Accelerated Climate Modeling for Energy (ACME): Ocean and Sea-Ice Processes
Project Team
Principal Investigator
A consortium of Department of Energy (DOE) laboratories have proposed to develop an Earth System modeling capability to be responsive to DOE Office of Science strategic objectives as well as broader climate change needs. The development of the Accelerated Climate Modeling for Energy (ACME) project is designed to be responsive to a suite of climate science drivers (see ACME proposal abstract) that aim, among others, to improve our understanding of how (1) the hydrological cycle and water resources, (2) biogeochemical cycles, and (3) rapid changes in cryospheric systems interact with the climate system. SIO proposes to partner with DOE Laboratory investigators to address questions related to the roles of the ocean and the cryosphere (sea-ice in the first instance) in the climate system produced by the new Earth System Model (ESM) in which an eddy-active ocean model is coupled to a weather-scale atmospheric model.
We propose to carry out assessments of the veracity of the ocean and sea-ice components at all development stages using a suite of performance metrics and analyses that were developed by the PI over the past decade. Emerging data sets will be included in the suite of validation analyses as they become available. We will provide alpha/beta testing of the climate model diagnostics package: UV-CDAT that is under further development as part of the DOE Laboratory effort. We will provide our analysis codes for transition into this package. The focus of our analyses will be on the representation of air-sea and upper-ocean processes responsible for the uptake of heat and carbon in the eddy active ocean component of the ESM. As well, we will examine atmosphere-ocean/sea-ice interactions in the marginal ice zones of both hemispheres and relate them to recent climate change and variability. Once land ice is implemented in the ESM, we will investigate ocean/sea-ice/land ice interactions around Greenland and over the Antarctic shelf, again in the context of climate change over the past decades.
The primary objective of this renewal project is to complete a study that evaluates the relative merits of two approaches used to initialize fine resolution present day transient climate model simulations representing the latter half of the 20th century. Unlike standard climate models, the horizontal resolution (0.1° ocean/sea-ice and 0.25° atmosphere/land) used here explicitly simulates oceanic mesoscale eddies and frontal structures in most of the global ocean, fine mean oceanic flows, atmospheric storms in mid- and high latitudes, and tropical cyclones. This increased realism is expected to lower uncertainty in subsequent climate projections. Standard initialization methodology involves first carrying out a long multi-century simulation under preindustrial conditions followed by an ensemble of transients from 1850 to the present day. This approach is computationally prohibitive at high resolution; so instead, we contrast an ensemble of three transients initialized from atmospheric reanalysis-forced 0.1° ocean/sea-ice states, and a fourth transient branched off a fine-resolution ~100-year 1850 preindustrial control simulation.
We evaluated upper ocean and sea-ice bias and drift in the transient simulations and determined the processes responsible for the most significant biases at the end of the 20th Century. We also considered the veracity of the simulations from which the initial conditions were drawn. Together these findings guide our understanding of the relative merits and weaknesses of the two initialization approaches. Unrealistically strong Southern Ocean winds led to entrenched warm sea surface temperature (SST) biases and excessive ocean ventilation of the lower latitudes of the southern hemisphere in all transients regardless of their initial conditions. However, in the high latitude North Atlantic Ocean, cold SST and shallow mixed layer biases in the forced ocean/sea-ice simulation were inherited by the transient ensemble, in strong contrast with the overall warm SST biases in the transient initialized from the preindustrial simulation. It is proposed to complete the study by calculating changes in the high-latitude North Atlantic Oceans of the transient ensemble between the end of the 20th Century and some 10 to 15 years later to determine whether initial condition influence is waning with time and whether the climate states of the ensemble are trending towards that of the transient initialized from the preindustrial simulation. Finally, the manuscript describing all these results will be completed.
It is also proposed to complete a second manuscript that examines the role of upper ocean vertical stratification, particularly ocean barrier layer thickness (BLT), in the southeast Indian Ocean on rainfall over the Indonesian-Australian Continent. The BLT relationship has not been explored in the past. It is proposed to complete the study by gaining a better understanding the processes governing the seasonal variability of the BLT to the west of Sumatra.