High-resolution Earth System Model Research Articles

Flux coupling approach on an exchange grid for the IOW Earth System Model (version 1.04.00) of the Baltic Sea region

Abstract. In this article the development of a high-resolution Earth System Model (ESM) for the Baltic Sea region is described. In contrast to conventional coupling approaches, the presented model features an additional (technical) component, the flux calculator, which calculates fluxes between the model components on a common exchange grid. This approach naturally ensures conservation of exchanged quantities, a locally consistent treatment of the fluxes, and facilitates interchanging model components in a straightforward manner. The main purpose of this model is to downscale global reanalysis or climate model data to the Baltic Sea region as typically, global model grids are too coarse to resolve the region of interest sufficiently. The regional ESM consists of the Modular Ocean Model 5 (MOM5) for the ocean and the COSMO model in CLimate Mode (CCLM, version 5.0_clm3) for the atmosphere. The bi-directional ocean–atmosphere coupling allows for a realistic air–sea feedback that outperforms the traditional approach of using uncoupled standalone models, as typically pursued with the EURO-CORDEX protocol. In order to address marine environmental problems (e.g., eutrophication and oxygen depletion), the ocean model is internally coupled with the marine biogeochemistry model, ERGOM, set up for the Baltic Sea's hydrographic conditions. The regional ESM can be used for various scientific questions such as climate sensitivity experiments, reconstruction of ocean dynamics, study of past climates, and natural variability, as well as investigation of ocean–atmosphere interactions. Therefore, it can serve for a better understanding of natural processes via attribution experiments that relate observed changes to mechanistic causes.

Design and evaluation of an efficient high-precision ocean surface wave model with a multiscale grid system (MSG_Wav1.0)

Abstract. Ocean surface waves induced by wind forcing and topographic effects are a crucial physical process at the air–sea interface, which significantly affect typhoon development, ocean mixing, etc. Higher-resolution wave modeling can simulate more accurate wave states but requires a huge number of computational resources, making it difficult for Earth system models to include ocean waves as a fast-response physical process. Given that high-resolution Earth system models are in demand, efficient high-precision wave simulation is necessary and urgent. Based on the wave dispersion relation, we design a new wave modeling framework using a multiscale grid system. It has the fewest number of fine grids and reasonable grid spacing in deep-water areas. We compare the performance of wave simulation using different spatial propagation schemes, reveal the different reasons for wave simulation differences in the westerly zone and the active tropical cyclone region, and quantify the matching of spatial resolutions between wave models and wind forcing. A series of numerical experiments show that this new modeling framework can more precisely simulate wave states in shallow-water areas without losing accuracy in the deep ocean while costing a fraction of the price of traditional simulations with uniform fine-gridding space. With affordable computational expenses, the new ocean surface wave modeling can be implemented into high-resolution Earth system models, which may significantly improve the simulation of the atmospheric planetary boundary layer and upper-ocean mixing.

Southern Ocean polynyas and dense water formation in a high-resolution, coupled Earth system model

Abstract. Antarctic coastal polynyas produce dense shelf water, a primary source of Antarctic Bottom Water that contributes to the global overturning circulation. This paper investigates Antarctic dense water formation in the high-resolution version of the Energy Exascale Earth System Model (E3SM-HR). The model is able to reproduce the main Antarctic coastal polynyas, although the polynyas are smaller in area compared to observations. E3SM-HR also simulates several occurrences of open-ocean polynyas (OOPs) in the Weddell Sea at a higher rate than what the last 50 years of the satellite sea ice observational record suggests, but similarly to other high-resolution Earth system model simulations. Furthermore, the densest water masses in the model are formed within the OOPs rather than on the continental shelf as is typically observed. Biases related to the lack of dense water formation on the continental shelf are associated with overly strong atmospheric polar easterlies, which lead to a strong Antarctic Slope Front and too little exchange between on- and off-continental shelf water masses. Strong polar easterlies also produce excessive southward Ekman transport, causing a build-up of sea ice over the continental shelf and enhanced ice melting in the summer season. This, in turn, produces water masses on the continental shelf that are overly fresh and less dense relative to observations. Our results indicate that high resolution alone is insufficient for models to properly reproduce Antarctic dense water; the large-scale polar atmospheric circulation around Antarctica must also be accurately simulated.

