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.
Sydney Aero averages for those months are actually 179/205/242, a 3-month average of 209 hours. And I will, as per posted already, continue to have my reservations about direct comparisons between US sunshine measurements and those in UK, Canada, Australia and New Zealand.
I’m certainly no expert in solar geography (lol)...but let me just mention a few things I’ve gathered as a weather observer. You generally seem interested, and I like responding to your posts anyway...
1) NWS SURFACE Climate data - is data observed at principal meteorological stations by trained observers or automated equipment that has been tested and accepted by the controlling agency. It is 100% incorrect (above post) that there is no standard producers at each NWS station...in fact the reverse is true: NWS has standard instrument housing, equipment, procedural data collection...etc. Data is collected on a wide range of time scales from one minute for some current automated equipment... to three observations per day for intermittent periods at some remote, part-time locations. However, normally...the majority of NWS stations collect hourly observations.
In the 1990’s NWS began automating surface weather observations through the new high tech Automated Surface Observing System (ASOS). The measurement of all weather elements included in conventional observations have not (yet) all been automated. Some of those weather elements that have not been automated include clouds above 12,000 ft, snowfall and snow depth, sunshine, and the identification of certain types of weather (hail, drizzle, smoke...etc).
2) The last time NOAA collected sun hr data was from 1961 to 1990 (these are the solar maps I posted with the final 1989 update). This is the reason (above post again) that NWS defers to National Climatic Data Center (NCDC)...its a historical item.
During this time NWS/NOAA was collecting solar data....a sun hr was logged when there was a measurement of direct radiation from the sun. The percentage is given without regard for the intensity of sunshine. That is, thin clouds, light haze, or other minor obstructions to direct solar rays may be present, but would not mitigate the full counting of an hour of sun. So logically, while the case can be made that the calculation of what is termed a “clear” or “party cloudy day” could have some variation to it - it is 100% incorrect that a cloudy day or hr with no sun.... would be counted as a sunny day/hr (or vice versa). NOAA/NWS collected this sun data for 30 years without interruption, and several times updated/changed/cleared errors/streamlined data...etc to produce the last set of 1989 sun hr maps.
So while I have no doubt that there is some error to all sun data...I think it is fair to say the NWS sun hr data collected from 1961 – 1990/and the sun maps final updates in August 1989 before AOS took over are fairly accurate. Additionally, I have seen a few thermal pictures (today modern research academics/advanced professionals in climate almost universally use thermal imaging data(color shaded algorithms ) to examine climate elements)...and noticed that the 4000 hrs a year that Yuma was calculated to have by NWS in the their 30 year data set, was quite close to the 3850 hrs that space based thermal senses generated. I think this is solid testment to the 1961 – 1990 data sun hr set that NOAA/NWS generated back then.
I'm certainly learning a lot in this thread and don't take offense being corrected. I'm not sure why NCDC could not tell me what machine was measuring sun hours in my city, or any city in the US. And I seem to recall a post from a former meterologist at the Detroit weather station. He claimed that the equipment didn't work some of the time and they used visual observations. I think he claimed that high thin clouds would count towards sun hours, while maybe in Windsor, OT it would have to be more clear to count as sun time. Why would Windsor, OT have almost two hundred less sun hours than Detroit? It seems the US equipment is more liberal in measuring sun hours.
I wish the NCDC made it easier for people to get this data for their locality in the US, and also wish we had more recent data than 1960 to 1990.
I’m certainly no expert in solar geography (lol)...but let me just mention a few things I’ve gathered as a weather observer. You generally seem interested, and I like responding to your posts anyway...
1) NWS SURFACE Climate data - is data observed at principal meteorological stations by trained observers or automated equipment that has been tested and accepted by the controlling agency. It is 100% incorrect (above post) that there is no standard producers at each NWS station...in fact the reverse is true: NWS has standard instrument housing, equipment, procedural data collection...etc. Data is collected on a wide range of time scales from one minute for some current automated equipment... to three observations per day for intermittent periods at some remote, part-time locations. However, normally...the majority of NWS stations collect hourly observations.
In the 1990’s NWS began automating surface weather observations through the new high tech Automated Surface Observing System (ASOS). The measurement of all weather elements included in conventional observations have not (yet) all been automated. Some of those weather elements that have not been automated include clouds above 12,000 ft, snowfall and snow depth, sunshine, and the identification of certain types of weather (hail, drizzle, smoke...etc).
2) The last time NOAA collected sun hr data was from 1961 to 1990 (these are the solar maps I posted with the final 1989 update). This is the reason (above post again) that NWS defers to National Climatic Data Center (NCDC)...its a historical item.
