Pith. sign in

REVIEW 4 major objections 5 minor 51 references

A deep-learning model for predicting daily PM2.5 concentration in response to emission reduction

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read CleanAir, a 3D U-Net trained on 2,416 CMAQ emission-reduction scenarios, reproduces the chemical transport model's daily, gridded PM2.5 responses over China at 36 km resolution with monthly delta R=0.999 and RMSE=0.281 μg/m³, while…

desk verdict CleanAir is a genuinely useful CMAQ surrogate for PM2.5 response to emission cuts over China, with a strong training dataset and honest held-out evaluation; the main gaps are missing artifacts and thin evidence that the four-month, single-year training transfers to all seasons and to 2060 extremes. read the letter →

arxiv 2506.18018 v1 pith:LHNRFEOK submitted 2025-06-22 physics.ao-ph

Shigan Liu , Guannan Geng , Yanfei Xiang , Hejun Hu , Xiaodong Liu , Xiaomeng Huang , Qiang Zhang This is my paper · ORCID
classification physics.ao-ph PACS 92.60.Sz
keywords PM2.5emissionreductionscenarioschemicaltransportmodelemulation3DU-NetdeeplearningsurrogateairqualitypolicyChinaCMAQ
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims that a deep-learning surrogate, CleanAir, can replace a chemical transport model for a specific but policy-critical task: predicting how daily PM2.5 and its chemical components across China respond to precursor emission reductions. Trained on 2,416 CMAQ scenarios spanning four seasons of 2017, it matches CMAQ's emission-induced concentration changes closely (monthly R=0.999, RMSE=0.281 μg/m³; daily R=0.998, RMSE=0.582 μg/m³) and generalizes to unseen meteorological years and lower-emission inventories. The speed gain is the point: one simulated year takes 9 seconds rather than about 4.5 days, so policymakers could screen hundreds of intervention scenarios in the time a single CTM run takes. If the claim holds, CleanAir offers a practical fast emulator for PM2.5 regulation in China across both short-term controls and multi-decade pathways.

What carries the argument

The central mechanism is a Residual Symmetric 3D U-Net that predicts concentration changes relative to a CMAQ baseline and adds them back to the baseline, so the network only learns the emission-response increment rather than the full concentration field. Inputs combine baseline emissions, scenario emissions in three vertical layers, WRF meteorology, biogenic NMVOCs, and pre-simulated baseline concentrations of PM2.5 components and reactive intermediates; an adaptive multi-task weighted loss balances learning across the ten output species and across absolute-vs-delta predictions. The training dataset is built by perturbing 2017 baseline emissions in a 15-dimensional space (five species across three emission layers) using Sobol quasi-random sampling plus targeted NOx/NMVOC and layer-specific scenarios, giving 2,416 scenarios and 74,292 daily samples.

What would settle it

Run CleanAir and CMAQ on a full calendar year of randomized emission-reduction scenarios for a meteorological year not in the training set, using emissions covering the full 0-100% reduction range, and compare daily gridded ΔPM2.5 with the same protocol as the paper's test set; if the grid-level daily R falls well below 0.998 or the RMSE substantially exceeds 0.582 μg/m³, the generalization claim is falsified.

Watch

Extended reading notes

Core claim

CleanAir establishes that a Residual Symmetric 3D U-Net can learn the emission-concentration response surface of CMAQ well enough to act as a fast surrogate for PM2.5 regulation scenarios. Given a CMAQ-simulated baseline concentration field, meteorological fields, biogenic emissions, and an anthropogenic emission scenario at or below 2017 levels, the model outputs daily concentrations of sulfate, nitrate, ammonium, organic matter, black carbon, and other PM2.5 components. On held-out scenarios it reproduces both absolute concentrations and changes relative to baseline, including nonlinear responses such as occasional increases in PM2.5 under emission cuts. The paper demonstrates generalization across 2017-2020 MEIC emission inventories with declining emissions, a February 2017 short-term control case over 57 cities, and DPEC v1.2 long-term pathways to 2060, with health-impact estimates closely tracking CMAQ (mortality R=0.998, NMB=-1.9%).

