Global Land Cover Mapping Looking Back Looking Forward

Global Land Cover Mapping Looking Back, Looking Forward Alan Belward Head of Land Resource Management Unit, European Commisison, Joint Research Centre Mexico City, 9 th November 2015

Photos A. S. Belward

A growing demand for global land cover information ① Earth system energy, water and material transport studies ② General Circulation Models ③ Biological process models ④ Land surface process models ⑤ MEA negotiations (UNFCCC, CBD, UNCCD, RAMSAR. . . ) ⑥ MEA monitoring, reporting, verification ⑦ SDG goal setting and progress indicators ⑧ Official Development Assistance programmes ⑨ Humanitarian and disaster reduction and relief operations ⑩. . . Extract GLC 2000 Africa Mayaux et al. JRC

. . . and therefore new constraints, new products ① Legally binding contexts increase requirements for full accountability, and accurate (documented) products ② Reliable on-demand product availability and delivery ③ A growing list of global land cover products; thematic maps for inventory and area estimates; biophysical products and direct parameterisation for use in models; measurement and monitoring of cover change; targeted classes (e. g. deforestation) for compliance verification ④ A growing range of time demands – seasonal, annual, decadal and spatial resolutions L to R: UMD, IGBP, MODIS, GLC 2000

Daily 2400 km 350 km 35 km Annually

The lifespan of all near-polar orbiting, land imaging, civilian satellites shown in chronological order Source JRC Belward, A. S. , & Skøien, J. O. (2015) http: //dx. doi. org/10. 1016/j. isprsjprs. 2014. 03. 009

36 Sovereign States flying Earth Observation Satellites with global land observing capabilities Source JRC Belward, A. S. , & Skøien, J. O. (2015) http: //dx. doi. org/10. 1016/j. isprsjprs. 2014. 03. 009

Two decades of global land cover mapping: from the 8 km AVHRR datasets of 1990 s to the 30 m ETM+ of the 2000 s Finer spatial resolution (but not necessarily finer temporal resolution) Map fusion (but not necessarily data fusion)

Source: Tsendbazar, N. E. , Mora, B. 100 160, 000 1 13, 20 Cl 2, 13 Cl 80 5, 17 Cl 70 2 3, 17 Cl 60 Accuracy (%) 7, 11 Cl 8, 22 Cl 17, 10 Cl 11, 20 Cl 16, 10 Cl 15, 11 Cl 9, 17 Cl 14, 22 Cl 10, 22 Cl 6, 22 Cl 16, 000 12, 10 Cl 4, 11 Cl 50 1, 600 3 40 4 5 7 11 15 6 9 8 30 13 10 17 14 160 20 10 12 0 1989 1994 1999 2004 Accuracy % Spatial resolution Legend: 1. AVHRR; 2. AVHRR; 3. IGBP; 4. UMD*; 5. MODIS 4; 6. GLC 2000; 7. SYNMAP*; 8. Globcover; 9. MODIS 5; 10. Globcover v 2; 11. GLCNMO 1; 12. FROM-GLC; 13. GLCNMO 2; 14. Land cover CCI; 15. GLCshare; 16. Globeland 30; 17. IIASA hybrid GLC. 2009 2014 16 16 2019 Spatial resolution (m) 90

FAO’s Land Cover Classification System; Setting standards, and a basis for regional classifications

Pesaresi, M. ; et al. , "A Global Human Settlement Layer From Optical HR/VHR RS Data: Concept and First Results doi: 10. 1109/JSTARS. 2013. 227144

Landsat GHSL Source M. Pesaresi (GHSL)

30 years' global surface water occurrence Source Joint Research Centre (JRC) and Google Earth Engine (GEE) 2015

0 occurrence as % of valid 100 obs 10 km Dubai’s new coastline Source: Global surface water occurrence, JRC/GEE

The views expressed in this presentation are purely those of the author and may not in any circumstances be regarded as stating an official position of the European Commission Conclusions ① Progress is very evident: Finer spatial resolution, improved accuracies, more detailed legends. . . ② But, one map won’t serve all users; this restricts use to specific modelling communities, compromises regional and national relevance, limits value for resource management, limits flexibility as a source of reference data for MEAs – can we identify converging user requirements though? ③ We still need to invest more in fusing data from multiple sensors ④ We still need product and processing standards – do we need harmonisation? ⑤ We still need to ensure data and products are available and accessible ⑥ We still need to establish operational infrastructures with clearly identified functions

Thank you! 17