annex 7.4.4
7.4.4 ECMWF Reanalysis 5th Generation (ERA5) Atmospheric Reanalysis

Overview

This document provides an overview of single-level wind products of ERA5, in the context of the larger ERA5 dataset. As background, ERA5 is the 5th generation of the global atmospheric reanalysis (the latest – it replaces the ERA-Interim reanalysis) produced by the Copernicus Climate Change Service at ECMWF, covering the period from January 1950 to present. It provides hourly data on many atmospheric, land-surface and sea-state parameters together with estimates of uncertainty.

Provider's contact information

ERA5 is produced by the Copernicus Climate Change Service (C3S) at ECMWF.

Copernicus User support (copernicus-support@ecmwf.int (external to C3S) ).

Licensing

Licence: Copernicus (Licence agreement information can be found here or here).

Dataset citable as: Copernicus Climate Change Service (C3S) (2017): ERA5: Fifth generation of ECMWF atmospheric reanalyses of the global climate . Copernicus Climate Change Service Climate Data Store (CDS), date of access. https://cds.climate.copernicus.eu/cdsapp#!/home

Variable name and units:

Single-level wind of ERA5 over northern Canada is the main focus of this document. The parameter description is provided in the table below.

Name Units Description
100 m u-component of wind m s-1 This parameter is the eastward component of the 100 m wind. It is the horizontal speed of air moving towards the east, at a height of 100 metres above the surface of the Earth, in metres per second. Care should be taken when comparing model parameters with observations, because observations are often local to a particular point in space and time, rather than representing averages over a model grid box. This parameter can be combined with the northward component to give the speed and direction of the horizontal 100 m wind.
100 m v-component of wind m s-1 This parameter is the northward component of the 100 m wind. It is the horizontal speed of air moving towards the north, at a height of 100 metres above the surface of the Earth, in metres per second. Care should be taken when comparing model parameters with observations, because observations are often local to a particular point in space and time, rather than representing averages over a model grid box. This parameter can be combined with the eastward component to give the speed and direction of the horizontal 100 m wind.
10 m u-component of neutral wind m s-1 This parameter is the eastward component of the "neutral wind", at a height of 10 metres above the surface of the Earth. The neutral wind is calculated from the surface stress and the corresponding roughness length by assuming that the air is neutrally stratified. The neutral wind is slower than the actual wind in stable conditions, and faster in unstable conditions. The neutral wind is, by definition, in the direction of the surface stress. The size of the roughness length depends on land surface properties or the sea state.
10 m u-component of wind m s-1 This parameter is the eastward component of the 10m wind. It is the horizontal speed of air moving towards the east, at a height of ten metres above the surface of the Earth, in metres per second. Care should be taken when comparing this parameter with observations, because wind observations vary on small space and time scales and are affected by the local terrain, vegetation and buildings that are represented only on average in the ECMWF Integrated Forecasting System (IFS). This parameter can be combined with the V component of 10m wind to give the speed and direction of the horizontal 10m wind.
10 m v-component of neutral wind m s-1 This parameter is the northward component of the "neutral wind", at a height of 10 metres above the surface of the Earth. The neutral wind is calculated from the surface stress and the corresponding roughness length by assuming that the air is neutrally stratified. The neutral wind is slower than the actual wind in stable conditions, and faster in unstable conditions. The neutral wind is, by definition, in the direction of the surface stress. The size of the roughness length depends on land surface properties or the sea state.
10 m v-component of wind m s-1 This parameter is the northward component of the 10m wind. It is the horizontal speed of air moving towards the north, at a height of ten metres above the surface of the Earth, in metres per second. Care should be taken when comparing this parameter with observations, because wind observations vary on small space and time scales and are affected by the local terrain, vegetation and buildings that are represented only on average in the ECMWF Integrated Forecasting System (IFS). This parameter can be combined with the U component of 10m wind to give the speed and direction of the horizontal 10m wind.
10 m wind gust since previous post-processing m s-1 Maximum 3 second wind at 10 m height as defined by WMO. Parametrization represents turbulence only before 01/10/2008; thereafter effects of convection are included. The 3 s gust is computed every time step and the maximum is kept since the last post processing.
Instantaneous 10 m wind gust m s-1 This parameter is the maximum wind gust at the specified time, at a height of ten metres above the surface of the Earth. The WMO defines a wind gust as the maximum of the wind averaged over 3 second intervals. This duration is shorter than a model time step, and so the ECMWF Integrated Forecasting System (IFS) deduces the magnitude of a gust within each time step from the time-step-averaged surface stress, surface friction, wind shear and stability. Care should be taken when comparing model parameters with observations, because observations are often local to a particular point in space and time, rather than representing averages over a model grid box.