ENSO Diversity and the Simulation of Its Teleconnections to Winter Precipitation Extremes Over the US in High Resolution Earth System Models

AbstractAccounting for the diversity in El Niño Southern Oscillation (ENSO)'s spatial pattern, with the novel ENSO longitudinal index (ELI), we evaluate the simulation of its teleconnections to US winter precipitation extremes by seven global high‐resolution (HR) Earth System Models (ESM). Six (four) HR ESMs simulate the observed increase in precipitation extremes over Southwest US (Southeast US) during ELI‐defined El Niño events better than their low‐resolution counterparts, which are low‐biased. The stronger ENSO‐dependence over the Southwest US and Southeast US in those models is associated with an improved simulation of moisture flux into the regions and/or storm track activity there. HR ESMs, however, generally overestimate the increase in precipitation extremes over the Pacific‐Northwest during La Niña events. Model bias there is associated with bias in moisture transport into the region during La Niña events, which is amplified by the enhanced vertical mass fluxes in HR.

Modelling feedbacks between the Northern Hemisphere ice sheets and climate during the last glacial cycle

Abstract. During the last glacial cycle (LGC), ice sheets covered large parts of Eurasia and North America, which resulted in ∼120 m of sea level change. Ice sheet–climate interactions have considerable influence on temperature and precipitation patterns and therefore need to be included when simulating this time period. Ideally, ice sheet–climate interactions are simulated by a high-resolution Earth system model. While these models are capable of simulating climates at a certain point in time, such as the pre-industrial (PI) or the Last Glacial Maximum (LGM; 21 000 years ago), a full transient glacial cycle is currently computationally unfeasible as it requires a too-large amount of computation time. Nevertheless, ice sheet models require forcing that captures the gradual change in climate over time to calculate the accumulation and melt of ice and its effect on ice sheet extent and volume changes. Here we simulate the LGC using an ice sheet model forced by LGM and PI climates. The gradual change in climate is modelled by transiently interpolating between pre-calculated results from a climate model for the LGM and the PI. To assess the influence of ice sheet–climate interactions, we use two different interpolation methods: the climate matrix method, which includes a temperature–albedo and precipitation–topography feedback, and the glacial index method, which does not. To investigate the sensitivity of the results to the prescribed climate forcing, we use the output of several models that are part of the Paleoclimate Modelling Intercomparison Project Phase III (PMIP3). In these simulations, ice volume is prescribed, and the climate is reconstructed with a general circulation model (GCM). Here we test those models by using their climate to drive an ice sheet model over the LGC. We find that the ice volume differences caused by the climate forcing exceed the differences caused by the interpolation method. Some GCMs produced unrealistic LGM volumes, and only four resulted in reasonable ice sheets, with LGM Northern Hemisphere sea level contribution ranging between 74–113 m with respect to the present day. The glacial index and climate matrix methods result in similar ice volumes at the LGM but yield a different ice evolution with different ice domes during the inception phase of the glacial cycle and different sea level rates during the deglaciation phase. The temperature–albedo feedback is the main cause of differences between the glacial index and climate matrix methods.

