Estimation of PM2.5 concentration in southern China using stacked machine learning models based on GNSS and radiosonde precipitable water vapor

Advances in Space Research, Journal Year: 2025, Volume and Issue: unknown

Published: May 1, 2025

Language: Английский

---

State-of-art in modelling particulate matter (PM) concentration: a scoping review of aims and methods

Environment Development and Sustainability, Journal Year: 2024, Volume and Issue: unknown

Published: April 2, 2024

Abstract Air pollution is the one of most significant environmental risks to health worldwide. An accurate assessment population exposure would require a continuous distribution measuring ground-stations, which not feasible. Therefore, efforts are spent in implementing air-quality models. However, complex scenario emerges, with spread many different solutions, and consequent struggle comparison, evaluation replication, hindering definition state-of-art. Accordingly, aim this scoping review was analyze latest scientific research on modelling, focusing particulate matter, identifying widespread solutions trying compare them. The mainly focused, but limited to, machine learning applications. initial set 940 results published 2022 were returned by search engines, 142 resulted analyzed. Three main modelling scopes identified: correlation analysis, interpolation forecast. Most studies relevant east south-east Asia. majority models multivariate, including (besides ground stations) meteorological information, satellite data, land use and/or topography, more. 232 algorithms tested across (either as single-blocks or within ensemble architectures), only 60 more than once. A performance comparison showed stronger evidence towards Random Forest particular when included architectures. it must be noticed that varied significantly according experimental set-up, indicating no overall best solution can identified, case-specific necessary.

Language: Английский

---

Exploring high-resolution near-surface CO concentrations based on Himawari-8 top-of-atmosphere radiation data: Assessing the distribution of city-level CO hotspots in China

Atmospheric Environment, Journal Year: 2023, Volume and Issue: 312, P. 120021 - 120021

Published: Aug. 11, 2023

Language: Английский

---

Estimation of the ground-level SO2 concentration in eastern China based on the LightGBM model and Himawari-8 TOAR

Frontiers in Earth Science, Journal Year: 2023, Volume and Issue: 10

Published: Jan. 5, 2023

Sulfur dioxide (SO 2 ) is one of the main pollutants in China’s atmosphere, but spatial distribution ground-based SO monitors too sparse to provide a complete coverage. Therefore, obtaining high resolution concentration great significance for pollution control. In this study, based on LightGBM machine learning model, combined with top-of-atmosphere radiation (TOAR) Himawari-8 and additional data such as meteorological factors geographic information, temporal TOAR-SO estimation model eastern China (97–136°E, 15–54°N) established. TOAR are two variables that contribute most both their feature importance values exceed 30%. The has performance estimating ground-level concentrations 10-fold cross validation R (RMSE) 0.70 (16.26 μg/m 3 ), 0.75 (12.51 0.96 (2.75 0.97 (2.16 (1.71 when hourly, daily, monthly, seasonal, annual average . Taking North study area, estimated. showed downward trend since 2016 decreased 15.19 2020. good agreement between ground measured estimated highlights capability advantage using monitor spatiotemporal variations Eastern China.

Language: Английский

---

Reconstruction of MODIS LST Under Cloudy Conditions by Integrating Himawari-8 and AMSR-2 Data Through Deep Forest Method

IEEE Transactions on Geoscience and Remote Sensing, Journal Year: 2024, Volume and Issue: 62, P. 1 - 17

Published: Jan. 1, 2024

Land Surface Temperature (LST) plays a crucial role in Earth's energy balance and ecosystems. Various gap-filling methods have been developed to reconstruct seamless LST datasets deal with the effect of data gaps caused by cloud cover, however, existing studies mainly focus on reconstruction under clear-sky conditions, rather than generating actual cloud-impacted LST. This study treats MODIS cloud-free pixels as known sample points. The deep forest (DF) algorithm is employed establish nonlinear relationship model between Himawari-8 cumulative downward surface shortwave radiation (DSSR), AMSR2 brightness temperature (TB) data, other influencing factors points, well applied cloud-covered obtain underlying pixels, thereby reconstructing real over Yellow River source region. feasibility this approach lies fact that DSSR incorporates impact coverage incoming solar radiation, there exists correlation TB results for January, April, July, October 2021 were validated against situ 0 cm measurements from five meteorological stations. show reconstructed exhibits high consistency in-situ measurements, R ² 0.86, Bias 0.62 K, RMSE 4.48 K. demonstrate effectiveness using microwave reconstruction, accurately representing

Language: Английский

---

A comprehensive review delineates advancements in retrieving particulate matter utilising satellite aerosol optical depth: Parameter consideration, data processing, models development and future perspectives

Atmospheric Research, Journal Year: 2024, Volume and Issue: 308, P. 107514 - 107514

Published: June 5, 2024

Language: Английский

---

The Reconstruction of FY-4A and FY-4B Cloudless Top-of-Atmosphere Radiation and Full-Coverage Particulate Matter Products Reveals the Influence of Meteorological Factors in Pollution Events