Load-bearing premise

The load-bearing premise is that CMAQ simulations of four months in 2017 (January, April, July, October) capture enough of the chemical and meteorological variability that a network trained on them reproduces CMAQ's emission-response relationship for all other seasons, years, and very low emission levels down to zero.

Editorial extensions

If this is right

  • Emission-reduction scenario analysis over China can be run at five orders of magnitude lower computational cost, enabling near-real-time assessment of emergency controls and large policy ensembles.
  • CleanAir's performance on 2017-2020 MEIC emissions suggests the trained response surface transfers across meteorological years, not just the four sampled months, as long as emissions stay at or below the 2017 baseline.
  • Long-term health impact calculations under 2020-2060 pathways can be reproduced quickly enough for iterative policy optimization, with mortality estimates matching CMAQ within about 2% normalized mean bias.
  • The architecture and training pipeline are not specific to PM2.5 or China in principle, so the approach could be retrained to emulate other CTM outputs, other regions, or other pollutants.
  • Because the model outputs component-level changes, it can support source-apportionment-style questions about which precursor cuts drive sulfate versus nitrate reductions.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension would replace the CMAQ baseline concentration input with a learned baseline, which would remove the model's remaining dependence on CTM runs and extend its applicability to periods without pre-simulated baselines.
  • The training range is strictly 0-100% reductions from a 2017 baseline, so emission-increase scenarios are outside the model's demonstrated domain; economic-growth or relaxed-regulation cases would require new training data before the speedup could be trusted.
  • The positive ΔPM2.5 values the model reproduces at some grids suggest it has absorbed real nonlinear chemistry; checking whether those locations coincide with CMAQ's known oxidant-limitation regimes would indicate whether the learned nonlinearity is physically grounded or an artifact.
  • The same emulation strategy could, in principle, be applied to ozone and other secondary pollutants if training scenarios were expanded, extending the speedup from PM2.5-only regulation to multi-pollutant co-control.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper presents CleanAir, a deep-learning emulator of the CMAQ chemical transport model for daily, gridded PM2.5 and its chemical components over China at 36 km resolution. The model is a Residual Symmetric 3D U-Net trained on 2,416 CMAQ-simulated emission-reduction scenarios for four months of 2017 (Jan, Apr, Jul, Oct), with inputs including baseline emissions, biogenic emissions, meteorology, and CMAQ-simulated baseline concentrations. The authors report excellent agreement with CMAQ on a held-out test set (monthly ΔPM2.5 R=0.999, RMSE=0.281 μg/m3; daily R=0.998, RMSE=0.582 μg/m3) and a speedup of over 40,000×. They also evaluate the model against observations for 2017–2020 using MEIC emissions, demonstrate a February 2017 short-term control case, and apply the model to 2020–2060 DPEC scenarios with CMAQ benchmarks at 2030 and 2060. The central claim is that CleanAir provides a fast, accurate surrogate for CMAQ across unseen meteorological years and emission scenarios.

Significance. If the generalization claims are substantiated, CleanAir would be a practically valuable tool for policy-oriented air quality scenario analysis over China, enabling rapid exploration of emission control strategies that would otherwise require thousands of CPU-days of CMAQ simulation. The paper's strengths include a large and carefully designed training dataset (Sobol sampling plus targeted scenarios), a physically informed input set (three emission layers, precursor species, baseline chemistry), a multi-task adaptive loss, and evaluation against independent observations in multiple application settings. The central limitations are that the headline accuracy metrics are computed on a test set sharing the same four months and same meteorological year as the training data, and that the evidence for generalization to unseen seasons and years is indirect or aggregated. For these reasons the result is plausible but not yet fully established as stated.

