Fast dimension-reduced climate model calibration and the effect of data aggregation
From MaRDI portal
Publication:400583
DOI10.1214/14-AOAS733zbMATH Open1454.62438arXiv1303.1382MaRDI QIDQ400583FDOQ400583
Authors: Won Chang, Murali Haran, Roman Olson, Klaus Keller 
Publication date: 22 August 2014
Published in: The Annals of Applied Statistics (Search for Journal in Brave)
Abstract: How will the climate system respond to anthropogenic forcings? One approach to this question relies on climate model projections. Current climate projections are considerably uncertain. Characterizing and, if possible, reducing this uncertainty is an area of ongoing research. We consider the problem of making projections of the North Atlantic meridional overturning circulation (AMOC). Uncertainties about climate model parameters play a key role in uncertainties in AMOC projections. When the observational data and the climate model output are high-dimensional spatial data sets, the data are typically aggregated due to computational constraints. The effects of aggregation are unclear because statistically rigorous approaches for model parameter inference have been infeasible for high-resolution data. Here we develop a flexible and computationally efficient approach using principal components and basis expansions to study the effect of spatial data aggregation on parametric and projection uncertainties. Our Bayesian reduced-dimensional calibration approach allows us to study the effect of complicated error structures and data-model discrepancies on our ability to learn about climate model parameters from high-dimensional data. Considering high-dimensional spatial observations reduces the effect of deep uncertainty associated with prior specifications for the data-model discrepancy. Also, using the unaggregated data results in sharper projections based on our climate model. Our computationally efficient approach may be widely applicable to a variety of high-dimensional computer model calibration problems.
Full work available at URL: https://arxiv.org/abs/1303.1382
Recommendations
- Computer Model Calibration Using High-Dimensional Output
- Computational methods for parameter estimation in climate models
- Climate projections using Bayesian model averaging and space-time dependence
- Bayesian Projected Calibration of Computer Models
- Uncertainty quantification for computer models with spatial output using calibration-optimal bases
Computational methods for problems pertaining to statistics (62-08) Factor analysis and principal components; correspondence analysis (62H25) Inference from spatial processes (62M30) Gaussian processes (60G15) Applications of statistics to environmental and related topics (62P12) Geostatistics (86A32) Meteorology and atmospheric physics (86A10)
Cites Work
- Algorithm 611: Subroutines for Unconstrained Minimization Using a Model/Trust-Region Approach
- Fixed Rank Kriging for Very Large Spatial Data Sets
- Probabilistic Forecasts, Calibration and Sharpness
- Design and analysis of computer experiments. With comments and a rejoinder by the authors
- Bayesian calibration of computer models. (With discussion)
- Gaussian Predictive Process Models for Large Spatial Data Sets
- Strictly and non-strictly positive definite functions on spheres
- Approximating Likelihoods for Large Spatial Data Sets
- Approximate Likelihood for Large Irregularly Spaced Spatial Data
- An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach
- Computer Model Calibration Using High-Dimensional Output
- Modularization in Bayesian analysis, with emphasis on analysis of computer models
- Finding structure with randomness: probabilistic algorithms for constructing approximate matrix decompositions
- Space and space-time modeling using process convolutions
- Augmented Implicitly Restarted Lanczos Bidiagonalization Methods
- Computer model validation with functional output
- Fast dimension-reduced climate model calibration and the effect of data aggregation
- On principal components and regression: a statistical explanation of a natural phenomenon
- A THEORETICAL STUDY OF THE DIFFUSION OF TRACER GAS IN AN AIRWAY
- V<scp>ERTICAL</scp> M<scp>IXING</scp>, E<scp>NERGY</scp>, <scp>AND THE</scp> G<scp>ENERAL</scp> C<scp>IRCULATION OF THE</scp> O<scp>CEANS</scp>
- Comment on article by Sansó et al.
- Parameter estimation for computationally intensive nonlinear regression with an application to climate modeling
Cited In (14)
- Computer Model Calibration with Time Series Data Using Deep Learning and Quantile Regression
- Temperatures in transient climates: improved methods for simulations with evolving temporal covariances
- Fast dimension-reduced climate model calibration and the effect of data aggregation
- Saving storage in climate ensembles: a model-based stochastic approach
- Axially symmetric models for global data: a journey between geostatistics and stochastic generators
- High-fidelity hurricane surge forecasting using emulation and sequential experiments
- Flood hazard model calibration using multiresolution model output
- Constructing a Simulation Surrogate with Partially Observed Output
- Ice model calibration using semicontinuous spatial data
- Efficient Stochastic Generators with Spherical Harmonic Transformation for High-Resolution Global Climate Simulations from CESM2-LENS2
- Large scale variable fidelity surrogate modeling
- Fast Nonseparable Gaussian Stochastic Process With Application to Methylation Level Interpolation
- A fast particle-based approach for calibrating a 3-D model of the Antarctic ice sheet
- Gaussian Process Prediction using Design-Based Subsampling
Uses Software
This page was built for publication: Fast dimension-reduced climate model calibration and the effect of data aggregation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q400583)