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(18), P. 3363 - 3363

Published: Sept. 10, 2024

By utilizing top-of-atmosphere radiation (TOAR) data from China’s new generation of geostationary satellites (FY-4A and FY-4B) along with interpretable machine learning models, near-surface particulate matter concentrations in China were estimated, achieving hourly temporal resolution, 4 km spatial 100% coverage. First, the cloudless TOAR matched modeled solar products ERA5 dataset to construct estimate a fully covered under assumed clear-sky conditions, which increased coverage 20–30% 100%. Subsequently, this was applied matter. The analysis demonstrated that (R2 = 0.83) performed better than original 0.76). Additionally, using feature importance scores SHAP values, impact meteorological factors air mass trajectories on increase PM10 PM2.5 during dust events investigated. haze indicated main driving changes included pressure, temperature, boundary layer height. concentration obtained exhibit high spatiotemporal resolution. Combined data-driven learning, they can effectively reveal influencing China.

Language: Английский

---

Real-time mapping of gapless 24-hour surface PM10 in China

National Science Review, Journal Year: 2024, Volume and Issue: 12(2)

Published: Dec. 9, 2024

Large-scale mapping of surface coarse particulate matter (PM10) concentration remains a key focus for air quality monitoring. Satellite aerosol optical depth (AOD)-based data fusion approaches decouple the non-linear AOD-PM10 relationship, enabling high-resolution PM10 acquisition, but are limited by spatial incompleteness and absence nighttime data. Here, gridded visibility-based real-time retrieval (RT-SPMR) framework China is introduced, addressing gap in seamless hourly within 24-hour cycle. This utilizes multisource inputs dynamically updated machine-learning models to produce 6.25-km Cross-validation showed that RT-SPMR model's daily accuracy surpassed prior studies. Additionally, through rolling iterative validation experiments, model exhibited strong generalization capability stability, demonstrating its suitability operational deployment. Taking record-breaking dust storm as an example, proved effective tracking fine-scale evolution intrusion process, especially under-observed areas. Consequently, provides comprehensive monitoring pollution China, has potential improve forecasting enhancing initial field.

Language: Английский

---

The Role of Machine Learning in Enhancing Particulate Matter Estimation: A Systematic Literature Review

Technologies, Journal Year: 2024, Volume and Issue: 12(10), P. 198 - 198

Published: Oct. 15, 2024

As urbanization and industrial activities accelerate globally, air quality has become a pressing concern, particularly due to the harmful effects of particulate matter (PM), notably PM2.5 PM10. This review paper presents comprehensive systematic assessment machine learning (ML) techniques for estimating PM concentrations, drawing on studies published from 2018 2024. Traditional statistical methods often fail account complex dynamics pollution, leading inaccurate predictions, especially during peak pollution events. In contrast, ML approaches have emerged as powerful tools that leverage large datasets capture nonlinear, intricate relationships among various environmental, meteorological, anthropogenic factors. synthesizes findings 32 studies, demonstrating techniques, ensemble models, significantly enhance estimation accuracy. However, challenges remain, including data quality, need diverse balanced datasets, issues related feature selection, spatial discontinuity. identifies critical research gaps proposes future directions improve model robustness applicability. By advancing understanding applications in monitoring, this seeks contribute developing effective strategies mitigating protecting public health.

Language: Английский

---

Predicting PM10 Concentrations Using Evolutionary Deep Neural Network and Satellite-Derived Aerosol Optical Depth

Mathematics, Journal Year: 2023, Volume and Issue: 11(19), P. 4145 - 4145

Published: Sept. 30, 2023

Predicting particulate matter with a diameter of 10 μm (PM10) is crucial due to its impact on human health and the environment. Today, aerosol optical depth (AOD) offers high resolution wide coverage, making it viable way estimate PM concentrations. Recent years have also witnessed in-creasing promise in refining air quality predictions via deep neural network (DNN) models, out-performing other techniques. However, learning weights biases DNN task classified as an NP-hard problem. Current approaches such gradient-based methods exhibit significant limitations, risk becoming ensnared local minimal within multi-objective loss functions, substantial computational requirements, requirement for continuous objective functions. To tackle these challenges, this paper introduces novel approach that combines binary gray wolf optimizer (BGWO) improve optimization models pollution prediction. The BGWO algorithm, inspired by behavior wolves, used optimize both weight bias DNN. In proposed BGWO, sigmoid function transfer adjust position wolves. This study gathers meteorological data, topographic information, PM10 satellite images. Data preparation includes tasks noise removal handling missing data. evaluated through cross-validation using metrics correlation rate, R square, root-mean-square error (RMSE), accuracy. effectiveness BGWO-DNN framework compared seven machine (ML) models. experimental evaluation method data shows superior performance traditional ML BGWO-DNN, CapSA-DNN, BBO-DNN achieved lowest RMSE values 16.28, 19.26, 20.74, respectively. Conversely, SVM-Linear GBM algorithms displayed highest levels error, yielding 36.82 32.50, algorithm secured R2 (88.21%) accuracy (93.17%) values, signifying Additionally, between predicted actual model surpasses observes relatively stable during spring summer, contrasting fluctuations autumn winter.

Language: Английский

---