A detailed analysis by an international team shows that recent global increases in terrestrial carbon stocks are largely stored in non-living parts of ecosystems, rather than in living biomass, as previously thought. This application is the subject of a recently published study, which was supplemented with a detailed method and supplementary materials, described in scientific document "The additional materials for recent gains in global carbon stocks on Earth are mostly stored in non-living reserves."

The researchers used comprehensive analyses of various data sources and models to examine how carbon has been stored over the past three decades (the 1990s, 2000s and 2010s). The results showed that while there has been a global increase in total carbon stock, the proportion of this increase stored in living terrestrial biomass is less than 50 %. Most of the gains remain stored in non-living stocks, such as carbon in soil, dead wood, sediments in river, lake and wetland systems, or long-term carbon stocks in wood products.

According to the study authors, the metric of greatest importance is living woody biomass, which accounts for more than 95% of all carbon stored in living organisms on land. Due to the difficulty of measurement, they also included the mass of soil fungi and bacteria in the non-living carbon groups. This approach provides a better balance of total modern global carbon, because changes in the microbiome are difficult to measure and relatively small.

The researchers combined various measurement techniques and approaches – from direct field inventory measurements, airborne laser scanning data, advanced satellite imaging and machine learning models. Satellite data included, for example, measurements using the MODIS optical sensor or the passive microwave sensors of the SMOS, AMSR-E or AMSR2 satellites. The specific vegetation optical depth data from the satellites (so-called L-VOD) were carefully calibrated and corrected to minimize problems with radio frequency interference.

The resulting data were then harmonized to account for differences in vegetation type coverage (forests and non-forest areas), to include belowground biomass, and to ensure both temporal and spatial scales. The analyses are performed within defined active regions (RECCAP) and include sensitive tests for the detectability of small increments in dense forest ecosystems.

Importantly, the research took into account the uncertainty associated with all approaches used and the results obtained were also confirmed by comparison with dynamic global vegetation models (DGVM models), which are independent of satellite observations.

The study authors also point to mechanisms that allow for the storage of carbon in non-living reservoirs, such as carbon stored as a result of soil erosion, changes in management, long-term carbon storage in wood products, and dead organic matter stored in wetlands and other water bodies.

Although the approach, the authors also highlight existing uncertainties, such as the possible inability of some inventors to detect smaller increments in dense intact forests, these aspects were also tested in detail and the study results show that the key conclusions of the study are robust and consistent with other studies.

This study requires a comprehensive view of the active carbon cycle and ensuring that the protection and stewardship of non-living carbon stocks is at least as important as the protection of living ecosystems. Detailed information and results of the supplementary analysis can be found in the comprehensive document "Supplementary Materials", which is publicly available and allows for a deeper understanding of this important issue determining the further development of climate policies. Spring

The work is published in the magazine Science .