Wednesday, January 12, 2011

Satellite Imagery for Forest Preservation Projects

Here at Global Green Carbon we are very excited about the potential of satellite data to monitor afforestation and avoided deforestation/degradation (REDD) projects. Satellite monitoring offers huge advantages compared with ground-based monitoring, by being able to cover the whole project area frequently and at a low cost.

Good quality satellite images of the Earth have been available since the early 1970s from the NASA’s Landsat satellite series [link http://landsat.gsfc.nasa.gov/]. These six satellites have provided high resolution images over the whole of the Earth’s surface several times per year, and since last year the whole database has been freely available [link http://glovis.usgs.gov/], providing an incredible resource. Up until now most tropical forestry projects have analysed Landsat (or similar) data themselves in order to create landcover maps for the present day and back into the past, as indeed we are in the process of doing for our projects.

However, such analyses have become under increased scrutiny with the advent of Reducing Emissions from Deforestation and Degradation (REDD) projects: because the payments for these forest-protection projects relies entirely on the calculation of a ‘baseline’ deforestation rate – the rate of carbon loss that would have happened in the absence of the project – accurately and conservatively calculating previous deforestation rates has become vital. Equally, COP16 at Cancun agreed in December that REDD+ is to be implemented by governments across the tropics [link http://unfccc.int/meetings/cop_16/items/5571.php], so independent estimates of deforestation and degradation over huge areas have become very important.

Monitoring at this scale requires the ability to analyse vast quantities of satellite data, so perhaps unsurprisingly Google.org [link www.google.org] (the not-for-profit arm of Google), in collaboration with GEO [link http://www.earthobservations.org], the Group on Earth Observations, a collaboration of all the major space agencies, is in the process of developing a real-time monitoring service for global deforestation [link http://blog.google.org/2009/12/seeing-forest-through-cloud.html]. While the final service has not yet been released, a step towards it occurred late last year when Google.org released the Google Earth Engine [link http://earthengine.googlelabs.com/#intro], an analysis toolbox with access to many terabytes of satellite imagery.

However, the Google system has some weaknesses from our point of view. It is entirely centred around deforestation, i.e. the conversion of forest into non-forest. However, many forestry and REDD projects rely on more complicated conversions than this, for example preventing degradation, the loss of just a portion of biomass from a forest or, to take an example from our projects, planting trees in degraded woodland or farmland. Equally there is no attempt within Google’s system to assess the biomass of forest, so the area of deforestation cannot easily be converted into tonnes of carbon lost.

An alternative service being developed that attempts to do this and more was recently launched by Ecometrica [link http://www.ecometrica.co.uk/], a UK-based company providing assessments of ecosystems and greenhouse gas emissions. This is called the Biocarbon Tracker [link http://tracker.biocarbontracker.com/interface], and currently provides a global map of aboveground biomass, and for the Amazon a unique ‘carbon risk’ map. In the near future they hope to include a real-time deforestation layer, allowing for emissions from deforestation within a particular area to be calculated.

As none of these systems are currently fully operational, we will continue to do our satellite analyses in-house. But we will continue to watch the development of these global monitoring solutions with interest.


No comments:

Post a Comment