Dallas Fort Worth airport is one of hundreds of GHCN reporting stations gone missing. GHCN stations are becoming an endangered species.
2010 Thermometer Langoliers Hit List
Guest post by E.M.Smith
Well, They Are At It Again
UPDATE 12 Feb 2010: Dallas has been found in the Failed QA file. Also, the “process” was discovered described in the ClimateGate emails by Phil Jones (see comments). The final update date is ill defined at best and this list ought to be treated as a “missing in action” list until the end of the month as we wait to see if further updates are applied. -E.M. Smith
Don’t know what to make of this list yet, other than it directly ‘gives the lie’ to the assertion that thermometer ‘drops’ were / are entirely an artifact of GHCN being a creation at a historical moment in time (i.e. made in 1990’s era so that’s why they drop out then in The Great Dying of Thermometers – which itself ignores The Lesser Dying in 2006).
It also shows that the excuse of things being dropped for not electronically reporting is pretty much a lie, too. I note that Dallas Fort Worth Airport is on this list and I’m pretty sure they have electronic reporting… From the NASA / GISS web site, as confirmation:
(*) Dallas-Fort W 32.9 N 97.0 W 425722590000 4,037,000 1947 – 2009
Note the end date of 2009.
And Strasbourg airport is on the list too, so it’s not just an America thing…
I’ve not examined this list for any patterns, nor re-done any of the prior “by latitude” and “by altitude” reports to see what the changes do to the world. For now, it’s just another “Dig Here” list. (Though a casual look at the altitude field shows a fair number of 1000m and 2000m stations died.)
Oh, and the list is also confirmation that the extraordinary hatred of thermometers shown by the managers of GHCN continues, unabated. Particular emphasis seems to have landed on Africa (already poorly covered) and Asia, with a modest effort to eradicate more of South America. By comparison, Europe is only slightly mauled…
[chiefio@Hummer data]$ wc -l 2009_uniq_station_list
1597 2009_uniq_station_list
[chiefio@Hummer data]$ wc -l 2010_uniq_station_list
1113 2010_uniq_station_list
So from 1597 we drop to 1113. That’s a drop of 30%.
Just shy of 1/3 of the stations, taken out back and shot this year.
But once you decided that you can just make up any missing data, then who needs to actually read the thermometers any more?
Ok, enough of my complaint. Here is the list. If anyone notices anything interesting about their part of the world, feel free to let us all know. Remember that the StationID (that first field) is structured as 1 digit of continent then 2 that tell the particular country, then 8 for the particular station and substation. So records that start with 1 are Africa, 2 Asia, 3 South America, 4 North America, 5 Pacific, 6 Europe, 7 Antarctica, 8 Ships at sea (a very few geographic spots with few records as a ship happens by and reports). I will break up the list into groups by continent, but notice that there are no “8″ stations on the list and only one from Antarctica. Oh, and there were 2 stations added, so I’ll list them here at the top:
Added Stations
11365563000 YAMOUSSOUKRO 6.90 -5.35 213 168U 50HIxxLA-9A15TROP. SEASONAL A 11365594000 SAN PEDRO 4.75 -6.65 30 0R -9FLFOCO 2A-9EQ. EVERGREEN C
Antarctic Deletion:
70089642000 DUMONT D'URVI -66.67 140.02 43 150R -9MVICCO 1x-9ANTARCTICA A
The Requiem List
Africa
10160355000 SKIKDA 36.93 6.95 7 18U 107HIxxCO 1x-9WARM DECIDUOUS C 10160403000 GUELMA 36.47 7.47 227 287S 47HIxxno-9x-9WARM CROPS C 10160430000 MILIANA 36.30 2.23 715 1167R -9MVDEno-9x-9WARM DECIDUOUS C 10160444000 BORDJ BOU ARR 36.07 4.77 928 1051U 57MVxxno-9x-9WARM FOR./FIELD C 10160445000 SETIF 36.18 5.42 1081 1060U 144MVxxno-9x-9WARM FOR./FIELD C 10160457000 MOSTAGANEM VI 35.88 0.12 137 157U 102HIxxCO 4A 2WARM CROPS B 10160468000 BATNA 35.55 6.18 1052 1015U 85MVxxno-9x-9WARM DECIDUOUS C 10160518000 BENI-SAF 35.30 -1.35 68 103R -9HIDECO 1x-9WARM CROPS B 10160535000 DJELFA 34.68 3.25 1144 1071U 51HIxxno-9x-9MED. GRAZING C 10160536000 SAIDA 34.87 0.15 750 802U 62MVxxno-9A 4MED. GRAZING C 10160540000 EL KHEITER 34.15 0.07 1000 1113R -9FLDEno-9x-9WARM DECIDUOUS B 10160549000 MECHERIA ALGERIA 33.60 -0.30 176 1501R -9MVDEno-9A-9MED. GRAZING A 10160555000 TOUGGOURT 33.12 6.13 85 106U 76FLxxno-9x-9HOT DESERT B 10160560000 AIN SEFRA 32.77 -0.60 1058 1811R -9MVDEno-9x-9WARM GRASS/SHRUBA 10160602000 BENI ABBES 30.13 -2.17 499 520R -9HIDEno-9A-9SAND DESERT C 10764910000 DOUALA OBS. 4.00 9.73 9 12U 458FLxxCO 5A 1MARSH, SWAMP C 10964655000 BRIA 6.53 21.98 584 587S 25FLxxno-9A 1TROP. SAVANNA A 11064700000 NDJAMENA 12.13 15.03 295 294U 179FLxxno-9A 2WARM GRASS/SHRUBC 11161968000 ILES GLORIEUS -11.58 47.28 4 0R -9FLxxCO 1A-9WATER A 11161970000 ILE JUAN DE N -17.05 42.70 10 0R -9FLxxCO 1A-9WATER A 11161972000 ILE EUROPA -22.32 40.33 13 0R -9FLxxCO 1A-9WATER A 11264400000 POINTE-NOIRE -4.82 11.90 17 7U 142FLxxCO 4A 2WATER B 11264401000 LOUBOMO -4.20 12.70 330 321S 20HIxxno-9A 1TROP. SAVANNA C 11264402000 MOUYONDZI -3.98 13.92 512 321R -9HIxxno-9x-9TROP. SAVANNA A 11264405000 SIBITI -3.68 13.35 531 481R -9HIxxno-9A-9TROP. SAVANNA A 11264450000 BRAZZAVILLE / -4.25 15.25 316 303U 299FLxxno-9A 2WARM CROPS C 11264454000 GAMBOMA -1.87 15.87 377 314R -9FLxxno-9A-9TROP. SEASONAL A 11365578000 ABIDJAN 5.25 -3.93 8 26U 686FLxxCO 1A 5COASTAL EDGES C 11365585000 ADIAKE 5.30 -3.30 39 25R -9FLxxCO 3x-9WARM FOR./FIELD A 11763331000 GONDAR 12.53 37.43 1966 2002S 39MVxxno-9A10TROP. MONTANE A 11763333000 COMBOLCHA 11.08 39.72 1916 1465U 50MVxxno-9A10HIGHLAND SHRUB B 11763402000 JIMMA 7.67 36.83 1676 1776S 40MVxxno-9A 2TROP. MONTANE B 11763403000 GORE 8.17 35.55 1974 1760R -9HIFOno-9A-9TROP. MONTANE A 11763450000 ADDIS ABABA 8.98 38.80 2324 2586U 1196MVxxno-9A 2WARM CROPS C 11763471000 DIRE DAWA 9.60 41.87 1146 1239U 64MVxxno-9A 2TROPICAL DRY FORC 12263612000 LODWAR 3.12 35.62 515 431R -9FLxxno-9A-9WARM GRASS/SHRUBA 12263740000 NAIROBI/KENYA -1.32 36.92 1624 1634U 509HIxxno-9A 2WARM FIELD WOODSC 12462002000 NALUT 31.87 10.98 621 570S 24HIxxno-9x-9WARM GRASS/SHRUBC 12462007000 ZUARA 32.88 12.08 3 5S 20FLxxCO 1x-9HIGHLAND SHRUB C 12462008000 YEFREN 32.08 12.55 691 575R -9HIDEno-9x-9WARM GRASS/SHRUBB 12462010000 TRIPOLI LIBYA 32.90 13.20 84 9U 550HIxxCO 3x-9MED. GRAZING C 12462012000 EL KHOMS 32.63 14.30 22 12S 17HIxxCO 2x-9WATER C 12462016000 MISURATA 32.42 15.05 32 5U 102FLxxCO 4x-9WARM GRASS/SHRUBC 12462019000 SIRTE 31.20 16.58 14 16S 23FLxxCO 1x-9WATER C 12462053000 BENINA 32.10 20.27 132 152U 287HIxxCO20A16WARM CROPS C 12462055000 AGEDABIA 30.72 20.17 7 6U 53FLxxCO25x-9WARM GRASS/SHRUBB 12462056000 SHAHAT 32.82 21.85 625 475S 17HIxxCO11x-9WARM CROPS C 12462059000 DERNA 32.78 22.58 26 138S 44HIxxCO 1x-9WATER B 12462103000 GHADAMES 30.13 9.50 347 273R -9HIDEno-9x-9WARM GRASS/SHRUBC 12462124000 SEBHA 27.02 14.45 432 424S 35FLxxno-9A 5SAND DESERT C 12462131000 HON 29.12 15.95 267 256R -9HIDEno-9A-9HOT DESERT A 12462161000 JALO 29.03 21.57 60 109R -9FLDEno-9x-9HOT DESERT A 12462176000 GIARABUB 29.75 24.53 -1 45R -9FLDEno-9x-9HOT DESERT B 12462271000 KUFRA 24.22 23.30 436 406R -9FLDEno-9A-9HOT DESERT C 12667693000 CHILEKA -15.68 34.97 767 802U 222MVxxno-9A10TROPICAL DRY FORB 12761214000 KIDAL 18.43 1.35 459 431R -9HIDEno-9A-9SUCCULENT THORNSA 12761270000 KITA 13.07 -9.47 334 381S 18HIxxno-9A 3WARM CROPS A 12761285000 KENIEBA 12.85 -11.23 132 306R -9HIFOno-9x-9TROP. SAVANNA A 12761297000 SIKASSO 11.35 -5.68 375 398S 47HIxxno-9A 2WARM CROPS B 12861421000 ATAR 20.52 -13.07 224 261S 16FLxxno-9x-9HOT DESERT B 12861437000 AKJOUJT 19.75 -14.37 120 134R -9FLDEno-9x-9HOT DESERT A 12861461000 BOUTILIMIT 17.53 -14.68 75 41R -9FLDEno-9x-9WARM GRASS/SHRUBA 13167215000 PORTO AMELIA -13.00 40.50 50 13R -9HIxxCO 2A-9WATER A 13167217000 VILA CABRAL -13.30 35.30 1365 1359S 10HIxxno-9x-9WARM CROPS A 13167237000 NAMPULA -15.10 39.28 441 435S 23HIxxno-9A 1TROP. SAVANNA C 13167261000 TETE -16.18 33.58 150 187R -9HIxxno-9x-9SUCCULENT THORNSA 13167283000 QUELIMANE -17.88 36.88 16 7S 11FLxxCO10x-9MARSH, SWAMP B 13167297000 BEIRA -19.80 34.90 16 6S 46FLxxCO 2A 5MARSH, SWAMP C 13167323000 INHAMBANE -23.87 35.38 15 22R -9FLxxCO 2A-9WARM CROPS B 13167341000 LOURENCO MARQUES/COUNTINHO -25.90 32.60 44 18U 755FLxxCO 7A 4MARSH, SWAMP C 13268312000 KEETMANSHOOP -26.53 18.12 1061 961S 10FLxxno-9x-9WARM GRASS/SHRUBA 13361090000 ZINDER 13.78 8.98 453 445U 58FLxxno-9A 1SUCCULENT THORNSA 13761679000 KAOLACK 14.13 -16.07 7 8U 107FLxxno-9x-9WARM CROPS B 14168262000 PRETORIA -25.73 28.18 1322 1377U 573HIxxno-9x-9WARM CROPS C 14168438000 KIMBERLEY -28.80 24.77 1200 1213U 105FLxxno-9A 3WARM GRASS/SHRUBC 14168588000 DURBAN (LOUIS -29.97 30.95 14 23U 975HIxxCO 2A 1WATER C 14168842000 PORT ELIZABET -33.98 25.60 61 63U 414HIxxCO 4A 1WATER C 14361997000 CROZET -46.43 51.87 143 0R -9HIxxCO 1x-9WATER A 14361998000 PORT-AUX-FRAN -49.35 70.25 30 173R -9HIxxCO 1x-9WATER A 14761901000 ST. HELENA IS. -16.00 -5.70 627 0R -9HIxxCO 1x-9WATER A 14862600000 WADI HALFA 21.92 31.32 126 192R -9FLDELA-9x-9WARM IRRIGATED A 14862640000 ABU HAMED 19.53 33.32 312 231R -9FLDEno-9x-9HOT DESERT A 14862641000 PORT SUDAN 19.58 37.22 2 16U 133FLxxCO 1x-9COASTAL EDGES C 14862650000 DONGOLA 19.17 30.48 226 229R -9FLDEno-9x-9WARM IRRIGATED C 14862660000 KARIMA 18.55 31.85 249 242S 13HIxxno-9x-9WARM IRRIGATED C 14862680000 ATBARA 17.70 33.97 345 288U 66FLxxno-9x-9WARM GRASS/SHRUBC 14862721000 KHARTOUM 15.60 32.55 380 239U 1334FLxxno-9A 1WARM IRRIGATED C 14862730000 KASSALA 15.47 36.40 500 518U 99HIxxno-9x-9WARM GRASS/SHRUBC 14862733000 HALFA EL GEDI 15.32 35.60 451 501R -9FLDEno-9x-9WARM IRRIGATED C 14862750000 ED DUEIM 14.00 32.33 378 261S 27FLxxno-9A 1WARM GRASS/SHRUBB 14862751000 WAD MEDANI 14.40 33.48 408 313U 107FLxxno-9x-9WARM GRASS/SHRUBC 14862752000 GEDAREF 14.03 35.40 599 544U 92FLxxno-9x-9SUCCULENT THORNSC 14862760000 EL FASHER 13.62 25.33 730 758U 52FLxxno-9A 1WARM GRASS/SHRUBB 14862762000 SENNAR 13.55 33.62 418 415S 10FLxxLA-9x-9SUCCULENT THORNSC 14862771000 EL OBEID 13.17 30.23 574 574U 90FLxxno-9A 1WARM GRASS/SHRUBC 14862772000 KOSTI 13.17 32.67 381 367U 57FLxxno-9x-9WARM GRASS/SHRUBC 14862795000 ABU NA'AMA 12.73 34.13 445 379R -9FLDEno-9x-9WARM GRASS/SHRUBA 14862805000 DAMAZINE 11.78 34.38 470 471R -9FLxxLA-9A-9SUCCULENT THORNSA 14862810000 KADUGLI 11.00 29.72 499 523S 18FLxxno-9x-9WARM GRASS/SHRUBC 14862880000 WAU 7.70 28.02 438 433U 53FLxxno-9x-9TROP. SAVANNA A 14963971000 MTWARA -10.27 40.18 113 74S 49HIxxCO 3x-9TROP. SEASONAL B 15061701000 BATHURST/YUNDUM 13.40 -16.70 26 6S 39FLxxCO 6A15COASTAL EDGES B 15165351000 DAPAON 10.87 0.25 330 279R -9HIxxno-9A-9WARM GRASS/SHRUBA 15165352000 MANGO 10.37 0.47 146 136R -9FLxxno-9A-9WARM GRASS/SHRUBA 15165355000 NIAMTOUGOU 9.77 1.10 343 340R -9HIxxno-9A-9WARM GRASS/SHRUBA 15165357000 KARA 9.55 1.17 341 308R -9HIxxno-9x-9WARM GRASS/SHRUBB 15165361000 SOKODE 8.98 1.15 387 419S 30HIxxno-9x-9TROPICAL DRY FORC 15165376000 ATAKPAME 7.58 1.12 402 427R -9HIxxno-9x-9TROP. SAVANNA A 15165380000 TABLIGBO 6.58 1.50 44 67R -9FLxxno-9x-9WARM FOR./FIELD A 15165387000 LOME 6.17 1.25 25 17U 148FLxxCO 5A 1WARM FOR./FIELD C 15567633000 MONGU -15.25 23.15 1053 1023R -9FLxxno-9A-9TROPICAL DRY FORB 15567663000 KABWE -14.45 28.47 1207 1195U 144HIxxno-9x-9SUCCULENT THORNSC 15567743000 LIVINGSTONE -17.82 25.82 986 970U 72FLxxno-9A 2SUCCULENT THORNSA 15761996000 ILE NOUVELLE-AMSTERDAM -37.80 77.50 28 0R -9HIxxCO 1x-9WATER A 15960010000 IZANA 28.30 -16.50 2368 1591R -9MTxxCO12x-9WATER B 16367005000 DZAOUDZI/PAMA -12.80 45.28 7 36R -9FLxxCO 1A-9WATER B 16561980000 SAINT-DENIS/G -20.88 55.52 25 239U 80HIxxCO 1A 1WATER C 16861976000 SERGE-FROLOW -15.88 54.52 13 0R -9FLxxCO 1A-9WATER A
Asia
20140948000 KABUL AIRPORT 34.55 69.22 1791 2290U 534MVxxno-9A 2WARM FIELD WOODSA 20140990000 KANDAHAR AIRP 31.50 65.85 1010 1008U 180HIxxno-9A15HOT DESERT A 20550527000 HAILAR 49.22 119.75 611 630U 120FLxxno-9A 1COOL FIELD/WOODSC 20550963000 TONGHE 45.97 128.73 110 477S 20HIxxno-9x-9COOL CROPS C 20551243000 KARAMAY 45.60 84.85 428 354R -9HIDEno-9x-9WARM GRASS/SHRUBC 20551644000 KUQA 41.72 82.95 1100 1300U 103HIxxno-9x-9WARM GRASS/SHRUBC 20551656000 KORLA 41.75 86.13 933 1189S 46FLxxno-9x-9SAND DESERT C 20552267000 EJIN QI 41.95 101.07 941 1220R -9HIDEno-9x-9HOT DESERT A 20552323000 MAZONG SHAN 41.80 97.03 1770 1906R -9HIDEno-9x-9HOT DESERT A 20552418000 DUNHUANG 40.15 94.68 1140 1066U 55FLxxno-9x-9COOL IRRIGATED B 20552495000 BAYAN MOD 40.75 104.50 1329 1220R -9HIDEno-9x-9HOT DESERT A 20552681000 MINQIN 38.63 103.08 1367 1520R -9FLDEno-9x-9SAND DESERT B 20552866000 XINING 36.62 101.77 2262 2376U 250MVxxno-9x-9WARM GRASS/SHRUBC 20553336000 HALIUT 41.57 108.52 1290 1317R -9HIDEno-9x-9HOT DESERT B 20553845000 YAN AN 36.60 109.50 959 1156R -9HIxxno-9A-9WARM GRASS/SHRUBC 20554026000 JARUD QI 44.57 120.90 266 300R -9FLDEno-9x-9COOL CROPS C 20554102000 XILIN HOT 43.95 116.07 991 1079S 40FLxxno-9x-9COOL GRASS/SHRUBC 20554161000 CHANGCHUN 43.90 125.22 238 283U 1500FLxxno-9x-9COOL FIELD/WOODSC 20554218000 CHIFENG 42.27 118.97 572 648U 90HIxxno-9x-9COOL CROPS C 20554662000 DALIAN 38.90 121.63 97 39U 1480HIxxCO 2x-9WATER C 20554823000 JINAN 36.68 116.98 58 60U 1500HIxxno-9x-9WARM FOR./FIELD C 20555228000 SHIQUANHE 32.50 80.08 4279 4597R -9MVxxno-9x-9WARM DECIDUOUS A 20555472000 XAINZA 30.95 88.63 4670 5205R -9MVxxno-9x-9TUNDRA A 20555591000 LHASA 29.67 91.13 3650 4813U 175MVxxno-9x-9SIBERIAN PARKS C 20556004000 TUOTUOHE 34.22 92.43 4535 4606R -9MVxxno-9x-9TUNDRA A 20556029000 YUSHU 33.02 97.02 3682 5078U 80MVxxno-9x-9TUNDRA A 20556046000 DARLAG 33.75 99.65 3968 4017R -9MVxxno-9x-9SIBERIAN PARKS A 20556079000 RUO'ERGAI 33.58 102.97 3441 3622R -9MVxxno-9x-9WARM CROPS A 20556106000 SOG XIAN 31.88 93.78 4024 5015R -9MVxxno-9x-9TUNDRA A 20556444000 DEQEN 28.50 98.90 3488 3349R -9MVxxno-9x-9TUNDRA A 20556964000 SIMAO 22.77 100.98 1303 1336R -9MVxxno-9x-9WARM DECIDUOUS B 20557127000 HANZHONG 33.07 107.03 509 611U 120MVxxno-9x-9WARM DECIDUOUS C 20557494000 WUHAN 30.62 114.13 23 60U 4250FLxxno-9x-9PADDYLANDS C 20557516000 CHONGQING 29.52 106.48 351 352U 3500HIxxno-9x-9PADDYLANDS C 20557816000 GUIYANG 26.58 106.72 1074 1289U 1500FLxxno-9x-9WARM GRASS/SHRUBC 20558027000 XUZHOU 34.28 117.15 42 60U 1500HIxxno-9A 1WARM CROPS C 20558238000 NANJING 32.00 118.80 12 100U 2000HIxxno-9x-9PADDYLANDS C 20558633000 QU XIAN 28.97 118.87 71 303U 60MVxxno-9x-9WARM MIXED C 20558666000 DACHEN DAO 28.45 121.88 84 0R -9HIxxCO 1x-9WATER A 20558847000 FUZHOU 26.08 119.28 85 199U 900HIxxCO30x-9PADDYLANDS