Hourly and monthly subsets are available at the links below:

- hourly data on single levels from 1950 to 1978,

- hourly data on single levels from 1979 to present,

- monthly averaged data on single levels from 1950 to 1978,

- monthly averaged data on single levels from 1979 to present,

The monthly data is pre-calculated as monthly-mean averages from hourly data.

Spatial coverage and resolution:

Wind in ERA5, like all ERA5 data, is a global dataset. The atmospheric data is available on a regular latitude-longitude grid at 0.25o x 0.25o resolution (converted from native reduced‐Gaussian grid resolution of approximately 31 km x 31 km), and on 37 pressure levels.

Temporal coverage and resolution:

Wind in ERA5, like all ERA5 data, is available from 1950 to present (split into two entries: primary from 1979 onwards and a back extension from 1950-1978). The back extension is a preliminary version that has been released in 2020, and an updated version (that corrects some issues in the tropics) will appear around the end of 2021.

The data is available as hourly and monthly data.

Wind in ERA5, like all ERA5 data, is updated daily with a latency of about 5 days in an early product and with a final release 2 to 3 months later.

Information about observations (number, homogeneity)

ERA5’s data assimilation uses observations for all geophysical quantities from about 0.75 million observations per day in 1979 and about 24 Million in 2018. The 2D-OI uses surface observations at 'screen level'. The online technical documentation provides tables with the satellite and in-situ observations used as input into ERA5.

The satellite measurements used in ERA5 are: temperature, humidity, ozone, column water vapour, cloud liquid water, precipitation, ocean surface wind speed, wind vector, soil moisture, wave height.

The in-situ data is provided by WMO WIS and consists in measurements for: surface pressure, temperature, humidity, wind, wind profiles and snow depth. Figure 4 from Hersbach et al. (2020) presents the conventional observations assimilated per day in ERA5 during the period 1979–2018.

ERA5 assimilates rain rates from ground-based radar–gauge composite observations from 2009, and snow cover (NH only) from NOAA/NESDIS IMS.

The time evolving nature of the assimilated observations means that caution should be employed when using ERA5 to evaluate long-term variability and trends.

Methodology

Like any other climate variable from a reanalysis product, ERA5 wind is strongly influenced by the data assimilation methodology. ERA5 is produced using 4D-Var data assimilation with the ECMWF’s Integrated Forecast System (IFS) model (CY41R2). The forecast model has 137 hybrid sigma/pressure (model) levels in the vertical, with the top level at 0.01 hPa. The IFS is coupled to a land-surface model and an ocean wave model. The model uses as boundary conditions the sea surface temperature, the sea ice cover, the greenhouse gases, the aerosols, and the total solar irradiance. Climate variables are offered from the atmospheric model, the surface model and the wave model.

The ERA5 dataset contains one (31 km) high resolution realization (HRES) and a reduced resolution 10-member ensemble (EDA). The model time step is 12 minutes for the HRES and 20 minutes for the Ensemble Data Assimilation (EDA), though occasionally these numbers are adjusted to cope with instabilities. Climate variables result from analyses and short (18 hours) forecasts, initialized twice daily from analyses at 06 and 18 UTC. Most of climate variables from the analyses are also available from the forecasts. However, there are several climate variables from forecast, e.g., mean rates and accumulations, that are not available from the analyses. More information on the differences between analysis, forecast, instantaneous, accumulated and mean rate parameters are provided on https://confluence.ecmwf.int/pages/viewpage.action?pageId=85402030.

