Technical Guide

A deeper look at how the WWF Traffic Workflow is built: the portal login & download automation, the multi-file CSV merge pipeline, GVL spatial filtering, map and chart methodology, and metric calculations.

Download the full Technical Guide (PDF)

Unlike the EarthRanger-connected workflows in this fleet, WWF Traffic has no EarthRanger data source — input data is downloaded directly from the TRAFFIC Wildlife Trade Portal via direct API calls (with automatic retry-with-backoff if the portal rate-limits a request), optionally merged with a local CSV, and the only other network dependency is a one-time Dropbox download of the GVL boundary file. This guide was written directly against the current spec.yaml.

Contents

1. Overview
2. Dependencies & Prerequisites
   1. Traffic Portal Connection & Download
   2. Input Data — Three Merged CSV Exports
   3. Grouping Strategy
   4. GVL 30 km Buffer Boundary
   5. Base Map Tile Layers
   6. Colour Palette
3. Data Ingestion Pipeline
   1. Column Mapping
   2. Trade Route De-duplication
   3. Numeric Conversion & Geometry
   4. Spatial Join Against the GVL Boundary
   5. Country Restriction
   6. Species & Category Mapping
   7. Temporal Index
   8. Optional Local Data Merge
4. Static Map Layers
   1. GVL Boundary Layer
5. Map Outputs — Methodology
   1. Incidents Map
   2. Species Map
   3. Filename Suffix Collision
6. Charts
   1. Incidents by Species / by Category
7. Summary Metrics
   1. Per-Year Aggregations
   2. Conviction Rate — Two Independent Calculations
   3. Ivory Seizures
   4. Most Trafficked Species
8. Summary Table & CSV Export
9. Interactive Dashboard
10. Output Files
11. Workflow Execution Logic
    1. Skip Conditions
    2. Data Flow Summary
12. Software Versions

Read it here

The guide is embedded below. If it doesn't render in your browser, use the download link above.