scholarly journals An Overview of Neural Network Methods for Predicting Uncertainty in Atmospheric Remote Sensing

2021 ◽  
Vol 13 (24) ◽  
pp. 5061
Author(s):  
Adrian Doicu ◽  
Alexandru Doicu ◽  
Dmitry S. Efremenko ◽  
Diego Loyola ◽  
Thomas Trautmann
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Inverse Problem ◽  
Radiative Transfer Model ◽  
Transfer Model ◽  
Input Noise ◽  
Cloud Parameters ◽  
First Case ◽  
Atmospheric Remote Sensing ◽  
Network Methods

In this paper, we present neural network methods for predicting uncertainty in atmospheric remote sensing. These include methods for solving the direct and the inverse problem in a Bayesian framework. In the first case, a method based on a neural network for simulating the radiative transfer model and a Bayesian approach for solving the inverse problem is proposed. In the second case, (i) a neural network, in which the output is the convolution of the output for a noise-free input with the input noise distribution; and (ii) a Bayesian deep learning framework that predicts input aleatoric and model uncertainties, are designed. In addition, a neural network that uses assumed density filtering and interval arithmetic to compute uncertainty is employed for testing purposes. The accuracy and the precision of the methods are analyzed by considering the retrieval of cloud parameters from radiances measured by the Earth Polychromatic Imaging Camera (EPIC) onboard the Deep Space Climate Observatory (DSCOVR).

scholarly journals Improving thermodynamic profile retrievals from microwave radiometers by including Radio Acoustic Sounding System (RASS) observations

2021 ◽  
Author(s):  
Irina V. Djalalova ◽  
David D. Turner ◽  
Laura Bianco ◽  
James M. Wilczak ◽  
James Duncan ◽  
...  
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Pearson Correlation ◽  
Radiative Transfer Model ◽  
Transfer Model ◽  
Iterative Approach ◽  
Linear Quadratic ◽  
Acoustic Sounding ◽  
Microwave Radiometers ◽  
Mean Square Errors

Abstract. Thermodynamic profiles are often retrieved from the multi-wavelength brightness temperature observations made by microwave radiometers (MWRs) using regression methods (linear, quadratic approaches), artificial intelligence (neural networks), or physical-iterative methods. Regression and neural network methods are tuned to mean conditions derived from a climatological dataset of thermodynamic profiles collected nearby. In contrast, physical-iterative retrievals use a radiative transfer model starting from a climatologically reasonable value of temperature and water vapor, with the model run iteratively until the derived brightness temperatures match those observed by the MWR within a specified uncertainty. In this study, a physical-iterative approach is used to retrieve temperature and humidity profiles from data collected during XPIA (eXperimental Planetary boundary layer Instrument Assessment), a field campaign held from March to May 2015 at NOAA's Boulder Atmospheric Observatory (BAO) facility. During the campaign, several passive and active remote sensing instruments as well as in-situ platforms were deployed and evaluated to determine their suitability for the verification and validation of meteorological processes. Among the deployed remote sensing instruments was a multi-channel MWR, as well as two radio acoustic sounding systems (RASS), associated with 915-MHz and 449-MHz wind profiling radars. Having the possibility to combine the information provided by the MWR and RASS systems, in this study the physical-iterative approach is tested with different observational inputs: first using data from surface sensors and the MWR in different configurations, and then including data from the RASSs. These temperature retrievals are also compared to those derived by a neural network method, assessing their relative accuracy against 58 co-located radiosonde profiles. Results show that the combination of the MWR and RASS observations in the physical-iterative approach allows for a more accurate characterization of low-level temperature inversions, and that these retrieved temperature profiles match the radiosonde observations better than all other approaches, including the neural network, in the atmospheric layer between the surface and 5 km AGL. Specifically, in this layer of the atmosphere, both root mean square errors and standard deviations of the difference between radiosonde and retrievals that combine MWR and RASS are improved by ~0.5 °C compared to the other methods. Pearson correlation coefficients are also improved.

scholarly journals Error simulations of uncorrected NDVI and DCVI during remote sensing measurements from UAS

2014 ◽  
Vol 18 (2) ◽  
pp. 35-45 ◽  
Author(s):  
Michał T. Chiliński ◽  
Marek Ostrowski
Keyword(s):  
Remote Sensing ◽  
Radiative Transfer ◽  
Vegetation Index ◽  
Vegetation Mapping ◽  
Radiative Transfer Model ◽  
Unmanned Aerial Systems ◽  
Transfer Model ◽  
Digital Cameras ◽  
Aerial Systems

