Guides
Weather model guides
Weather models are easier to use once you know what the labels mean. These guides explain how to read a model map, how forecast accuracy is measured, what a skew-T and a 500 mb map show, how ensembles express uncertainty, and where the 40/70 benchmark comes from. Each one links to the model pages and, where it helps, to the 4070 tool that shows the same thing.
Which weather model is the most accurate?
How forecast accuracy is measured, why the answer changes with lead time, variable, region and season, and how to judge models for yourself.
How to read weather models
Runs, valid time, forecast hours, resolution, ensembles, the common map fields and a checklist for reading a model map without being fooled by it.
How to read a skew-T
The axes, the temperature and dew point lines, wind, moist and dry layers, warm noses and precipitation type, and CAPE and CIN, with a step-by-step read.
How to read a 500 mb map
What 500 mb is, how to read heights in decameters, ridges, troughs and shortwaves, how they relate to surface storms, and what height anomalies add.
Ensembles and spaghetti plots, explained
Members, spread, plumes and spaghetti plots, how to use them for tropical and winter storms, and how spread tells you how much to trust a forecast.
What is the 40/70 benchmark?
The point at 40°N, 70°W that forecasters use to judge a nor’easter’s track, what it tells you about snow and rain, and where the rule of thumb stops working.
See the models for yourself.
Model Explorer, Compare Models, soundings and plumes are part of 4070 Pro, with a seven-day free trial.