Synergistic fusion of aerosol parameters from multi-sensor measurements (satellite and ground-based) is crucial for integrating diverse data sources and generating spatially consistent representations of aerosol distribution for accurate climate impact assessment. In this study, a two-stage Universal Kriging (UK) framework is employed. In the first stage, UK is used for spatial interpolation to fill missing values in individual satellite datasets (MODIS and MISR). In the second stage, Kriging is formulated as a fusion model by incorporating spatial covariance structures derived from variogram models of the satellite data, thereby producing fused AOD estimates from both satellite and ground-based (ARFINET) observations. Following this, seasonal fused AOD maps are generated for winter, pre-monsoon, and post-monsoon periods. Leave-one-out cross-validation (LOOCV) shows that the 95% confidence interval (±2σ) of the fused AOD values accommodate more than 80% of the ground-based observations, effectively capturing regional variations. This also highlights the influence of number of ground measurement points in the generation of fused map. To address this, a Residual Kriging with Machine Learning (RK-ML) approach is explored. The RK-ML framework captures stable spatial patterns and yields LOOCV scores comparable to those of the UK method, even under sparse ground-based coverage. These findings demonstrate the suitability of both UK and RK-ML approaches (with adequate ground-based observations) for producing reliable and near-instantaneous fused AOD fields over the Indian region.