Update on 09/03/2022: This paper has been retracted by the authors after an error was discovered in their approach, which “alter[ed] some results”.
In their retraction note on the Nature website, the authors explain:
“In this article, we calculated global land evapotranspiration for 2003 to 2019 using a mass-balance approach. To do this, we calculated evapotranspiration as the residual of the water balance, using an ensemble of datasets for precipitation, discharge and total water storage change. We made an error in calculating the global mean precipitation: we used arithmetic averaging to calculate the mean, instead of calculating a spatially weighted mean to account for the changing grid box size with latitude. As a result, the magnitudes of the global mean precipitation time series were underestimated. This impacted the subsequent calculation of global mean evapotranspiration, resulting in the mean evapotranspiration values being underestimated and altering some results. We are therefore retracting this article.”
Evapotranspiration – the transfer of water from the ground into the air through a combination of evaporation and transpiration – increased globally by 10% between 2003 and 2019, according to new research.
The study, published in Nature, finds that the change is mainly driven by the world’s land surface heating up. This confirms a long-standing theory that, as the climate continues to warm in the future, the water cycle will gain energy and “intensify”.
The study introduces a novel way of tracking changes in the mass of water from different parts of the planet by using satellite data. This method allows the authors to “complete the water cycle picture for evapotranspiration”, a scientist who was not involved in the research tells Carbon Brief. “This closure has eluded scientists for decades,” he adds.
However, if the observed trends continue, “the associated decreases in river flow and increases in wildfire risk will present challenges for water managers and emergency planners across the world”, another scientist tells Carbon Brief.
After precipitation, evapotranspiration is the second largest component of the global water cycle. It is a term that describes the movement of water from the Earth’s surface to the atmosphere – through evaporation from the soil and bodies of water and transpiration from vegetation.
Scientists have long thought that, as global warming drives up temperatures, the increase in energy to the climate system will drive higher levels of evapotranspiration. This would result in an accelerated, more “intense” global water cycle.
To measure changes in evapotranspiration, scientists have typically used techniques including models, remote sensing and in-situ observations. However, the study notes, each method has its limitations when making estimates for large regions or the entire world.
Dr Michael Byrne – a climate scientist and lecturer at the University of St Andrews, who was not involved in the paper – tells Carbon Brief that in-situ measurements are “typically sparse, strongly influenced by local conditions and, by themselves, insufficient for global estimates of evapotranspiration”.
The authors of the paper add that there are “extensive uncertainties” linked to some of the other methods – for example, in scaling local observations up to a global level.
Dr Joshua Fisher, a research scientist at NASA’s Jet Propulsion Laboratory, who also was not involved in the study, tells Carbon Brief that evapotranspiration can be measured using satellite images. However, the satellites need a clear line of sight to take readings, Fisher notes, and clouds often “get in the way”.
In this study, instead of trying to measure evapotranspiration directly, the authors use satellites to measure the related components of the water cycle. Furthermore, instead of using traditional satellite images, the authors use data from the Gravity Recovery and Climate Experiment (GRACE) and GRACE-Follow On (GRACE-FO) satellite missions to measure changes in the position of large amounts of water.
GRACE is “unique” because the satellites do not need to see the surface of the Earth directly to gather data, Fisher tells Carbon Brief. He explains that the satellites take advantage of the pull of gravity, adding that this sidesteps the issue of clouds because they do not have much mass:
“GRACE/FO literally and physically moves when it is pulled by gravity. Gravity is caused by mass. So, the gravitational field is something that one can feel without needing to physically see it. The mass, in this case, is land and water. But, the land more or less stays the same, while the surface water changes – there’s not much mass in clouds.”
To measure evapotranspiration, the authors assume that all rainfall is divided into evapotranspiration, river discharge and land-based water storage. By measuring the three other components of the water cycle individually, the authors can, therefore, calculate evapotranspiration.
Byrne tells Carbon Brief that replacing direct measurements with this “water-budget” method is “a creative approach to estimating global evapotranspiration trends”. He adds:
“A range of independent measurements of these quantities were combined to provide a robust estimate of the global trend in evapotranspiration and, crucially, its uncertainty.”
The study uses this approach to produce an “ensemble”, or collection, of 20 estimates of global land evapotranspiration for 2003-19.
The images below show how evapotranspiration (top), precipitation (second from top), discharge (second from bottom) and change in ground water storage (bottom) have varied over 2003-19. The black line shows the average trend and the shading shows the confidence range, where red regions indicate a high confidence.
