In a February 7 article ("A likely record, but experts will get back to us") Frank Roylance wrote:
"The contractor paid to make snow measurements at BWI for the weather service (the NWS has none of its own personnel there) evidently failed to follow NWS protocols in measuring the snow. Those rules say the observer must allow snow to fall on an official 'snow board' for six hours, then wipe it clear and repeat the procedure every six hours until the snow ends."
This statement is inaccurate and misleading in that it gives the impression that the weather observers (contractors) at Baltimore-Washington Thurgood Marshall Airport did not accurately perform their job in measuring snowfall. The weather observers at BWI have a combined experience on the job of well over 100 years of service with some of the individual observers having worked more than 30 years in the field. These workers know well the correct procedures to follow regarding snow measurement and followed them during the time period in question.
Our company employs the weather observers at BWI. We contract services for the Federal Aviation Administration (FAA) and not the National Weather Service (NWS). The guidance we follow at BWI in making snow measurements is given by the FAA and not the NWS. The guidance is provided in the FAA publication 7900.5. The relevant paragraph states:
"If practicable these measurement shall be made on a surface that has been cleared of previous snowfall. "
There is no further guidance given in the FAA manual 7900.5 on the frequency of measurement, and it certainly does not provide guidance to require only measurements every six hours of snow fall. In fact, the FAA manual 7900.5 sections 11-28 provides no definition regards "previous snowfall" to indicate a time period that "previous" is considered to define.
Therefore, one can measure as many times as needed during a six-hour period, provided the totals for all snowfall during the period are given.
Moreover, the FAA requires observers to keep track of the snow increasing every hour and requires hourly measurements.
The weather observers use panels of wood or plastic called snow boards on which snow accumulates to do their measurements. Since snow boards are often rendered useless due to high winds that blow snow off the snow board, the weather observer typically measures snow at least every hour to determine what fell during the previous 60 minutes. The FAA manual 7900.5 implies that if the observer is measuring snow every hour to keep track of the snowfall that is being lost to the wind that the area should be cleaned and measured every hour. These procedures were followed at BWI.
In fact even the NWS in its snow training program declares that during windy conditions snow must be measured more often to compensate for wind loss. As such the weather observers at BWI in fact followed NWS protocol. It is therefore inaccurate to claim that in every instance the NWS policy states that snow must be measured only every six hours.
During the dates of the storm over February 5 and 6, BWI reported a snow depth of 24 inches. Dulles (IAD) reported 21, and Ronald Reagan National Airport (DCA) reported 18. Given their geographic proximity, the snow they reported to have fallen at each location seem to coincide.
While the totals at the airport appear to coincide, however, the report of the NWS at Sterling, Va. of 32 inches of snowfall that left only 23 inches of snow depth on the ground (they had 2 inches snow depth on the ground before the storm for a net gain of 21 inches of snow depth) seems to vary from acceptable criteria when compared with the totals of IAD -- whose office is just across the airport from Sterling.
IAD had 20 inches of snowfall and had a snow depth of 21 inches of snow on the ground after the storm. (They had 1 inch of snow on the ground before the storm, a net gain of 20 inches). The totals at IAD which indicate that they added 20 inches of snow depth for 20 inches of snowfall seem much more consistent than the NWS Sterling that reported 32 inches of snowfall but only an increase of snow depth of 21 inches.
It is therefore odd that the NWS would also accuse the weather observers at IAD of providing erroneous snow fall measurements during that storm because they were measuring the snowfall more often than once every six hours because the NWS snow training guidance states that hourly snow measurements could erroneously increase the estimated snowfall, not decrease them. Therefore, if IAD measured snowfall every hour -- according to the NWS assertions -- they should have had a snowfall total much greater than reported by the NWS in Sterling and not in fact more than 10 inches less than the NWS at Sterling reported.
Richard Carlson, Port Angeles, Wash.
The writer is vice president of Pacific Weather Inc.