ALBERT

Description: Stochastic subgrid parameterizations enable ensemble forecasts of fluid dynamic systems and ultimately accurate data assimilation (DA). Stochastic advection by Lie transport (SALT) and models under location uncertainty (LU) are recent and similar physically based stochastic schemes. SALT dynamics conserve helicity, whereas LU models conserve kinetic energy (KE). After highlighting general similarities between LU and SALT frameworks, this paper focuses on their common challenge: the parameterization choice. We compare uncertainty quantification skills of a stationary heterogeneous data-driven parameterization and a non-stationary homogeneous self-similar parameterization. For stationary, homogeneous surface quasi-geostrophic (SQG; QG) turbulence, both parameterizations lead to high-quality ensemble forecasts. This paper also discusses a heterogeneous adaptation of the homogeneous parameterization targeted at a better simulation of strong straight buoyancy fronts.

Description: We present a new framework for global ocean–sea-ice model simulations based on phase 2 of the Ocean Model Intercomparison Project (OMIP-2), making use of the surface dataset based on the Japanese 55-year atmospheric reanalysis for driving ocean–sea-ice models (JRA55-do). We motivate the use of OMIP-2 over the framework for the first phase of OMIP (OMIP-1), previously referred to as the Coordinated Ocean–ice Reference Experiments (COREs), via the evaluation of OMIP-1 and OMIP-2 simulations from 11 state-of-the-science global ocean–sea-ice models. In the present evaluation, multi-model ensemble means and spreads are calculated separately for the OMIP-1 and OMIP-2 simulations and overall performance is assessed considering metrics commonly used by ocean modelers. Both OMIP-1 and OMIP-2 multi-model ensemble ranges capture observations in more than 80 % of the time and region for most metrics, with the multi-model ensemble spread greatly exceeding the difference between the means of the two datasets. Many features, including some climatologically relevant ocean circulation indices, are very similar between OMIP-1 and OMIP-2 simulations, and yet we could also identify key qualitative improvements in transitioning from OMIP-1 to OMIP-2. For example, the sea surface temperatures of the OMIP-2 simulations reproduce the observed global warming during the 1980s and 1990s, as well as the warming slowdown in the 2000s and the more recent accelerated warming, which were absent in OMIP-1, noting that the last feature is part of the design of OMIP-2 because OMIP-1 forcing stopped in 2009. A negative bias in the sea-ice concentration in summer of both hemispheres in OMIP-1 is significantly reduced in OMIP-2. The overall reproducibility of both seasonal and interannual variations in sea surface temperature and sea surface height (dynamic sea level) is improved in OMIP-2. These improvements represent a new capability of the OMIP-2 framework for evaluating process-level responses using simulation results. Regarding the sensitivity of individual models to the change in forcing, the models show well-ordered responses for the metrics that are directly forced, while they show less organized responses for those that require complex model adjustments. Many of the remaining common model biases may be attributed either to errors in representing important processes in ocean–sea-ice models, some of which are expected to be reduced by using finer horizontal and/or vertical resolutions, or to shared biases and limitations in the atmospheric forcing. In particular, further efforts are warranted to resolve remaining issues in OMIP-2 such as the warm bias in the upper layer, the mismatch between the observed and simulated variability of heat content and thermosteric sea level before 1990s, and the erroneous representation of deep and bottom water formations and circulations. We suggest that such problems can be resolved through collaboration between those developing models (including parameterizations) and forcing datasets. Overall, the present assessment justifies our recommendation that future model development and analysis studies use the OMIP-2 framework.

Description: This paper presents global comparisons of fundamental global climate variables from a suite of four pairs of matched low- and high-resolution ocean and sea ice simulations that are obtained following the OMIP-2 protocol (Griffies et al., 2016) and integrated for one cycle (1958–2018) of the JRA55-do atmospheric state and runoff dataset (Tsujino et al., 2018). Our goal is to assess the robustness of climate-relevant improvements in ocean simulations (mean and variability) associated with moving from coarse (∼ 1∘) to eddy-resolving (∼ 0.1∘) horizontal resolutions. The models are diverse in their numerics and parameterizations, but each low-resolution and high-resolution pair of models is matched so as to isolate, to the extent possible, the effects of horizontal resolution. A variety of observational datasets are used to assess the fidelity of simulated temperature and salinity, sea surface height, kinetic energy, heat and volume transports, and sea ice distribution. This paper provides a crucial benchmark for future studies comparing and improving different schemes in any of the models used in this study or similar ones. The biases in the low-resolution simulations are familiar, and their gross features – position, strength, and variability of western boundary currents, equatorial currents, and the Antarctic Circumpolar Current – are significantly improved in the high-resolution models. However, despite the fact that the high-resolution models “resolve” most of these features, the improvements in temperature and salinity are inconsistent among the different model families, and some regions show increased bias over their low-resolution counterparts. Greatly enhanced horizontal resolution does not deliver unambiguous bias improvement in all regions for all models.