C 20559211000 BAISE 23.90 106.60 242 268R -9HIxxno-9x-9PADDYLANDS C 20559948000 YAXIAN 18.23 109.52 7 48R -9HIxxCO 1A-9WATER C 20559981000 XISHA DAO 16.83 112.33 5 0R -9FLxxCO 1x-9WATER A 20647014000 CHUNGGANG 41.78 126.88 332 543R -9HIxxno-9x-9COOL MIXED A 20647016000 HYESAN 41.40 128.17 714 882S 20MVxxno-9x-9WARM FOR./FIELD C 20647025000 KIMCHAEK 40.67 129.20 19 272U 150HIxxCO 1x-9WARM GRASS/SHRUBC 20647035000 SINUIJU 40.10 124.38 7 30U 300HIxxno-9x-9WARM MIXED C 20647055000 WONSAN 39.18 127.43 36 29U 275HIxxCO 1x-9WARM GRASS/SHRUBB 20647058000 PYONGYANG 39.03 125.78 38 22U 1250FLxxno-9x-9WARM CROPS C 20647069000 HAEJU 38.03 125.70 81 93U 140MVxxCO 1x-9WARM CROPS C 20742071000 AMRITSAR 31.63 74.87 234 217U 458FLxxno-9x-9WARM IRRIGATED C 20742475000 ALLAHABAD/BAM 25.45 81.73 98 90U 513FLxxno-9A 4WARM CROPS C 20742587000 DALTONGANJ 24.05 84.07 221 279S 43HIxxno-9x-9WARM CROPS C 20840706000 TABRIZ 38.08 46.28 1361 1479U 599MVxxno-9x-9WARM GRASS/SHRUBC 20840712000 ORUMIEH 37.53 45.08 1312 1402U 164MVxxno-9x-9HOT DESERT C 20840729000 ZANJAN 36.68 48.48 1663 1752U 100MVxxno-9x-9HIGHLAND SHRUB C 20840731000 GHAZVIN 36.25 50.00 1278 1284U 139FLxxno-9x-9COOL FOR./FIELD C 20840738000 GORGAN 36.82 54.47 155 280U 88MVxxno-9x-9WARM MIXED C 20840745000 MASHHAD 36.27 59.63 980 1035U 670MVxxno-9x-9HIGHLAND SHRUB C 20840747000 SANANDAJ 35.33 47.00 1373 1650U 96MVxxno-9x-9WARM GRASS/SHRUBC 20840754000 TEHRAN-MEHRAB 35.68 51.35 1191 1230U 4496MVxxno-9A 1HIGHLAND SHRUB C 20840757000 SEMNAN 35.55 53.38 1171 1406S 31MVxxno-9x-9HIGHLAND SHRUB C 20840766000 KRMANSHAH 34.27 47.12 1322 1482U 291MVxxno-9x-9WARM CROPS B 20840769000 ARAK 34.10 49.40 1720 1941U 115MVxxno-9x-9WARM GRASS/SHRUBB 20840798000 SHAHRE-KORD 32.33 50.85 1991 2436S 24MVxxno-9x-9WARM FIELD WOODSC 20840800000 ESFAHAN 32.47 51.72 1550 1620U 672HIxxno-9A 1HOT DESERT B 20840809000 BIRJAND 32.87 59.20 1491 1528S 26MVxxno-9x-9WARM FIELD WOODSC 20840841000 KERMAN 30.25 56.97 1754 2096U 140MVxxno-9A 5COOL GRASS/SHRUBC 20840848000 SHIRAZ 29.53 52.58 1491 1909U 416MVxxno-9A 3HIGHLAND SHRUB C 20840856000 ZAHEDAN 29.47 60.88 1370 1580U 93HIxxno-9x-9WARM FIELD WOODSC 20840875000 BANDARABBASS 27.22 56.37 10 149U 89FLxxCO 3A 3WARM GRASS/SHRUBC 21047582000 AKITA 39.72 140.10 21 12U 261FLxxCO 3x-9PADDYLANDS C 21128952000 KUSTANAI 53.22 63.62 156 180U 165FLxxno-9x-9COOL CROPS C 21135746000 ARALSKOE MORE 46.78 61.65 62 33S 38FLxxLA-9x-9WARM GRASS/SHRUBC 21135925000 SAM 45.40 56.12 82 117R -9FLDEno-9x-9HOT DESERT A 21136859000 PANFILOV 44.17 80.07 645 953S 19MVxxno-9x-9WARM GRASS/SHRUBC 21138001000 FORT SEVCENKO 44.55 50.25 -25 1S 12FLxxCO 1x-9WATER B 21448930000 LUANG-PRABANG 19.88 102.13 305 673R -9MVxxno-9A-9TROP. SEASONAL B 21544203000 RINCHINLHUMBE 51.12 99.67 1583 1720R -9MVxxno-9x-9SOUTH. TAIGA C 21544207000 HATGAL 50.43 100.15 1668 1816R -9MVxxLA-9x-9SOUTH. TAIGA A 21544213000 BARUUNTURUUN 49.65 94.40 1232 1318R -9MVDEno-9x-9TUNDRA A 21544214000 UIGI 48.93 89.93 1715 2256S 15MVxxno-9x-9HOT DESERT A 21544215000 OMNO-GOBI 49.02 91.72 1590 1903R -9MVxxLA-9x-9COOL DESERT A 21544218000 HOVD 48.02 91.57 1405 1574S 25MVxxno-9x-9HOT DESERT A 21544225000 TOSONTSENGEL 48.73 98.28 1723 2062R -9MVxxno-9x-9COOL DESERT A 21544230000 TARIALAN 49.57 102.00 1235 1317R -9MVxxno-9x-9SOUTH. TAIGA A 21544231000 MUREN 49.57 100.17 1283 1659S 20MVxxno-9x-9COOL DESERT A 21544232000 HUTAG 49.38 102.70 938 1280R -9MVxxno-9x-9COOL DESERT A 21544237000 ERDENEMANDAL 48.53 101.38 1510 1801R -9MVxxno-9x-9COOL DESERT A 21544239000 BULGAN 48.80 103.55 1208 1484S 15HIxxno-9x-9COOL GRASS/SHRUBA 21544241000 BAYAN-GOL, SELENGE 48.90 106.10 807 914R -9HIxxno-9x-9COOL GRASS/SHRUBA 21544259000 CHOIBALSAN 48.08 114.55 747 910S 30FLxxno-9x-9COOL GRASS/SHRUBB 21544272000 ULIASTAI 47.75 96.85 1759 2525S 15MVxxno-9x-9TUNDRA A 21544277000 ALTAI 46.40 96.25 2181 2716S 14MVxxno-9x-9WARM GRASS/SHRUBA 21544282000 TSETSERLEG 47.45 101.47 1691 2290S 28HIxxno-9x-9COOL DESERT A 21544284000 GAIUUT 46.70 100.13 2126 2135S 10MVxxno-9x-9TUNDRA A 21544285000 HUJIRT 46.90 102.77 1662 1898R -9HIDEno-9x-9COOL GRASS/SHRUBA 21544287000 BAYANHONGOR 46.13 100.68 1859 1980S 10MVxxno-9x-9COOL DESERT A 21544288000 ARVAIHEER 46.27 102.78 1813 1819S 12HIxxno-9x-9COOL GRASS/SHRUBA 21544292000 DAUUNMOD, CENTRAL 47.80 106.80 -999 1520S 12HIxxno-9x-9COOL GRASS/SHRUBA 21544294000 MAANTI 47.30 107.48 1430 1520R -9FLDEno-9x-9COOL GRASS/SHRUBA 21544298000 CHOIR 46.45 108.22 1286 1520R -9FLDEno-9x-9COOL GRASS/SHRUBA 21544302000 BAYAN-OVOO 47.78 112.12 926 996R -9FLDEno-9x-9COOL GRASS/SHRUBA 21544304000 UNDERKHAAN 47.32 110.63 1033 1220S 14FLxxno-9x-9COOL GRASS/SHRUBC 21544305000 BARUUN-URT 46.68 113.28 981 910S 12FLxxno-9x-9COOL GRASS/SHRUBA 21544336000 SAIKHAN-OVOO 45.45 103.90 1316 1370R -9FLDEno-9x-9WARM GRASS/SHRUBA 21544341000 MANDALGOVI 45.77 106.28 1393 1228S 10FLxxno-9x-9WARM GRASS/SHRUBC 21544347000 TSOGT-OVOO 44.42 105.32 1298 1271R -9FLDEno-9x-9WARM GRASS/SHRUBA 21544352000 BAYANDELGER 45.73 112.37 1101 1064R -9FLDEno-9x-9COOL GRASS/SHRUBA 21544354000 Sainshand 44.90 110.10 -999 926S 14FLxxno-9x-9WARM GRASS/SHRUBA 21544358000 ZAMYN-UUD 43.73 111.90 964 929R -9FLDEno-9x-9COOL GRASS/SHRUBB 21544373000 DALANZADGAD 43.58 104.42 1465 1556S 10MVxxno-9A 1WARM GRASS/SHRUBA 21744454000 KATHMANDU AIR 27.70 85.37 1337 1538U 354MVxxno-9A 2WARM FIELD WOODSC 21941560000 PARACHINAR 33.87 70.08 1726 2198R -9MVxxno-9A-9COOL GRASS/SHRUBA 22041170000 DOHA INTERNAT 25.25 51.57 10 10U 250FLxxCO 3A 1WATER C 22223711000 TROICKO-PECER 62.70 56.20 139 106R -9FLxxno-9A-9MAIN TAIGA C 22223921000 IVDEL' 60.68 60.45 95 190S 15HIxxno-9x-9BOGS, BOG WOODS C 22225744000 KAMENSKOE 62.43 166.08 10 256R -9MVxxno-9x-9WOODED TUNDRA A 22228138000 BISER 58.52 58.85 463 388R -9MTxxno-9x-9COOL MIXED B 22228434000 KRASNOUFIMSK 56.65 57.78 206 240S 40HIxxno-9x-9COOL GRASS/SHRUBC 22228552000 SADRINSK 56.07 63.65 89 121U 82FLxxno-9x-9COOL CROPS C 22229807000 IRTYSSK 53.35 75.45 94 120R -9FLxxno-9x-9COOL IRRIGATED C 22232411000 ICA 55.58 155.58 6 3R -9FLxxCO 1x-9SIBERIAN PARKS A 22340356000 TURAIF 31.68 38.73 852 816R -9FLDEno-9A-9HOT DESERT C 22340405000 GASSIM 26.30 43.77 650 582U 70FLxxno-9A15SAND DESERT C 22340438000 RIYADH 24.72 46.73 620 696U 1380FLxxno-9A 1WARM IRRIGATED C 22340439000 YENBO 24.02 38.22 11 71S 25HIxxCO 7A 3HOT DESERT C 22443424000 PUTTALAM 8.03 79.83 2 3S 18FLxxCO 1x-9WATER B 22443466000 COLOMBO 6.90 79.87 7 9U 852FLxxCO 1x-9WATER C 22443473000 NUWARA ELIYA 6.97 80.77 1880 1543S 16MVxxno-9x-9WARM FOR./FIELD B 22443497000 HAMBANTOTA 6.12 81.13 20 42S 11FLxxCO 1x-9WARM GRASS/SHRUBA 22848462000 ARANYAPRATHET 13.70 102.58 49 61R -9HIFOno-9x-9TROPICAL DRY FORA 23041196000 SHARJAH INTER 25.33 55.52 33 30U 266FLxxCO10A 5WARM GRASS/SHRUBC 23248820000 HA NOI 21.02 105.80 6 45U 2571FLxxno-9x-9PADDYLANDS C 23248826000 PHU LIEN 20.80 106.63 119 60U 1279FLxxCO14x-9PADDYLANDS B 23248855000 DA NANG 16.03 108.18 7 233U 492MVxxCO 2A 1WATER B 23248877000 NHA TRANG 12.25 109.20 10 20U 216MVxxCO 1A 1WATER C
South America
30187934000 RIO GRANDE B. -53.80 -67.75 22 32S 13FLxxCO 3A 2WATER B 30382397000 FORTALEZA -3.77 -38.60 26 0U 648FLxxCO 1x-9WARM CROPS C 30382578000 TERESINA -5.08 -42.82 74 112U 339FLxxno-9A 1WARM CROPS C 30382861000 CONCEICAO DO -8.25 -49.28 157 157R -9HIxxno-9A-9WARM GRASS/SHRUBC 30383096000 ARACAJU -10.92 -37.05 5 29U 288FLxxCO 2x-9WARM FOR./FIELD C 30383229000 SALVADOR -13.02 -38.52 51 3U 1496HIxxCO 2x-9WATER C 30383361000 CUIABA -15.55 -56.12 151 170U 167HIxxno-9x-9MARSH, SWAMP C 30383552000 CORUMBA -19.08 -57.50 130 171U 66FLxxno-9x-9TROP. SAVANNA A 30383618000 TRES LAGOAS -20.78 -51.70 313 353S 45HIxxLA-9x-9WARM FIELD WOODSC 30383702000 PONTA PORA -22.53 -55.73 650 629S 20HIxxno-9A 2TROP. SEASONAL C 30489056000 CENTRO MET.AN -62.42 -58.88 10 0R -9HIICCO 1x-9ANTARCTICA A 30886033000 BAHIA NEGRA -20.22 -58.17 96 86R -9FLMAno-9A-9SEMIARID WOODS A 30886065000 PRATS-GIL -22.70 -61.50 220 224R -9FLxxno-9A-9SUCCULENT THORNSA 30886068000 MARISCAL -22.02 -60.60 181 189R -9FLxxno-9A-9SUCCULENT THORNSC 30886086000 PUERTO CASADO -22.28 -57.87 87 80R -9FLMAno-9A-9TROPICAL DRY FORA 30886097000 PEDRO JUAN CA -22.58 -55.65 662 631S 20HIxxno-9A 2TROP. SEASONAL A 30886134000 CONCEPCION -23.42 -57.30 74 85S 19FLxxno-9x-9TROPICAL DRY FORA 30886218000 ASUNCION/AERO -25.27 -57.63 101 90U 388FLxxno-9A 5TROPICAL DRY FORC 30886233000 SAN JUAN BAUTISTA/MISIONES -25.80 -56.30 155 233R -9HIFOno-9x-9TROP. SEASONAL A 30886255000 PILAR -26.85 -58.32 56 60S 15FLxxno-9A 1MARSH, SWAMP B 30886260000 SAN JUAN BAUT -26.67 -57.15 126 90R -9HIxxno-9x-9TROPICAL DRY FORB 30886297000 ENCARNACION -27.32 -55.83 91 91S 23FLxxno-9A 6WARM FIELD WOODSC 30984370000 TUMBES -3.55 -80.40 27 35S 48FLxxCO 5A 5WARM GRASS/SHRUBA 30984377000 IQUITOS -3.75 -73.25 126 90R -9FLFOno-9x-9EQ. EVERGREEN A 30984401000 PIURA -5.18 -80.60 55 67U 186HIxxno-9A 1HOT DESERT C 30984452000 CHICLAYO -6.78 -79.83 34 43U 280FLxxCO15A 1WARM IRRIGATED C 30984455000 TARAPOTO -6.45 -76.38 282 995R -9HIFOno-9A-9EQ. EVERGREEN B 30984501000 TRUJILLO -8.10 -79.03 30 231U 355MVxxCO 6x-9WATER C 30984515000 PUCALLPA -8.42 -74.60 149 180U 92HIxxLA-9A 3EQ. EVERGREEN B 30984628000 LIMA-CALLAO/A -12.00 -77.12 13 20U 376MVxxCO 2A 1WATER C 30984686000 CUZCO -13.55 -71.98 3249 3693U 181MVxxno-9x-9TUNDRA B 30984691000 PISCO -13.75 -76.28 7 5U 53FLxxCO 1A 1WATER B 30984735000 JULIACA -15.48 -70.15 3827 3833U 78MVxxno-9A 1COOL CROPS C 30984782000 TACNA -18.07 -70.30 469 385U 93MVxxCO30A 2HOT DESERT B 31281225000 ZANDERIJ 5.45 -55.20 15 30R -9FLxxno-9A-9COOL CROPS B 31386330000 ARTIGAS -30.38 -56.50 120 140S 29FLxxno-9A 2WARM GRASS/SHRUBB 31386350000 RIVERA -30.88 -55.53 241 254U 50FLxxno-9x-9WARM GRASS/SHRUBC 31386360000 SALTO -31.38 -57.95 33 52U 73FLxxno-9x-9WARM GRASS/SHRUBC 31386430000 PAYSANDU -32.33 -58.03 61 58U 62FLxxno-9x-9WARM GRASS/SHRUBB 31386440000 MELO -32.37 -54.22 100 142S 38HIxxno-9A 5WARM GRASS/SHRUBA 31386560000 COLONIA -34.45 -57.83 22 0S 17FLxxCO 2x-9WATER B 31386565000 ROCHA -34.48 -54.30 18 46S 22HIxxCO20A 1COASTAL EDGES C 31386580000 CARRASCO -34.83 -56.00 32 16U 1173FLxxCO 3A 2WARM CROPS C 31480403000 CORO 11.42 -69.68 17 19U 69HIxxCO 6A 1TROPICAL DRY FORC 31480410000 BARQUISIMETO 10.07 -69.32 614 551U 331HIxxno-9x-9WARM GRASS/SHRUBC 31480413000 MARACAY - B.A 10.25 -67.65 437 640U 255MVxxLA-9A 1WARM GRASS/SHRUBC 31480415000 CARACAS/MAIQU 10.60 -66.98 48 239U 1035MVxxCO 1A10WATER C 31480416000 CARACAS/LA CARLOTA 10.50 -66.90 865 1135U 1035MVxxCO12x-9WARM CROPS C 31480419000 BARCELONA 10.12 -64.68 7 62U 78HIxxCO 3A 1WATER C 31480423000 LA GUIRIA VENEZUE 10.58 -62.30 8 136S 15HIxxCO 1A 1COASTAL EDGES C 31480435000 MATURIN 9.75 -63.18 66 70U 98FLxxno-9x-9WARM GRASS/SHRUBC 31480438000 MERIDA 8.60 -71.18 1498 2555U 74MVxxno-9x-9WARM GRASS/SHRUBC 31480444000 CIUDAD BOLIVA 8.15 -63.55 48 62U 104FLxxno-9A 1TROP. SAVANNA C 31480447000 SAN ANTONIO D 7.85 -72.45 378 474U 220MVxxno-9A 2TROP. MONTANE C 31480450000 SAN FERNANDO 7.90 -67.42 48 55S 39FLxxno-9A 5WARM GRASS/SHRUBC 31480453000 TUMEREMO 7.30 -61.45 181 183R -9FLxxno-9A-9WARM GRASS/SHRUBA 31480457000 PUERTO AYACUC 5.60 -67.50 74 162S 10FLxxno-9A10TROP. SAVANNA A 31480462000 SANTA ELENA D 4.60 -61.12 907 934R -9HIxxno-9x-9TROP. MONTANE C 31581401000 SAINT-LAURENT 5.50 -54.03 9 34R -9FLxxCO30x-9EQ. EVERGREEN C 31581405000 CAYENNE/ROCHA 4.83 -52.37 9 109S 37HIxxCO10A10MARSH, SWAMP B 31581408000 SAINT GEORGES 3.88 -51.80 7 46R -9FLxxno-9x-9EQ. EVERGREEN A 31581415000 MARIPASOULA 3.63 -54.03 106 268R -9HIxxno-9A-9EQ. EVERGREEN A
North America
40278583000 BELIZE/PHILLI 17.53 -88.30 5 16U 51FLxxCO 3A10TROP. SEASONAL A 40578762000 JUAN SANTAMAR 10.00 -84.22 939 1060S 33MVxxno-9A 2TROP. SEASONAL C 40578767000 PUERTO LIMON 10.00 -83.05 3 60S 30FLxxCO 1x-9TROP. SEASONAL C 40678367000 GUANTANAMO,OR 19.90 -75.13 23 3R -9HIxxCO 1A-9WATER C 41278705000 LA CEIBA (AIR 15.73 -86.87 26 240S 39MVxxCO 2A 5WARM FOR./FIELD B 41278708000 LA MESA 15.45 -87.93 31 43U 151MVxxno-9A10WARM DECIDUOUS C 41278720000 TEGUCIGALPA 14.05 -87.22 1007 1047U 305MVxxno-9A 2WARM FOR./FIELD C 41476160000 HERMOSILLO,SO 29.07 -110.95 211 225U 233HIxxno-9x-9WARM GRASS/SHRUBC 41476220000 TEMOSACHIC,CH 28.95 -107.83 1870 1944R -9MVxxno-9x-9WARM DECIDUOUS B 41476225000 UNIV. DE CHIH 28.63 -106.08 1435 1528U 327MVxxno-9x-9WARM GRASS/SHRUBC 41476243000 PIEDRAS NEGRA 28.70 -100.52 250 227S 21FLxxno-9A 1WARM GRASS/SHRUBC 41476311000 CHOIX,SIN. 26.72 -108.28 238 403R -9HIxxno-9x-9TROP. SAVANNA A 41476342000 MONCLOVA,COAH 26.88 -101.42 615 768U 78MVxxno-9x-9SUCCULENT THORNSC 41476373000 TEPEHUANES,DG 25.35 -105.75 1810 2061R -9MVxxno-9x-9WARM DECIDUOUS B 41476382000 TORREON,COAH. 25.53 -103.45 1124 1339U 244HIxxno-9x-9WARM GRASS/SHRUBC 41476390000 SALTILLO,COAH 25.45 -100.98 1790 1594U 201MVxxno-9x-9SUCCULENT THORNSC 41476393000 MONTERREY,N.L 25.87 -100.20 512 548U 1923MVxxno-9x-9WARM IRRIGATED C 41476405000 LA PAZ, B.C.S 24.27 -110.42 18 71S 46HIxxCO 3x-9WATER A 41476458000 MAZATLAN 23.20 -105.40 3 1642U 147FLxxCO 1x-9TROP. SAVANNA A 41476525000 ZACATECAS,ZAC 22.78 -102.57 2612 2421U 50HIxxno-9x-9WARM DECIDUOUS C 41476548000 TAMPICO, TAMP 22.22 -97.85 9 32U 212FLxxCO 5x-9COASTAL EDGES C 41476556000 TEPIC,NAY. 21.52 -104.90 922 927U 109MVxxCO30x-9WARM CROPS C 41476577000 GUANAJUATO,GT 21.02 -101.25 1999 2244S 37HIxxno-9x-9WARM FIELD WOODSC 41476581000 RIO VERDE,S.L 21.85 -100.00 990 1038S 17HIxxno-9x-9COOL DESERT A 41476632000 PACHUCA,HGO. 20.13 -98.73 2417 2508U 84MVxxno-9x-9WARM CROPS C 41476640000 TUXPAN.VER. 20.95 -97.40 28 27S 34FLxxCO 7x-9WARM CROPS C 41476644000 AEROP.INTERNA 20.98 -89.65 9 10U 234FLxxno-9A 2WARM CROPS C 41476647000 VALLADOLID,YU 20.70 -88.22 22 15S 15FLxxno-9x-9TROP. SAVANNA C 41476654000 MANZANILLO,CO 19.05 -104.33 3 30S 21HIxxCO 1x-9TROPICAL DRY FORC 41476662000 ZAMORA,MICH. 19.98 -102.32 1562 1733R -9MVxxno-9A-9WARM CROPS C 41476665000 MORELIA,MICH. 19.70 -101.18 1913 1979U 199MVxxno-9x-9WARM FIELD WOODSC 41476680000 MEXICO (CENTR 19.40 -99.20 2303 2307U13994MVxxno-9x-9WARM CROPS C 41476683000 TLAXCALA,TLAX 19.32 -98.23 2248 2342S 10HIxxno-9x-9TROP. MONTANE C 41476685000 PUEBLA,PUE. 19.05 -98.17 2179 2151U 466MVxxno-9x-9TROP. MONTANE C 41476687000 JALAPA,VER. 19.53 -96.92 1389 1423U 161MVxxno-9x-9WARM CROPS C 41476692000 HACIENDA YLAN 19.15 -96.12 13 6U 256FLxxCO 1x-9TROP. SEASONAL C 41476695000 CAMPECHE,CAMP 19.85 -90.55 5 8U 70FLxxCO 1x-9WATER C 41476726000 CUERNAVACA,MO 18.88 -99.23 1618 1720U 240MVxxno-9x-9WARM CROPS C 41476741000 COATZACOALCOS 18.15 -94.42 23 3U 70FLxxCO 1x-9WATER C 41476750000 CHETUMAL,Q.RO 18.48 -88.30 9 3S 24FLxxCO 1x-9TROP. SEASONAL C 41476775000 OAXACA,OAX. 17.07 -96.72 1550 1858U 115MVxxno-9x-9TROP. SAVANNA C 41476805000 ACAPULCO,GRO. 16.83 -99.93 13 113U 309MVxxCO 1x-9COASTAL EDGES C 41476845000 SN. CRISTOBAL 16.73 -92.63 2276 2336S 26MVxxno-9x-9TROP. SEASONAL C 41476903000 TAPACHULA, CH 14.92 -92.27 118 281U 60MVxxCO20A 3TROPICAL DRY FORC 41578741000 MANAGUA, NICARAGUA 12.10 -86.20 56 107U 405HIxxLA-9A 5WARM FIELD WOODSC 42572259000 DALLAS-FORT W 32.90 -97.03 182 161U 4037FLxxno-9A 5WARM FIELD WOODSC 42572597000 MEDFORD/MEDFO 42.37 -122.87 405 415S 47MVxxno-9A 2WARM FOR./FIELD C 42978384000 OWEN ROBERTS 19.28 -81.35 3 0R -9FLxxCO 1A-9WATER C 43104220000 EGEDESMINDE 68.70 -52.75 41 20R -9HIxxCO 1x-9WATER A 43104250000 GODTHAB NUUK 64.17 -51.75 70 0R -9MVxxCO 1A-9TUNDRA B 43104312000 NORD ADS 81.60 -16.67 34 13R -9HIxxCO 1x-9ICE A 43104320000 DANMARKSHAVN 76.77 -18.67 12 265R -9FLxxCO 1x-9WATER A 43104360000 ANGMAGSSALIK 65.60 -37.63 52 275R -9MVxxCO 1x-9WATER A 43104390000 PRINS CHRISTI 60.05 -43.17 74 0R -9HIxxCO 1x-9TUNDRA A 43278897000 LE RAIZET,GUA 16.27 -61.52 11 36S 25HIxxCO 2A 3WATER C 43378925000 LAMENTIN/MARTINIQUE/FT DE 14.60 -61.10 144 64U 98HIxxCO 1x-9WATER C 43478866000 JULIANA AIRPO 18.05 -63.12 9 24R -9HIxxCO 1A-9WATER C 43478988000 HATO AIRPORT, 12.20 -68.97 67 0U 95FLxxCO 1A10WATER C 43871805000 SAINT-PIERRE 46.77 -56.17 5 10R -9HIxxCO 1A-9WATER B
