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.
I'll grant that I've been having trouble finding the FSU station specifically, so I looked up what the WeatherSTEM standard was and it seemed to be masted stations. But thanks for finding that.
It's pretty blurry from my end....is the thing on the triangle roof the station? If it is, I would actually say it is likely to provide accurate readings for its environment because it meets the standard of being at the edge (together with its colder highs and National Weather Service recommendations strongly suggesting it is accurate).
See: https://www.city-data.com/forum/64770372-post31.html. This would meet these standards of representing the environment - although I did put in an email to Ed Mansouri about this.
Quote:
Besides, putting a thermometer on top of a tall mast creates all sorts of new problems. For starters, you are no longer measuring the temperature at 2 meters above the ground so comparisons with other sites are not possible.
Masts don't solve the problem, they create new problems. How tall is very tall? For a mast to be tall enough to reduce the radiation effect of a 1.4 acre 7 storey parking garage like the FSU one would mean a very tall mast indeed. There's a reason standards for measurement of temperature exist and why those standards do not include "very tall" masts.
I disagree. The reason for this standard is for maximum representativity of the urban environment - meaning it may not be directly comparable in all aspects to a 2m off the ground station, but can be compared in terms of representing its environment and the result it gets.
I realize they are not comparable in terms of setup (as the Rhodes thread said), but this is a necessity. You can absolutely compare the sufficiently representative methods of obtaining the weather of each - it's like comparing the indoor thermometer to the outdoor weather station to see how the temperatures in each place compare with the proper methods of each, simple as that.
National Weather Service Tallahassee seems to recognize that this can be done, given their usage of the stations to compare w/the airport.
When choosing a time period for climate normals there is a balance between making it long enough to smooth year-to-year variation and short enough to give an indication of the climate as it is now, rather than what it was in the past due to longer term changes. 30 years seems a good balance between the two.
If we look at say Sydney you can see what kind of a difference it makes. I've only chosen Sydney as the BOM has records going back a long way that are easily accessible.
Using 1858-2020 data the annual mean maximum is 21.8C. For 1991-2020 it is 22.8C
1858-2020 has a coldest month mean maximum of 16.4C. For 1991-2020 it is 17.9C
So you can easily see that if we use an inappropriately long period, the averages will be a poorer predictor of what this year, next year or a year a few years in the future will be like than a more appropriate period such as 1991-2020.
In the case of Sydney-- the new site is a wildly different setup to the old. By no account should they ever be compared with one another. Nowhere else in Australia do we see such a drastic mean midwinter change...just look at Bathurst (1858-1983 vs 1991-2020). I thus have reason to believe the equipment at the old Sydney site, was faulty. Likewise @ Newcastle Nobbys Head site. 16.4 C avg max is outrageously cold for a leeward climate like Sydney-- use some common sense, and it becomes clear that something isn't right. Adelaide West Terrace site (1887-2023), averaged only 0.6 C different from the new 1991-2020 Kent Town site (15.0 vs 15.6 C). You expect me to believe that Adelaide, being fully windward to the subpolar westerlies, somehow only averaged one degree cooler than Sydney in midwinter? Yeah, nah.
Last edited by WesterlyWX; 06-27-2023 at 06:00 PM..
Depends on how accurate you want your data to be.
For instance, did you know that the "Little Ice Age" ran from about 1300AD to 1850? (Google it!)
So there's been a period of general, natural warming since it ended. If you only look at 30 years, you might only see the warming, and not the cold period that went before.
From the 'purest' climatological standpoint, yep, I agree that 30 years come short to summarize the influence of some big scale process.
But I think that after all, the thing about a 30 year period has to do more with human geography and let's say, 'sociology'. And in that case it's very appropiate IMO. If you think twice 30 years is roughly the median age of the population, so that amount of years matches perfectly with what a generation perceives as the climate during their times and their daily lives. So, that's my take.
I remember that not only dramatic events, but day today weather from as early as the late 1960s. I even remember droughts from the mid 1960s. 30 years is just not that long. Certainly not long enough to wash out the effect of extreme events on averages such as very heavy rain in 1972 and 1983.
It's pretty blurry from my end....is the thing on the triangle roof the station? If it is, I would actually say it is likely to provide accurate readings for its environment because it meets the standard of being at the edge (together with its colder highs and National Weather Service recommendations strongly suggesting it is accurate).
Placing a sensor on the edge of a building is not a NWS recommendation, nor is the paper referenced in your link a NWS document. That paper simply describes how they placed the sensors for that particular project in Greece. Hopefully that provided some internal consistency for their data but we simply do not know if measurements made by those sensors are generalizable.
I said it before in the other thread and I'll say it again here: More than likely the sensor is not measuring the air temperature but some bizarre hybrid of the air temperature and radiative temperature of the nearby structures. This cannot be compared to anything.
Quote:
Originally Posted by Can't think of username
I disagree. The reason for this standard is for maximum representativity of the urban environment - meaning it may not be directly comparable in all aspects to a 2m off the ground station, but can be compared in terms of representing its environment and the result it gets.
I realize they are not comparable in terms of setup (as the Rhodes thread said), but this is a necessity. You can absolutely compare the sufficiently representative methods of obtaining the weather of each - it's like comparing the indoor thermometer to the outdoor weather station to see how the temperatures in each place compare with the proper methods of each, simple as that.
