Workflow Structure (Part 2)

[!NOTE] 📹 Video Reference 16:40 Review the Workflow Structure Part 2

The second phase of the workflow (Steps 5-8) handles the cleaning, harmonization, and the core outlier detection logic powered by Specleaner.

Phase 2: Processing & Detection

Step Goal Details
Step 5: Merge & Harmonize Columns Combine online and offline data into a unified structure. Maps divergent column names to a standard set: Species, Decimal Latitude, Decimal Longitude, Date, Country.
Step 6: Harmonize Species Names Ensure taxonomic consistency. Connects to FishBase. Resolves scientific synonyms and ensures all records use the currently accepted scientific name.
Step 7: Extract Environmental Predictors Add environmental context to each occurrence point. Uses WorldClim raster files. For every coordinate pair, it extracts environmental values (e.g., Mean Annual Temperature).
Step 8: Flag Outliers (The Core) Apply Specleaner algorithms. Tests a chosen variable (e.g., Temp) against selected univariate and multivariate methods. Flagged by outlier strength (Poor, Fair, Strong, Perfect).