10:00 AM - 10:15 AM
[ACG44-04] Geostationary-Satellite-Based Study of Diurnal Variation in Aerosol-Cloud Interactions
Keywords:Himawari-8/9, Aerosol-Cloud Interactions, Twomey Effect, Diurnal Variation
The latest climate assessment report identifies aerosol-cloud interactions (ACI) as one of the largest sources of uncertainty in current climate evaluations, a situation largely attributable to the inaccuracies in observational estimates of ACI. Presently, ACI estimates primarily rely on polar-orbiting satellites (e.g., MODIS), whose observational characteristics inherently limit the data to represent the cloud and aerosol conditions at specific times within the observed region (e.g., at 10:30 and 13:30 local time (LT)). However, the impact of diurnal variations in aerosols and clouds on ACI estimates, the underlying physical mechanisms, and the potential for using these variations to refine existing quantitative estimates have yet to be systematically investigated. In this context, this study utilizes the geostationary Himawari-8/9 satellite (2016–2023, 10-minute temporal resolution, 5 km spatial resolution) to examine the impact of diurnal variations in clouds and aerosols on ACI estimates. A critical variable in estimating ACI is the sensitivity of cloud droplet number concentration (Nd) to variations in the aerosol index (AI), named SNd-AI, which serves as a key parameter linking anthropogenic aerosol changes to cloud radiative properties. Therefore, this study primarily investigates the influence of diurnal variations on SNd-AI.
Given that Himawari-8 observations are based on visible light, data from 08:00 LT to 18:00 LT within the 30°S–30°N and 90°E–150°E region are utilized. The findings reveal that SNd-AI exhibits significant diurnal variation (-0.002 to 0.152), with land regions showing a notably stronger variation (-0.068 to 0.065) compared to oceanic regions (0.146 to 0.203). When comparing SNd-AI values derived from the daytime with those corresponding to MODIS overpass times (10:30 LT and 13:30 LT), which are 0.109, 0.128, and 0.135, respectively, it is evident that relying solely on a single time point for calculating SNd-AI introduces a positive bias of 17%–24% in ACI estimates, thereby overestimating the cooling effect of ACI on global warming. Mechanistically, the study shows that both Nd and AI exhibit pronounced diurnal variability, and the changes are non-coherent, jointly contributing to the observed variation in SNd-AI. Leveraging ERA5 reanalysis data, the study further investigates the relationship between environmental variable changes and SNd-AI with preliminary results suggesting that variations in SNd-AI are influenced by entrainment mixing, land-ocean transport, and convective transport. To incorporate these findings into ACI estimates, two approaches are proposed: one involves applying a correction factor (e.g., 117% or 124%) to the existing ACI estimates, thus extending the results from specific times to the entire daytime. The other approach directly calculates SNd-AI by considering full-day global data, utilizing the newly compiled global geostationary satellite dataset by our research group. Relevant results will be presented at an upcoming conference.
In conclusion, this study quantitatively evaluates the impact of diurnal variations in cloud/aerosol properties on ACI estimates using the Himawari-8/9 observations, explores the underlying causes of this impact, and proposes methods to incorporate diurnal effects into global ACI assessments, a step that is critical for reducing uncertainty in climate evaluations.
Given that Himawari-8 observations are based on visible light, data from 08:00 LT to 18:00 LT within the 30°S–30°N and 90°E–150°E region are utilized. The findings reveal that SNd-AI exhibits significant diurnal variation (-0.002 to 0.152), with land regions showing a notably stronger variation (-0.068 to 0.065) compared to oceanic regions (0.146 to 0.203). When comparing SNd-AI values derived from the daytime with those corresponding to MODIS overpass times (10:30 LT and 13:30 LT), which are 0.109, 0.128, and 0.135, respectively, it is evident that relying solely on a single time point for calculating SNd-AI introduces a positive bias of 17%–24% in ACI estimates, thereby overestimating the cooling effect of ACI on global warming. Mechanistically, the study shows that both Nd and AI exhibit pronounced diurnal variability, and the changes are non-coherent, jointly contributing to the observed variation in SNd-AI. Leveraging ERA5 reanalysis data, the study further investigates the relationship between environmental variable changes and SNd-AI with preliminary results suggesting that variations in SNd-AI are influenced by entrainment mixing, land-ocean transport, and convective transport. To incorporate these findings into ACI estimates, two approaches are proposed: one involves applying a correction factor (e.g., 117% or 124%) to the existing ACI estimates, thus extending the results from specific times to the entire daytime. The other approach directly calculates SNd-AI by considering full-day global data, utilizing the newly compiled global geostationary satellite dataset by our research group. Relevant results will be presented at an upcoming conference.
In conclusion, this study quantitatively evaluates the impact of diurnal variations in cloud/aerosol properties on ACI estimates using the Himawari-8/9 observations, explores the underlying causes of this impact, and proposes methods to incorporate diurnal effects into global ACI assessments, a step that is critical for reducing uncertainty in climate evaluations.