Simulation of ENSO Teleconnections to Precipitation Extremes over the United States in the High-Resolution Version of E3SM

Abstract We evaluate the simulated teleconnection of El Niño–Southern Oscillation (ENSO) to winter season precipitation extremes over the United States in a long (98 years) 1950 control high-resolution version (HR; 25-km nominal atmosphere model horizontal resolution) of the U.S. Department of Energy’s (DOE) Energy Exascale Earth System Model version 1 (E3SMv1). The model bias and spatial pattern of ENSO teleconnections to mean and extreme precipitation in HR overall are similar to the low-resolution model’s (LR; 110 km) historical simulation (four-member ensemble, 1925–59). However, over the southeastern United States (SE-U.S.), HR produces stronger El Niño–associated extremes, reducing LR’s model bias. Both LR and HR produce weaker than observed increase in storm track activity during El Niño events there, but HR improves the ENSO-associated variability of moisture transport over SE-U.S. During El Niño, stronger vertical velocities in HR produce stronger large-scale precipitation, causing larger latent heating of the troposphere that pulls in more moisture from the Gulf of Mexico into the SE-U.S. This positive feedback also contributes to the stronger mean and extreme precipitation response in HR. Over the Pacific Northwest, LR’s bias of stronger than observed La Niña associated extremes is amplified in HR. Both models simulate stronger than observed moisture transport from the Pacific Ocean into the region during La Niña years. The amplified HR bias there is due to stronger orographically driven vertical updrafts that create stronger large-scale precipitation, despite weaker La Niña–induced storm track activity. Significance Statement New high-resolution Earth system models (ESMs) solve mathematical equations of fluid flow at much smaller spatial scales than prevalent ESMs, and thus are prohibitively expensive to compute. However, they can be useful for simulating accurate details of regional climate extremes that are driven by naturally occurring climate oscillations like El Niño–Southern Oscillation (ENSO). Here, we evaluate the simulation of ENSO-driven precipitation extremes over the United States in the high-resolution version of the U.S. Department of Energy’s new Energy Exascale Earth System Model version 1. We find that the high-resolution model improves upon its low-resolution counterpart over the southeastern United States by producing a better transport of moisture into the region from the Gulf of Mexico during El Niño. Over the U.S. Pacific Northwest, the high-resolution model simulates the atmospheric flow in more detail over the complex mountainous terrain. However, it also brings in more moisture from the Pacific Ocean just like the low-resolution model. This causes it to produce precipitation extremes during La Niña years there that are stronger than that observed in the real world.

Earth system models for regional environmental management of red tide: Prospects and limitations of current generation models and next generation development.

Earth system models (ESMs) serve as a unique research infrastructure for quality climate services, yet their application for environmental management at regional scale has not yet been fully explored. The unprecedented resolution and model fidelity of the Coupled Model Intercomparison Project Phase 6 (CMIP6) simulations, especially of the High-Resolution Model Intercomparison Project (HighResMIP) focusing on regional phenomena, offer opportunities for such applications. This article presents the first venture into using the HighResMIP simulations to tackle a regional environmental issue, the Florida Red Tide. This is a harmful algae bloom caused by the dinoflagellate Karenia brevis, a toxic single-celled microscopic protist. We use CMIP6 historical simulations to establish a causal agreement between the position of Loop Current, a warm ocean current that moves into the Gulf of Mexico, and the occurrence of K. brevis blooms on the Western Florida shelf. Results show that the high-resolution ESMs are capable of simulating the phenomena of interest (i.e., Loop Current) at the regional spatial scale with generally adequate data-model agreement in the context of the relation between Loop Current and red tide. We use this case study to elaborate on the prospects and limitations of using publicly available CMIP data for regional environmental management. We highlight the current gaps and the developmental needs for the next generation ESMs, and discuss the role of stakeholder participation in future ESMs development to facilitate the translation of scientific understanding to better inform decision-making of regional environmental management.

Mediterranean Mercury Assessment 2022: An Updated Budget, Health Consequences, and Research Perspectives.

