NorCPM1 (Bethke et al., 2021; https://doi.org/10.5194/gmd-14-7073-2021) is a CMIP6-type global climate model (GCM) designed for generating climate reanalyses and conducting seasonal-to-decadal climate predictions. It integrates the Norwegian Earth System Model version 1 (NorESM1) with CMIP6 external forcings and utilizes the ensemble Kalman filter (EnKF) data assimilation technique. NorESM1 features an isopycnic-coordinate ocean component that includes biogeochemistry. The EnKF (Evensen, 2003; https://doi.org/10.1007/s10236-003-0036-9) is one of the most advanced data assimilation methods, offering uncertainty quantification through Monte-Carlo ensembles. The forecast error covariances, which propagate information from observed to non-observed variables, are flow-dependent and evolve with the climate system's variability. NorCPM1 uses anomaly assimilation to prevent prediction drift caused by changes in model climatology. The atmospheric and land components have a horizontal resolution of 1.9° latitude and 2.5° longitude, with the atmosphere consisting of 26 hybrid sigma-pressure levels up to 3 hPa, indicating that the stratosphere is not fully resolved. The ocean and sea ice components have a horizontal resolution of about 1°. The ocean component includes 51 isopycnic vertical layers and a bulk mixed layer that represents two layers with dynamically changing thicknesses and densities.

This dataset features a prediction-like pacemaker simulation using NorCPM1, which imposes the development and evolution of the cold and fresh anomaly in the subpolar gyre (SPG) region throughout the pacemaker period. The "pacemaking" is applied from initialization on October 15, 2014, through December 31, 2019. After this period, the pacemaker simulation continues without restoring or assimilation until December 31, 2024. The dataset includes 30 members. All ocean physical state variables in the SPG are constrained by assimilating (i) temperature and salinity anomalies from EN4 Objective Analysis (only data below the mixed layer with an observation weight greater than 0.8 are assimilated) and (ii) SST anomalies from OISSTV2. During each assimilation time step, observations update both observed and unobserved variables using the EnKF method, with forecast error covariances estimated from the ensemble members (Bethke et al., 2021). The climatology reference period is 1980-2010 for temperature and salinity, and 1982-2010 for SST. The pacemaker region is defined as [0-80°W, 30-60°N] with 5° linear transition zones to zero at the southern and northern boundaries.