Uncertainty quantification and global sensitivity analysis of the Los Alamos sea ice model

Urrego-Blanco, J. R.; Urban, Nathan M.; Hunke, Elizabeth; Turner, Adrian K.; Jeffery, Nicole

doi:10.1002/2015jc011558

Cited by 54 publications

(87 citation statements)

References 81 publications

(82 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The model baseline simulation underestimates irradiance below the sea ice. Changing the default value of the sigma coefficient for snow grain (R_snow)—a delta‐Eddington parameter that gives the standard deviation of the snow grain size [ Urrego‐Blanco et al , ]—has a strong effect on simulated light intensity. The best fit was obtained with R_snow = 0.8 (Figure a).…”

Section: Resultsmentioning

confidence: 99%

“…This work does not include a thorough sensitivity analysis, which generally implies changing each parameter at a time, running the model and comparing the results with a reference simulation. However, as already noted by Urrego‐Blanco et al [], such an approach cannot identify interactions among parameters and assumes linearity and additivity. The same authors used a global variance approach with Sobol sequences to efficiently sample the parameter space of the CICE model.…”

Section: Discussionmentioning

confidence: 99%

“…SYI_Sim_1: standard CICE parameter values, except for the sigma coefficient for snow grain (R_snow) [ Urrego‐Blanco et al , ] that was changed from 1.5 to 0.8 (see section 4.4); SYI_Sim_2: “random” algal motion implemented; SYI_Sim_3: “deterministic” algal motion implemented; SYI_Sim_4: “deterministic” algal motion combined with lowered Si:N ratio and reducing the half saturation constant for silicon uptake as above…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Sea ice thermohaline dynamics and biogeochemistry in the Arctic Ocean: Empirical and model results

Duarte

Meyer

Olsen

et al. 2017

J. Geophys. Res. Biogeosci.

Self Cite

View full text Add to dashboard Cite

Large changes in the sea ice regime of the Arctic Ocean have occurred over the last decades justifying the development of models to forecast sea ice physics and biogeochemistry. The main goal of this study is to evaluate the performance of the Los Alamos Sea Ice Model (CICE) to simulate physical and biogeochemical properties at time scales of a few weeks and to use the model to analyze ice algal bloom dynamics in different types of ice. Ocean and atmospheric forcing data and observations of the evolution of the sea ice properties collected from 18 April to 4 June 2015, during the Norwegian young sea ICE expedition, were used to test the CICE model. Our results show the following: (i) model performance is reasonable for sea ice thickness and bulk salinity; good for vertically resolved temperature, vertically averaged Chl a concentrations, and standing stocks; and poor for vertically resolved Chl a concentrations. (ii) Improving current knowledge about nutrient exchanges, ice algal recruitment, and motion is critical to improve sea ice biogeochemical modeling. (iii) Ice algae may bloom despite some degree of basal melting. (iv) Ice algal motility driven by gradients in limiting factors is a plausible mechanism to explain their vertical distribution. (v) Different ice algal bloom and net primary production (NPP) patterns were identified in the ice types studied, suggesting that ice algal maximal growth rates will increase, while sea ice vertically integrated NPP and biomass will decrease as a result of the predictable increase in the area covered by refrozen leads in the Arctic Ocean.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Sea ice thermohaline dynamics and biogeochemistry in the Arctic Ocean: Empirical and model results

Duarte

Meyer

Olsen

et al. 2017

J. Geophys. Res. Biogeosci.

Self Cite

View full text Add to dashboard Cite

show abstract

“… Note. Optimized CICE parameters (ksno, rsnw_mlt, R_snw) are based largely on Urrego‐Blanco et al () and Urrego‐Blanco et al (). The E3SMv0‐HiLAT parameter values for ksno and rsnw_mlt are final values; see Table for details of their adjustment during the course of integration.…”

