Problems with precip_accum_24_hour_set_1

[SoonaHabibi]

Please can you provide information on the logic used to calculate the precip_accum_24_hour_set_1? We are trying to obtain an hourly time series of rainfall. We have tried to convert the data in multiple ways, validating it against the precip_accum_since_local_midnight_set_1, but we cannot get the correct hourly rainfall no matter the logic we apply. Thank you, Luciana

[joeyoun9]

Hi Luciana,

precip_accum_24_hour_set_1 is a variable reported by the station, so unfortunately it is up to the individual station data logging system to determine how it calculates that value. There is not necessarily any other variable which directly relates to that one depending on how the site measures and reports precipitation.

The Precipitation service is an API tool that meshes together all the various forms of precipitation reports from stations into a single integrated precip measurement over various time period options. If you would like to do more complex things (such as hourly time series), you can try that. Do note that even the precipitation service cannot produce precipitation amounts at a finer time resolution than what the station reports.

If this resolves your issue, we will close this in a few days.

Regards,

Joe Young

[jbpechmann]

Luciana,

precip_accum_24_hour_set_1 is a rolling 24 hour total. It may be what is reported by the stations or calculated. In either case, if precip_accum_24_hour_set_1 is given hourly then taking the difference between two hours gives you the amount of precipitation within that hour minus the amount in the hour 24 hours previous. Therefore, it is not straightforward to obtain the hourly amount from the rolling 24 hour total. If precip_accum_since_local_midnight_set_1 is available, taking the difference of that between hours will give you the hourly total. If you have additional questions, please let us know what station(s) you are interested in.

Judy Pechmann MesoWest

On 8/20/2019 4:17 PM, soona1989 wrote:

Please can you provide information on the logic used to calculate the precip_accum_24_hour_set_1? We are trying to obtain an hourly time series of rainfall. We have tried to convert the data in multiple ways, validating it against the precip_accum_since_local_midnight_set_1, but we cannot get the correct hourly rainfall no matter the logic we apply. Thank you, Luciana

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/mesowx/issues/issues/13?email_source=notifications&email_token=AASTVOSUUT3B5QFK5SDQBX3QFRUPBA5CNFSM4IN5VC42YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HGLVCZA, or mute the thread https://github.com/notifications/unsubscribe-auth/AASTVOU4GUM27JPDI7IQIALQFRUPBANCNFSM4IN5VC4Q.

[SoonaHabibi]

Hi Judy,

We actually tried the method that you explained before. But, the problem was that sometimes we get negative value for the rainfall!! Do you know what could be the reason? Unfortunately the precip_accum_since_local_midnight_set_1 is only available after mid 2015 and we need data from 2011. Station that we are working with are D1628, D1439, C1063 (DW1628 Glide, DW1439 Eugene, CW1063 Reedsport).

Thanks, Soona Do

[jbpechmann]

hi Soona,

Taking the difference between two values of precip_accum_24_hour which are separated by one hour can give a negative result. For example, suppose we are taking the difference between 0:00 and 1:00. If it didn't rain between 0:00 and 1:00 on the current day but it did rain 1" between 0:00 and 1:00 on the previous day, the difference you will get is -1". If, in that hour, it rained 1" on the current day and 2" on previous day, you would still get -1" difference. This problem is why it is so difficult to get the hourly totals from a rolling 24 hr accumulation.

Also, since the stations you are using are from the CWOP network, which are owned and operated by private citizens, the siting and maintenance of the stations may not be optimal. The data may be good, but should be used with caution.

Judy

[SoonaHabibi]

Thanks Judy!

Soona

On Mon, Aug 26, 2019 at 9:56 AM jbpechmann notifications@github.com wrote:

hi Soona,

Taking the difference between two values of precip_accum_24_hour which are separated by one hour can give a negative result. For example, suppose we are taking the difference between 0:00 and 1:00. If it didn't rain between 0:00 and 1:00 on the current day but it did rain 1" between 0:00 and 1:00 on the previous day, the difference you will get is -1". If, in that hour, it rained 1" on the current day and 2" on previous day, you would still get -1" difference. This problem is why it is so difficult to get the hourly totals from a rolling 24 hr accumulation.

Also, since the stations you are using are from the CWOP network, which are owned and operated by private citizens, the siting and maintenance of the stations may not be optimal. The data may be good, but should be used with caution.

Judy

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/mesowx/issues/issues/13?email_source=notifications&email_token=AJOD55YWI6GHZSZU4X4KQSLQGQDJ5A5CNFSM4IN5VC42YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD5E65PY#issuecomment-524938943, or mute the thread https://github.com/notifications/unsubscribe-auth/AJOD556GXSNOMPW2YUHIBDLQGQDJ5ANCNFSM4IN5VC4Q .