Satellite Remote Sensing of Global Land Surface Temperature: Definition, Methods, Products, and Applications

Reviews of Geophysics, Journal Year: 2022, Volume and Issue: 61(1)

Published: Dec. 24, 2022

Abstract Land surface temperature (LST) is a crucial parameter that reflects land–atmosphere interaction and has thus attracted wide interest from geoscientists. Owing to the rapid development of Earth observation technologies, remotely sensed LST playing an increasingly essential role in various fields. This review aims summarize progress estimation algorithms accelerate its further applications. Thus, we briefly most‐used thermal infrared (TIR) algorithms. More importantly, this provides comprehensive collection widely used TIR‐based products offers important insights into uncertainties these with respect different land cover conditions via systematic intercomparison analysis several representative products. In addition discussion on product accuracy, address problems related spatial discontinuity, spatiotemporal incomparability, short time span current by introducing most effective methods. With aim overcoming challenges available products, much been made developing seamless data, which significantly promotes successful applications field evapotranspiration soil moisture estimation, agriculture drought monitoring, environment anomaly climate change. Overall, encompasses recent advances state‐of‐the‐art at temporal scales, identifies critical research needs directions advance optimize retrieval methods, application improve understanding dynamics exchanges.

Language: Английский

---

Significant changes in the chemical compositions and sources of PM2.5 in Wuhan since the city lockdown as COVID-19

The Science of The Total Environment, Journal Year: 2020, Volume and Issue: 739, P. 140000 - 140000

Published: June 5, 2020

Language: Английский

---

Separating emission and meteorological contributions to long-term PM<sub>2.5</sub> trends over eastern China during 2000–2018

Atmospheric chemistry and physics, Journal Year: 2021, Volume and Issue: 21(12), P. 9475 - 9496

Published: June 23, 2021

Abstract. The contribution of meteorology and emissions to long-term PM2.5 trends is critical for air quality management but has not yet been fully analyzed. Here, we used the combination a machine learning model, statistical method, chemical transport model quantify meteorological impacts on pollution during 2000–2018. Specifically, first developed two-stage prediction with synthetic minority oversampling technique improve satellite-based estimates over highly polluted days, thus allowing us better characterize effects haze events. Then two methods examine PM2.5: generalized additive (GAM) driven by full-coverage daily retrievals Weather Research Forecasting/Community Multiscale Air Quality (WRF/CMAQ) modeling system. We found good agreements between GAM estimations CMAQ monthly scale (correlation coefficient 0.53–0.72). Both revealed dominant role emission changes in trend concentration China 2000–2018, notable influence from condition. interannual variabilities meteorology-associated were dominated fall winter conditions, when regional stagnant stable conditions more likely happen events frequently occurred. From 2000 2018, became unfavorable across North Plain central beneficial control southern part, e.g., Yangtze River Delta. meteorology-adjusted eastern (denoted East figures) peaked 2006 2011, mainly peaks primary gas precursors these years. Although trends, meteorology-driven anomalies also contributed −3.9 % 2.8 annual mean concentrations estimated GAM. contributions even higher regionally, −6.3 4.9 Beijing-Tianjin-Hebei region, −5.1 4.3 Fenwei Plain, −4.8 Delta, −25.6 12.3 Pearl Considering remarkable possible worsening northern part where severe population clustered, stricter clean actions are needed avoid future.

Language: Английский

---

Variational Bayesian Network with Information Interpretability Filtering for Air Quality Forecasting

Mathematics, Journal Year: 2023, Volume and Issue: 11(4), P. 837 - 837

Published: Feb. 7, 2023

Air quality plays a vital role in people’s health, and air forecasting can assist decision making for government planning sustainable development. In contrast, it is challenging to multi-step forecast accurately due its complex nonlinear caused by both temporal spatial dimensions. Deep models, with their ability model strong nonlinearities, have become the primary methods forecasting. However, because of lack mechanism-based analysis, uninterpretability makes decisions risky, especially when decisions. This paper proposes an interpretable variational Bayesian deep learning information self-screening PM2.5 Firstly, based on factors related concentration, e.g., temperature, humidity, wind speed, distribution, etc., multivariate data screening structure was established catch as much helpful possible. Secondly, layer implanted network optimize selection input variables. Further, following implantation layer, gated recurrent unit (GRU) constructed overcome distribution achieve accurate The high accuracy proposed method verified Beijing, China, which provides effective way, multiple determined using technology.

