Caio Bustani Andrade
Caio Bustani Andrade has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Alexandre Ten Caten
Federal University of Santa Catarina
2 shared publications - Gustavo Brunetto
Department of Soil Science, Federal University of Santa Maria, Santa Maria, Brazil
1 shared publication - Jean Michel Moura-Bueno
Soil Science Department, Federal University of Santa Maria (UFSM), Santa Maria 97105-900, Brazil
1 shared publication - Jucinei José Comin
Universidade Federal de Santa Catarina
1 shared publication - Lucas Kreutzfeld
Universidade Federal de Santa Catarina
1 shared publication - Luiz Carlos Pittol Martini
Universidade Federal de Santa Catarina
1 shared publication
Latest Publications
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Data Descriptions
- Vol. 3, 880-886
Descriptor: SensNexus Spectral Dataset (SensNexusDat)
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Data Descriptions
- Vol. 3, 880-886
Open AccessDescriptor: SensNexus Spectral Dataset (SensNexusDat)
- 1 January 2026
- Published by Revista Brasileira de Ciencia do Solo in Revista Brasileira de Ciência do Solo
- Vol. 50 (spe1)
Microplastics in terrestrial ecosystems: Soil contamination and its implications for pedology
- 1 January 2026
- Published by Revista Brasileira de Ciencia do Solo in Revista Brasileira de Ciência do Solo
- Vol. 50 (spe1)
Microplastics in terrestrial ecosystems: Soil contamination and its implications for pedology
- Open Access
- 30 November 2023
- Published by MDPI AG in Horticulturae
- Vol. 9 (12) , 1294
Grape Yield Prediction Models: Approaching Different Machine Learning Algorithms
Efficient marketing of winegrapes involves negotiating with potential buyers long before the harvest, when little is known about the expected vintage. Grapevine physiology is affected by weather conditions as well as by soil properties and such information can be applied to build yield prediction models. In this study, Partial Least Squares Regression (PLSR), Cubist (CUB) and Random Forest (RF) algorithms were used to predict yield from imputed weather station data and soil sample analysis reports. Models using only soil variables had the worst general results (R2 = 0.15, RMSE = 4.16 Mg ha−1, MAE = 3.20 Mg ha−1), while the use of only weather variables yielded the best performance (R2 = 0.52, RMSE = 2.99 Mg ha−1, MAE = 2.43 Mg ha−1). Models built with CUB and RF algorithms showed signs of overfitting, yet RF models achieved the best average results (R2 = 0.58, RMSE = 2.85 Mg ha−1, MAE = 2.24 Mg ha−1) using only weather variables as predictors. Weather data imputation affected RF and CUB models more intensely while PLSR remained fairly insensitive. Plant age, yield level group, vineyard plot, May temperatures, soil pH and exchangeable concentrations of Zn, Cu, K and Mn were identified as important predictors. This exploratory work offers insights for future research on grape yield predictive modeling and grouping strategies to obtain more assertive results, thus contributing to a more efficient grapevine production chain in southern Brazil and worldwide.
- 30 November 2023
- Published by MDPI AG in Horticulturae
- Vol. 9 (12) , 1294
Open AccessGrape Yield Prediction Models: Approaching Different Machine Learning Algorithms
Efficient marketing of winegrapes involves negotiating with potential buyers long before the harvest, when little is known about the expected vintage. Grapevine physiology is affected by weather conditions as well as by soil properties and such information can be applied to build yield prediction models. In this study, Partial Least Squares Regression (PLSR), Cubist (CUB) and Random Forest (RF) algorithms were used to predict yield from imputed weather station data and soil sample analysis reports. Models using only soil variables had the worst general results (R2 = 0.15, RMSE = 4.16 Mg ha−1, MAE = 3.20 Mg ha−1), while the use of only weather variables yielded the best performance (R2 = 0.52, RMSE = 2.99 Mg ha−1, MAE = 2.43 Mg ha−1). Models built with CUB and RF algorithms showed signs of overfitting, yet RF models achieved the best average results (R2 = 0.58, RMSE = 2.85 Mg ha−1, MAE = 2.24 Mg ha−1) using only weather variables as predictors. Weather data imputation affected RF and CUB models more intensely while PLSR remained fairly insensitive. Plant age, yield level group, vineyard plot, May temperatures, soil pH and exchangeable concentrations of Zn, Cu, K and Mn were identified as important predictors. This exploratory work offers insights for future research on grape yield predictive modeling and grouping strategies to obtain more assertive results, thus contributing to a more efficient grapevine production chain in southern Brazil and worldwide.
