Stock di Carbonio Attuale - Biomassa Sotterranea (BGB)

Land Usev0

Responsible:Niccolò Calandri

Current Carbon Stock — Below-Ground Biomass (BGB)

Overview

The Current Carbon Stock — Below-Ground Biomass (BGB) layer quantifies the carbon stored in root systems beneath the soil surface, expressed in tonnes of carbon per hectare (tC/ha). Below-ground biomass encompasses all living root tissue, including coarse roots (> 2 mm diameter), fine roots, rhizomes, and root-associated mycorrhizal networks. While less visible than above-ground biomass, BGB plays a crucial role in the global carbon cycle.

Root carbon typically constitutes 20-30% of total living biomass in temperate forests and up to 40-60% in grasslands and shrublands, where root allocation is proportionally higher. BGB is essential for accurate total biomass carbon accounting, as omitting roots can underestimate living biomass carbon by 20-40%. The dataset derives from Walker et al. (2022), with the BGB pool isolated and downscaled using Random Forest regression with satellite-derived features.

BGB is particularly important for:

Complete carbon inventories — required by IPCC Tier 2+ methodologies
Soil stability assessment — root systems prevent erosion and stabilize soil carbon
Ecosystem resilience — BGB enables vegetation recovery after disturbances
Root:shoot ratio analysis — comparing BGB to AGB indicates ecosystem health

Calculation Methodology

Random Forest downscaling pipeline for BGB extraction:

Baseline extraction — The Walker et al. 2022 BGB carbon density raster is loaded from S3 and clipped to the area of interest. The BGB pool is isolated from the total carbon map.
Feature engineering — For each pixel:
- NDVI from Sentinel-2 — correlates with above-ground vegetation which is allometrically related to root biomass
- Land cover class from Dynamic World — different vegetation types have characteristic root:shoot ratios
- Elevation from SRTM DEM — altitudinal gradients affect root allocation
Training — Random Forest regressor trained on baseline BGB values
Prediction — Model predicts BGB at ~200 m resolution
Validation — Cross-validation with R-squared and RMSE

Output unit: tC/ha (tonnes of carbon per hectare, BGB pool only)

Data Sources

Code	Name	Provider	Type	Coverage	Resolution	Reference Year
Walker 2022	Global Carbon Density (BGB)	Walker et al. / S3	Raster (COG)	Global	~300 m	2016
`COPERNICUS/S2_SR_HARMONIZED`	Sentinel-2 L2A	ESA/Copernicus	Multispectral	Global	10 m	2017 — present
`GOOGLE/DYNAMICWORLD/V1`	Dynamic World	Google/WRI	Land cover	Global	10 m	2015 — present
`USGS/SRTMGL1_003`	SRTM DEM	NASA/USGS	Elevation	60N-56S	30 m	2000

Quality Range and Levels

Indicator	Unit	Range	Inverted
`carbon_stock_current_bgb`	tC/ha	[0, 2, 8, 20, 40, 80]	No

Inverted = No: a higher value indicates greater below-ground carbon storage.

Level	Thresholds	Interpretation
E (Very Low)	< 2 tC/ha	Bare soil, urban, desert — negligible root biomass
D (Low)	2-8 tC/ha	Sparse grasslands, annual croplands with shallow roots
C (Moderate)	8-20 tC/ha	Perennial grasslands, young forests, agroforestry
B (High)	20-40 tC/ha	Mature temperate forests with deep root systems
A (Very High)	> 40 tC/ha	Tropical forests, old-growth forests with extensive root networks

Technical Notes

BGB estimation relies heavily on allometric root:shoot ratios since roots cannot be directly observed by satellite sensors
Fine root turnover is rapid (months to years), so BGB represents a snapshot of standing root carbon rather than cumulative root production
Grassland and shrubland BGB may be proportionally higher relative to AGB compared to forests
Uncertainty in BGB estimates is generally higher than for AGB due to the indirect estimation approach

References

Walker, W.S. et al. (2022). "The global potential for increased storage of carbon on land". PNAS, 119(23), e2111312119.
Spawn, S.A. et al. (2020). "Harmonized global maps of above and belowground biomass carbon density in the year 2010". Scientific Data, 7, Article 112.
Breiman, L. (2001). "Random Forests". Machine Learning, 45(1), pp. 5-32.

Data Sources & References

Sources

WRD_CARBN_22

Methodology

Il layer Carbon Stock Attuale - BGB viene calcolato estraendo la componente di biomassa sotterranea (Below-Ground Biomass) dalla mappa globale di densita di carbonio di Walker et al. (2022). La biomassa sotterranea comprende il carbonio presente nei sistemi radicali: radici grossolane (> 2 mm), radici fini, rizomi e reti micorriziche. Il pool BGB viene isolato dalla mappa di carbonio totale e raffinato tramite Random Forest con NDVI (Sentinel-2), copertura del suolo (Dynamic World) ed elevazione (SRTM) come predittori. La stima BGB si basa su rapporti allometrici radice:parte aerea specifici per tipo di vegetazione. Output in tC/ha a ~200 m di risoluzione.