Alaska is a “weather-poor” state

Alaska pilots are poor (impoverished) when it comes to the amount of weather data available to make critical go/no-go flight decisions. According to the FAA’s surface weather observation stations website, Alaska has 133 AWOS or ASOS weather station locations. In comparison, the “contiguous 48 states” have over 1,800 similar sites. Based on average density of stations nationwide, Alaska would need 183 additional stations to be on par with the rest of the country. That is 2.4 times as many observations as we have today. I am not expecting to see that number of conventional stations in Alaska, but it does point to the need for Alaska pilots to be creative, weather-vigilant, and look to non-conventional sources of information. But first, let’s dig a little deeper into our weather observing system of today.

An overview of the over 1,800 aviation weather stations providing data for pilots and forecasters across the "contiguous 48 states"

An overview of over 1,800 aviation weather stations that provide data for pilots and forecasters across the “contiguous 48 states.”

At approximately the same scale as the previous map, note the density of aviation weather stations providing coverage for Alaska.  Some 180 more stations would be needed to provide a comparably dense network to that enjoyed by the rest of the country.

At approximately the same scale as the map above, note the density of aviation weather stations providing coverage for Alaska. Some 180 additional stations would be needed to provide a comparably dense network to that enjoyed by the rest of the country.

Not all weather reports are equal
Not all weather observations are the same quality. The standard weather observation today is an unattended FAA Automated Weather Observing System (AWOS) and its National Weather Service counterpart the Automated Surface Observing System (ASOS). These devices operate 24 hours a day, and report weather based on sensors that measure wind speed and direction, temperature, dew point, altimeter, ceiling and visibility. Some models may detect precipitation type and accumulation and/or thunderstorms. Advanced as they may be, the unattended stations have some significant limitations. The ceiling is measured using a small laser beam directly overhead while a computer calculates the cloud cover based on a 30 minute average of readings. If, for example, a low fog bank is creeping up on the airport, the unit won’t know about it until the field has gone IFR. Another well-known limitation of these devices is the visibility sensor, which measures the particles within a 1 meter beam of light, and calculates the “up to 10 miles” visibility value we see in the reports. A frustration with this sensor at rural Alaska airports results when a four-wheeler parks next to the sensor (perhaps waiting for an arriving aircraft) and its exhaust drifts into the visibility sensor’s “view,” reducing the reported visibility to 1/8 mile. It’s a mere annoyance to most pilots flying under Part 91, but a commercial pilot flying under Part 135 regulations can’t even shoot the approach with reported conditions lower than the allowable minimums. Automated stations operating unattended contain the word AUTO in the report to alert pilots to that fact. The omission of that term lets the pilot know that either a human is making the observation in the first place, or the observation is being augmented by an observer.

Augmented Weather Stations
Given these limitations in automated stations, the FAA has contract weather observers who augment the equipment at select locations. Airports with significant volumes of traffic, such as Anchorage and Fairbanks, are augmented. One of our Alaska adaptations has been that when the network of Flight Service Stations was reduced in the mid-1990’s, locations that were identified as important strategic locations were provided with a contract weather observer to ensure that the known limitations of automated units didn’t catch a pilot off guard. In the summer of 2013, the weather augmentation contract at Gulkana was cancelled. I am concerned that in the interest of budget reductions, other stations may be on the chopping block.

Five more weather stations closed
One of the means of collecting weather information at remote locations that don’t have an automated station is to contract with a local resident using the A-PAID Program. Under this program, an interested person is trained and certified by the NWS to make a set number of weather reports per day the old fashioned way—by looking at the sky and making manual observations, such as using the distance to local landmarks to estimate visibility. A-PAID observers don’t report 24 hours a day, and if the observer has to travel, is sick or otherwise not available, no report gets filed. A-PAID observers also don’t file Special reports to alert pilots when conditions change, but often they are the only source of weather information in remote areas, or along VFR routes, that help pilots make informed decisions on whether to initiate a flight. A few days ago I learned that the FAA had cancelled the contracts for the last five stations that they had funded for years, leaving us with no weather reports from Farewell Lake, Merrill Pass West, Manley Hot Springs, Nabesna and Chandalar Lake. Five more points, of our already sparse weather network, went dark.

