Comment on acp-2022-634 DOI Creative Commons
Boming Liu, Xin Ma, Jianping Guo

et al.

Published: Oct. 7, 2022

Abstract. Accurate estimation of wind speed at turbine hub height is significance for energy assessment and exploitation. Nevertheless, the traditional power law method (PLM) generally estimates hub-height by assuming a constant exponent between surface speed. This inevitably leads to significant uncertainties in estimating profile especially under unstable conditions. To minimize uncertainties, we here use machine learning algorithm known as random forest (RF) estimate heights such 120 m (WS120), 160 m class="inline-formula">160), 200 m class="inline-formula">200). These go beyond mast limit 100–120 m. The radar profiler synoptic observations Qingdao station from May 2018 August 2020 are used key inputs develop RF model. A deep analysis model construction has been performed ensure its applicability. Afterwards, PLM retrieve WS120, class="inline-formula">160, class="inline-formula">200. comparison analyses both models against radiosonde measurements. At 120 m, shows relatively higher correlation coefficient R 0.93 smaller RMSE 1.09 m s−1, compared with 0.89 1.50 m s−1 PLM. Notably, metrics determine performance decline sharply model, opposed stable variation suggests exhibits advantages over because considers well factors friction heat transfer. diurnal seasonal variations class="inline-formula">200 then analyzed. hourly class="inline-formula">120 large during daytime 09:00 16:00 local solar time (LST) reach peak 14:00 LST. spring winter low summer autumn. class="inline-formula">160 similar those class="inline-formula">120. Finally, investigated absolute percentage error (APE) density different heights. In vertical direction, APE gradually increased increases. Overall, some limitations height. which combines more or auxiliary data, suitable estimation. findings obtained have great implications development utilization industry future.

Language: Английский

Estimating hub-height wind speed based on a machine learning algorithm: implications for wind energy assessment DOI Creative Commons
Boming Liu, Xin Ma, Jianping Guo

et al.

Atmospheric chemistry and physics, Journal Year: 2023, Volume and Issue: 23(5), P. 3181 - 3193

Published: March 10, 2023

Abstract. Accurate estimation of wind speed at turbine hub height is significance for energy assessment and exploitation. Nevertheless, the traditional power law method (PLM) generally estimates hub-height by assuming a constant exponent between surface speed. This inevitably leads to significant uncertainties in estimating profile especially under unstable conditions. To minimize uncertainties, we here use machine learning algorithm known as random forest (RF) estimate heights such 120 m (WS120), 160 (WS160), 200 (WS200). These go beyond mast limit 100–120 m. The radar profiler synoptic observations Qingdao station from May 2018 August 2020 are used key inputs develop RF model. A deep analysis model construction has been performed ensure its applicability. Afterwards, PLM retrieve WS120, WS160, WS200. comparison analyses both models against radiosonde measurements. At m, shows relatively higher correlation coefficient R 0.93 smaller RMSE 1.09 s−1, compared with 0.89 1.50 s−1 PLM. Notably, metrics determine performance decline sharply model, opposed stable variation suggests exhibits advantages over because considers well factors friction heat transfer. diurnal seasonal variations WS200 then analyzed. hourly WS120 large during daytime 09:00 16:00 local solar time (LST) reach peak 14:00 LST. spring winter low summer autumn. WS160 similar those WS120. Finally, investigated absolute percentage error (APE) density different heights. In vertical direction, APE gradually increased increases. Overall, some limitations height. which combines more or auxiliary data, suitable estimation. findings obtained have great implications development utilization industry future.

Language: Английский

Citations

39
Validation, inter-comparison, and usage recommendation of six latest VIIRS and MODIS aerosol products over the ocean and land on the global and regional scales DOI Creative Commons
Xin Su,

Mengdan Cao,

Lunche Wang

et al.

The Science of The Total Environment, Journal Year: 2023, Volume and Issue: 884, P. 163794 - 163794

Published: April 29, 2023

Language: Английский

Citations

19
Identification of dust aerosols, their sources, and the effect of soil moisture in Central Asia DOI Creative Commons
Jie Liu, Jianli Ding, Xiaohang Li

et al.

The Science of The Total Environment, Journal Year: 2023, Volume and Issue: 868, P. 161575 - 161575

Published: Jan. 11, 2023

Language: Английский

Citations

18
Validation and diurnal variation evaluation of MERRA-2 multiple aerosol properties on a global scale DOI
Xin Su, Yuhang Huang, Lunche Wang

et al.

Atmospheric Environment, Journal Year: 2023, Volume and Issue: 311, P. 120019 - 120019

Published: Aug. 10, 2023

Language: Английский

Citations

18
A Two-Stage Machine Learning Algorithm for Retrieving Multiple Aerosol Properties Over Land: Development and Validation DOI

Mengdan Cao,

Ming Zhang, Xin Su

et al.

IEEE Transactions on Geoscience and Remote Sensing, Journal Year: 2023, Volume and Issue: 61, P. 1 - 17

Published: Jan. 1, 2023

Satellite-based aerosol optical property retrieval over land, especially size-related parameters, is challenging. This study proposed a novel two-stage machine learning (ML) algorithm for retrieving depth (AOD), Ångström exponent (AE), fine mode fraction (FMF), and AOD (FAOD)) land using MODIS observed reflectance. The new ML consists of three steps: (1) first, all samples extracted from AERONET measurements were used to train the model, (2) then, reduce extreme estimation bias divided low-value high-value models, respectively, (3) finally, models integrated into final based on weight interpolation. Independent site network validation results show that has Pearson correlation coefficient (R) 0.894 (0.638, 0.661, 0.865) root mean square error (RMSE) 0.146 (0.258, 0.245, 0.153) (AE, FMF, FAOD) retrieval, which significantly outperforms metrics operational products, with RMSE 0.130-0.156 (0.536-0.569, 0.313, 0.191). inter-comparison products shows spatial patterns AOD, AE, FAOD are in good agreement those POLDER products. These illustrate performance transferability indicate ability methods be applied multispectral instruments (such as MODIS) retrieve multiple properties.

