Data strategies
An overview of the different data strategies for each DeepMIP group is given on the following pages:
The DeepMIP model database
Model data for all DeepMIP groups are organised in a common database structure. The output from each modelling group is initially uploaded to a central data storage at the University of Bristol, where new data is automatically checked for consistency across the existing ensemble. CF-compliant metadata is added to all model fields and the quality-checked data is moved to a long-term storage to be accessible to the wider community following the dissemination strategy of each group.
Uploading data
Please follow these minimum requirements to streamline the upload, quality check and distribution of your data:
- we strongly encourage you to upload the full monthly mean time series data of the last 100 model years for each simulation (1200 time steps)
- each variable should be stored in a separate file following the naming convention below
- as a minimum, please change variable names and units to be consistent with the CMIP6 conventions listed below
- other metadata do not need to be modified and will be added to all data before wider dissemination
- once you finished uploading new data, please use this quick form to trigger the further processing of the data
An example of the final directory and file naming structure for the Eocene phase 1 model data is available here.
Getting access (for model data providers)
- ask Seb to be added as a ‘collaborator’ to the DeepMIP model database
- you will receive an invitation link to register with the University of Bristol – no personal data necessary!
- confirm email address and we will approve your application
- access the “DeepMIP_Model_output_shared” database either by WebDAV or SFTP
You only need to register once and can use the same account to upload data to different DeepMIP groups.
Naming Convention
Output file names should follow the DeepMIP model naming convention outlined in the reference publication preprint. Each output variable is stored in a separate file according to the following structure:
directory = /<Project>/<Family>/<Model>/<Experiment>/<Version>/<Averaging>/
filename = <Variable>_<Model>_<Experiment>_<Version>.<Statistic>.nc
where:
-
<Project>is currently either:deepmip-eocene-p1: Eocene phase 1deepmip-miomip2-p1: MioMIP2 phase 1
<Family>and<Model>can be chosen by each modelling group (please use hyphens instead of underscores), e.g. “/CESM/CESM1.2-CAM5”<Experiment>options are listed in the individual “Data strategies” linked at the top of the page<Variable>names are listed in the first column in Tables 3 and 4-
<Statistic>is either:mean: climatological monthly means over last 100 model years (12 timesteps)std: climatological monthly standard deviation over last 100 model years (12 timesteps)time_series: monthly mean data of last 100 model years (1200 timesteps)<Statistic>is omitted for the time-independent boundary conditions
- the smaller
meanandstdfiles are stored in the<Averaging>=climatologydirectory and are separated from the largertime_seriesfiles in the<Averaging>=time_seriesdirectory to enable more granular download options
Storing all relevant information in the file name itself allows all phases of coordinated DeepMIP experiments to be integrated into a single database/directory in the future.
To facilitate the generation and analysis of the standardised model output, up-to-date Python dictionaries containing all experiment and variable names are available in this repository .
Variable names
Variable names, units and signs of fluxes follow the naming convention of the Coupled Model Intercomparison Project 6 (CMIP6) data request wherever possible, to adhere to community standards and to allow the integration into existing processing and analysis workflows.
Table 3: Atmosphere output variables following the conventions of the CMIP6 data request.
