Seeking consensus between Eulerian and Lagrangian moisture tracking methods for precipitation origin analysis in Atmospheric Rivers

Global warming is increasingly aggravating hydro-climate extremes, such as floods and droughts. In this context it is essential to understand the complex dynamics of the atmospheric branch of the water cycle, including the link between evaporation and precipitation. For this reason, many studies have investigated the origin of the moisture that feeds precipitation, which has led to a better understanding of atmospheric water transport. However, the lack of observations has prevented a direct validation of the different moisture tracking tools used for this purpose, and it is common to find large discrepancies between the results they provide.

To fill this gap, we compare two different Lagrangian methodologies for moisture tracking based on the FLEXible PARTicle dispersion model (FLEXPART) against the Eulerian “Water Vapor Tracers” technique based on WRF (WRF-WVTs). Considering the results of WRF-WVTs as “ground truth”, we explore the discrepancies between the Eulerian and Lagrangian approaches for five precipitation events associated with ARs and, based on that, propose some physics-based adjustments to the Lagrangian tools. Our results show that Lagrangian methodologies using evaporation data instead of specific humidity data provide results much closer to those of WRF-WVTs. Specifically, they reduce large biases that underestimate remote sources (such as tropical ones), while overestimating local contributions. When we introduce our physical corrections, both methods improve remarkably, which means that these biases are strongly reduced and the results provided by the different techniques reach a consensus.