Pacific Region
50194259000 BURKETOWN -17.73 139.53 8 7R -9FLxxCO30x-9WARM FIELD WOODSA 50194968000 LAUNCESTON AI -41.53 147.20 178 146S 31HIxxno-9A 8COOL FIELD/WOODSB 50291652000 UNDU POINT -16.13 -179.98 63 0R -9HIxxCO 1x-9WATER A 50291680000 NANDI -17.75 177.45 18 65R -9HIxxCO 1A-9TROP. MONTANE C 50291683000 NAUSORI -18.05 178.57 7 88U 64HIxxCO 8A20WATER B 50291699000 ONO-I-LAU -20.67 -178.72 28 0R -9HIxxCO 1A-9WATER A 50396109000 PAKANBARU/ 0.47 101.45 31 83U 186FLxxno-9A 3EQ. EVERGREEN C 50396633000 BALIKPAPAN/SE -1.27 116.90 3 19U 281FLxxCO 1A 1EQ. EVERGREEN C 50396745000 JAKARTA/OBSER -6.18 106.83 8 27U 6503FLxxCO 6x-9PADDYLANDS C 50396797000 TEGAL -6.85 109.15 10 0U 132FLxxCO 1x-9PADDYLANDS C 50396925000 SANGKAPURA -5.85 112.63 3 0R -9HIxxCO 1x-9WATER A 50396973000 KALIANGET(MAD -7.05 113.97 3 70R -9FLxxCO 1x-9COASTAL EDGES A 50397048000 GORONTALO/JAL 0.52 123.07 2 75U 98MVxxCO 3x-9WATER B 50397182000 UJANG PANDANG -5.07 119.55 -999 30U 709HIxxCO 7A15WARM CROPS C 50397240000 AMPENAN/SELAP -8.53 116.07 3 35S 47MVxxCO 2A 3WARM FOR./FIELD B 50397796000 KOKONAO/TIMUK -4.72 136.43 3 0R -9FLMACO 1x-9WATER A 50548674000 MERSING 2.45 103.83 45 0S 18FLxxCO 1x-9WARM FOR./FIELD B 50998755000 HINATUAN 8.37 126.33 3 50R -9FLxxCO 1x-9TROP. SEASONAL A 51891643000 FUNAFUTI -8.52 179.22 2 0R -9FLxxCO 1A-9WATER A 52091554000 PEKOA -15.52 167.22 56 286R -9HIxxCO 5A-9WATER A 52091568000 ANEITYUM -20.23 169.77 7 148R -9HIxxCO 1A-9WATER A 52191765000 PAGO PAGO/INT -14.33 -170.72 3 0R -9HIxxCO 1A-9WATER C 52791334000 TRUK, 7.47 151.85 2 0R -9HIxxCO 1A-9WATER C 52791348000 PONAPE, 6.97 158.22 46 0R -9HIxxCO 1A-9WATER C 52791413000 YAP, CAROLINE 9.48 138.08 17 0R -9HIxxCO 1A-9WATER A 52891925000 ATUONA -9.80 -139.03 52 0R -9HIxxCO 4A-9WATER A 52891938000 TAHITI-FAAA -17.55 -149.62 2 0S 23MVxxCO 1A 2WATER C 52891943000 TAKAROA -14.48 -145.03 3 0R -9FLxxCO 1x-9WATER A 52891945000 HEREHERETUE -19.87 -145.00 3 0R -9FLxxCO 1x-9WATER A 52891948000 TOTEGEGIE, GAMBIER IS. -23.10 -134.90 3 0R -9FLxxCO 1A-9WATER A 52891954000 TUBUAI -23.35 -149.48 3 0R -9HIxxCO 1A-9WATER A 52891958000 RAPA -27.62 -144.33 2 0R -9HIxxCO 1x-9WATER A 53191366000 KWAJALEIN/BUC 8.73 167.73 8 0R -9FLxxCO 1A-9WATER B 53191376000 MAJURO/MARSHA 7.08 171.38 3 0R -9FLxxCO 1x-9WATER B 53291577000 KOUMAC -20.57 164.28 18 42R -9MVxxCO 1x-9WATER A 53291592000 NOUMEA -22.27 166.45 72 0U 56HIxxCO 2A 1WATER C 53691408000 KOROR, PALAU 7.33 134.48 33 0R -9HIxxCO 1A-9WATER B 53991245000 WAKE ISLAND A 19.28 166.65 4 0R -9FLxxCO 1A-9WATER A 54091753000 HIHIFO -13.23 -176.17 27 0R -9HIxxCO 2A-9WATER A
Europe
60237789000 YEREVAN 40.13 44.47 907 1067U 1019MVxxno-9A 1WARM GRASS/SHRUBC 61111464000 MILESOVKA 50.55 13.93 -999 409R -9MVxxno-9x-9COOL CROPS A 61111518000 PRAHA/RUZYNE 50.10 14.25 365 322U 1161HIxxno-9A 3COOL FOR./FIELD C 61111520000 PRAHA-LIBUS 50.02 14.45 304 309U 1161HIxxno-9x-9COOL CROPS C 61111723000 BRNO/TURANY 49.15 16.70 246 302U 336HIxxno-9A 4COOL FOR./FIELD B 61111782000 OSTRAVA/MOSNO 49.68 18.12 256 278U 294HIxxno-9A15COOL CONIFER A 61206030000 ALBORG 57.10 9.87 13 10U 155FLxxCO30A 5WARM CROPS C 61206186000 KOBENHAVN/ 55.68 12.55 9 5U 1328FLxxCO 1x-9WATER C 61206190000 RONNE 55.07 14.75 16 81S 15FLxxCO 1A 3WATER A 61507015000 LILLE 50.57 3.10 52 33U 171FLxxno-9x-9WARM CROPS C 61507037000 ROUEN 49.38 1.18 157 131U 114HIxxno-9A 3WARM CROPS C 61507110000 BREST 48.45 -4.42 103 78U 164FLxxCO 7A 3WARM CROPS C 61507190000 STRASBOURG 48.55 7.63 154 170U 252FLxxno-9A 3WARM DECIDUOUS C 61507222000 NANTES 47.17 -1.60 27 51U 253FLxxno-9A 3WARM CROPS C 61507255000 BOURGES 47.07 2.37 166 152U 75HIxxno-9A 1WARM CROPS C 61507280000 DIJON 47.27 5.08 227 241U 150HIxxno-9A 4WARM FOR./FIELD C 61507434000 LIMOGES 45.87 1.18 402 335U 136HIxxno-9A 5WARM CROPS C 61507460000 CLERMONT-FERR 45.78 3.17 330 473U 153MVxxno-9x-9WARM CROPS C 61507510000 BORDEAUX/MERI 44.83 -0.70 61 44U 220FLxxCO30A 3WARM DECIDUOUS C 61507560000 MONT AIGOUAL 44.12 3.58 1565 1019R -9MTxxno-9x-9MED. GRAZING A 61507630000 TOULOUSE/BLAG 43.63 1.37 153 160U 371FLxxno-9A 3WARM GRASS/SHRUBC 61507643000 MONTPELLIER 43.58 3.97 6 38U 178HIxxCO 8x-9WARM CROPS C 61507650000 MARSEILLE/MARIGNANE FRANCE 43.30 5.40 8 95U 901HIxxCO10A10WATER C 61507690000 NICE 43.65 7.20 10 142U 331MVxxCO 1A 5WARM CROPS C 61507747000 PERPIGNAN 42.73 2.87 48 45U 101FLxxCO12x-9WARM CROPS C 61507761000 AJACCIO 41.92 8.80 9 80S 47MVxxCO 1A 3MED. GRAZING C 61710020000 LIST/SYLT 55.02 8.42 29 0R -9FLxxCO 1x-9WATER A 61710348000 BRAUNSCHWEIG 52.30 10.45 88 74U 269FLxxno-9x-9WARM CROPS C 61710381000 BERLIN-DAHLEM 52.47 13.30 58 41U 3021FLxxno-9x-9WARM CONIFER C 61710384000 BERLIN-TEMPEL 52.47 13.40 49 41U 3021FLxxno-9A 1WARM CONIFER C 61710410000 ESSEN 51.40 6.97 161 113U 7452HIxxno-9A 3WARM FIELD WOODSC 61710444000 GOETTINGEN 51.50 9.95 171 214U 124HIxxno-9x-9WARM DECIDUOUS A 61710739000 STUTTGART/ 48.83 9.20 311 301U 600HIxxno-9x-9WARM CROPS C 61816622000 THESSALONIKI 40.52 22.97 4 107U 482HIxxCO 1A 6MED. GRAZING C 61816641000 KERKYRA (AIRP 39.62 19.92 4 39S 29HIxxCO 2A 1WATER C 61816648000 LARISSA 39.63 22.42 74 101U 72HIxxno-9x-9MED. GRAZING C 61816714000 ATHINAI/OBSER 37.97 23.72 107 100U 2567HIxxCO 6x-9WARM CROPS C 61816716000 ATHINAI (AIRP 37.90 23.73 15 19U 2567HIxxCO 1A 2WARM CROPS C 61816723000 SAMOS (AIRPOR 37.70 26.92 7 73R -9HIxxCO 1A-9WARM CROPS B 61816726000 KALAMATA 37.07 22.02 8 99S 39HIxxCO 4A 6WARM CROPS C 61816734000 METHONI 36.83 21.70 34 58R -9HIxxCO 1x-9WATER A 61816746000 SOUDA (AIRPOR 35.48 24.12 151 27R -9HIxxCO 3A-9WATER B 61816754000 HERAKLION (AI 35.33 25.18 39 88U 78HIxxCO 1A 3WARM CROPS C 62316310000 CAPO PALINURO 40.02 15.28 185 42R -9MVxxCO 1x-9WATER A 62316459000 CATANIA/SIGON 37.40 14.92 22 40U 403HIxxCO15A20MED. GRAZING B 62440250000 H-4 'IRWAISHE 32.50 38.20 688 691R -9FLDEno-9x-9HOT DESERT B 62440310000 MA'AN 30.17 35.78 1070 1042S 11HIxxno-9A 3WARM GRASS/SHRUBA 63822165000 KANIN NOS 68.65 43.30 49 5R -9FLxxCO 1x-9WATER A 63822602000 REBOLY 63.83 30.82 182 188R -9HIxxLA-9x-9MAIN TAIGA B 64214015000 LJUBLJANA/BEZ 46.07 14.52 298 319U 169MVxxno-9x-9WARM CROPS C 64308075000 BURGOS/VILLAF 42.37 -3.63 891 894U 118HIxxno-9A 3WARM FIELD WOODSC 64308215000 NAVACERRADA 40.78 -4.02 1888 1698R -9MVxxno-9x-9WARM CROPS B 64502196000 HAPARANDA 65.83 24.15 6 5R -9FLxxCO 3x-9COASTAL EDGES C 64740007000 ALEPPO 36.18 37.22 393 408U 639FLxxno-9A 2WARM CROPS C 64740022000 LATTAKIA 35.53 35.77 7 13U 126FLxxCO 1x-9WATER C 64740030000 HAMA 35.13 36.72 309 317U 137FLxxno-9A 2WARM IRRIGATED C 64740045000 DEIR EZZOR 35.32 40.15 212 221U 293FLxxno-9A 4WARM GRASS/SHRUBC 64917250000 NIGDE 37.97 34.68 1208 1364S 32MVxxno-9x-9MED. GRAZING C 64917255000 KAHRAMANMARAS 37.60 36.93 549 960U 136MVxxno-9x-9MED. GRAZING C 64917260000 GAZIANTEP 37.08 37.37 855 930U 300HIxxno-9x-9MED. GRAZING C 64917270000 URFA 37.13 38.77 547 630U 133FLxxno-9x-9MED. GRAZING C 64917280000 DIYARBAKIR 37.88 40.18 677 665U 170FLxxno-9A 1WARM GRASS/SHRUBB 64917282000 BATMAN 37.87 41.17 540 546U 64HIxxno-9x-9WARM GRASS/SHRUBB 64917285000 HAKKARI 37.57 43.77 1720 2446S 12MVxxno-9x-9WARM GRASS/SHRUBB 64917292000 MUGLA 37.20 28.35 646 790S 24MVxxCO25x-9MED. GRAZING B 64917300000 ANTALYA 36.70 30.73 57 40U 130HIxxCO 2x-9WATER A 64917340000 MERSIN 36.82 34.60 3 69U 152MVxxCO 1x-9MED. GRAZING C 64917370000 ISKENDERUN 36.58 36.17 3 213U 107MVxxCO 1x-9WARM MIXED C 64917375000 FINIKE 36.30 30.15 3 65R -9MVxxCO 1x-9WATER B 65206011000 THORSHAVN 62.02 -6.77 55 90S 12HIxxCO 1x-9TUNDRA B
Off the right hand edge of the table (not visible on some browsers) are some of the technical fields, like the A flag for “Airstation” (where you often see 1x-9 for a rural non-airport and 1A-9 for a “rural” airport. The A is airport while x is not.
You can shrink the font size or just look at the page source if you wish to see the rest of the record. The only ‘interesting bit’ other than the A flag is the imagination applied to the ‘terrain type’; where what used to be in a surrounding region when a map was made ages ago, is what is asserted to be present today (where Dallas Fort Worth Airport is described as warm fields and woods… in the midst of one of the more extended urban areas on the planet… ). If folks really care, I’ll put the same data in as ‘ragged right’ so you can see it easily.
Oh, and those first two numbers after the name are LAT and LON followed by reported elevation and elevation from a map grid. Then the Urban Suburban Rural flag. Also visible ought to be population in thousands and then the start of a block of codes (that includes the airstation flag). A -9 population is rural.
Wasn’t there already a dearth of stations in Africa already?
The only station that will remain is the one above Karl’s and Hansen’s Bunsen burner in their labs.
OT – I know. 🙂
What am I missing? I thought we want to drop airports because they show artificial heating…
How does this list stack up next to the lists of stations that meet or do not meet the technical standards noted in the surface stations report?
In other words, are these good stations or bad stations? or is there any other correlation?
I didn’t realize North America was dominated by such tropical locals. 🙂
Oh well. It’s snowing in FL also right now.
OT Gordon Brown is at it again – numpty
http://www.guardian.co.uk/environment/2010/feb/12/gordon-brown-climate-change-fundraising
“But Mr Brown brushed aside the sceptics’ challenge during a UN webcast to launch the group today.
“Those people who have become global warming deniers and those people who have become climate change deniers are against the grain of all the evidence that has been assembled that global warming and climate change are indeed challenges that the world must meet and that can only be met together,” he said. He has previously denounced what he described as “anti-science, flat-earth climate sceptics”.”
Some are located where there is just too much cold and snow and the checkers have to travel in the mud away from all weather black asphalt surfaces. Pruning makes the data stronger (More robust)
Did anyone else notice “Batman?!”
Question number five answered by Dr. Gavin Schmidt, a climate researcher at NASA’s Goddard Institute for Space Studies (GISS) in New York City.
“5. What about the meteorological stations? There have been suggestions that some of the stations are located in the wrong place, are using outdated instrumentation, etc.
Global weather services gather far more data than we need. To get the structure of the monthly or yearly temperature changes over the United States, for example, you’d need just a handful of stations, but there are actually some 1,100 of them. You could throw out 50 percent of the station data or more, and you’d get basically the same answers. Individual stations do get old and break down, since they’re exposed to the elements, but this is just one of things that NOAA [the National Oceanic and Atmospheric Administration] has to deal with. One recent innovation is the set up of a climate reference network alongside the current stations so that they can look for potentially serious issues on the large scale — and they haven’t found any yet.”
http://climate.nasa.gov/news/index.cfm?FuseAction=ShowNews&NewsID=248
“… haven’t found any yet.” Hmmm, Thank you God, for making our foes so easy to take down.
Mike Ramsey
In the Africa list, I note that stations 80-99 are all in the Republic of Sudan spanning the northern tropical zone from about 8-22N latitude. The sites cover the capital city and all the major regional towns throughout this vast country, from the desert north through the Sahel and savannah zones south to the edge of the rain forest. In effect, Sudan has been deleted from the climate recording map. It is possible/likely that the meteorological recording system has broken down completely, given the frequent internal political instability/conflicts across much of Sudan over ecent decades.
Many other deleted stations refer to north-west African countries.
I would be interested in seeing if there is any sort of pattern to the station dropouts… Is it possible to tell if a person is selecting whether or not to include a station.
I know that everytime these guy get caught, they cliam that the changes are not sinister in nature, but simply a mistake. But if the dropouts are nefarious in nature, it is possible that simply looking at the individual station droput data might not show any trends, but I bet they allow other nearby stations that show more recent warming (or earlier cooler temps) to extend their effect.
One thing is sure, there is an agenda here, it just needs to be teased out.
Well, it will be a lot easier to to the “adjustments” to fewer stations now !!!
Somebody should take EM’s work and do a google map before and after.
Interesting to see that they dropped Haparanda. You can get the hourly readings in real time here:
http://www.smhi.se/vadret/vadret-i-sverige/observationer#
Haparanda is also one of the sites with longest continuous record at high latitudes anywhere in the World (it has records back to 1802), and is, while not exactly rural, a rather small town.
