CLIMATE AND WEATHER-BASED PREDICTIVE MODEL FOR ZIKA EPIDEMIC INCIDENCE IN BRAZIL AND COLOMBIA

The Zika virus (ZIKV) is a member of the Flavivirus genus and is primarily transmitted through the bite of Aedes mosquitoes. Its transmission is influenced by environmental conditions, making climate and weather factors crucial in understanding its spread. Key climatic factors affecting ZIKV transmission include precipitation, Niño 3.4 index, maximum and minimum temperatures, relative humidity, surface soil wetness, and all-sky surface shortwave downward irradiance, particularly in Brazil and Colombia. This study aims to analyze and monitor ZIKV transmission patterns, necessitating a predictive model to forecast future trends. 

This study employs a time-series analysis by integrating climate variables with Zika case data. The research methodology consists of three main stages: (1) Data Processing, which includes data loading, cleaning, and outlier detection and removal to ensure data quality. (2) Analysis and Modeling, which involves feature selection, lag variable creation, and the application of predictive models such as Linear Regression. Model evaluation is conducted using K-Fold Cross Validation to enhance accuracy and prevent overfitting. (3) Model Evaluation and Visualization, where model performance is assessed using R² and RMSE. Predictions are compared with observational data through scatter plots and time-series graphs to interpret the patterns of Zika virus transmission.

The spread of the Zika virus is influenced by historical factors, with Zika Lag 1, 2, and 3 making dominant contributions to the model. This indicates that case numbers from previous weeks serve as strong predictors for subsequent weeks, likely due to the virus’s incubation period, transmission patterns, and vector mosquito dynamics. Among various climatic factors, maximum temperature, minimum temperature, and relative humidity emerge as the most influential. Temperature affects mosquito development, while relative humidity influences mosquito survival and breeding conditions. This study highlights the importance of climate-based analysis in predicting Zika transmission, serving as a foundation for mitigation efforts and environmentally driven prevention strategies. These findings are expected to contribute to more effective public health policies for controlling future Zika outbreaks.

Keywords: Zika Virus, Predictive Modeling, Climatic Factors, Temperature, Relative Humidity, Public Health Policy