major comments (4)
  1. [Model performance; Training, validation, and test splits (Methods)] The headline metrics (Fig. 2a, Supplementary Fig. 5a) are computed on a test set drawn from the same four months (January, April, July, October) and the same meteorological year (2017) as the training data. Because each test scenario shares WRF meteorology, biogenic emissions, and CMAQ baseline fields with training scenarios from the same month, this evaluation measures interpolation within the training meteorological conditions, not generalization across seasons or years. To support the abstract's claim that CleanAir 'generalizes well across unseen meteorological years and emission patterns,' the paper should report grid-level daily ΔPM2.5 R and RMSE between CleanAir and CMAQ for months not represented in training (e.g., February, March, May, June, August, September, November, December) and for years other than 2017. The February 2017 case study reports only city-level peak concentrations, and the 2017–2020 MEIC evaluation reports daily agreement against observations (R≈0.6) and annual/population-weighted aggregates against CMAQ, neither of which directly validates the emission-response mapping at grid-daily scale for unseen months.
  2. [Modeling skills with different meteorological conditions and emission inventories; Evaluation for MEIC-based…] In the 2017–2020 evaluation, CleanAir's daily PM2.5 correlation against CNEMC observations is reported as R over 0.6 (Section 'Modeling skills...'), which is far lower than the R=0.998 test-set metric. The paper attributes this gap to CMAQ's own bias, but this implies that the model's daily accuracy on unseen conditions is tightly coupled to the accuracy of the CMAQ training target. The comparison between CleanAir and CMAQ for 2017–2020 is presented only as annual spatial distributions and population-weighted annual means (Fig. 3b,c), not as grid-level daily ΔPM2.5 statistics. Without such a comparison, the reader cannot judge whether the emission-response mapping transfers to other meteorological years. Please provide grid-level daily (or at least monthly) ΔPM2.5 R/RMSE between CleanAir and CMAQ for 2018, 2019, and 2020, or explicitly state that such validation was not performed.
  3. [Model capability in long-term pollution intervention] The 2060 projections are benchmarked against CMAQ at only two time points (2030 and 2060) and only for national population-weighted means and spatial patterns (Fig. 5a–c). Because the DPEC scenarios can involve emissions substantially lower than the 2017 baseline, and the training data only cover reductions from that baseline, the claim that CleanAir captures nonlinear secondary-aerosol behavior at very low emission levels is not directly demonstrated. Reporting the full trajectory of CMAQ benchmark comparisons (e.g., at 2030, 2040, 2050, 2060) and, if available, component-level comparisons (sulfate, nitrate, ammonium) would materially strengthen this application. As written, the two-point comparison is a weak test of the model's long-term behavior.
  4. [Data availability (throughout)] No code, training data, model weights, or evaluation scripts are provided, and the manuscript contains no data availability statement. Since the model is an emulator trained on a proprietary CMAQ dataset, the reported results cannot be reproduced or independently evaluated by the community. The authors should either release the trained model and the dataset (or a representative subset), or clearly document the restrictions on availability. This is a load-bearing issue for a paper whose main contribution is a trained model.
minor comments (5)
  1. [Throughout] There are several typographical errors, including 'inevntory' in Section 'Modeling skills...', 'addtion' in Methods, 'condictions' and 'dimentional' in the scenario sampling section, 'matrics' in Evaluation methods, and 'exporsure' in the long-term evaluation section. The manuscript would benefit from a careful proofreading pass.
  2. [Methods, Model inputs and outputs] The list of meteorological 2D variables in Supplementary Table 1 includes 'wind speed at 10 m' twice; one of these is presumably a different variable (e.g., wind direction). Please clarify.
  3. [Methods, Adaptive weighted loss function] The loss function definitions in the equations would be clearer if the normalization constants and summation limits were explicitly defined in the main text, especially the use of N and M as grid dimensions. The current text refers to M and N but the exact shapes are only inferred from the model description.
  4. [Model performance] All reported metrics (R, RMSE, NMB) are point estimates without confidence intervals, bootstrap uncertainties, or any other measure of sampling variability. Given the large number of grid cells and days, even small differences between models may appear significant; providing uncertainty bounds would help readers assess the robustness of the comparisons.
  5. [Discussion] The Discussion correctly lists the limitation that CleanAir cannot handle emission increases and still requires a CMAQ-simulated baseline field for the target year. However, the abstract and introduction do not mention these constraints, and the phrase 'generalizes well across unseen meteorological years and emission patterns' is stronger than the evidence presented. I recommend softening the abstract to reflect that generalization is demonstrated for 2017–2020 MEIC conditions and for a limited set of long-term scenarios.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: CleanAir is an explicit CMAQ emulator, and its headline metrics are computed on held-out scenarios, with independent observational checks.

