From my experienceI'd like to sketch a typical setup, how a pilot might be dissatisfied with MET:
Pilot: "I'll be landing at A tomorrow around sunrise, what do I have to expect ?"
MET: "weak easterly surface winds, a few clouds at 5000 ft amsl and a chance of shallow but dense fog patches that may at times cover parts of the runway system around sunrise. It will be gone in the first few hours of the day."
Pilot: "I can't work with that. I need a precise forecast."
MET : "Fog with 400 m RVR will come at 4 and be gone at 7 am local time."
Fog comes shallow at 2:30, grows 10 m high at 4 in the meadows along the runway, briefly rolls over the concrete with RVR 300 m, weakens to harmless state at 5 while a bank of cloud is hovering over the field and the last remaining fog patches finally dissipate at 6:30, one hour after sunrise.
Pilot lands at 5:30 and calls next day: "I knew you would be wrong, there was absolutely nothing, I could see the field from 30 miles out. You guys really need to get your act together."
(Exchange the fog for a thunderstorm scenario during the hot season )
My point being, with whatever meteorological sources you have, it's much better to evaluate and - if you like - discuss the risks and chances from a healthy mix of raw data and interpreted forecasts and then establish a strategy to cover those risks, rather than insisting on a precise and perfect weather forecast , which for obvious reasons is impossible.