Language: Английский

Citations

15
Towards long-term, high-accuracy, and continuous satellite total and fine-mode aerosol records: Enhanced Land General Aerosol (e-LaGA) retrieval algorithm for VIIRS DOI
Lunche Wang, Xin Su, Yi Wang

et al.

ISPRS Journal of Photogrammetry and Remote Sensing, Journal Year: 2024, Volume and Issue: 214, P. 261 - 281

Published: July 1, 2024

Language: Английский

Citations

5
Global evaluation of Fengyun-3 MERSI dark target aerosol retrievals over land DOI Creative Commons
Leiku Yang,

Weiqian Ji,

Pei Xin

et al.

International Journal of Digital Earth, Journal Year: 2024, Volume and Issue: 17(1), P. 1 - 24

Published: April 25, 2024

The Medium Resolution Spectral Image (MERSI) is a MODIS-like sensor aboard Fengyun-3 satellite. first version of MERSI aerosol algorithm has been developed based on MODIS dark target (DT) algorithm, with modified models for estimating surface reflectance and an adjusted inland water masking method to release haze aerosols. This study applies DT the global observations from upgraded (MERSI-II) Fengyun-3D. And then, Aerosol Optical Depth (AOD) results year 2019–2020 are validated against Robotic Network (AERONET) data. In addition, analyses spatial distribution error characteristics MERSI-II retrievals presented. overall validation demonstrates that perform well globally, correlation coefficient 0.877 67.1% matchups within Expected Error envelope ± (0.05 + 0.2τ), which close statistic metrics products. AODs exhibit similar trends dependence. Moreover, two revealed in retrieval performance at site regional scales, as analysis monthly averages. These findings indicate success ported algorithm.

Language: Английский

Citations

5
Research on the distribution and influencing factors of fine mode aerosol optical depth (AODf) in China DOI
Haifeng Xu, Jinji Ma,

Wenhui Luo

et al.

Atmospheric Environment, Journal Year: 2024, Volume and Issue: 334, P. 120721 - 120721

Published: Oct. 1, 2024

Language: Английский

Citations

4
Global estimates of gap-free and fine-scale CO2 concentrations during 2014–2020 from satellite and reanalysis data DOI Creative Commons
Lingfeng Zhang, Tongwen Li, Jingan Wu

et al.

Environment International, Journal Year: 2023, Volume and Issue: 178, P. 108057 - 108057

Published: June 24, 2023

Carbon dioxide (CO2) is a crucial greenhouse gas with substantial effects on climate change. Satellite-based remote sensing commonly used approach to detect CO2 high precision but often suffers from extensive spatial gaps. Thus, the limited availability of data makes global carbon stocktaking challenging. In this paper, gap-free column-averaged dry-air mole fraction (XCO2) dataset resolution 0.1° 2014 2020 generated by deep learning-based multisource fusion, including satellite and reanalyzed XCO2 products, vegetation index data, meteorological data. Results indicate accuracy for 10-fold cross-validation (R2 = 0.959 RMSE 1.068 ppm) ground-based validation 0.964 1.010 ppm). Our has advantages fine compared reanalysis as well that other studies. Based dataset, our analysis reveals interesting findings regarding spatiotemporal pattern over globe national-level growth rates CO2. This fine-scale potential provide support understanding cycle making reduction policy, it can be freely accessed at https://doi.org/10.5281/zenodo.7721945.

Language: Английский

Citations

9
Assessment of the high-resolution estimations of global and diffuse solar radiation using WRF-Solar DOI Creative Commons

Yunbo Lu,

Lunche Wang, Jiaojiao Zhou

et al.

Advances in Climate Change Research, Journal Year: 2023, Volume and Issue: 14(5), P. 720 - 731

Published: Oct. 1, 2023

Compared with physical models, WRF-Solar, as an excellent numerical forecasting model, includes abundant novel cloud and dynamical processes, which enablesenable the high-frequency output of radiation components are urgently needed by solar energy industry. However, popularisation WRF-Solar in a wide range applications, such estimation diffuse radiation, suffers from unpredictable influences aerosol optical property parameters. This study assessed accuracy improved weather prediction (WRF-Solar) model simulating global radiation. Aerosol properties at 550 nm, were provided moderate resolution imaging spectroradiometer, used input to analyse differences accuracies obtained with/without input. The sensitivity zenith angle (SZA) was analysed. results show superiority WRF-Dudhia terms their root mean square error (RMSE) absolute (MAE). coefficients determination between revealed no statistically significant difference, values greater than 0.9 for parent nested domains. In addition, relative RMSE (RRMSE%) reached 46.60%. experiment on negative bias but attained slightly lower higher correlation coefficient WRF-Dudhia. WRF-Solar-simulated under clear sky conditions poorer, RMSE, RRMSE, percentage MAE 181.93 W m−2, 170.52%, 93.04% 138 respectively. Based Himawari-8 data, statistical thickness (COT) cloudy days that overestimated COTs 20. Moreover, when depth or equal 0.8, also difference 58.57 m−2. errors simulations exhibited dependence SZA. dispersion degree deviation increased gradually decrease Thus, serves tool can provide high temporal high-spatial-resolution data photovoltaic power. Studies should explore improvement cumulus parameterisation schemes enhance component conditions.

Language: Английский

Citations

9