full rationale

The paper's derivation chain is: CMAQ simulates 2,416 emission reduction scenarios in four sampled months; CleanAir is trained on 60% of these scenarios and evaluated on the held-out 20% test split made at the scenario level. The central comparison (CleanAir vs. CMAQ ΔPM2.5, R=0.999 monthly, RMSE=0.281 μg/m3; R=0.998 daily, RMSE=0.582 μg/m3) is therefore a genuine out-of-sample test of the learned emission-response mapping, not a fitted parameter renamed as a prediction. The model's use of a CMAQ-simulated baseline concentration field as an input is an explicit design choice, with final concentrations obtained by adding a predicted change to the baseline; the predicted change is the learned quantity and is not copied from CMAQ by construction. The 2017-2020 evaluation against CNEMC observations and CMAQ, plus the February 2017 short-term case and the DPEC long-term scenarios, provide independent checks: February and the 2018-2020 years were not in the training set. Several cited datasets (MEIC/MEIC-HR, DPEC, TAP, and chemical-composition observations from Geng et al.) originate from the same group, but they are used as input data or evaluation references rather than as the evidence that the learned mapping is correct, and the CMAQ training target itself is separately evaluated against external CNEMC observations. The remaining concern, that training months (January, April, July, October of 2017) may not cover all seasons, meteorological years, or very-low-emission nonlinear regimes, is a generalization and transferability risk rather than a circularity, because the held-out test set does not by construction determine the cross-season and cross-year results.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the validity of CMAQ as a ground truth, on the representativeness of the four-month training sample, and on the transferability of the emission-response mapping across years and emission levels. The model introduces no new physical entities, but it introduces a large set of fitted parameters and design choices that are not independently documented.

free parameters (4)
  • Neural network weights and biases = not disclosed
    Trained on 74,292 daily samples from 2,416 CMAQ scenarios; the model's predictive accuracy depends entirely on these fitted weights, which are not released.
  • Adaptive loss weights (sigma_conc, sigma_delta_conc, sigma_i) = not reported
    Learned during training to balance tasks; chosen by the algorithm rather than theory.
  • Input history length = 8 days (target day plus previous 7)
    Chosen by hand based on pollutant lifetimes; not optimized or justified with sensitivity analysis.
  • Architecture hyperparameters (depth, channels, kernel sizes) = see Methods
    Chosen based on prior U-Net designs; no ablation study is shown to justify the specific choices.
assumptions (4)
  • domain assumption CMAQ v5.2 with CB05/AERO6 and the stated WRF setup produces accurate enough PM2.5 and precursor fields to serve as training data and evaluation benchmarks.
    The paper validates CMAQ against observations (Supp. Figs 7-9), but the training target is CMAQ itself, so any systematic CMAQ error is inherited by CleanAir.
  • domain assumption The four sampled months (January, April, July, October 2017) capture the full range of meteorological and chemical regimes needed for multi-year generalization.
    Generalization to all months and years relies on this coverage; no data from other months are used in training.
  • domain assumption The emission-response relationship is transferable from 2017 meteorology and MEIC-HR to other years and to MEIC inventories with lower emissions.
    The 2017-2020 tests support this for that period, but the model is applied to 2060 scenarios in Fig. 5 without a mechanistic guarantee.
  • domain assumption The learned delta mapping is additive on top of the provided baseline concentration field; the effect of an emission reduction does not depend on the baseline level in a way the model fails to capture.
    The model architecture computes a change and adds it to the baseline concentration; this additivity is assumed, not derived or stress-tested across very low baseline levels.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A deep-learning model for predicting daily PM2.5 concentration in response to emission reduction." pith.science (2026). https://pith.science/paper/LHNRFEOK

@misc{pith2026250618018,
  author       = {Pith},
  title        = {Pith review of: A deep-learning model for predicting daily PM2.5 concentration in response to emission reduction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LHNRFEOK}},
  note         = {Machine review of arXiv:2506.18018}
}
read the original abstract