Description: We quantify skill and uncertainty in observing the statistics of natural variability using observing system simulation experiments on an ensemble of climate simulations and an observing strategy of in situ measurements and objective mapping. The targeted statistic is the 0–700 m global ocean heat content anomaly as observed by the In Situ Analysis System 2013 (ISAS13) strategy of a long, equilibriated simulation of the Community Climate System Model (CCSM) version 3.5. Subannual variability is found to be significantly contaminated by the observing strategy, especially before 2005, primarily due to the sparseness and seasonality in the number and location of pre-Argo observations. However, one-year running means from 2005 onward are found to faithfully capture the natural variability of the model's true ocean heat content variability. During these years, synthetic observed annual running means are strongly correlated with the actual annual running means of the model, with a median correlation of 95 %, versus only 60 % for the observational record before 2005. When scaled to account for the fact that the real ocean is more variable than the model, root mean square errors in observing the annual-running mean natural variability of the global ocean heat content are estimated to be 6.2 ZJ for the pre-Argo era (1990–2005) and 2.1 ZJ for the Argo era (2005–2013) with relative signal-to-noise ratios of 1.9 and 14.7. Combining the estimated, scaled uncertainties of the observing strategy with its estimated trend, the 1990–2013 trend in global ocean heat content is found to be 5.3 ± 1.0 ZJ/yr.

Description: Marine sediments have greatly improved our understanding of the climate system, but their interpretation often assumes that certain climate mechanisms operate consistently over all timescales of interest and that variability at one or a few sample sites is representative of an oceanographic province. In this study, we test these assumptions using modern observations in an idealized manner mimicking paleo-reconstruction to investigate whether sea surface temperature and productivity proxy records in the Southern California Current System can be used to reconstruct Ekman upwelling. The method uses extended empirical orthogonal function (EEOF) analysis of the covariation of alongshore wind stress, chlorophyll, and sea surface temperature as measured by satellites from 2002 to 2009. We find that EEOF1 does not reflect an Ekman upwelling pattern but instead much broader California Current processes. EEOF2 and 3 reflect upwelling patterns, but these patterns are timescale dependent and regional. Thus, the skill of using one site to reconstruct the large-scale dominant patterns is spatially dependent. Lastly, we show that using multiple sites and/or multiple variables generally improves field reconstruction. These results together suggest that caution is needed when attempting to extrapolate mechanisms that may be important on seasonal timescales (e.g., Ekman upwelling) to deeper time but also the advantage of having multiple proxy records.

Description: In sea-ice-covered areas, the sea ice floe size distribution (FSD) plays an important role in many processes affecting the coupled sea–ice–ocean–atmosphere system. Observations of the FSD are sparse – traditionally taken via a painstaking analysis of ice surface photography – and the seasonal and inter-annual evolution of floe size regionally and globally is largely unknown. Frequently, measured FSDs are assessed using a single number, the scaling exponent of the closest power-law fit to the observed floe size data, although in the absence of adequate datasets there have been limited tests of this “power-law hypothesis”. Here we derive and explain a mathematical technique for deriving statistics of the sea ice FSD from polar-orbiting altimeters, satellites with sub-daily return times to polar regions with high along-track resolutions. Applied to the CryoSat-2 radar altimetric record, covering the period from 2010 to 2018, and incorporating 11 million individual floe samples, we produce the first pan-Arctic climatology and seasonal cycle of sea ice floe size statistics. We then perform the first pan-Arctic test of the power-law hypothesis, finding limited support in the range of floe sizes typically analyzed in photographic observational studies. We compare the seasonal variability in observed floe size to fully coupled climate model simulations including a prognostic floe size and thickness distribution and coupled wave model, finding good agreement in regions where modeled ocean surface waves cause sea ice fracture.

Description: The Ocean Model Intercomparison Project (OMIP) is an endorsed project in the Coupled Model Intercomparison Project Phase 6 (CMIP6). OMIP addresses CMIP6 science questions, investigating the origins and consequences of systematic model biases. It does so by providing a framework for evaluating (including assessment of systematic biases), understanding, and improving ocean, sea-ice, tracer, and biogeochemical components of climate and earth system models contributing to CMIP6. Among the WCRP Grand Challenges in climate science (GCs), OMIP primarily contributes to the regional sea level change and near-term (climate/decadal) prediction GCs.OMIP provides (a) an experimental protocol for global ocean/sea-ice models run with a prescribed atmospheric forcing; and (b) a protocol for ocean diagnostics to be saved as part of CMIP6. We focus here on the physical component of OMIP, with a companion paper (Orr et al., 2016) detailing methods for the inert chemistry and interactive biogeochemistry. The physical portion of the OMIP experimental protocol follows the interannual Coordinated Ocean-ice Reference Experiments (CORE-II). Since 2009, CORE-I (Normal Year Forcing) and CORE-II (Interannual Forcing) have become the standard methods to evaluate global ocean/sea-ice simulations and to examine mechanisms for forced ocean climate variability. The OMIP diagnostic protocol is relevant for any ocean model component of CMIP6, including the DECK (Diagnostic, Evaluation and Characterization of Klima experiments), historical simulations, FAFMIP (Flux Anomaly Forced MIP), C4MIP (Coupled Carbon Cycle Climate MIP), DAMIP (Detection and Attribution MIP), DCPP (Decadal Climate Prediction Project), ScenarioMIP, HighResMIP (High Resolution MIP), as well as the ocean/sea-ice OMIP simulations.