The ERA5 atmospheric analysis is based on a hybrid incremental 4-dimensional variational data assimilation (4D-Var) system including variational bias correction (VarBias). The method finds the best estimate of the state of the atmosphere/land/surface ocean within an assimilation time window, given a background forecast valid at the start of the window and observations falling within that window. The 4D-Var data assimilation uses 12 hour windows from 09 UTC to 21 UTC and 21 UTC to 09 UTC (the following day).

Uncertainty estimate: An uncertainty estimate is sampled by a 10-member lower-resolution Ensemble Data Assimilation (EDA) which provides background-error estimates for the deterministic HRES 4D-Var Data Assimilation system. The analysis method is the same for each EDA member and follows that of the HRES. Each member (except the control) is run with different random perturbations added to the observations. Likewise, the model physical tendencies are perturbed in the short forecasts that link subsequent analysis windows. Ensemble mean and spread have been pre-computed for convenience. Such uncertainty estimates are closely related to the information content of the available observing system which has evolved considerably over time. They also indicate flow-dependent sensitive areas. To facilitate many climate applications, monthly-mean averages have been pre-calculated too, though monthly means are not available for the ensemble mean and spread.

A strength of reanalysis (including ERA5) is the use of a consistent assimilation/forecast methodology throughout the analysis cycle. Thus, even though the observations assimilated are evolving in time (see above), the data assimilation approach can be considered fixed throughout the products analysis period, which adds to the homogeneity of the dataset.

Information about the technical and scientific quality

ERA5 wind represents one of the products of the latest global atmospheric reanalysis produced by Copernicus Climate Change Service at ECMWF. It is archived at a shorter (hourly) time step, has a finer spatial resolution, uses a more advanced assimilation system and includes more sources of data than previous versions. It is accompanied by extensive technical documentation and two principal scientific documentation papers. A list of ‘known issues’ is maintained at the online documentation (https://confluence.ecmwf.int/display/CKB/ERA5%3A+data+documentation#ERA5:datadocumentation-Knownissues).

A prerelease quality control revealed some problems affecting the performance in the tropics (tropical cyclones are too intense) and that the deep soil moisture tends to be too dry for the 1950-1978 dataset. A new version of the data should gradually become available by late 2021. This issue will be of little direct relevance to ERA5 wind in Canada’s north, but the user should be aware of the reason for this update.

Information on model improvement: The forecast model of the ERA5 is the IFS Cycle 41r2. In the ten-year period between ERA-Interim (Cy31r2) and ERA5 (Cy41r2), many significant improvements have been made to the representation of atmospheric physical processes (see Section 4 of Hersbach et al. (2020)). ERA5 wind will be influenced by several changes to ERA5’s land-surface model and parameterization schemes. ERA5’s HTESSEL land surface scheme (Balsamo et al., 2015) accounts for seasonally varying monthly vegetation maps specified from a MODIS-based satellite dataset. In addition, an enhanced snowpack parameterization allows a more realistic timing of runoff and terrestrial water storage variations and a better match of the albedo to satellite products. The chosen parameterization for lakes (FLake), allows consideration of both subgrid and resolved water bodies, which is potentially relevant for the lake-enriched Canadian sub-Arctic. This series of changes contributes to significant improvements in the soil moisture and land surface fluxes consistency, which allowed for the usage of satellite data in ERA5 to analyze soil moisture. This will influence the surface energy budget. Some important improvements in the wave model include: an updated model bathymetry with a more recent version of ETOPO2 and a revised unresolved bathymetry scheme. Some of these changes will also affect coastal regions as well as better accounting for wave propagation along coastlines and better modelling of the impact of previously unresolved features like islands and narrow embayments.

Limitations and strengths for application in North Canada

ERA5 is a new atmospheric reanalysis and there are not available scientific evaluations of the dataset dedicated specifically to northern Canada. However, it should be noted that in northern Canada, there are currently no sub-daily records over a long historical period for many weather stations. Reanalyses data with hourly output cover this gap and, with suitable investigation and calibration (downscaling), could be valuable resources.

As for all gridded data, care should be taken when comparing 10m wind with observations, because wind observations vary on small space and time scales and are affected by the local terrain, vegetation and buildings that are represented only on average in the reanalyses.

