How aviation weather forecasts are written

By Adam Hunt

Our November Flight 8 meeting was presented by Merv Jamieson, who is with the Meteorological Service of Canada (MSC), which is part of Environment Canada (EC). Jamieson is part of the Ottawa team that manages the contract with Nav Canada to write aviation weather forecasts.

Until 2008 he worked as an aviation weather forecaster in Edmonton, writing forecasts for the west coast, arctic and the prairies.

Historically aviation weather forecasting was part of the Department of Transport, now called Transport Canada (TC). Then in 1971 weather forecasting was moved to the Department of the Environment (now EC) under a memorandum of understanding with both the Transport and Defence departments. This continued until Nav Canada was formed in 1996 and MSC started providing the new organization with contract weather services, while continuing to write forecasts for the Department of National Defence as well.

In a bid to reduce costs MSC consolidated their regional operations into a single forecast centre, with two offices, in 2003.

Jamieson started by explaining that MSC forecasters work under contract to Nav Canada to produce terminal (TAF) and area (GFA) forecasts as well as put out SIGMETs, AIRMETs, Sigwx and Local Graphical Forecasts.

Forecasters work at the Canadian Meteorological Aviation Centre, which has its two offices located in Edmonton and Montreal.

The Edmonton office writes forecasts for western Canada and the arctic, while the Montreal office produces forecasts for central and eastern Canada and the North Atlantic.

At any given time the two offices have a total of 10-12 forecasters on duty, with fewer at night, when some airports do not have TAFs written for them.

The regulations that govern forecast writing include CAR 804.01 which specifies that forecasts shall be written to the standards of the Manual of Standards and Procedures for Aviation Weather Forecasts (MANAIR). There are currently 189 civil airport locations for which TAFs are written, plus 15 military and two other locations. TAFs are required to be under constant review against the actual weather observations to see if the TAF needs amending.

Jamieson explained the process by which a forecaster writes a TAF. In general this requires assessing and diagnosing the causes of the existing weather conditions and then extrapolating these, writing the forecast from the existing weather conditions, plus the Numerical Weather Prediction (NWP) model predictions.

The Canadian NWP is one of many weather models available, but is optimized for Canadian latitudes and conditions. Presently it allows one kilometre resolution. Climatology data is also useful in writing forecasts as it shows the probability of conditions forming and persisting, based on what it usually has done in the past in that location.

When written, the new forecast has to also be consistent with other forecasts for that area and bordering areas.

Jamieson went on to explain the “Forecast Funnel”, the process of starting with the broadest look at large area weather and then focusing it down to the TAF area around the airport. This is done at three levels: hemispheric, synoptic and meso scales.

At the hemispheric level the forecaster looks at large scale effects, like the jet stream location and strength, using tools like the pressure charts compiled from balloon soundings and jet aircraft automatic reporting about the upper atmosphere.

At the synoptic level they consider medium scale factors, like the location of highs and lows. Finally, at the meso scale level they look at tools like weather radar, lightning, METARs and the NWP model predictions. Sometimes “ensemble forecasting” is employed, using multiple models or one model with differing starting conditions. Information from both Canadian and U.S. weather radar can be used in border areas, although with its 5.5 cm wavelength the Canadian radar is more effective in showing mixed precipitation, while the U.S. 10 cm radar shows thunderstorm development better.

At the present time TAFs are fully hand written, but there is work being done to automate TAF production, which will free up forecasters to spend more time on severe weather instead. TAFs are assessed for their accuracy, using performance measurement statistics. These auto matic reports provide feedback to the forecaster based on ceiling and visibility, although wind direction and speed is in the process of being added. The tool calculates reliability indices and false alarm rates, which forecasters can then use to improve their accuracy.

Graphical Area Forecasts (GFAs) are written every six hours and consist of six panels, three for clouds and weather and three for icing and turbulence panels. At the present there is no means to assess the accuracy of the GFA, largely due to the graphical nature of the forecast.

Jamieson concluded his presentation with a look at changes coming to the SIGMET and AIRMET formats which will bring Canadian use in line with ICAO standards. ICAO identified in 2003 that Canada was not complying with its standards and recommended practices. In 2008 EC, along with TC and Nav Canada agreed to start using the ICAO format by November 2013.

The AIRMET and SIGMET changes will result in using the same format for both reports, as ICAO does. It will also change the areas covered to use Flight Information Regions (FIRs), in place of the current GFA boundaries used.

The text order will also change, putting the affected area further down the report, although all the existing details will be retained. Included text will still link the reports to the associated GFA. The new format will provide locations of interest in both lat/long and bearing and distance from aerodromes, which is more useful to aircraft in flight.

Between now and November 2013 the new AIRMET and SIGMET formatting will be publicized in COPA Flight as well as in the AIM and through the issue of AICs as well. TC will make changes to the pilot training program and EC will amend MANAIR to complete the documentary requirements.