This was the year that business added its clout to the global effort to add trillions of trees to the arsenal of weapons against climate change. The goal is to use these trees to restore land and absorb climate-warming carbon out of the atmosphere, which could keep global warming below 1.5 degrees C (2.7 degrees F), the level the Paris Agreement says is needed to avoid the most damaging effects of climate change.

But planting the trees in the ground is just the beginning. We need to track if these trees keep growing, how much carbon they store and what other types of benefits (e.g., food, economic, biodiversity) they provide to people and the environment. There’s one major problem to start: How do we track how and where these trees are growing?

Why we need to track tree growth

Just as not all trees grow to maturity, not all tree-planting campaigns succeed. Tracking tree growth helps governments, companies and NGOs to understand progress on their pledges, encourages people to replicate successful projects and tweak struggling ones and inspires funders to continue investing where they can see past progress.

Measuring progress can also put the spotlight on farmers and others who have restored their land with little external help. Across the dry Sahel, for example, farmers have restored millions of hectares of land, boosting crop yields and stopping the southward creep of the Sahara Desert – with little outside aid or acknowledgment. These success stories and the lessons they hold can’t be shared unless we can systematically measure changes in the landscape.

Challenges of tracking tree growth

High-quality satellite data and platforms like Global Forest Watch track deforestation, but they don’t work well for tracking tree growth. Why?

Growing trees to maturity takes time, often 15 years or more. And with freely available satellite imagery, it is harder to see a tiny sapling slowly growing in a field than to see the blank space where a tree once stood. Very-high-resolution imagery that could see that change is available, but it is prohibitively expensive to purchase and analyze for the entire world every year.

Some individual projects do track tree growth, but each project follows its own method, and the type and quality of data vary greatly. Much of this data is not independently verified, and only spans the first few years of the project (before the trees fully mature). Third-party independent verification that certifies a project's carbon or biodiversity benefits is often prohibitively expensive, especially for local farmers working in the developing world. All of that makes project managers and funders alike skeptical of tree restoration success stories.

Building globally consistent datasets

That is why we want to solve some of these problems before the UN Decade on Ecosystem Restoration starts in 2021, and corporate pledges to protect and grow a trillion trees start in earnest. As a member of the UN Decade’s task force on restoration monitoring and co-chair of the Global Restoration Observatory, WRI is fully engaged in this challenge.

Recent studies have found that there are billions of sparsely scattered trees in rural areas that we couldn’t see with conventional remote sensing approaches. By including those trees in global data, we can paint a more comprehensive picture of tree cover gain and loss.

A pilot study that WRI conducted as part of the five-year progress assessment of the New York Declaration on Forests showed that in Southeast Asia’s Mekong region, the number of trees on farms and pasture – where restoration has the most benefit for people – increased, even as forest loss expanded. Building on that study, we are working with new data from the University of Maryland’s GLAD lab to expand that work to Latin America, aiming to develop a globally consistent, annual dataset that can track where trees are growing at the global, national and subnational scales.

Harnessing the power of artificial intelligence, we have also begun analyzing raw satellite data across more than 30 landscapes of around 10,000 hectares (25,000 acres) each. The goal is to build accurate wall-to-wall maps of millions of trees that official statistics, plans and models have rarely considered. Combining this approach with the global data can help us see more trees more accurately than before and track the overall progress of Trillion Trees and the tree restoration movement.

Establishing a protocol and community of practice

Globally consistent, independent data is important for identifying where “hotspots” of tree restoration success (or failure) are occurring, so that we can better understand the impact of countries’, NGOs’ and companies’ investments.

We also need to make sure that the many individual projects that get funding from platforms like TerraMatch are part of a well-supported community of practice, where consistent, independent monitoring protocols are in place to support both project implementers and funders.

Adapting to the local context and goals is important. To build this community of practice, we need a consistent, scientifically sound, cost-effective protocol that practitioners across the world can use, and investors can trust. That’s what WRI is building for projects that restore trees within agricultural, forest and urban landscapes. The protocol focuses on key indicators – tree count, tree cover and tree species – that will be measured for up to 15 years after planting.

Because this protocol will combine field data collection with remote monitoring approaches like satellite imagery interpretation and artificial intelligence, the costs will be kept as low as possible. When projects use the same protocol, their aggregated data will help us understand whether those trillion trees are growing and whether our efforts to fight climate change are on the right track.

Measuring, understanding and investing in tree restoration

We will need a robust global atlas to track tree restoration progress that mirrors the role that Global Forest Watch plays for tree cover loss – a “Restoration Watch.” We need a platform that will deliver lessons learned from the robust work that WRI and many others have carried out at the national and landscape levels. We also need an adaptable yet consistent protocol for projects.

Measuring tree growth is only one benchmark of restoration success. We need to invest in measuring the economic and social impact of restoration too, to ensure that people stand to benefit.

For now, in a world where corporations and governments are increasingly interested in proving their success scientifically, tracking progress is essential. Funders invest in measuring success and insist that their planting partners do too. They must realize that the true price of tree restoration success goes beyond tree planting to include the cost of engaging local people, preparing and caring for trees and measuring progress. With this new momentum spurred by corporate pledges, platforms like TerraMatch and the latest scientific advances, now is the time to start tracking progress.

Originally appeared on WRI Insights.