In this paper, we define the method that is used to merge high-resolution multisensor chlorophyll-a (chl-a) data on the Ireland-Biscay-Iberia Regional Ocean Observing System area from 1998 to the present at a resolution of 1.1 km. The method is based on geostatistics and is known as kriging. The merged variable is the daily anomaly of chl-a, with the anomaly being defined as the difference between the daily image and the mean historical field for the considered day. For each day, the continuous anomaly image is generated using the kriging method, and the mean historical field is then added to obtain the cloudless field of chl- a. The initial satellite chl-a data set used in the merging procedure is derived from the daily level-2 water leaving radiances of three ocean color sensors: the Sea-Viewing Wide Field of View Sensor on the Orbview platform, the Moderate Resolution Imaging Spectroradiometer on the Aqua platform, and the Medium Resolution Imaging Spectrometer Instrument on the ENVISAT platform. The chl-a concentration is obtained using a specific algorithm developed by Ifremer, known as OC5 product. Before merging, each satellite-derived chl-a product has been compared to in situ data and has been validated using a matchup data set. After this validation against in situ data, intercomparisons between the satellite data sets have been performed. As the chl-a anomaly variability depends, in this region, on the season and the distance from the shore, local space-time semivariograms have been calculated to estimate the spatiotemporal dependence or covariance of the chlorophyll anomalies. The semivariograms, used in the estimation of the kriged anomaly, are defined by their nuggets (noise), their spatial and temporal range (maximum distance for a nonnull covariance between the anomalies), and their sill (maximum variance). The spatial range of the semivariogram has been approximated locally on a regular grid. The nuggets and the sills have been deduced from the squa- - re of the mean of the chl-a concentration (the climatological reference) as we have observed a classical proportionality effect between the square of the chl-a mean, the variance of the distribution, and the parameters of the semivariograms. Compared to each original product, the analysis shows a complete coverage and differences with the in situ data that are statistically equivalent to those observed with the initial satellite data set. The merged product offers a number of applications for environmental monitoring such as the monitoring of the eutrophication risk required by the Water Framework Directive of the European Union.