During this time NWS/NOAA was collecting solar data....a sun hr was logged when there was a measurement of direct radiation from the sun. The percentage is given without regard for the intensity of sunshine. That is, thin clouds, light haze, or other minor obstructions to direct solar rays may be present, but would not mitigate the full counting of an hour of sun. So logically, while the case can be made that the calculation of what is termed a “clear” or “party cloudy day” could have some variation to it - it is 100% incorrect that a cloudy day or hr with no sun.... would be counted as a sunny day/hr (or vice versa). NOAA/NWS collected this sun data for 30 years without interruption, and several times updated/changed/cleared errors/streamlined data...etc to produce the last set of 1989 sun hr maps.
So while I have no doubt that there is some error to all sun data...I think it is fair to say the NWS sun hr data collected from 1961 – 1990/and the sun maps final updates in August 1989 before AOS took over are fairly accurate. Additionally, I have seen a few thermal pictures (today modern research academics/advanced professionals in climate almost universally use thermal imaging data(color shaded algorithms ) to examine climate elements)...and noticed that the 4000 hrs a year that Yuma was calculated to have by NWS in the their 30 year data set, was quite close to the 3850 hrs that space based thermal senses generated. I think this is solid testment to the 1961 – 1990 data sun hr set that NOAA/NWS generated back then.
.
To cut a long story short, where you have described collection methods in the period 1961-1990, there is little or no doubt from the above that the C-S technology would record significantly less sunshine in those typical hours than the above methods would. Those "full hours" might thin down to quite small fractions in some cases. I think a general reduction of at least 10% is conservative, and I stand by my comments about Chicago. Ironically, future-past consistency will be compromised by automation projects underway in the UK and New Zealand, where in NZ's case at least the automated equipment yields totals significantly higher than the older technology unless some downward calibration is done.
You say that automation in the US case is not available for sunshine data yet, so we have no information about that. I'm not sure how the space based sensing would compare with C-S.
To cut a long story short, where you have described collection methods in the period 1961-1990, there is little or no doubt from the above that the C-S technology would record significantly less sunshine in those typical hours than the above methods would.
I like the NCDC methods because it sounds like it fits very closely to what I'd register as "sunshine". If I can still fun the sun on me, it still feels like sun to me whether there's a bit of cloud in the way.
But what's important is that the readings are consistent. If wavehunter's description is accurate, perhaps all we need to do is shift the NCDC by about 200 hours lower to match C-S derived ones (or the other way), assuming that the Windsor - Detroit difference is the correct difference between the two methods. Or is it more random?
We need to find as many Canadian – American city pairs as possible and then we can answer this very important question.
I like the NCDC methods because it sounds like it fits very closely to what I'd register as "sunshine". If I can still fun the sun on me, it still feels like sun to me whether there's a bit of cloud in the way.
But what's important is that the readings are consistent. If wavehunter's description is accurate, perhaps all we need to do is shift the NCDC by about 200 hours lower to match C-S derived ones (or the other way), assuming that the Windsor - Detroit difference is the correct difference between the two methods. Or is it more random?
We need to find as many Canadian – American city pairs as possible and then we can answer this very important question.
Quite a few US/Canada border towns we can use... although I'm not sure all have sunshine data...
Vancouver, BC (Surrey, BC actually) - Blaine, WA
Osoyoos, BC - Oroville, WA
Coutts, AB - Sweetgrass, MT
Fort Frances, ON - International Falls, MN
Sault Ste. Marie, MI - Sault Ste. Marie, ON
Niagara Falls, ON - Niagara Falls, NY
St Stephen, NB - Calais, ME
These are just a few...
I'd be interested in knowing why Canadian values are so much less than their nearby US counterparts... I guess there really is a lack of standardization in this regard which is too bad.
Since the US data only goes until 1990, perhaps this would be a good opportunity to update equipment and make it more in line with other countries so it would be easier to compare like with like.
Quite a few US/Canada border towns we can use... although I'm not sure all have sunshine data...
Vancouver, BC (Surrey, BC actually) - Blaine, WA
Osoyoos, BC - Oroville, WA
Coutts, AB - Sweetgrass, MT
Fort Frances, ON - International Falls, MN
Sault Ste. Marie, MI - Sault Ste. Marie, ON
Niagara Falls, ON - Niagara Falls, NY
St Stephen, NB - Calais, ME
These are just a few...
I'd be interested in knowing why Canadian values are so much less than their nearby US counterparts... I guess there really is a lack of standardization in this regard which is too bad.
Since the US data only goes until 1990, perhaps this would be a good opportunity to update equipment and make it more in line with other countries so it would be easier to compare like with like.
