Monitoring forest biomass is essential for understanding the global carbon cycle because:
- Forests account for around 45 % of terrestrial carbon, and deforestation accounts for 10% of greenhouse gas emissions
- Deforestation and forest degradation release approximately identical amounts of greenhouse gases as all the worldâ€™s road traffic
- Forests sequester significant amounts of carbon every year
The United Nations (UN) intergovernmental Reducing Emissions from Deforestation and forest Degradation in developing countries (REDD+) programme, was secured in 2013 during the 19th Conference of the Parties to the UN Framework Convention on Climate Change. It requires countries to map and monitor deforestation and forest degradation, together with developing a system of sustainable forest management. Remote sensing can play a great role in helping to deliver these requirements, and below are three fantastic remote sensing initiatives in this area.
Firstly, the Real Time System for Detection of Deforestation (DETER) gives monthly alerts on potential areas of deforestation within Amazon rainforests. It uses data from MODIS, at 250 m pixel resolution, within a semi-automated classification technique. A computer model detects changes in land use and cover such as forest clearing that are then validated by interpreters. It has been valuable helping Brazil to reduce deforestation rates by around 80% over the last decade; however, it takes two weeks to produce the output of this computer model.
A similar initiative is FORest Monitoring for Action (FORMA), which also use MODIS data. FORMA is fully automated computer model which combines vegetation reflectance data from MODIS, active fires from NASAâ€™s Fire Information for Resource Management and rainfall figures, to identify potential forest clearing. Like DETER it produces alerts twice a month, although it works on tropical humid forests worldwide.
A third initiative aims to provide faster alerts for deforestation using the research by Hansen et al, published in 2013. The researchers used successive passes of the current Landsat satellites to monitor land cover, and when gaps appear between these passes it is flagged. These will be displayed on an online map, and the alerts will be available through the Word Resources Instituteâ€™s Global Forest Watch website, starting in March 2016. With the 30 m resolution of Landsat, smaller scale changes in land use can be detected than is possible for sensors such as MODIS. Whilst this is hoped to help monitor deforestation, it doesnâ€™t actually determine it, as they could be other reasons for the tree loss and further investigation will be required. Being an optical mission, Landsat has problems seeing both through clouds and beneath the forestry canopy, and so itâ€™s use will be limited in areas such as tropical rain forests.
Finally, one way of combat the weather and satellite canopy issue is to use radar to assess forests, and the current AfriSAR project in Gabon is doing just that â€“ although itâ€™s with flights and Unmanned Aerial Vehicles (UAV) rather than satellites. It began in the 2015 with overflights during the dry season, and the recent flights in February 2016 captured the rainy season. This joint ESA, Gabonese Space Agency and Gabon Agency of National Parks initiative aims of the project is to determine the amount of biomass and carbon stored in forests, by using the unique sensitivity of P-band SAR, the lowest radar frequency used in remote sensing at 432â€“438 MHz. NASA joined the recent February missions adding its Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) and the Land, Vegetation and Ice Sensor (LVIS) instrument, which are prototypes of sensors to be used on future NASA missions. Overall, this is giving a unique dataset on the tropical forests.
These are just four example projects of how remote sensing can contribute towards towards understanding what is happening in the worldâ€™s forests.