5
$\begingroup$

I've been searching for a 2017 or 2018 or 2019 (or any combo) gridded dataset for surface temperature for the globe with a 0.1 degree resolution. Unfortunately even after hours of searching, for this time range, I've only been able to find ones with at most 0.5 degree resolution. Does anyone have any recommendations of where I might be able to find this kind of data?

If gloabal temperature isn't available, I'm specifically searching for Africa, South Asia, and South America (or any combo of those) - so if anyone knows where this data is available, it would be much appreciated!

$\endgroup$
  • $\begingroup$ Do you mean the average over the entire year? You could calculate this from ERA-5 data, but 0.1 degree may be a little too high resolution to ask for; that would be in the order of 10 km. Will 0.25° do? I don't know if there are regional models for the continents you ask for, as most regional models are centered at the richer areas of the world. $\endgroup$ – gerrit Jan 22 at 8:13
  • $\begingroup$ I don't know if the data are available, but Ashrit et al. (2020) describe a regional reanalysis over the Indian subcontinent. Does that help? $\endgroup$ – gerrit Jan 22 at 10:09
  • $\begingroup$ Not an expert, bit the ICON weathermodel and the ECMWF (since 2016?) pride themselves of such a global resolution. But i don't know about details and public availability, only the colourful images on the internet :-) Sources: ecmwf.int/en/about/media-centre/news/2016/… and dwd.de/EN/research/weatherforecasting/num_modelling/… $\endgroup$ – user18607 Jan 22 at 10:38
  • 2
    $\begingroup$ @ebv even if some of the modeling behind could be similarly, weather forecasts and global reanalyses are two different products (ecmwf.int/en/research/climate-reanalysis). Climate reanalyses use high resolution models (that could be the same as the ones used in the forecasts) to reproduce recent climate in accordance to the observations from the meteorological stations on the ground. The resolution of the models could be much finer than 0.1 degrees, but the outputted data at global level at that resolution would be enormous and therefore these datasets very expensive to manage $\endgroup$ – Nemesi Jan 22 at 11:37
  • $\begingroup$ Ocean sea surface temperature is available from GOFS 3.1 (hycom.org/dataserver/gofs-3pt1/analysis). The most recent configuration is 0.08 deg in longitude and 0.04 deg in latitude. $\endgroup$ – John Jan 22 at 20:50
9
$\begingroup$

The most important reanalysis datasets covering until present time (I'm excluding the others that stop some years before present time) currently available are:

  1. ERA5 by ECWMF: 1979-present (soon 1950-present), temporal resolution of 1 hour and spatial resolution quarter degree (0.25 x 0.25);
  2. CRU by the UEA: 1900-present (some variables 1850-present), temporal resolution 3 or 6h, spatial resolution half degree (0.5 x 0.5);
  3. GLDAS by NASA: 1948-present, temporal resolution 3 or 6h, spatial resolution quarter degree (0.25 x 0.25).

(please, feel free to edit the answer to add more)

You could download one of those and process these temperature data using a simple resampling procedure bringing it at the resolution you are interested in by using CDO or another similar software. The procedure could be based on a bilinear interpolation of the original data with a finer grid. This would bring uncertainty because the correct procedure would be to use the reanalysis as boundary conditions in a regional climate model, but it would give a good approximation of the data you are looking for.

The only dataset I am aware of that is providing global data at a resolution of a 0.1 deg. resolution is the MSWEP by JRC and Princeton, but present data are not available for free, and it includes only precipitation data.

$\endgroup$
  • $\begingroup$ This is awesome, and if 0.1 degree resolution doesn't exist, I think 0.25 will work! Thanks for your answer; I'll leave the question open for a couple more days, and then mark as best! $\endgroup$ – F16Falcon Jan 22 at 14:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.