Moisture Recycling over the Iberian Peninsula: The Impact of 3DVAR Data Assimilation

Abstract

In this paper, we have estimated the spatiotemporal distribution of moisture recycling over the Iberian Peninsula (IP). The recycling ratio was computed from two simulations over the IP using the Weather Research and Forecasting (WRF) model with a horizontal resolution of 15 km spanning the period 2010–2014. The first simulation (WRF N) was nested inside the ERA-Interim with information passed to the domain through the boundaries. The second run (WRF D) is similar to WRF N, but it also includes 3DVAR data assimilation every six hours (12:00 a.m., 6:00 a.m., 12:00 p.m. and 6:00 p.m. UTC). It was also extended until 2018. The lowest values of moisture recycling (3%) are obtained from November to February, while the highest ones (16%) are observed in spring in both simulations. Moisture recycling is confined to the southeastern corner during winter. During spring and summer, a gradient towards the northeastern corner of the IP is observed in both simulations. The differences between both simulations are associated with the dryness of the soil in the model and are higher during summer and autumn. WRF D presents a lower bias and produces more reliable results because of a better representation of the atmospheric moisture.

Changes in the simulation of atmospheric instability over the Iberian Peninsula due to the use of 3DVAR data assimilation

“…The bootstrap analysis with resampling supports this result. Additionally, this is also in agreement with previous studies by the authors, whereby these two simulations were validated against in situ or gridded observational datasets of integrated water vapour, precipitation, evaporation, or wind over coastal areas (Ulazia et al, 2017;González-Rojí et al, 2018). Particularly for integrated water vapour, a variable closely related to the parameters analysed in this study, our verification against an independent satellitebased dataset that has not been assimilated (MODIS) showed that this quantity was substantially better simulated across the whole IP for the D simulation (González-Rojí et al, 2018).…”

“…A detailed analysis for Barcelona (Fig. S8) shows that there is a substantial underestimation of moisture at lowest levels by the N simulation, something which is consistent with findings in González-Rojí et al (2018) in a verification with independent non-assimilated MODIS-integrated water vapour. This is also observed in Murcia (figure not shown).…”

“…Data availability. These results can be reproduced using the post-processed outputs from the model available at https://doi.org/10.5281/zenodo.3611343 (González-Rojí et al, 2020b).…”

“…Both simulations were already presented and validated in previous studies by the authors. Integrated water vapour, precipitation, and evaporation over the IP were validated against station measurements and gridded datasets including independent satellite data in González-Rojí et al (2018), and the outputs produced by D were always superior to N and the driving reanalysis ERA-Interim (for the latter, at least comparable for some variables). The closure of the water balance was also better for D. Additionally, the precipitation from D exhibited similar capabilities to the one downscaled with statistical methods .…”

“…The effect of data assimilation on moisture and temperature was measured by the analysis increments (analysis minus background) in González-Rojí et al (2018). The effect of the data assimilation is more intense at 12:00 UTC compared to the other times and particularly for summer (see their Fig.…”