Mercury (Hg) and especially its methylated species (MeHg) are toxic chemicals that contaminate humans via the consumption of seafood. The most recent UNEP Global Mercury Assessment stressed that Mediterranean populations have higher Hg levels than people elsewhere in Europe. The present Critical Review updates current knowledge on the sources, biogeochemical cycling, and mass balance of Hg in the Mediterranean and identifies perspectives for future research especially in the context of global change. Concentrations of Hg in the Western Mediterranean average 0.86 ± 0.27 pmol L-1 in the upper water layer and 1.02 ± 0.12 pmol L-1 in intermediate and deep waters. In the Eastern Mediterranean, Hg measurements are in the same range but are too few to determine any consistent oceanographical pattern. The Mediterranean waters have a high methylation capacity, with MeHg representing up to 86% of the total Hg, and constitute a source of MeHg for the adjacent North Atlantic Ocean. The highest MeHg concentrations are associated with low oxygen water masses, suggesting a microbiological control on Hg methylation, consistent with the identification of hgcA-like genes in Mediterranean waters. MeHg concentrations are twice as high in the waters of the Western Basin compared to the ultra-oligotrophic Eastern Basin waters. This difference appears to be transferred through the food webs and the Hg content in predators to be ultimately controlled by MeHg concentrations of the waters of their foraging zones. Many Mediterranean top-predatory fish still exceed European Union regulatory Hg thresholds. This emphasizes the necessity of monitoring the exposure of Mediterranean populations, to formulate adequate mitigation strategies and recommendations, without advising against seafood consumption. This review also points out other insufficiencies of knowledge of Hg cycling in the Mediterranean Sea, including temporal variations in air-sea exchange, hydrothermal and cold seep inputs, point sources, submarine groundwater discharge, and exchanges between margins and the open sea. Future assessment of global change impacts under the Minamata Convention Hg policy requires long-term observations and dedicated high-resolution Earth System Models for the Mediterranean region.

Natural hazards and extreme events in the Baltic Sea region

Abstract. A natural hazard is a naturally occurring extreme event that has a negative effect on people and society or the environment. Natural hazards may have severe implications for human life and can potentially generate economic losses and damage ecosystems. A better understanding of their major causes, probability of occurrence, and consequences enables society to be better prepared to save human lives as well as to invest in adaptation options. Natural hazards related to climate change are identified as one of the Grand Challenges in the Baltic Sea region. Here, we summarize existing knowledge about extreme events in the Baltic Sea region with a focus on the past 200 years as well as on future climate scenarios. The events considered here are the major hydro-meteorological events in the region and include wind storms, extreme waves, high and low sea levels, ice ridging, heavy precipitation, sea-effect snowfall, river floods, heat waves, ice seasons, and drought. We also address some ecological extremes and the implications of extreme events for society (phytoplankton blooms, forest fires, coastal flooding, offshore infrastructure, and shipping). Significant knowledge gaps are identified, including the response of large-scale atmospheric circulation to climate change and also concerning specific events, for example, the occurrence of marine heat waves and small-scale variability in precipitation. Suggestions for future research include the further development of high-resolution Earth system models and the potential use of methodologies for data analysis (statistical methods and machine learning). With respect to the expected impacts of climate change, changes are expected for sea level, extreme precipitation, heat waves and phytoplankton blooms (increase), and cold spells and severe ice winters (decrease). For some extremes (drying, river flooding, and extreme waves), the change depends on the area and time period studied.

Threat by marine heatwaves to adaptive large marine ecosystems in an eddy-resolving model

Marine heatwaves (MHWs), episodic periods of abnormally high sea surface temperature, severely affect marine ecosystems. Large marine ecosystems (LMEs) cover ~22% of the global ocean but account for 95% of global fisheries catches. Yet how climate change affects MHWs over LMEs remains unknown because such LMEs are confined to the coast where low-resolution climate models are known to have biases. Here, using a high-resolution Earth system model and applying a ‘future threshold’ that considers MHWs as anomalous warming above the long-term mean warming of sea surface temperatures, we find that future intensity and annual days of MHWs over the majority of the LMEs remain higher than in the present-day climate. Better resolution of ocean mesoscale eddies enables simulation of more realistic MHWs than low-resolution models. These increases in MHWs under global warming pose a serious threat to LMEs, even if resident organisms could adapt fully to the long-term mean warming.

