Saiteja380/EY_Data-Challenge

EY Optimizing Clean Water Supply (South Africa): Predict Alkalinity, Electrical Conductance, and Phosphorus levels in South African rivers. # 🌊 EY Open Science Data Challenge 2026: River Health Predictor **Solo Operator:** [Your Name] **Professional Profile:** Mechanical Engineering Graduate / Junior Research Analyst **Timeline:** Jan 30 – March 13 (43 Days) **Project Status:** 🏗️ Phase 0: Infrastructure Setup --- ## 🎯 Project Vision & Global Impact Developing an automated, high-precision forecasting model for river flow and environmental health indicators. This project aligns with the **United Nations Sustainable Development Goals**: - **Goal 6:** Clean Water and Sanitation. - **Goal 13:** Climate Action. - **Goal 15:** Life on Land (via NDVI/Vegetation monitoring). ### ⏱️ The Solo Operator Routine (24-Hour Cycle) To maintain progress alongside a 10-7 professional schedule, this project follows a disciplined execution loop: * **07:30 PM - 10:30 PM:** **Deep Work** (Coding, Model Tuning, Logic). * **10:30 PM - 08:30 AM:** **Automation** (Data Ingestion, Cloud Processing). * **08:30 AM - 07:30 PM:** **Monitoring** (Passive status checks via GitHub Mobile). --- ## 🏗️ Technical Architecture & Methodology I leverage a modern, cloud-native stack to process massive geospatial datasets efficiently: ```mermaid graph LR A[NASA/Copernicus API] --> B{Snowflake Ingest} B --> C[H3 Spatial Grid] C --> D[Feature Engineering] D --> E[Hybrid LSTM Model] E --> F[Streamlit Dashboard] F --> G[Cortex AI Insights] Data Warehouse: Snowflake (SQL & Python Snowpark). Spatial Indexing: Uber H3 Hexagonal Grid (Resolutions 7 & 8). Engineering Logic: Applying hydrological lags (7/14/30-day rainfall) and vegetation health indices (NDVI/NDWI) to model catchment retention. AI Model: Hybrid LSTM for time-series forecasting.