Seems I can’t type today, cliam should have been claim…
You are confusing cause and effect. The thermometers are going extinct *because* of Global Warming. It will be in AR5. In fact, at this rate all temperature stations will likly be extinct by 2035 according to WWF.
Maybe they realize that satellite data is superior, so they are slowly winding down the surface thermometer mess.
It will be pretty hard to pull all the reading centers as we still have uncontrolled…ops CBC, ummm, controlled televison stations.
Ah well we still have the uncencored…oh no the climate cops….
censored internet.
They killed my hometown, Braunschweig, Germany. They must have found out i’m a skeptic. Or maybe it didn’t warm enough: The station is at a small airport that is not tarmacked, and it’s outside the city near a forest, so not much UHI.
It can be found easily by searching for this in google maps:
Flughafen Braunschweig-Wolfsburg
Crikey – they’ve taken out most of France!
Don’t they like the Fwench?
There doesn’t seem to be reason behind this, like taking out all ‘U’, or all ‘R’ stations: the thermometers are dying across the board.
DanD (10:22:28) :
Did anyone else notice “Batman?!”
I think that Batman is a city in Turkey.
Mike Ramsey
I don’t see 1 station in the UK or Ireland! Am I missing something?
I guess this comes as no surprise as it transgresses every rule of data collection. In my simplicity, I would reckon that for a global temperature average one would take as many stations as one could get and correct for the number of stations per unit area so that well covered areas did not skew the mean. Adjustment for UHI would be reasonable but that’s it. Dealing with missing values is standard statistical practice and would not trouble a decent statistician.
Even if it is reasonable to drop some stations for incomplete data, the values from those stations should be reported to check that they are representative of the retained stations. Epidemiology 101.
This is what we jokingly used to call data enrichment. I never thought I would see it for real from supposedly “top” research establishments
just because NOAA and GCHN have stopped using their data doesn’t mean they have stopped gathering and reporting. We need the “real” raw data, not just the filtered raw data. That would allow us to correlate the dropped staion data against the remaining to see if there is a discrenable bias. By keep the raw data out of the worlds hands they are setting the playing field in their favor.
I notice that Durban Airport has been given the chop. I cannot imagine why. Its surrounds are described as “water”. Well, yes, it has the Indian Ocean along one side, but on the other it has the outskirts of a city with a reputed population of 5M. (probably only 3M or 4M or so in fact).
I guess surfacestations.org will have to hire a detective team. This is no longer about metereology, so its becoming less interesting.
The trouble is what kind of experts will be needed in the future for finding all missing data; archeologists, cryptologists or criminologits?☺. Well, real data is lost for ever anyway.
I havde to add there is no political unrest in my hometown ATM, unlike Sudan, and, glancing at google i see the Braunschweig airport does have one tarmacked runway now – probably the people at Volkswagen paid for one, their managers use this airport with a Learjet or something like that.
No Canada… why mostly sweaty places?
Mike Ramsey (10:24:13), that reminds me of Isaac Asimov’s 1955 short story “Franchise”. In that story the supercomputer “Multivac” decides all US elections by interviewing a single voter who’s so average that he represents the entire electorate.
Hey, the “average surface temperature of the Earth” is a number so nebulous and utterly useless that NASA might just as well derive it from a single thermometer and save us all a few bucks.
Thinking more about this. I guess, now that the science is settled and the end is (Bill) Nye unless we do something, there is no need to rely on temperature stations.
I saw a map a few days ago (maybe here, maybe not) where all the poorly or non-measured areas showed up as warmer including the Arctic, Siberia and N. Africa.
Of course in the past few months Europe, China, USA etc must have had a lot of cold readings. Chances are they are for the ax next.
However roll forward a few years. What’s the chance the die hard AGW supporters will use 2009/2010 as a new baseline of cold to show ‘unprecedented’ warming over the next few years?
Schmidt and his legacy thermometers. The old ones are tired and need a little help. We have snowfall in all 57 states. Is that showing up on a map?
Don’t you mean northeast?
Plato Says (10:19:31) :
OT Gordon Brown is at it again – numpty
(…)
Curious. “Global warming deniers” and “climate change deniers” are now identified as being separate groups. Perhaps there are enough of them now to differentiate based on the slight but noticeable differences, thus leading to declaring them separate species. Evolution in action, as populations grow to fill enticing and expanding niches.
Didn’t Michael Mann, with his hockey stick, deny there was either climate change or global warming before Modern Industrialization? Thus it is allowable to be “anti-science, flat-earth climate sceptics” with regards to historical times, but not more modern ones?
The actual number of records is just way to low. If the professionals are unable to collect the temp, or think it can be restricted by dislosure agreements, there are other sources. Many amateurs like to collect temp readings, and business do for commercial reasons also. It should be practical to do an open call to the public to provide long station records. I suspect there are schools, sea ports, ice cream makers, farms, ski resorts which have 20 years records. Even if you set the limit at 50 years, I think we would find significant numbers Of course, you still need QA, but this is easier if there is a direct dialogue with the supplier.
Mike Ramsey (10:50:03) :
“I think that Batman is a city in Turkey.”
Ridiculous, bats are mammals not birds.
Did I miss it? What does the A, B, or C represent at the end of the row? And why is Mexico such a target? Are the thermometers getting kidnapped too??!!!
Dallas Fort Worth airport is one of hundreds of GHCN reporting stations gone missing.
Because it’s too cold in Dallas now?
Yikes!!! First Bolivia and now they’re going after Paraguay!!!
Mark Young (10:07:42) :
Wasn’t there already a dearth of stations in Africa already?
Yup. The large holey areas just got more holey…
Leif Svalgaard (10:13:22) : What am I missing? I thought we want to drop airports because they show artificial heating…
I don’t want to drop any stations. I want the data to clearly state what is happening and for change to be reportable by group (such as Airports vs rural vs Urban vs …) so see what’s contributing to the means.
With that said: Dropping an airport can still induce an upward bias. Say, for example DFW was in the ‘baseline to now’ period (and so it STAYS in the baseline). And say it was cooler in the baseline (pre -Jets) and ‘warmed the planet’ until now. But since, oh, 1990 or so it has not changed to be much warmer (i.e. traffic is down and they are not adding more tarmac). It’s just not adding any new ‘lift’. So you drop it. But if you leave in a nearby place that is still growing, it will now be showing increases of “anomaly” when compared to the prior values of DFW and those increased values will be used to “fill in” DFW that will then be compared to it’s prior self… and found to be warming.
And yes, GIStemp compares “old basket of thermometers” to “new basket of thermometers” when computing “grid box anomalies”. It does not compute anomalies as ‘thermometer to self”… except by accident if that’s all it’s got in a box.
So change itself is a problem.
Someone may have asked.
I dunno…it just seems that vast areas are under represented a la the Bolivia Syndrome.
Are these stations pinpointed on a world map somewhere?
Just curious.
Thanks,
Clive
According to Gavin Schmidt, you only need a handful of thermometers to accurately represent the mean temperature of the planet earth. Conveniently, these handful of thermometers are all located in urban environments, and what they read doesn’t matter anyway, because Schmidt will just aimlessly wander over to Hansen’s office and they will adjust the earlier times down and the more recent times up.
Gotta keep that global warming afloat!
extinction or culling?
E.M.Smith (11:45:29) :
There you go again E.M., explaining things so it’s easy to understand.
🙂
In plots, the UHI seems to be more of a step than a ramp. In the Russian stations, for instance, It appeared to be a slight ramp for 10 years, then a steep increase for ten years, then a levelling off afterward.
If you trim data from UHI sites like airports _after_ the UHI step (using the UHI excuse), then you maximize the effect of UHI on the aggregate series rather than minimize it.
On the other hand, if you trim data from UHI sites like airports by removing the site and its entire history, you will get the most accurate picture for a number of sites that might be to small to be counted as a global proxy.
On the other hand, if you accept UHI error by not correcting (by deletion) for it at all, then the average sum of the staggered anomoly steps will form a ramp that starts around the 1950s, then correlates well with fossil fuel use (a proxy for urbanization), then stabilizes when the bulk of your sites has passed the “step”. You would then approach an accurate anomoly measurement again (with readings offset from pre-UHI by a static error).
I think (conspiracy theory) the thermometer team has been gaming these three scenarios, and has maybe painted itself into a corner by exhausting sources of UHI step.
Global weather services gather far more data than we need. To get the structure of the monthly or yearly temperature changes over the United States, for example, you’d need just a handful of stations, but there are actually some 1,100 of them. You could throw out 50 percent of the station data or more, and you’d get basically the same answers. — Gavin
If the distribution is even and random, you might get the same answers. If you carefully select which half to throw out one could introduce confirmation bias. Since the advent of modern data processing, when has too much data ever been a problem.
Mike Ramsey (10:24:13) :….Anthony
Has anyone taken a look at the reference network products, or their siting yet? I’d be curious if these actually fit with the GISS products or not, assuming they’re not adjusted or otherwise shenaniganized…
http://www.ncdc.noaa.gov/crn/
There are 10 stations from Greece. I wonder if any are left because the ones taken out seem to cover the country!
We have had a cool summer and a medium winter. Maybe it is the cool summer that axed them?
So are they now set up to declare Africa has become a hotter and more arid desolate wasteland in the years to come, exactly as forecast? Have they properly pared down the weather stations elsewhere so their predictions will be proven true?
James Sexton (11:39:32) : Did I miss it? What does the A, B, or C represent at the end of the row? And why is Mexico such a target? Are the thermometers getting kidnapped too??!!!
For Mexico, a traunch was added some years back in The Megathermal Zone. A casual look at the Mexican deletions look like more up toward the Sierra Madre mountains. They don’t like Mountains at NOAA… Japan has no thermometer above 300 M any more, so, I suspect, will the rest of the world once they are done…
I found where they kept the data and descriptions once, and kept track of it here:
http://chiefio.wordpress.com/2009/02/24/ghcn-global-historical-climate-network/
For those wondering what all the misc flag characters mean, the “documentation” provided as described on that link (BEGIN QUOTE):
A detailed description of GHCN’s Quality Control can be
found through http://www.ncdc.noaa.gov/ghcn/ghcn.html.
So, there you go. Some pretty good pointers to where to get bits and what they mean. But what about these “read.inv.f” and “read.data.f” programs it mentions? Well, I didn’t see them. But I did see one named “v2.read.data.f” that seems to do the same thing.
The comment block from down in the guts of that program does a nice job of telling you what the fields are:
c ic=3 digit country code; the first digit represents WMO region/continent
c iwmo=5 digit WMO station number
c imod=3 digit modifier; 000 means the station is probably the WMO
c station; 001, etc. mean the station is near that WMO station
c name=30 character station name
c rlat=latitude in degrees.hundredths of degrees, negative = South of Eq.
c rlong=longitude in degrees.hundredths of degrees, – = West
c ielevs=station elevation in meters, missing is -999
c ielevg=station elevation interpolated from TerrainBase gridded data set
c pop=1 character population assessment: R = rural (not associated
c with a town of >10,000 population), S = associated with a small
c town (10,000-50,000), U = associated with an urban area (>50,000)
c ipop=population of the small town or urban area (needs to be multiplied
c by 1,000). If rural, no analysis: -9.
c topo=general topography around the station: FL flat; HI hilly,
c MT mountain top; MV mountainous valley or at least not on the top
c of a mountain.
c stveg=general vegetation near the station based on Operational
c Navigation Charts; MA marsh; FO forested; IC ice; DE desert;
c CL clear or open;
c not all stations have this information in which case: xx.
c stloc=station location based on 3 specific criteria:
c Is the station on an island smaller than 100 km**2 or
c narrower than 10 km in width at the point of the
c station? IS;
c Is the station is within 30 km from the coast? CO;
c Is the station is next to a large (> 25 km**2) lake? LA;
c A station may be all three but only labeled with one with
c the priority IS, CO, then LA. If none of the above: no.
c iloc=if the station is CO, iloc is the distance in km to the coast.
c If station is not coastal: -9.
c airstn=A if the station is at an airport; otherwise x
c itowndis=the distance in km from the airport to its associated
c small town or urban center (not relevant for rural airports
c or non airport stations in which case: -9)
c grveg=gridded vegetation for the 0.5×0.5 degree grid point closest
c to the station from a gridded vegetation data base. 16 characters.
c A more complete description of these metadata are available in
c other documentation
Unfortunately, it does not tell you just what that ‘other documentation’ might be nor where to find it…
So you will notice that the far right A,B,C is um, er, “not clearly defined” but you can go look in “other documentation”… somewhere…
Viv Evans (10:49:41) :
Crikey – they’ve taken out most of France!
Don’t they like the Fwench?
Yes, they must really hate the French because also gone is the station at St. Pierre, on the North American list. The islands of St. Pierre et Miquelon off the southern coast of Newfoundland belong to France.
The NOAA cafeteria must still be serving Freedom Fries.
If my memory is accurate, (Colonel?) Batman was the bloke who founded Melbourne, Aus.
Sean Peake (10:56:42) : No Canada… why mostly sweaty places?
IMHO, Canada is already showing warming (from prior deletions?) so no need to change anything. Africa is just flat (in temperatures) and hot, so it will take more work to make it show warming. For Latin America, warming is a bit slow too (the Andes were deleted a long time ago…) and some places like Peru had too many headlines with children freezing to death or the snowfall in Southern Brazil and the crop failures in Argentina from cold. So it would need better headlines… /sarcoff>
Or maybe it isn’t sarcastic… maybe in our Orwellian world… No, no, must not think that way. Where did I leave my soma?…
“According to Gavin Schmidt, you only need a handful of thermometers to accurately represent the mean temperature of the planet earth.”
In that case, why do we need the 3,300 submersibles that make up the Argo network to measure the ocean temperatures?
Honestly, I lately I feel like every day is Christmas. We wake up, we get to check WUWT, and then see the latest news that shows what a complete joke the idea of catastrophic global warming is. Every day more and more lies are exposed. Lies, fake data, and everything in between.
I sincerely hope that someone is keeping track of all of the talking heads, politicians, and pseudo-scientists who perpetuate the myth that “the science is settled.” We should forward all of these things to them, daily.
Sincerely, the current idea of global warming is only being kept alive by the know-nothings who keep repeating the same lies over and over. There is just nothing of substance left.
” BrianSF (10:50:34) :
I don’t see 1 station in the UK or Ireland! Am I missing something?”
I was also looking for UK stations and found none. This is remarkable as the UK temperatures do differ from the Continent. My old home country Denmark on the other hand has three station on perhaps 1/10th the area.
E.M.Smith (12:26:21) : Where did I leave my soma?…. BTW there are already several pharmaceutical products which have been invented for that purpose: The drugs for oblivion. Some even innocent ones, and very popular, like known paracetamol/tylenol.
E.M.Smith: I forgot. I think we have about 7 stations left here in Canada: 3 are beside steel mills, 2 are near charcoal manufacturing plants, and the rest are indoors so they are easier to reach and read in bad weather.
“John Batman (21 January 1801 – 6 May 1839) was an Australian farmer and businessman who was one of the first settlers of the Melbourne area and known for founding Victoria.”
So says the ever-authoritative Wikipedia. If you disagree, please take it up with them, I’m just relaying the info. And find another source for your term papers.
Leif:
“What am I missing? I thought we want to drop airports because they show artificial heating…”
Even with that they haven’t managed to get any heating in the last 12 years. The ultimate goal is to extrapolate from the thermometer on James Hansen’s outdoor grill to the world. Will you take your steak well done or carbon only, Leif?
E.M.Smith (12:14:46) :……..So you will notice that the far right A,B,C is um, er, “not clearly defined” but you can go look in “other documentation”… somewhere…
Yeh, sigh, and so it is as it always was. Obfuscation is key for these people. Thanks for the clarity you provided though.
Obviously, our English speaking thermometers are way better than the others!!!!!
The Global Historical Climatology Network (GHCN) is a database of temperature, precipitation and pressure records managed by the National Climatic Data Center, Arizona State University and the Carbon Dioxide Information Analysis Center (CDIAC).
CDIAC focus on obtaining, evaluating and distributing data related to climate change and greenhouse gas emissions.
i.e. there’s climate change – we stay in business or there’s no climate change, so what’s the point of CDIAC?
Yes no UK or Ireland, strange to leave out the country with the oldest temp records.
Well not strange as these show there is no warming as temps are close to what they were 100 years ago.
Dr. Robert (11:51:43) :
According to Gavin Schmidt, you only need a handful of thermometers to accurately represent the mean temperature of the planet earth. Conveniently, these handful of thermometers are all located in urban environments, and what they read doesn’t matter anyway, because Schmidt will just aimlessly wander over to Hansen’s office and they will adjust the earlier times down and the more recent times up.
Gotta keep that global warming afloat!
_______________________
Reply:
Better yet, with just a handful of thermometers, he can just tuck them in his shirt pocket and read them at his convenience:
“Yup… these all say 98.6 degrees again! Dang, the earth’s getting intolerably warm, just as we predicted!”
“E.M.Smith (12:26:21) :
[…]
Or maybe it isn’t sarcastic… maybe in our Orwellian world.”
NOAA are swindlers. They cannot justify this. Every economist, every physicist and every engineer can see this.
OT: WUWT was shown in Swedish TV in Two different news shows tonight! 🙂
The tide is turning here and MSM are now reporting on prime time about Himalayagate (they are lagging a bit as you understand).
On sunday, one of Swedens most active sceptic bloggers, Maggie of http://www.theclimatescam.se, will participate in a live debate against our environmental minister!
And this is in what is probably the most self righteous country in the world where it’s (up until now) been regarded as blasphemy or lunacy to question AGW.
lol, IT’S A LANGUAGE BARRIER!!!! The NOAA doesn’t know how to translate temperature in other languages to English temperature!!!!! That must be it!!!
It does seem a reasonable question to ask that, if we are to believe we can control temperature through taxation, how are we going to measure the ROI on that approach?
However, I have noticed that ANY questions on the “science” seems to get you labelled as a skeptic.
This can’t go on much longer, can it?
Slightly OT, here is a follow up comment by Andrew Lacis, on the previous piece posted here a while ago. http://dotearth.blogs.nytimes.com/ For me this statement “Anthropogenic warming of the climate system can be detected in temperature observations taken at the surface, in the troposphere and in the oceans.” Is precisely the root argument is it not? Is he implying that the measurements are 100% drop dead certain and reliable – I think not.
I have converted the station list into a google earth kml file that is available for download at
http://cid-78d84fe53bde5e59.skydrive.live.com/self.aspx/Public/RIPtempStations.kml
I can sort the data into different groupings/latitude if anyone is interested.
I live just off Dallas-Fort Worth Airport. Remember it’s huge enough to be an environment of its own, even though it’s surrounded by suburbs.
I’d consider their land surrounding the airport proper woodlands. The last remaining stretch of tree covered country lane is on airport property and I drove it on the way home yesterday my flight last night. The trees were completely flocked with snow, a scene right out of Currier and Ives.
Not something you see every day around here.
Vincent (12:36:46)
“In that case, why do we need the 3,300 submersibles that make up the Argo network to measure the ocean temperatures?”
There is a fundamental question as to ‘Where is all the heat going’. It doesn’t show up in the surface temperature record, it doesn’t show up in the lower and upper atmosphere records. So it must be hiding in the ocean. So 3300 Buoy’s to find the missing heat in the ocean.
So, they’re discarding/exempting data from certain temperature stations, and this somehow increases their confidence in their predictions and further settles the science.. Isn’t that like a sportscaster reporting who won a baseball game by only adding up the run totals from the 1st, 5th, and 9th innings?
(You think bookies in Vegas would buy that..?)
Cheers,
BTW, in response to the KUSI broadcast where the issue of thermometers being dropped was aired, there was a response posted at Yale that said:
http://www.yaleclimatemediaforum.org/2010/01/kusi-noaa-nasa/
When glancing at the chart showing the number of temperature stations used over time, it does appear rather odd to see the number of stations used in the GHCN network drop dramatically between the 1970s and present. D’Aleo and Smith point to purposeful elimination of those stations.
However, as Thomas Peterson and Russell Vose, the researchers who assembled much of GHCN, have explained:
It’s common to think of temperature stations as modern Internet-linked operations that instantly report temperature readings to readily accessible databases, but that is not particularly accurate for stations outside of the United States and Western Europe. For many of the world’s stations, observations are still taken and recorded by hand, and assembling and digitizing records from thousands of stations worldwide is burdensome.
During that spike in station counts in the 1970s, those stations were not actively reporting to some central repository. Rather, those records were collected years and decades later through painstaking work by researchers. It is quite likely that, a decade or two from now, the number of stations available for the 1990s and 2000s will exceed the 6,000-station peak reached in the 1970s.
For the 2010 deletions, “That dog won’t hunt”… since BY DEFINITION those stations were reporting up until now. I’m sure someone can find a Somalia or Sudan set that went flakey for political reasons (much as Indonesia did for a few years during their troubles, but was not dropped, just had missing data flags put in…) but, well, DFW and Strasbourg… And last I looked Mexico was not having a revolution…
BTW, the link for the “explanation” from NOAA is:
http://www.ncdc.noaa.gov/oa/climate/research/Peterson-Vose-1997.pdf
Guess they will need some time to come up with a new explanation…
EGEDESMINDE
GODTHAB NUUK
NORD ADS
DANMARKSHAVN
ANGMAGSSALIK
PRINS CHRISTI
Looks like Greenland has been turfed. Are there even any stations left there at all?? I couldn’t find any on the quick pass I just did. Must be all that pesky ice. Odd considering they’re so worried about the place melting and drowning us all…
Maybe somebody can answer the following:
1. I keep reading that temps have risen(as an example: .7C) over a certain time period, such as the last 100 years…
a. how are these temps measured, by what kind of instrument, and are all of them worldwide the same?
b. If these current instruments are digital readouts, capable of measuring to a few tenths, what was used prior to digital?
c. how are all of them worldwide calibrated?
