An Improved Method Combining ANN and 1D-Var for the Retrieval of Atmospheric Temperature Profiles from FY-4A/GIIRS Hyperspectral Data

Huang, Po-Hao; Guo, Qiang; Han, Changpei; Zhang, Chunming; Yang, Tianhang; Huang, Shuo

doi:10.3390/rs13030481

Cited by 17 publications

(10 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Retrieving atmospheric temperature profiles from satellite remote sensing data is crucial for numerical weather prediction (NWP)-based forecasting and climate research [1]. With the development of hyperspectral infrared (IR) sounders with thousands of narrow spectral channels [2], high-spatial-resolution vertical detection of the atmosphere can be achieved [3][4][5]. The hyperspectral IR sounder is capable of sensing changes in thermodynamic variables and tracing gas species as well as critical climate variables [6], and has shown a significant positive impact on global numerical weather prediction (NWP) applications [7,8].…”

Section: Introductionmentioning

confidence: 99%

“…Cai et al [40] retrieved atmospheric temperature and humidity profiles based on ANN using the GIIRS hyperspectral data, and achieved good results. Huang et al [4] used 1D-Var and ANN to retrieve atmospheric temperature profiles from the GIIRS hyperspectral data. The results showed that using ANN yields higher accuracy than using 1D-Var in the pressure range of 800-1000 hPa.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Retrieval of Atmospheric Temperature Profiles from FY-4A/GIIRS Hyperspectral Data Based on TPE-MLP: Analysis of Retrieval Accuracy and Influencing Factors

Xu,

Han,

Gao

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

In this study, a novel method for retrieving atmospheric temperature profiles with tree-structured Parzen estimator (TPE) and multilayer perceptron (MLP) algorithms was proposed, using FY-4A/GIIRS (Geosynchronous Interferometric Infrared Sounder) and ERA5 data. Firstly, by adding solar altitude angle, satellite zenith angle, 2m temperature, and surface temperature to the input layer of MLP, there is an improvement in retrieval accuracy. Secondly, TPE is effective in optimizing the hyper-parameters of MLP, and a set of optimized hyper-parameters is obtained through iterative optimization. Thirdly, comparing the retrieved temperature profiles with ERA5 data, we found that retrieval accuracy is influenced by detector, signal-to-noise ratio, terrain, solar altitude angle, satellite zenith angle, and the horizontal temperature gradient. The mean biases of the two adjacent detectors show significant differences, and the retrieval accuracy of the center detectors is greater than that of the north and south sides. The retrieval accuracy is relatively poor in areas with high terrain and large satellite zenith angle. There is a monthly variation in the retrieval accuracy due to the horizontal temperature gradient and signal-to-noise ratio and a significant diurnal variation due to solar altitude angle and signal-to-noise ratio. Compared to in situ sounding data, the mean biases vary from −0.56 K to 0.60 K, and the standard deviations vary from 1.26 K to 2.17 K. The analysis of factors influencing retrieval accuracy provides important insights into improving the ability to retrieve atmospheric temperatures from geostationary hyperspectral IR sounder observations for near real-time (NRT) applications.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Retrieval of Atmospheric Temperature Profiles from FY-4A/GIIRS Hyperspectral Data Based on TPE-MLP: Analysis of Retrieval Accuracy and Influencing Factors

Xu,

Han,

Gao

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Besides, the channel centers and Instrumental Spectral Response Functions (ISRFs) of the FY‐4A/GIIRS and FY‐4B/GIIRS are identical for the same wave number channels. The observations from the FY‐4A/GIIRS and FY‐4B/GIIRS can be used for the retrieval of three‐dimensional atmospheric temperature and humidity fields (Huang et al ., 2021), the data assimilation for the numerical weather prediction (Yin et al ., 2021), and the monitoring of trace gases like ammonia (NH 3 ) (Clarisse et al ., 2021). Moreover, the European Meteorological Satellite (EUMESAT) plans to launch their next generation of geostationary satellites in 2024, that is, the Meteosat Third Generation's Sounding satellites (MTG‐S).…”

Section: Introductionmentioning

confidence: 99%

Performances between the FY‐4A/GIIRS and FY‐4B/GIIRS Long‐Wave InfraRed (LWIR) channels under clear‐sky and all‐sky conditions

