sodustus/CLIMATELAB

Ethiopia Climate Lab is an interactive platform built to visualize and simulate climate data across Ethiopia's regions (1991–2020). Using a CDS API Python bot, SVG maps, and JavaScript, it translates temperature, rainfall, and stress metrics into actionable visual timelines and dynamic survivability simulations for regional planning. ...a brief explanation on a project bridging climate data, visualization and simulation. All built from scratch for Ethiopia's Diverse regions. In the Benninging During my review of blockchain-based agricultural finance proposals, it became clear that microfinance and insurance solutions often overlooked the seasonal and geographic realities of smallholder farms. To translate abstract ideas into actionable insights, I developed Ethiopia Climate Lab: an interactive platform that contextualizes climate data for Ethiopian regions. It combines visualization, simulation and analysis. The goal is to make climate effects tangible, show agricultural vulnerabilities and experiment with what-if scenarios across temperature and rainfall variations. The Data From day one, the challenge was both technical and conceptual. Interactive SVG maps pushed me to refine hover interactions, tooltips and sidebars. Managing 1650 data points across 11 regions, 3 variables (converted to 5 after calculations) and 30 years showed the limitations of manual entry, so I built a Python bot to automate data extraction via the CDS (Climate Data Store) API. Each iteration taught me new ways to integrate programming with climate science. Inspiration came from unexpected sources: video games, institutional climate research and the logic of incremental statistics. The Timeline I used the average of the climate data from 1991-2020 to squeeze out averages shown in the visualizer page. It did not require further data fetching to prepare for the timeline version. The difficult part was converting the data from default state of .nc (netCDF) files to JSON entries that could be read by the JavaScript to make the timeline version come alive. For this purpose, I used the online tool MyGeoData. Even after conversion, restructuring the data in a way that allows the Javascript to read it was a hurdle of its own. Once the data pipeline was stable, the Timeline came alive. This finally gave me some room to exp …