| Name | Long Name | Units | Dimensions |
|---|---|---|---|
| tas | Near-Surface Air Temperature | K | time×lat×lon |
| ts | Surface Temperature | K | time×lat×lon |
| pr | Precipitation | kg m-2 s-1 | time×lat×lon |
| evspsbl | Evaporation Including Sublimation and Transpiration | kg m-2 s-1 | time×lat×lon |
| hfls | Surface Upward Latent Heat Flux | W m-2 | time×lat×lon |
| hfss | Surface Upward Sensible Heat Flux | W m-2 | time×lat×lon |
| ps | Surface Air Pressure | Pa | time×lat×lon |
| psl | Sea Level Pressure | Pa | time×lat×lon |
| snc | Snow Area Fraction | % | time×lat×lon |
| rsds | Surface Downwelling Shortwave Radiation | W m-2 | time×lat×lon |
| rlds | Surface Downwelling Longwave Radiation | W m-2 | time×lat×lon |
| rsus | Surface Upwelling Shortwave Radiation | W m-2 | time×lat×lon |
| rlus | Surface Upwelling Longwave Radiation | W m-2 | time×lat×lon |
| rsdt | TOA Incident Shortwave Radiation | W m-2 | time×lat×lon |
| rsut | TOA Outgoing Shortwave Radiation | W m-2 | time×lat×lon |
| rlut | TOA Outgoing Longwave Radiation | W m-2 | time×lat×lon |
| rsdscs | Surface Downwelling Clear-Sky Shortwave Radiation | W m-2 | time×lat×lon |
| rldscs | Surface Downwelling Clear-Sky Longwave Radiation | W m-2 | time×lat×lon |
| rsuscs | Surface Upwelling Clear-Sky Shortwave Radiation | W m-2 | time×lat×lon |
| rluscs | Surface Upwelling Clear-Sky Longwave Radiation | W m-2 | time×lat×lon |
| rsutcs | TOA Outgoing Clear-Sky Shortwave Radiation | W m-2 | time×lat×lon |
| rlutcs | TOA Outgoing Clear-Sky Longwave Radiation | W m-2 | time×lat×lon |
| tauu | Surface Downward Eastward Wind Stress | Pa | time×lat×lon |
| tauv | Surface Downward Northward Wind Stress | Pa | time×lat×lon |
| uas | Eastward Near-Surface Wind | W m-2 | time×lat×lon |
| vas | Northward Near-Surface Wind | W m-2 | time×lat×lon |
| clh | High Level Cloud Fraction | % | time×lat×lon |
| clm | Mid Level Cloud Fraction | % | time×lat×lon |
| cll | Low Level Cloud Fraction | % | time×lat×lon |
| clt | Total Cloud Cover Percentage | % | time×lat×lon |
| cl | Percentage Cloud Cover | % | level×time×lat×lon |
| hus | Specific Humidity | 1 | level×time×lat×lon |
| ta | Air Temperature | K | level×time×lat×lon |
| ua | Eastward Wind | m s-1 | level×time×lat×lon |
| va | Northward Wind | m s-1 | level×time×lat×lon |
| wap | Omega (=dp/dt) | Pa s-1 | level×time×lat×lon |
| zg | Geopotential Height | m | level×time×lat×lon |
| orog | Surface Altitude | m | lat×lon |
| sftlf | Percentage of the Grid Cell Occupied by Land | % | lat×lon |
Table 4: Ocean output variables following the conventions of the CMIP6 data request.
| Name | Long Name | Units | Dimensions |
|---|---|---|---|
| tos | Sea Surface Temperature | °C | time×lat×lon |
| siconc | Sea-Ice Area Percentage (Ocean Grid) | % | time×lat×lon |
| mlotst | Ocean Mixed Layer Thickness Defined by Sigma T | m | time×lat×lon |
| zos | Sea Surface Height Above Geoid | m | time×lat×lon |
| hfds | Downward Heat Flux at Sea Water Surface | W m-2 | time×lat×lon |
| wfo | Water Flux Into Sea Water | kg m-2 s-1 | time×lat×lon |
| tauuo | Sea Water Surface Downward X Stress | N m-2 | time×lat×lon |
| tauvo | Sea Water Surface Downward Y Stress | N m-2 | time×lat×lon |
| msftbarot | Ocean Barotropic Mass Streamfunction | kg s-1 | time×lat×lon |
| msftmz | Ocean Meridional Overturning Mass Streamfunction | kg s-1 | time×depth×lat |
| so | Sea Water Salinity | 0.001 | depth×time×lat×lon |
| thetao | Sea Water Potential Temperature | °C | depth×time×lat×lon |
| uo | Sea Water X Velocity | m s-1 | depth×time×lat×lon |
| vo | Sea Water Y Velocity | m s-1 | depth×time×lat×lon |
| wo | Sea Water Vertical Velocity | m s-1 | depth×time×lat×lon |
| difvmo | Ocean Vertical Momentum Diffusivity | m-2 s-1 | depth×time×lat×lon |
| difvtrbo | Ocean Vertical Tracer Diffusivity Due to Background | m-2 s-1 | depth×time×lat×lon |
| deptho | Sea Floor Depth Below Geoid | m | lat×lon |