Aerosols are the combination of fine solid and liquid particles in the atmosphere that significantly modulates the total energy budget of the earth and atmosphere (Fan et al. 2021). These particles may considerably influence the climate system and cause changes in cloud properties, water cycle, ecosystem, environment, agriculture, and most importantly, can affect human health (Gupta et al. 2013; Middleton and Kang 2017). Moreover, it is reported in numerous studies that aerosols dominantly contribute to global climate change (Schwartz and Andreae 1996; Haywood 2021). Underscoring the fact, Arabian Peninsula (AP) region is of high importance for its contribution to atmospheric aerosols in the Northern Hemisphere of the world, including both natural and anthropogenic aerosols (Lihavainen et al. 2017). The aerosol concentrations from natural phenomena in AP, especially from dust outbreaks and storms, are typically high in the summer season due to the peak dust storm activities in this season. For instance, it is reported that the AP region receives fifteen to twenty major dust storms per year (Jish Prakash et al. 2015). Whereas anthropogenic aerosols in this region originated mainly from traffic pollution, oil and gas industries, construction works, etc., (Khodeir et al. 2012). Thus, it is indispensable to understand the aerosol concentration, properties, types, and their interaction with the earth's climate system in this region.
Aerosol optical depth (AOD) is an important parameter that quantified the aerosol loading in the atmosphere and comprehensively describes the impact on climate. The AOD information is generally collected through ground-based stations, satellite remote sensing, and model simulations. Global and regional ground-based station networks using Sun photometers such as AERONET (Holben et al. 1998), AEROCAN (McArthur et al. 2003), GAW PFR (Wehrli 2000), BoM (Mitchell and Forgan 2003), SKYNET (Kim et al. 2004), NOAA/ESRL (Dutton et al. 1994), CARSNET (Che et al. 2009), and SONET (Li et al. 2018) monitor the AODs over different regions worldwide. Due to limitations, this widespread ground-based station network cannot provide global coverage of AOD measurements. To overcome this, satellites with large-swath, and near-polar orbits can provide frequent aerosol information over the entire globe (King et al. 1999). The satellite-derived AOD products are retrieved from different sensors such as the “VIIRS (Visible Infrared Imaging Radiometer Suite) (Liu et al. 2014), MERIS (Medium Resolution Imaging Spectrometer) (Kosmale et al. 2017), TOMS (Total Ozone Monitoring-Instrument) (Torres et al. 2002), OMI (Ozone Monitoring Instrument) (Torres et al. 2007), MISR (Multi-angle Imaging Spectroradiometer) (Kahn et al. 2011), SeaWiFS (Sea Viewing-Wide field of View Sensor) (Sayer et al. 2012), MODIS (Remer et al. 2005: Levy et al. 2013). The AOD retrievals from satellites measures radiances based on numerous assumptions related to aerosols type and surface features. Therefore, the satellite-derived product is modeled, and its accuracy is validated by ground-based station data. In addition, climate models can predict the aerosol properties and describe the AOD variability and trends for a certain period. Many climate change assessments, including Intergovernmental Panel on Climate Change (IPCC) assessment reports, have discussed different effects of aerosols on projected climate changes which is based on the results of climate model simulations. For example, the IPCC`s 5th (AR5) and 6th (AR6) assessment reports highlighted the results of Coupled Model Intercomparison Project (CMIP) Phase 5 (CMIP5) and Phase 6 (CMIP6), respectively. Particularly in CMIP6 project, the “global emissions trajectories” are broadly categorized from two major sources i.e., greenhouse Gases (GHG) and aerosols which describe the contributions of these emissions to the global climate system. Even though having uncertainties in climate models, these models are considered a reliable tool to assess long term climate projections and the behavior of AOD over certain regions (Ghan and Schwartz 2007; Shalaby et al. 2015; Gidden et al. 2019; Mortier et al. 2020; Tokarska et al. 2020; Ramachandran et al. 2022; Zhao et al. 2022).
Against this background, we focused on the AOD variability, trends and evaluate the performance of CMIP6 individual models relative to satellite-observation MODIS DTB over the Arabian Peninsula in the present study. MODIS is a key measuring instrument onboard Terra and Aqua satellites that is continuously providing spectral retrievals twice a day (cloud-free conditions) for the last 21 years along with a spatial resolution of 10km². Moreover, it is capable to provide AOD retrievals with good accuracy over dark and dense vegetated areas, whereas the quality of AOD retrieval decreases in the bright and irregular surfaces (e.g., desert, and urban regions). In addition, the Dark Target (DT) algorithm is used to retrieve AODs over vegetated lands and oceans, while the Deep Blue (DB) algorithm is employed over brighter surfaces. Both the DT and DB algorithms advance over time. Several studies investigated the different perspectives of AOD utilizing MODIS DT, DB, and combined DT-DB algorithms over both global and regional scales (Kumar et al. 2018; Ali et al. 2017; Li et al. 2020; Zhao et al. 2022; De Leeuw et al. 2015; Georgoulias et al. 2016; Wang et al. 2017; Pierce et al. 2009). Similarly, there are few studies available that explore the AOD variability, types, and properties over Arabian Peninsula. Shalaby et al. (2015) analyzed the dust aerosol climatology and compare the RegCM4 with ground stationed AERONET and different satellite AOD products for the period 1999 to 2012. Another study discussed the AOD variation in Saudi Arabia (which covers 80% of the Arabian Peninsula) employed MODIS deep blue product from 2000 to 2013 (Butt et al. 2017). Consequently, Kumar et al. (2018) assessed the AOD variability over Arabian Peninsula by using AERONET, MODIS, and MISR from 2003 to 2012. Moreover, the Ali et al. (2017) investigated the annual AOD trends and temporal variation in Arabian Peninsula by utilizing the MODIS combined DT and DB algorithm AOD product during the period 2003 to 2017. Though, the CMIP6 AOD output is less explored over both global and regional domains including AP. There are very few studies published recently that evaluated CMIP6 models AOD output, for example, Li et al. (2020) assessed the 16 CMIP6 simulations with respect to satellite-derived AOD from 2010 to 2019 over China. Furthermore, Ramachandran et al. (2022) examined the AOD trends over Asia using observations and CMIP6 simulations for the period 2000 to 2018. In addition, Zhao et al. (2022) analyzed the performance of CMIP6 dust aerosols simulations relative to CAMS and MERRA2 reanalysis products over the entire globe from 2005 to 2014. However, there is no study available yet that simultaneously evaluates the AOD variability, trends, and CMIP6 simulations against MODIS DTB over AP. Thus, the main objectives of the present work are: (i) to investigate the average behavior of AODs during summer (JJA) for the period 2000 to 2014 (ii) to calculate the long-term spatiotemporal variation and trends (ii) to evaluate the CMIP6 simulations against MODIS DTB.