Estimation of all-weather land surface temperature through correcting cloud-shadowing bias simulated by hourly cloud information
Explore the source record for details and available documents.
Peilin Song↗
SEARCH · Search NASA
Results for “Cloud shadow”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Cloud shadow effects on remote sensing
On relatively clear days, the shadow cast by an isolated, optically dense cloud produces an effect which is similar in extent and magnitude to the adjacency effects associated with high-contrast ground albedo boundaries. A Monte Carlo radiative transfer model has been used to examine the intensity and its components, comparing the cloud shadow case with that for a Lambertian albedo boundary. Although the behavior of the total intensity is nearly the same for both cases, the behaviors of the intensity components differ significantly.
Pearce, W. A.↗
Satellite-Sensor Calibration Verification Using the Cloud-Shadow Method
An atmospheric-correction method which uses cloud-shaded pixels together with pixels in a neighboring region of similar optical properties is described. This cloud-shadow method uses the difference between the total radiance values observed at the sensor for these two regions, thus removing the nearly identical atmospheric radiance contributions to the two signals (e.g. path radiance and Fresnel-reflected skylight). What remains is largely due to solar photons backscattered from beneath the sea to dominate the residual signal. Normalization by the direct solar irradiance reaching the sea surface and correction for some second-order effects provides the remote-sensing reflectance of the ocean at the location of the neighbor region, providing a known 'ground target' spectrum for use in testing the calibration of the sensor. A similar approach may be useful for land targets if horizontal homogeneity of scene reflectance exists about the shadow. Monte Carlo calculations have been used to correct for adjacency effects and to estimate the differences in the skylight reaching the shadowed and neighbor pixels.
Reinersman, P.↗
AVIRIS calibration using the cloud-shadow method
More than 90 percent of the signal at an ocean-viewing, satellite sensor is due to the atmosphere, so a 5 percent sensor-calibration error viewing a target that contributes but 10 percent of the signal received at the sensor may result in a target-reflectance error of more than 50 percent. Since prelaunch calibration accuracies of 5 percent are typical of space-sensor requirements, recalibration of the sensor using ground-base methods is required for low-signal target. Known target reflectance or water-leaving radiance spectra and atmospheric correction parameters are required. In this article we describe an atmospheric-correction method that uses cloud shadowed pixels in combination with pixels in a neighborhood region of similar optical properties to remove atmospheric effects from ocean scenes. These neighboring pixels can then be used as known reflectance targets for validation of the sensor calibration and atmospheric correction. The method uses the difference between water-leaving radiance values for these two regions. This allows nearly identical optical contributions to the two signals (e.g., path radiance and Fresnel-reflected skylight) to be removed, leaving mostly solar photons backscattered from beneath the sea to dominate the residual signal. Normalization by incident solar irradiance reaching the sea surface provides the remote-sensing reflectance of the ocean at the location of the neighbor region.
Carder, K. L.↗
Solar reflection from interacting and shadowing cloud elements
An attempt is made to relate the reflection from a broken cloud field to that from a plane parallel cloud of the same optical properties by using Monte Carlo simulations for a very simple model of an array of regularly spaced, infinitely long bar clouds. Reflection from the array is represented by an effective cloud fraction, and it is shown that for many cases this may be obtained by treating the cloud elements to be noninteracting azimuthally symmetric reflectors. The effective cloud fraction is a function of solar zenith angle, the aspect ratio of the individual cloud elements, and the normal or zenith cloud fraction of the array. Model computations for reflection from an array of randomly distributed identical cylinders are also presented as an example of a possible application of this technique.
HARSHVARDHAN↗
Cloud shadows drive vertical migrations of deep-dwelling marine life
Explore the source record for details and available documents.
