search

monocongo / climate_indices

Climate indices for drought monitoring
https://monocongo.github.io/climate_indices/
Other
343 stars 168 forks source link

SPI: days of No Data cause yield SPI with no data #368

Open aleccourt opened 4 years ago

aleccourt commented 4 years ago

When trying to calculate SPI (daily in this case) with the new "spi" argument, any day with no data yields SPI values of no data for any scale that spans that one day of No Data. This shouldn't happen since there is plenty of data to calculate an SPI value.

monocongo commented 4 years ago

Hey Alec can you please post or link to a small file with example data that causes the problem? Also is this for Pearson or gamma or both?

aleccourt commented 4 years ago

Hey James,

Please see attached for an example file. The issue happens for both Pearson and Gamma, so I believe it may be an issue that occurs before the fitting - potentially some issue when the input data is aggregated to the specified scale. cmorph_0.25deg_DLY_nanDay_example.nc https://drive.google.com/file/d/1VJgF1NW_9PgllZn1qSGN0hfS2nddbsgc/view?usp=drive_web

Cheers, Alec

On Tue, Feb 11, 2020 at 7:34 PM James Adams notifications@github.com wrote:

Hey Alec can you please post or link to a small file with example data that causes the problem? Also is this for Pearson or gamma or both?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/monocongo/climate_indices/issues/368?email_source=notifications&email_token=ALQIYRBI2HDK5GWESDVR3BDRCNU4LA5CNFSM4KTKZMF2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOELPI2PQ#issuecomment-585010494, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALQIYRG7MDZERLUDC6TTN23RCNU4LANCNFSM4KTKZMFQ .

-- Alec Courtright Co-Founder | Destination SPACE GIS Analyst | ISciences, L.L.C. NOAA National Centers for Environmental Information Cell: (864)-607-3377

monocongo commented 4 years ago

Thanks @aleccourt

Is there a particular lat/lon you know of where this happens or is there a certain day that's missing for all lats/lons (looks like this is the case for Jan 28th, 2020), etc.? I'm trying to narrow down where/when this happens so I can follow the computation path and see where it goes off the rails. Also if you can provide the command you're using that triggers the error that might be helpful as well.

aleccourt commented 4 years ago

Hey James,

There’s not a particular location. Unfortunately the intermediate CMORPH data has some full days without any data globally. I know that Oct. 1st is missing data.

Alec

On Wed, Feb 12, 2020 at 2:41 PM James Adams notifications@github.com wrote:

Thanks @aleccourt https://github.com/aleccourt

Is there a particular lat/lon you know of where this happens or is there a certain day that's missing for all lats/lons (looks like this is the case for Jan 28th, 2020), etc.? I'm trying to narrow down where/when this happens so I can follow the computation path and see where it goes off the rails.

— You are receiving this because you were mentioned.

Reply to this email directly, view it on GitHub https://github.com/monocongo/climate_indices/issues/368?email_source=notifications&email_token=ALQIYRGLK2G37BFWAUJHIU3RCR3JRA5CNFSM4KTKZMF2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOELSVL3I#issuecomment-585455085, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALQIYRBNCZEERDJ6CAPDWIDRCR3JRANCNFSM4KTKZMFQ .