Niu

Zhang

Yang³

et al. 2023

Quart J Royal Meteoro Soc

View full text Add to dashboard Cite

The observation‐minus‐background (O − B) bias characteristics are compared for the Long‐Wave InfraRed (LWIR) channels of the Geostationary Interferometric Infrared Sounder (GIIRS) on board Fengyun‐4A (FY‐4A) and Fengyun‐4B (FY‐4B) satellites under clear‐sky and all‐sky conditions. Under the clear‐sky condition, the data quality of the FY‐4B/GIIRS is better than that of the FY‐4A/GIIRS. Specifically, the standard deviations (<1 K) of the O − B values for FY‐4B/GIIRS LWIR channels are smaller than those (>1 K) of the FY‐4A/GIIRS, especially in carbon dioxide (CO2) and ozone (O3) absorption bands. The diurnal variations of the mean O − B biases for the O3 band of the FY‐4B/GIIRS are less than those of FY‐4A/GIIRS. To our surprise, the diurnal variations of O − B biases for the FY‐4B/GIIRS and FY‐4A/GIIRS channels show an obvious antiphase characteristic in the CO2 band, which could be caused by diurnal environmental field variations from the different geostationary (GEO) platforms. In addition, the FY‐4B/GIIRS channel 7 (703.75 cm−1) has smaller pixel‐dependent biases than those of FY‐4A/GIIRS. Under the all‐sky condition, the observation and cloudy simulation (Bcloudy$$ {\mathrm{B}}_{\mathrm{cloudy}} $$) for channels 7 (703.75 cm−1), 36 (721.875 cm−1), and 320 (900 cm−1) of the FY‐4B/GIIRS are more consistent than those of FY‐4A/GIIRS with smaller root‐mean‐square errors (RMSEs), which also indicates a better observation quality in the cloudy area.

show abstract

“…The fine spectral resolution of images is obtained through remote sensing sensors which utilize multiple adjacent narrow spectral bands to construct an HSI image [3]. Hyperspectral sensors generate wavelength bands in the visible and infrared spectra [4,5] and analyzing these spectral bands will provide target information in a better manner [6]. The HSI image pixels are composed of electromagnetic radiation bands [7] and hypercubes are used to represent the HSI in three dimensions.…”

Section: Introductionmentioning

confidence: 99%

Dimensionality Reduction Using Optimized Self-Organized Map Technique for Hyperspectral Image Classification

Srinivasan¹,

Rajakumar²

2023

Computer Systems Science and Engineering

View full text Add to dashboard Cite

The high dimensionalhyperspectral image classification is a challenging task due to the spectral feature vectors. The high correlation between these features and the noises greatly affects the classification performances. To overcome this, dimensionality reduction techniques are widely used. Traditional image processing applications recently propose numerous deep learning models. However, in hyperspectral image classification, the features of deep learning models are less explored. Thus, for efficient hyperspectral image classification, a depth-wise convolutional neural network is presented in this research work. To handle the dimensionality issue in the classification process, an optimized self-organized map model is employed using a water strider optimization algorithm. The network parameters of the self-organized map are optimized by the water strider optimization which reduces the dimensionality issues and enhances the classification performances. Standard datasets such as Indian Pines and the University of Pavia (UP) are considered for experimental analysis. Existing dimensionality reduction methods like Enhanced Hybrid-Graph Discriminant Learning (EHGDL), local geometric structure Fisher analysis (LGSFA), Discriminant Hyper-Laplacian projection (DHLP), Group-based tensor model (GBTM), and Lower rank tensor approximation (LRTA) methods are compared with proposed optimized SOM model. Results confirm the superior performance of the proposed model of 98.22% accuracy for the Indian pines dataset and 98.21% accuracy for the University of Pavia dataset over the existing maximum likelihood classifier, and Support vector machine (SVM).

show abstract

An Improved Method Combining ANN and 1D-Var for the Retrieval of Atmospheric Temperature Profiles from FY-4A/GIIRS Hyperspectral Data

Cited by 17 publications

References 25 publications

Retrieval of Atmospheric Temperature Profiles from FY-4A/GIIRS Hyperspectral Data Based on TPE-MLP: Analysis of Retrieval Accuracy and Influencing Factors

Retrieval of Atmospheric Temperature Profiles from FY-4A/GIIRS Hyperspectral Data Based on TPE-MLP: Analysis of Retrieval Accuracy and Influencing Factors

Performances between the FY‐4A/GIIRS and FY‐4B/GIIRS Long‐Wave InfraRed (LWIR) channels under clear‐sky and all‐sky conditions

Dimensionality Reduction Using Optimized Self-Organized Map Technique for Hyperspectral Image Classification

Contact Info

Product

Resources

About