Identifying global favourable habitat for early juvenile loggerhead sea turtles.

Loggerhead sea turtles (Caretta caretta) nest globally on sandy beaches, with hatchlings dispersing into the open ocean. Where these juveniles go and what habitat they rely on remains a critical research question for informing conservation priorities. Here a high-resolution Earth system model is used to determine the biophysical geography of favourable ocean habitat for loggerhead sea turtles globally during their first year of life on the basis of ocean current transport, thermal constraints and food availability (defined here as the summed lower trophic level carbon biomass). Dispersal is simulated from eight major nesting sites distributed across the globe in four representative years using particle tracking. Dispersal densities are identified for all turtles, and for the top 15% 'best-fed' turtles that have not encountered metabolically unfavourable temperatures. We find that, globally, rookeries are positioned to disperse to regions where the lower trophic biomass is greatest within loggerheads' thermal range. Six out of the eight nesting sites are associated with strong coastal boundary currents that rapidly transport hatchlings to subtropical-subpolar gyre boundaries; narrow spatial migratory corridors exist for 'best-fed' turtles associated with these sites. Two other rookeries are located in exceptionally high-biomass tropical regions fuelled by natural iron fertilization. 'Best-fed' turtles tend to be associated with lower temperatures, highlighting the inverse relationship between temperature and lower trophic biomass. The annual mean isotherms between 20°C and the thermal tolerance of juvenile loggerheads are a rough proxy for favourable habitat for loggerheads from rookeries associated with boundary currents. Our results can be used to constrain regions for conservation efforts for each subpopulation, and better identify foraging habitat for this critical early life stage.

Optimizing high-resolution Community Earth System Model on a heterogeneous many-core supercomputing platform

Abstract. With semiconductor technology gradually approaching its physical and thermal limits, recent supercomputers have adopted major architectural changes to continue increasing the performance through more power-efficient heterogeneous many-core systems. Examples include Sunway TaihuLight that has four management processing elements (MPEs) and 256 computing processing elements (CPEs) inside one processor and Summit that has two central processing units (CPUs) and six graphics processing units (GPUs) inside one node. Meanwhile, current high-resolution Earth system models that desperately require more computing power generally consist of millions of lines of legacy code developed for traditional homogeneous multicore processors and cannot automatically benefit from the advancement of supercomputer hardware. As a result, refactoring and optimizing the legacy models for new architectures become key challenges along the road of taking advantage of greener and faster supercomputers, providing better support for the global climate research community and contributing to the long-lasting societal task of addressing long-term climate change. This article reports the efforts of a large group in the International Laboratory for High-Resolution Earth System Prediction (iHESP) that was established by the cooperation of Qingdao Pilot National Laboratory for Marine Science and Technology (QNLM), Texas A&M University (TAMU), and the National Center for Atmospheric Research (NCAR), with the goal of enabling highly efficient simulations of the high-resolution (25 km atmosphere and 10 km ocean) Community Earth System Model (CESM-HR) on Sunway TaihuLight. The refactoring and optimizing efforts have improved the simulation speed of CESM-HR from 1 SYPD (simulation years per day) to 3.4 SYPD (with output disabled) and supported several hundred years of pre-industrial control simulations. With further strategies on deeper refactoring and optimizing for remaining computing hotspots, as well as redesigning architecture-oriented algorithms, we expect an equivalent or even better efficiency to be gained on the new platform than traditional homogeneous CPU platforms. The refactoring and optimizing processes detailed in this paper on the Sunway system should have implications for similar efforts on other heterogeneous many-core systems such as GPU-based high-performance computing (HPC) systems.

