DATE2022-06-22 00:35:56
TITLESimulation of the Moroccan climate using a variable resolution GCM
AUTHORSSaloua Balhane (1) ,Fatima Driouech (1) ,Frédérique Cheruy (2,3) ,Adriana Sima (2) ,Abderrahmane Idelkadi (2) ,Étienne Vignon (2,3) ,Abdelghani Chehbouni (1) ,Philippe Drobinski (2,4)
  1. 1) Mohammed Vi Polytechnic University, Iwri, Benguerir (Morocco) ,2) Laboratoire De Météorologie Dynamique (lmd)/ipsl, Paris (France) ,3) Sorbonne Université/cnrs, Umr, Paris (France) ,4) École Polytechnique/institut Polytechnique De Paris/ens/psl Research University, Palaiseau (France)
ABSTRACTLying within the influence of the Atlantic, the Mediterranean, and the Saharan desert, together with very steep orography, Morocco is one of the most vulnerable territories to climate change in the Mediterranean and North Africa, where precipitation is expected to decline considerably and temperatures are getting warmer. The ongoing and projected climate changes can threaten the stability of many climate-sensitive sectors, including water and agriculture. A better understanding and assessment of the regional climate variability and change in Morocco is, therefore, necessary for supporting effective risk management in such sectors. Dynamical downscaling is classically done through limited-area Regional Climate Models (RCM) driven by large-scale fields from the global models. RCMs can improve the representation of many processes, such as mesoscale circulation and orographic effects but they also have flaws that can affect the reliability of climate change projections. For instance, potential inconsistencies between the physical parameterizations of the RCMs and their forcing GCMs or an incomplete description of some climate forcings. In this work, we are using a variable resolution global general circulation model, LMDZ (Laboratoire Météorologie Dynamique, Z stands for zoom), in a coupled configuration (atmospheric/land-surface component of the IPSL climate model) forced with observed SST and SIC. We developed a new configuration reaching 35km resolution over Morocco (using the model’s grid stretching capacity: “zoom”) in order to study regional characteristics of Moroccan precipitation and temperature events and their response to global warming. Our approach suggests a regional climate evaluation framework, where the model's behavior is isolated from any external inherited biases issued from driving models. A 36-year-long (1979-2014) simulation is produced and compared with a similar simulation where state variables are nudged towards reanalysis in order to constrain the large-scale dynamics and assess the model’s physics-related biases. The regional climate simulations are then compared to a hierarchy of simulations, which includes intermediate resolution global simulations (50km) and low-resolution AMIP simulations(250km) generated within the framework of the CMIP6 exercise. They are also compared to several observational datasets (stations and satellite-based). Our results show good consistency in the mean global circulation fields and improvements due to the model's increased resolution over Morocco. For instance, the refined grid simulation reduces the wet bias (present in the other configurations) over the Atlas by reducing convective precipitation. On a regional scale, the seasonal cycle of precipitation is also improved in most regions. We also show improvements in terms of humidity transport especially compared to the intermediate global resolution (50km) which is 20% more expensive than our configuration. The developed model configuration and the obtained results will be used for the assessment of climate change projections in the region and the related uncertainties.