Mapping Weather Impacts on Economy using Geospatial & Remote Sensing Data (Central Africa)

Project Overview

As lead analyst for this component of a UK-based international research project, my role was to bridge the gap between large-scale remote sensing weather data and ground-level economic realities in a Central African country.

Work Done

• Methodology Development: Designed innovative techniques to spatially and temporally align vast remote sensing datasets (e.g., rainfall estimates, temperature grids) with disparate economic survey data, administrative boundaries, infrastructure maps, and development indicators.
• Custom R Functions: Wrote computationally efficient R code for handling large geospatial datasets, including optimized functions for geodesic distance calculations relevant to market access modeling.
• Data Pipeline Management: Architected and executed the full data pipeline – extracting raw data from multiple sources (APIs, files, databases), performing rigorous cleaning and validation, structuring data for analysis, and integrating datasets into a unified analytical framework.
• Analysis & Reporting: Conducted exploratory data analysis and provided key derived datasets and visualizations to the project’s econometrics team.

Deliverables

• Integrated datasets mapping weather variables to economic/geographic features.
• A library of documented, reproducible R scripts and functions for the analysis.
• Cleaned and structured datasets suitable for advanced statistical modeling.

Solution Overview

Developed novel methodologies in R to map large-scale remote sensing weather datasets (rainfall, temperature anomalies) onto fine-grained economic/geographic features (e.g., crop yields, market access, regional GDP proxies) and development indices. Created computationally efficient geodesic distance calculations in R. Managed the complex ETL process for diverse data sources (satellite imagery, surveys, administrative data).