- Open Access
- 16 March 2023
- Published by Empresa de Pesquisa Agropecuaria e Extensao Rural de Santa Catarina in Agropecuária Catarinense
- Vol. 36 (1) , 30-36
Estimativa da área de cebola por meio de imagens de satélites para os municípios de Ituporanga e Imbuia, SC, Brasil
Este estudo utilizou imagens Landsat 8-OLI e Sentinel 2A/2B-MSI para estimar a área destinada à produção de cebola em dois municípios do estado de Santa Catarina (Ituporanga e Imbuia) durante a safra 2020/2021. Uma vez que as lavouras de cebola na região de estudo ocupam pequenas áreas com cerca de 2 hectares, os usos das terras foram identificadosusando a interpretação visual de composições coloridas multi temporais geradas pela combinação de bandas espectrais do vermelho, infravermelho próximo e infravermelho de ondas curtas. As imagens de satélite mostraram que a cebola cobriu 3.136 hectares em Ituporanga e 934 hectares em Imbuia na safra 2020/2021. Em ambos os municípios avaliados, as estimativas obtidas a partir de imagens de satélite foram menores do que as esti mati vas obti das por levantamentos indiretos baseados em declarações das partes locais envolvidas na ati vidade ou na quantidade comercializada de insumos. O estudo mostrou que as imagens Landsat 8-OLI e Sentinel 2A/2B-MSI são adequadas para identificar e estimar áreas cultivadas com cebola na região produtora de Santa Catarina. Dessa forma, as imagens de satélite podem auxiliar os levantamentos estatísticos convencionais utilizados pelos organismos oficiais responsáveis pelas estatísticas agrícolas.
- 16 March 2023
- Published by Empresa de Pesquisa Agropecuaria e Extensao Rural de Santa Catarina in Agropecuária Catarinense
- Vol. 36 (1) , 30-36
Open AccessEstimativa da área de cebola por meio de imagens de satélites para os municípios de Ituporanga e Imbuia, SC, Brasil
Este estudo utilizou imagens Landsat 8-OLI e Sentinel 2A/2B-MSI para estimar a área destinada à produção de cebola em dois municípios do estado de Santa Catarina (Ituporanga e Imbuia) durante a safra 2020/2021. Uma vez que as lavouras de cebola na região de estudo ocupam pequenas áreas com cerca de 2 hectares, os usos das terras foram identificadosusando a interpretação visual de composições coloridas multi temporais geradas pela combinação de bandas espectrais do vermelho, infravermelho próximo e infravermelho de ondas curtas. As imagens de satélite mostraram que a cebola cobriu 3.136 hectares em Ituporanga e 934 hectares em Imbuia na safra 2020/2021. Em ambos os municípios avaliados, as estimativas obtidas a partir de imagens de satélite foram menores do que as esti mati vas obti das por levantamentos indiretos baseados em declarações das partes locais envolvidas na ati vidade ou na quantidade comercializada de insumos. O estudo mostrou que as imagens Landsat 8-OLI e Sentinel 2A/2B-MSI são adequadas para identificar e estimar áreas cultivadas com cebola na região produtora de Santa Catarina. Dessa forma, as imagens de satélite podem auxiliar os levantamentos estatísticos convencionais utilizados pelos organismos oficiais responsáveis pelas estatísticas agrícolas.