documented data sets and machine learning models could support technology transfer, reduce costs with fertilizers and yield gaps, and sustain the Brazilian garlic production.Natural Sciences and Engineering Research Council of CanadaCaçador Experimental Station Agricultural Research and Rural Extension Agency of Santa Catarina (Epagri), CaçadorDepartment of Soils and Agrifood Engineering Laval UniversityDepartment of Soil Federal University of Santa Maria, Rio Grande do SulExecutive Secretariat for the Environment Government Administrative Center, Santa CatarinaState University of Paulista “Julio Mesquita Filho” Campus RegistroUniversity of Alto Vale do Rio do Peixe, CaçadorState University of Paulista “Julio Mesquita Filho” Campus RegistroNatural Sciences and Engineering Research Council of Canada: NSERC-2254Agricultural Research and Rural Extension Agency of Santa Catarina (Epagri)Laval UniversityFederal University of Santa MariaGovernment Administrative CenterUniversidade Estadual Paulista (UNESP)University of Alto Vale do Rio do PeixeHahn, LeandroParent, Léon-ÉtiennePaviani, Angela CristinaFeltrim, Anderson LuizWamser, Anderson FernandoRozane, Danilo Eduardo [UNESP]Ender, Marcos MatosGrando, Douglas LuizMoura-Bueno, Jean MichelBrunetto, Gustavo2023-03-01T20:42:12Z2023-03-01T20:42:12Z2022-05-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlehttp://dx.doi.org/10.1371/journal.pone.0268516PLoS ONE, v. 17, n. 5 May, 2022.1932-6203http://hdl.handle.net/11449/24099210.1371/journal.pone.02685162-s2.0-85130100101Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengPLoS ONEinfo:eu-repo/semantics/openAccess2024-05-03T13:20:53Zoai:repositorio.unesp.br:11449/240992Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-05-03T13:20:53Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Garlic (Allium sativum) feature-specific nutrient dosage based on using machine learning models |
title |
Garlic (Allium sativum) feature-specific nutrient dosage based on using machine learning models |
spellingShingle |
Garlic (Allium sativum) feature-specific nutrient dosage based on using machine learning models Hahn, Leandro |
title_short |
Garlic (Allium sativum) feature-specific nutrient dosage based on using machine learning models |
title_full |
Garlic (Allium sativum) feature-specific nutrient dosage based on using machine learning models |
title_fullStr |
Garlic (Allium sativum) feature-specific nutrient dosage based on using machine learning models |
title_full_unstemmed |
Garlic (Allium sativum) feature-specific nutrient dosage based on using machine learning models |
title_sort |
Garlic (Allium sativum) feature-specific nutrient dosage based on using machine learning models |
author |
Hahn, Leandro |
author_facet |
Hahn, Leandro Parent, Léon-Étienne Paviani, Angela Cristina Feltrim, Anderson Luiz Wamser, Anderson Fernando Rozane, Danilo Eduardo [UNESP] Ender, Marcos Matos Grando, Douglas Luiz Moura-Bueno, Jean Michel Brunetto, Gustavo |
author_role |
author |
author2 |
Parent, Léon-Étienne Paviani, Angela Cristina Feltrim, Anderson Luiz Wamser, Anderson Fernando Rozane, Danilo Eduardo [UNESP] Ender, Marcos Matos Grando, Douglas Luiz Moura-Bueno, Jean Michel Brunetto, Gustavo |
author2_role |
author author author author author author author author author |
dc.contributor.none.fl_str_mv |
Agricultural Research and Rural Extension Agency of Santa Catarina (Epagri) Laval University Federal University of Santa Maria Government Administrative Center Universidade Estadual Paulista (UNESP) University of Alto Vale do Rio do Peixe |
dc.contributor.author.fl_str_mv |
Hahn, Leandro Parent, Léon-Étienne Paviani, Angela Cristina Feltrim, Anderson Luiz Wamser, Anderson Fernando Rozane, Danilo Eduardo [UNESP] Ender, Marcos Matos Grando, Douglas Luiz Moura-Bueno, Jean Michel Brunetto, Gustavo |
description |
Brazil presents large yield gaps in garlic crops partly due to nutrient mismanagement at local scale. Machine learning (ML) provides powerful tools to handle numerous combinations of yield-impacting factors that help reducing the number of assumptions about nutrient management. The aim of the current study is to customize fertilizer recommendations to reach high garlic marketable yield at local scale in a pilot study. Thus, collected 15 nitrogen (N), 24 phosphorus (P), and 27 potassium (K) field experiments conducted during the 2015 to 2017 period in Santa Catarina state, Brazil. In addition, 61 growers’ observational data were collected in the same region in 2018 and 2019. The data set was split into 979 experimental and observational data for model calibration and into 45 experimental data (2016) to test ML models and compare the results to state recommendations. Random Forest (RF) was the most accurate ML to predict marketable yield after cropping system (cultivar, preceding crops), climatic indices, soil test and fertilization were included features as predictor (R2 = 0.886). Random Forest remained the most accurate ML model (R2 = 0.882) after excluding cultivar and climatic features from the prediction-making process. The model suggested the application of 200 kg N ha-1 to reach maximum marketable yield in a test site in comparison to the 300 kg N ha-1 set as state recommendation. P and K fertilization also seemed to be excessive, and it highlights the great potential to reduce production costs and environmental footprint without agronomic loss. Garlic root colonization by arbuscular mycorrhizal fungi likely contributed to P and K uptake. Well-documented data sets and machine learning models could support technology transfer, reduce costs with fertilizers and yield gaps, and sustain the Brazilian garlic production. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-05-01 2023-03-01T20:42:12Z 2023-03-01T20:42:12Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1371/journal.pone.0268516 PLoS ONE, v. 17, n. 5 May, 2022. 1932-6203 http://hdl.handle.net/11449/240992 10.1371/journal.pone.0268516 2-s2.0-85130100101 |
url |
http://dx.doi.org/10.1371/journal.pone.0268516 http://hdl.handle.net/11449/240992 |
identifier_str_mv |
PLoS ONE, v. 17, n. 5 May, 2022. 1932-6203 10.1371/journal.pone.0268516 2-s2.0-85130100101 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
PLoS ONE |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1799965739379064832 |