Understanding the variability of global land aerosol optical depth (AOD) and its teleconnection with sea surface temperature (SST) is essential for assessing aerosol–climate interactions. This study developed a high-resolution (0.05°), gap-free daily AOD dataset for 2000–2023 by integrating MAIAC AOD retrievals with meteorological, reanalysis, and geographical predictors using a CatBoost model. Validation against independent AERONET observations shows good agreement (R² = 0.72, RMSE = 0.09), with significantly improved spatial completeness compared to reanalysis products. Empirical Orthogonal Function (EOF) analysis indicates that the leading mode exhibits predominantly positive loadings over most global land areas, with the strongest signals in Northern Hemisphere mid- and high latitudes. The associated temporal evolution shows a shift from above-average AOD in the early 2000s to below-average levels after the mid-2010s, consistent with widespread AOD declines reported since the early 2000s and most pronounced in high-loading regions. Singular Value Decomposition (SVD) analysis between monthly land AOD and SST reveals that the leading coupled mode (36.28 % covariance) is characterized by persistent SST warming and a marked reduction in AOD over major emission regions since the early 2010s, while subsequent modes capture interannual signals related to El Niño-Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO) as well as decadal-scale patterns. These findings provide new evidence of large-scale SST–AOD covariability and offer a basis for exploring potential physical linkages, contributing to improved representation of aerosol–climate interactions in coupled models.