Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
It is important to note....that no temperature data is official, regardless of source (i.e. coooperative station, awos/asos faa station, etc) until data is published by the NCDC. The NCDC will smooth over any possible errors and correct them to fit a smooth curve. So it is very unlikely that hot temp in Morgantown willl survive in the history books.
We have even seen minor temp changes occur from what occurred versus what was published locally. If the Grand Forks airport had a low of say 3 and the NWS cooperative site had a low of 15 (in town reading)...the end result of what is published in the final records would have the airport at 7 and our office at 11. Smoothing the curve and smoothing out differences...happens all the time.
From what I understand, it just makes for an easier to manage database over time. At least that is what I am led to believe. The changes over a period of time are not significant and wouldnt signifcantly alter averages. Also same thing with snowfall. The snowfall network in the U.S. suffered terribly once the automated sites were installed in at airports that once had human observers in the early to mid 90s. Only in the last 5 years or so have steps been taken to supplement data on snowfall lost from these stations.
Smoothing with snowfall data is done quite a bit....as most snowfall data is obtained by cooperative observers from around the area. While each person is educated on how to measure, some dont do so very well. Thus you can often get very wide disparities in a short distance, even though the storm system doesnt warrant that. Thus the folks at the NCDC upon receiving the monthly reports look them over and smooth over the differences which makes a nice clean database of snowfall observations.
Now....bad data such as temp spikes at automated gauges would be removed from the database before final certification and that is a good thing. I looked at the data from the Morgantown obs and saw it jump to the mid 80s....then the next ob went blank, then it returned to normal. Having a warm spike does occur for some unknown reason, but the obs is QC'd at the NWS office level and beyond so it in the long run doesnt affect the national temp database.
By the way, I love reading about the weather in NZ and Australia. The oceans there moderate the climate so much..... where I live which is in the middle part of North America, we have an extreme continental climate with strong seasonal swings....from 80s and 90s (F) in the summer for highs to lows of -30F in the winter. So we have the A/C for the summer and heat for the winter time. Many areas up where I live have plug ins for their vehicles....as we do at my work place at the National Weather Service. On real cold days (-10F or lower) we can use cords to plug into the car block heaters (if installed) to help keep the engine a bit warmer and easier to start.
The NCDC will smooth over any possible errors and correct them to fit a smooth curve. So it is very unlikely that hot temp in Morgantown willl survive in the history books.
Yes, I was aware of quality control but Not that I dont believe you, I just doubt anyone goes back and checks if that was corrected. I'm also not sure if the surroundings of each station are known.
I dont see how the 80s and 90s for hours will be adjusted to 60s and 70s.
Here's another one from June in Connecticut. This was a true spike whereas the WV one was a warming trend for hours which then came back down.
I checked the prelim F6 for Morgantown on the WFO Pittsburgh site this morning (before bed as I worked mids ) and the Pittsburgh office went M (missing) for Morgantown temps on the 13th.
I checked... and the Morgantown site is an ASOS, which has very tight quality control and maintenance from the local NWS office. I have seen such spikes occur in the past at ASOS's (and AWOS's) and usually what happens is something with the fan unit on the system goes bad and shuts down allowing the temperature unit to heat up. There is a place called AOMC, which is the ASOS monitoring center, which calls each office when the ASOS unit detects errors. A priority is given, and then NWS maintenance is notified. Many times a fan issue can correct itself with little intervention, sometimes all is needed is maintenance to log into the system and do some re-boot. Now AWOS stations have a bit less real-time QC issue and often us forecasters have to notice an error and then notify the responsible maintenance folks. In Minnesota it is the FAA, which can then send techs.
Here is the written message on the prelim F6 on the web and what we tell anyone wanting climate data for a certain date/month...
These data are preliminary and have not undergone final quality control by the National Climatic Data Center (NCDC). Therefore, these data are subject to revision. Final and certified climate data can be accessed at the NCDC - NCDC - National Climatic Data Center.
Thus for right now....the climate data for Morgantown will go missing until after the end of October at which time the data for the month is sent to the NCDC for review. NCDC will use the good hourly data and surrounding observation points to come up with a scientific "guess" at what the high was in Morgantown on the 13th and that will be what is certified and published.
The process from preliminary to certified data takes a few months. Same process goes with storm data....each office compiles a list of local storm reports (snow, ice, hail, damaging wind, tornadoes, etc) and a prelim report is due at the NCDC by a certain date. The NCDC runs a check and drops duplicate reports and then finalizes the reports and publishes them as certified storm data records. That process takes a while.
Now for cooperative observer data, some observers call in their data or send it in via weather coder to the NWS office they are under. Some do not and send in their monthly obs sheet to the NWS after each month. We QC it and then it is QC'd again at the NCDC by comparing the observations to surrounding sites and smoothing the data field, and then it is published as certified.
So there are multiple layers of QC that occurs at least for the NWS in the United States.
QC in other countries may not be a lengthy. I do know from chatting with folks at the Prairie and Arctic Storm Prediction Center in Winnipeg Manitoba Canada that their QC is a bit less stringent.
By the way, I love reading about the weather in NZ and Australia. The oceans there moderate the climate so much..... where I live which is in the middle part of North America, we have an extreme continental climate with strong seasonal swings....from 80s and 90s (F) in the summer for highs to lows of -30F in the winter. So we have the A/C for the summer and heat for the winter time. Many areas up where I live have plug ins for their vehicles....as we do at my work place at the National Weather Service. On real cold days (-10F or lower) we can use cords to plug into the car block heaters (if installed) to help keep the engine a bit warmer and easier to start.
It's certainly a different climate there. It's hard to imagine such variation over the year. I like that no major changes have to be made here for different seasons, but would like to experience more extremes - just not here.
Nothing extreme in NZ temps, but there are some very high rainfall areas (not where I live fortunately), and the Alps could possibly have the largest annual snowfall on the planet.
85/65F is average as far as I can tell, and the average so far this month is 87F, with is clearly above normal. Lows are averaging 72F, which is 7F above normal, but that huge discrepancy could be due to the NWS station being closer to the coast than whatever station The Weather Channel takes its 85/65F averages from. I'm not aware of any weather feature in Florida that could be causing such an extreme anomaly.
At any rate, yesterday the high temperature dropped to a clearly below normal 81F, so perhaps that is what was being referred to.
I think it's amazing that you can tell if it's been above or below normal at all. There is so little variation in the high temperatures in Sarasota. 89F - 87F - what's the difference?
As for the observations and smoothing, all that talk makes me envy how they observe the weather so precisely at Mount Washington.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
) and the Pittsburgh office went M (missing) for Morgantown temps on the 13th. 
