Evaluation of the arbitrary Lagrangian–Eulerian vertical coordinate method in the MPAS-Ocean model

Abstract

The vertical coordinate of the Model for Prediction Across Scales-Ocean (MPAS-Ocean) uses the Arbitrary Lagrangian-Eulerian (ALE) method, which offers a variety of configurations. When fully Eulerian, the vertical coordinate is fixed like a z-level ocean model; when fully Lagrangian there is no vertical transport through the interfaces so that the mesh moves with the fluid; additional options for vertical coordinates exist between these two extremes, including z-star, z-tilde, sigma, and isopycnal coordinates. Here we evaluate spurious diapycnal mixing in MPAS-Ocean in several idealized test cases as well as real-world domains with full bathymetry. Mixing data is compared to several other ocean models, including the Parallel Ocean Program (POP) z-level and z-star formulations. In three-dimensional domains, MPAS-Ocean has lower spurious mixing that other ocean models. A series of simulations show that this is likely due to MPAS-Ocean's hexagon-type horizontal grid cells combined with a flux-corrected transport tracer advection scheme designed for these unstructured meshes. The frequency-filtered vertical coordinate of Leclair and Madec (2011) (also called z-tilde) has been implemented and analyzed in MPAS-Ocean. This addition allows low-frequency vertical transport to pass through the vertical interface in an Eulerian manner, while high-frequency vertical oscillations, such as internal gravity waves, are treated in a Lagrangian manner.

OMIP contribution to CMIP6: experimental and diagnostic protocol for the physical component of the Ocean Model Intercomparison Project

“…However, the alternative paradigms of terrain-following coordinates and isopycnal coordinates are gaining traction in the climate community, along with even more general features offered by the arbitrary Lagrangian-Eulerian (ALE) method (e.g. Bleck, 2002;Donea et al, 2004;Adcroft and Hallberg, 2006;Petersen et al, 2015). Following the CMIP5 protocol, the CMIP6 ocean diagnostics should be regridded to depth/pressure vertical levels for those models not based on one of the vertical coordinates z, z * , p, or p * , with details of these vertical coordinates provided in Appendix C2.…”

“…Studies such as Gregory (2000), Piecuch andPonte (2011), Palter et al (2014), Buckley et al (2015), Griffies et al (2015), Exarchou et al (2015), Kuhlbrodt et al (2015), and Morrison et al (2016) illustrate the value for physical interpretation provided by budget terms. The bulk of the science resulting from these budgets has thus far come from use of the annual mean fields, hence we request annual means, with these fields being Priority = 1 for FAFMIP (Gregory et al, 2016) and Priority = 3 for other experiments.…”

In Situ Eddy Analysis in a High-Resolution Ocean Climate Model

“…It uses unstructured horizontal meshes based on Voronoi tessellations, which has applications in regional climate modeling [31]. The vertical grid is arbitrary Lagrangian-Eulerian to reduce spurious mixing [29]. The oceanclimate module, MPAS-Ocean is developed at Los Alamos National Laboratory and it has been shown to have nearly linear scalability to the 15 km grid cell resolution needed to resolve eddies, on many thousands of cores, seen in Figure 3.…”

“…The oceanclimate module, MPAS-Ocean is developed at Los Alamos National Laboratory and it has been shown to have nearly linear scalability to the 15 km grid cell resolution needed to resolve eddies, on many thousands of cores, seen in Figure 3. MPAS-Ocean has been validated against analytic solutions and other models in numerous test cases, from idealized to realistic [29]. It has been shown to produce an ocean climate similar to observations, and with performance similar to other ocean models [31].…”

Research priorities in support of ocean monitoring and forecasting at the Met Office

“…This neatly allows the model to limit the amount of adaptation required whilst removing the primary source of spurious vertical mixing. Petersen et al (2015) described the impact of a range of vertical coordinates, including fully ALE and the z subset of ALE and concluded that the use these coordinates worked well at reducing spurious mixing under many scenarios. These formulations should be introduced into operational configurations once they are sufficiently mature.…”