Predictive Analytics for Climate-Sensitive Sectors

• Global overview of climate-sensitive sectors and their vulnerabilities

• Introduction to predictive analytics: regression, classification, time series forecasting

• Understanding and accessing open climate data sources (NASA EarthData, NOAA, Copernicus, IPCC)

• Data preparation techniques: missing values, temporal formatting, spatial tagging

• Visualizing climate trends using Python and Jupyter Notebooks

🛠️ Hands-on Tools: Python (Pandas, NumPy, Plotly), Jupyter Notebook

• Machine learning techniques for climate data: ARIMA, Random Forest, XGBoost

• Predictive modeling for rainfall, temperature, and crop yield

• Introduction to geospatial data: raster vs vector, spatial layers, map overlays

• Climate zoning and risk area detection (e.g., drought, floods)

• Model development using real datasets, including training, testing, and validation

🛠️ Hands-on Tools: Scikit-learn, GeoPandas, Folium, QGIS, Google Earth Engine

• Review of global case studies: agriculture forecasting, hydrology, health risk models

• Building a full predictive pipeline: data ingestion → modeling → evaluation → visualization

• Model evaluation metrics: MAE, RMSE, R², bias assessment

• Designing interactive climate dashboards using Streamlit

• Group-based capstone project: Presenting a predictive solution for a real-world climate issue

🛠️ Hands-on Tools: Streamlit, GitHub, Python (Seaborn, Altair, Matplotlib)

Edit

Edit