The authors find a “statistically significant” increase in evapotranspiration of 2.3mm per year over 2003-19 – corresponding to an increase of around 10% above the long-term average. According to the study, these findings “are consistent with the hypothesis that global evapotranspiration should increase in a warming climate.”
By comparison, precipitation increased by 3% and discharge decreased by 6% over the same period, relative to their long-term averages, the study says.
Meanwhile, Dr Madeleine Pascolini-Campbell from the Jet Propulsion Laboratory at NASA – the lead author of the study – tells Carbon Brief that the trend confirms the suggestions of other models and data. However, she adds that the change is “larger than we expected based on current estimates”.
The authors also identify the drivers behind the changes they identified. They find that more than half of the increase is driven by global land surface temperature, with hotter temperatures driving increases in evapotranspiration. Meanwhile, 17% of the variability is due to the natural climate phenomenon, the El Nino Southern Oscillation.
The authors note that, as the climate warms, the amount of rainfall re-entering the atmosphere through evapotranspiration is increasing, while “surface runoff” is decreasing. Runoff is the amount of rainfall that flows over land and into streams and rivers, rather than soaking to the soil.
Pascolini-Campbell explains that this drop in runoff could have implications for water availability:
“Increasing evapotranspiration affects the amount of surface water available for agriculture and water supply. In the future, increasing evapotranspiration could also mean the increased loss of water from land in some regions – leading to droughts – while other areas experience more intense rainfall as atmospheric circulation is affected.”
Byrne notes that higher evapotranspiration could also lead to increased risks of wildfires by drying out vegetation:
“If these evapotranspiration trends continue in line with expectations, the associated decreases in river flow and increases in wildfire risk will present challenges for water managers and emergency planners across the world.”
Willett adds that the trends found in this study also explain “part of the story” behind the decrease in relative humidity that scientists have observed over the land:
“The study finds that this [increase in evapotranspiration] is largely temperature driven, and linked to a small increase in precipitation and a larger decrease in water storage. This suggests a drying out of the land surface that is also likely to be part of the story behind decreasing relative humidity over land.”
Meanwhile, authors also note that higher rates of evapotranspiration can have a cooling effect on the land surface, as the evaporating water takes energy away from the surface of the Earth. This “could create a negative feedback mechanism” that partially counteracts the warming from climate change, according to the study.
‘Complete the water cycle picture’
Evapotranspiration is a “critical” component of the water cycle, Pascolini-Campbell tells Carbon Brief, but has historically “been very difficult to measure at large scales”. She adds that this study introduces “a new line of evidence that the water cycle is becoming more intense with warming temperatures”.
Fisher tells Carbon Brief that the paper is “excellent” and that the satellite data provides “awesome closure to the global water cycle”. He tells Carbon Brief:
“By combining the ocean measurements for river discharge along with good precipitation measurements, GRACE/GRACE-FO can now complete the water cycle picture for evaporation. This closure has eluded scientists for decades. [The study has] solved it. And, sure enough, they can now see through the fog of uncertainty to confirm that evaporation has indeed been increasing with increased trapped energy from rising CO2.”
Dr Alexis Berg is a research associate at Harvard University and was not involved in the study. He tells Carbon Brief that the study is “an original approach and a great effort”, but adds that he also has concerns about the authors’ attribution of increasing evapotranspiration to increasing land temperature.
He tells Carbon Brief that, over land, higher temperatures can drive greater evapotranspiration in wet regions. However, he adds that in dry, “water-limited” regions, the relationship “goes the other way”:
“Even though global-mean changes in evapotranspiration are going to be dominated by wet regions, I think the global picture, in terms of processes, is a bit more complicated than just ‘more warming, more evapotranspiration’; indeed, temperature and evapotranspiration are coupled in complex ways.”
Berg adds that the authors do not mention the “global greening” trend that has taken place in recent decades, which “could lead to increased evapotranspiration, independent of warming”.
Meanwhile, the authors of the study note that the research presents a “bulk estimate” of evapotranspiration – and cannot be used to pinpoint specific hotspots of climate change.
Byrne tells Carbon Brief that, for a global-scale measurement, “this kind of ‘top down’ estimate of evapotranspiration is insightful. However, he adds that it could “potentially be further developed to measure evapotranspiration trends at regional scales”.
Fisher agrees that the next step is to “fuse these big picture measurements with local-scale measurements”.
Pascolini-Campbell, M. et al. (2021) A 10 per cent increase in global land evapotranspiration from 2003 to 2019, Nature, doi:10.1038/s41586-021-03503-5