Coastal ecosystems are among the most biologically and biogeochemically active and diverse systems on Earth. Because they act as important linkages between terrestrial ecosystems and the open ocean, their incorporation in Earth system models (ESMs) is critical to predicting coastal and global responses to environmental changes. However, they vary greatly in the magnitude of tides and the volume and timing of freshwater input from land, making it challenging to model the major biogeochemical reactions that control productivity and greenhouse gas emissions across coastal terrestrial aquatic interfaces. Here, we test a land surface model coupled to a biogeochemical model (ELM-PFLOTRAN) to predict marsh biogeochemical reactions as well as CH₄ and CO₂ fluxes along an estuarine salinity gradient. The model has been validated for fresh and saline endmembers, and we further assess the model’s sensitivity to temporally varying salinity in intermediate ranges on marsh processes. The main input variable, surface water salinity, varies primarily as a function of discharge along the estuary. We use sensor data collected along the Parker River (MA) from a ‘normal’, ‘extreme dry', and ‘extreme wet’ year to drive the model, which captures differences in timing, duration, and magnitude of saltwater intrusion along the estuary. ‘Normal’ years are characterized by high spring and low summer discharge, interspersed by rainfall events. Consequently, river water salinity at each location along the estuary varies by 10–15 psu in a normal year but varies more strongly under extreme conditions. Porewater measurements, used for model validation, indicate a clear seasonal increase from 0–12 ppt in salinity in the upper part of the estuary, while the more saline marsh displays relatively constant values of about 30 ppt. In addition, we assessed the model’s capability to predict water level variation during non-flooding neap tides. Predicted fluxes were validated against three Ameriflux sites and recently updated thresholds of porewater salinity, above which CH4 fluxes decrease substantially (e.g., 21psu, Arias-Ortiz et al., accepted). Finally, we will make first assessments on the capability of ELM-PFLOTRAN to predict marsh productivity for varying salinities.