A carbon credit is financial proof that a project has reduced or avoided the release of greenhouse gases, such as those caused by tree-plantings or solar farms. When used correctly, it can help companies meet their goals to reduce climate impact. But a fragmented and confusing carbon market is producing credits with dubious environmental value. And some buyers are worried they’re not getting a fair price for their carbon credits.

A few large carbon.credit have generated interest from investors and companies seeking to offset their climate footprints, but the market is still in its early stages. As the number of companies pursuing net-zero emissions grows, it’s crucial to develop a transparent and robust voluntary carbon market that’s free from fraudulent projects.

One of the biggest obstacles is the difficulty of verifying carbon-capture claims. Soil samples must be taken and analyzed at least three times to ensure they are accurate. This is expensive and labor-intensive, and many projects are struggling to cover costs. This year, some developers are working to cut sampling expenses by partnering with farmers and land owners to share the cost of bringing in equipment like satellites and remote-sensing vehicles.

The new approaches are boosting farmers’ returns and helping them recoup their setup costs, but they come with risks. A mistake in calculating the carbon content of a forest or cropland can lead to misguided project decisions and even fraud. A few years ago, a developer sold credits based on the false claim that trees growing in a certain area had more carbon than the regional average. That decision produced tens of millions of credits of dubious climate value.

In the case of New Forests, a project on the Mescalero Apache Tribe’s reservation in New Mexico, an error boosted the amount of credits the company could earn. Using the wrong forest-carbon calculations, New Forests’ project was able to claim that it could have heavily logged its forestland and earned an extra 3 million credits.

Other forest projects have also erred, and some have failed to capture the carbon they claim, according to research by CarbonPlan, a San Francisco nonprofit that analyzes carbon-reduction efforts. New Forests and a developer called Finite Carbon, which BP bought last year, declined to comment on the research, but both said their projects were in compliance with California’s program.

More companies are pledging to limit their climate impacts, but they often find it is impossible to entirely eliminate their own carbon emissions or lessen them as quickly as they’d like. So they’re turning to carbon projects to help them reach their goals by addressing the remaining scope 1, 2, and 3 emissions that can’t be eliminated or lessened. Increasingly, those projects are being sourced from the world’s poorest people, with some relying on unsustainable practices that can backfire. To be effective, those projects must do more to uphold the sovereignty and self-determination of Indigenous Peoples and local communities. The success of carbon projects depends on their ability to do that.