Skip to content
Example images of EO based datasets used in the classification

Examples of datasets used in the project area classification (From left: Sentinel-2 with GEDI height samples, ALOS-2 PALSAR-2 annual mosaic and Colombian land cover map).

It’s time to step into the Colombian Amazon forest!

Perhaps one of the most exciting demonstrations to be conducted during the Forest Carcon Monitoring (FCM) continuation is the Colombian use case. The overall aim is to highlight the feasibility of the FCM methods to meet the requirements of the new Verra Verified Carbon Standard (VCS) for REDD+ projects The demonstration is conducted in the context of an ongoing REDD+ project, ‘Pueblos Indígenas Resguardando la Selva’ (VCS2297), developed by South Pole.

The key requirements in the Verra's REDD methodology (VM0048) include stratification of the project area and implementation of field sampling. In the future, the project’s forest reference emission levels need to be estimated using Verra-allocated activity data analysed by strata, including, for example, risk classes and forested wetlands. The project developer must create a stratification map of the area with known strata characteristics. To meet these requirements, we defined a two-step sampling approach with the following five key activities:

  1. Forest classification and design of a field sampling
  2. Field measurement campaign
  3. Visual interpretation plot sampling on NICFI Planet imagery
  4. Statistical estimation of forest classes and biomass using the information gathered in the previous steps
  5. Creation of benchmark biomass and stratification maps for project Activity Data (AD) monitoring

The progress has been so fast over the project's first three months that we are ready to start the field measurements. We first classified the project area's  forest/land cover into six classes (see below). This classification was based on a combination of Sentinel-2 and ALOS2-PALSAR-2, with several auxiliary information sources (for example, GEDI height data and several Colombian GIS datasets).

Classification map and subsets of the Colombia demo area

Project area classification map with six land cover classes. Subsets highlighting the details of the landscape.

Subsequently, a field sampling scheme was designed for the project area considering the land cover distribution, the difficult working conditions in the Colombian Amazon and the prerequisites for statistical reasoning. Around 50 field sample plots will be measured. The information gathered from this survey will allow us to plan the subsequent NICFI Planet visual interpretation. Together, these two samplings enable the creation of the project area stratification map and estimation of average biomass per strata.

It requires a major expedition to conduct a field measurement campaign in the pristine rainforests of the Colombian Amazon. Two field crews will work for around a month, based on riverboats. The field protocols follow the Colombian National Forest Inventory (NFI) sample plot and field working guide. The field crews, supported by local people, are about to start their work in the coming days. With great excitement, we will follow the progress of the expeditions and look forward to seeing the results of the field measurements in October.

Wishing the best of luck to the field crews!