Paleoclimate data assimilation with CLIMBER-X: An ensemble Kalman filter for the last deglaciation

Abstract

Using the climate model CLIMBER-X, we present an efficient method for assimilating the temporal evolution of surface temperatures for the last deglaciation covering the period 22000 to 6500 years before the present. The data assimilation methodology combines the data and the underlying dynamical principles governing the climate system to provide a state estimate of the system, which is better than that which could be obtained using just the data or the model alone. In applying an ensemble Kalman filter approach, we make use of the advances in the parallel data assimilation framework (PDAF), which provides parallel data assimilation functionality with a relatively small increase in computation time. We find that the data assimilation solution depends strongly on the background evolution of the decaying ice sheets rather than the assimilated temperatures. Two different ice sheet reconstructions result in a different deglacial meltwater history, affecting the large-scale ocean circulation and, consequently, the surface temperature. We find that the influence of data assimilation is more pronounced on regional scales than on the global mean. In particular, data assimilation has a stronger effect during millennial warming and cooling phases, such as the Bølling-Allerød and Younger Dryas, especially at high latitudes with heterogeneous temperature patterns. Our approach is a step toward a comprehensive paleo-reanalysis on multi-millennial time scales, including incorporating available paleoclimate data and accounting for their uncertainties in representing regional climates.

Fig 1. Online DA.

Schematic view of our DA system for the first two cycles.

Fig 2. Effect of DA using different radius.

Effect of DA increment at first analysis step on ST field using 2500, 5000, 7500, and 10000 km radius.

Fig 3. Ensemble members before DA.

Ensemble members (coloured lines) and ensemble mean before DA in (a) Exp_GLAC1D and (b) Exp_PaleoMist.

Fig 4. GMST from free runs of the last deglaciation and comparison with the proxy-based reconstruction.

a) GMST from free runs of the last deglaciation using the different ice sheet reconstructions, GLAC-1D and PaleoMist. b) Comparison of GMST anomaly from the early Holocene of our free runs compared to [9].

Fig 5. ΔGMST calculated by raw averaging over the observation locations.

Mean surface temperature change (ΔMST) calculated by averaging over the proxy locations for (a) Exp_GLAC1D and (b) Exp_PaleoMist. The orange, green, and black lines illustrate trajectories for observation, DA ensemble, and prior ensemble means, respectively.

Fig 6. Climate variables trajectories in Exp_GLAC1D.

North Atlantic FW (a), SSS (b), AMOC at 26° north (c), and ΔGMSTs based on the model field (d) for Exp_GLAC1D. Different colours in (a), (b), and (c) correspond to ensemble members. The red line in (d) represents ΔGMST for the free run. The orange line in (d) is the proxy reconstruction of ΔGMST by [9] projected to 5° × 5° resolution. The blue line in (d) is the DA-based reconstruction of ΔGMST by [25].

Fig 7. Climate variables trajectories in Exp_PaleoMist.

North Atlantic FW (a), SSS (b), AMOC at 26° north (c), and ΔGMSTs based on the model field (d) for Exp_PaleoMist. Different colours in (a), (b), and (c) correspond to ensemble members. The red line in (d) represents ΔGMST for the free run. The orange line in (d) is the proxy reconstruction of ΔGMST by [9] projected to 5° × 5° resolution. The blue line in (d) is the DA-based reconstruction of ΔGMST by [25].

Fig 8. Effect of DA on ST fields in Exp_GLAC1D and Exp_PaleoMist.

ΔST anomaly (DA minus Prior) in LGM, BA, and YD. (a), (b), and (c) show anomaly for Exp_GLAC1D and (d), (e), and (f) for Exp_PaleoMist. The green dots indicate the observation locations.

Fig 9. Comparison of ST anomalies in Exp_GLAC1D before and after DA.

ΔST anomalies field for different time intervals in Exp_GLAC1D for before DA’s implementation (a), (b), and (c) and after DA (d), (e), and (f).

Fig 10. Comparison of ST anomalies in Exp_PaleoMist before and after DA.

ΔST anomalies field for different time intervals in Exp_Paleomist for before DA’s implementation (a), (b), and (c) and after DA (d), (e), and (f).