Language: Английский

---

Global synthesis of two decades of research on improving PM2.5 estimation models from remote sensing and data science perspectives

Earth-Science Reviews, Journal Year: 2023, Volume and Issue: 241, P. 104461 - 104461

Published: May 29, 2023

Language: Английский

---

Application of the Lasso regularisation technique in mitigating overfitting in air quality prediction models

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 2, 2025

Language: Английский

---

Long-Term Retrospective Predicted Concentration of PM2.5 in Upper Northern Thailand Using Machine Learning Models

Toxics, Journal Year: 2025, Volume and Issue: 13(3), P. 170 - 170

Published: Feb. 27, 2025

This study aims to build, for the first time, a model that uses machine learning (ML) approach predict long-term retrospective PM2.5 concentrations in upper northern Thailand, region impacted by biomass burning and transboundary pollution. The dataset includes PM10 levels, fire hotspots, critical meteorological data from 1 January 2011 31 December 2020. ML techniques, namely multi-layer perceptron neural network (MLP), support vector (SVM), multiple linear regression (MLR), decision tree (DT), random forests (RF), were used construct prediction models. best was selected considering root mean square error (RMSE), (MPE), relative (RPE) (the lower, better), coefficient of determination (R2) bigger, better). Our found model-based RF technique using PM10, CO2, O3, air pressure, rainfall, humidity, temperature, wind direction, speed performs when predicting concentration with an RMSE 6.82 µg/m3, MPE 4.33 RPE 22.50%, R2 0.93. this research could further studies effects on human health related issues.

Language: Английский

---

Construction of a virtual PM2.5 observation network in China based on high-density surface meteorological observations using the Extreme Gradient Boosting model

Environment International, Journal Year: 2020, Volume and Issue: 141, P. 105801 - 105801

Published: May 30, 2020

With increasing public concerns on air pollution in China, there is a demand for long-term continuous PM2.5 datasets. However, it was not until the end of 2012 that China established national observation network. Before that, satellite-retrieved aerosol optical depth (AOD) frequently used as primary predictor to estimate surface PM2.5. Nevertheless, AOD often encounter incomplete daily coverage due its sampling frequency and interferences from cloud, which greatly affect representation these AOD-based Here, we constructed virtual ground-based network at 1180 meteorological sites across using Extreme Gradient Boosting (XGBoost) model with high-density observations major predictors. Cross-validation XGBoost showed strong robustness high accuracy estimation (monthly) 2018, R2, root-mean-square error (RMSE) mean absolute values 0.79 (0.92), 15.75 μg/m3 (6.75 μg/m3) 9.89 (4.53 μg/m3), respectively. Meanwhile, find visibility plays dominant role terms relative importance variables model, accounting 39.3% overall importance. We then use data year 2017 assess predictive capability model. Results capable accurately hindcast historical monthly (R2 = 0.80, RMSE 14.75 seasonal 0.86, 12.28 annual 0.81, 10.10 levels. In general, newly based shows great potential reconstructing ~1000 China. It will be benefit filling gaps data, well other environmental studies including epidemiology.

Language: Английский

---

Quantifying influences of natural and anthropogenic factors on vegetation changes using structural equation modeling: A case study in Jiangsu Province, China

Journal of Cleaner Production, Journal Year: 2020, Volume and Issue: 280, P. 124330 - 124330

Published: Sept. 29, 2020

Language: Английский

---

Estimation of hourly full-coverage PM2.5 concentrations at 1-km resolution in China using a two-stage random forest model

Atmospheric Research, Journal Year: 2020, Volume and Issue: 248, P. 105146 - 105146

Published: July 18, 2020

Language: Английский

---