An Advanced Radiative Transfer and Neural Network Scheme and Evaluation for Estimating Water Vapor Content from MODIS Data

Abstract

This work made an improvement upon and a further evaluation of previous work for estimating water vapor content from near-infrared around 1 µm from MODIS data. The accuracy of RM-NN is determined by the complicated relationship of the geophysical parameters. An advanced scheme is proposed for building different training databases for different seasons in different regions to reduce the complexity. The training database includes three parts. The first part is a simulation database by MODTRAN for different weather conditions, which is made as a basic database; the second part is reliable field measurement data in observation stations; and the third part is the MYD05_L2 product on clear days, which is produced by the standard product algorithm for water vapor content. The comparative analyses based on simulation data indicate that maximum accuracy of single condition could be improved by about 34% relative to the "all conditions" results. Two study regions in China and America are selected as test areas, and the evaluation shows that the mean and the standard deviation of estimation error are about 0.08 g cm −2 and 0.09 g cm −2 , respectively. All the analysis indicates that the advanced scheme can improve the retrieval accuracy of water vapor content, which can make full use of the advantages of previous methods.

An Algorithm to Retrieve Total Precipitable Water Vapor in the Atmosphere from FengYun 3D Medium Resolution Spectral Imager 2 (FY-3D MERSI-2) Data

“…The values obtained for f i are f 16 = 0.208, f 17 = 0.433 and f 18 = 0.359, respectively. Finally, Equation ( 10) can be written as: W = 0.208W 16 + 0.433W 17 + 0.359W 18 (13) Equation ( 13) is a final equation that we developed to estimate the total PWV from MERSI-2 images.…”

“…where σ[R i ] is the standard deviation of radiance ratios (Ri) for all surface covers which are considered in simulation; the denominator Ri (3.5 g/cm 2 ) and Ri (0.3 g/cm 2 ) are the mean value of radiance ratios for wet and dry atmospheres from surface covers considered. To this end, the sensitivity analysis applied to Equation (13) gives a standard deviation of 0.356 g/cm 2 for wet atmospheres to 0.11 g/cm 2 for dry atmospheres.…”

“…Using the data from the above-mentioned remote sensing satellites, we can estimate the PWV at both regional and global scales for various remote sensing applications. Several studies have been published on PWV retrieval from MODIS data over land, cloud, and oceans [13][14][15].…”

“…There are several methods to calculate PWV from satellite data, such as ratio technique, regression slope, split-window difference of the thermal bands, look-up table derived from atmospheric radiative transfer models and the HITRAN2000 spectroscopic database [1,11,13,14,16]. As stated in Sobrino et al [14], the methods for PWV estimation from remote sensing data should meet the following three essential requirements: first, it can minimize the sensibility among various types of the ground surfaces; second, it can minimize the sensibility to the noise resulted from the variations of different channels; third, it can minimize the sensibility caused by variations in other components of the atmosphere [14].…”

Water Vapor Retrieval From MODIS NIR Channels Using Ground-Based GPS Data

“…Inter-comparisons among multi-source water vapor products over other places also show overestimation of MOD05 product [21], [31]- [33], indicating that a better model is needed for water vapor retrieval in MODIS NIR channels. Further development of retrieval algorithm of MODIS was performed using either a pre-calculated look-up table, regression method or an artificial neural network [17], [28], [34], [35]. Empirical correction coefficients were introduced for transmittance calculation to eliminate the wet bias of MODIS NIR products and the RMS deviations were between 0.9 and 2 mm against ground-based observations [36].…”