Skip to content
Ecological Intelligence

Ecological evidence.
Independent perspective.
Explore Wild Lands ↗

Ecological Intelligence / Subject

forest restoration

Peer-reviewed Asia

Mixed forests scored highest—and still left important functions behind

Evidence
Peer-reviewed
Geography
Four subtropical regions of southern China
System
Plantation and secondary forest

WHAT THE EVIDENCE SAYS

Across 36 plots in 32–39-year-old stands, mixed Cunninghamia-broadleaf plantations had the highest mean ecosystem multifunctionality (0.559, against 0.403 secondary and 0.383 pure plantations). They also contained the most strongly imbalanced service pairs; regeneration remained relatively weak and water retention was stronger in secondary forests.

WHY IT MATTERS / OUR READ

An average restoration score is not a balanced ecological service bundle. Track regeneration, hydrology, diversity, soil and carbon separately to judge whether site-specific management goals are actually advancing.

CAVEATS / UNCERTAINTY

Observational, middle-aged-stand comparison; it cannot isolate causal benefits of mixing species or demonstrate long-term resilience.

SOURCE

Frontiers in Plant Science — Wu et al. (2026)
View original evidence ↗

A Chinese forest comparison shows how an impressive multifunctionality score conceals service imbalances.

Read the Wild Lands assessment →
Institutional report Global

FAO’s 2026 forest outlook treats restoration as a landscape system, not a tree-counting exercise

Evidence
Institutional / NGO report
Geography
Global
System
Forests and multifunctional landscapes

WHAT THE EVIDENCE SAYS

FAO’s State of the World’s Forests 2026 reports more than 460 million hectares of degraded forest and about 1.56 billion hectares of degraded land globally. It says meeting 2030 goals requires at least 470 million hectares to be brought under restoration and presents pathways that include natural regeneration, agroforestry, hydrologic repair, native vegetation, production-landscape changes and community-led approaches rather than equating restoration with planting alone.

WHY IT MATTERS / OUR READ

The useful signal is architectural: restoration at scale is a portfolio of ecological processes, land uses, institutions and incentives. The report is strongest where it treats hydrology, livelihoods, production, governance and biodiversity as parts of the intervention instead of assuming that adding trees is the outcome.

CAVEATS / UNCERTAINTY

This is a global institutional assessment and synthesis, not a single controlled experiment. Global area estimates depend on definitions and underlying datasets, and the case studies demonstrate approaches rather than providing standardized causal effect sizes across regions.

SOURCE

FAO — The State of the World’s Forests 2026
View original evidence ↗

The new State of the World’s Forests report puts more than 460 million hectares of degraded forest inside a broader 1.56-billion-hectare land-degradation problem—and emphasizes multiple restoration pathways rather than one universal intervention.

Read the Wild Lands assessment →
Peer-reviewed

Forest carbon can recover on a similar trajectory while the ecosystem underneath takes a different path

Evidence
Peer-reviewed
Geography
China
System
Subalpine secondary and planted forests

WHAT THE EVIDENCE SAYS

Researchers compared secondary and planted subalpine forests across a recovery chronosequence, separating total above-ground tree carbon into ectomycorrhizal (EcM) and arbuscular-mycorrhizal (AM) components and testing stand structure, functional composition and soil variables. Total tree carbon and EcM-associated carbon increased with recovery stage in both strategies, with no significant strategy-by-stage interaction for either; AM-associated carbon did show a significant interaction. Secondary forests were more strongly associated with mycorrhizal dominance and functional reassembly, while planted forests were more strongly associated with stand structure and soil conditions.

WHY IT MATTERS / OUR READ

The useful result is the null as much as the difference: two restoration approaches can converge on a similar headline carbon trajectory while rebuilding different ecological machinery. Carbon stock alone can therefore hide differences in composition, structure and belowground relationships that may matter for durability and future function.

CAVEATS / UNCERTAINTY

This is a chronosequence and structural-equation analysis rather than a randomized restoration experiment, so inferred pathways remain observational. The findings concern above-ground tree carbon in one subalpine forest context and do not establish that secondary and planted forests are equivalent across biodiversity, soil carbon or long-term resilience.

SOURCE

Journal of Applied Ecology · Original evidence
View original evidence ↗

A subalpine forest chronosequence found similar total above-ground carbon recovery in secondary and planted forests, but the mycorrhizal, structural and soil pathways behind that recovery differed.

Read the Wild Lands assessment →
Peer-reviewed North America

Recovered tree cover did not bring periodical cicadas back

Evidence
Peer-reviewed
Geography
Johnson County, Iowa, USA
System
Temperate deciduous forest and regrown woodland

WHAT THE EVIDENCE SAYS

Researchers compared historical and contemporary tree cover with 2024 periodical cicada distributions and found breeding populations were better predicted by forest cover in 1930 than by current cover. Many reforested areas remained unoccupied despite decades of habitat recovery, and an experimental relocation of 20,126 adult cicadas to apparently suitable restored forest failed to establish a chorusing population.

WHY IT MATTERS / OUR READ

The counter-signal is that habitat recovery is not necessarily population recovery. For species with extreme dispersal constraints and long generation times, historical fragmentation can create a recovery debt lasting many decades or centuries even when the vegetation metric looks repaired.

CAVEATS / UNCERTAINTY

Periodical cicadas have unusually long life cycles and limited dispersal, so the result should not be generalized to all species or forest restorations. Much of the empirical work is from one Iowa landscape, and the relocation result is one experiment shaped by strong predator aggregation.

SOURCE

Ecology · Original evidence
View original evidence ↗

In Iowa, present-day cicada populations were better predicted by 1930 forest cover than by today’s habitat, and a relocation of more than 20,000 adults failed.

Read the Wild Lands assessment →

Keep exploring

Follow the systems. Find the connections.

WildlifeLandWaterOceansClimateCitiesOne HealthGovernanceMethods / Monitoring
Browse the archive →