Abstract Remote sensing from unmanned aerial systems (UAS) has been gaining popularity in the last few years. In the field of vegetation mapping, digital cameras converted to calculate vegetation index (DCVI) are one of the most popular sensors. This paper presents simulations using a radiative transfer model (libRadtran) of DCVI and NDVI results in an environment of possible UAS flight scenarios. The analysis of the results is focused on the comparison of atmosphere influence on both indices. The results revealed uncertainties in uncorrected DCVI measurements up to 25% at the altitude of 5 km, 5% at 1 km and around 1% at 0.15 km, which suggests that DCVI can be widely used on small UAS operating below 0.2 km.

scholarly journals Effect of OH emission on the temperature and wind measurements derived from limb-viewing observations of the 1.27 µm O<sub>2</sub> dayglow

2020 ◽  
Vol 13 (4) ◽  
pp. 1817-1824
Author(s):  
Kuijun Wu ◽  
Weiwei He ◽  
Yutao Feng ◽  
Yuanhui Xiong ◽  
Faquan Li
Keyword(s):  
Remote Sensing ◽  
Radiative Transfer ◽  
Spectral Region ◽  
Emission Lines ◽  
Radiative Transfer Model ◽  
Transfer Model ◽  
Altitude Range ◽  
Band Emission ◽  
Near Space ◽  
Wind Measurements

Abstract. The O2(a1Δg) emission near 1.27 µm is well-suited for remote sensing of global wind and temperature in near-space by limb-viewing observations to its bright signal and extended altitude coverage. However, vibrational–rotational emission lines of the OH dayglow produced by the hydrogen–ozone reaction (H+O3→OH•+O2) overlap the infrared atmospheric band emission (a1Δg→X3Σg) of O2. The main goal of this paper is to discuss the effect of OH emission on the wind and temperature measurements derived from the 1.27 µm O2 dayglow limb-viewing observations. The O2 dayglow and OH dayglow spectrum over the spectral region and altitude range of interest is calculated by using the line-by-line radiative transfer model and the most recent photochemical model. The method of four-point sampling of the interferogram and sample results of measurement simulations are provided for both O2 dayglow and OH dayglow. It is apparent from the simulations that the presence of OH dayglow as an interfering species decreases the wind and temperature accuracy at all altitudes, but this effect can be reduced considerably by improving OH dayglow knowledge.

scholarly journals An Approach to Minimize Atmospheric Correction Error and Improve Physics-Based Satellite-Derived Bathymetry in a Coastal Environment

2020 ◽  
Vol 12 (17) ◽  
pp. 2752
Author(s):  
Christopher O. Ilori ◽  
Anders Knudby
Keyword(s):  
Remote Sensing ◽  
Environmental Factors ◽  
Complete Solution ◽  
Atmospheric Correction ◽  
Florida Keys ◽  
Radiative Transfer Model ◽  
Inversion Method ◽  
Transfer Model ◽  
Landsat 8 ◽  
Depth Range

Physics-based radiative transfer model (RTM) inversion methods have been developed and implemented for satellite-derived bathymetry (SDB); however, precise atmospheric correction (AC) is required for robust bathymetry retrieval. In a previous study, we revealed that biases from AC may be related to imaging and environmental factors that are not considered sufficiently in all AC algorithms. Thus, the main aim of this study is to demonstrate how AC biases related to environmental factors can be minimized to improve SDB results. To achieve this, we first tested a physics-based inversion method to estimate bathymetry for a nearshore area in the Florida Keys, USA. Using a freely available water-based AC algorithm (ACOLITE), we used Landsat 8 (L8) images to derive per-pixel remote sensing reflectances, from which bathymetry was subsequently estimated. Then, we quantified known biases in the AC using a linear regression that estimated bias as a function of imaging and environmental factors and applied a correction to produce a new set of remote sensing reflectances. This correction improved bathymetry estimates for eight of the nine scenes we tested, with the resulting changes in bathymetry RMSE ranging from +0.09 m (worse) to −0.48 m (better) for a 1 to 25 m depth range, and from +0.07 m (worse) to −0.46 m (better) for an approximately 1 to 16 m depth range. In addition, we showed that an ensemble approach based on multiple images, with acquisitions ranging from optimal to sub-optimal conditions, can be used to estimate bathymetry with a result that is similar to what can be obtained from the best individual scene. This approach can reduce time spent on the pre-screening and filtering of scenes. The correction method implemented in this study is not a complete solution to the challenge of AC for satellite-derived bathymetry, but it can eliminate the effects of biases inherent to individual AC algorithms and thus improve bathymetry retrieval. It may also be beneficial for use with other AC algorithms and for the estimation of seafloor habitat and water quality products, although further validation in different nearshore waters is required.

scholarly journals Characterization of aerosol hygroscopicity using Raman lidar measurements at the EARLINET station of Payerne

