What problem does it solve? Operating the SUMMA hydrologic model requires deep operational expertise: preparing GRU/HRU domains, converting forcing units correctly, choosing among 35 physics decision categories, generating consistent NetCDF inputs, and recovering from silent failures like unit mismatches or vegetation-table mis-mappings. This Skill packages that expertise so an AI agent can run the real SUMMA Fortran binary end-to-end without substituting approximations. ## Core Features & Use Cases - Full 7-stage pipeline: domain setup (GRU/HRU from shapefile + DEM), VIC or reanalysis forcing conversion, decision configuration with validation, ROSETTA-based soil parameters, cold-start initial conditions, execution with fileManager validation, and automated multi-physics comparison with plots. - Diagnostic recovery: 18+ documented diagnostic triplets map symptoms (e.g., zero runoff, ET bias, CRS mismatch, Fortran path truncation) to diagnoses and validated remedies. - Calibration support: a commissioned calibration.yaml defines parameter ranges, constraints, and a runner for NSE-based calibration against gauged discharge. - Use Case: Compare snow-physics options on a mountain basin by running full-factorial decision variants and scoring simulated discharge against a HYDAT gauge. ## Quick Start Ask the agent to run the preflight check and then execute the SUMMA pipeline for your basin shapefile, DEM, and forcing data using the default decision set.
