1. Copernicus for nature policy – recent work undertaken by EEA & ETC/BD
Based on work prepared by ETCBD partner S4E: Stefan Kleeschulte, Karl Ruf
Sophie Condé 22th EGR meeting Brussels

2. What is COPERNICUS?
Description of the Copernicus components and swot analysis made by ETCBD
ETCBD analysis related to Natura 2000 & Copernicus

3. What is Copernicus ?
Copernicus is the European Union's Earth Observation Programme, looking at our planet and its environment for the ultimate benefit of all European citizens. It offers information services based on satellite Earth Observation and in situ (non-space) data
Copernicus services address six main thematic areas:
Atmosphere
Marine Environment
Land
Security
Climate Change
Emergency Management

4. Copernicus land monitoring service
The Copernicus land monitoring service provides geographical information on land cover and on variables related, for instance, to the vegetation state or the water cycle
It supports applications in a variety of domains such as spatial planning, forest management, water management, agriculture and food security, etc.
 The service became operational in 2012.
 It consists of three main components:
 A global component
A Pan-European component
A local component
The global component is coordinated by the European Commission DG Joint Research Centre (JRC). It produces data across a wide range of biophysical variables at a global scale (i.e. worldwide), which describe the state of vegetation (e.g. leaf area index, fraction of green vegetation cover, vegetation condition index), the energy budget (e.g. albedo, land surface temperature, top of canopy reflectance) and the water cycle (e.g. soil water index, water bodies).

5. Copernicus Pan-European component
The Pan-European component is coordinated by the European Environment Agency (EEA)
Produce 5 high resolution data layers (HRL) describing the main land cover types:
imperviousness (artificial surfaces)
forest,
grasslands,
wetlands (water & wetness)
small water bodies
The aim of these layers is to provide more frequent updates on land cover and to serve as an early warning system for land cover changes
All HRLs are produced from 20 m resolution satellite imagery and then aggregated to 1 ha grid cells
The pan-European wall-to-wall products cover the EEA39 countries
The High Resolution Layers are designed as mono-thematic raster data sets (imperviousness, forest, grassland, water & wetness), which are to be derived from satellite data in a highly automated process.
The aim of these layers is to provide more frequent updates on land cover and to serve as an early warning system for land cover changes.

6. Copernicus local component
The local component is coordinated by the EEA
Aims to provide specific and more detailed information complementary to the information obtained through the Pan-European component
Focused on areas that are specifically sensitive with environmental challenges and problems
Do not cover the complete territory but only selected areas of interest, with a specific challenge in terms of pressures or spatial management
Provide a different level of spatial and thematic detail, going down from 1ha to 0.25 ha MMU based on very high resolution imagery (2.5 x 2.5 m pixels)
Urban Atlas: provides detailed LC & LU information
Riparian Zones: provide biodiversity in areas around rivers,
Natura 2000 sites rich in grassland
The local component of Copernicus focuses on areas that are specifically sensitive and prone to environmental challenges and problems. Contrary to the products of the pan-European component, the local component products do not cover the complete territory in a wall-to-wall manner, but only selected areas of interest, with a specific challenge in terms of either environmental pressures and problems or spatial management. Examples include: cities above a given number of inhabitants, protected areas under the Natura2000 instrument, riparian zones along the hydrographic network or coastal zones etc….
As opposed to the pan-European component, which for national, regional or local applications is hampered by a relatively coarse spatial resolution (CLC 25ha MMU, HRLs 1ha grid), the local component provides a different level of spatial and thematic detail, going down from 1ha to 0.25 ha MMU. However, the offset is that these products are only mapped for specific areas of interest. The products are based on very high resolution imagery (2.5 x 2.5 m pixels).

