Weather Station Questions
56 views
Skip to first unread message
John Doyle
Jun 7, 2022, 1:19:10 PM6/7/22
to wx...@googlegroups.com
My weather station (D7756) currently is QC'd by MADIS. When I open it up on your site and check the MADIS ratings, I get two thumbs up but also get a red X for my daytime temperature. The station is mounted on the roof with a tripod with an extension pole that elevates the station to about 35' total elevation and about 10' above the roof. What is more bizarre is when I pull the data from the last 3, 7 and 14 days, it all shows as two thumbs up with a green checkmark. When I pull data from 4, 8 and 13 weeks I get the two thumbs up with a red X. But when I pull it for 26 weeks I get two thumbs up with a green check. So, basically, it's all over the place. That doesn't give me a lot of confidence in the MADIS tool. Can you please advise as to how I should interpret this data ?
Additionally, I am also getting two thumbs down for the wind vector due to failing the L3 spatial consistency. I checked the other stations in my area, to include the NASA Shuttle Landing Facility, and everyone is failing the same thing. Can you please also advise as to how I should interpret that data, as I just don't see how all of us can fail the same way all the time.
Finally, I have attached some pics of my station as mounted on my house. I was hoping you could add them to my station on your site. I couldn’t find any other way to do so. If there is another process for this please let me know and I will be happy to oblige.
Thank you for your assistance and for doing all the extra work to make all this possible. You folks are nothing short of fantastic.
Regards,
John W. Doyle
tucso...@gmail.com
Jun 7, 2022, 1:33:29 PM6/7/22
to wxqc
Hello John
Your temperature sensor and many others are sitting correctly. Reference:
Hope this helps,
James B
DW4536
tucso...@gmail.com
Jun 7, 2022, 1:38:25 PM6/7/22
to wxqc
Hello John
My last post should have said, "not sitting correctly".
Thanks,
James B
DW4536
On Tuesday, June 7, 2022 at 10:19:10 AM UTC-7 bluco...@gmail.com wrote:
googl...@tedlum.com
Jun 7, 2022, 4:08:35 PM6/7/22
to wxqc
So, MADIS performs 3 levels of quality checks (L1, L2, & L3), beginning with whether data is even with a valid range, then whether it's rate of change is rational, then finally whether it is statistically similar to other stations. There is also another check for how many individual readings readings fail over time. Those checks result in a bitmap value being added to each row of data you submit and become part of that record. In addition MADIS produces a QA data set that includes the analysis values derived for other stations.
The thumbs are tied to the 3 levels of quality checks, which are pass fail, and simply tracks how MADIS tagged the data. The green-check/red-X are based on the Q/A data set which Philip analyses locally and is also what drives the red analysis line in the graphs. Philip set a specific variance threshold for each parameter where the average falls outside, which will vary depending on how many days you include in the average.
So, now I have to give you the standard siting talk:
The rain gauge should always be as low to the ground as possible, not in the shadow of tall objects, and cleaned frequently to remove debris. When placed where you have it, you will experience poor data quality due to wind driven under-catch, and due to it's location it will not be cleaned as often as need be.
The temperature/humidity sensors should be placed at about 4' above the predominant natural terrain - in the east that may be grass, in the (semi-)arid west that may be just sand/gravel. Locations that have a line-of-sight view (radiation) of, or are likely to experience convection from, uncharacteristic surfaces like asphalt streets/parking lots or roofs, is strongly discouraged. When placed where you have them, you will experience temperature/humidity errors, particularly in the daytime and extending into the evening, until those surfaces cool, resulting from both convection and direct radiation. Convection is primarily an issue in the summer but you'll still experience errors from radiation in the winter.
The wind instruments should be placed at 10 meters (33'), and not within the influence of obstacles which block, or alter the wind speed and direction, like trees and roofs. Where you have them located appears to be below the roof peak. You're likely to experience complex errors from the eddys caused by the building walls and sloped roof plains.
Recommendation would be to separate the wind instruments from the rest of the station, moving the tripod to the roof peak, and the rest of the station to a post or poll over native vegetation.
The thumbs are tied to the 3 levels of quality checks, which are pass fail, and simply tracks how MADIS tagged the data. The green-check/red-X are based on the Q/A data set which Philip analyses locally and is also what drives the red analysis line in the graphs. Philip set a specific variance threshold for each parameter where the average falls outside, which will vary depending on how many days you include in the average.
So, now I have to give you the standard siting talk:
The rain gauge should always be as low to the ground as possible, not in the shadow of tall objects, and cleaned frequently to remove debris. When placed where you have it, you will experience poor data quality due to wind driven under-catch, and due to it's location it will not be cleaned as often as need be.
The temperature/humidity sensors should be placed at about 4' above the predominant natural terrain - in the east that may be grass, in the (semi-)arid west that may be just sand/gravel. Locations that have a line-of-sight view (radiation) of, or are likely to experience convection from, uncharacteristic surfaces like asphalt streets/parking lots or roofs, is strongly discouraged. When placed where you have them, you will experience temperature/humidity errors, particularly in the daytime and extending into the evening, until those surfaces cool, resulting from both convection and direct radiation. Convection is primarily an issue in the summer but you'll still experience errors from radiation in the winter.
The wind instruments should be placed at 10 meters (33'), and not within the influence of obstacles which block, or alter the wind speed and direction, like trees and roofs. Where you have them located appears to be below the roof peak. You're likely to experience complex errors from the eddys caused by the building walls and sloped roof plains.
Recommendation would be to separate the wind instruments from the rest of the station, moving the tripod to the roof peak, and the rest of the station to a post or poll over native vegetation.
Reply all
Reply to author
Forward
0 new messages