Copernicus 2: The future of the Copernicus programme

Copernicus 2 Illustration (Source: https://marine.copernicus.eu/)

The iconic European Programme that provided scientists with the Sentinel missions has something else up its sleeve and this for the decade to come. This article will overview each of the currently studied missions and present which problem they try to address and how they fit in EU policy.

I. Copernicus — What’s that all about ?

The Copernicus Programme, co-managed by the European Commission and the European Space Agency, aims at providing high-quality remote sensing data, on a global scale, for all. This free-of-charge aspect of the data made the programme extremely popular among the community. The development of visualisation platforms such as EO Browser makes it even easier for everyone to manipulate Sentinel data.

Fig. 1 — EO Browser Interface. Visualisation of Sentinel-5P column average of dry air mixing ratio of Methane over Australia

The Copernicus missions are defined within a multi-annual financial framework of 7 years. The first cycle lasted from 2014 & 2020 and saw the planned launch of a LOT of satellites, as shown in figure 2.

Fig. 2 — Copernicus First Cycle Launches

Two satellites designed during Copernicus’ first cycle are planned to be launched in the coming years: Sentinel-5 (2021) & Sentinel-6B (2023).

Fig. 3 — Applicability of 1st cycle’s satellites

Each of these satellites had its own objective and can fit in one of these 3 applicative categories, as presented in figure 3: they can be used, either for land monitoring, ocean monitoring or atmospheric monitoring. No matter the target application, the Sentinels cover it with more than one satellite imagery.

Thus, Copernicus 2.0 appears in this context of vast range of Earth observation capacity offered by its predecessor.

II. Copernicus 2.0 — Seeing Earth in a new light

The same way as scientific priorities matured over the last 7 years, the objectives of the 2nd cycle of Copernicus missions involve a more explicit design of its missions toward solving matters such as Climate Change.

It is in this context that the 2021–2027 cycle involves the study of 6 “high-priority candidate missions” in order to “address EU policy and gaps in Copernicus user needs” (esa.int).

Fig. 4 — 2nd Cycle High-Priority Copernicus Candidate Missions (esa.int)

The 6 aforementioned missions, presented in figure 4, display a wide variety of sensor technologies, each designed for a specific task.

Fig6 — CHIME Mission Illustration (esa.int)

Developed with the perspective of supporting Copernicus Sentinel-2 for applications such as land-cover mapping, the CHIME mission is expected to provide routine hyperspectral observations. Primarily aimed at monitoring Agriculture & Raw Materials, the CHIME sensor is expected to help to manage natural resources alongside Sentinel-2 measurements.

For that matter, the CHIME sensor must have a spatial resolution between 20 & 30m that would then be resampled to Sentinel-2 20m grid for more straightforward multimodal computation.

Fig 5 — CHIME Spectral Range

Besides, the spectral range of the CHIME Mission is expected to cover from visible light (400nm) to shortwave infrared (2500nm), with a maximum spectral resolution of around 10nm, as presented in figure 5.

Fig 6 — CIMR Mission Illustration (esa.int)

A recent expansion of the Copernicus Programme includes new High Priority Copernicus Missions (HPCM) with the notable example of Priority 2, which concerns the monitoring of Polar regions, specifically of sea ice/floating ice concentration and surface elevation.

Thus, in this context, the aim of the CIMR mission is, formally speaking, to:

provide high-spatial resolution microwave imaging radiometry measurements and derived products with global coverage and sub-daily revisit in the polar regions and adjacent seas to address Copernicus user needs.(esa.int)

In details, this corresponds to 3 primary objectives, as presented at esa.int:

  • (PRI-OBJ-1) To measure Sea Ice Concentration (SIC) and Sea Ice Extent (SIE)
  • (PRI-OBJ-2) To measure Sea Surface Temperature (SST)
  • (PRI-OBJ-3) Ensure European operational continuity of L-band (e.g., SMOS/SMAP)

To meet these objectives, the CIMR mission consists of a global multi-frequency imaging microwave radiometry, focusing on high-latitude regions, in adequation with the aforementioned HPCM setting. The main difference between radiometry and RADAR is that radiometry carries out passive measurements.

The 5 allocated bands and centre frequencies of the CIMR mission are detailed in figure 7.

Fig 7 — CIMR Allocated bands & centre frequency

Thanks to all these characteristics and of its adequate spatial resolution (<5km), the CIMR mission will allow for the monitoring of floating sea ice surface parameters.

Fig 8 — CO2M Mission Illustration (esa.int)

Monitoring anthropogenic CO2 at a global scale can help assess the effectiveness of the Paris Agreement and COP21 decisions. For that matter, ESA and the European Commission jointly established a CO2 monitoring Task Force. Among the outputs of the Task Force is the creation of a CO2 monitoring system embedded within the Copernicus ecosystem: the Copernicus Anthropogenic Carbon Dioxide Monitoring Mission.

