|Provider:||National Centers for Environmental Prediction, NOAA (USA)|
|Update frequency:||every 6 hours|
|Resolution:||0.5°, 30.0nm, 55.6km|
|Model duration:||41 forecasts starting at 0 hr, ending at 16 days|
|Parameters:||pressure, wind, rain, cloud, temperature, humidity, vertical velocity, 250 mb, 500 mb, 850 mb|
|GRIB model date:||Sun Jan 20 00:00:00 2019 UTC|
|Download date:||Sun Jan 20 05:33:47 2019 UTC|
|Download delay:||5hr 33min|
This is an ensemble model, and in this case, it is the result of averaging 21 slightly different weather models together. The idea in an ensemble model is that there is instability in many weather systems and that capturing the initial state of the atmosphere for the weather models can not be done perfectly. The ensemble process slightly alters the initial inputs and/or the numerical model in order to generate a range of different results, which are then averaged to create the final result.
Some people feel that an ensemble model is better at longer term forecasts. It is a useful exercise to download both an ensemble model as well as a normal weather model and compare them. If the two forecasts are similar, you may be able to feel more confident in their prediction ability.
The following description has been taken directly from the official documentation
The Global Ensemble Forecast System (GEFS), previously known as the GFS Global ENSemble (GENS), is a weather forecast model made up of 21 separate forecasts, or ensemble members. The National Centers for Environmental Prediction (NCEP) started the GEFS to address the nature of uncertainty in weather observations, which is used to initialize weather forecast models. The proverbial butterfly flapping her wings can have a cascading effect leading to wind gusts thousands of miles away. This extreme example illustrates that tiny, unnoticeable differences between reality and what is actually measured can, over time, lead to noticeable differences between what a weather model forecast predicts and reality itself. The GEFS attempts to quantify the amount of uncertainty in a forecast by generating an ensemble of multiple forecasts, each minutely different, or perturbed, from the original observations.