A microphone recording six straight days, hanging two meters high, cannot distinguish between a pretty forest and a functional one. But if you leave it there long enough, and compare what it recorded with a pasture and an intact forest, the sound begins to reveal something verifiable about how well that ecosystem has recovered.

That is exactly what a team led by Giacomo Delgado, from ETH Zurich, accomplished alongside researchers from the Smithsonian Tropical Research Institute, the Royal Society for the Protection of Birds, and Costa Rican government agencies (FONAFIFO, SINAC, SETENA). The findings were published on February 4, 2026, in Global Change Biology, with an open-access version available on PubMed Central.

What Was Measured and How

The team installed AudioMoth acoustic recorders across 119 sites in the Nicoya Peninsula:

• 18 reference forests (never logged)
• 39 natural regeneration sites
• 43 plantations
• 19 pastures

In total, 16,658 hours of audio were analyzed to construct a soundscape comparison across the four land-cover types.

All of the natural regeneration and plantation sites belong to Costa Rica’s Payment for Environmental Services (PES) program (Pago por Servicios Ambientales). In effect since 1996, the program currently protects over 200,000 hectares with payments of nearly $70 per hectare per year, according to the study. It stands as one of the world’s oldest and most researched PES schemes, and this study is among the first to ask—using sound data rather than just canopy cover—whether it has truly restored biodiversity or merely brought back trees.

Natural Regeneration Sounds Different Than a Plantation—and That Matters

The central finding: Sites under natural regeneration sounded between 1.4 and up to 4 times more similar to reference forests than to pastures, depending on the time of day. In contrast, plantations were barely 1.24 times closer in sound to a reference forest than to a pasture—a far more modest advancement.

This distinction is particularly noticeable during the dusk chorus, the time of day with peak bird and insect activity: this is where natural regeneration most closely mirrors the pattern of an intact forest. Plantations, by contrast, consistently exhibited reduced biological activity throughout the entire daily cycle—morning, afternoon, and night.

Why Sound Matters (And Not Just Satellite Imagery)

A satellite vegetation map can confirm that an area “is forest” without revealing anything about how functional that forest actually is for the wildlife expected to inhabit it. A monoculture plantation also looks green from the air.

A soundscape captures a different layer of information: the physical presence of vocalizing birds, insects, amphibians, and mammals—not just plant tissue.

This makes acoustic monitoring a relatively inexpensive verification tool for PES and restoration programs. While it doesn’t replace a biologist’s field inventory, it allows teams to cover significantly more sites with fewer person-hours, detecting differences between “it looks green” and “it sounds alive” that would otherwise go unnoticed.

What This Means for the Design of Restoration Programs

The most uncomfortable outcome of the study is not that natural regeneration outperforms tree plantations (something restoration ecology has pointed out for years), but rather the sheer magnitude of the difference: up to 4 times greater acoustic similarity to reference forests.

For any country or organization designing an offset or PES program—facing the decision of whether to pay for letting a forest regenerate naturally or for establishing a plantation—that difference ought to carry real weight in the decision-making process, well beyond cost per hectare or canopy growth rates.

It also raises an uncomfortable question for active programs: How many are measuring success solely by reforested hectares or canopy coverage, without ever verifying if the wildlife actually returned?

Conclusion

A single microphone cannot replace a complete biological inventory. However, 16,658 hours of recordings across 119 sites are more than enough to reveal something the human eye—or a satellite—cannot capture with equal clarity: allowing a forest to regenerate naturally is not a passive or second-rate option compared to planting trees.

- Advertisement -