Predictability of weather and climate

Abstract:

Programmable hardware, in particular Field Programmable Gate Arrays (FPGAs), promises a significant increase in computational performance for simulations in geophysical fluid dynamics compared with CPUs of similar power consumption. FPGAs allow adjusting the representation of floating-point numbers to specific application needs. We analyze the performance-precision trade-off on FPGA hardware for the two-scale Lorenz '95 model. We scale the size of this toy model to that of a high-performance computing application in order to make meaningful performance tests. We identify the minimal level of precision at which changes in model results are not significant compared with a maximal precision version of the model and find that this level is very similar for cases where the model is integrated for very short or long intervals. It is therefore a useful approach to investigate model errors due to rounding errors for very short simulations (e.g., 50 time steps) to obtain a range for the level of precision that can be used in expensive long-term simulations. We also show that an approach to reduce precision with increasing forecast time, when model errors are already accumulated, is very promising. We show that a speed-up of 1.9 times is possible in comparison to FPGA simulations in single precision if precision is reduced with no strong change in model error. The single-precision FPGA setup shows a speed-up of 2.8 times in comparison to our model implementation on two 6-core CPUs for large model setups.

Abstract:

Abstract. Data from the High Resolution Dynamics Limb Sounder (HIRDLS) instrument on NASA's Aura satellite are used to investigate the relative numerical variability of observed gravity wave packets as a function of both horizontal and vertical wavenumber, with support from the Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) instrument on TIMED. We see that these distributions are dominated by large vertical and small horizontal wavenumbers, and have a similar spectral form at all heights and latitudes, albeit with important differences. By dividing our observed wavenumber distribution into particular subspecies of waves, we demonstrate that these distributions exhibit significant temporal and spatial variability, and that small-scale variability associated with particular geophysical phenomena such as the monsoon arises due to variations in specific parts of the observed spectrum. We further show that the well-known Andes/Antarctic Peninsula gravity wave hotspot during southern winter, home to some of the largest wave fluxes on the planet, is made up of relatively few waves, but with a significantly increased flux per wave due to their spectral characteristics. These results have implications for the modelling of gravity wave phenomena.