Chlorophyll-a concentration (Chl-a) is used as a proxy for phytoplankton biomass to track blooms and monitor the state of the ecosystem. This dataset contains daily satellite Chl-a since Sep 1997. The data is at a 4.64x4.64 km spatial resolution covering an area of the Northwest Atlantic from 39-82°N, 42-95°W. Regionally-tuned coefficients for POLY4 Chl-a (Clay et al., 2019, see resources) were obtained by matching level-3 gridded 1km-resolution satellite pixels to in situ HPLC (High Performance Liquid Chromatography) Chl-a samples within the region of interest. POLY4 Chl-a was then optimized by forcing the linear regression of POLY4 against HPLC to the 1:1 line to reduce the negative bias observed in higher concentrations in the default global Chl-a models. Daily Chl-a images were generated by applying the optimized POLY4 algorithm on level-3 binned remote sensing reflectance (Rrs) data from the OC-CCI multisensor product generated by the Plymouth Marine Laboratory (PML) as part of the European Space Agency (ESA) climate change initiative. The Chl-a data were then projected onto an Equidistant Cylindrical grid (4.64x4.64 km per pixel) using a nearest neighbour method. This model and the Northwest Atlantic coefficients are available in the R package oceancolouR (https://github.com/BIO-RSG/oceancolouR).