Restoring grassland starts with fungi in the soil

Habitat restoration usually comes down to sowing plants. It less often asks whether the soil holds what native plants cannot manage without — in grasslands, the arbuscular mycorrhizal fungi, whose community is often rebuilt or impoverished after decades of cultivation.

The experiment described by Liz Koziol and colleagues in Restoration Ecology was established in 2016 on a restored tallgrass prairie. It tested a density gradient of native mycorrhizal inoculum, from zero to 8,192 kilograms per hectare. First-year results showed that native plant establishment benefited from high but not low densities: abundance, richness and diversity all improved.

The key question, though, was whether the effect would last. The site was resampled in 2020, after the fourth growing season, and analysed the same way. The effect held: inoculation still favoured native vegetation and suppressed weeds four years on.

That is an important methodological distinction. Many restoration treatments produce an effect in the first season and lose it by the third, because they change starting conditions rather than how the system works. A difference persisting after four years suggests the inoculation changed something durable in the soil.

Transferring the conclusion to European grasslands calls for caution — different continent, different plants, different fungal communities. The general principle travels well and is rarely applied, however: a restoration plan that says nothing about soil fungi is silently assuming they are present and in good order.

Dose, time and local inoculum

A tallgrass-prairie restoration experiment applied eight densities of native arbuscular-mycorrhizal fungal inoculum, from 0 to 8,192 kg/ha. In the first growing season, high densities improved native-plant abundance, richness and diversity; very low densities were ineffective. In the fourth season, greater diversity and lower non-native abundance persisted, while lower doses that showed no initial benefit also began suppressing non-native plants. This is not evidence that “any fungi repair any grassland”: the experiment used locally developed native inoculum in a Kansas prairie. Microbial transfer elsewhere requires habitat matching, pathogen-risk assessment and biosecurity, and cannot substitute for restoring hydrology, nutrient regimes or appropriate disturbance.