Assisted Forest Regeneration Lab

Multi-lingual lit review to accelerate global progress toward best practices for forest carbon capture and biodiversity recovery.

There is tremendous interest in mitigating the impacts of climate change with forest restoration. Many strategies are used to regain tree cover on deforested lands, ranging from low-cost approaches like natural regeneration (letting forest grow back on its own) to resource-intensive approaches such as assisted restoration (e.g., tree planting). Given the limited resources to achieve ambitious international goals, determining the most efficient and effective restoration approach at specific sites is critical to scaling the global restoration movement.

While there have been significant advances in understanding where forests are likely to naturally regenerate, the rates of recovery under natural vs. assisted restoration are extremely variable and often sparsely documented. Current remote sensing techniques and models, while useful in a broad sense, fall short of delivering regionally specific recommendations that can directly guide the choices of restoration practitioners on the ground. Moreover, in situations where assisted restoration approaches are necessary, projects have a high failure rate, often because inappropriate approaches are selected for site-level conditions.

The CE AFR Lab helped address this problem by conducting a systematic literature review synthesizing existing published sources, unpublished data from field partners, and many non-english sources which are often not included in similar efforts. A diverse consortium of reviewers read through thousands of identified texts to pull out important details like restoration methodologies, location, size, and outcomes.

What began as a standard literature review evolved into an opportunity to expand how restoration knowledge is collected. As the scale of the task came into focus, tech-savvy team members began scraping data, writing code, and adapting the open-source machine learning tool ASReview to prioritize and sift through the vast body of material. This work, initially formed in this lab, was later developed further in the IDEAS microlab, which you can read more about here.

The information gleaned from this lab will contribute to a global analysis of forest restoration outcomes that will be submitted for publication by partners at ETH Zurich.