Full text
| |
| Author(s): |
Lucas Eduardo de Oliveira Aparecido
Total Authors: 1
|
| Document type: | Doctoral Thesis |
| Press: | Jaboticabal. 2019-10-10. |
| Institution: | Universidade Estadual Paulista (Unesp). Faculdade de Ciências Agrárias e Veterinárias. Jaboticabal |
| Defense date: | 2019-08-30 |
| Advisor: | Glauco de Souza Rolim |
| Abstract | |
Coffee is the most consumed beverage in the world, but phytosanitary problems are amongst the main causes of reduced productivity and quality. The application of foliar fungicides and insecticides is the most common strategy for controlling these diseases and pests, depending on their intensity in a region. This traditional method can be improved by using alert systems with models of disease and pest indices. This work has as OBJECTIVES: A) To calibrate the meteorological variables: air temperature and rainfall of the European Center for Medium Range Weather Forecast (ECMWF) in relation to the real surface data measured by the national meteorological system (INMET) for the state of Minas Gerais; B) To evaluate which meteorological elements, and at what time, have a greater influence on the main pests (coffee borer and coffee miner) and diseases (coffee rust and cercosporiosis) of Coffee arabica in the main coffee regions of the South of Minas Gerais and Cerrado Mineiro; C) To develop agrometeorological models for pest and disease prediction in function of the meteorological variables of the South of Minas Gerais and Cerrado Mineiro using algorithms of machine learning with sufficient temporal anticipation for decision making. MATERIAL AND METHODS: To achieve goal "A" we used monthly climatic data (T, ºC) and rainfall (P, mm) from the ECMWF and INMET from 1979 to 2015. Potential evapotranspiration was estimated by Thornthwaite (1948) and water balance by Thornthwaite and Mather (1955). The comparisons between the ECMWF and INMET were performed by the indexes: mean absolute percentage error (MAPE) and precision mean (R2adj). To achieve the goal "B" we use climatic and phytosanitary data of Boa esperança, Carmo de Minas, Muzambinho and Varginha, located in the South of Mines (SOMG) and the Araxá, Araguari and Patrocínio located in the Cerrado Mineiro region (CEMG). We simulate the trend of disease and pest progression over time using nonlinear models as a function of the accumulated thermal index. And we estimated levels of pest infestation and disease severity using multiple linear regression. The dependent variable was the levels of diseases and pests and the independent variables: cumulative days (DD), coffee leafage estimated by DD and number of nodes estimated by DD. To achieve the "C" objective we use the climatic and phytosanitary data of SOMG and CEMG. The algorithms calibrated and tested for the prediction of coffee pests and diseases were: 1) Multiple linear regression, 2) K-Neighbors Regressor, 3) Random Forest Regressor and 4) Artificial Neural Networks. The best models were selected using the MAPE, Willmott's 'd', RMSE and R² adj. RESULTS AND DISCUSSION: The largest deviations between PINMET and PECMWF were 75 mm mo-1 and occurred in the summer. The coffee plant implanted in CEMG has higher rates of diseases and pests in relation to SOMG coffee. The random forest algorithm was more accurate in the prediction of coffee rust, cercospora, coffee miner and coffee borer in both regions. CONCLUSION: The climatic variables from the ECMWF are accurate and can be used in modeling the climatological water balance. It is possible to simulate the trend and to predict coffee pests and diseases using as regressive variables the climatic data and machine learning methodology. (AU) | |
| FAPESP's process: | 15/17797-4 - Agrometeorological models for forecasting pests and diseases in Coffea arabica L. in the State of Minas Gerais |
| Grantee: | Lucas Eduardo de Oliveira Aparecido |
| Support Opportunities: | Scholarships in Brazil - Doctorate |