2019 ◽  
Vol 19 (18) ◽  
pp. 11651-11668 ◽  
Author(s):  
Francisco Navas-Guzmán ◽  
Giovanni Martucci ◽  
Martine Collaud Coen ◽  
María José Granados-Muñoz ◽  
Maxime Hervo ◽  
...  
Keyword(s):  
Lower Troposphere ◽  
Radiative Transfer Model ◽  
Transfer Model ◽  
Raman Lidar ◽  
Hygroscopic Growth ◽  
Backscatter Coefficient ◽  
Radiative Balance ◽  
First Case ◽  
The Impact ◽  
Aerosol Hygroscopicity

Abstract. This study focuses on the analysis of aerosol hygroscopicity using remote sensing techniques. Continuous observations of aerosol backscatter coefficient (βaer), temperature (T) and water vapor mixing ratio (r) have been performed by means of a Raman lidar system at the aerological station of MeteoSwiss at Payerne (Switzerland) since 2008. These measurements allow us to monitor in a continuous way any change in aerosol properties as a function of the relative humidity (RH). These changes can be observed either in time at a constant altitude or in altitude at a constant time. The accuracy and precision of RH measurements from the lidar have been evaluated using the radiosonde (RS) technique as a reference. A total of 172 RS profiles were used in this intercomparison, which revealed a bias smaller than 4 % RH and a standard deviation smaller than 10 % RH between both techniques in the whole (in lower) troposphere at nighttime (at daytime), indicating the good performance of the lidar for characterizing RH. A methodology to identify situations favorable to studying aerosol hygroscopicity has been established, and the aerosol hygroscopicity has been characterized by means of the backscatter enhancement factor (fβ). Two case studies, corresponding to different types of aerosol, are used to illustrate the potential of this methodology. The first case corresponds to a mixture of rural aerosol and smoke particles (smoke mixture), which showed a higher hygroscopicity (fβ355=2.8 and fβ1064=1.8 in the RH range 73 %–97 %) than the second case, in which mineral dust was present (fβ355=1.2 and fβ1064=1.1 in the RH range 68 %–84 %). The higher sensitivity of the shortest wavelength to hygroscopic growth was qualitatively reproduced using Mie simulations. In addition, a good agreement was found between the hygroscopic analysis done in the vertical and in time for Case I, where the latter also allowed us to observe the hydration and dehydration of the smoke mixture. Finally, the impact of aerosol hygroscopicity on the Earth's radiative balance has been evaluated using the GAME (Global Atmospheric Model) radiative transfer model. The model showed an impact with an increase in absolute value of 2.4 W m−2 at the surface with respect to the dry conditions for the hygroscopic layer of Case I (smoke mixture).

scholarly journals Airborne Hyperspectral Imaging for Submerged Archaeological Mapping in Shallow Water Environments

2019 ◽  
Vol 11 (19) ◽  
pp. 2237 ◽  
Author(s):  
Alexandre Guyot ◽  
Marc Lennon ◽  
Nicolas Thomas ◽  
Simon Gueguen ◽  
Tristan Petit ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Shallow Water ◽  
Water Column ◽  
Sea Level ◽  
Hyperspectral Imagery ◽  
Airborne Lidar ◽  
Radiative Transfer Model ◽  
Invasive Technique ◽  
Transfer Model ◽  
Water Bottom

Nearshore areas around the world contain a wide variety of archeological structures, including prehistoric remains submerged by sea level rise during the Holocene glacial retreat. While natural processes, such as erosion, rising sea level, and exceptional climatic events have always threatened the integrity of this submerged cultural heritage, the importance of protecting them is becoming increasingly critical with the expanding effects of global climate change and human activities. Aerial archaeology, as a non-invasive technique, contributes greatly to documentation of archaeological remains. In an underwater context, the difficulty of crossing the water column to reach the bottom and its potential archaeological information usually requires active remote-sensing technologies such as airborne LiDAR bathymetry or ship-borne acoustic soundings. More recently, airborne hyperspectral passive sensors have shown potential for accessing water-bottom information in shallow water environments. While hyperspectral imagery has been assessed in terrestrial continental archaeological contexts, this study brings new perspectives for documenting submerged archaeological structures using airborne hyperspectral remote sensing. Airborne hyperspectral data were recorded in the Visible Near Infra-Red (VNIR) spectral range (400–1000 nm) over the submerged megalithic site of Er Lannic (Morbihan, France). The method used to process these data included (i) visualization of submerged anomalous features using a minimum noise fraction transform, (ii) automatic detection of these features using Isolation Forest and the Reed–Xiaoli detector and (iii) morphological and spectral analysis of archaeological structures from water-depth and water-bottom reflectance derived from the inversion of a radiative transfer model of the water column. The results, compared to archaeological reference data collected from in-situ archaeological surveys, showed for the first time the potential of airborne hyperspectral imagery for archaeological mapping in complex shallow water environments.

