Mapping Earth’s underground fungal networks for the first time, estimated to stretch a distance roughly equivalent to traveling to the sun and back nearly a billion times.
Mapping Earth’s underground fungal networks required solving a problem unrelated to counting individual organisms. Arbuscular mycorrhizal, or AM, fungi form symbiotic partnerships with roughly 70% of plant species on Earth, trading soil nutrients such as phosphorus and nitrogen for carbon that plants produce through photosynthesis.
The fungi extend the reach of plant roots by up to 100 times through thread-like structures called hyphae, each about a tenth to a twentieth the width of a human hair. In just the top 15 centimeters of soil, researchers estimate every square centimeter packs roughly 4.4 meters of hyphae.
Published in Science and led by researchers at the Society for the Protection of Underground Networks, the study mapping Earth’s underground fungal networks drew on more than 16,000 soil cores collected worldwide, combined with machine-learning models trained on data spanning deserts, tundra, forests, and grasslands.
Researchers calibrated those models using robotic imaging of over 300,000 living AM fungal hyphae grown in laboratory conditions in collaboration with the AMOLF Biophysics Institute in Amsterdam. The resulting estimate: roughly 110 quadrillion kilometers of hyphae carrying about 300 megatons of carbon.
That carbon figure matters beyond its sheer scale. AM fungal networks move an estimated 4 billion tons of carbon dioxide equivalent into soils every year, roughly 11 percent of all human-related carbon dioxide emissions. Justin Stewart, the study’s lead author, called the fungi one of Earth’s circulatory systems, moving carbon, water, and nutrients across underground ecosystems largely invisible to conservation policy built around what’s visible aboveground.
Mapping Earth’s underground fungal networks reveals where these systems face the greatest risk. The pattern is stark. Wild grasslands hold an estimated 40% of the planet’s AM fungal biomass, with particularly dense networks predicted in South Sudan’s flooded grasslands, Florida’s Everglades, and the Tibetan Plateau. Grasslands are also among Earth’s least protected ecosystem types and are being converted to farmland roughly four times faster than forests are being cleared.
Wild grasslands like the Everglades and South Sudan’s flooded plains hold an estimated 40% of Earth’s fungal biomass but remain among the least protected ecosystems. Photo by K on Pexels.
The agricultural comparison is direct and troubling. Croplands showed roughly half the fungal network density of wild ecosystems, a 47% reduction where native grasslands had been converted to farmland.
Researchers suspect tilling physically disrupts hyphal networks underground, while synthetic fertilizer may reduce a plant’s reliance on fungal partnerships for nutrient uptake in the first place, since the plant no longer needs to trade carbon for nutrients it’s already getting from a bag. Stewart suggests that Indigenous farming methods, no-till practices, and organic growing may preserve more of this underground infrastructure, though the researchers caution that definitively linking specific farming practices to fungal health requires further study.
The conservation gap is the study’s most consequential finding. It reinforces earlier SPUN research published in Nature, which showed that fewer than 10% of AM fungal biodiversity hotspots fall within existing protected areas.
Mapping the physical infrastructure of these networks now, rather than just their biodiversity, shows that exposure extends to the sheer mass and density of fungal networks themselves, not merely species richness. Protected area boundaries drawn to safeguard visible wildlife and forest cover largely miss the underground systems those same ecosystems depend on.
Toby Kiers, SPUN’s executive director, frames the stakes bluntly: fungi have been left out of climate and conservation planning for too long, and this body of research is meant to change that by giving policymakers concrete action. Whether that translates into altered protected-area boundaries or agricultural policy remains to be seen; a map, however detailed, doesn’t compel land-use decisions on its own.
Real uncertainty remains embedded in the methodology. Regions like the Sahara Desert and Greenland lack sufficient soil sampling data for the model to confidently estimate fungal presence, leaving significant blank spots in an otherwise remarkably precise-sounding global figure. The 110-quadrillion-kilometer estimate is a machine-learning extrapolation from thousands of sampled points, not a direct measurement, and should be read with the same caution applied to any large-scale ecological model built substantially on interpolation.
Mapping Earth’s underground fungal networks doesn’t, by itself, slow grassland conversion or change fertilizer subsidies that drive reduced fungal density on farmland. What it does provide, for the first time, is a baseline detailed enough that future soil disturbance, land conversion, or ecosystem loss can be measured against a known starting point, turning a previously invisible planetary system into something scientists and policymakers can actually track.