Replacements for A-PAID stations
In 2011 the National Weather Service announced its intention to phase out the A-PAID program, and for those areas that they felt they needed continued observations, replace them with an automated observations similar but not identical to AWOS units. The package they selected is called a Modular Automated Weather Station (MAWS). It is built by a company that makes AWOS systems, and the sensors used are all certified by FAA for use in an AWOS system. MAWS stations record the main elements we need for aviation weather, including ceiling and visibility, but don’t have a VHF radio to transmit the data to an aircraft. They are not certified by FAA as an AWOS, and cost about half as much as a fully certified unit. Due to the lack of certification, at least so far, the FAA and NWS haven’t been able to agree on a basis to consider the reports as METARs, and distribute them through the normal FAA weather channels. This is a real problem for John & Suzy Q Pilot, because unless they know exactly where to look, these observations don’t exist. To date these stations have been deployed in the Central/Circle Hot Springs area, Healy and at Whittier. AOPA and other Alaskan aviation groups are pushing both NWS and FAA to find a way to distribute these observations through the normal channels, given that they are intended for use at VFR airports, or non-airport locations along key VFR routes. Given the lack of progress solving this issue between two federal agencies, we have asked Senator Begich, who sits on the Senate Commerce, Science and Transportation Committee, for help getting the two agencies to come up with a practical solution to this issue to make the observations available. We will continue to push to make this weather fully available to pilots.

Weather Cameras
The FAA Weather Camera Program is the one bright spot that adds weather information to a pilot’s flight kit when it comes to making go/no-go flight decisions. A set of cameras looking multiple directions, updated every 10 minutes, available on the internet from 221 locations across the state provides a tremendous amount of information for flight planning and decision making. As just one example, the camera at the McKinley Park airstrip is co-located with the AWOS unit there. The camera has helped me “interpret” the AWOS report, which one morning was reporting 1/8 mile visibility. A look at the weather camera revealed blue skys in multiple directions with a few wisps of ground fog in the foreground.  This image let me know I was good to launch for a flight thorough Windy Pass. On another occasion, while the AWOS was reporting “clear below 12,000,” a look at the big, ugly, towering cumulus clouds both to the north and south of the station let me know that this was not a good time to expect smooth sailing through the mountains. The station at the airstrip is located in the of the valley between two sets of ridges, which are often where the clouds form, outside the “view” of the AWOS cloud sensor.

As valuable as the network of cameras is, there is a very serious limitation. Currently, the cameras are good during daylight hours only. Great in the summer, but as days shorten, pilots are back “in the dark” having to make go/no-go decisions before camera observations are available. Even in mid-October a local pilot told me this past week he had to wait until 10 a.m. to get a usable image from an interior camera to tell if he could conduct a flight down the Tanana and middle Yukon Rivers. And we aren’t yet into really short winter days! There are now low-light level cameras on the market that might extend the utility of the camera network, however we need a serious research and development effort to evaluate available sensors, and consider the human factors of how to present other than standard color video data for pilots to use in their decision making process.

Alaska forecasts also have limitations
The sparse network of weather observations impacts pilots in more ways than one. In addition to our own weather interpretation, the NWS forecasters are a major consumer of surface observations. They count on them to make and verify the Area and Terminal Forecasts that we use to anticipate what conditions will be like in a few hours, along a cross-country route of flight. Or how fast a weather system is approaching that will impact even local operations. At a recent conference a map was presented (see below) showing how the Alaska weather forecast areas correspond to a similar size area “outside.” NWS forecasters in three weather offices (Anchorage, Fairbanks and Juneau) turn out forecasts for areas that would be covered by 68 forecast offices in the lower 48. Even if you discount the marine areas, the three forecast offices are covering an area equal to 30 offices down south. Another way to look at it is that about 50 forecasters in Alaska issue products for an area that is covered by about 400 forecasters “outside.”

The three NWS forecast offices in Alaska cover the an area that overlays 68 forecast areas in the middle of the country.

The three NWS forecast offices in Alaska cover the an area that overlays 68 forecast areas in the middle of the country.

