outliers in global projections

Home Page Forums C3S_422_Lot1_SMHI forum outliers in global projections

This topic contains 5 replies, has 3 voices, and was last updated by  paugallesisardsat 1 week, 5 days ago.

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • #3923

    paugallesisardsat
    Participant

    Dear all,

    We have downloaded daily climate projection global maps for the tmax (maximum temperature) and prepared some plots to summarize its content.
    There are some anomalous values and we are not sure whether if they are errors in the execution or expected.

    The figure below is plotting the minimum, mean and maximum TMAX values along time. (One value for each day and for each global map)

    TMAX-min-mean-max

    We have observed that all maximum values are lower than ~57Celsius degrees except 2.
    These 2 anomalies are 67 and 96 Celsius degrees respectively.

    Our aim is to downscale from these models. Are these values expected? Should we remove and treat them?

    Regards,

    Pau

    #3965

    KKlehmet
    Participant

    Dear Pau,

    can you please specify what model, realization, scenario and time period you have used. What do you mean by ‘each global map’?
    Have you downloaded the data from the ESGF portal?

    Best,
    Katharina

    • This reply was modified 1 month ago by  KKlehmet.
    #4051

    paugallesisardsat
    Participant

    Dear Katharina,

    This has been indeed downloaded from ESGF portal, the filter used was

    Project > CMIP5
    Model > EC-EARTH
    Experiment > rcp45
    Time Frequency > day
    Ensamble > r1i1p1
    Variable > tasmax

    The procedure for the plot was to
    1. load all the data in a matrix* with 3 axis (time, latitude, longitude)
    2. Then plot the average for each day (along the time axis)
    In my previous explanation ‘each global map’ refers to each latitudes-longitudes slide in the matrix*. Therefore the average was for each slide.

    Regards,

    Pau

    #4141

    KKlehmet
    Participant

    Dear Pau,

    Unfortunately we cannot give further support to answer questions related to the quality of the data from the ESGF portal. The previous information from our side was only an interim solution before the data will be available on the Climate Data Store. From now on we can only provide support for the data presented here on the web demonstrator. We used EC-EARTH r12i1p1 in the model ensemble for producing climate impact indicators.

    However, in order to ask further questions related to EC-EARTH you might have a look in the metadata and check the contact/email address where you could pose you question to. Another option would be to check the ESGF website (https://esgf.llnl.gov/) or the website where the data was downloaded from as e.g. https://esg-dn1.nsc.liu.se/projects/esgf-liu/.

    Best,
    Katharina

    #4180

    FulcoLudwig
    Participant

    Dear Pau,
    the WMO record for the highest official temperature on Earth is 57.8 °C so you can safely assume that these outliers are errors and you should disregard them.
    Regards
    Fulco

    #4183

    paugallesisardsat
    Participant

    Thanks Fulco, that makes sense. I will then interpolate the conflictive pixels with the neighbor values.

Viewing 6 posts - 1 through 6 (of 6 total)

You must be logged in to reply to this topic.