Abstract: Wetland ecosystems perform a wide range of critical ecological functions, including regulating regional and global climate, purifying water, and conserving biodiversity. Among the various ecological indicators used to assess wetland health, aboveground biomass (I_AGB) of vegetation has been widely recognized as a fundamental parameter, as it directly reflects vegetation productivity, nutrient cycling, and carbon storage capacity. Accurately estimating integrated aboveground biomass (I_AGB) is therefore of great significance for wetland protection, ecological restoration, and long-term monitoring of ecosystem services. Traditional ground-based measurement methods, although precise at local scales, are constrained by high cost, labor intensity, and limited applicability for capturing the large-scale and temporally dynamic characteristics of wetlands. This challenge highlights the necessity of developing remote sensing-based approaches that can provide spatially explicit, cost-efficient, and repeatable biomass estimates. In this study, we focus on the wetlands of the Yellow River Delta, one of the most ecologically important coastal wetlands in China, characterized by complex hydrological processes, tidal influences, and rapid land-sea interactions. To address the limitations of traditional field surveys, we integrate multi-source satellite remote sensing datasets from Landsat-8 and Sentinel-2 with the computational capacity of the Google Earth Engine (GEE) platform to estimate I_AGB across this dynamic wetland region. Ground-based I_AGB sampling data collected in the Yellow River Delta were used as reference measurements to calibrate and validate biomass prediction models. A comparative analysis of five machine learning algorithms was conducted to assess their predictive performance in estimating I_AGB. These models incorporated a diverse set of predictors, including raw spectral bands, vegetation indices, and water-related indices, which are particularly relevant in wetland ecosystems due to the strong influence of hydrological conditions on vegetation distribution and growth. The results revealed significant spatial variation in I_AGB across different wetland regions. Specifically, in the Diaokou River area, mean I_AGB values ranged from 112.05 to 170.09 g/m², while in the Qingshuigou area, mean I_AGB values ranged from 114.72 to 165.47 g/m². Across the broader Yellow River Delta wetlands, I_AGB values ranged from 0 to 746.00 g/m², illustrating the wide heterogeneity of biomass distribution under varying environmental conditions. Spatially, areas with low I_AGB values (≤ 304.00 g/m²) were mainly distributed along tidal flats near the sea, where vegetation coverage is sparse due to high salinity and frequent tidal inundation. Medium I_AGB regions (304.00 g/m² < I_AGB≤ 436.00 g/m²) were the most widespread, representing zones of moderate vegetation productivity. In contrast, high I_AGB values (> 436.00 g/m²) were primarily observed along riverbanks, reflecting favorable hydrological conditions, nutrient availability, and soil moisture that support dense vegetation growth. These findings not only provide new insights into the spatial distribution of wetland vegetation biomass but also underscore the ecological drivers of biomass variability, particularly the influence of proximity to the sea and rivers. Importantly, this study demonstrates the effectiveness of integrating multi-source remote sensing data with advanced machine learning algorithms for biomass estimation in wetland ecosystems. By combining spectral and index-based variables with ground-truth data, the study improves the accuracy and reliability of biomass predictions. Moreover, the methodological framework developed here shows strong potential for application beyond the Yellow River Delta, as it can be adapted to other wetland regions worldwide for biomass estimation, ecological assessment, and carbon stock evaluation. In conclusion, this research highlights the importance of accurate I_AGB estimation for understanding the ecological functions of wetlands, including carbon sequestration, biodiversity conservation, and ecosystem service provision. The results provide valuable scientific evidence to inform wetland ecological protection and restoration strategies in the Yellow River Delta, while also supplying essential data for wetland carbon storage assessment and climate change research. Moreover, the study demonstrates that integrating remote sensing technologies, machine learning approaches, and field-based validation constitutes a powerful means of advancing wetland monitoring and management. By linking large-scale remote observations with detailed field measurements, this research makes a significant contribution to both the theoretical understanding and practical application of biomass estimation in wetland ecosystems, ultimately supporting sustainable wetland management and global climate change mitigation.