Assessing terrestrial biogeochemical feedbacks in a strategically geoengineered climate

Geoengineering by injecting sulfur dioxide (SO2) into the lower stratosphere has been suggested to reduce anthropogenically induced warming. While impacts of such geoengineering on climate have been investigated in recent decades, few modeling studies have considered biogeochemical feedbacks resulting from such intervention. This study comprehensively characterizes responses and feedbacks of terrestrial ecosystems, from an ensemble of coupled high-resolution Earth system model climate change simulations, under the highest standard greenhouse gas scenario with an extreme geoengineering mitigation strategy. Under this strategy, temperature increases beyond 2020 levels due to elevated anthropogenic carbon dioxide (CO2) were completely offset by the SO2 injection. Carbon cycle feedbacks can alter the trajectory of atmospheric CO2 levels by storing or releasing additional carbon on land and in the ocean, thus moderating or amplifying climate change. We assess terrestrial biogeochemical feedbacks to climate in response to geoengineering, using model output from the Stratospheric Aerosol Geoengineering Large Ensemble (GLENS) project. Results indicate terrestrial ecosystems become a stronger carbon sink globally because of lower ecosystem respiration and diminished disturbance effects under geoengineering. An additional 79 Pg C would be stored on land by the end of the twenty-first century, yielding as much as a 4% reduction in atmospheric CO2 mole fraction without marine biogeochemical feedbacks, compared to the high greenhouse gas scenario without geoengineering.

Bottom-up drivers of global patterns of demersal, forage, and pelagic fishes

Large-scale spatial heterogeneity in fisheries production is predominantly controlled by the availability of zooplankton and benthic organisms, which have a complex relationship with primary production. To investigate how cross-ecosystem differences in these drivers determine fish assemblages and productivity, we constructed a spatially explicit mechanistic model of three fish functional types: forage, large pelagic, and demersal fishes. The model is based on allometric scaling principles, includes basic life cycle transitions, and has trophic interactions between the fishes and with their pelagic and benthic food resources. The model was applied to the global ocean, with plankton food web estimates and ocean conditions from a high-resolution earth system model. Further, a simple representation of fishing was included, and led to moderate matches with total, large pelagic, and demersal catches, including re-creation of observed variations in fish catch spanning two orders of magnitude. Our results highlight several ecologically meaningful model sensitivities. First, coexistence between forage and large pelagic fish in productive regions occurred when forage fish survival is promoted via both favorable metabolic allometry and enhanced predator avoidance in adult forage fish. Second, the prominence of demersal fish is highly sensitive to the efficiency of energy transfer to benthic invertebrates. Third, the latitudinal distribution of the total catch is modulated by the temperature dependence of metabolic rates, with increased sensitivity pushing fish biomass toward the poles. Fourth, forage fish biomass is suppressed by strong top-down controls on temperate and subpolar shelves, where mixed assemblages of large pelagic and demersal fishes exerted high predation rates. Last, spatial differences in the dominance of large pelagics vs. demersals is strongly related to the ratio of pelagic zooplankton production to benthic production. We discuss the potential linkages between model misfits and unresolved processes including movement, spawning phenology, seabird and marine mammal predators, and socioeconomically driven fishing pressure, which are identified as priorities for future model development. Ultimately, the model and analyses herein are intended as a baseline for a robust, mechanistic tool to understand, quantify, and predict global fish biomass and yield, now and in a future dominated by climate change and improved fishing technology.

Impact of air–sea coupling on the simulated global tropical cyclone activity in the high-resolution Community Earth System Model (CESM)

