A comparison of correlation-length estimation methods for the objective analysis of surface pollutants at Environment and Climate Change Canada
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
UNLABELLED: An objective analysis is one of the main components of data assimilation. By combining observations with the output of a predictive model we combine the best features of each source of information: the complete spatial and temporal coverage provided by models, with a close representation of the truth provided by observations. The process of combining observations with a model output is called an analysis. To produce an analysis requires the knowledge of observation and model errors, as well as its spatial correlation. This paper is devoted to the development of methods of estimation of these error variances and the characteristic length-scale of the model error correlation for its operational use in the Canadian objective analysis system. We first argue in favor of using compact support correlation functions, and then introduce three estimation methods: the Hollingsworth-Lönnberg (HL) method in local and global form, the maximum likelihood method (ML), and the [Formula: see text] diagnostic method. We perform one-dimensional (1D) simulation studies where the error variance and true correlation length are known, and perform an estimation of both error variances and correlation length where both are non-uniform. We show that a local version of the HL method can capture accurately the error variances and correlation length at each observation site, provided that spatial variability is not too strong. However, the operational objective analysis requires only a single and globally valid correlation length. We examine whether any statistics of the local HL correlation lengths could be a useful estimate, or whether other global estimation methods such as by the global HL, ML, or [Formula: see text] should be used. We found in both 1D simulation and using real data that the ML method is able to capture physically significant aspects of the correlation length, while most other estimates give unphysical and larger length-scale values. IMPLICATIONS: This paper describes a proposed improvement of the objective analysis of surface pollutants at Environment and Climate Change Canada (formerly known as Environment Canada). Objective analyses are essentially surface maps of air pollutants that are obtained by combining observations with an air quality model output, and are thought to provide a complete and more accurate representation of the air quality. The highlight of this study is an analysis of methods to estimate the model (or background) error correlation length-scale. The error statistics are an important and critical component to the analysis scheme.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it