I doubt you will get sunshine values for those locations. I'm thinking obstructions can have something to do with it as well as failures of equipment. S-C was developed in the 1880's, pretty old technology.
The US is actually still measuring it, and more accurately. It's just not being done by the NWS, instead by the CRN. See my next post. The new radiation machines are being calibrated apparently to use the same threshold of what is considered "bright sunshine", which is 120W/ meter sq. They don't want to have inconsistencies with past data. It is just the new methods are more reliable and accurate.
To cut a long story short, where you have described collection methods in the period 1961-1990, there is little or no doubt from the above that the C-S technology would record significantly less sunshine in those typical hours than the above methods would. Those "full hours" might thin down to quite small fractions in some cases. I think a general reduction of at least 10% is conservative, and I stand by my comments about Chicago. Ironically, future-past consistency will be compromised by automation projects underway in the UK and New Zealand, where in NZ's case at least the automated equipment yields totals significantly higher than the older technology unless some downward calibration is done.
You say that automation in the US case is not available for sunshine data yet, so we have no information about that. I'm not sure how the space based sensing would compare with C-S.
I found a very interesting article about the testing of US sun sensors in 2001 for the ASOS deployment. Basically, the US used Foster-Foskett for almost 60 years (around 1953 on). Here is a quote from the article written around 2000:
Currently, sunshine duration measurements inthe United States are reported from ~100 NWS surface weather observing stations using the Foster-Foskett
sunshine switch (A-081). Testing by Hughes STX has
shown that the potentiometer in the Foster-Foskett
requires frequent adjustment to maintain a consistent
voltage threshold for sunshine (Hughes STX, 1996).
Additionally I found this article from a local US paper about their Foster-Fosket machine. It uses a threshold of 120watts/sq meter to detect "bright sunshine". When the machine was offline observers would do the work.
Currently it seems NWS offices are not recording it, but the data is now recorded using more accurate methods. This from the Jacksonville NWS office:
EFFECTIVE THURSDAY...JULY 1 2010...THE FOSTER-FOSKETT SUNSHINE SWITCH SENSOR AT JACKSONVILLE INTERNATIONAL AIRPORT WILL BE DEACTIVATED. THE RECORDING OF DAILY MINUTES OF SUNSHINE...AND THUS ITS REPORTING IN THE LOCAL CLIMATOLOGICAL DATA /LCD/ PUBLICATION...WILL CEASE WITH THE DEACTIVATION OF THE SENSOR.
THE FOSTER-FOSKETT SUNSHINE SWITCH SENSOR IS NEARLY 60 YEARS OLD...OBSOLETE...AND INCREASINGLY DIFFICULT...AND COSTLY...TO MAINTAIN AND CALIBRATE. IN ADDITION...THE NATIONAL WEATHER SERVICE /NWS/ CURRENTLY HAS NO REQUIREMENT TO RECORD MINUTES OF SUNSHINE...AND HAS ALREADY STOPPED THIS PRACTICE AT NUMEROUS LOCATIONS ACROSS THE COUNTRY.
SUNSHINE DATA...IN THE FORM OF RADIATION DATA WHICH CAN BE CONVERTED TO MINUTES OF SUNSHINE...IS AVAILABLE AT THE NATIONAL CLIMATE DATA CENTER /NCDC/ FROM SEVERAL NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION /NOAA/ OWNED NETWORKS...INCLUDING THE CLIMATE REFERENCE NETWORK /CRN/ AND SURFACE RADIATION /SURFRAD/ NETWORK. THESE DATA ARE CONSIDERED TO BE MORE ACCURATE AND RELIABLE THAN THE MINUTES OF SUNSHINE CURRENTLY OBTAINED BY THE FOSTER-FOSKETT SUNSHINE SENSOR. THE CRN ALONE HAS 114 SITES ACROSS THE COUNTRY. THE SITE CRN SITE CLOSEST TO JACKSONVILLE IS KNOWN AS GA BRUNSWICK 23 S – CUMBERLAND ISLAND NATIONAL SEASHORE (STAFFORD FIELD). IT IS LOCATED 28 MILES NORTHEAST OF JACKSONVILLE INTERNATIONAL AIRPORT.
And finally from a journal that discussed how the world meteo organization came to use Stokes-Campbell's bright sunshine threshold of 120Watts/sq meter in order to standardize world sunshine levels:
"With regard to the latter, the settlement of a direct solar threshold
irradiance corresponding to the burning threshold of the Campbell-Stokes recorders was strongly advised. Investigations at different
stations showed that the threshold irradiance for
burning the card varied between 70 and 280W m-2 (Bider, 1958; Baumgartner, 1979). However,further investigations, especially performed with the IRSR in France, resulted in a mean value of 120W m–2, which was finally proposed as the threshold of direct solar irradiance to distinguish bright
sunshine. "
It seems that both the Foster-Fosket and the Stokes-Campbell used 120Watts/Sq meter as the threshold. And in fact, 120Watts was selected because that was the burning threshold for the card used in S-C.
