Ecological restoration can deplete water storage

Successful land restoration comes at a cost of freshwater resources if not done in a sustainable manner.

Like Comment
Read the paper
Local people preparing to plant grass to stabilize sand dunes at the edge of the Mu Us Sandyland in Lingwu, in northwest China's Ningxia Hui Autonomous Region, in 2007. To stabilize dune fields, local people create "checkboards" of straw or sticks, within which they plant vegetation. Image credit: Reuters/Stringer

When our co-author Jien Zhang first visited the Mu Us Sandyland of northern China on a field trip, local people complained that they had to dig deeper and deeper wells to get drinking water. The first that came to his mind was that restored vegetation in this region probably consumed too much water by evapotranspiration. This is worrisome because China has been restoring its Gobi Desert and similarly arid regions in the north by paying local people to plant trees, grasses, and shrubs. This could worsen water scarcity. On a global scale, large-scale ecological restoration is an increasingly popular human practice to combat land degradation and climate change. For instance, the Bonn Challenge, launched in 2011 by the Government of Germany and the International Union for Conservation of Nature, will “bring 150 million hectares of degraded and deforested landscapes into restoration by 2020 and 350 million hectares by 2030”[1]. The sustainability of these efforts requires a better understanding of their impacts on water resources.

When Jien and I met in an American Geophysical Union meeting, we performed a quick analysis of the rainfall change. We were surprised to find that rainfall had increased since restoration, which left us wondering about the net impact on total water resources. We then turned to the Gravity Recovery and Climate Experiment (GRACE) for total terrestrial water storage (TWS) observations. We find a substantial depletion of TWS after restoration, consistent with local people's accounts. However, due to GRACE's large footprint and associated errors, we need to carefully interpret the GRACE trend. With the help of co-authors Geruo A and Isabella Velicogna, we conclude that restoration efforts are depleting land total water resources at an alarming rate, comparable to other groundwater withdrawal hotspots such as the Central Valley in California. This depletion trend will likely continue if local stakeholders opt to maintain or elevate the level of restoration efforts.

Our findings provide a framework that directly informs the total water cost of these ecological restoration programs, which will help recalibrate future restoration plans' cost-benefit matrix. 

Our paper is accessible at

Greening in the desert. Image credit: Erik Solheim



Meng Zhao

Ph.D., University of California Irvine