This project presents an exploratory data analysis (EDA) of daily climate data from the Aeroparque Weather Station (Buenos Aires, Argentina) for the period 2002–2022.
The main goal of the project is to showcase an exploratory data analysis workflow in Python, focusing on data cleaning, transformation, exploratory analysis, and visualization.
Climate data are used as a practical case study to explore time series behavior, basic statistical patterns, and environmental variables.
Author: Joaquín Llarena
Year: 2026
Note: This project's documentation and analysis are conducted in Spanish.
The README is provided in English for international accessibility.
- Build a clean and consistent dataset from raw daily records provided by the Argentine National Meteorological Service (SMN).
- Explore seasonal and interannual variability of precipitation and wind.
- Identify and characterize extreme precipitation and wind events.
- Produce clear and reproducible metrics and visualizations, suitable for inclusion in a Data Analysis portfolio.
- Source: Argentine National Meteorological Service (SMN)
- Station: Aeroparque Weather Station
- Location: Buenos Aires (CABA), Argentina
- Period: 2002–2022
- Temporal resolution: Daily records
- Daily precipitation
- Wind speed (scalar mean and daily maximum)
- Wind direction
The original data are provided in raw tabular text formats and are not directly suitable for analysis.
Significant preprocessing was required, including parsing, cleaning, quality control, and integration into a consistent and analysis-ready dataset.
- Python
- pandas – data manipulation, cleaning, and aggregation
- numpy – numerical operations
- matplotlib – data visualization
- Jupyter Notebook – analysis, documentation, and reproducible workflows
- Data loading from raw SMN files.
- Data cleaning and quality control.
- Construction of derived variables and basic aggregations (monthly and annual).
- Exploratory analysis of precipitation and wind behavior.
- Identification of extreme events using percentile-based thresholds.
- Visualization and summary of results.
The analysis was developed and executed using Jupyter Notebook.
All notebooks are fully reproducible using the provided data and standard Python libraries for data analysis and visualization.
- Monthly and interannual analysis of precipitation.
- Identification of extreme precipitation events by month and by year type.
- Seasonal and directional characterization of wind behavior.
- Directional analysis of extreme wind events based on daily maximum wind speed.
- Comparative analysis of wind direction and extreme events between two sub-periods (first vs. last 10 years).
- Precipitation and wind regimes show clear seasonal patterns, with higher intensity and variability during warmer months.
- Precipitation exhibits strong interannual variability, with alternating wet and dry years and a slight shift toward drier conditions in recent years.
- Extreme precipitation events are highly seasonal and are more frequent during wet years, although they can also occur under normal or dry annual conditions.
- Wind direction is dominated by easterly flows in terms of frequency, while the strongest wind events are primarily associated with southern sectors.
- The comparison between the first and last decades of the record highlights small differences in wind behavior, suggesting possible changes in the relative contribution of certain wind directions.
This study is limited to a single weather station and focuses on exploratory analysis rather than trend attribution or predictive modeling.
/
├── notebook/
│ └── analisis_climatico_aeroparque.ipynb
├── data/
│ └── raw_data/
├── figures/
└── README.md