I would think the only time the measurements are impaired is thru obstructions and equipment breakages. Then observers take over. I would think that happens all over the world. I'm now not sure why Detroit and Windsor varied, maybe the equipment in Windsor was more obstructed than Detroit (which seems a common problem).
I couldn't find any article that mentioned S-C giving less sun hours than Foster-Fosket. It may be the case, but I think now the US data more reliable than before. And certainly what I found is that S-C machines give different values all over the map as well. Until I find some article that gives credible evidence of US machines giving more sun hours than S-C, I think I'm going to stick with the sun hours listed. I respect your feelings about your time in Chicago, but seriously, that would never stand up to scientific scrutiny.
Sydney is back to 2500 hours, nothing special. No wonder a local friend of mine (from England originally) who lived in Sydney before Philly likes our summers here much better than Sydney, and he thinks the ocean water here warmer as well. And Melbourne is not all that sunny.
Before this is done and dusted, I am going to consult an expert in the UK who has been studying these issues all his life. I feel that there must be some kind of basis for the assertion made in an old manual about US comparability. I don't think you have sufficiently explained Canada-US discrepancies for sites a relatively short distance apart, either - merely speculated. I may have to repeal my opinion on all of this, but not before I get some feedback from this other source. And rememember that there are some parts of Sydney at least that have averaged 2625 hours over a fairly substantial period - around the airport area, anyway.
It would be interesting to know which technology is being used to get radiation data (direct solar only) that can be converted to sunshine duration.
In the interim, as a check, you could ask these sources if the F-F machines registered sunshine in the first and last 15 minutes of daylight in temperate latitudes. The C-S instruments in NZ were known to be below the threshold for burning
in those timespans, so the measureable portion of the day (before allowing for any obstructions at other times) was reduced by 30 minutes for percentage calculation purposes.
I like the NCDC methods because it sounds like it fits very closely to what I'd register as "sunshine". If I can still fun the sun on me, it still feels like sun to me whether there's a bit of cloud in the way.
But what's important is that the readings are consistent. If wavehunter's description is accurate, perhaps all we need to do is shift the NCDC by about 200 hours lower to match C-S derived ones (or the other way), assuming that the Windsor - Detroit difference is the correct difference between the two methods. Or is it more random?
We need to find as many Canadian – American city pairs as possible and then we can answer this very important question.
Also, one other interesting consideration…don’t discount how rapidly solar gradients can change:
An old college professor I knew used to try this trick T/F question on his first year physical geography students every year; He would say with a smile “Pittsburgh is only 4 hrs by motor car away from Baltimore, MD….does Baltimore, MD get 25% more sun hrs annually and 35% more sun hrs in winter months than Pittsburgh? Almost universally everyone fell for it - and firmly stated FALSE.
I realize that Windsor and Detroit are much closer…but little physiographic, climatic, and even manmade differences can add up to alot when a weather element is calculated annually.
Before this is done and dusted, I am going to consult an expert in the UK who has been studying these issues all his life. I feel that there must be some kind of basis for the assertion made in an old manual about US comparability. I don't think you have sufficiently explained Canada-US discrepancies for sites a relatively short distance apart, either - merely speculated. I may have to repeal my opinion on all of this, but not before I get some feedback from this other source. And rememember that there are some parts of Sydney at least that have averaged 2625 hours over a fairly substantial period - around the airport area, anyway.
It would be interesting to know which technology is being used to get radiation data (direct solar only) that can be converted to sunshine duration.
In the interim, as a check, you could ask these sources if the F-F machines registered sunshine in the first and last 15 minutes of daylight in temperate latitudes. The C-S instruments in NZ were known to be below the threshold for burning
in those timespans, so the measureable portion of the day (before allowing for any obstructions at other times) was reduced by 30 minutes for percentage calculation purposes.
I'm really enjoying this as I love science and am learning a lot. It would be great to hear what your friend in the UK thinks. He may have another twist on this. I recall reading that the F-F, and also from the email I got from NCDC, is that they consider daylight hours based (to be used in the % calc's) on the Navy's time of official sunrise and sunset (disc above horizon). Also that a visible disc on a clear day will burn the SC card at 120W per sq meter> They also make adjustments for the fact the yearly daylight hours are not the same from the equator to the poles. There is a slight variation.
Here is Canada:
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.
