Chlorophyll-a is a vital water quality parameter used to quantify concentrations of algal biomass in freshwater systems. However, insufficient field data in the Ohio River Basin has resulted in limited understanding of the development of algal blooms. We built a 38-year (1984 – 2022) dataset of satellite derived chlorophyll-a predictions to support research efforts which aim to quantify the frequency and intensity of river algal blooms. We developed our model by leveraging coinciding in situ chlorophyll-a data and surface reflectance extracted from Landsat Collection 2 Tier 1, referred to as matchups. Matchups were used to train and test our machine learning model. We also extracted Landsat surface reflectance over 6,116 NHD river reaches using similar methods. We then applied our model to this reach-level data to create a comprehensive dataset of chlorophyll-a predictions. This dataset includes the following files: 1) data used to train and test the model (matchups), 2) the model infrastructure, 3) satellite derived chlorophyll-a predictions aggregated over NHD river reaches, and 4) a shapefile of NHD river reaches.