The 2008 hurricane season, one of the most active on records, as simulated by the FV3-powered GFDL model at 13-km resolution. FV3 improves representation of small-scale weather features such as hurricanes while maintaining the quality of large-scale global circulation. Image from NOAA.
The NOAA CPO Modeling, Analysis, Predictions, and Projections (MAPP) program hosted a webinar on the topic of High resolution modeling: Working toward improved process representation and simulations of the Earth system on Monday, March 13, 2017. The announcement is provided below.
Jim Kinter - In the nearly 9 years since the 2008 World Modeling Summit, at which it was proposed to initiate a major effort to develop and test much higher resolution global climate models, there have been attempts by several different groups to do just that. These efforts have been directed at simulation, projection and prediction. While convection-permitting global models are only beginning to be developed, weather-resolving and eddy-permitting models have been exercised in several large-scale experiments. There are some points that are common across the several results the various groups have obtained. Lessons learned from these experiments and potential avenues for further study will be described.
Justin Small - This presentation describes recent experiences with high resolution climate modeling at NCAR. Firstly, a long coupled simulation with 0.1deg ocean and 0.25deg atmosphere showed improved representation of coastal upwelling zones, western boundary currents, ENSO and Southern Ocean deep mixed layers. Comparable runs with lower resolution ocean or atmosphere are used to attribute these improvements. Secondly, new efforts of data assimilation with a high resolution ocean model are described, in the context of a broader project on decadal climate prediction.
Eric Swenson - The occurrence of boreal winter Rossby wave breaking (RWB) along with quantitative role of synoptic transient eddy momentum and heat fluxes directly associated with RWB are examined during the development of Euro-Atlantic circulation regimes using the ERA Interim Reanalysis. Results are compared to those from seasonal re-forecasts made using the Integrated Forecast System model of ECWMF (at T319, T639 and T1279 spectral resolution) coupled to the NEMO ocean model. The development of both Scandinavian Blocking and the Atlantic Ridge is directly coincident with anticyclonic wave breaking (AWB), however the associated transient eddy fluxes do not contribute to (in fact oppose) ridge growth, as indicated by the local Eliasson-Palm (EP) flux divergence. Evidently other factors drive development, and it appears that wave breaking assists more with ridge decay. The growth of the North Atlantic Oscillation (NAO) in its positive phase is independent of RWB in the western Atlantic, but strongly linked to AWB further downstream. During AWB, the equator-ward flux of cold air at upper-levels contributes to a westerly tendency just as much the poleward flux of momentum. The growth of the negative phase of the NAO is almost entirely related to cyclonic wave breaking (CWB), during which equator-ward momentum flux dominates at jet-level, yet low-level heat fluxes dominate below. The re-forecasts yield realistic frequencies of CWB and AWB during different regimes, as well as realistic estimates of their roles during development. However, a slightly weaker role of RWB is simulated, consistent with a modest underestimation in RWB frequency.
Francisco Doblas-Reyes - The European SPECS project has delivered a new generation of climate forecast systems, with improved forecast quality and efficient regionalization tools to produce reliable, local climate information over land at seasonal-to-decadal time scales. The improved understanding of the sources of predictability have offered better estimates of the future frequency of high-impact, extreme climatic events. The project has brought into the climate prediction field knowledge from the weather forecasting and climate change communities to leverage both knowledge and model developments that boost the forecast quality. Driven by needs identified by a sister project on climate services, EUPORIAS, SPECS provided an enhanced communication protocol and services to satisfy the climate information needs of a wide range of public and private stakeholders. This has also become a coordinated European response to some of the GFCS components.
Lucas Harris - The increasing demands for high-resolution weather and climate models require new models that can produce useful simulation of smaller-scale motions while also making skillful predictions of the driving larger-scale dynamics. The GFDL Finite-Volume Cubed-Sphere dynamical core (FV3) has been developed to allow the development of new models able to unify global and limited-area prediction. FV3 is capable of non-hydrostatic modeling to allow explicit representation of deep convection, and also has grid-nesting and grid-stretching capabilities to be able to efficiently refine the model grid over areas of interest. This dynamical core is being used in several models---GFDL HiRAM, AM4, and fvGFS, NASA GEOS, and elsewhere---for a range of applications, from coarse-resolution climate modeling to convection-resolving weather forecasts, and is being incorporated as the replacement for the spectral core in GFS.
Results are demonstrated for several forecast applications. We show preliminary subseasonal predictions of TC activity in HiRAM nested to 8-km over the North Atlantic. Several variable-resolution configurations able to explicitly resolve convective clouds are being developed, with a focus upon severe weather and hurricane forecasts. Forecasts of high-impact weather events, such as the 2012 Derecho and Hurricane Matthew, are presented. We also show some early results from global convection-resolving fvGFS and GEOS forecasts, and discuss the feasibility of a global cloud-scale forecast model within the next decade.
Ben Kirtman - There is a continually increasing demand for near-term (i.e., lead times of 2-4 weeks up to a couple of decades) climate information. This demand is partly driven by the need to have robust forecasts to support adaptation and response strategies, and is partly driven by the need to assess how much of the ongoing climate change is due to natural variability and how much is due to anthropogenic increases in greenhouse gases or other external factors. Here we discuss results from a set of state-of-the-art climate model prediction and predictability experiments in comparison with observational estimates that show that an assessment of predictability, and indeed, robust predictions from days to decades, requires models that capture the variability of major oceanic fronts and ocean eddies, which are, at best, poorly resolved and may even be absent in current sub-seasonal to interannual prediction systems, and in the decadal climate predictions experiments made as part of Intergovernmental Panel on Climate Change.
Poll Questions Results (click the gallery thumbs below)