Geophysical data

Where to find them ?

Cyril Gadal

Institut de Mécanique des Fluides de Toulouse (IMFT), France

About me

  • 2021 – now: Postodctoral researcher, IMFT, Toulouse
    • Particle-laden (turbidity) currents
    • Experimental work

Turbidity current building a deposit over time. Generated by the steady injection of a suspension

About me

  • 2021 – now: Postdoctoral researcher, IMFT, Toulouse
    • Particle-laden (turbidity) currents
    • Experimental work
  • 2017 – 2020: PhD Student, IPGP, Paris

    • Sand dune patterns

    • Theory, experiments, numerical simulations and geophysical data

      Dunes gathering at the bottom of the Mazar Tagh (small mountain) in the Taklamacan desert, in China

Various data types

The Blue Marble (NASA)

  • 9 topics (EarthData – NASA):

Atmopshere

Biosphere

Cryosphere

Human dimension

Land surface

Ocean

Solid Earth

Sun-Earth interactions

Terrestrial hydrosphere

Where to find them ?

Two global repositories

NASA (US)

Copernicus (EU)

Different levels of Data processing (repository dependent)

  • Level 0: Unprocessed instrument data
  • Level 1: Time-referenced, georeferenced, annotated data
  • Level 2: Derived geophysical variables at the same resolution and location as L1 source data.
  • Level 3: Variables mapped on uniform space-time grid scales, usually with some completeness and consistency.
  • Level 4: Model output or results from analyses of lower-level data (e.g., variables derived from multiple measurements).

Today

Topographical data

Lidar-derived Digital Elevation Model over Zion National Park, Utah (USGS)

Meteorological data

Current wind flow over Europe (ECMWF forecast).

Topographical data

Lidar-derived Digital Elevation Model over Zion National Park, Utah (USGS)

DEMs: Digital Elevation Models

  • Gridded elevation data as \(Z(x, y)\), \(Z(lat, lon)\), etc..
  • Not cloud points (already ‘interpolated/gridded’)
  • Not Discret Element Method :)

Different origins (linked to spatial scale/resolution)

  • Satellite data
  • LIDAR-derived data (plane, drone, hand instruments)
  • Hand measurements (Theodolite, GPS, photogrammetry, …)

Topographical data

Best global coverage: SRTM

  • Shuttle Radar Topography Mission
  • horizontal resolution: \(1'' \sim 30~\textrm{m}\)
  • only land
  • vertical precision, complicated:
    • depend on the place
    • systematic bias, always positive on forested areas
    • \(\sim 10~\textrm{m}\)
  • easy access: EarthData or Map dowload

Illustration of the space shutlle Endeavour orbiditng above the Earth (NASA)

successor: NASADEM

  • released in 2020
  • less holes in dataset, less vertical error
  • more extensive (spatial vertical precision, calculation of slope/curvature)
  • access: EarthData

Formats: .hgt or NETCDF

  • .hgt: specific to SRTM mission
    • not a format, just an extension (see here)
    • “raw” format (no headers and not compressed)
    • no metadata included


  • NETCDF (.nc): Network Common Data Form
    • open format
    • can be compressed, optimized for gridded data
    • include metadata
    • not always available yet

Topographical data: an example

Meteorological data

Different origins

  • In situ observations
  • Satellite observations
  • model forecasts
  • Reanalyses:
    • data assimilation + models
    • past (+ forecasts)

Meteorological data: Reanalyses

Global datasets

  • ERA series (EU):
    • latest: ERA5-Land, hourly, 9 km mesh
  • MERRA series (US):
    • latest: MERRA-2, hourly, 30 km mesh
  • GLDAS (US):
    • 3 hours, 30 km mesh

Local datasets

  • CERRA:
    • Europe, 3 hours, 5 km mesh
  • CARRA:
    • Arctic, 3 hours, 3.5 km mesh
  • NLDAS:
    • USA, 13 km mesh
  • many others …

Formats

  • GRIB:
    • NOT a file format, just a file with GRIB messages
    • no metadata included
    • often difficult to read (must reshape data, etc ..)
  • NETCDF (.nc): Network Common Data Form
    • open format
    • can be compressed, optimized for gridded data
    • include metadata
    • not always available yet

Meteorological data: an example

Final notes

  • Same principle for other data types
  • protocol depends on repository
  • especially true in Europe/Copernicus
  • API access available most of time
  • many other repositories and models



Questions, other examples ?