Also, as mentioned above, changes in the amounts and types of observational data that are assimilated may produce artificial trends or variability in reanalysis variables. For ERA5 this has been observed for wind in the boundary layer (Hersbach et al., 2020).

Up to once or twice per year, the analyzed near-surface (e.g., 10 m) winds in ERA5 suffer from a problem of extremely large wind speeds; the largest speeds seen so far are of order 300 ms −1.

References to documents describing the methodology or/and the dataset

Hersbach, H., B. Bell, P. Berrisford, G. Biavati, A. Horányi, J. Muñoz Sabater, J. Nicolas, C. Peubey, R. Radu, I. Rozum, D. Schepers, A. Simmons, C. Soci, D. Dee, and J-N. Thépaut, 2018: ERA5 hourly data on single levels from 1979 to present. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). (Accessed on < 29-Apr-2019 >), https://doi.org/10.24381/cds.adbb2d47

Hersbach, H., B .Bell, P. Berrisford, S. Hirahara, A. Horányi, J. Muñoz‐Sabater, J. Nicolas, C. Peubey, R. Radu, D. Schepers, A. Simmons, C. Soci, S. Abdalla, X. Abellan, G. Balsamo, P. Bechtold, G. Biavati, J. Bidlot, M. Bonavita, G. Chiara, P. Dahlgren, D. Dee, M. Diamantakis, R. Dragani, J. Flemming, R. Forbes, M. Fuentes, A. Geer, L. Haimberger, S. Healy, R.J. Hogan, E. Hólm, M. Janisková, S. Keeley, P. Laloyaux, P. Lopez, C. Lupu, G. Radnoti, P. Rosnay, I. Rozum, F. Vamborg, S. Villaume, and J.-N. Thépaut, 2020: The ERA5 global reanalysis. Quarterly Journal of the Royal Meteorological Society, 146(730), 1999–2049. https://doi.org/10.1002/qj.3803.

Online technical documentation: https://confluence.ecmwf.int/display/CKB/ERA5%3A+data+documentation

Link to download the data and format of data:

Data Access: Copernicus | NCAR | ECMWF

ERA5 is available in GRIB1 and NetCDF formats

Link to download hourly and monthly data on Copernicus:

- hourly data on single levels from 1950 to 1978,

- hourly data on single levels from 1979 to present,

- hourly data on pressure levels from 1950 to 1978,

- hourly data on pressure levels from 1979 to present,

- monthly averaged data on single levels from 1950 to 1978,

- monthly averaged data on single levels from 1979 to present,

- monthly averaged data on pressure levels from 1950 to 1978,

- monthly averaged data on pressure levels from 1979 to present.

Publications including dataset evaluation or comparison with other data in Canada

Tarek, M., F.P. Brissette, and R. Arsenault, 2020: Evaluation of the ERA5 reanalysis as a potential reference dataset for hydrological modelling over North America. Hydrology and Earth System Sciences, 24(5), 2527-2544. (It compares ERA5 and ERA-Interim with stations in US and Canada south of 60ᵒ latitude).

Sheridan, S.C., C.C. Lee, and E.T. Smith, 2019: A comparison between station observations and reanalysis data in the identification of extreme temperature events. Geophysical Research Letters, 47(15), e2020GL088120. (It compares observations, ERA5, ERA5-LAND, and NARR, in the United States and Canada- only a very small number of stations are in North Canada).

Betts, A.K., D.Z. Chan, and R.L. Desjardins, 2019: Near-surface biases in ERA5 over the Canadian Prairies. Frontiers in Environmental Science, 7 (129). (ERA5 is compared with hourly data for 4 stations in Saskatchewan, Canada).

Cao, B., X. Quan, N. Brown, E. Stewart-Jone, and S. Gruber, 2019: GlobSim (v1.0): deriving meteorological time series for point locations from multiple global reanalyses. Geosci. Model Dev., 12, 4661–4679, https://doi.org/10.5194/gmd-12-4661-2019 (2 m temperature from ERA5 is compared with ERA-Interim, JRA-55 and MERRA-2 at a site located near the north shore of Lac de Gras in the Northwest Territories, Canada)