Forecast: Caught out by stratus

South African Weather Service.

Moderator: Moderators

User avatar
Too Tousand
Too Tousand
Posts: 2223
Joined: Thu Jul 21, 2005 10:12 am
Closest Airfield: Port Elizabeth
Location: Port Elizabeth
Has liked: 0
Been liked: 0

Forecast: Caught out by stratus

Unread post by FAPE » Sun Jan 08, 2012 11:11 am

On Friday evening 6 Jan 2012 I worked the afternoon shift. 5 pilots from the same flight school visited me during the afternoon to plan their night navs at 18Z from PE to Grahamstown. The afternoon satellite imagery (black and white image) showed no cloud enroute. In fact the closest cloud was stratocumulus about 100km eastward in the East London area (16Z).

Within 30 minutes past 15:00 SAST, the first RGB images ('Night Time Fog') became available (bluish images). These images are updated for SAWS forecasters every 15 minutes. The 17Z image shows a small trace of stratus along the coast midway between FAPE and FAGT. As the take-off time was 18Z, I was not too concerned as the full moon conditions and the clearer conditions just inland would have still made the flight possible, given the evidence at my disposal.

The 18Z image showed a different picture with the stratus now sitting in valleys enroute while spreading eastwards along the coast to Port Alfred. By now it became clear that the pilots needed to divert slightly north of their intended direct track to make it to Grahamstown. Conditions became worse by 19Z as the stratus started to occupy large areas of the route. My forecast had gone horribly pear shaped. All 5 aircraft turned back and landed safely at PE airport.

I share this experience as some valuable lessons can be learned from it:

1. In this case even armed with satellite images updated every 15 minutes, I did not forecast the low cloud enroute. I put too much trust in the late afternoon visible images which showed crystal clear skies. In addition to that the PE dew point remained a conservative 14C by 17Z with the dew point spread a safe 8C. This lead to an oversight of the drop in dew point spread for both Grahamstown and Port Alfred. This was available to me on the hour. I would have been able to prevent those blokes from taking off at least 1 hour before their planned departure, but not seeing those dew points (for whatever reason) robbed me of that opportunity. So as a lesson to myself: Check every bit of evidence all the time. Do not get side-tracked by good-news data.

2. Was this preventable? Yes.

3. Would I have been able to do a proper forecast without the satellite tools and hourly data of SAWS automatic stations? No. The key to overcoming similar obstacles in the future is to monitor dew point hourly. Sticking to the synoptic chart analysis routine, forces me to look at the data 3-hourly - simply not good enough. Another issue is the fact that all SAWS data must be WMO compliant. The data is therefore checked and calibrated at about 60 day intervals and all stations meet the international standards. Websites displaying weather station data which is not linked to a similar WMO standard of calibration is far less reliable in terms of temp/dew point data.

Moral of the story:

I walk away having learned something of my own imperfection but am at the same time reminded of the power of the tools at my disposal.
You do not have the required permissions to view the files attached to this post.
Deon van der Mescht

Return to “SAWS”