LAUSR.org creates dashboard-style pages of related content for over 1.5 million academic articles. Sign Up to like articles & get recommendations!

Using Artificial Neural Networks to Estimate Cloud-Base Height From AERI Measurement Data

Photo by lukaszlada from unsplash

A new cloud-base height (CBH) inversion algorithm based on infrared hyperspectral radiation using a machine learning algorithm is proposed in this letter. We use the LBLRTM and DISORT model for… Click to show full abstract

A new cloud-base height (CBH) inversion algorithm based on infrared hyperspectral radiation using a machine learning algorithm is proposed in this letter. We use the LBLRTM and DISORT model for forward research. The minimal-redundancy-maximal-relevance (mRMR) algorithm is used to extract the sensitive channels of CBH as the feature vectors. The CBHs measured by Vaisala CL31 ceilometer (VCEIL) are taken as the reference values. The artificial neural network (ANN) method with two hidden layers of 50 and 10, respective is applied to construct the mapping relationship between atmospheric emitted radiance interferometer (AERI) radiation and CBH (ANN-CBH algorithm). The dataset is collected during the period from January 2012 to December 2017 at the Atmospheric Radiation Measurement (ARM) SGP- and NSA-site. Among them, the data from 2012 to 2014 are used as the training set, while the data of 2015–2017 of each site are respectively used as the testing set. Compared with the traditional physical algorithm, the ANN-CBH algorithm has higher accuracy. The correlation coefficients (CCs) between the inversion results of CBH from the ANN-CBH algorithm and the measurement results of the VCEIL are about 0.9 at SGP site and 0.85 at NSA site, while the CC of the CBH inversion results between CO2 slicing algorithm and VCEIL is only about 0.7 and 0.65, respectively. In addition, the experimental results indicate that the ANN-CBH algorithm is less affected by precipitable water vapor (PWV).

Keywords: cbh; cloud base; base height; algorithm; artificial neural; measurement

Journal Title: IEEE Geoscience and Remote Sensing Letters
Year Published: 2022

Link to full text (if available)


Share on Social Media:                               Sign Up to like & get
recommendations!

Related content

More Information              News              Social Media              Video              Recommended



                Click one of the above tabs to view related content.