The objectives of this mission could be summed to the ability to measure CO2 with “the resolution, accuracy, time sampling and spatial coverage required” (esa.int) to conclude on the state of anthropogenic CO2 emissions at different scales (local, regional, national, continental & global). In more details, the mission objectives are:

  1. to detect emitting hot spots (megacities/power plants);
  2. to monitor and quantify hot spot emissions (for reductions/increases detection);
  3. to put side-by-side local emission changes against local reduction targets.

To monitor the amount of CO2 at various scale, the CO2M Mission will have a spatial resolution of around 4km², allowing it to have precise measures of cities emissions but also of emissions of a larger scale (as a matter of illustration, Paris area is around 105 km²).

Despite being a hyperspectral mission, the main objective of CO2M implies the use of a particular spectral range to retrieve XCO2 measures (XCO2 being a column-averaged measure of CO2 in the atmosphere). For that, 3 subranges, presented in figure 9, are introduced.

Fig 9 — XCO2 Product Spectral Range (esa.int)

For a more accurate retrieval of XCO2, the estimation of parameters interfering in the measures is critical. For that matter, CO2M also provides an estimation of other quantities such as Methane, Solar-induced Fluorescence of Vegetation, or Surface Pressure.

Fig 10— CRISTAL Mission Illustration (esa.int)

Coincidently with CIMR, the CRISTAL Mission is developed in the new HPMC regarding Polar Ice monitoring. However, while the CIMR Mission develops tools for monitoring sea ice surface parameters, the CRISTAL Mission focuses on sea ice thickness and land ice elevation measurements.

For that matter, CRISTAL would carry a high spatial resolution dual Ku/Ka-band SAR altimeter plus a passive microwave radiometer in an optimised polar orbit.

The two main objectives of the CRISTAL mission, as stated in esa.int, are:

  1. To measure and monitor the variability of Arctic and Southern Ocean sea-ice thickness and its snow depth. The focus is here on analysing the seasonal sea ice cycles that are meant to occur and to extract from it inter-annual variability of sea ice as an indicator of climate condition.
  2. To measure and monitor the surface elevation and changes in polar glaciers, ice caps, and the Antarctic and Greenland ice sheets. The monitoring of surface elevations and changes provides tools to quickly identify and track emerging instabilities.
Fig 11 - Effect of dual frequency altimetry (AltiKA & CryoSat-2) for sea ice thickness retrieval (Left: CryoSat-2 with Warren 99 climatology, Middle: CS-2 sea ice thickness using snow estimate from SARAL/AltiKA, Right: Difference in sea ice thickness measure between the two methods) (ESA Arctic+ Snow on Sea Ice, FMI/UCL)

The task of snow depth retrieval will be mostly carried out by the dual Ku/Ka-band SAR sensor as multiple studies have already shown the ability of dual-frequency rdar for such tasks (use of SARAL KA-band and CryoSat-2 Ku-band frequencies at crossing points). As seen in figure 11, snow cover over sea ice can lead to perturbated altimetry measures for CryoSat-2. Hence, once combined with the snow cover estimation from SARAL/AltiKA, we can correct the perturbation that led to estimation variations, in the worst cases, of almost 1 meter.

Fig 12— LSTM Mission Illustration (esa.int)

The Long Short Term Memory model is a… wait.. no that is the wrong LSTM. This LSTM is the 5th high-priority candidate of the Copernicus Programme, and it involves the observation of surface temperature at a high spatio-temporal resolution.

With the idea of complementing the existing Copernicus family of satellites, the LSTM Mission aims at having “direct synergies with Sentinel-1, Sentinel-2, Sentinel-3 missions” (esa.int). It comprises of two objectives:

  1. A primary objective of monitoring the evapotranspiration rate at the European field scale using Land Surface Temperature (LST) variations.
  2. A secondary objective to support the mapping of a wide range of other services would benefit from the high spatio-temporal resolution of LST.

The generation of LST mappings requires the use of a measure with a sensibility to ground temperature, which corresponds to Thermal Infra Red (TIR) bands (cf figure 12).

Fig 13 — Atmospheric transmittance as a function of the wavelength (Sabins et al, 1998)

Thermal infrared sensors, as shown in figure 13, are already on board multiple satellites, with the most notable being Landsat 7. However, with a resolution of 120m, it becomes harder to make conclusions on the scale of crops.

The resolution objective of the LSTM mission is to estimate evapotranspiration at field scale for parcel sizes between 0.5ha (~5000m²) and 1.0ha (~10000m²). Hence, the targeted Spatial Sampling Distance of LSTM’s Level-1c product lies is at 30m.

Fig 14 — ROSE-L Mission Illustration (esa.int)

Copernicus’ current SAR satellite currently in circulation, Sentinel-1, is a C-band radar mission. Its higher frequency than L-band SAR prevents it from penetrating many natural materials such as forest canopy, dry snow, or ice. Hence, standing as a complementary mission of Copernicus Sentinel-1, ROSE-L would enable retrieving information hidden under the materials that C-band SAR cannot penetrate.