...
No, you cannot do that. The laws of thermodynamics say so.
I said it before in the other thread and I'll say it again here: Comparing bizarre rooftop measurement A with bizarre rooftop measurement B is of questionable generalizability.
In the case of Sydney-- the new site is a wildly different setup to the old. By no account should they ever be compared with one another. ...
If you're referring to the Sydney Observatory Hill site, observations began there in 1859.
From 1859 to 1917 the temperature measurements were made just outside the observatory building using a shed as a screen. Certainly unreliable.
In 1917 the thermometer was moved ~135 meters to the south to a yard on the grounds of the neighbouring Fort Street school; it was placed in a standard Stevenson screen. In 1990 the instrument was changed from manual to automatic but the location remained the same. Measurements were made continuously until 1600h August 31, 2020.
Thus the period 1917-2020 is quite consistent. The thermometer was about 5 meters from a hedge, 8 meters from the closest tree, and 10 meters from the closest building. Not perfect but not bad for an urban site. Regardless, comparisons from 1991-2020 to any thirty-year period all the way back to 1917 are reasonable.
In 2017 a new station ~200 meters to the north was opened back on the grounds of the actual observatory. This is the station which has been used since 1600h August 31, 2020. Is this the new site to which you refer? Its data are only included in the last 4 months of the 1991-2020 averages.
If you're referring to the Sydney Observatory Hill site, observations began there in 1859.
From 1859 to 1917 the temperature measurements were made just outside the observatory building using a shed as a screen. Certainly unreliable.
In 1917 the thermometer was moved ~135 meters to the south to a yard on the grounds of the neighbouring Fort Street school; it was placed in a standard Stevenson screen. In 1990 the instrument was changed from manual to automatic but the location remained the same. Measurements were made continuously until 1600h August 31, 2020.
Thus the period 1917-2020 is quite consistent. The thermometer was about 5 meters from a hedge, 8 meters from the closest tree, and 10 meters from the closest building. Not perfect but not bad for an urban site. Regardless, comparisons from 1991-2020 to any thirty-year period all the way back to 1917 are reasonable.
In 2017 a new station ~200 meters to the north was opened back on the grounds of the actual observatory. This is the station which has been used since 1600h August 31, 2020. Is this the new site to which you refer? Its data are only included in the last 4 months of the 1991-2020 averages.
Yes, and those observations from the original siting 1859-1917, were comically cold and greatly skew the average down for the entire 1859-2020 period.
I mean just look at this nonsense .. 14.8 C (!!) Jul avg max 1881-1910 lol ! Clearly faulty. The other 30 year periods surrounding it also feature 15 C avg maxes for July. Laughable. We can safely discard Sydney OH.
OK I didn't realise Sydney's record was going to be controversial and bog down the discussion, diverting it from the principle that in an era of rapid change, using an inappropriately long period is going to tell you more about what the climate was rather than what it is. Eg if you're a farmer and use normals to decide the most appropriate time to plant your crop, 1991-2020 is almost certainly going to be a better guide than say 1931-2020. In other words, next year is more likely to me more similar to 1991-2020 averages than 1931-2020.
Placing a sensor on the edge of a building is not a NWS recommendation, nor is the paper referenced in your link a NWS document. That paper simply describes how they placed the sensors for that particular project in Greece. Hopefully that provided some internal consistency for their data but we simply do not know if measurements made by those sensors are generalizable.
I said it before in the other thread and I'll say it again here: More than likely the sensor is not measuring the air temperature but some bizarre hybrid of the air temperature and radiative temperature of the nearby structures. This cannot be compared to anything.
No, you cannot do that. The laws of thermodynamics say so.
I said it before in the other thread and I'll say it again here: Comparing bizarre rooftop measurement A with bizarre rooftop measurement B is of questionable generalizability.
I just got the email from Ed Mansouri which would fix all that. He said:
"I’m not too worried about the temperature readings - I think it’s high enough off the structure where it’s not going to have its readings overly amplified by ground-based irradiance.
I said it before in my other comment and I'll say it again here: the colder highs and recommendation to me by the National Weather Service are discordant as can be with a bizarre hybrid of air temperature and thermal radiation. Ed (lol) just confirmed that for me.
@Jbgusa, sorry for derailing this thread - the heat island topic got way more out of proportion than I intended it to be. I have a question about normal year bases that I hope someone knows the answer to.
A recommendation I have seen by the WMO is that these normals must start on a year ending with 1, and the end at a year ending with 0. I understand the usefulness of this for human purposes but is there any evidence this is actually a better period than, say, a 6 and 5?
Last edited by Can't think of username; 06-28-2023 at 09:20 PM..
@Jbgusa, sorry for derailing this thread - the heat island topic got way more out of control than I intended it to be. I have a question about normal year bases that I hope someone knows the answer to.
No need for an apology. It's all good.
Quote:
Originally Posted by Can't think of username
A recommendation I have seen by the WMO is that these normals must start on a year ending with 1, and the end at a year ending with 0. I understand the usefulness of this for human purposes but is there any evidence this is actually a better period than, say, a 6 and 5?
I actually have no idea and I'm a history buff. I would guess it would facilitate saying "30's, 40's and 50's" or something arbitrary like that.
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.
) just confirmed that for me.