2. My point is: How can anyone say that temperatures have risen by X amount, when in all probability that amount is within the variance for error of all instruments used, especially prior to digital readouts, and especially since the number and location of sites seems to vary constantly, and since many are exposed to urban growth around them.
In 2006 the UK experienced a very hot July. In all of over 300 years of record keeping, the UK had never experienced a month where the average temperature was 20 deg c. We waited for the final figure from the Met Office to discover that it was 19.7 deg c. It was, to paraphrase one of our prime ministers, a damn close run thing. It turned out to be 2.82 deg c above the average July temperatures for the previous 10 years. Inevitably it was touted as more evidence of global warming and its impact on the UK. The BBC journalists almost wet themselves with excitement.
It will be interesting then, to see what the media makes of the fact that the UK Met Office have just re-adjusted the January temperature – which turns out to be just 1.4 deg c for the month – which is 3.62 deg c LOWER than an average of the past 10 years of Jan temps.
BBC: I’ll see your +2.82 and raise you -3.62. But I’m betting that NOTHING will be said.
I’m thinking the next step will be to say:
“We don’t need no stinking reporting stations. We have enough historical data to interpolate global temperature without wasting time reading and manipulating the data.”
Both “Nature” and “Science” will have predominate “peer reviewed” articles showing that this is the best way to provide accurate information to the world while also describing how “carbon neutral” the process is.
anna v (12:08:09) :
There are 10 stations from Greece. I wonder if any are left because the ones taken out seem to cover the country!
We have had a cool summer and a medium winter. Maybe it is the cool summer that axed them?
Bad Greece! No Baklava for you! 😉
[chiefio@Hummer data]$ grep ^618 2010_uniq_station_list
[chiefio@Hummer data]$
None in 2010, but here they are in 2009:
[chiefio@Hummer data]$ grep ^618 2009_uniq_station_list
61816622000
61816641000
61816648000
61816714000
61816716000
61816723000
61816726000
61816734000
61816746000
61816754000
[chiefio@Hummer data]$
And checked directly against the v2.mean file to assure no ‘screw up’ of my process:
[chiefio@Hummer data]$ grep ^618 v2.mean | grep 2010
[chiefio@Hummer data]$
We’ll need to see if a warm Italy or hot Turkey warms you up or not…
When I tried the GISS map setting it to Jan 2010 it said:
“Surface Temperature Analysis: Maps
Error
I cannot construct a plot from your input. Time interval: data not available yet.”
So we won’t know for a while yet if you got warmer… No fair looking out the window 😉
Well Gavin Schmidt says he doesn’t need all the data he has now; and could throw out a lot more stations.
I can even buy that; he may in fact have enough stations in the US to represent the group of stations he has in the US.
But unfortunately; the public is led to believe, and the really important public news media, is led to believe that what they are reporting actually represents the whole world; which sadly it does not. So yes they may be oversampling the areas of the USA that they are sampling; but they aren’t oversampling the whole planet.
By the way, E.M. I understand that somebody we both know used to work for you years ago; he speaks highly of his experience working for you.
George
E.M Smith. Much thanks for your tireless, concise, and easy to understand analysis. fm
I am amazed and outraged with the two deletions affecting Spain:
64308075000 BURGOS/VILLAF 42.37 -3.63 891 894U 118HIxxno-9A 3WARM FIELD WOODSC
64308215000 NAVACERRADA 40.78 -4.02 1888 1698R -9MVxxno-9x-9WARM CROPS B
I almost don’t know where to start with these!
Amongst all the provincial capitals in Spain, Burgos holds the record for the lowest average temperature (9.9C). Two other provinces which are widely known as being very cold, are second and third with… Avila: 10.4C, Soria: 10.5C.
Burgos also happens to hold the record for the lowest average minimum during summer, at 10.6C. The absolute minimum for Burgos was -22C in 1971.
In fact, “Villafría” (Burgos) is explicitly mentioned as having achieved the lowest temp. in Spain (-17.1C) during last December’s cold spell here.
So they have just wiped Spain’s coldest city off the record. Nice one!
With regards to the “Airport” rating, well… sort of. The official statistics for the whole of 2009 state that it handled 27,710 passengers, 3,569 flights, and exactly zero cargo. In terms of number of flights, that makes it Spain’s 4th airport… from the bottom of the table!! The average number of passengers per flight is 7.76, which should give a good idea as to the type of aircraft it handles.
This is anecdotal, but interesting nonetheless : the full unabbreviated name for the station should be: “Burgos/Villafría” because it is actually nearest “Villafría de Burgos”, a small town on the outskirts of Burgos city. Now any of you with a bit knowledge of spanish will immediately recognise that “fría” part of the name as meaning: cold (“frío”, but in feminine form because the first part, “Villa” is also feminine) . So an appropriate translation of this small town’s name would be something like “Coldville”. No kidding!
Now, if the first station’s removal is outrageous, the second is even worse. There are very few places where you can go skiing in central Spain, and yes you guessed it, Navacerrada is one of them. Actually, it is THE main place where you can ski in central Spain (nearby “La Pinilla” has a rather patchy record for snow cover) The nearest ski run (piste) is less than 3 Km away from the station’s official coordinates, and about the same height above sea level (~1900m)
Despite the station’s name, it is not particularly near the small town called “Navacerrada”. The name has to do with the nearby mountain pass known as “Puerto de Navacerrada”.
The “WARM CROPS” attribute had me rolling on the floor for a good while: for chrisssssake, this is subalpine vegetation!!! It is probably Spain’s most extensive and best conserved population of Pinus sylvestris (Scots pine). There is even a small breeding population of Bluethroat (Luscinia svecica) in the area, a bird much more typical of more northerly climates (Norway, Russia and Poland, with smaller and more localised populations in central Europe)
What are these people playing at????!!!!
It is refreshing to see, among all the doom and gloom, the sun is on the go. Hopefully, more global warming in the pipeline.
http://stereo-ssc.nascom.nasa.gov/beacon/latest/behind_euvi_195_latest.jpg
In my hundreds of discussions with people who support the theory that there is significant global warming caused by man made CO2 in the atmosphere, what I find most striking is that all their reasonings are “a priori” and they seem to have nothing to say about experiment. As far as I can tell, there are no interesting experiments being conducted by climate scientists. Their disdain for the experimental extends to monitoring stations for atmospheric temperature and CO2 concentrations. Some have argued that the one CO2 monitoring station at Mauna Loa is not only enough but actually superior to having multiple stations on the surface and throughout the atmosphere. They seem to buy into Al Gore’s belief that the science is settled, that it has transcended experiment.
The list is a list of Thermometers DELETED. There are UK and Irish thermometers kept:
[chiefio@Hummer 2010]$ country IRELAND
621 IRELAND
[chiefio@Hummer 2010]$ country UNITED
230 UNITED ARAB EMIRATES
425 UNITED STATES OF AMERICA
651 UNITED KINGDOM
[chiefio@Hummer 2010]$ cd data
[chiefio@Hummer data]$ grep ^621 2010_uniq_station_list
62103953000
62103955000
62103962000
62103967000
62103969000
62103976000
62103980000
[chiefio@Hummer data]$ grep ^651 2010_uniq_station_list
65103005000
65103026000
65103091000
65103100000
65103162000
65103257000
65103302000
65103377000
65103862000
65103917000
[chiefio@Hummer data]$
So it’s a good thing that you are not finding Irish or UK stations on the lists!
Anthony, WUWT gets an honourable mention in an article by Christopher Pearson in today’s The Australian newspaper.
“World wide web of doubt.”
http://www.theaustralian.com.au/politics/opinion/world-wide-web-of-doubt/story-e6frgd0x-1225829874281
Congratulations on being a burr under the saddle of AGW.
EMS, have you (or anyone else) checked to see if V1raw = V2raw?
lol, while my other comments await moderation…….honestly, it seems the NOAA has a bias for English speaking thermometers. U.S., Canada, U.K., Australia, and New Zealand. Lost a total of 5. India seems fairly unscathed also. They lost 3 in the mountainous northern area. Russia and what used to be known as the Soviet bloc in Europe is also fairly unscathed(A few were lost in what appears to be the Urals.). While I can’t attach any meaning to it, it seems very strange to me. I know people don’t do things out of randomness. One does things for a purpose. This may take me a twelve pack or so to sort out.
Leif Svalgaard (10:13:22) :
Leif, you should know that Ålborg & Köbenhavn are both ports, not airports.
DaveE.
Warren Berman (13:29:48) :
Warren, it’s a travesty. First of all, it’s complicated to get accurate measurements in its own right because instruments change over time, environment of weather stations changes through land use changes and UHI, weather stations move, time series are often incomplete. But worse, the guardians of the data collection do their best to mess with the data. NOAA drops more and more thermometers and GISS takes the remainder and extrapolates from them the now missing holes up to 1200 km away.
Search here at WUWT for GISTEMP and read the musings of the chiefio (a.k.a. E.M. Smith) at http://chiefio.wordpress.com/
Data gets tortured with bad algorithms, i won’t call them statistics because that gives statistics a bad name it doesn’t deserve. It’s blatant disregard for the Nyquist theorem in temporal and spatial dimensions, it’s below any standard. If one would compute all this with confidence intervals or error range propagation or any sound numerics i’m sure the alleged 0.6 deg C warming in the 20th century would drown in the error bars.
Here’s a guy who did a very simple analysis of raw data who comes to the conclusion that there is no discernible trend:
http://crapstats.wordpress.com/2010/01/21/global-warming-%e2%80%93-who-knows-we-all-care/
Any views on the Pacific — it looks to me like a lot of island thermometers have gone… again I thought the Pacific ocean area was one that was not well covered.
Is there not a case in some key areas to keep collecting even if the data doesn’t come electronically because of the GDP of the country concerned?
I really think that this community should be thinking about doing something about the great dying. Can’t we organise (surfacestation style) to resurrect some of them by collecting the data and presenting it electronically to NOAA so they have no excuses for not using it?
One other thought — if we were going to do a collection, I would have thought it was sensible to collect all the records for the stations that were going before the great die-off began ie from the early 1990s or whatever that it.
It goes without saying such an operation would need to be well-organised and well peer reviewed so it would need one of you lot who knows about these things to kick it off.
Deb (13:28:27) : Looks like Greenland has been turfed. Are there even any stations left there at all?? I couldn’t find any on the quick pass I just did. Must be all that pesky ice. Odd considering they’re so worried about the place melting and drowning us all…
[chiefio@Hummer 2010]$ country GREEN
431 GREENLAND (DENMARK)
[chiefio@Hummer 2010]$ cd data
[chiefio@Hummer data]$ grep ^431 2010_uniq_station_list
[chiefio@Hummer data]$
Sheesh. Greenland gone too?
OK, this got me worried that maybe I’d blown the file download, so just to be sure, I’ve double checked. The “ls -l” on the file shows it is in the ‘usual’ size range of about 45 MB and with a date stamp from when I downloaded it. New records are scattered through it (sorted by stationID then by year) and it ought to end with the highest station IDs as the last records, and it does. Those “8” records:
[chiefio@Hummer data]$ ls -l
-rw-rw-r– 1 chiefio chiefio 45959144 Feb 8 16:28 v2.mean
-rw-rw-r– 1 chiefio chiefio 103258856 Feb 8 16:47 v2.mean.inv
-rw-rw-r– 1 chiefio chiefio 22529199 Feb 8 16:27 v2.min
-rw-rw-r– 1 chiefio chiefio 742560 Feb 8 16:10 v2.temperature.inv
[chiefio@Hummer data]$ tail v2.mean
8009991400101962 149 141 154 159 172 219 242 256 246 231 204 169
8009991400101963 158 129 143 148 171 189 231-9999 245 226 194 169
8009991400101964 148 123 148 163 167 192 240 258 253 231 194 152
8009991400101965 142 141 156 140 172 199-9999 247 256 237 196 172
8009991400101966 143 146 149 145 176 200-9999-9999 257 231 192 169
8009991400101967 152 144 137 149 158 194 244 260 257 237 185 170
8009991400101968 143 123-9999 161 174 196 239 253 241 244 186 172
8009991400101969 159 151 142 164 197 207 246 251-9999 215 183-9999
8009991400101970 141 136 123-9999 183 197-9999 259 243 225 196 165
8009991400101971 150 128 149 165-9999 206 232-9999 245 221 200 170
[chiefio@Hummer data]$
Though on a re-visit to the ftp site, the meta data for the file give a last modification time of ‘today’… Wonder if they are fixing things again 😉
So if there were a truncation, the “8”s would be gone, and with the last records there (and gunzip not barfing on the unpack) then all the lower station IDs ought to have their 2010 records in place…
So, looks to me like I did the download fine. Though with that mod time of ‘today’ I’m going to do another download and compare them later to see ‘what changed’…
For folks wanting to “try this at home” the ftp URL is:
ftp://ftp.ncdc.noaa.gov/pub/data/ghcn/v2
Tilo Reber (12:51:38) :
Leif:
“What am I missing? I thought we want to drop airports because they show artificial heating…”
Even with that they haven’t managed to get any heating in the last 12 years. The ultimate goal is to extrapolate from the thermometer on James Hansen’s outdoor grill to the world. Will you take your steak well done or carbon only, Leif?
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Tilo:
Outdoor barbeque? Are you mad? We can’t have random weather events affecting the thermometer. A safer place would be inside in front of Hansen’s air conditioner with his hand resting on the thermostat. Foolish man… what were you thinking? 😉
Leif:
I think we could file the dropping of airports under “Be careful what you wish for…” 😀
Would it be possible to create a web site that had three instances of Google Earth, one with gold halo markers on the saved sites and one with red pitchfork markers for the not-saved sites; then one map with just the halo markers and another map with just the pitchfork markers.
I think that it would be a very good illustration of the situation and if it was possible to include a surface station photo of the site that would be informative; in absence of a photo of the site a generic photo of the area would serve a similar purpose. having the last known trend and anomaly would also be nice.
Is that Kobenhaven ( Copenhagen) on the list of dropped stations ?
That will teach them
I will endeavour to obtain the raw data from the Zambian stations mentioned (Livingstone, Mongu and Kabwe) as well as others I know only too well are still in existence and from which data is still obtained (Chirundu, Lusaka {both airport and University} and Chilanga {experimental agricultural station}) I think that the farmers in Mazabuka also keep records, if only for their own benefit.
I will report back in due course!
Burketown Australia was one of a very small number of stations well less than twenty) which had continuous data from the 19th century till now. The Bureau of Meteorology seems to have deleted it in 2009. Only about nine remain; most are cities. In Victoria the only rural stations which still exist are Cape Otway and Wilson’s Promontory. Both show a pronounced cooling since the 19th century.
Heads up – CNN reporting a school shooting at UAH Shelby Center. Isn’t that where John Christy works? Hope things aren’t too bad over there. If I was the type, I’d be saying prayers.
http://www.reuters.com/article/idUSTRE61B59V20100212
Warren Berman (13:29:48) : Maybe somebody can answer the following:
The bits I know:
a. how are these temps measured, by what kind of instrument, and are all of them worldwide the same?
They vary a lot. Early on was mercury in glass, recent is electronic. There have been questions raised about some of the calibration and cross calibration…
b. If these current instruments are digital readouts, capable of measuring to a few tenths, what was used prior to digital?
Plain old liquid in glass. Read (for the USA) to 1/10 F then ROUNDED by the observer to whole degrees F that were sent in on paper sheets. Lacking a reading, the NOAA guidelines say the observer is allowed to GUESS.
In other countries, I don’t know the rounding policy nor do I know if the electronic ones also get rounded.
c. how are all of them worldwide calibrated?
Some are good, some are lousy. Hawaii was “hot” for a few months before anyone figured it out. I think they did not bother to remove the ‘new record’ that was set…
2. My point is: How can anyone say that temperatures have risen by X amount, when in all probability that amount is within the variance for error of all instruments used, especially prior to digital readouts, and especially since the number and location of sites seems to vary constantly, and since many are exposed to urban growth around them.
BINGO!
All that 1/10 C and especially the 1/100 C numbers are just playing in the error bands of the process.
Now you will get folks who try to assert that the law of large numbers and the central limit theorem will let you know the AVERAGE to 1/100 C even if the data are dodgy. But in GIStemp, for example, you carry temperature data all the way to the ‘fill in the GRID / BOX’ step where the (now 1000 ish) thermometer records fill in 8000 boxes. Yes, 8 boxes per thermometer. So you don’t get a lot of ‘large numbers’ in that average. You get, for the monthly anomaly, ONE highly processed number for many boxes. Said number starting life as an average of Daily (MAX+MIN)/2 so you have at most 31 of them that then get averaged.
The “Central Limit Will Save Us” folks like to trot out 7000+ thermometers with a gazillon records and start calculating the theoretical precision on the average of all of them. They get grumpy when you say the code doesn’t do that…
They get grumpier when you say it’s only 1000 real thermometers (now) and that many of the things they are averaging are just made up “fill in” data based on that other crappy data and a dodgy method.
They get very much grumpier when you say that after the code has processed it a dozen times or more they will be lucky if the original 1 C can be held steady as “scientific programming” has been found to do a terrible job of keeping precision in their calculations:
http://chiefio.wordpress.com/2010/02/07/of-hypothetical-cows-and-real-program-accuracy/
(For example, in GIStemp I found a single line that ‘warmed’ 1/10 of the records by 1/10 C due to a bit shift in the order in which they did the F to C conversion… )
So I’d suggest not talking about it too much unless you really want to have a lot of grumpy “scientists” in your face claiming you don’t understand statistics (when you understand them just fine…) 8-}
But you’re right… 😉
Just normal bureaucratic SNAFU, too many chiefs and not enough indians. Must save money, layoff low level workers. Must reduce load on remaining workers, reduce data entry and cover up with software. Not a problem, we only need a few data entrys, the rest are simulated in the software.
The raw data is likely still out there and still being gathered. Just not being entered into the computer data base. The bureaucratic system to record the local weather data was set up to operate with very little or no on budget cost and oversight, so it is most likely still in operation. Bureaucratic inertia is one of the most powerful forces in the world.
Getting that information into the computer database is the bottleneck as that costs large amounts of money in one place and budget. Now there are only 3 data bases that need to be controlled to set policy. As long as they can control the output, they control the argument, and the raw database entries are made at one time and shared to save time and money. Each of the 3 data bases are then modified to suit the needs of the holder. Do I correctly understand the process?
Before everyone hyperventilates on this topic, there are some questions to ask.
Is there a principled method for either selecting or deselecting a reporting station?
Does the principled method make sense scientifically/statistically-how was it derived and tested?
Is the method as described actually being practiced?
If these questions can be answered affirmatively, all the huffing and puffing doesn’t mean too much. The temperature compilers should be credited for creating an accurate collection method that reduces workload and complexity.
If the answer to any of the questions is a negative, or if there is no available elaboration of the methods and justifications, then there are lots of worthy questions and alternative analysis to be done.
If one wishes to have a conversation about science, one should adhere to some form of scientific approach…a stepwise analysis of the factors and methods one wishes to discuss and criticise. Otherwise it is just a bunch of barstool blowhards bloviating.
E.M.Smith (14:37:46) :
For folks wanting to “try this at home” the ftp URL is:
ftp://ftp.ncdc.noaa.gov/pub/data/ghcn/v2
Dang it E.M.!!!! Now I’m running a muck in their ftp site. You know they’ve actually a file named virusnt.exe? Now I’m in a quandary. I’d like to run it and see what it does, but if it loads a virus, how would I complain? It says virus…..lol
http://1.bp.blogspot.com/_Xns3mMGw2fU/S2ipbBw-EyI/AAAAAAAACCY/jAU3IZ4HYfk/s1600-h/tree+ring+circus.jpg
“E.M.Smith (15:09:21) :
[…]
(For example, in GIStemp I found a single line that ‘warmed’ 1/10 of the records by 1/10 C due to a bit shift in the order in which they did the F to C conversion… )”
They do know about this thing called floating point numbers, do they, those “scientist” people?
hswiseman (15:10:56) :
See the E.M.Smith (13:23:08) : post. There are actually 2 different explanations but neither make much sense and in these recent cases the actions (deleting from database) are contradictory to the explanation. E.M. has provided the links in his post.
A bit of analysis of the deleted stations
Urban 214
Suburban 105
Rural 166
Median elevation 160 metres
Airport 196
Non Airport 289
“hswiseman (15:10:56) :
[…]
If one wishes to have a conversation about science, ”
Maybe you can contribute a convincing idea as to why they drop the thermometers instead of denouncing this debate? Do you think Gavin Schmidt’s explanation that they have far more data than they needed is even slightly convincing? Or in any way scientific?
Tom in Texas (14:13:14) : EMS, have you (or anyone else) checked to see if V1raw = V2raw?
Well, they are not exactly “raw”…. The official term is “unadjusted” for GHCN. For USHCN they are explicitly adjusted and USHCN is significantly colder than USHCN.v2 but for GHCN I think there have been some spot checks showing GHCN.v2 “unadjusted” is close to the original reports. Don’t know about GHCN.v1 but I think it can be downloaded from the same ftp site listed in comments above (just go above the ‘v2’ part of the link 😉
Oh, and there is also a “GHCN Adjusted” that tells you how NOAA would like to re-imagine the world and that is highly different from ‘unadjusted’ with the past made much colder…
Frank Mosher (13:52:48) : E.M Smith. Much thanks for your tireless, concise, and easy to understand analysis. fm
You are most welcome. I attribute the ‘tireless’ part to massive infusions of coffee (Irish and otherwise 😉 along with tea… 😉
@george: Say “Hi” for me! And there is an email address buried in the ‘about’ box on chiefio.wordpress.com if they want to send me an email.
@James Sexton: Be Advised: Russia has a European number (635?) for everything West of the Urals and an Asian number (222?) for everything to the east… so you have to check both continents to see WUWT for Russia…
@Margaret: http://www.wunderground.com/ has the reports for most of the GHCN dropped thermometers. I’d also wager there are METARS and other on line sources. A lot of a ‘new collection’ can be done from a comfortable keyboard at home…
E.M.Smith (14:37:46) :
Sheesh. Greenland gone too?
