Nutrient Diagnosis of Eucalyptus at the Factor-Specific Level Using Machine Learning and Compositional Methods

Paula, Betânia Vahl de; Arruda, Wagner Squizani; Parent, Léon E.; Araújo, Elias Frank de; Brunetto, Gustavo

doi:10.3390/plants9081049

Cited by 20 publications

(21 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The size and diversity of the dataset and number of features selected in the model should be commensurate to avoid overfitting. The GPR, ANN and Random Forest were found to be useful ML models for small-size yet complex agronomic datasets [18,21,22,43]. Random Forests may return staircase response curves, while GPR and ANN may return smooth response curves.…”

Section: Machine Learningmentioning

confidence: 98%

“…On the other hand, proportions show Dirichlet distribution, and are thus multiplicative (products). While clr and ilr variables can be used as features in ML models, the accuracy of ML models appears to be little influenced by nutrient expressions [43]. Using raw concentrations as features may be preferable, because log ratio computation does not permit missing values, unless imputed or replaced by 0.65 times the detection limit [47], if the number of missing values is relatively small in the dataset.…”

Section: Compositional Data Analysis (Coda)mentioning

confidence: 99%

See 1 more Smart Citation

Nutrient Diagnosis of Fertigated “Prata” and “Cavendish” Banana (Musa spp.) at Plot-Scale

Neto

Deus

Filho³

et al. 2020

Plants

Self Cite

View full text Add to dashboard Cite

Fertigation management of banana plantations at a plot scale is expanding rapidly in Brazil. To guide nutrient management at such a small scale, genetic, environmental and managerial features should be well understood. Machine learning and compositional data analysis (CoDa) methods can measure the effects of feature combinations on banana yield and rank nutrients in the order of their limitation. Our objectives are to review ML and CoDa models for application at regional and local scales, and to customize nutrient diagnoses of fertigated banana at the plot scale. We documented 940 “Prata” and “Cavendish” plot units for tissue and soil tests, environmental and managerial features, and fruit yield. A Neural Network informed by soil tests, tissue tests and other features was the most proficient learner (AUC up to 0.827). Tissue nutrients were shown to have the greatest impact on model accuracy. Regional nutrient standards were elaborated as centered log ratio means and standard deviations of high-yield and nutritionally balanced specimens. Plot-scale diagnosis was customized using the closest successful factor-specific tissue compositions identified by the smallest Euclidean distance from the diagnosed composition using centered or isometric log ratios. Nutrient imbalance differed between regional and plot-scale diagnoses, indicating the profound influence of local factors on plant nutrition. However, plot-scale diagnoses require large, reliable datasets to customize nutrient management using ML and CoDa models.

show abstract

Section: Machine Learningmentioning

confidence: 98%

Section: Compositional Data Analysis (Coda)mentioning

confidence: 99%

Nutrient Diagnosis of Fertigated “Prata” and “Cavendish” Banana (Musa spp.) at Plot-Scale

Neto

Deus

Filho³

et al. 2020

Plants

Self Cite

View full text Add to dashboard Cite

show abstract

“…We found that relating berry yield to nutrient composition of leaves and stems and other features collected in the same year resulted in the highest model accuracy where all yield-impacting features documented in the dataset were included in the model. There is thus a need for paradigm change toward factor-specific nutrient diagnosis and site-specific fertilizer recommendations supported by large and diversified datasets [35,54,55].…”