Description: Correlation does not necessarily imply a causation, but in climatology and paleoclimatology, correlation is used to identify potential cause-and-effect relationships because linking mechanisms are difficult to observe. Confounding by an often unknown outside variable that drives the sets of observables is one of the major factors that lead to correlations that are not the result of causation. Here we show how autoregressive (AR) models can be used to examine lead-lag relationships – helpful in assessing cause and effect – of paleoclimate variables while addressing two other challenges that are often encountered in paleoclimate data: unevenly spaced data; and switching between regimes at unknown times. Specifically, we analyze multidimensional paleoclimate proxies, sea surface temperature (SST), C37, ∂15N, and %N from the central Peru margin to find their correlations and changes in their variability over the Holocene epoch. The four proxies are sampled at high-resolution but are not synchronously sampled at all possible locations. The multidimensional records are treated as evenly spaced data with missing parts, and the missing values are filled by the Kalman filter expected values. We employ hidden Markov models (HMM) and autoregressive HMM (AR-HMM) to address the potential that the degree of variability and the correlations between in these proxies appears to show changes over time. The HMM, which is not autoregressive, shows instantaneous correlations between observables in two regimes. However, our investigation of lead-lag relationships using the AR-HMM shows that the cross-correlations do not indicate a causal link. Each of the four proxies has predictability on decadal timescales, but none of the proxies is a good predictor of any other, so we hypothesize that a common unobserved variable – or a set of variables – is driving the instantaneous relationships among these four proxies, revealing probable confounding without prior knowledge of potential confounding variable(s). These findings suggest that the variability at this site is remotely driven by processes such as those causing the Pacific Decadal Oscillation, rather than locally driven by processes such as increased or decreased vertical mixing of nutrients.

Description: In sea-ice-covered areas, the sea ice floe size distribution (FSD) plays an important role in many processes affecting the coupled sea-ice-ocean-atmosphere system. Observations of the FSD are spare – traditionally taken via a pain-staking analysis of ice surface photography – and the seasonal and inter-annual evolution of floe size regionally and globally is largely unknown. Frequently, measured FSDs are assessed using a single number, the scaling exponent of the closest power law fit to the observed floe size data, although in the absence of adequate datasets there have been limited tests of this power-law hypothesis. Here we derive and explain a mathematical technique for deriving statistics of the sea ice FSD from polar-orbiting altimeters, satellites with sub-daily return times to polar regions with high along-track resolutions. Applied to the CryoSat-2 radio altimetric record, covering the period from 2010–2018, and incorporating 11 million individual floe samples, we produce the first climatology and seasonal cycle of sea ice floe size statistics. We then perform the first pan-Arctic test of the power law hypothesis, finding limited support in the range of floe sizes typically analyzed in photographic observational studies. We compare the seasonal variability in observed floe size to fully coupled climate model simulations including a prognostic floe size and thickness distribution and coupled wave model, finding good agreement in regions where modeled ocean surface waves cause sea ice fracture.

Description: Marine sediments have greatly improved our understanding of the climate system, but their interpretation often assumes that certain climate mechanisms operate consistently over all timescales of interest and that variability at one or few sample sites is representative of an oceanographic province. In this study, we test these assumptions using modern observations in an idealized manner mimicking paleo-reconstruction to investigate whether sea surface temperature and productivity proxy records in the Southern California Current System can be used to reconstruct Ekman upwelling. The method uses Extended Empirical Orthogonal Function (EEOF) analysis of covariation of alongshore windstress, chlorophyll and sea surface temperature as measured by satellites from 2002 to 2009. We find that EEOF1 does not reflect an Ekman upwelling pattern, but instead much broader California Current processes. EEOF2 and 3 reflect upwelling patterns, but these patterns are timescale dependent and are regional. Thus, the skill of using one site to reconstruct the large scale dominant patterns is spatially dependent. Lastly, we show that using multiple sites and/or multiple variables generally improve field reconstruction. These results together suggest caution is needed when attempting to extrapolate mechanisms that may be important on seasonal time scales (e.g. Ekman upwelling) to deeper time, but also the advantage of having multiple proxy records.