segunda-feira, 19 de abril de 2010

Arctic heat

Following the fascinating story of the missing "M" minus sign on METAR data, by Anthony Watts, I spent yesterday afternoon digging some interesting Arctic weather data. It seems that wherever you dig, you find something interesting.

Take Alert Airport, in Nunavut, Canada, for instance. It's part of the Global Historical Climatology Network (GHCN). The temperatures for April 2, 2010, were clearly below -20ºC. But slightly after 6am, surprise! In comes a zero value, and temperature skyrockets, as can be seen in the graphic below, with a special SPECI message. After three hours, cold temperatures resume, but that day should have established a new maximum day temperature, with 0ºC.

Switching to Greenland, where ice is said to be melting, we find more interesting hot temperatures. Take Qaanaaq, and the first week of last February. As can be seen in the following graph, some good warming occurred between late February 3 and early morning next day. From the METAR data, in the table below, it doesn't seem to be a normal "M" missing problem. Even if it was an M2 or M3, it would still be strange...

METAR BGQQ 040050Z AUTO 05001KT //// NCD M11/M11 Q1007
METAR BGQQ 040150Z AUTO VRB02KT //// NCD M10/M10 Q1006
METAR BGQQ 040250Z AUTO 13025KT //// NCD 02/M05 Q1007
METAR BGQQ 040350Z AUTO 13026KT //// NCD 02/M05 Q1007
METAR BGQQ 040450Z AUTO 11027KT //// NCD 02/M06 Q1008
METAR BGQQ 040650Z AUTO 11017KT //// NCD 02/M09 Q1008
METAR BGQQ 040750Z AUTO 10017KT 080V140 //// NCD 03/M09 Q1008
METAR BGQQ 040850Z AUTO 11017KT //// NCD 03/M10 Q1009
METAR BGQQ 040950Z AUTO 11016KT //// NCD 03/M10 Q1010
METAR BGQQ 041050Z AUTO VRB04KT //// NCD M08/M09 Q1011
METAR BGQQ 041150Z 28002KT 9999 SKC M11/M11 Q1012

Going back to Nunavut, in Canada, we can look at Cape Hooper temperatures, for February 23, 2010. Now, this was a quite cold day, but something interesting happened slightly after lunch time. Temperatures soared by more than 10ºC, during only one hour. But wait a minute: the graph has two different temperatures for the same hour! How can that be? You have to take a look at the raw data to find out that the METAR and AAXX data don't agree! Looking at the other data at Weather Underground, it seems that it also got quite windy during that period. Might it be an Arctic tornado, or, who wonders, simply a plane taking off?

METAR CWUP 231500Z AUTO 03005KT M19/ A3015 RMK SLP260
METAR CWUP 231600Z AUTO 06007KT M19/ A3013 RMK SLP256
METAR CWUP 231700Z AUTO 08008KT M20/ A3012 RMK SLP252
AAXX 23184 71093 460// /2541 11073 39510 49999 52004 6///1
METAR CWUP 231800Z AUTO 14003KT M18/ A3012 RMK SLP252
METAR CWUP 231900Z AUTO 12005KT M17/ A3012 RMK SLP251

While METAR and AAXX seem to agree most of the time, there are many other situations where they disagree by much more than one degree. Don't really know why, but here is another example. From Longstaff Bluff, also in Nunavut, the following is an example of February 23, 2010.

AAXX 23154 71091 45/// /0921 11111 39946 40158 53009
METAR CWLX 231500Z AUTO ///// M11/ A2993 RMK SLP158
METAR CWLX 231600Z AUTO ///// M10/ A2993 RMK SLP155
METAR CWLX 231700Z AUTO ///// M09/ A2994 RMK SLP158
AAXX 23184 71091 460// ///// 11055 39857 40063 55003 6///1
METAR CWLX 231800Z AUTO ///// M08/ A2995 RMK SLP163
METAR CWLX 231900Z AUTO ///// M09/ A2999 RMK SLP176
METAR CWLX 232000Z AUTO ///// M10/ A3001 RMK SLP183
AAXX 23214 71091 45/// /1711 11113 39980 40192 51028
METAR CWLX 232100Z AUTO ///// M11/ A3004 RMK SLP192

Now, these are examples of nice places in the Arctic, where you can get a little warmth once in a while. But if you are interested in a nice dark place, all day long, but very warm, you should take a look at Phippsoya, in Norway. It's way up in Svalbard, and looks like a hidden treasure on Earth. Polar bears seem to like it, maybe because of Global Warming. March had a mean temperature of 18ºC, and a maximum of 33ºC!!! Almost as good as Portugal in the Summer!