Combining Satellite Imagery, Airborne and Ground-based Measurements with Numerical Models for the Monitoring of Hydrometeorological Processes over the UAE

Abstract

Accurate and timely monitoring of precipitation remains a challenge, particularly in arid regions such as the United Arab Emirates (UAE), characterized by complex land-atmosphere dynamics and inadequate hydrological monitoring networks. The overarching goal of this dissertation is to improve precipitation monitoring and better understand the underlying generation mechanisms specific to the UAE and similar arid environments – from regional to microscale processes. This research is divided into three work packages aiming to (i) assess the role of land surface-atmospheric interactions through soil moisture-precipitation feedback, (ii) develop the first multisource, gap-filled and ground-calibrated rainfall dataset and (iii) quantify the role of aerosol-cloud-precipitation interactions over the UAE. The methodology incorporates numerical modeling, remote sensing, ground-based and in situ airborne measurements of key hydro-meteorological variables. First, the Weather Research and Forecasting (WRF) model, along with its enhanced hydrological land surface model (Hydro), are fully-coupled in a nested configuration and compared to standalone WRF simulations to assess the impact of land surface processes on rainfall generation. The physics-based WRF model results are then used to guide the data-driven fusion of the UAE's first high-resolution rainfall dataset merged from radar, satellite and ground-based parameters through Geographically Weighted Regression (GWR) and Artificial Neural Networks (ANNs). Finally, unresolved microscale aerosol-cloud-precipitation interactions are studied using in situ airborne measurements collected over the UAE during August 2019. Storm-scale precipitation forecasts from WRF-Hydro show reductions of 24% and 13% in root-mean-square error (RMSE) and relative bias (rBIAS) measures, respectively, compared to standalone WRF results. The spatiotemporal evolution of WRF-Hydro-simulated soil moisture compares well to satellite observations, which indicates the model's capability to simulate surface drainage, and the importance of resolving soil moisture perturbations to account for local rainfall recycling through surface evaporation over desert environments. The inclusion of satellite-based soil moisture as an explanatory variable in the merged rainfall product gave improved results from the ANN model compared to GWR, with relative increases in Nash–Sutcliffe Efficiency (NSE) coefficients of 56% (and 25%) during summer (and winter) seasons. In addition to the local-scale control of soil moisture, airborne measurements show the active role of background dust aerosols as cloud condensation nuclei (CCN) with the onset of their deliquescence in the sub-cloud region. Despite the presence of giant CCN (d>15 µm) from local pollution, particularly over the southwest, the low concentrations of intermediate-size drops (10–15 µm) hinders the development of an active collision-coalescence process. The role of background dust aerosols as precipitation suppressants with diameters up to 100 µm is shown to be a critical factor to account for in ongoing operational cloud seeding activities over the UAE.

Date of Award	Sep 2020
Original language	American English