New Forest Land Classification | News | Rezatec

New Forest Land Classification Data Set Launched


New Forest Land Classification Data Set Launched

British satellite imaging company DMC International Imaging Ltd (DMCii) today announced the completion of its Flagship project to develop a global system using Earth Observation (EO) satellite data to measure land carbon storage and how it changes over time.

The project, supported by Innovate UK (formerly known as the Technology Strategy Board) was developed with consortium partners Rezatec (Landscape Intelligence data services provider) and University College London (UCL), world-renowned remote sensing and carbon sequestration researchers.

The consortium was able to develop and deliver a unique approach to assimilating and transforming EO data from different sources and resolutions to calculate tropical forest carbon stock worldwide and provide a platform for carbon fluctuation modelling.
The project developed an online model representation of the tropical forest land class on a global scale.

The model uses Enhanced Vegetation Index (EVI) and Normalised Difference Vegetation Index (NDVI) outputs, from the NASA MODIS (Moderate Resolution Imaging Spectroradiometer) instrument, combined with ground data to emulate contemporary forest land classification distribution across the tropical portions of the globe.

The model was designed to form the baseline for monitoring trends in forest cover and associated carbon stock quantification over time. The model software environment has been developed to assimilate ground data from multiple sources so that carbon stock calculations for a given area of interest can be further trained for enhanced local accuracy using minimal ground plots.

The challenge the project sought to overcome was in reducing the high levels of error and uncertainty inherent in using coarse resolution EVI/NDVI inputs to drive quantitative assessments of carbon stock.

Using a combination of highly optimised statistical processing algorithms developed by Rezatec on the CEMS (Climate, Environment and Monitoring from Space) facility at the Satellite Applications Catapult in Harwell and EO data modelling approaches developed by UCL and DMCii, error and uncertainty in this area has been substantially reduced.

Accurate carbon stock measurement is critical to effective landscape management in the bio-fuels, agriculture and forestry sectors. Using the model for online processing of user ground data can significantly lower the costs of carbon stock measurements and overall landscape monitoring. This is of particular economic benefit for use in supporting auditing mechanisms such as MRV (Monitoring, Reporting and Verification) where physical audit costs are high relative to the tradable value of the underlying asset.

DMCii, focused on the data transformation element of the project, developing an EO processing system to produce high resolution surface reflectance data supported by a data and metadata repository which interfaced through an API to the main platform.

Rezatec was responsible for the construction of a global tropical forest carbon stock model using surface reflectance satellite data at varying resolutions as a key input for processing alongside other data sets such as digital elevation model outputs and biomass data
UCL focused on the scientific analysis of the carbon data at both the model level and the data sources used as inputs to the model to quantify the uncertainties involved and supply users with valuable quality assurance information.

DMCii Managing Director, Dave Hodgson, says: “We are committed to enabling better monitoring of global change from space. Together with a great team we’ve made big steps in pushing forward real products that can be applied to monitoring and measuring land carbon with commercial and national satellites.”

Patrick Newton, Chief Executive Officer of Rezatec, commented, “We are very pleased to have been invited to participate in this highly innovative project. The carbon stock data we have developed as a result of completing this initiative represent key components in our overall library of landscape intelligence data products and have use across all the sectors in which we operate.”

Professor Mark Maslin of University College London, concluded, “This project has allowed us to develop an accurate and cost effective means of annually monitoring tropical forest carbon storage and fluctuation. This will not only stimulate the global market in land carbon credits but will provide a means of measuring our effectiveness in protecting existing forest and reforestation. Both of which are essential if we are to prevent environmental degradation and reduce the effects of climate change.”

See how we can help

Take a look at geospatial AI in action. Explore how satellite data can provide a view of your entire asset base across vast areas. Discover how analytics drill down to a level of detail you never knew existed.