A study by Gebrechorkos et al. (published in Nature in 2025) claiming climate change is worsening global drought severity has been retracted, reports ScienceFiles here (Das nächste Klimawandel-Junk-Paper wurde zurückgezogen). 

Gross errors throughout…

The ScienceFiles article reports that the study was heavily flawed, citing errors in calculations, miscoded data, misaligned comparison periods, and flawed weightings.

The now retracted paper claimed that drought severity had increased significantly since 1981, that “atmospheric evaporative demand” (PET/AED) had worsened droughts by 40% from 2018–2022 compared to previous decades, and that 30% of global land area had suffered moderate-to-extreme drought since 2022. It now appears all these findings were incorrect and based on shoddy methodology.

Reality: no detectable global drought acceleration

The ScienceFiles article highlights a follow-up study by Xu et al. (2026, published in Communications Earth & Environment), which challenged the original study. The main issue revolves around how Potential Evapotranspiration (PET) was calculated: In the retracted paper, soil dryness was overstated by assuming uniform moisture dynamics based on air dryness. When re-calculated by Xu et al., the drought severity ended up being less than one sixth of that reported in the retracted paper, and found no detectable recent acceleration in global drought.

The ScienceFiles author accuses major mainstream media outlets (such as Die Zeit and The New York Times) of uncritically spreading alarmist “climate apocalypse” headlines based on flawed science without issuing corrections after retractions.

ScienceFiles criticizes the journal Nature and peer reviewers for allowing low-quality science to be published in the first place.