Ah, but it seems there might still be a couple thermometers left in Newfoundland within that 1200km band that can correlate north. I mean if the beaches in Peru and the Amazon jungle are good enough for Bolivia then surely the rocky harbours of Nfld/Labrador or the woods of Quebec are a good enough match for a giant ice cube. 😉 /sarc
Thanks for checking. You’ve been doing really excellent work sorting through all this data.
I really don’t understand how these people think they can get away with this sort of thing. Like many of us here, if I ignored and deleted data like this while doing my job I’d be fired!
Anthony:
I had a look at the two A-stations in Germany:
61710020000 LIST/SYLT 55.02 8.42 29 0R -9FLxxCO 1x-9WATER A
Looking at Google Earth there seems to be no stucture (51.50 9.95) which could represent a Stevenson Screen. Even if so it would be plain rural and nowhere near an airport (A for Airport). The Sylt airport is more to the south near Westerland: 54,91N 8.34E
61710444000 GOETTINGEN 51.50 9.95 171 214U 124HIxxno-9x-9WARM DECIDUOUS A
Looking at Google Earth there seems to be a stucture (51.50N 9.95E) which could represent a Stevenson Screen. Even if so it is plain rural and nowhere near an airport (A for Airport). The city of Goettingen doesn’t know anything about an airport (website).
Please look and correct me.
Are they removing stations from fictional airports just to tell us “We removed the airport stations and left the rural ones”?
I’ve just made a quick scan of the comments and worthy almost all are but didn’t see any reference to your superb headline.
You can run, you can hide the data but The Langoliers will come through the trees and munch it all.
(That will have lots of people saying “What’s this guy on?”)
From the Yale article’s explanation:
For many of the world’s stations, observations are still taken and recorded by hand, and assembling and digitizing records from thousands of stations worldwide is burdensome.
If they’ve been digitizing the records, that makes the claim that the earlier temperature records were dumped questionable. Either the raw data was digitized and is still available, or someone is telling tall tales. Yale dropped the ball by not following up on that assertion.
I have another question: when someone reads a temperature from a mercury thermometer, and it’s, say, 65.7° F, or 18.7°C [as close as the divisions on a glass thermometer can be eyeballed], do they just average up and record 66° or 19°? What temperature do they record, if it looks like exactly 65.5° F, or 18.5°C?
A change in policy could account for this divergence: click
And “adjusting” the raw data can dramatically alter the trend line: click
Also, the comment mentioning that digital thermometers are accurate to within a few tenths of a degree is technically accurate. That would be for a thermometer in calibration, with its calibration traceable to N.I.S.T. [formerly the National Bureau of Standards].
But in practice there are ascending levels of calibration. The calibration interval can be lengthened or reduced. And the thermometer can be calibrated to an instrument standard, or to a primary physical standard like the triple point of water.
Under the current system, a worldwide temperature change of a fraction of a degree can not be observed with any confidence. And the record can be manipulated in other ways, such as those mentioned in this article, making the temperature record unreliable.
For those who haven’t seen it, here’s a blink gif of NOAA raw vs adjusted temperatures going back to 1880: click [give it a few seconds to load]
So, no thermometers left for Czech Republic at all??
(Oldest temprature record -regular since 1775- in Central Europe is Praha-Klementinum record – incorporated by GHCN in the record of Praha/Ruzyne – or more exactly “connected” into international airport Praha-Ruzyne record – now also dropped?)
I was just looking at Giss and found maybe interesting things:
1. they state they produce their world anomaly maps from GHCN “Global Maps from GHCN Data” (http://data.giss.nasa.gov/gistemp/maps/)
2. I looked absolutely randomly in the 2009 data for the stations from Chiefio list at
(http://data.giss.nasa.gov/gistemp/station_data/)
and
NEW YORK CENTRAL PARK – data AUG-DEC missing
BARCELONA – data SEP-OCT missing
ILES GLORIEUS – data SEP-OCT missing
NALUT – all except JAN missing (very interesting record…)
CHIFENG – APR, SEP-DEC missing
KRMANSHAH – MAY, AUG, DEC missing (very interesting record too)
SAIKHAN-OVOO – APR missing
TERESINA – JUL, AUG, NOV, DEC missing (aso very interesting record)
LAMENTIN/MARTINIQUE/FT DE – SEP-NOV missing
etc. etc.
– one wonders how anybody on earth can – in the era of permanent manned space missions – use such incomplete records for global temperatures analysis?? …and now I’m fully convinced Hansen urgently needs a custody…
In all cases the Giss 2009 global anomaly and temperature maps are just a big nonsense or lie.
One hardly finds a station without some 2009 data missing. What this people at NOAA, NASA are doing? What for the taxpayers and others pay them? To have records even a high school student would be ashamed of? Unbelievable!!
This whole thing stinks like whole that prospering bunch of polar bears around the “mostly no GHCN, GISS thermometers at all” Arctica…
Simply unbelievable. 1500 stations are too much to track? BS. I work at a very small electric utility. We have about 3100 meters. I can graph each meters hourly use each day. Of course, from time to time, I don’t get reads on all my meters all the time. I don’t throw out the usage data. And I don’t average or extrapolate some of the meters usage to give my total usage to my boss. I’m expected to be correct. I would assume the NOAA has a bit more resources at their disposal than my little COOP. I suppose I could volunteer to give them a bit of instruction as to how to properly manage such a huge(snicker, snicker, lol, rotflmao, rotflmaopmp!!!) database. Sigh, now to clean up and go drink some beer contemplating the case of the disappearing thermometers. Cheers to all!!!
Removal of entire regions (like the Arctic) makes for -9999’s in the data set.
They show up as pegged in the scale on the homogenized map.
Rather ingenious method of creating ‘hot spots’ in lieu of no data.
Following the trend of station removal/relocation you get an entirely new data set that cannot be calibrated against the old.
Satellite data needs to be calibrated to ground. Simply alter the ground stations and spice to taste. Move data collection equatorward for the winter, homogenize areas lying poleward to new improved warmer locations.
Walla …. It’s getting hotter faster than previously arranged.
Who said “Climate data ‘not well organised’ “?
Who “agreed that two periods in recent times had experienced similar warming. And he agreed that the debate had not been settled over whether the Medieval Warm Period was warmer than the current period.”?
http://news.bbc.co.uk/1/hi/sci/tech/8511701.stm
I hear fingernails on the granite face
You think bookies in Vegas would buy that..?
If they were in charge of the adding.
DirkH (15:22:38) : They do know about this thing called floating point numbers, do they, those “scientist” people?
They know about them, they just don’t use them very well…
Details:
http://chiefio.wordpress.com/2009/07/30/gistemp-f-to-c-convert-issues/
Oh, and this one too:
http://chiefio.wordpress.com/2009/11/21/hadley-hack-and-cru-crud/
where the CRU crew ‘fix’ a bad handling of overflow via removing a bad data item from the input and leave the program broken…
In today’s OBAMA 2010 DEFENSE REVIEW on security threats to America, it did not include the word ‘Iran’ or Islamic. The entire report is 128 pages long. There were 8 pages devoted to the dire threat to the USA from global warming.
Maybe the stations that had to be shot were naughty for telling the truth?
Here some more info on the IPCC made up temperatures for Scandinavia:
http://hidethedecline.eu/
The site also has an A – Z of sceptic viewpoints and knowledge for all newcomers that want fast effective introduction to the climate debate.
K.R. Frank
I am not denouncing debate in the least, merely saying that debating the deletion of thermometers without debating the underlying rationales is an utter waste of time. I don’t presume that the reasoning is either perfect or utterly flawed. I don’t know. I try to avoid ascribing motives (it is getting harder and harder). And it is also possible that once pointed to the methodology that I don’t have the science chops to do the analysis myself. In which case I contribute to the debate how? Unless someone does the work for my benefit I am just making guesses. I can (and sometimes do) throw out a layman’s observation or common sense analysis, and like a blind squirrel, I find an acorn now and then. Some very smart folks have been kind enough to challenge or validate my reasoning from time to time and I have learned alot.
The deletion of thermometers and reducion of observational science is interesting. Check. Now proceed from first principles. It is alot more interesting to debate the validity of method, reasoning, theory, process, etc, than the fact that any particular station has been included or excluded absent context. E.M. Smith says he has done the analysis of the justifications/methods and has found them unconvincing or otherwise unsatisfactory. I appreciate his undertakings but is he the last word on this? Smith makes no such claim himself, and I suspect he would be happy to discuss, debate and argue the matter with someone representing the the GHCN perspective. If they showed that they were right, I believe he would agree with them. But Smith and anyone else with the temerity to even ask a question has been excoriated and treated as someone beneath contempt, unworthy of acknowledgement or engagement. I don’t care to perpetuate the disgraceful attitude of the climate establishment with my own prejudice, paranoia and contempt. I have curiosity about nature and I think that it is important to understand what impacts we are having on the planet with the discharge of CO2 and other pollutants. Kneejerk conspiracy theories don’t advance anyone’s knowledge, nor does a presumption that I know all the answers before the questions are asked.
b.poli (15:50:58) : Are they removing stations from fictional airports just to tell us “We removed the airport stations and left the rural ones”?
No, just that the meta-data are really crappy. But that’s OK, it’s not like they use the URS, elevation, population, etc. data to control the processing or that they expect to direct the world economy based on the results…Oh, wait, they are 8-{
“E.M.Smith (16:13:44) :
[…]
Details:
http://chiefio.wordpress.com/2009/07/30/gistemp-f-to-c-convert-issues/
”
I see this incredible sloppy line
nint( 50.*(temp-32.)/9 )
and your analysis about the compiler dependency. For non programmers: they failed to define explicitly whether the program shall multiply with 50.0 first or divide by the integer value 9 first. Worse, why even mix integer and float constants here? This is a precision-killing bug, a reliability foulup, a numeric SNAFU. Probably nobody ever reviewed the code and nobody ran a detailed test.
Strange that they put Indonesia in the Pacific area. They must have had pretty old maps if they class Jakarta as Paddy Fields. Wikipedia (I know I should not quote that but it puts in the direction) states that the greater Metropolitan area has 23 million people (2004) and is second largest in the world. I know that it takes a couple of hours on main roads and freeways to get out. Once out, you get into the hills and mountains. There are many mountains over 2000m.
I would image that Bandung (an old colonial Dutch settlement with a catholic cathedral) in the centre of the south-east of Java should have a long weather record dating back a couple of hundred years. I think it was the seat of Dutch government and that there is a University there.
To go along with all the talk of global temps. the January record at UAH is making the rounds in the blogosphere, like here
http://www.techeblog.com/index.php/tech-gadget/hottest-january-on-record
This blog has a lot of interesting stuff, the bad news is that they decided to get their information/explanation from Joe Romm.
DirkH (16:30:04) :
“nint( 50.*(temp-32.)/9 )”
Is this a temperature scale conversion, F to C, if so should the 50.0 not actually be 5.0?
Fascinating quote from a Phil Jones-BBC interview today. This is part of his answer to the first question that the interviewer asked. Note the first sentence.
“…Temperature data for the period 1860-1880 are more uncertain, because of sparser coverage, than for later periods in the 20th Century. The 1860-1880 period is also only 21 years in length. As for the two periods 1910-40 and 1975-1998 the warming rates are not statistically significantly different (see numbers below)….”
If the data are more uncertain because of sparser coverage, it follows that the same rule must apply to deleting stations today. Did Jones go “off message” here?
news.bbc.co.uk/2/hi/science/nature/8511670.stm
A caution on this one: I get 1179 stations
http://www.dplot.com/agw/station_count.png
from ftp://ftp.ncdc.noaa.gov/pub/data/ghcn/v2/zipd/v2.mean.zip, not 1113. And since our 2009 figures agree (1597 stations) I assume we’re using the same source. Mr. Smith may be right and stations are dropping like flies… but maybe the count will go up tomorrow and the day after that and the day after that. It doesn’t seem unreasonable to assume that some stations are just slow to report.
one would think that as AGW is such a professed ‘crisis’ that the number of temp measuring stations would be increasing rather than decreasing. Just goes to show the way GISS think, that the opposite is the case.
The Langoliers! Now that is sooooooo apropo! In that schlocky B grade flick, the langoliers were chewing up chunks of space/time or some such, as I remember, and threatening to cut the woeful aircraft and passengers off from a return to their reality. It works really well as analogy for the unreal mastication of GHCN reporting stations and data sets by the NOAA Langley-Liers! As more recording sites are dropped and the data sets backfilled with approximations from the remaining sites, we get analyses that are more and more estranged from scientific reality.
A very good analogy, indeed! Esoteric… but Good!!!
hswiseman (16:21:50)
“I have curiosity about nature and I think that it is important to understand what impacts we are having on the planet with the discharge of CO2 and other pollutants. ”
What is your basis for concluding that CO2 is a pollutant? How does a gas essential for all flora growth and, as flora is fodder to fauna, the root of all fauna growth on the planet as well, justify classification as a contaminate?
Regarding DFW. There are 10ths deg C:
rbroberg@ubuntu910:~/Downloads$ ls *failed*
v2.max.failed.qc v2.mean.failed.qc v2.min.failed.qc
rbroberg@ubuntu910:~/Downloads$ grep 425722590000 *failed* | grep 2010
v2.max.failed.qc:425722590000 2010 1 -22
v2.mean.failed.qc:425722590000 2010 1 -62
v2.min.failed.qc:425722590000 2010 1 -103
So NOAA was showing a mean monthly average of -6.2C.
NOAA GHCNv2 did the right thing tossing DFW out of the data set.
But what caused this obviously erroneous data range?
NOAA processing or DFW reporting?
Don’t have an answer to that.
And just fyi,
here is my station count …
(using v2.mean.Z not v2.mean.zip)
rbroberg@ubuntu910:~/Downloads$ cat v2.mean | grep 2010 | cut -b 4-9 | sort -n -u | wc
1187 1187 8309
I can understand removing some stations in countries with endless wastes of flat land where there is nothing much in the way of geographical features to affect temperature-chaging microclimates. But in countries with many mountanous areas it’s incomprehensible.
I’m not sure where exactly the deleted temp station is/was at Clermont Ferrand, but I used to live about 10 miles S/W of the city. It’s set on the edge of a plain to the East, with a quite sizeable mountain range immediately to the West, which retains its snow cover til early summer (esp on the Puy de Dome). I lived in a South facing village built into a rock face, which has been occupied since Mesolithic times. We could sit on our terrace in tee shirts in mid winter bathed in sunshine whilst across the village valley on the facing slope the temp stayed well below freezing all day, every day, for months.
The entire area has any number of wildly differing microclimates within a few sq miles: siting of a temp station would be an interesting exercise! Gavin Schmidt’s remarks as quoted above have left me with even less respect for his integrity than I had before.
The terrifying thing is that these arrogant people still have the ear of policymakers worldwide. And their ‘useful idiots’ – teachers and politicians – have infected all our children
p.g.sharrow “PG” (15:09:50) :
With the bajillions of $$$ being spent on “climate studies” and with the massive “WE’RE ALL GOING TO BAKE!!!” group-think among the college crowd one would think they (GISS & Co) could nab a few (dozen) interns to do some of the work.
David Hyde (17:18:51) : It doesn’t seem unreasonable to assume that some stations are just slow to report.
There are only a handful of stations listed as having failed QC in 2010. I suspect that you are right to expect to see more of the ‘missing’ stations show up in the v2.mean file.
rbroberg@ubuntu910:~/Downloads$ grep ” 2010 1 ” v2.mean.failed.qc | cut -d ” ” -f 1 | sort -u -n
109646550002
122637400001
137616790003
207420710003
207424750003
304890560004
314804160003
314804230003
425722590000
425725970000
643082150001
E.M.Smith,
I think you might want to keep your powder dry on the number of stations. A lot of stations may not show up until a few months into the year, due to late reporting. The list you have now would only include those having already reported January temperature.
You need to do a file compare and list any new additions as well.
Andrew30 (16:57:24) : Is this a temperature scale conversion, F to C, if so should the 50.0 not actually be 5.0?
The data are in 1/10 C increments, so 102 is 10.2 F.
Ron Broberg (17:59:43) : here is my station count …
(using v2.mean.Z not v2.mean.zip)
rbroberg@ubuntu910:~/Downloads$ cat v2.mean | grep 2010 | cut -b 4-9 | sort -n -u | wc
1187 1187 8309
Two bits of ‘finesse’. One is that you need to grep for …………2010 to avoid getting any station ID that has 2010 in the middle of it counted as a ‘year’ (that is, you don’t want the hypothetical station 42520101000 to be counted as the year 2010) and I conflated any stations with the same station ID (8 digit) but different ‘modification flags’. That is, I count 425123450001 and 425123450002 as the same station.
Also, on the ‘failed QA’: This is speculation, but do you think that snow in Pensacola Florida will cause it’s February data to “fail QA”? The simple fact is that we’re getting historic snow (and the attendant cold) from our shift to a colder regime (IMHO due to a sleepy sun and the PDO flip to the cold phase) and if we’ve got gatekeepers on the data that don’t believe those values are ‘real’ then they will ‘fail QA’.
So I’d like to have assurance that tossed records are really bad, and not just being ‘unbelieved’.
Inconceivable! May statistics are reported years later, as in 2006 is the latest available data. The IPCC/NOAA/NASA should try to use every resource available. The most recent 1/3 to die were presumably reporting in real time otherwise they would have been purged in the early 1990’s; so what has occurred now to warrant this latest extinction?
They could/should have a preliminary report based on the real time stations complete with their manipulations and/or corrections. They should then report based on all stations, using corrections for missing data where required at some point after most of the data has been received. In addition, since we are talking computers here there should be no problem adding a late reporter.
The *bonus* from an approach like this would be that the favored real time corrected/manipulated data set temperatures from a given month could be compared to the *actual* reported a few months later. This would verify or repute the ever smaller real time data, and allow corrections to the smaller data set to make the real time data much more accurate. But until we have this comparison we skeptics have no faith in the shrinking real time data set.
Why is this not done? It’s not like we’re going to implement something drastic that has to have up to the minute data.
Bob Koss (18:24:27) : I think you might want to keep your powder dry on the number of stations. A lot of stations may not show up until a few months into the year, due to late reporting. The list you have now would only include those having already reported January temperature.
The claim is that once the due date passes for a month they don’t go back and put stations in who report late. That was one of the excuses I ran into on Bolivia (that they reported a few days after the ‘cutoff’ for the month, each month). While I’d love to hear that they were accepting late updates, I’d also want to see the changed procedure published… and Bolivia could then be put back in.
The ‘readme’ on the ftp site says:
“New monthly data are added to GHCN a few days after the end of
the month. Please note that sometimes these new data are later
replaced with data with different values due to, for example,
occasional corrections to the transmitted data that countries
will send over the Global Telecommunications System.”
So they say they put the data set up a few days after the end of the month ( I got it on the 8 th which seems to be about when they first put things up) and the only caveat on it is that individual values might change from corrections later, not that whole countries might be slipped in later. “Occasional” does not mean 30% to me… and ‘data replaced’ is not “whole countries put in”.
Further down there is a statement that ‘QA failed’ data might be put back in if someone can convince them the data are real. So folks in Dallas might want to consult the January newspaper reports and see if it was frozen the whole month 😉
So, no, I don’t see much reason to ‘keep power dry’. I could easily see a single digit number of thermometers change from “QA failed” being itself QA’d but if they start putting 5% or more of the records ‘back in’ I think they are violating their published methodology… (though I’d welcome it). I’d also welcome them putting the thermometer record in the v2.mean data set with the required -9999 missing data flag if a site has failed QA but has not been deleted. That is what they are supposed to be doing, but the DFW case implies that a site is simply left out until (whenever) in a year it gets it’s first valid datum, then the whole year-to-date of missing data flags get written. Sloppy at best.
Oh, and GIStemp would need to put a caveat on all those monthly horror reports that they are being made on partial data and not to be trusted.
But basically, I’d be fine with them putting in the readme file that the data are not to be trusted when published for a few months… if that’s what they are doing.
Mac
The monotonic increase in C02 is well established and the preponderance of the evidence supports a human source.
http://www.esrl.noaa.gov/gmd/webdata/ccgg/trends/co2_data_mlo.pdf
A discharged material can be pollutant even if it is naturally occuring and not otherwise harmful. We regularly emit C0 in small, non-lethal amounts into the atmosphere. Like C02, in high concentrations C0 will kill you. Not too many people would argue that tailpipe C0 is not a pollutant. The world at large has made a cost-benefit determination that creating additional C0 provides so many benefits to man-kind that it tolerates and adapts to the small burdens of potential C0 poisioning in closed spaces. The cost-benefit analysis for C02 cannot be conducted accurately if the science through which cost is determined is faulty or incomplete. Burn fossil fuels and create energy, plant food and some as yet undetermined increment of infrared radiation absorbtion. The benefits attached to the creation of C02 are pretty obvious. The costs are in dispute, but most likely less than the doomsday crowd would have you believe. Whatever the costs, I would argue that the adaptation burden should be borne by the parties with the lowest cost of risk management. Displacing a 1000 inundated islanders makes more sense than beggering the entire world population. The actual cost of displacement should be borne by the parties that garner the most benefit. I would I respectfully disagree if your argument is that we should not even bother to do the calculus.
It’s a bit disturbing when the only weather stations that remain are in the CBD of a city.
I found another local source of weather data in the local Department of Agriculture
(http://www.agric.wa.gov.au/PC_93316.html) here in Western Australia.
Although the nearest station (http://agspsrv34.agric.wa.gov.au/climate/clig/climinfo/awsdata/MD.htm) doesn’t have any really old data online from before the station was automated, th period since 1995 to present shows no warming at all. Temperature follows insolation very closely (R^2 > 0.8); delayed by between 1 and 2 months; presumably due to thermal capacity.
E.M.Smith (19:29:57) : Two bits of ‘finesse’…
Yup. Good catch. I stand corrected.
