Changes in global water vapor from observations and reanalysis products

Understanding the long-term changes in the global water vapor content is critical for assessing natural vs. human-caused climate change. Despite the strong thermodynamical relationship between temperature and water vapor changes, substantial discrepancies still exist between observations, reanalysis products, and climate model simulations.

In this work, we assess the consistencies and discrepancies of total column water vapor (TCWV) estimates between three observational techniques and three reanalysis products. The observations include satellite-borne microwave radiometers (MWR) over the oceans, GPS–Radio Occultation (GPS-RO) observations from low-orbiting satellites over both ocean and land, and ground-based GNSS receivers over land and on islands. The three reanalyses are ERA5, MERRA-2, and JRA-55. They all assimilate radiances from the satellite microwave radiometers and bending angles produced from GPS-RO measurements. Ground-based GNSS measurements are not assimilated and serve as a fully independent validating data set.

We examine the overall agreement in global TCWV trends in the different data sets over the period from 1980 to the present. We highlight strong features of global climate variability such as the El-Niño Southern Oscillation (ENSO). We focus on the past few years which were characterized by a persistent strong La Niña period (2020-2022), followed by a strong El Niño event (2023/2024). Both ENSO phases had a tremendous impact on regional climate extremes, leading to extended heat waves and wildfires or heavy precipitation and flooding in many places around the world.