The spatial granularity of Alaska products is also different. Just looking at the winds-aloft product,  Alaska forecasts are reported using a 90 kilometer grid in contrast to a 30 kilometer grid used elsewhere in the country. Pilots flying in Alaska have to bear in mind that while the forecast products look the same across the nation, the informational content of our forecasts are lower than if we were planning a route across other portions of the country.

From this flight planning program screen shot, one can see the difference in density between winds aloft forecast values generated for Alaska versus the rest of the country.

From this flight planning program screen shot, one can see the difference in density between winds aloft forecast values generated for Alaska versus the rest of the country.

We need all the observations we can get
Given the size of Alaska, our reliance on the airplane to provide basic transportation, the diversity of terrain and climate, and paucity of emergency landing areas, we need all the weather observations we can lay our hands on. The lack of conventional weather stations enjoyed by pilots in the rest of the country means that:

a)      We need to continue to have augmented weather in key regional locations.

b)      It is essential that observations from lower cost MAWS sites are fully distributed.

c)      We need fully certified AWOS units at airports with instrument approaches.

d)     It is important to expedite research into expanding the use of weather cameras beyond daylight hours, to obtain better utilization of this innovative program.

AOPA is working with the Alaska Airmen’s Association, Air Carriers Association and the Alaskan Aviation Safety Foundation on these issues, and is engaging both the FAA and National Weather Service to express our concerns. We addressed the Senate GA Caucus meeting held by Senator Begich last spring and have also asked for the help of the entire Alaska congressional delegation. In the current budget climate it will not be an easy sell, but for aviation safety and access, we must make the effort.

Experimental Winds Aloft graphic for Alaska

As pilots, we are very interested in the weather.  An early lesson one gets while learning to fly is not to put total faith in weather forecasts.  I believe it was President Reagan who made famous the phrase– trust, but verify. That certainly applies to forecasts and flying.  For the last year-and-a-half AOPA has been working with our friends at the National Weather Service in Alaska to bringing together groups of seasoned pilots from different parts of Alaska to sit down with forecasters and have a discussion about aviation weather needs, primarily focused on VFR flying.  Questions asked in these sessions typically start with, “What route do you fly to get from Fairbanks to Eagle?” followed by, “Where along that route do you encounter adverse weather?”  A lively discussion regarding the nature of the weather conditions normally follows.

Don Moore manages the Alaska Aviation Weather Unit, located on Sand Lake Road, just south of the Anchorage International Airport, and has led these discussions.  After listening to pilots describe some of the conditions that plagued them, he pulled up an experimental forecast product the weather service is working on, and asked if we thought it might be helpful.  Following a look at the product, heads started to nod around the table.  A few weeks later, an experimental winds aloft forecast was added to the AAWU website, and is available for pilots to use.

sample winds aloft graphic 1

Sample output from the experimental product, showing winds at 6,000 feet for the 12 hour time period. Users can select the altitude, set through time periods, and toggle features on and off.

This product is based on a computer model, but has finer resolution in time and space than current products we are used to seeing.  The arrows indicate the direction of the wind at an altitude selected by the pilot, but the intensity is displayed as a color.  Temperature is also displayed as a contour line, with its own color scheme. The legend at the bottom provides the color codes for each feature.  Several details about this product are worth noting:

1)      The user selects the altitude at the top of the page
2)      The tabs across the top allow you to step through different forecast periods
3)      The + and – symbols on the top left corner of the image allow you to zoom in (only one step, currently)
4)      The + symbol on the upper right edge of the product lets you toggle features on and off (click to expand)
5)      The color patches represent the area forecast for each wind speed, the vectors merely show direction.

Please give this product a try.  You will find this graphic by clicking a link at the bottom of the Winds Aloft page on the AAWU’s website (see yellow arrows, below).

page to find experimental productThis product is still in development.  For now, the National Weather Service would really appreciate receiving pilot reports to help validate this product, as well as their other forecasts.  So when you are headed out to fly, please take a few minutes and file PIREPs enroute, including an estimate of the winds aloft.  Remember– trust, but verify!