HOUGHTON - When snow fell early and hard in December, the number of inches immediately became a topic of conversation among Keweenaw residents, who for years have compared snowfall totals from winter to winter.
High numbers confirm a sense of Yooper pride for having survived the shoveling and winter roads, and offer concrete evidence to draw locals and tourists alike to ski areas and snowmobile trails. They've even brought Hancock national attention as the Weather Channel's third snowiest city in America.
But while the tools of the trade are simple, recording and tracking accurate data remains challenging, says meteorologist Mike Dutter from the National Weather Service Negaunee Township office.
Dan Roblee/Daily Mining Gazette
Mark Osborne of the Keweenaw County Road Commission holds a daily snow total measuring stick and at the commission’s observation site near Phoenix. Behind him is the stake that measures accumulation on the ground.
"We have pretty strict standards of how to do snow measuring, but it's still more of an art than a science," Dutter said. "Sometimes when snow is blowing it's real hard to take good measurements."
Dutter said the NWS provides its official Cooperative Observers with the equipment they'll need - a thermometer, rain gauge, oversized ruler and board that goes on a table or above existing snow to serve as a daily measurement base. NWS also provides training materials and quality controls to check for clerical errors or unreasonable reports, but it doesn't regularly check up on observers and their methods.
The organization also collects and uses data from many non-official observers, who provide their own gear.
"A lot of it is based on who wants to volunteer," Dutter said.
Looking at data historically brings more challenges when paper records and memories combine data from different weather stations as recorders of record change over the years.
When Debbie Pindral retired from taking measurements near her home in Painesdale two winters ago - her 30-year-old son Adam had started the family project as a grade schooler - the Mining Gazette switched from listing her readings to represent Houghton County to using readings from the Michigan Tech Keweenaw Research Center near the Houghton County Memorial Airport.
Many residents have since noted lower Houghton County totals in winters they felt were equally harsh, and Painesdale's location in the widely acknowledged snow belt that surrounds the Range towns backs up that intuition.
Dutter said the NWS avoids that trap by gathering data from several different sites within a county and not giving any of them an official designation.
The KRC faces unique difficulties due to its weather station being located in a very windy area.
Observers there have improved the situation by putting up snow fences around their snow measurement table, but they've also posted a disclaimer along with daily totals on their web site stating that "on days of high winds the snow readings recorded here are significantly lower than actual snowfall."
Snow accumulation totals present even more of a problem. At one point in late December the KRC was reporting over 50 inches of snow on the ground - an unusually high number, especially when listed alongside about 74 inches of total snowfall. At the same time, Keweenaw County was reporting just 36 inches on the ground for over 140 inches fallen.
In the past week, the snow on the ground totals on the KRC's site were adjusted downward retroactively, all the way back to mid-December.
KRC's Mark Osborne, who's been doing the hands-on data collecting and reporting for the past few years, explained in an email that their station's windswept location makes accurate on-the-ground readings nearly impossible.
"No matter where we put the stake the readings for snow depth on the ground are very low," he wrote. "I have seen where we show 12 inches of snow on the ground stake and I could walk in the woods on our test course that is wooded enough to be protected by trees and we could measure 36 inches."
So, for the last few years he's been "guesstimating" on-the-ground totals, he wrote.
This year, when snow was falling hard and fast in December, he didn't take into account the snow's settling until later in the month. When his KRC colleague Bob Baratono, who oversees the weather station, passed along comments that the on-the-ground totals didn't seem reasonable, Osborne used daily temperature readings to estimate settling, then adjusted his figures and re-posted them online.
Currently, the KRC lists 38 inches on the ground, compared to Keweenaw County's 31 inches as of Jan. 12.
Despite the location's challenges, readings taken at or near the airport have been the most commonly used on charts listing annual Houghton County snowfall since the '60s.
Keweenaw County's official snow totals are gathered by the county road commission, and they're what the Gazette uses for its daily page 2A updates. The commission averages data from two sites in Delaware and just outside Phoenix before publishing results, and takes measurements in low-wind areas chosen to meet specifications published by the National Oceanic and Atmospheric Association, a federal organization that works hand-in-hand with the NWS.
But unusual conditions can still require judgment calls and ingenuity, according to road commission assistant engineer Steve Defour, who takes many of the readings.
"If our table's blown clean, I'll find a road nearby that hasn't been plowed yet and isn't subject to the wind and take a reading there," he said.
As for Hancock's claim to fame as the nation's third snowiest city - and snowiest east of the Rockies - that's based on NOAA data taken from various stations and accumulated by the National Climactic Data Center.