Atmosphere–ocean coupling is critical for tropical cyclones (TC) formation and development. TCs derive their energy from the upper ocean, and the associated TC-ocean interactions can in turn modulate storm evolution. This study assesses the impact of ocean coupling on directly-simulated global TC activity in the high-resolution “TC-permitting” Community Earth System Model (CESM). Model-simulated global TC activity is evaluated in a 30-year fully-coupled CESM simulation (CPL), in which the 0.25° atmosphere component is coupled to the nominal 1° dynamic ocean (with ~ 0.27° horizontal grid spacing in the tropics). An atmosphere-only simulation (ATM) is branched from CPL, with sea surface temperature (SST) specified from CPL, which we use to isolate the effect of ocean coupling on TC activity. We find that the two-way ocean coupling can affect global TC frequency, geographical distribution, storm intensity, and interannual variability. ATM on average simulates 27% more major TC events than CPL globally, and the TC power dissipation is higher than CPL poleward of 12° latitude in both hemispheres. The lack of negative SST feedbacks in ATM allows TCs to have a longer intensification period and reach the maximum intensity at a higher latitude. In CPL, TC interannual variability is heavily influenced by El Nino/La Nina events. This relationship can be captured in ATM under strong events but is less predictable during weak and neutral years. Results help to better understand the connections and feedbacks linking air–sea coupling, tropical variability, and the directly simulated TC activity within the high-resolution Earth System Models.

Preconditioning and Formation of Maud Rise Polynyas in a High-Resolution Earth System Model

AbstractOpen-ocean polynyas (OOPs) in the Southern Ocean are ice-free areas within the winter ice pack that are associated with deep convection, potentially contributing to the formation of Antarctic Bottom Water. To enhance the credibility of Earth system models (ESMs), their ability to simulate OOPs realistically is thus crucial. Here we investigate OOPs that emerge intermittently in a high-resolution (HR) preindustrial simulation with the Energy Exascale Earth System Model, version 0.1 (E3SMv0), an offspring of the Community Earth System Model (CESM). While low-resolution (LR) simulations with E3SMv0 show no signs of OOP formation, the preindustrial E3SMv0-HR simulation produces both large Weddell Sea polynyas (WSPs) as well as small Maud Rise polynyas (MRPs). The latter are associated with a prominent seamount in the eastern Weddell Sea, and their preconditioning and formation is the focus of this study. The steep flanks of the rugged topography in this region are in E3SMv0-HR sufficiently well resolved for the impinging flow to produce pronounced Taylor caps that precondition the region for convection. Aided by an accumulation of heat in the Weddell Deep Water layer, the ultimate trigger of convection that leads to MRPs is the advection of anomalously high upper-ocean-layer salinity. The crucial difference to WSP-producing LR ESM simulations is that in E3SMv0-HR, WSPs are realistically preceded by MRPs, which in turn are a result of the flow around bathymetry being represented with unprecedented detail.

Evaluating lossy data compression on climate simulation data within a large ensemble

Abstract. High-resolution Earth system model simulations generate enormous data volumes, and retaining the data from these simulations often strains institutional storage resources. Further, these exceedingly large storage requirements negatively impact science objectives, for example, by forcing reductions in data output frequency, simulation length, or ensemble size. To lessen data volumes from the Community Earth System Model (CESM), we advocate the use of lossy data compression techniques. While lossy data compression does not exactly preserve the original data (as lossless compression does), lossy techniques have an advantage in terms of smaller storage requirements. To preserve the integrity of the scientific simulation data, the effects of lossy data compression on the original data should, at a minimum, not be statistically distinguishable from the natural variability of the climate system, and previous preliminary work with data from CESM has shown this goal to be attainable. However, to ultimately convince climate scientists that it is acceptable to use lossy data compression, we provide climate scientists with access to publicly available climate data that have undergone lossy data compression. In particular, we report on the results of a lossy data compression experiment with output from the CESM Large Ensemble (CESM-LE) Community Project, in which we challenge climate scientists to examine features of the data relevant to their interests, and attempt to identify which of the ensemble members have been compressed and reconstructed. We find that while detecting distinguishing features is certainly possible, the compression effects noticeable in these features are often unimportant or disappear in post-processing analyses. In addition, we perform several analyses that directly compare the original data to the reconstructed data to investigate the preservation, or lack thereof, of specific features critical to climate science. Overall, we conclude that applying lossy data compression to climate simulation data is both advantageous in terms of data reduction and generally acceptable in terms of effects on scientific results.