7. General info on Copernicus
Copernicus land service pan-European & local components
Sealed soil
Corine Land Cover 2012
Forest type
Tree cover density
Grassland
Urban Atlas
Wetlands
Water bodies

8. Description of the Pan-European & local components
Layer name
Year of production
Data format
Coverage
Pan-European
HRL natural and semi-natural grasslands
2012
Raster
EEA 39
HRL forest type / density
HRL wetlands
HRL permanent water bodies
HRL imperviousness
2006, 2009, 2012
Local
Urban atlas
2006, 2012
Vector
Thematic
Riparian zones
Green Linear Elements*
Natura 2000 rich in grassland

9. HRL natural and semi-natural grasslands
Pan European - HRL natural and semi-natural grasslands
HRL natural and semi-natural grasslands
The 2012 product aimed at mapping natural & semi-natural grasslands as opposed to cultivated / managed grasslands used for grazing or hay production :
This was based on absence of signs of management, non-regular shapes of the patches, absence of artificially limited parcels and location in remote areas often on slopes or poor soils
Binary layer showing the presence / absence of natural grasslands
Due to the weakness of the 2012 product, a new product based on a grassland (grass and non-woody vegetation) mask with several supporting layers: a ploughing indicator to support the separation of grasslands from croplands and a grassland probability layer
These two outputs will not be comparable

10. Pan European - HRL forest type / density
The forest HRL actually consists of two individual products: Tree Cover Density (TCD) and Forest Type (i.e. coniferous / broadleaved/mixed)
The tree cover density product maps the level of tree cover density in a range from 0-100%
The forest type product comes as close as possible to the FAO forest definition

11. Pan European - HRL wetlands / water & wetness
Misunderstanding with the name of the 2012 layer: as the layer does not map “wetlands” in the ecological understanding of the idea, but rather different degrees of wet lands (i.e. wetness of the surface) and their permanence
Updated specifications of the 2015 HRL on “water & wetness” : to identify areas that are characterised by differently high degrees of soil water content and open water over different times during the year
Eventually this layer could be used to support the identification of (ecological) wetlands together with other ancillary information

12. Pan European - HRL permanent water bodies
This layer maps permanent lakes and ponds, rivers and coastal water surfaces, such as lagoons and estuaries
It does not include the sea and ocean, nor liquid dumpsites
Use seasonal time series of medium resolution images to determine the “permanence” of the water bodies
The final product is a binary layer showing the presence / absence of permanent water bodies

13. Local Component - Riparian zones
Areas occurring between land and freshwater ecosystems, characterised by distinctive hydrology, soil and biotic conditions and strongly influenced by stream water
Providing a wide range of riparian functions (e.g. chemical filtration, flood control, bank stabilization, aquatic life and riparian wildlife support, etc.) and ecosystem services
The Riparian LC/LU product is providing a detailed LC/LU dataset for areas along a buffer zone of selected rivers covering EEA39
Two complementary products: Delineation of the zones, Green Linear Elements

14. Local Component - Natura 2000 – grasslands
Detailed LC/LU map for a selection of 750 Natura2000 sites and a surrounding 2 km buffer zone
Initial exercise with a selection of five semi-natural & species rich grassland habitat types: Semi natural dry grasslands and scrublands facies on calcareous substrates 6240 Sub-pannonic steppic grasslands 6250 Pannonic steppic grasslands 6510 Lowland hay meadows 6520 Mountain hay meadows
The mapping was conducted for 2006 & 2012 and change analysis is available
The full dataset of N2K sites rich in grassland will be available in 2019

15. Description of the Pan-European & local components
Layer name
Positive
Negative
HRL natural & semi- natural grasslands
Mapping of all grass-covered areas
Based on time series to improve separation from arable land (ploughing indicator)
No direct information on management intensity
No information on grassland density
HRL forest type / density
High reliability
Comparability of different forest definitions (based on density)
No time series available yet
No species information (coniferous and deciduous only)
No information on forest condition
HRL water & wetness (wetlands)
Joint production of water and wetness product
Wetness is not directly visible in any EO data
Mapping of water / wetness under vegetation only if detectable
HRL permanent water bodies
Lower reliability at coastal inlets (transition between fresh & saltwater)
Issue with temporarily flooded riverbeds
Large variation in quality/reliability between different MS
Riparian zones
Very high thematic resolution
Spatially restricted to riparian areas (complex technical definition)
Natura (grassland rich sites)
Land cover product specifically targeting N2K sites
Currently only available for a small sample of grassland rich sites
For all PEC: (+) spatial resolution and pan european coverage

