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.
Okay now I found one location where the data is completely incorrect and doesn't even remotely match the actual data there.
NYC in 2004 and the years preceding. Says it reached -22C with several highs below -10C in January. Complete BS there. Last time NYC reached -20C was 1949 I think. https://weatherspark.com/history/310...-United-States
So yes, Weatherspark can have inaccuracies in the data, but in my experience they're few and far between.
Weatherspark isn't very accurate for here; it says that the hottest day of the year has an average high of 23c, when the entire months of July and August have warmer average highs. It's been proven that the average high peaks at 24.7c.
I'm not sure the way you are figuring out the average high for each day is actually the way met offices do it. Taking daily highs and dividing them over a 30 year period. For example, according to the NCDC, MDW has several days in a row during the month of July where the average high is 29.6C which is statistically impossible using your method.
Though if I average out the actual high temps for those days over 30 years I get a different figure entirely. So I chose July 5th at random, added the actual high temps between 1981-2010 and divided by 30 years and got 86.1F/30.1C
Last edited by chicagogeorge; 02-22-2016 at 08:35 AM..
Don't know what you're talking about. I don't see any 23C in February. The data looks perfectly consistent with what you say is correct.
Weatherspark gets its data directly from the obs in the official weather station in these places, so it can be off by a degree or two, but generally not more. Sometimes it also gets some weird spikes and breaks in the data but usually for the well-known locations it doesn't happen.
The hottest day of the last 12 months was July 31, with a high temperature of 38°C. For reference, on that day the average high temperature is 31°C and the high temperature exceeds 34°C only one day in ten. The hottest month of the last 12 months was July with an average daily high temperature of 32°C.
Relative to the average, the hottest day was February 15. The high temperature that day was 23°C, compared to the average of 15°C, a difference of 8°C. In relative terms the warmest month was February, with an average high temperature of 18°C, compared to an typical value of 15°C.
The data on there & also Tutiempo often has lots of errors, the site I use which seems to be the most accurate is Ogimet, as I check the monthly reports from there with the data released from the Malta Met Office at Luqa airport each month...
The data on there & also Tutiempo often has lots of errors, the site I use which seems to be the most accurate is Ogimet, as I check the monthly reports from there with the data released from the Malta Met Office at Luqa airport each month...
The 23C happened on February 15th of this year, not last year. Same with the 18C average high. If you look at the data strictly for 2015 (not "last 12 months") you will see that the data is correct.
The 32C in July as opposed to 33C comes from the fact that Weatherspark computes averages from obs reports as opposed to official highs. In this sense Ogimet is indisputably more accurate but I wouldn't use the two sites interchangeably in the first place.
I'm not sure the way you are figuring out the average high for each day is actually the way met offices do it. Taking daily highs and dividing them over a 30 year period. For example, according to the NCDC, MDW has several days in a row during the month of July where the average high is 29.6C which is statistically impossible using your method.
Though if I average out the actual high temps for those days over 30 years I get a different figure entirely. So I chose July 5th at random, added the actual high temps between 1981-2010 and divided by 30 years and got 88.7F/31.5C
I also averaged the highs for the period of 23 Jul - 7 Aug to try to smooth out any spikes, and the average high for those 2 weeks still comes out at 24.2c. That's at least 2 weeks that average out to be warmer than weatherspark's supposed peak hottest day of the year, by over 1c.
I also averaged the highs for the period of 23 Jul - 7 Aug to try to smooth out any spikes, and the average high for those 2 weeks still comes out at 24.2c. That's at least 2 weeks that average out to be warmer than weatherspark's supposed peak hottest day of the year, by over 1c.
I went back and rechecked my math. The 30 year average is 30.1C for July 5th. Still higher than what the NCDC has it. I used official NWS temperature data. Don't really understand why
Also Weathersparks uses 1974-2012 day for London Heathrow. I don't know if that makes any difference in the averages...
Last edited by chicagogeorge; 02-22-2016 at 09:06 AM..
The 23C happened on February 15th of this year, not last year. Same with the 18C average high. If you look at the data strictly for 2015 (not "last 12 months") you will see that the data is correct.
The 32C in July as opposed to 33C comes from the fact that Weatherspark computes averages from obs reports as opposed to official highs, and also (as chicagogeorge pointed out) Malta's met office may or may not have a different way of computing average highs. In this sense Ogimet is indisputably more accurate but I wouldn't use the two sites interchangeably in the first place.
I didn't expect them to be basing it on this month that hasn't even finished yet & it is still wrong as Luqa was 23.6C on the 15th of this month, which should be rounded up not down... Also saying an average high of 32C when it was 33.1C is quite a big difference really when it comes to an average. Malta's Met Office work their averages out the same way everyone else does, I think it is Weatherspark who obviously do it the "wrong" way... I don't use Weatherspark anyway because it isn't completely accurate, but had noticed the errors yesterday when someone else linked to the site...
I went back and rechecked my math. The 30 year average is 30.1C for July 5th. Still higher than what the NCDC has it. I used official NWS temperature data. Don't really understand why
Also Weathersparks uses 1974-2012 day for London Heathrow. I don't know if that makes any difference in the averages...
It wouldn't make any difference for the 2-week average, individual daily averages might change by 0.1-0.2c either direction.
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.
