A consistent framework for quantifying global energy budgets

Accurate quantification of energy exchanges within the coupled climate system from scales of a climate model grid column to the global scale contribute key elements to our understanding of Earths climate. High quality input data sets, most notably reanalyses, are a necessary prerequisite for this. Atmospheric reanalyses are gridded data sets combining meteorological observations with a dynamical model in a statistically optimal way. Similar products exist for the ocean and sea ice as well as land. However, current physical frameworks are inconsistent and numerical procedures for energy budget evaluations are suboptimal. This project aims to expand a recently developed consistent coupled energy budget framework that is independent of reference temperature in the steady state to the non-steady state. The goal is to minimize ambiguities that arise from local mass variations. The envisaged framework will ensure consistency between atmosphere, ocean, sea ice, and land heat budgets, and will be applied to new reanalysis data sets. In addition to global assessments, regional budget studies will be carried out for high latitudes, where valuable mooring-derived oceanic flux data are available to the working group, and the tropical Pacific, where strong air-sea coupling associated with El Nino-Southern Oscillation is present. Numerous technical advances are envisaged: inclusion of so-far neglected terms (e.g. heat fluxes associated with precipitation and evaporation; water in all its states), investigations towards optimal choice of reference temperatures that minimize ambiguities in the non-steady state, a novel three-dimensional-wind adjustment, and improvement of numerical methods to minimize noise especially over high topography. The developed methods will be implemented in a numerically optimal way for the new atmospheric reanalysis ERA5, which is currently being produced by Copernicus Climate Change Service (C3S). Budgets from ERA5, its ocean-sea-ice counterpart ERA5-OCEAN, and its land surface derivative ERA5-LAND will be used to perform dedicated coupled energy cycle analyses with unprecedented accuracy. Results will be verified for the first time in the central Arctic over an annual cycle with aggregated surface fluxes from the MOSAiC (Multidisciplinary drifting Observatory for the Study of Arctic Climate) expedition, which is carried out during winter 2019/20. In addition, energy fluxes from ERA5 will be compared to other reanalysis products and climate models. This work is crucial to provide a framework that i) is physically consistent and ii) seizes the full potential of ERA5. A significant gain in useful horizontal resolution and accuracy of energy budget terms such as 3D energy flux divergence and net surface energy flux is expected. The improved diagnostics will finally be implemented into the standard output software of future C3S reanalyses, which will ensure maximum quality of the budgets distributed by C3S.

Diagnosing global coupled (atmosphere-ocean-sea ice-land) energy budgets using observationally constrained data is essential for monitoring and understanding global climate as well as validation of climate model simulations. High quality input data sets, most notably reanalyses, are a prerequisite for this. Atmospheric reanalyses are gridded data sets combining meteorological observations with a dynamical model in a statistically optimal way. Similar products exist for the ocean and sea ice as well as land. This project (i) addressed shortcomings in previously existing diagnostic budget frameworks, (ii) evaluated energy-budget-related aspects of the most recent atmospheric reanalysis effort by the Copernicus Climate Change Service (C3S; called ERA5), and (iii) supported several scientific studies that employed the energy budget data generated using ERA5. Previous studies of the global energy budgets employed diagnostics frameworks which exhibited inconsistencies (e.g., inconsistent choices of the used reference temperature in atmosphere and ocean) and neglected some potentially sizeable terms (e.g., latent heat of frozen water in the atmosphere), deteriorating the quality of the evaluations. This is particularly problematic when a hard-to-measure quantity like net surface energy exchange is inferred as a residual from other energy budget terms: diagnostic inconsistencies will inevitably project on the inferred quantity. This project supported development and use of a more consistent diagnostic framework. Improvements could be demonstrated by reduced budget residuals and reduced biases of inferred net surface energy flux (<1 W/m2 for the global ocean mean). In addition to the diagnostic framework, considerable efforts went into improvement of the numerical methods. The divergence of lateral atmospheric energy transports is most critical, since it suffers from spurious divergent winds in atmospheric reanalyses and spatially noisy fields. Within this project, original software used in production of ERA5 have been used for divergence computations in spectral space to minimize noise. In conjunction with general quality improvements of ERA5, this helped to significantly reduce noise levels compared to previous standards and increase useful resolution to ~1x1, now matching the resolution of state-of-the-art satellite data widely used in energy budget studies. The generated ERA5 energy budget data has been made publicly available via the C3S climate data store. The applied part of this project used the created energy budget data in a range of climate studies. Focus regions were the North Atlantic and the Arctic, where different aspects of the observed energy budget were investigated (e.g., decadal trends in North Atlantic net surface energy flux, a critical quantity related to the Atlantic Meridional Overturning Circulation) and employed for evaluation of the latest generation of climate model simulations. In total, the work within this project fed into 29 first- and co-authored peer-reviewed publications.