Search

Coffee production is significantly impacted by various diseases, predominantly those caused by fungi. One such notable disease in coffee crops is caused by the fungus Phoma spp. This pathogen leads to several symptoms detrimental to coffee plants, such as leaf lesions, drying of branches, and rotting of flowers and fruits. These symptoms often result in the dropping of the affected parts, subsequently leading to a decrease in the overall yield of the coffee crop. In response to this challenge, our objective was to develop a forecasting model for the incidence of Phoma leaf spot in Brazilian coffee crops, utilizing advanced machine learning algorithms. This approach is intended to predict disease outbreaks, thereby enabling timely and effective management strategies to mitigate the impact on coffee yield. The study was conducted in two stages: (1) calibration of machine learning models for locations (Boa Esperança, Carmo de Minas, Muzambinho, Varginha, Araxá, Araguari, and Patrocínio) with field data between 2010 and 2022; (2) Phoma leaf spot incidence forecast in municipalities of coffee-producing states in Brazil [Paraná (PR), São Paulo (SP), Rio de Janeiro (RJ), Espírito Santo (ES), Minas Gerais (MG), Goiás (GO), and Bahia (BA)]. Thirty-year climate data were retrieved from the NASA/POWER platform. Reference evapotranspiration was estimated by the Penman–Monteith method, generating water balance according to Thornthwaite and Mather (1955). To understand the effect of climate variables on the disease incidence, Pearson's univariate correlation was performed for each location. We used six algorithms to forecast the disease incidence, considering a 7-day latency period to define input variables. It is noteworthy that the evaluated locations present similar climatic conditions. Summer was the hottest and rainiest period, while winter was the coldest and driest. Annual averages of air temperature, cumulative rainfall, potential evapotranspiration, soil water storage, and incident radiation were 21.1 °C, 1208.9 mm, 1283.2 mm, 58.0 mm, 435.7 mm, and 18.1 MJ m2 day−1, respectively. The XGBoost model demonstrated superior performance for both high- and low-yield coffee trees, achieving an impressive precision (R2fit) of 0.46 and 0.51, respectively. Additionally, it exhibited high accuracy, with Root Mean Square Error (RMSE) values of 3.45% for high-yielding and 3.16% for low-yielding trees. In contrast, the multilayer perceptron (MLP) model displayed suboptimal results under both yield conditions. Given these findings, the XGBoost model proves effective in predicting the incidence of the disease at least 7 days ahead, based on the parameters applied in this study. [ABSTRACT FROM AUTHOR]

Brown eye spot (Cercospora coffeicola) is one of the main fungal diseases of coffee, leading to a significant drop in crop productivity and beverage quality in Brazil. The identification of potential risk areas for the development of the disease provides promising information for the management of the pathogen. This study aimed to elaborate an agroclimatic zoning for the incidence of brown eye spot on coffee under climate change scenarios, as suggested by IPCC (IPCC-AR5), in the main coffee-growing regions. Climate data of air temperature, precipitation, and relative humidity were collected from the National Aeronautics and Space Administration/Prediction of Worldwide Energy Resources (NASA/POWER) platform from 1989 to 2020 for 46 municipalities in the states of Paraná, São Paulo, Minas Gerais, Espírito Santo, Goiás, and Bahia. The ideal climate for brown eye spot occurrence consists of an air temperature (Tmean) between 18 and 30 °C, relative humidity (RHmean) > 90%, and leaf wetness duration (LWD) > 9 h. The number of hours of leaf wetness was determined by the sum of hours with relative humidity > 90%. Relative humidity was estimated for each hour of the day using air temperature and estimated data of mean dew point temperature, maximum relative humidity, and minimum relative humidity. Climate change scenarios were designed based on sets of climate simulations for the twenty-first century. Scenario S1 is the current scenario without changes, while scenarios S2 and S3 show Tmean + 1.5 °C and 3.0 °C, respectively, with RHmean without changes. Scenarios S4 and S5 present RHmean varying from − 30% to + 30%, respectively, with Tmean unchanged. In the current scenario (S1), Minas Gerais presented a predominance of 100% for low climate risk to brown eye spot in September. Paraná presented a medium risk in 76.15% of the state in April. Scenarios S2 and S3 showed significant changes, increasing the average fitness class in the study region, mainly in the states of São Paulo, Minas Gerais, Rio de Janeiro, and Paraná. Scenario S4 showed 100% predominance of the low-risk class. In contrast, S5 showed the occurrence of the high-risk class for the study region with a + 30% increase in relative humidity. Minas Gerais presented a predominance of the high-risk class for the development of C. coffeicola in 76%, 100%, 97.83%, 89.30%, 93.46%, 80.64%, and 57.77% from November to May, respectively. The presence of high relative humidity represents the main factor for the expansion of the high-risk class for the development of C. coffeicola. Producers knowing the months of the year and the places in Brazil with the highest incidence of brown eye spot will be able to prevent the disease in a more sustainable way, using more ecological products, such as the early application of copper. [ABSTRACT FROM AUTHOR]

