Ground Data are Essential for Biomass Remote Sensing Missions

Several remote sensing missions will soon produce detailedcarbon maps over all terrestrial ecosystems. These missions are dependent on accurate and representative in situ datasets for the training of their algorithms and product validation. However, long-term ground-based forest-monitoring systems are limited, especially in the tropics, and to be useful for validation, such ground-based observation systems need to be regularly revisited andmaintained at least over the lifetime of the planned missions. Here we propose a strategy for a coordinated and global network of in situ data that would benefit biomass remote sensing missions. We propose to build upon existing networks of long-term tropical forest monitoring.To produce accurate ground-based biomass estimates, strict data quality must be guaranteed to users. It is more rewarding to invest ground resources at sites where there currently isassurance of a long-term commitment locally and where a core set of data is already available. We call these supersites'. Long-term funding for such an inter-agency endeavour remains an important challenge, and we here provide costing estimates to facilitate dialogue among stakeholders. One critical requirement is to ensure in situ data availability over the lifetime of remote sensing missions. To this end, consistent guidelines for supersite selection and management are proposed within the Forest Observation System, long-term funding should be assured, andprincipal investigators of the sites should be activelyinvolved.