fog at airports

Pretty good point here.

Autoland is not the same as “land without seeing anything”. Though lots of laymen and dumb-ass news reporters confuse or link the two ideas.

There absolutely is “Autoland with no requirement to see anything prior to auto touchdown”.

But much more common is “Autoland but with a requirement to spot the runway a few seconds before the auto-touchdown and intervene if necessary at that point”. There’s also “Autoland to practice the procedures and exercise the equipment on a nice sunny day.”

In the case of the first option, there is still a requirement for a certain amount of visibility. Otherwise once HAL landed the jet there’d be no way for the humans to taxi off the runway. There would also be some challenges getting slowed properly when you can’t see the end of the runway until you’re passing it.
But …

Bzzt. As a fraction of all landings it’s not a huge percentage,maybe 2 or 3%. But every aircraft needs to be exercised every month or so, and every crew needs to practice every so often as well. For a short-haul airplane that flies 8 times per day, a month is 240 flights with at least one practice autoland. For a long-haul aircraft that flies 1-1/2flights per day, a month is 45 flights with at least one practice Autoland. So the percentages can differ a bunch.

Except in the worst weather part of the year, typically Jan & Feb in the US, practice autolands far outnumber real bad-weather ones.

My old house is across the river from NAS Jacksonville (map). 3 years ago I went out there (neighbors still let me visit), and the entire airport was embedded in a fog bank, tho it was clear on my side and clear above 200 feet.

Note the intersecting runways at the eastern end. I see one plane emerge from the fog (a Navy airliner) from runway 090-THEN a P-3 from runway 300. The P-3 immediately banked hard to the left to avoid what would have been an insane collision-missed each other by what I would estimate to be no more than 100 feet.

Bolding mine.

I fly business jets, and just the other day I had a passenger ask this very question:

Me: “I’m sorry, we can’t take off yet. There’s fog down to the ground at our destination.”

Him: “But… but… don’t you have radar!!!”

Yes we do, but onboard radar has f@$& all to do with landing the plane. It’s for weather avoidance. Radar from the ground can be used for approaches, but that’s not a typical procedure in civilian aviation.

Yeah, radar isn’t some kind of magic fix for every aeronautical problem we have. It is one of many tools that are available to the aviation system. In a nutshell, on-board radar is for weather avoidance, and ATC radar is for collision avoidance.

So far most of the discussion has focused on landing. Now for the other half …

Foggy weather also slows down the rate of takeoffs at an airport. When everybody is having a hard time seeing we just need to do everything more slowly and deliberately with more space between aircraft.

If the fog gets thick enough then takeoffs stop completely. The exact degree of “thick enough” depends, as with landing, on the airport equippage, aircraft type and equippage, and crew certification. Broadly speaking, takeoff and landing limitations are similar. So if a particular combination of airport, aircraft, and crew can legally land, they could also legally take off under the same conditions. Generally speaklng; there are always exceptional cases.

Takeoff is a completely manual process and it’d be very hard to remain aligned with the runway if you can’t see past your hood ornament. Running off the side of the runway at 130+ mph would really suck. Plus if an engine fails during takeoff, the airplane really wants to swerve pretty violently. Correcting for that and staying aligned with the runway would be very difficult if you can’t see what you’re doing.
Obviously aircraft in the air need to land fairly soon; nobody carries infinite fuel. As such when weather gets marginal and the total rate of permissible take offs and landings goes down, there’s a natural tendency to prioritize landings over takeoffs. At least until the airborne backlog is dealt with.

Which also means each foggy airport is slowly filling up with aircraft. Gridlock at the terminal or gridlock on the taxiways can happen. And is a slow and frustrating mess to clean up.
All in all, dense fog makes for a long and frustrating day for everyone. Snowfall also causes very low visibilities, plus all the extra work to keep the runways, taxiways, and ramps clean enough to use. Plus the delays to de-ice aircraft before take-off. Gaah, I love winter! Not.