We developed models for simulating trends over time as functions of the thermal index and models for estimating the levels of infestation of the coffee leaf miner and coffee berry borer and the severity of disease for coffee leaf rust and cercospora, the main phytosanitary problems in coffee crops around the world. We used historical series of climatic data and levels of pest infestation and disease severity in Coffea arabica for high and low yields for seven locations in the two main coffee-producing regions in the state of Minas Gerais in Brazil, Sul de Minas Gerais and Cerrado Mineiro. We conducted two analyses: (a) we simulated the trends of the progress of diseases and pests over time using non-linear models. We only used the thermal index because air temperature is commonly measured by farmers in the regions. (b) We estimated the levels of pest infestation and disease severity using multiple linear regression, with the levels of diseases and pests as dependent variables and accumulated degree days (DD), coffee foliage (LF) estimated by DD and the number of nodes (NN) estimated by DD as independent variables. We used DD and LF = f (DD) and NN = f (DD) to predict diseases and pests with accuracy. MAPEs were 19.6, 5.7, 9.5, and 15.8% for rust, cercospora, leaf miner, and berry borer, respectively, for Sul de Minas Gerais. Establishing phytosanitary alerts using only air temperature was possible with these models. [ABSTRACT FROM AUTHOR]

The natural quality (NQ) of a coffee drink is defined as the one that is obtained in the year of bean production with the standard postharvest treatments of the region. The NQ varies with the climatic conditions during the crop cycle and with the drying period of beans in the sun. A well-formed bean can provide a good drink if well processed, but a bean grown during unfavourable environmental conditions will have poor quality regardless of the postharvest options. The goals of this study were to (1) increase our understanding of the relationship between meteorological elements (MEs) and NQ in coffee-producing regions, (2) identify the ranges of the MEs during the crop cycle that optimise quality, and (3) develop models to predict NQ based on the MEs. We used the two major regions of coffee production in the Brazilian state of Minas Gerais, the Cerrado Mineiro (CEMG), and the southern Minas Gerais (SOMG). We mapped the influences of ME on NQ successfully. Air temperature in November and December for SOMG and precipitation from November to January for CEMG were generally the most important MEs for NQ. Water deficiency, water storage, and rainfall became increasingly more important during winter (June to September) than during other seasons. The crop models were accurate, with errors < 7.9% for predicting NQ for all regions. These models used precipitation in June and December, the actual evapotranspiration in May, and the water deficit in April as the MEs for CEMG, and the rainfall in June and December, the water storage in April, and the actual evapotranspiration in May as the MEs for SOMG. [ABSTRACT FROM AUTHOR]

Disease and pest alert models are able to generate information for agrochemical applications only when needed, reducing costs and environmental impacts. With machine learning algorithms, it is possible to develop models to be used in disease and pest warning systems as a function of the weather in order to improve the efficiency of chemical control of pests of the coffee tree. Thus, we correlated the infection rates with the weather variables and also calibrated and tested machine learning algorithms to predict the incidence of coffee rust, cercospora, coffee miner, and coffee borer. We used weather and field data obtained from coffee plantations in production in the southern regions of the State of Minas Gerais (SOMG) and from the region of the Cerrado Mineiro; these crops did not receive phytosanitary treatments. The algorithms calibrated and tested for prediction were (a) Multiple linear regression (RLM); (b) K Neighbors Regressor (KNN); (c) Random Forest Regressor (RFT), and (d) Artificial Neural Networks (MLP). As dependent variables, we considered the monthly rates of coffee rust, cercospora, coffee miner, and coffee tree borer, and the weather elements were considered as independent (predictor) variables. Pearson correlation analyses were performed considering three different time periods, 1–10 d (from 1 to 10 days before the incidence evaluation), 11–20 d, and 21–30 d, and used to evaluate the unit correlations between the weather variables and infection rates of coffee diseases and pests. The models were calibrated in years of high and low yields, because the biannual variation of harvest yield of coffee beans influences the severity of the diseases. The models were compared by the Willmott's 'd', RMSE (root mean square error), and coefficient of determination (R2) indices. The result of the more accurate algorithm was specialized for the SOMG and Cerrado Mineiro regions using the kriging method. The weather variables that showed significant correlations with coffee rust disease were maximum air temperature, number of days with relative humidity above 80%, and relative humidity. RFT was more accurate in the prediction of coffee rust, cercospora, coffee miner, and coffee borer using weather conditions. In the SOMG, RFT showed a greater accuracy in the predictions for the Cerrado Mineiro in years of high and low yields and for all diseases. In SOMG, the RMSE values ranged from 0.227 to 0.853 for high-yield and 0.147 and 0.827 for low-yield coffee in the coffee borer forecasting. [ABSTRACT FROM AUTHOR]

Copyright of Revista em Agronegócio e Meio Ambiente is the property of Revista em Agronegocio e Meio Ambiente and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)