(also, the field range for the 5 char station id is 4-8)
grep ………..2010 v2.mean | cut -b 4-8 | sort -n -u | wc
1179 1179 8253
So I am now in agreement with David Hyde.
So I’d like to have assurance that tossed records are really bad, and not just being ‘unbelieved’.
Not hard to figure out.
Just grab the daily data at WU for January and calculate.
Here is the link for Jan 1, 2010
http://www.wunderground.com/history/airport/KDFW/2010/1/1/DailyHistory.html?format=1
I tweaked the scripts here for Tmean using Tmax,Tmin presented here:
http://rhinohide.wordpress.com/2010/02/07/charles-pierce-methods-of-monthly-means/
I calculate the monthly mean for DFW in Jan 2010 to be 44.4F = 6.9C
That is significantly higher than the -6.2C that NOAA threw out. (sign inversion?)
How does that compare to previous NOAA v2.mean for DFW for January?
(recall that v2.mean is in 10thsC so the number we are comparing is 69) …
for i in `seq 2001 2009`; do grep 425722590000 v2.mean | grep “$i ” | cut -b 1-21; done
4257225900002001 59
4257225900002002 86
4257225900002003 64
4257225900002004 92
4257225900002005 98
4257225900002006 128
4257225900002007 57
4257225900002008 83
4257225900002009 90
It obviously lies in the low end of the range for the decade, but not the lowest. Out of curiousity, here are the previous years with lower DFW Jan mean temps per NOAA v2.mean ….
grep 425722590000 v2.mean | cut -b 1-21 | sort -r -n -k2
4257225900001968 69
4257225900001997 67
4257225900001960 67
4257225900001964 66
4257225900001974 65
4257225900001957 65
4257225900002003 64
4257225900001983 64
4257225900001996 61
4257225900001991 60
4257225900002001 59
4257225900001973 59
4257225900001959 58
4257225900002007 57
4257225900001988 57
4257225900001961 49
4257225900001970 48
4257225900001966 46
4257225900001962 42
4257225900001984 41
4257225900001949 39
4257225900001948 35
4257225900001985 32
4257225900001963 32
4257225900001979 19
4257225900001977 15
4257225900001978 10
This still doesn’t tell us which end of the pipe broke with DFW data reported to NOAA. I did a quick peek into the GHCN daily ftp directory.
ftp://ftp.ncdc.noaa.gov/pub/data/ghcn/daily/all/USC00412242.dly
Only 10 days of data with Tmax and Tmin with GHCN daily. But these 10 days also yield a Tmean of 4.6C. So two independent methods are converging on 4C as being the about right for DFW for Jan 2010. Still not sure how the -6.2 ended up in the v2.mean. But NOAA GHCN was correct to reject it. Nothing in the GHCN daily file resembles the numbers that GHCN holds in v2.max and v2.min for DFW.
Bah. I need an editor. Strike this: So two independent methods are converging on 4C as being the about right for DFW for Jan 2010.
The point about converging on 4C is in error. The Weather Underground wx data yields 6.9C. I had earlier mistakenly used the dew point as Tmin – that erroneous calc yielded a bad Tmean of 4.somethingC.
E.M.Smith (19:29:57) :… and I conflated any stations with the same station ID (8 digit) but different ‘modification flags’. That is, I count 425123450001 and 425123450002 as the same station.
My method does the same. By selecting all 2010 records, but then cutting just the 5 char station id and removing dupes, I only count station ids and multiple records for a station are discarded. Returning the same number of 2010 stations as David Hyde. Maybe you can post your station counting code …
grep ………..2010 v2.mean | cut -b 4-8 | sort -n -u | wc
1179 1179 7074
@ Deb (13:28:27) :
Nah. All these stations remain in GHCN for Jan 2010.
grep ^431 v2.mean | grep ^…………2010 | cut -b 4-8
04220
04250
04312
04320
04360
04390
To get their names, we can grep it out of v2.temperature.inv
for i in `grep ^431 v2.mean | grep ^…………2010 | cut -b 4-8`; do grep ^…$i v2.temperature.inv | cut -b 1-40; done
43104220000 EGEDESMINDE
43104250000 GODTHAB NUUK
43104312000 NORD ADS
43104320000 DANMARKSHAVN
43104360000 ANGMAGSSALIK
43104390000 PRINS CHRISTI
E.M.Smith (19:52:09) :
I have several of their databases from the past couple years. Here are some of the download dates and number of stations containing data for 2008.
02-08-2008 928
11-14-2008 1233
01-20-2010 1613
I mentioned in a post at CA a couple weeks ago they added 100’s of stations back in at the beginning of 2009.
http://climateaudit.org/2010/01/30/lazarus-of-the-thermometers/
The two databases from 2008 indicate they add them as they get data throughout the year.
E.M.Smith (19:52:09) :
“Bob Koss (18:24:27) : I think you might want to keep your powder dry on the number of stations.”
So, no, I don’t see much reason to ‘keep power dry’.
Well, in thinking about this I decided that I was depending on NOAA:
1) Following their own statements.
2) Doing things that make sense and are consistent.
3) Following proper professional data set update standards.
4) Assuring that a broken (i.e. un-ripe) data set is not released.
5) Having a largely automated and standardized process for doing things.
Basically, I’m expecting professional standards of behaviour that may not be in evidence… The more I pondered, the more I realized there is little reason to expect any of those 5 behaviours given the things we have seen from CRU, GISS, GIStemp, et. al.
So I decided to go looking for the published “availability date” standard.
I could not find one.
If they have one, it’s well hidden from a casual user of their data.
But what I did find was a letter in the FOI set at this web site:
http://junkscience.com/FOIA/mail/1226959467.txt
That paints a rather haphazard picture of the ‘update’ process. Given that haphazard method, I’ve decided there really IS no reason to think that the GHCN data set is ever “done” or “ready”. It is just a “work in progress” and is a bit “slapdash” at any time…
Some quotes:
From: Phil Jones
To: Gavin Schmidt
Subject: Re: GHCN
Date: Mon Nov 17 17:04:27 2008
Gavin,
First the figures are just for you – don’t pass on!!! I don’t normally see these. I just asked my MOHC contact – and he’s seen the furore on the blogs.
So about a year and 3 months ago. Probably still what happens…
These 3 paras (below) are from the GHCN web site. They appear to be the only mention I can see of the WMO CLIMAT network on a web site.
I could not find much on the web site either, but perhaps searching with chunks of the material you are looking for as keys would find it? Something to explore later…
The rigorous QC that is being talked about is done in retrospect.
They don’t do much in real time – except an outlier check.
OK, so QC is sort of an after the fact glue on… and the web pages are doing some sellers puff about “rigor”.
Anyway – the CLIMAT network is part of the GTS. The members (NMSs) send their monthly averages/total around the other NMSs on the 4th and the 18-20th of the month afterwards.
So, by the 8th the data set OUGHT to have been complete. But it comes around again on the 18-20th. That means we might get the data ‘fixed’ next week… but maybe by the end of the month for sure?
Few seem to adhere to these dates much these days, but the aim is to send the data around twice in the following month.
Or NOT…
Data comes in code like everything else on the GTS, so a few centres (probably a handful, NOAA/CPC, MOHC, MeteoFrance, DWD, Roshydromet, CMA, JMA and the Australians) that are doing analyses for weather forecasts have the software to pick out the CLIMAT data and put it somewhere.
“put it somewhere”… that’s comforting…
At the same time these same centres are taking the synop data off the system and summing it to months – producing flags of how much was missing. At the MOHC they compare the CLIMAT message with the monthly calculated average/total. If they are close they accept the CLIMAT. Some countries don’t use the mean of max and min (which the synops provide) to calculate the mean, so it is important to use the CLIMAT as this is likely to ensure continuity.
“how much is missing” flags… “If they are close”… now there is a fine standard metric of acceptable error band. /sarcoff>
If they don’t agree they check the flags and there needs to be a bit of human intervention. The figures are examples for this October. What often happens is that countries send out the same data for the following month.
“a bit of human intervention”… “often happens” “same data”. So, we have no idea if there are loads of bad data that was just broken in both the dailies and the monthlies (if they are both broken by, oh, reading a sign wrong on the dailies, it will just sail through?!) or if blocks are just repeated because what ‘often happens’ happens? And we have no idea what a “bit of human intervention” is, or if there are standards for it or for how long it might take. AND we wait until the 18-20th to get a second bite at the apple and hope one of them works out to be right…
This happens mostly in developing countries, as a few haven’t yet got software to produce the CLIMAT data in the correct format. There is WMO software to produce these from a wide variety of possible formats the countries might be using. Some seem to do this by overwriting the files from the previous month. They add in the correct data, but then forget to save the revised file. Canada did this a few years ago – but they sent the correct data around a day later and again the second time, after they got told by someone at MOHC.
So if someone notices a really bad screw up, they can get the data sent around again. Sort of whenever. In some format or other. Possibly overwritten with old values, if they remembered to save the file..
My guess here is that NOAA didn’t screw up, but that Russia did. For all countries except Russia, all data for that country comes out together. For Russia it comes out in regions – well it is a big place! Trying to prove this would need some Russian help – Pasha Groisman? – but there isn’t much point. The fact that all the affected data were from one Russian region suggests to me it was that region.
Probably not of much use to an FAQ!
Cheers
Phil
And some chunks of countries could get screwed up too, but hey, it’s not like you can find out if it was screwed up or when or by whom or whatever “there isn’t much point”… so why bother. It’s only data…
There are then three paragraphs of ‘sellers puff’ quoted from the web site that looks oddly familiar, vapid, and empty.
Then the Gavin statement to which the reply was sent, and the original complaint from Jones that caused Gavin’s reply. I’ll just include Gavin’s bit:
At 12:56 17/11/2008, you wrote:
thanks.
Actually, I don’t think that many people have any idea how the NWS’s
send out data, what data they send out, what they don’t and how these
things are collated. Perhaps you’d like to send me some notes on this
that I could write up as a FAQ? Won’t change anything much, but it
would be a handy reference….
gavin
Where we find out that not “many people have any idea” how it all works and they didn’t even have a FAQ about it but could use one as ‘a handy reference’.
Well, with those kinds of “standards” and “procedures” and “documentation” I find I must now recant my statement that I didn’t see much reason to “keep my powder dry”.
It would seem that there is no way to ever know what stations are in, what stations are out, WHEN they are in, and WHEN they are out, or even if what is IN is really what is supposed to be in. As long as it’s self consistent between the dailies and monthlies, it can be in, unless it’s a very bad “outlier” (but even that might be an error if we are going to be having “extreme weather events” as they ought to be, by definition, outliers…)
At any rate, given their “process”, it looks like we can have some confidence that the glue-on QA will be done eventually and most of the time a lot of the data will be available by the middle of the following month, except when it’s the end of the month, or maybe the next month… or whatever.
Just unbelievable…
Ron Broberg (22:32:06) : Maybe you can post your station counting code …
grep ………..2010 v2.mean | cut -b 4-8 | sort -n -u | wc
Not that much different from yours. I’m prone to parameters in scripts, though, so I have work directories and years as passed parameters with defaults. That way the ‘process’ is consistent and I don’t have as much to worry about in the typo / qa process. Since the v2.mean file is already sorted by stationID, I don’t have a sort step:
WRK=${2-./data}
grep …………${1-2010} $WRK/v2.mean | cut -b 1-11 | uniq > $WRK/$1_uniq_station_list
and then I do things like “wc -l” on the ouput list or diff or…
I suspect that the difference in results is due to provenance of the data. It does have a ‘today’ update date on the ftp server and my copy is from 4 days ago.
And, given their rather ersatz ‘process’ as described by the folks who ought to know this the most (the leader of the unit in the UK) it looks like updates can happen at any time…
Oh yes – all Czech thermometers were scraped. Why couldn’t they leave at least Milesovka – it’s nice professional station at the mountain top with a record from the end of 20’s.
http://travel.webshots.com/photo/1038138769028455201hytgZu
Wait a month or two, then get back to us.
I make that – Stations Dropped:
Africa – 118
Asia – 137
S America – 62
N&C America – 59
Pacific Reg – 39
Europe – 70
Antarctica – 1
For Europe – using database-derived trends for GISS adjusted data:
65 of the 70 dropped stations were in the database – the 5 missing ones may be below the QC threshold
Average trend for ALL European Stations: 0.56 Deg C/Century warming
Average trend for Dropped European Stations: 1.23 Deg C/Century warming
Average trend for Remaining European Stations: 0.5 Deg C/Century warming
So for Europe at least, dropping these thermometers MAY not have a warming effect. However, these trends are for all years – it depends on the forward trends at the remaining stations.
@mILLrAT (13:04:40) :
“I have converted the station list into a google earth kml file that is available for download at
http://cid-78d84fe53bde5e59.skydrive.live.com/self.aspx/Public/RIPtempStations.kml”
Thank you Sir!
(the file ext. is found as *.xml – just rename to *.kml)
…3 stations left in Denmark, 1 in Copenhagen (lagest city), 1 in Aalborg (4 th. lagest city), and the last (Roenne, Bornholm) on a smal iland in the middle of a parking lot
..an on a funny note: they left out the one in Tarm, a Danish syn. for “anus”…
again, your work is highly appreciated!
“E.M.Smith (19:29:57) :
Andrew30 (16:57:24) : Is this a temperature scale conversion, F to C, if so should the 50.0 not actually be 5.0?
The data are in 1/10 C increments, so 102 is 10.2 F.”
You’ve got a typo, you mean “102 is 10.2 C”. Andrew, sometimes programmers multiply a float with a value that is usually a power of 10 or power of 2 like 256 and store the product in an integer. From there they use fixed-point arithmetic. Often done in economy where you want exact dollar/cents (you multiply by 100 in this case) and avoid the inability of floating point numbers to exactly represent certain fractions – sometimes you get a pocket calculator throwing something like 0.999999999 at you when you know it should say 1 – this is such an artefact.
This is legit, only the multiplication/division they did was buggy.
Offhand it looks like I surmised correctly. The goal is to establish there is a long-term strong warming trend. They have adjusted individual station data downward in the far past but upward in the near past. However this was done clumsily and has been detected at Darwin etc. Their ignoring of UHI and the proper adjustments has also been noted.
Realizing they will be facing increased scrutiny, they are placing their bets on stations that will show an increased warming trend. The high altitude stations will not show that, so they are eliminated. Urban thermometers that have bottomed-out on UHI and will now report anomalies more consistent with real global temperatures, were removed. For example, note the dropping of the Dallas-Fort Worth airport. It seems unlikely there will be a large increase in air travel anytime soon, and between current economic conditions and theoretical upcoming “green taxes” air travel may likely decrease, thus it will not show a great warming trend, thus it is dropped.
What is left appear to be stations where nearby development with UHI will bias the readings to a warm trend in the coming years. They went light on the suburban deletions, as increased populations and future economic development will make suburbs warmer. There are rural stations that are unlikely to see nearby development anytime soon, may even be showing cooling, thus they are good candidates for deletion. But rural stations that will see development and thus increases would be kept, including those that now or in the future should not be labeled “rural.”
Of course far more research needs to be done as to the specific deletions and retentions to positively conclude the result will likely yield an increased artificial warming trend, in line with AGW predictions. But offhand, to me, I can believe that until shown otherwise.
… it looks like updates can happen at any time…
Which is unsurprising to me when dealing with international data sets
I also occasionally take a look at the EIA world oil production numbers. While interim updates are published more-or-less on a monthly schedule, the data effected can go back two years. Estimates are used for oil production in countries that don’t report in a timely fashion, and those estimates are updated whenever a particular country gets around to providing more accurate figures.
Those estimates can affect things like total world oil production. Early figures in 2009, for instance, suggested that annual oil production reached a maximum in 2008. Current figures suggest 2005. (See 4.1d). (And the ‘max’ statement is a statement of past history, not a future prediction)
http://www.eia.doe.gov/ipm/supply.html
So I’m used to seeing international data sets – even on “very important topics” – shift a bit. Not everyone reports in a timely fashion. The major difference here seems to be that NOAA has enough thermometers in play that it does not need to make estimates for stations that haven’t reported yet.
My prediction: more stations in the data set for 2010 next time it is updated.
“DirkH (08:35:35) :
[…]
This is legit, only the multiplication/division they did was buggy.”
Oh well, and they should have thought of the fact that nint truncates, not rounds, so they should have added 0.5 before the truncation so as to not introduce bias.
“DirkH (08:43:01) :
[..]
Oh well, and they should have thought of the fact that nint truncates, not rounds, so they should have added 0.5 before the truncation so as to not introduce bias.”
Dang! correcting myself. It rounds. Sorry not being a Fortran specialist and talking out of my ass again. So that’s a wrong assertion of mine, thank you, google.
Co isn’t analogous to CO2, because CO2 is much more inert. CO is taken up by red blood corpuscles in a way that is effectively poisonous. It’s qualitatively bad, unlike CO2.
E.M., i just saw on your blog that the people who try to rewrite GISTEMP in Python continue the tradition of omitting brackets where they can. That means they will get different undefined artefacts than the Fortran original but undefined artefacts it will be. Good luck Python guys. Go to E.M.’s blog to the thread he linked to to see my comment.
Well, I’ve just completed a new download of the v2.mean.Z file followed by unpacking and a “diff” with the 8 Feb 2010 version.
The most interesting difference is this one:
< 4347898800032009 264 260 261 274 278 281-9999 289 295 290 288-9999
> 4347898800032009 264 260 261 274 278 281-9999 289 295 290 288 278
> 4347898800032010 274-9999-9999-9999-9999-9999-9999-9999-9999-9999-9999-9999
Notice that, in addition to putting IN the 2010 record for this station, the 2009 record has had a December value added.
So between 8 Feb 2010 and now, Dec 2009 was “discovered” somewhere and put in. So even a month and a half after the fact, lost thermometers can be found.
There were 67 other “adds” to the file
So at this point, it looks like when NOAA make a file available, it’s only a suggestion not an endorsement, and it will be constantly changed, day by day, for months. Given that each month there will be a new month of data: any single day the image of the file may well be different from any other day but with a ’spike’ of newness near the 2nd week of the month.
How does one assure any consistency of processing or comparison of results between researchers when the base data is constantly mutated? How does one “measure” the process or even standardize the error detection? Unless you know the DAY that GISS chooses to do their copy of the GHCN input, you can not duplicate their results (though you could come “close” most of the time.) But given the tendency to ’spread’ a value 1200 km in GIStemp, that single December added value could influence a box of space 2400 km x 2400 km and that could also shift hemispheric values, etc.
Just amazing.
Here’s the monthly history for DFW AP
JAN 2010
http://www.wunderground.com/history/airport/KDFW/2010/1/13/MonthlyHistory.html
DEC 2009
http://www.wunderground.com/history/airport/KDFW/2009/12/31/MonthlyHistory.html
JAN mean is 45F
DEC mean is 43F
Here’s Chiefio’s GHCN station reference: (*) Dallas-Fort W 32.9 N 97.0 W 425722590000 4,037,000 1947 – 2009
Checking NCDC’s MMS database it shows it to be current, but curiously, shows a start date of 1974 NOT 1947 so I wonder if the start date is a typo? According to AP history records, DFW Airport opened for commercial service on January 13, 1974. Sure looks like a typo in the GHCN database to me.
I thought maybe the issue has to do with METAR reporting, as I have
previously pointed out to have data transmission errors. So I went through every hourly METAR for Dec09
Starting here and working backwards through Dec09 and forwards through Jan10:
http://www.wunderground.com/history/airport/KDFW/2009/12/31/DailyHistory.html
(be sure to click “show full METARS”)
Every METAR report for every hour of the month of December checks out. As does every one on January 2010
So if they are using DFW AP data, and the Lat/Lon given ( 32.9 N 97.0 W) places it there, see…
http://maps.google.com/maps?q=32.9+N+97.0+W&oe=utf-8&client=firefox-a&ie=UTF8&hq=&hnear=%2B32%C2%B0+54%27+0.00%22,+-97%C2%B0+0%27+0.00%22&gl=us&ei=6dJ2S_v5BIPMsgPH073LCA&ved=0CAgQ8gEwAA&ll=32.9,-97&spn=0.119052,0.128574&z=13
…then there’s nothing wrong with the raw data that I can find.
So it puts the onus on NCDC’s algorithms.
I don’t see why this is so surprising to you. The most recent month(s) continue getting updated over time, as the data come in. I find it strange that your first thought was that it’s all due to a “hatred” of thermometers.
You must have other, older versions of v2.mean lying around. Just do the same comparisons on them.
“you can not duplicate their results (though you could come “close” most of the time”
Has GISS even updated their webpage for Jan 2010 results yet? Not that I can see. I do think they wait until they think enough stations are in. But yes, the numbers change do change a hair as more updates come in. You can see this for yourself, and I think it’s been noted on this website.
“It also shows that the excuse of things being dropped for not electronically reporting is pretty much a lie, too. I note that Dallas Fort Worth Airport..”
Care to re-word that?
@ E.M. Smith
It also shows that the excuse of things being dropped for not electronically reporting is pretty much a lie, too. I note that Dallas Fort Worth Airport is on this list and I’m pretty sure they have electronic reporting…
For the 2010 deletions, “That dog won’t hunt”… since BY DEFINITION those stations were reporting up until now.
I notice your “update” statement that DFW is indeed a failed QC station.
Are you going to retract your “pretty much a lie” statement?
Ok, enough of my complaint …
Nah. Say it aint so … 😉
Or maybe it isn’t sarcastic… maybe in our Orwellian world… No, no, must not think that way. Where did I leave my soma?…
So, no, I don’t see much reason to ‘keep power dry’. I could easily see a single digit number of thermometers change from “QA failed” being itself QA’d but if they start putting 5% or more of the records ‘back in’ I think they are violating their published methodology… (though I’d welcome it).
Really? You would “welcome it?”
So at this point, it looks like when NOAA make a file available, it’s only a suggestion not an endorsement, and it will be constantly changed, day by day, for months. …
… Just amazing.
I’ll assume your ‘amazement’ comes with an undertone of ‘welcoming’ 😉
—————
kadaka wrote:The goal is to establish there is a long-term strong warming trend.
