Back

At Space Intelligence we have 20 years of experience working in academic forest research, and specialising in the remote sensing of forest properties and processes. We leverage this experience with technology to provide high quality products and services, all underpinned by world-class research.

We understand that there are a range of products and services available to natural resource managers, and will guide our customers toward the best solution available on the market. We generate the most value by focusing on our core strengths, the automated analysis of satellite data using the latest algorithms and machine-learning technologies.

In particular we are concentrating on producing analyses of deforestation and forest degradation. The need  for this service is increasing as more companies make Zero Deforestation Commitments. Under these commitments, and in accordance with Sustainable Development Goal 15, companies are pledging to reduce the level of deforestation in their commodity supply chains to zero by 2020. Any such commitments need to be monitored to ensure that they are being met. However this represents a technical challenge. There are existing products which allow for deforestation monitoring at a global level. Pre-eminent among these is Global Forest Watch (GFW), whose data you can interact with here: https://www.globalforestwatch.org/.  This is ideal for first order analyses, drawing large scale observations and trends, and has revolutionised our view of forest monitoring: free data available to all at a global level.

Nevertheless, the data is not ideal for all situations, and we are able to differentiate ourselves on a number of different important points:

  • GFW have a significant time delay between  a forest loss event happening and publishing (~2 years). We can provide results with effectively no delay. (However, please note that for a limited amount of countries (22 countries and counting),  GFW provides GLAD tree cover loss alerts, which are provided weekly at 30×30 meter resolution).
  • Separately, GFW uses optical data so is limited by cloud cover. In much of the world, especially the most forested areas (e.g. tropics, northern latitudes), cloud free satellite images at the resolution GFW uses may only appear every few years, meaning the data may be more out of date still. If no image is available GFW still reports for that pixel, stating ‘no change’ – this may bias analyses against seeing recent upticks in deforestation until too late.
  • GFW sees only ‘forest loss’ – not forest degradation
  • ‘Forest loss’ also differs from true deforestation – as they do not have a land cover type map for the previous year so cannot compare the trees they see being removed to the relevant (e.g. national) forest definition)
  • GFW is a global product. It is a fantastic advance, and really useful for regional and global statistics – but inevitably it will be less accurate than a custom product produced for your area of interest.
  • As GFW is an optical product it just sees the top of the forest canopy. It is inevitable that it therefore sometimes is confused by tree crops and other landcover types. Our radar product can distinguish tree crops and forest types, therefore producing more accurate statistics.
  • GFW’s global 30 m product is limited to annual timesteps, January-December (~2 years out of date as stated). We can provide statistics at a much finer grain (e.g. monthly, quarterly), or annually with perhaps a more appropriate starting month (e.g. July-July for southern hemisphere).

If you’re interested in how forest monitoring products differ, we have published a blog post which starts to explore the differences between GFW and a 5 m resolution RapidEye validation data set tested in Brazil.

Please contact us if you would like to discuss monitoring are area of forest, or simply to learn more about what we can offer. We are happy to help.