Air pollution remains a leading global health threat, with fine particulate matter (PM2.5) contributing to millions of premature deaths annually. Chemical transport models (CTMs) are essential tools for evaluating how emission controls improve air quality and save lives, but they are computationally intensive. Reduced form models accelerate simulations but sacrifice spatial-temporal granularity, accuracy, and flexibility. Here we present CleanAir, a deep-learning-based model developed as an efficient alternative to CTMs in simulating daily PM2.5 and its chemical compositions in response to precursor emission reductions at 36 km resolution, which could predict PM2.5 concentration for a full year within 10 seconds on a single GPU, a speed five orders of magnitude faster. Built on a Residual Symmetric 3D U-Net architecture and trained on more than 2,400 emission reduction scenarios generated by a well-validated Community Multiscale Air Quality (CMAQ) model, CleanAir generalizes well across unseen meteorological years and emission patterns. It produces results comparable to CMAQ in both absolute concentrations and emission-induced changes, enabling efficient, full-coverage simulations across short-term interventions and long-term planning horizons. This advance empowers researchers and policymakers to rapidly evaluate a wide range of air quality strategies and assess the associated health impacts, thereby supporting more responsive and informed environmental decision-making.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

51 extracted references · 45 canonical work pages

  1. [1]

    S., Fnais, M., Giannadaki, D

    Lelieveld, J., Evans, J. S., Fnais, M., Giannadaki, D. & Pozzer, A. The contribution of outdoor air pollution sources to premature mortality on a global scale. Nature 525, 367–371 (2015)

    work page 2015

  2. [2]

    & Unger, N

    Yue, X. & Unger, N. Fire air pollution reduces global terrestrial productivity. Nat. Commun. 9, 5413 (2018)

    work page 2018

  3. [3]

    Cheesman, A. W. et al. Reduced productivity and carbon drawdown of tropical forests from ground-level ozone exposure. Nat. Geosci. 17, 1003–1007 (2024)

    work page 2024

  4. [4]

    Murray, C. J. L. et al. Global burden of 87 risk factors in 204 countries and territories, 1990 – 2019: A systematic analysis for the global burden of disease study 2019. Lancet 396, 1223– 1249 (2020)

    work page 2020

  5. [5]

    West, J. J. et al. Co-benefits of mitigating global greenhouse gas emissions for future air quality and human health. Nat. Clim. Change 3, 885–889 (2013)

    work page 2013

  6. [6]

    Lelieveld, J. et al. Effects of fossil fuel and total anthropogenic emission removal on public health and climate. Proc. Natl. Acad. Sci. U.S.A. 116, 7192–7197 (2019)

    work page 2019

  7. [7]

    Geng, G. et al. Efficacy of China’s clean air actions to tackle PM2.5 pollution between 2013 and 2020. Nat. Geosci. 17, 987–994 (2024)

    work page 2024

  8. [8]

    Bey, I. et al. Global modeling of tropospheric chemistry with assimilated meteorology: Model description and evaluation. Journal of Geophysical Research: Atmospheres 106, 23073–23095 (2001)

    work page 2001

Show all 51 references
  1. [9]

    Appel, K. W. et al. Description and evaluation of the community multiscale air quality (CMAQ) modeling system version 5.1. Geosci. Model Dev. 10, 1703–1732 (2017)

    2017

  2. [10]

    J., Santillana, M., Wang, X

    Shen, L., Jacob, D. J., Santillana, M., Wang, X. & Chen, W. An adaptive method for speeding 14 up the numerical integration of chemical mechanisms in atmospheric chemistry models: application to GEOS-Chem version 12.0.0. Geosci. Model Dev. 13, 2475–2486 (2020)

    2020

  3. [11]

    Muller, N. Z. & Mendelsohn, R. Measuring the damages of air pollution in the united states. J. Environ. Econ. Manage. 54, 1–14 (2007)

    2007

  4. [12]

    Amann, M. et al. Cost-effective control of air quality and greenhouse gases in Europe: Modeling and policy applications. Environ. Model. Softw. 26, 1489–1501 (2011)

    2011

  5. [13]

    Zhang, D. et al. Machine learning approaches reveal highly heterogeneous air quality co - benefits of the energy transition. iScience 26, 107652 (2023)

    2023

  6. [14]

    Baker, K. R. et al. Source–Receptor Relationships Between Precursor Emissions and O3 and PM2.5 Air Pollution Impacts. Environ. Sci. Technol. (2023) doi:10.1021/acs.est.3c03317