Hence, this mission is, by design, a great opportunity in support of forest management or for crop types discrimination for agriculture. Besides, the ROSE-L mission may lend a hand to the CRISTAL & CIMR to monitor polar ice sheets & caps.

Officially, 6 main objectives for the ROSE-L sensor arise from its requirements document (esa.int):

  1. Monitoring European & Global geohazards: events related to land motion, such as Urban subsidence, or Landslides, are shown to be easier to monitor when working with L-Band SAR. It is thanks to its “interferometric information being more robust to phase unwrapping errors” (esa.int).
  2. Land use, land-use change, forestry & agriculture: as mentioned above, L-band SAR is a great candidate for monitoring land use and forestry thanks to its canopy penetration abilities. For the example of forestry, L-band SAR can be used to map forest/non-forest areas and broad forest types. Another use case is the repeat-pass interferometric acquisitions that were shown to be sensitive to forest height (Lavalle and Hensley, 2015). Other information regarding forest biomass can also be extracted from L-Band SAR.
  3. Soil Moisture: Compared to C-band SAR, L-band imagery has been shown to display “higher sensitivity to soil moisture for bare soils and medium vegetation cover densities” (El Hajj et al, 2018) while, for highest densities, illustrated by NDVI values above 0.7, “C-band data have no sensitivity at all, whilst L-band data continues to be sensitive to soil moisture” (esa.int).
  4. Cryosphere & Arctic Observations: an example of application of L-band SAR to Cryosphere & Arctic observations can be ice type mapping as L-band, and C-band SAR show high complementarity. Studies involving the use of ALOS data showed “the potential for improvements in ice type distinction when combining already available C-band data with L-band data” (esa.int).
  5. Marine & Maritime Surveillance: typical maritime applications of SAR data include oil spill monitoring, fisheries monitoring or simply maritime intelligence. However, no noticeable advantage of L-band SAR is found for now, and more investigations of C-band SAR capabilities for said applications need to be carried out. The only difference can be illustrated by the lower impact of wind field variability over the sea backscattering for L-band SAR than for C-band radar.

As illustrated by multiple points above, this new L-band satellite is meant to be used coincidently with existing C-band Sentinel-1 satellites. It also shows through the requirements for ROSE-L, which include a time interval between L-band and C-band acquisitions of 1 minute or less, for the maximisation of the complementarity between C- and L-band.

Fig 15 — Illustration of improved access through the combination of Sentinel-1 and ROSE-L missions (top) access based on Sentinel-1 c/d only and (bottom) access based on Sentinel-1 c/d + ROSE-L (assuming two spacecraft). Simulations performed by Michel Tossaint (ESA). (esa.int)

In addition, the ROSE Mission shall support resolutions equal to or higher than 50m², with repeat coverage of 1 day (Arctic), 3 days (Europe) and 6 days (Global). Combined with the future S1-C/D missions, ROSE-L A/B will significantly improve the revisit time of every part of the world, as shown in figure 15.

This high standard of spatio-temporal resolution follows the footsteps of Sentinel-1 and makes the promise for grandiose new discoveries regarding the combined potential of C- and L-band SAR imagery.

III. Conclusion

As a matter of conclusion, we can summarise this article as a sign of the bright future of the Copernicus Remote Sensing missions. The Copernicus Programme, too little acclaimed in my opinion, is a fantastic European success story. Its cost between 1998 & 2020 is estimated at €6.7 billion, €4.3 billion spent over its aforementioned first cycle, while it is estimated that it benefited the EU economy with roughly €30 billion through 2030. Apart from the purely financial aspect, the overwhelming scientific excitement generated by the exploitation of the Sentinels’ data and the spread of the open-data philosophy to the field of Remote Sensing make the Copernicus programme something that both ESA & the European Commission should be highly proud of.

IV. References

Oppenheimer, C. (1998). SABINS, F. F. 1997. Remote Sensing. Principles and Interpretation, 3rd ed. xiii 494 pp. New York: W. H. Freeman & Co. Price £32.95 (hard covers). ISBN 0 7167 2442 1. Geological Magazine, 135(1), 143–158. doi:10.1017/S0016756897318251

Lavalle M., and S. Hensley, “Extraction of Structural and Dynamic Properties of Forests From Polarimetric-Interferometric SAR Data Affected by Temporal Decorrelation,” in IEEE Transactions on Geoscience and Remote Sensing, vol. 53, no. 9, pp. 4752–4767, Sept. 2015.

El Hajj, Mohamad & Baghdadi, Nicolas & Bazzi, Hassan & Zribi, Mehrez. (2018). Penetration Analysis of SAR Signals in the C and L Bands for Wheat, Maize, and Grasslands. Remote Sensing. 11. 31. 10.3390/rs11010031.

As a French PhD student, I am passionate to whatever comes close to Artificial Intelligence and Earth Observation.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store