scholarly journals Analysis of Sentinel-2 and RapidEye for Retrieval of Leaf Area Index in a Saltmarsh Using a Radiative Transfer Model

2019 ◽  
Vol 11 (6) ◽  
pp. 671 ◽  
Author(s):  
Roshanak Darvishzadeh ◽  
Tiejun Wang ◽  
Andrew Skidmore ◽  
Anton Vrieling ◽  
Brian O’Connor ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Radiative Transfer ◽  
Leaf Area Index ◽  
Plant Traits ◽  
Remote Sensing Data ◽  
Radiative Transfer Model ◽  
Transfer Model ◽  
Area Index ◽  
Field Campaigns ◽  
Sentinel 2

The Sentinel satellite fleet of the Copernicus Programme offers new potential to map and monitor plant traits at fine spatial and temporal resolutions. Among these traits, leaf area index (LAI) is a crucial indicator of vegetation growth and an essential variable in biodiversity studies. Numerous studies have shown that the radiative transfer approach has been a successful method to retrieve LAI from remote-sensing data. However, the suitability and adaptability of this approach largely depend on the type of remote-sensing data, vegetation cover and the ecosystem studied. Saltmarshes are important wetland ecosystems threatened by sea level rise among other human- and animal-induced changes. Therefore, monitoring their vegetation status is crucial for their conservation, yet few LAI assessments exist for these ecosystems. In this study, the retrieval of LAI in a saltmarsh ecosystem is examined using Sentinel-2 and RapidEye data through inversion of the PROSAIL radiative transfer model. Field measurements of LAI and some other plant traits were obtained during two succeeding field campaigns in July 2015 and 2016 on the saltmarsh of Schiermonnikoog, a barrier island of the Netherlands. RapidEye (2015) and Sentinel-2 (2016) data were acquired concurrent to the time of the field campaigns. The broadly employed PROSAIL model was inverted using two look-up tables (LUTs) generated in the spectral band’s settings of the two sensors and in which each contained 500,000 records. Different solutions from the LUTs, as well as, different Sentinel-2 spectral subsets were considered to examine the LAI retrieval. Our results showed that generally the LAI retrieved from Sentinel-2 had higher accuracy compared to RapidEye-retrieved LAI. Utilising the mean of the first 10 best solutions from the LUTs resulted in higher R2 (0.51 and 0.59) and lower normalised root means square error (NRMSE) (0.24 and 0.16) for both RapidEye and Sentinel-2 data respectively. Among different Sentinel-2 spectral subsets, the one comprised of the four near-infrared (NIR) and shortwave infrared (SWIR) spectral bands resulted in higher estimation accuracy (R2 = 0.44, NRMSE = 0.21) in comparison to using other studied spectral subsets. The results demonstrated the feasibility of broadband multispectral sensors, particularly Sentinel-2 for retrieval of LAI in the saltmarsh ecosystem via inversion of PROSAIL. Our results highlight the importance of proper parameterisation of radiative transfer models and capacity of Sentinel-2 spectral range and resolution, with impending high-quality global observation aptitude, for retrieval of plant traits at a global scale.

scholarly journals Radiometric Microwave Indices for Remote Sensing of Land Surfaces

2018 ◽  
Vol 10 (12) ◽  
pp. 1859 ◽  
Author(s):  
Simonetta Paloscia ◽  
Paolo Pampaloni ◽  
Emanuele Santi
Keyword(s):  
Remote Sensing ◽  
Radiative Transfer ◽  
Snow Water Equivalent ◽  
Hydrological Cycle ◽  
Microwave Remote Sensing ◽  
Dense Medium ◽  
Single Frequency ◽  
Radiative Transfer Model ◽  
Transfer Model ◽  
Land Surfaces

This work presents an overview of the potential of microwave indices obtained from multi-frequency/polarization radiometry in detecting the characteristics of land surfaces, in particular soil covered by vegetation or snow and agricultural bare soils. Experimental results obtained with ground-based radiometers on different types of natural surfaces by the Microwave Remote Sensing Group of IFAC-CNR starting from ‘80s, are summarized and interpreted by means of theoretical models. It has been pointed out that, with respect to single frequency/polarization observations, microwave indices revealed a higher sensitivity to some significant parameters, which characterize the hydrological cycle, namely: soil moisture, vegetation biomass and snow depth or snow water equivalent. Electromagnetic models have then been used for simulating brightness temperature and microwave indices from land surfaces. As per vegetation covered soils, the well-known tau-omega (τ-ω) model based on the radiative transfer theory has been used, whereas terrestrial snow cover has been simulated using a multi-layer dense-medium radiative transfer model (DMRT). On the basis of these results, operational inversion algorithms for the retrieval of those hydrological quantities have been successfully implemented using multi-channel data from the microwave radiometric sensors operating from satellite.

Sign in / Sign up

Export Citation Format

Share Document