    2023 doi

  7. [15]

    Xing, J. et al. Deep Learning for Prediction of the Air Quality Response to Emission Changes. Environ. Sci. Technol. 54, 8589–8600 (2020)

    2020

  8. [16]

    Fang, L. et al. Development of a regional feature selection -based machine learning system (RFSML v1.0) for air pollution forecasting over china. Geosci. Model Dev. 15, 7791–7807 (2022)

    2022

  9. [17]

    & Huang, X.-F

    Cheng, Y ., He, L.-Y . & Huang, X.-F. Development of a high -performance machine learning model to predict ground ozone pollution in typical cities of China. J. Environ. Manage. 299, 113670 (2021)

    2021

  10. [18]

    Liu, Z. et al. Interaction Patterns between Climate Action and Air Cleaning in China: A Two- Way Evaluation Based on an Ensemble Learning Approach. Environ. Sci. Technol. 56, 9291– 9301 (2022)

    2022

  11. [19]

    X., Jang, C., Zhu, Y

    Xing, J., Wang, S. X., Jang, C., Zhu, Y . & Hao, J. M. Nonlinear response of ozone to precursor 15 emission changes in China: a modeling study using response surface methodology. Atmos. Chem. Phys. 11, 5027–5044 (2011)

    2011

  12. [20]

    W., Hill, J

    Tessum, C. W., Hill, J. D. & Marshall, J. D. InMAP: A model for air pollution interventions. PLoS ONE 12, e0176131 (2017)

    2017

  13. [21]

    Kurth, T. et al. FourCastNet: Accelerating global high -resolution weather forecasting using adaptive fourier neural operators. in Proceedings of the Platform for Advanced Scientific Computing Conference 1–11 (ACM, Davos Switzerland, 2023). doi:10.1145/3592979.3593412

    2023

  14. [22]

    Lam, R. et al. Learning skillful medium-range global weather forecasting. Science eadi2336 (2023) doi:10.1126/science.adi2336

    2023 doi

  15. [23]

    Bi, K. et al. Accurate medium -range global weather forecasting with 3D neural networks. Nature 619, 533–538 (2023)

    2023

  16. [24]

    Zhang, Y . et al. Skilful nowcasting of extreme precipitation with NowcastNet. Nature 619, 526– 532 (2023)

    2023

  17. [25]

    Kochkov, D. et al. Neural general circulation models for weather and climate. Nature 632, 1060–1066 (2024)

    2024

  18. [26]

    Bodnar, C. et al. A foundation model for the Earth system. Nature (2025) doi:10.1038/s41586- 025-09005-y

    2025 doi

  19. [27]

    & Seung, H

    Lee, K., Zung, J., Li, P., Jain, V . & Seung, H. S. Superhuman Accuracy on the SNEMI3D Connectomics Challenge. Preprint at http://arxiv.org/abs/1706.00120 (2017)

    2017 arXiv

  20. [28]

    Zheng, B. et al. Mapping anthropogenic emissions in China at 1 km spatial resolution and its application in air quality modeling. Sci. Bull. 66, 612–620 (2021)

    2021

  21. [29]

    Guenther, A. B. et al. The Model of Emissions of Gases and Aerosols from Nature version 2.1 16 (MEGAN2.1): an extended and updated framework for modeling biogenic emissions. Geoscientific Model Development 5, 1471–1492 (2012)

    2012

  22. [30]

    An, Z. et al. Severe haze in northern China: A synergy of anthropogenic emissions and atmospheric processes. Proc. Natl. Acad. Sci. U.S.A. 116, 8657–8666 (2019)

    2019

  23. [31]

    The distribution of points in a cube and the accurate evaluation of integrals (in Russian) Zh

    Sobol, I. The distribution of points in a cube and the accurate evaluation of integrals (in Russian) Zh. Vychisl. Mat. i Mater. Phys 7, 784–802 (1967)

    1967

  24. [32]

    & Cipolla, R

    Kendall, A., Gal, Y . & Cipolla, R. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics. Preprint at https://doi.org/10.48550/arXiv.1705.07115 (2018)

    2018 doi arXiv

  25. [33]