Section: Nutrient Standardsmentioning

confidence: 99%

Current and next-year cranberry yields predicted from local features and carryover effects

et al. 2021

Self Cite

View full text Add to dashboard Cite

Wisconsin and Quebec are the world leading cranberry-producing regions. Cranberries are grown in acidic, naturally low-fertility sandy beds. Cranberry fertilization is guided by general soil and tissue nutrient tests in addition to yield target and vegetative biomass. However, other factors such as cultivar, location, and carbon and nutrient storage impact cranberry nutrition and yield. The objective of this study was to customize nutrient diagnosis and fertilizer recommendation at local scale and for next-year cranberry production after accounting for local factors and carbon and nutrient carryover effects. We collected 1768 observations from on-farm surveys and fertilizer trials in Quebec and Wisconsin to elaborate a machine learning model using minimum datasets. We tested carryover effects in a 5-year Quebec fertilizer experiment established on permanent plots. Micronutrients contributed more than macronutrients to variation in tissue compositions. Random Forest model related accurately current-year berry yield to location, cultivars, climatic indices, fertilization, and tissue and soil tests as features (classification accuracy of 0.83). Comparing compositions of defective and successful tissue compositions in the Euclidean space of tissue compositions, the general across-factor diagnosis differed from the local factor-specific diagnosis. Nutrient standards elaborated in one region could hardly be transposed to another and, within the same region, from one bed to another due to site-specific characteristics. Next-year yield and nutrient adjustment could be predicted accurately from current-year yield and tissue composition and other features, with R2 value of 0.73 in regression mode and classification accuracy of 0.85. Compositional and machine learning methods proved to be effective to customize nutrient diagnosis and predict site-specific measures for nutrient management of cranberry stands. This study emphasized the need to acquire large experimental and observational datasets to capture the numerous factor combinations impacting current and next-year cranberry yields at local scale.

show abstract

“…Espécies do gênero Eucalyptus, são amplamente cultivadas no mundo. O Brasil é o maior produtor mundial, com 7,6 milhões de hectares plantados e áreas em constante expansão (Binkley et al, 2017;IBGE, 2019;Vahl de Paula et al, 2020).…”

Section: Introductionunclassified

Efeito da umidade do solo na fisiologia do eucalipto em competição com plantas daninhas

Alencar

Ferreira

Duque

et al. 2021

RSD

View full text Add to dashboard Cite

A competição com plantas daninhas por água ou luz pode afetar parâmetros fisiológicos das culturas de interesse. A partir desse estudo, objetivou-se avaliar o efeito da umidade do solo e a competição de duas espécies de plantas daninhas em diferentes combinações nas características fisiológicas de plantas de eucalipto. O delineamento experimental utilizado foi ao acaso com quatro repetições em esquema fatorial 4 x 4. O primeiro fator correspondeu às combinações de plantas do híbrido clonal de eucalipto cultivadas em competição com picão-preto e braquiária: 1 - Uma planta de eucalipto + duas plantas de braquiária por vaso; 2 - Uma planta de eucalipto + duas de picão-preto por vaso; 3 - Uma planta de eucalipto + uma planta de picão-preto + uma planta de braquiária por vaso, e; 4 - uma planta de eucalipto por vaso. O segundo correspondeu aos níveis de água no solo: 100, 74, 67 e 60% da capacidade de campo. Foram feitas avaliações dos teores de clorofila a, b, a/b e total, bem como, variáveis relacionadas a fluorescência da clorofila. O eucalipto crescendo isoladamente mostrou maiores valores de clorofila a clorofila b e clorofila total. A relação clorofila a/ b foi inferior em todos os níveis de umidade do solo. A fluorescência inicial mostrou tendência de decréscimo com o aumento da umidade. Nas menores umidades testadas as plantas de eucalipto se encontravam em condição de estresse com valores de fluorescência variável/fluorescência máxima inferiores a 0,75.

show abstract

Nutrient Diagnosis of Eucalyptus at the Factor-Specific Level Using Machine Learning and Compositional Methods

Cited by 20 publications

References 44 publications

Nutrient Diagnosis of Fertigated “Prata” and “Cavendish” Banana (Musa spp.) at Plot-Scale

Nutrient Diagnosis of Fertigated “Prata” and “Cavendish” Banana (Musa spp.) at Plot-Scale

Current and next-year cranberry yields predicted from local features and carryover effects

Efeito da umidade do solo na fisiologia do eucalipto em competição com plantas daninhas

Contact Info

Product

Resources

About