Evaluating the performance of PATMOS-x in global and regional analyses of clouds as they relate to Earth’s planetary albedo

Abstract

In 2023, the Clouds and the Earth’s Radiant Energy System - Energy Balanced and Filled (CERES-EBAF) dataset revealed a record low planetary albedo. Due to the dominant role of clouds in Earth’s albedo, we hypothesize that the negative anomaly identified in 2023 is mostlikely a consequence of a similar anomaly in cloud fraction and, subsequently, effective reflectance. Since the CERES-EBAF dataset cannot be used for in-depth analyses of cloud behavior, as it is limited by a narrow selection of cloud property variables, the visible reflectance parameter featured in the Pathfinder Atmospheres-Extended (PATMOS-x) v6.0 dataset – a 1980-present global satellite record with a primary focus on cloud properties – is proposed as asupplement for investigating the relative contribution of clouds to the 2023 albedo anomaly.However, because PATMOS-x incorporates data from Advanced Very-High Resolution Radiometers (AVHRRs) flown on sixteen separate satellite platforms, each of which has its own set of calibration coefficients, local overpass times, and degrees of orbital drift, synthesizing a stable, long-term record of global visible reflectance is not possible without radiometric normalization and frequent updates to the intercalibration. Presented here is a novel approach to validating the most recent PATMOS-x v6.0 visible reflectance calibration, wherein tropical deep convective clouds (DCCs) are used as climatologically predictable targets to evaluate reflective stability over time and develop a solar zenith angle-dependent normalization method for handling the impacts of orbital drift and local overpass time discrepancies. While a robust validation of the record’s visible intercalibration is obtained in a strictly-controlled context, the normalization methodology applied to DCCs in the Tropics does not necessarily handle orbital drift and local overpass time discrepancies to an adequate extent when it is applied (a) to a broader array of cloud types, (b) on a global scale, and (c) for the entire duration of the record. Alternatively, in a case study of the Central Indian Ocean,the PATMOS-x record demonstrates a reasonable capacity to produce accurate ocean-based regional analyses of cloud effective reflectance between 2000 and 2023. Impacts owed to orbital drift and local overpass differences are better-resolved by the solar zenith angle normalization when analysis is centered over a relatively localized oceanic region; it is, thus, more feasible to draw conclusive results of cloud fraction and, consequently, effective reflectance under this circumstance. It is ultimately determined that, over the Central Indian Ocean, the local planetary albedo anomaly in 2023 is the result of a pronounced negative anomaly in upper-level cloud fraction, primarily contributed to by clouds in the 3.6-23 range of optical depth and ~30-45% range of mean reflectance.