    Huang, X. et al. Solving partial differential equations with point source based on physics - informed neural networks. Preprint at https://doi.org/10.48550/arXiv.2111.01394 (2021)

    2021 doi arXiv

  26. [34]

    Emery, C. et al. Recommendations on statistics and benchmarks to assess photochemical model performance. Journal of the Air & Waste Management Association 67, 582–598 (2017)

    2017

  27. [35]

    Huang, L. et al. Recommendations on benchmarks for numerical air quality model applications in China – Part 1: PM2.5 and chemical species. Atmospheric Chemistry and Physics 21, 2725– 2743 (2021)

    2021

  28. [36]

    Wang, H. et al. Declining short-term emission control opportunity for major events in Chinese cities. Nat. Cities 2, 434–446 (2025)

    2025

  29. [37]

    Cheng, J. et al. A synergistic approach to air pollution control and carbon neutrality in China can avoid millions of premature deaths annually by 2060. One Earth 6, 978–989 (2023)

    2023

  30. [38]

    Reichstein, M. et al. Deep learning and process understanding for data -driven Earth system science. Nature 566, 195–204 (2019)

    2019

  31. [39]

    Karniadakis, G. E. et al. Physics-informed machine learning. Nat. Rev. Phys 3, 422–440 (2021). 17

    2021

  32. [40]

    & Mauzerall, D

    Wang, X. & Mauzerall, D. L. Evaluating impacts of air pollution in China on public health: Implications for future air pollution and energy policies. Atmos. Environ. 40, 1706–1721 (2006)

    2006

  33. [41]

    Zhang, Q. et al. Drivers of improved PM2.5 air quality in China from 2013 to 2017. Proc. Natl. Acad. Sci. U.S.A. 116, 24463–24469 (2019)

    2019

  34. [42]

    Zhang, Q. et al. Transboundary health impacts of transported global air pollution and international trade. Nature 543, 705–709 (2017)

    2017

  35. [43]

    Zhang, Q. et al. Asian emissions in 2006 for the NASA INTEX-B mission. Atmos. Chem. Phys. (2006)

    2006

  36. [44]

    Baek, J., Hu, Y ., Odman, M. T. & Russell, A. G. Modeling secondary organic aerosol in CMAQ using multigenerational oxidation of semi -volatile organic compounds: SECONDARY ORGANIC AEROSOL IN CMAQ. J. Geophys. Res.: Atmos. 116, n/a-n/a (2011)

    2011

  37. [45]

    Zheng, B. et al. Heterogeneous chemistry: a mechanism missing in current models to explain secondary inorganic aerosol formation during the January 2013 haze episode in North China. Atmos. Chem. Phys. 15, 2031–2049 (2015)

    2015

  38. [46]

    & Henze, G

    Burhenne, S., Jacob, D. & Henze, G. P. Sampling based on sobol′ sequences for monte carlo techniques applied to building simulations. in 1816–1823 (Sydney, 2011)

    2011

  39. [47]

    R., Millar, J

    Renardy, M., Joslyn, L. R., Millar, J. A. & Kirschner, D. E. To sobol or not to sobol? The effects of sampling schemes in systems biology applications. Math. Biosci. 337, 108593 (2021)

    2021

  40. [48]

    Wallace, J. M. & Hobbs, P. V . Atmospheric Science: An Introductory Survey. vol. 92 (Elsevier, 2006)

    2006

  41. [49]

    Leong, D. et al. World population prospects 2019. Department of economic and social affairs population dynamics. New york (NY): United nations; 2019 (https://population. un. 18 org/wpp/download/, accessed 20 september 2020). The decade of healthy ageing. Geneva: world health o...

    2018

  42. [50]

    is used to reduce the grid size to 127 × 172. The following Padding layer then adjusts the 224 grid to 128 × 176 to meet the max -pooling requirement; (2) The second branch consists of a 225 single Conv3D layer, where all input data are merged and passed through this layer. Th...

  43. [51]

    conc ” and “ ∆conc

    266 The total loss function ( Ltotal) consists of three terms, as shown below: (1) a loss function 267 that evaluates the accuracy of concentration estimation task ( Lconc); (2) a loss function that 268 evaluates the accuracy of concentration change estimation task (L∆conc); a...

    2017

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.