GLM Calibration for Lake Temperature
SkillMonitoring & opsCalibration guidance for GLM tasks. Often most effective after glm-basics has clarified the setup; glm-output is the companion skill for exact final metric computation.
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the GLM Calibration for Lake Temperature skill
What this skill tells your AI
The instructions your AI receives, as published by cxcscmu/skilllearnbench in skills/human_authored/temperature-simulation/glm-calibration/SKILL.md and read by ahel’s review.
Key Calibration Parameters (Lake Mendota)
| Parameter | Range | Effect |
|---|---|---|
Kw | [0.1, 0.5] | Light extinction; higher = less deep heating, stronger stratification |
coef_mix_hyp | [0.3, 0.7] | Hypolimnetic mixing; higher = more deep mixing, warmer hypolimnion |
wind_factor | [0.7, 1.3] | Wind speed multiplier; higher = more surface mixing |
lw_factor | [0.7, 1.3] | Longwave radiation multiplier; affects surface energy balance |
ch | [0.0005, 0.002] | Sensible heat transfer coefficient |
Calibration Strategy
- Start with defaults, run, compute RMSE
- Adjust
Kwfirst (strongest control on stratification) - Then
coef_mix_hyp(controls deep temperatures) - Fine-tune
wind_factorandlw_factorfor surface/overall bias chhas moderate effect on surface heat exchange
RMSE Computation
- Match observations to simulation by exact datetime and rounded depth
- depth_sim = round(lake_depth - z) to get depth from surface
- Overall RMSE, deep (>=13m) RMSE, summer deep (Jun-Sep, >=13m) RMSE
Signals
- GitHub stars
- 83
- Forks
- 5
- Last commit
- Jul 2026
Advanced
- Catalog kind
- skill
- Gateway key
glm-calibration- Source
- github.com/cxcscmu/skilllearnbench