16. ETCBD analysis related to Natura 2000 & Copernicus
Spatial overlaps between N2000 and Copernicus
First step towards elaborating applications for Copernicus in the monitoring of habitats was to establish how much surface area of Natura 2000 actually overlaps with Copernicus products:
For Riparian Zones the coverage is about a mean of 13% at EU level
The calculated overlaps for HRLs indicate that most of Natura 2000 areas are covered by forested area (EU 46 %)
The remaining HRLs (Grassland, Water and Wetlands) all show vastly smaller overlaps ranging between 3-6% on average per country (EU 4%).
As expected, there is generally only a very small amount of sealed area detected within protected sites (EU 0,5 %)

17. ETCBD analysis related to Natura 2000 & Copernicus
Suitability of Copernicus layers for potential applications

18. ETCBD analysis related to Natura 2000 & Copernicus
Changes within the N2K sites directly & within the associated 2km buffer zone
A selection of N2K grassland-rich sites including a 2km buffer covering approx km2
The sites were analysed for the 2006 and 2012 reference years and a change analysis was performed
Conclusions on the quality and degree of impact within sites only drawn from land-cover based approaches ( no information as species composition, abiotic factors)
Time constraint to use the N2K database which could provide additional information on habitat types and local pressures
Methodological exercice and results based on a limited number of sites !

19. ETCBD analysis related to Natura 2000 & Copernicus
Count of analysed sites per biogeographic region
Due to splitting of sites located along region borders the actual count of sites is higher than specified within the N2K grassland mapping.

20. ETCBD analysis related to Natura 2000 & Copernicus
Changes inside and outside of N2K sites between 2006 and 2012 in ha and percent at MAES 1 level
Grassland areas showed a decline in the period between 2006 and 2012 both inside & outside
The decrease outside was roughly five times as high as inside

21. ETCBD analysis related to Natura 2000 & Copernicus
Grassland loss and gain in period by biogeographic region
Within the change period more than ha were converted to other land uses
Losses were considerably higher outside of sites (except BLS & BOR)
The most severe losses were located within buffer zones of the ATL which features large amounts of pastures
Inside BOR sites, mainly due to afforestation
BLS due to sample size effect (n=4)

22. ETCBD analysis related to Natura 2000 & Copernicus
Land cover change groups calculated as proportion of total land change per biogeographic region
Outside, the largest environmental threat is urbanisation
Inside Agricultural intensification and urbanisation are the dominant pressures
Extensification processes occur at consistently higher rates inside protected areas over all regions
In boreal region afforestation takes place at extremely elevated rates inside sites
How to read : Change groups affiliated with land-use intensity increase located on the left side, those which indicate land-use intensity decrease located on the right

23. Conclusions
Copernicus products are mainly suitable to refine a range of different analyses with a better spatial resolution and a more frequent updating as compares to the traditional CLC updates from 6 to 3 years
Applications, such as the combination of HRLs with Art. 17 data, will require more extensive testing and development
When using remote sensing products as well as European databases for ecological assessments, important to bear in mind the limitations:
spatial scale of the data sets
remote sensing data can provide information on quantity (e.g. area change) rather than providing information on quality (e.g. how many species are present)
Ideas for the future:
Incorporating additional datasets and information from the N2K database
Evaluation of the relationship between the percentage of grassland changes inside N2K sites and their percentage changes in the surrounding buffers

24. Thank you for your attention !
And some links…