Section: Model Descriptionmentioning

confidence: 99%

“…They ranked the configurations using a variance‐based distance metric, to capture the spatial character of model skill as compared to different observational data sets of sea ice concentration and thickness. Using this distance metric and building on the work of Urrego‐Blanco et al (), who identified parameters for which the simulated sea ice response is more sensitive, the baseline parameter values for the sea ice model component in the coupled E3SMv0‐HiLAT system were taken to be those that had allowed the simulation to best match observations of sea ice concentration (EUMETSAT, NASA Team 1 and 2, and Bootstrap data sets) and sea ice thickness (Unified Sea Ice Thickness Data Record), as documented in Urrego‐Blanco et al ().…”

Section: Model Descriptionmentioning

confidence: 99%

E3SMv0‐HiLAT: A Modified Climate System Model Targeted for the Study of High‐Latitude Processes

Hecht

Veneziani

Weijer

et al. 2019

J Adv Model Earth Syst

Self Cite

View full text Add to dashboard Cite

We document the configuration, tuning, and evaluation of a modified version of the Community Earth System Model version 1 (Hurrell et al., 2013, https://doi.org/10.1175/BAMS-D-12), introduced here as E3SMv0‐HiLAT, intended for study of high‐latitude processes. E3SMv0‐HiLAT incorporates changes to the atmospheric model affecting aerosol transport to high northern latitudes and to reduce shortwave cloud bias over the Southern Ocean. An updated sea ice model includes biogeochemistry that is coupled to an extended version of the ocean model's biogechemistry. This enables cloud nucleation to depend on the changing marine emissions of aerosol precursors, which may be expected in scenarios with strongly changing sea ice extent, oceanic stratification and associated nutrient availability, and atmospheric state. An evaluation of the basic preindustrial state of E3SMv0‐HiLAT is presented in order to ensure that its climate is adequate to support future experimentation. Additional capability is not achieved without some cost, relative to the extraordinarily well‐tuned model from which it was derived. In particular, a reduction of bias in cloud forcing achieved over the Southern Hemisphere also allows for greater Southern Ocean sea ice extent, a tendency that has been partially but not fully alleviated through experimentation and tuning. The most interesting change in the behavior of the model may be its response to greenhouse gas forcing: While the climate sensitivity is found to be essentially unchanged from that of Community Earth System Model version 1, the adjusted radiative forcing has increased from within one standard deviation above that of Coupled Model Intercomparison Project Phase 5 models to nearly two standard deviations.

show abstract

Validation of sea ice models using an uncertainty‐based distance metric for multiple model variables

et al. 2017

Self Cite

View full text Add to dashboard Cite

We implement a variance‐based distance metric (Dn) to objectively assess skill of sea ice models when multiple output variables or uncertainties in both model predictions and observations need to be considered. The metric compares observations and model data pairs on common spatial and temporal grids improving upon highly aggregated metrics (e.g., total sea ice extent or volume) by capturing the spatial character of model skill. The Dn metric is a gamma‐distributed statistic that is more general than the χ2 statistic commonly used to assess model fit, which requires the assumption that the model is unbiased and can only incorporate observational error in the analysis. The Dn statistic does not assume that the model is unbiased, and allows the incorporation of multiple observational data sets for the same variable and simultaneously for different variables, along with different types of variances that can characterize uncertainties in both observations and the model. This approach represents a step to establish a systematic framework for probabilistic validation of sea ice models. The methodology is also useful for model tuning by using the Dn metric as a cost function and incorporating model parametric uncertainty as part of a scheme to optimize model functionality. We apply this approach to evaluate different configurations of the standalone Los Alamos sea ice model (CICE) encompassing the parametric uncertainty in the model, and to find new sets of model configurations that produce better agreement than previous configurations between model and observational estimates of sea ice concentration and thickness.

show abstract

Uncertainty quantification and global sensitivity analysis of the Los Alamos sea ice model

Cited by 54 publications

References 81 publications

Sea ice thermohaline dynamics and biogeochemistry in the Arctic Ocean: Empirical and model results

Sea ice thermohaline dynamics and biogeochemistry in the Arctic Ocean: Empirical and model results

E3SMv0‐HiLAT: A Modified Climate System Model Targeted for the Study of High‐Latitude Processes

Validation of sea ice models using an uncertainty‐based distance metric for multiple model variables

Contact Info

Product

Resources

About