Karhunen-Loeve expansion analysis of uncertainties in cloud microphysical property retrievals
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
- Beijing Normal Univ. (China)
Here, this study proposes a methodology of quantifying uncertainties for cloud retrievals on model resolution to facilitate the comparison with model outputs. Primary component analysis is applied to reduce the dimension of random variables (up to a factor of 50) and reveal the cross correlations in the input data, making large sampling computationally feasible and uncertainty quantification accurate and reliable. Our approach has the capability of parameterizing input uncertainties and attributing the uncertainties in the retrieval output to each individual source, which allows sensitivity analysis of cloud retrieval algorithms and provides directions for improving observation instruments as well as strategies. We applied the method to characterize uncertainties in cloud ice water content (IWC) retrieved from the DOE Atmospheric Radiation Measurement (ARM) programs baseline cloud microphysical retrieval algorithm (MICROBASE). We test it with a selected ice cloud case observed on 9 March 2000 at the ARM Southern Great Plains site during its 2000 cloud intensive observing period. The test results indicate that (1) uncertainties in the output retrieved by MICROBASE are comparable amongst different retrievals; (2) The mean values obtained by our UQ method are closer to the aircraft data with less errors compared to the direct ensemble average; (3) Ice water path (IWP) generally incurred larger uncertainty in optically thin ice clouds and there was more variability in vertical in the retrieved IWC; and (4) Uncertainties in the output are mainly due to the interactions among different modes of ARM radar pro files.
- Research Organization:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE National Nuclear Security Administration (NNSA)
- DOE Contract Number:
- AC52-07NA27344
- OSTI ID:
- 1557071
- Report Number(s):
- LLNL-JRNL--655829; 776673
- Journal Information:
- Proposed Journal Article, unpublished, Journal Name: Proposed Journal Article, unpublished Vol. 2014; ISSN 9999-9999
- Publisher:
- See Research Organization
- Country of Publication:
- United States
- Language:
- English
Similar Records
A variance-based decomposition and global sensitivity index method for uncertainty quantification: Application to retrieved ice cloud properties
A comparison of cloud microphysical quantities with forecasts from cloud prediction models