Land Use Regression Modelling: An Air Pollution Monitor Location Optimization Approach
Bibliographic record
Abstract
Land use regression (LUR) modelling associates measured pollution levels to land cover characteristics for spatial interpolation. LUR is commonly applied in the domain of air pollution and less commonly for noise, soil and water pollution. Model outputs can be applied to estimate human or environmental exposure. The model’s predictor variables include land use attributes, such as land cover and transportation network characteristics that are calculated within spatial buffers of the monitoring locations. LUR models can be used to predict values at unobserved locations.The application of spatial interpolation attempts to ensure that data are not extrapolated beyond the bounds of the observed values. However, our research identifies that without guaranteeing monitoring data be collected in all land use classes and conditions, it is possible with LUR to actually be extrapolating data and still be within the 2-dimensional spatial boundaries of the monitoring locations. This potential extrapolation occurs because the interpolation is based on a multi-dimensional space that sits upon the 2-D plane, which creates a new set of boundaries for the interpolation.In this paper, we define and demonstrate the potential problem of ensuring LUR models interpolate within both the 2-D spatial domain and the multi-dimensional space that is applied in LUR modelling. We then demonstrate a solution to this problem in a simulated dataset and in an empirical dataset. First, we identify all possible monitoring locations. Second, the objective function is defined with the goal of selecting monitoring locations to maximize the variation across land use conditions. A heuristic search technique is applied to identify good potential solutions. The location of monitors is identified using an ensemble of potential optimum solutions.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".