See vjones’ work below.
vjones wrote:Average trend for ALL European Stations: 0.56 Deg C/Century warming
Average trend for Dropped European Stations: 1.23 Deg C/Century warming
Average trend for Remaining European Stations: 0.5 Deg C/Century warming
Thanks for that update. I appreciate it.
A small shadow cast on the thesis that: NOAA is selectively removing cooling stations.
In this case, the stations dropped for Europe were warming stations.
“…then there’s nothing wrong with the raw data that I can find.
So it puts the onus on NCDC’s algorithms.”
Or, somebody/something messed up when coding in the CLIMAT report. You can look it up, and it shows the weird numbers that caused it to fail QC.
What I find amazing is that any scientist would not be horrified by this process. All data should be held out and made official on a specific date. With billions going for climate research, a new website to promote AGW and a new computer in WY, you’d think someone might consider updating the methodology to at least the 16th century.
I guess not.
Some airports going away, eh? Isn’t it pretty much SOP for this bunch when criticised with logic and facts, to 1). Tell the public it is unfair criticism by a bunch of skunks-at-the-garden-party that obviously has no substantive impact. . . and then 2). Run around behind the scenes going “Hmm, that does have some merit after all” and address it without comment or admission of why?
Ron Broberg (11:44:27) :
(…)
A small shadow cast on the thesis that: NOAA is selectively removing cooling stations.
In this case, the stations dropped for Europe were warming stations.
And amazingly enough, that wasn’t even my thesis. Or my hypothesis.
As you presented, the European stations that have shown great warming have been dropped. As I surmised. They have already been milked for the greatest positive change in rate of warming. UHI and other factors have taken place, they will now show a rate consistent with the rate for global average temperatures. The ones retained, I surmised, have the greatest potential to show the greatest positive change in rate of warming in the future.
The first set has been milked dry, now they are moving on to the next set. By creative selection of stations, they will have their great rate of warming without resorting to messy and detectable mis-adjustments of the raw data.
Try again.
@Anthony: The GISS web site shows a “GHCN + USHCN” record that starts in 1947:
http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=425722590000&data_set=0&num_neighbors=1
A look in the v2.mean file shows it is in GHCN:
Though that 1947 record is just a December value.
So I’m left to assume there was something recording there long ago. Perhaps some older rural airport prior to DFW being created? If so, that begs the question of how one mod flag spans so much time and change. If not, that begs the question of where the data came from. In either case, it’s an interesting question of station and data quality and provenance.
If there are any Dallas locals who know the history, perhaps they could speak up.
Oh, and in 2009 it had a couple of data dropouts in Sept and Oct as well.
Or has them now? Or might have them in the future, but maybe not?
In the mean time, having established that GHCN v2.mean is a perpetual “work in progress”: that complicates picking a new ‘poster child’ for electronic reporting and “dropped”. It will be easy to identify an electronic reporting and “missing in action”, but then there is that problem of prodigal thermometers returning months later.
The other interesting path this leads to is the question of just how much GIStemp products have “stability”. If the data can be MIA for a few months, then show up, how much “faith” can be put in any claims of “Warmest” or “Most Extreme” anything? WHEN does a GIStemp product become ‘stable’?
Best I can think of is to get a copy of GHCN at the moment any GISS claim is made and do a “missing data” analysis on it. What fraction of stations in the current month have “missing data” compared to ??? ( 6 months before? A year before? ) but even that becomes a bit of a random act. If the claim is made the 22 of the month, was the data downloaded by GISS on the 20th? The 15th? Did NOAA provide updates on the 21st?
For example:
http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=434789880003&data_set=0&num_neighbors=1
that station with the December Return. It has a 2009 “average” on the GISS graph. It does not include December. In a few (days? weeks?) it will. How would you know that the value in a few weeks is based on different data from the value now without a detailed data mining that just about nobody on the planet will do? Does it matter?
How often do the GISS published charts change? And by how much? Is it significant?
I don’t see an easy way to answer those QA questions. ( And “trust me” or “guess it’s OK” are not valid QA processes.)
(In accounting, there are similar issues. So you have a moment in time where you issue your annual report. You may ‘restate your numbers’ after an audit in some future year or in next year’s annual report, but that becomes part of the record. it stands out if every year you have “restated last years earnings lower 10%” for a few years. Here in GHCN / GISS that ‘restatement’ is an ongoing and un-audited process in perpetuity… )
But at least now we start to have a handle on why so many people have saved images of various GISS graphs or done data plots only to find they don’t quite match later. It’s all about the stability and provenance of the base data, and those are ill defined.
Then blend that with the GIStemp homogenizing that re-imagines the past based on changes in the present and you have a foundation of the past made of sand.
Really, what’s the points of these posts on mass dying of thermometers, when all “being eaten” means is that the stations weren’t among the first batch to report for the month? So folks speculate on why Greenland is being dropped, on the basis that there weren’t many stations in the list reporting Jan results by 8 Feb. Then we look on 13 Feb and amazing, there are six more! Good, so we can have the same discussion next month.
what’s the points of these posts on mass dying of thermometers
I’m assuming this is rhetorical, but just in case anyone needs it spelled out …
The point is to create an atmosphere of doubt and uncertainty. Two prongs of the stratagem known as: FUD
Fear is fed by the stories of ‘one world government carbon taxing us into the stone age.’
One response to FUD is to examine the details. In this case, DFW was raised as a rhetorical point by EM Smith to cast doubt on NOAA’s claims that stations are dropped if they aren’t reporting electronically.
I subsequently ‘found’ DFW in the v2.mean.failed.qc file. All I had to do was look. It was never lost. It was processed by NOAA and found wanting. E.M Smith was apparently too unfamiliar with the NOAA GHCN data set to be aware of the file.
Why NOAA v2.mean had a processed monthly mean of -6C is an interesting question. It is unlikely to be resolved in this thread. But NOAA GHCN was correct to reject their monthly mean for DFW (which was -6C). As I noted and Anthony followed on, METAR data is available for DFW in Jan 2010. It provides a monthly mean of +6C. I don’t think that tells us much about why NOAA did not receive their CLIMAT data updates or were not able to process them. DFW has values for onlu the first 10 days of January in the GHCN daily file for that station.
Of course, DFW was just an exemplar. Long lists of other stations were held out as more evidence of deliberate manipulation by an agency which was accused of ‘pretty much lying.’ E.M. Smith suggested he would ‘welcome’ seeing that list made shorter, missing stations added. When it became obvious that is in fact what happened in the 4-5 days between Feb 8 and Feb 12, he responded with pointed questions which suggest his ‘welcome’ was less than warm and friendly. From here, it appears that it is more important for him to hold on to the “theme” – to bolster the stratagem – to spread as much ‘uncertainty and doubt’ as possible – then to welcome the additional data sets. Hopefully, E.M. Smith’s welcome will be warmer as the additional stations accumulate next month and throughout the rest of the year.
As I said, all this seems fairly obvious. I’m just spelling it out on the off-chance the question was not rhetorical.
E.M.Smith (14:44:19) :
“If the data can be MIA for a few months, then show up, how much “faith” can be put in any claims of “Warmest” or “Most Extreme” anything?”
I’ve seen GISS monthly global anomalies get revised by upto +/- 0.02 C. You should realise there are uncertainty bounds on these numbers. If 2 years or months are within those bounds of each other, you can’t say one is warmer than the other. For example, a lot of breath is wasted here about whether 1934 or 1998 was warmer in the US, when they’re really just statistically tied. On the other hand, the difference between the 2000s, 1990s, 1980s and 1970s averages are significant.
But again, you’ll notice that GISS hasn’t put out Jan 2010 numbers yet. But you’ve got the GISS code running. Why don’t you just run it for the two different v2.means, and see what happens? Maybe you’ll see the reason why GISS is holding back on Jan 2010; I don’t know.
But in any case, I’m still wondering if you care to reword certain rash statements made above.
Ron Broberg (17:07:40) :
Thank you to you for actually looking into things. As for DFW, you can look up the CLIMAT report and see it for yourself. It has the numbers GHCN logged. I can’t speculate what went wrong where, but maybe we can watch it to see if it gets corrected.
The Giss monthly update normally occurs the middle of each month.
Hey carrot eater, I’ll bite.
How does one review the CLIMAT data?
Or are you referring to the data that was in v2.mean.failed.qc?
But this affect only the NOAA dataset, right? CRU/GISS still use stations whichever they want? I know that CRU dataset, besides its problems, uses much more stations that two latter datasets and is better (or less worse) than those.
By the way, EM Smith:
This is what happens when you remove high latitude stations from the record:
http://data.giss.nasa.gov/gistemp/graphs/IntegArea.pdf
Removing the SH high latitude makes little difference.
Removing the NH high latitude results in a slightly cooler record in the last few years, but that shouldn’t be a surprise, given how GISS features a warming Arctic over that time period.
So yes, anomalies do work. Absolute temperatures don’t matter when you work with anomalies; it’s the changes in temperature that matter.
OK, I bet we can figure out what happened to DFW if we work together a bit.
From the CLIMAT summary, I get:
Monthly mean temp: -6.2 deg C with at standard deviation of 5.2 C deg
Max temp: -2.2 deg C
Min temp: -10.3 deg C
And zero days of missing data for temperature.
On 29 days, the min temp was below 0 C.
On 21 days, the max temp was below 0 C.
This latter bit looks totally wrong. Either something got messed up at DFW when coding this in, or another station filed their report using DFW’s station code by accident.
And some more info from the DFW CLIMAT report
The day with the highest mean temperature: +2.5 C, on Jan 7 and also at least one other day
The day with the lowest mean temperature: -16.1 C on Jan 1
The maximum maximum temperature: + 4.4 C on Jan 7
The minimum minimum temperature: -23.8 C on Jan 2
This clearly isn’t right, so the NOAA QC was perfectly correct in tossing it out. Maybe somebody at DFW is very confused, or this is the report from some station in Siberia, with somehow the wrong station ID number.
Bob Koss (20:42:22) :
Well, there you go. This is perhaps why they usually wait a couple weeks – they need to wait for enough stations to report.
Juraj V. (01:02:55) :
GISS draws its land stations from GHCN and USHCN. Only ocean and some Antarctica come from elsewhere.
But again, note that GISS didn’t put out Jan numbers yet..
“carrot eater (03:47:45) :
[…]
Removing the SH high latitude makes little difference.
Removing the NH high latitude results in a slightly cooler record in the last few years, but that shouldn’t be a surprise, given how GISS features a warming Arctic over that time period.”
The problem is that they change the basket of thermometers over time, yet make comparisons between their final average that they get with the basket-du-jour with one they got with a very different basket of thermometers in say 1979. This opens up the room for arbitrary manipulations. If you call that scientific, that’s entirely your problem, work on your understanding of what sound statistics would require.
And if Ron Broberg calls this concern spreading of FUD, that doesn’t make it true, it’s a simple deconstruction of a botched way of working.
DirkH (06:40:44) :
Your complaint is vague to the point it lacks meaning. Point out exactly what you’re afraid of, and how it would impact the numbers. Then suggest a mathematical test to see if it occurs.
But I think the plot speaks for itself. Dropping the high latitudes from the analysis causes a slight cooling, if anything, because the NH high latitudes contribute some recent warming to GISS. Those studying the difference between GISS and CRU will be well aware of that.
Ron Broberg (11:44:27) :
Your glee at my Europe data was obvious but premature. Now that I have had time to look at N&C America, and S. America, you’ll be surprised to hear that the trends there for removed stations are primarily negative.
Average trend for Dropped N&C American Stations: -0.732 Deg C/Century
Average trend for Dropped S American Stations: -1.28 Deg C/Century
carrot eater (03:47:45) :
That graph you link to that removes high latitude stations….
I took a look at the stations in the GISS record that are removed by that analysis and at the temperature trends for each station.
It only removes 9 stations for >70S but 37 from >70N.
Those 37 have an unweighted average trend of 4.396 deg C/Century warming trend for the UNADJUSTED data set, but a 7.039
deg C/Century warming trend in the ADJUSTED data set. Given that the GISS adjustments are intended to correct for UHI, I
would say that is a very large average adjustment for a part of the world with few urban areas. I’m not surprised leaving the
>70N data out of an analysis cools the data.
@ Mr. Stokes, Mr. Eater and Mr. Broberg
Can you all think of an explanation of what happened to all 484 stations that are not included? As for DFW, given the extreme accuracy and precision claimed for the land record, it is troubling that such a prominent station could produce data that is apparently so wrong. It would seem that more digging would be necessary to find out how DFW got its figures so wrong. That might shed some light on things that might perturb the land temperature records.
P.S. The issues I am raising are meant to be sincere and not rhetorical. It is not my intent to foment a FUD fight. Please forgive the (intentional) pun.
Phil (11:13:18) : Can you all think of an explanation of what happened to all 484 stations that are not included?
I accept the possibility that those station have simply not reported January numbers to NOAA in a timely manner and/or those station have not yet been processed at NOAA.
vjones (10:48:26) : Your glee at my Europe data was obvious but premature. Now that I have had time to look at N&C America, and S. America, you’ll be surprised to hear that the trends there for removed stations are primarily negative.
Glee?
Nor am I surprised. I don’t believe that NOAA is selectively removing stations. That some regions could show ‘warming stations’ being dropped and other regions are showing ‘cooling stations’ being dropped seems to fall into line as evidence that no intentional selection is being applied. Randomness has that characteristic: some high, some low.
But more importantly, I am glad that someone is taking the first steps towards analyzing the effects – rather than just pointing to an event and making wild-eyed assertions.
Of course, the analysis isn’t over until someone checks how ‘dropped’ stations affect the final products: gridded anomalies and global averages.
I’ve followed EMS for a while now and add my vote of admiration to those above.
As an accounting type, I’m simply flabbergasted that, in an age where terabytes of SQL data can be interrogated with return times of milliseconmds, these guys are still sticking with text files and multiple passes, not to say, munges.
For my tax dollar I’d want to see a proper accounting-style transactional record for each and every read of each and every station. And I’d want to see every single adjustment, up, down or sideways, to each and every such record, type-grouped and categorised up the wazoo.
Hell, if us accounting types neglect to journal things properly, and don’t leave a proper transactional audit trail, the IRS won’t accept “well, we just filled ‘er in from nearby invoices”.
Neither, as taxpayers, should we demand anything less from our climate accountants.
A made-up sample of the sort of thing I’m urging:
Type Station Date ToD (Zulu) Tempx10 Process Who
RAW 4500234765 20100113 0850Zulu 120 AutoRead Therm
RAW 4500234765 20100113 1204Zulu 150 AutoRead Therm
UHI 4500234765 20100131 2359Zulu -15 UHI adj V2.1 EOM UHI
MIN 4500234765 20100113 0400Zulu -35 AutoRead Therm
MAX 4500234765 20100113 1320Zulu 157 Autoread Therm
CAL 4500234765 20100113 2359Zulu -5 Calibrate QA Crew
vjones (10:48:26) : You might not be surprised that leaving out high latitude NH cools the data (and really, nobody should be), but yet EM Smith thinks that leaving them out would warm the data. So far as I can tell, the reason for his thinking is that he’s using absolute temperatures in place of anomalies.
Phil (11:13:18) : Patience. Wait for GISS to put out its Jan 2010 numbers, then see what’s in v2.mean. Then check again a month later. Not every station sends out its CLIMAT report on the first day of the next month.
As for DFW, well, this is why GHCN has a quality control step. My uneducated guess is still that it’s a mislabeled station number, but that’s just an idle guess.
Re: Phil (Feb 14 11:13),
Mr. Stokes, Mr. Eater and Mr. Broberg
Can you all think of an explanation of what happened to all 484 stations that are not included?
Easy. Their reports of Jan 2001 figs didn’t get into the online file by 8 Feb, when EMS did his count. But by 12th Feb, the number was down to 415. I’m sure there’ll be another “resurrection” in a few days. The number reporting will probably be up to about 1500 by end Feb. Then I guess someone can look at the Feb figures on 8 March, and we can have another mass dying article.
@Ron Broberg (14:40:19) :
Do you take the time to realize what you are saying before you post it?
(…)
Nor am I surprised. I don’t believe that NOAA is selectively removing stations. That some regions could show ‘warming stations’ being dropped and other regions are showing ‘cooling stations’ being dropped seems to fall into line as evidence that no intentional selection is being applied. Randomness has that characteristic: some high, some low.
(…)
Do you think someone is hitting a “Purely Random Delete” button somewhere? Whereby both “good” and “bad” stations all have an equal chance of deletion therefore it’s a completely neutral unbiased process? Otherwise there is some form of selection criteria involved, therefore NOAA is selectively removing stations. From there one has to figure out what the selection criteria are, whether purely “quality control” or if something else is involved and to what degree.
I accept the possibility that those station have simply not reported January numbers to NOAA in a timely manner and/or those station have not yet been processed at NOAA.
(…)
Aren’t there stations with automatic electronic reporting that are missing? Why would it take longer to process them than the ones without such?
For those who are interested, I’ve posted on my blog a list of the latest v2.mean stations, in order of most recent reporting month (as at 12 Feb 2010), and with the countries marked (in groupings).
kadaka (00:20:42) :
“Aren’t there stations with automatic electronic reporting that are missing? Why would it take longer to process them than the ones without such?”
There’s an incorrect assumption embedded in your statement. Just because they’re sent electronically doesn’t mean they’re generated automatically. There will be some human interaction there (keying in data, checking for errors like the DFW weirdness above)
If you dig around, I think WMO guidelines ask for the CLIMAT reports to be sent in by the 5th or 8th of the following month.
carrot eater (05:01:21) :
There’s an incorrect assumption embedded in your statement. Just because they’re sent electronically doesn’t mean they’re generated automatically. (…)
There’s an incorrect assumption embedded in your reply. I know dang well what I meant by “automatic electronic reporting.” The weather station info gets sent into a computer where it is logged. Then compiling it into a report and sending it in becomes a normal scheduled job for the computer. No human action to initiate anything should be needed, making it fully automated and done automatically. Don’t they have stations set up to do that?
kadaka (07:01:26) :
You made the exact same error again. It’s electronically transmitted, yes. But who told you it was always automatically generated and sent?
Read about CLIMAT reports on the WMO website sometime; you’ll find a lot of material there. They’ve made available software that generates the CLIMAT report in the required format, but somebody still has to sit there and manually key in the data, in order to use that software. You can see screenshots of the software. Maybe in some countries, the databases are fancy enough that the CLIMATs can be written with no human interaction, but this is not the rule. Judging from the illustrations of common errors, some countries were probably writing the reports manually.
Me:
Sure looks to me like simple enough wording, easy to understand. The question at the end, don’t they have stations set up for full automation?
Which got me:
Me: Don’t they have fully automatic stations?
carrot eater: Who told you they were always automatic?
Logic error detected. Aborting process. Error report: Systemic lack of comprehension with intended recipient.
kadaka (13:42:06) :
Ok, I’m sorry, maybe you were asking a question instead of making an assumption.
Either which way, the answer is no. It is not a fully automated process, at least in many countries. Some African countries are especially bad at reporting on a timely and regular basis, though some others do really well.
All the active stations in any given country tend to get reported and added in chunks. EM Smith noticed Strasbourg hadn’t gotten in yet, but in fact none of France was in yet. One of these days you’ll look and probably see 17 French stations pop up for Jan 2010, all showing up at once.
Looking a bit more, using Nick Stokes’s breakdown:
EM Smith’s missing ~400 stations includes a lot of countries that don’t report regularly at all. For example, it includes Libya, which has had better things to do than send numbers in, since Jan 2009. That’s 17 stations there. Given their unreliability, who knows when they’ll report again.
A better predictor of what might show up in the next week is the difference between Dec 09 and Jan 10. This is 179, minus I think 11 in QC. So where are these missing ~168?
North Korea, for 7.
Mongolia, for 34.
Paraguay, for 10.
Peru, for 12.
France, for 17.
Greece, for 10.
French Polynesia for 7
Mozambique for 5
That’s a sum of 102, and then little bits here and there. All of the above countries are entirely missing as of now.
So maybe those will show up by month’s end, though there’s a chance some of these are also sporadic reporters, and just happened to file reports last month.
So if I had to guess, maybe 80 more stations will show up soon, while the rest of the ~400 could take several months.
Gavin Schmidt: “Considering the enormous success that we achieved with our global climate models, we decided to dropping the weather stations completely. From now on, all GHCN data will be interpolated from the output of our models, which provides a highly accurate representation of the Earth’s climate and what’s better, allows us to get surface temperatures from any point on the globe, instead of only those where weather stations are located.”
After spending a large part of last year in Central America (officially part of North America in this data set) and becoming familiar with all of the major airports in the region, I am very surprised to see deletions of:
Belize City, BLZ
Managua, NIC
Tegucigalpa, San Pedro Sula, and La Ceiba, HND
San Jose (Juan Santamaria), CRI
I haven’t yet taken the time to see what is left in the data set from CENTAM, but there are only a couple of other airports in the entire region (San Salvador, SLV; Guatemala City, GTM; and Panama City, PAN). There may also be some rural sites throughout the region, but I can’t imagine there are many. Perhaps it has been decreed from on high that all CENTAM “climate” can be represented by the data provided by the USAF at Soto Cano Air Base, situated in a valley around 2500ft (~800m) right beside mile-long strips of concrete and asphalt – perfect representation of the tropics.
Oh wait, Soto Cano doesn’t appear either, since there is only hourly data from there since the early 80s. Perhaps the CENTAM information is being interpolated from surrounding areas like Houston, New Orleans, Miami, and Antarctica.
Is there an analysis that shows that the changing fortunes of weather stations over time has an impact on the measured temperature record?
Ron Broberg (08:40:38) : My prediction: more stations in the data set for 2010 next time it is updated.
grep ^…………2010 v2.mean | cut -b 1-8 | sort -u -n | wc
1348 1348 12132
1348 stations for 2010 in the March 5 3:38AM v2.mean.Z
EM Smith: Just shy of 1/3 of the stations, taken out back and shot this year.
“I’m getting better!”