How successfully can we restore the world’s degraded lands? A new global meta-analysis seeks to understand what determines forest landscape restoration success and recovery rates.
By Karen D. Holl, Paula Meli, José M. Rey Benayas, and the SESYNC/iDIV Restoration Synthesis Working Group
Over the past five years there have been numerous global, regional, and national targets set for large-scale forest landscape restoration. Most notable among these are the nearly 60 million hectares of restoration commitments to the Bonn Challenge to restore 150 million hectares of degraded and deforested land worldwide by 2020 – commitments arriving from nearly a dozen countries and institutions spread across three continents. Restoration ambition is high, but many unknowns still exist. We know that restoration can conserve biodiversity, provide a range of ecosystem services, and support the well-being of human communities. But, we still know very little about what makes restoration successful and, in particular, to what degree ecosystems can recover from disturbance or how long it will take them to do so. Answering these questions will help guide the decision-makers now responsible for implementing large-scale forest landscape restoration, who often have large goals but limited resources.
To this end we are conducting a meta-analysis of restoration studies to determine which factors affect the degree of forest recovery across the world. We are concerned particularly with the recovery of plant and animal populations (looking at both diversity and abundance) and nutrient cycling functions. The factors we are considering include the type of past disturbance to the land (was the area mined, logged, or used for agriculture?), the existing forest type (tropical or temperate, wet or dry), the time since the disturbance has ceased, and whether humans have actively intervened to restore the degraded land.
To be clear, there are plenty of existing scientific studies on forest regeneration, reforestation, and forest recovery rates that offer limited answers to these questions. But the results of these studies are notoriously site-specific, making it difficult to draw from them general and practical conclusions. Our meta-analysis looks across these studies to find conditions that determine restoration recovery rate and success in a way that can inform the current worldwide restoration movement.
We have compiled 166 primary studies from the peer-reviewed published literature with 1,805 ecological response variables (e.g., measurements of abundance, diversity or nutrient cycling functions). These studies include a broad range of examinations from temperate and tropical and wet and dry forests covering 41 different countries (Figure 1). Each study includes measurements of forest quality after degradation and after restoration, as well as measurements from nearby minimally degraded forest (so called “reference” measures). We are comparing each of these measures to determine how deeply each type of disturbance (i.e., mining, logging, or agriculture) degrades a forest’s health and, on the other hand, how well restoration returns that forest to a state similar to nearby reference forests.
Our preliminary results on degradation suggest that agricultural use typically degrades forest ecosystems more than logging (Figure 2). In its destruction of the land mining falls somewhere between agriculture and logging. And logging proves to be the least destructive. Regardless of the kind of disturbance, our findings suggest that all post-disturbance lands prove significantly degraded when compared to reference forests.
After disturbance most logged sites are restored naturally, we have found, and these recover generally without human intervention (Figure 3). In contrast, mined sites are nearly always actively restored, usually by a mixture of interventions including the reconstruction of original land topography and the replanting of vegetation. Former agricultural sites are restored by a mix of approaches, including both the simple removal of the disturbance (e.g., cattle) to allow for natural regeneration and the active planting of trees.
We are in the process of determining the extent to which different kinds of degraded land recover from disturbance and, indeed, if the type of forest and style of restoration determine the rate of recovery. We will present more of our findings in a second post, here at IUCN, when our analyses are complete later this summer. Stay tuned.
This project is part of a larger study synthesizing results of ecosystem recovery and restoration across a range of ecosystem types that is funded by the U.S. National Socio-Environmental Synthesis Center and the German Centre for Integrative Biodiversity Research. This study on forest recovery is funded by the Know-For-FLR project of IUCN, made possible through support from UK Aid, by the British Government.
Bonner, MTL, S Schmidt & LP Shoo. 2013. A meta-analytical global comparison of aboveground biomass accumulation between tropical secondary forests and monoculture plantations. Forest Ecology and Management 291:73-86.
Rey Benayas, JM, AC Newton, A Diaz & JM Bullock. 2009. Enhancement of biodiversity and ecosystem services by ecological restoration: a meta-analysis. Science 325:1121-1124.