Despite taking up just about a quarter of Earth’s landmass, the Southern Hemisphere endured up to 95% of water loss over the past 2 decades, according to a new review. South America, southwestern Africa, and northwestern Australia are the most affected regions.
“We are seeing exactly what the IPCC reports predicted.”
According to the review published in Science, rainfall and evaporation levels roughly cancel each other out in the Northern Hemisphere, which largely explains why the region’s influence on global water loss is so small.
“We expected to see a higher water decline in the Southern Hemisphere, but not to that extent. That was a bit of a surprise,” said hydrologist and lead author Yongqiang Zhang from the Institute of Geographic Sciences and Natural Resources Research at the Chinese Academy of Sciences.
“We are seeing exactly what the IPCC [Intergovernmental Panel on Climate Change] reports predicted,” said Tércio Ambrizzi, a meteorologist at the University of São Paulo, Brazil, who did not take part in the study. Because it has less landmass, the Southern Hemisphere ends up being more vulnerable to phenomena that happen between the atmosphere and oceans, which explains why the region is more affected by climate variability, Ambrizzi explained.
Climate change is partly to blame, according to the study. The El Niño–Southern Oscillation—which causes the waters of the Pacific Ocean’s tropical region to warm and cool, also dramatically influencing rain patterns in South America and Australia—also plays an important role in the Southern Hemisphere’s water loss.
But human activity makes a significant contribution as well. Human-made reservoirs account for 57% of the change in global water availability, research revealed, and in arid and semiarid regions, human influence on water loss increases to nearly 100%. In terms of river flow, the Middle East and Central Asia bear most of the brunt: 70% of the global river water loss takes place in these regions, the study reports, mostly because of droughts and human activities such as the construction of dams and the diversion of rivers.
The results show the effect of many years of overexploitation of water resources on and under the ground.
To Ana Elisa Abreu, a geohydrologist at the State University of Campinas in Brazil who did not take part in the study, the results show the effect of many years of overexploitation of water resources on and under the ground. “Aquifers such as the Urucuia [between Brazil’s northeastern and northern regions and connected to the São Francisco River, one of the country’s largest basins] are being superexploited because of agribusiness expansion in that region. There’s a real worry it can be exhausted soon,” she said.
Hard Work, Challenges, and Perspectives
To get to their results, the authors waded into a sea of data and studies published in the past 2 decades.
“The definition of water availability was very mixed” in the literature, said Zhang. Some researchers used streamflow as an indicator, whereas others used the drought severity index, for example.
Systematizing time periods was also a challenge. “Some studies looked into 2 decades, others looked into 6,” so conclusions varied a lot, Zhang said.
Zhang and his fellow researchers combined remote sensing data with data sets of ground-based precipitation and evapotranspiration but needed to find a consistent definition of water availability. So they used statistical modeling and machine learning to get to an estimate that subtracted evaporation from rainfall levels (precipitation minus evapotranspiration) at a yearly scale.
The team then compared that estimate to data on water availability in different river basins around the world. To be included, the rivers had to have more than 30,000 square kilometers of basin area and more than 10 years of observation between 2003 and 2016; 134 basins met the criteria. “The test showed the water availability we estimated corresponded very nicely with the absolute values from satellite and on-the-ground data [of these rivers],” Zhang said.
“They definitely had a lot of work,” said Abreu. “But as there are limitations in measuring evapotranspiration with remote sensing, it’s important to validate [remote sensing] data with on-the-ground observations,” she explained.
The techniques the researchers used, Abreu said, are very advanced, and “this is cutting-edge work that needs further confirmation with other studies in the future.”
Zhang said one thing he and his team wanted to do but could not do with this study was to extend the analysis for a longer period of time. “It was challenging to find solid data to validate the results further back than 2002,” Zhang said. “But this is something we could do probably in the future by using a model or another technology to reduce uncertainty, but again, this is also very challenging.”
To Abreu, the results clearly show the need for South American countries like Brazil to invest in more numerous and more precise monitoring networks for precipitation and river flow. “We can’t stop investing in these networks,” she said, “so we can have a time series that can prove or disprove this tendency.”
—Meghie Rodrigues (@meghier), Science Writer