I did think that list was oddly short, but I did pull it directly from the FAA. I wonder if its a list for a very specific set of conditions/aircraft. There is a Cat IIIc but from what I remember no one actually uses it because when visibility is that bad (zero DH, zero everything!) you can’t even see the follow me truck and thus would not be able to get off the runway safely. I don’t know if that’s changed any.

Here is another link I found that lists CAT III approaches that looks to be a lot more extensive:

Here’s that FAA list I found yesterday:

https://www.faa.gov/about/office_org/headquarters_offices/avs/offices/afs/afs400/afs410/status_lists/media/Restrictedairports.xls

That list at your first link feels about right to me. That second list is of airports that have low vis approaches that are abnormally *more *restrictive (= *less *capable) than typical. So it’s a small subset of the total list of low-vis approaches.

As I explained for the laymen in post #21, autoland is not synonymous with zero visibility. We autoland out of most Cat II approaches with visibilities of around a quarter mile or slightly less (down to RVR 1000/600/300).

We can do this at darn any airport with scheduled big jet service and most places with small jet service other than small towns in the desert southwest where it’s just not needed.

The old distinctions between Cat III a, b, and c are now obsolete. There got to be enough exceptions and but-ifs that they simply relabeled the whole lot as plain old “Cat III” but with differing minimums based on local circumstances. You’ll still see the older labeling on charts that haven’t been recently revised.

Getting back to the OP, I think a lot of us seem to be dancing around the same answer.

Assuming that during fog we’re using only ILS (Instrument Landing System) approaches*, there are three main styles: Category I, II, and III. Cat I approaches are good down to 1800’, Cat II are used for less than 1800 but more than 1200, and Cat III are good from less than 1200 to as low as 150’.

The problem is that the components on the ground to support the ILS (a localizer, glideslope, and DME) are expensive to operate and maintain. The more precise the system is, the more expensive it is to run. Dorjän’s first link gives a list of all of the Cat III ILS systems in the US. Notice that there aren’t a lot of them - an ILS costs money, and no one wants to spend it unless they have to. Boston-Logan only has one Cat III (Runway 4R), even though it’s a foggy airport. The same with Providence** (Rwy 5). There are other ILS approaches at those airports, but they are only Cat I systems. Even so, not all of the runways at an airport even have an ILS approach. Boston-Logan has no ILS at all for 4L/23R, nor Rwy 16 at Providence.

Boston-Logan makes a great example. Normally, during calm winds, there are three runways in use: 4L, 4R, and 9. Takeoffs are on 9 (and sometimes 4R for heavies), landings on 4L and 4R. If you can land an aircraft every 2 minutes on each runway, that gives you an acceptance rate of one per minute (2 mins / 2 runways = 1/minute). But remember, 4L has no ILS, so if it’s foggy, only 4R is getting used. That cuts you down to one arrival every two minutes, which is half of your normal rate. Half of the aircraft landing in the scheduled time creates backups. The rest of the air traffic control system will stop aircraft who are flying to the affected airport from taking off, which called a ground stop (or ground delay). Aircraft that were supposed to be on the ground at Boston are now still on the ground in Newark or Charlotte. If they’re not here yet, the passengers can’t get on, and it’s a delay. Aircraft already in the air can be slowed or put into a hold to wait for their turn in line to land. Using a portion of the normally-used runways creates a bottleneck in the system, which creates delays.

Snow does the same thing. The airports can’t plow everything at once, so the primary runways get priority. This reduces the acceptance rate, which backs everything else up.

The FAA has an online tool to show what the current problems are in the ATC system. If you go to National Airspace System and click on “NAS Status,” it will display the airports currently having issues - updated every 5 minutes. The “AAR” column is the Airport Acceptance Rate, which will drop for low visibility, storms, or snow/ice.

  • There are many runways that have an RNAV (GPS) approach, which doesn’t use any equipment on the ground and is much cheaper. The FAA really likes RNAV approaches.
    **Full disclosure: I work there