Melissa M. Omand↗
Reduction of EUVE Guest Observer Data
We report on the observation of the HI filament LVC 88+36-2 with the Deep Survey/Spectrometers of EUVE. Detecting the shadow of this cloud in absorption might have proven the existence of a diffuse EUV background behind the cloud, and constrain the emission geometries of the hot plasma in the Local Interstellar Medium (LISM). By detecting diffuse emission in the EUVE spectrometers, it would also be possible to constrain the temperature and pressure of the emitting regions as well as determine whether or not the hot phase of the LISM is in equilibrium with cosmic elemental abundances. Unfortunately, we did not detect a cloud shadow with this set of observations. This is due to a combination of circumstances, but mostly due to the fact that EUV diffuse emission is weak compared to the background levels of EUVE. If a shadow exists in the EUVE, then it is less than 1% of the EUVE detector background of 1e-3 cts arcmin(exp -2) in the Deep Survey detector.
Vallerga, John↗
Reduction of EUVE Guest Observer Data
This report documents the observations of the HI filament LVC 88+36-2 with the Deep Survey/Spectrometers of EUVE. Detecting the shadow of this cloud in absorption might have proven the existence of a diffuse EUV background behind the cloud, and constrain the emission geometries of the hot plasma in the Local Interstellar Medium (LISM). By detecting diffuse emission in the EUVE spectrometers, it would also be possible to constrain the temperature and pressure of the emitting regions as well as determine whether or not the hot phase of the LISM is in equilibrium with cosmic elemental abundances. Unfortunately, we did not detect a cloud shadow with this set of observations. This is due to a combination of circumstances, but mostly due to the fact that EUV diffuse emission is weak compared to the background levels of EUVE. If a shadow exists in the EUVE, then it is less than 1% of the EUVE detector background of 1e-3 cts arcmin(sup -2) in the Deep Survey detector.
Vallerga, John↗
Comparison of Cloud Detection Algorithms for Sentinel-2 Imagery
Accurate, automated cloud and cloud shadow detection is a key component of the processing needed to prepare optical satellite imagery for scientific analysis. Many existing cloud detection algorithms rely on temperature information to identify clouds, making detection difficult for imagers that lack a thermal band, like Sentinel-2. To get maximum benefit from Sentinel-2 products it is critical to understand which algorithms best identify clouds and their shadows in images. We examined the relative performance of five different cloud-masking algorithms (Sen2Cor, MAJA, LaSRC, Fmask and Tmask) in 6 Sentinel-2 scenes (28 total images) distributed across the Eastern Hemisphere. Expanding on these comparisons, we tested ensemble approaches to improve results. We tested three ensemble approaches to cloud and shadow classification based on the outputs of the five initial algorithms using the cloud masks in: (1) a majority prediction model; (2) a random forests model; and (3) a conditional logic model. Accuracy assessments show a trade-off between omission and commission errors in cloud detection for individual algorithms across all sites, and some algorithms are better at detecting either clouds or cloud shadows. No single algorithm outperforms the others for both clouds and shadows. Aggregating the results from multiple algorithms produces fewer undetected clouds and higher overall accuracy than any single algorithm, with as high as 2.7% improvement over the top-performing algorithm, suggesting an ensemble approach may be the most useful for processing of Sentinel-2 data.
Sentinel-2↗
Improving IPATS Channel-to-Channel Registration Assessment
The Image Navigation and Registration (INR) Performance Assessment Tool Set (IPATS) is a primary tool for assessing INR performance of GOES-R series ABI images. IPATS assesses five INR metrics: navigation, channel-to-channel registration (CCR), frame-to-frame registration, within-frame registration, and swath-to-swath registration. It was discovered that CCR assessment results between Visible-Near-Infrared (VNIR) channels and Infrared (IR) channels exhibits an annual oscillation in the north-south (NS) direction and a diurnal oscillation in the east-west (EW) direction, with an amplitude of approximately 5 μrad and 2.5 μrad, respectively. However, differences of navigation assessment results between VNIR and IR channels do not exhibit the annual or diurnal oscillations observed in CCR results. This indicates that the observed oscillations are due to measurement errors. The characteristics of the oscillations imply that cloud shadows are a possible cause of these measurement errors. In this study, several methods are explored to minimize the impact of cloud shadows on VNIR to IR CCR assessments: a) assessment at landmark locations only; b) using navigation assessment results to filter CCR assessments; c) using the ABI clear-sky-ratio product as a cloud mask; and d) smaller CCR assessment windows. In this paper, each method and a combination of several methods are evaluated based on assessment accuracy and the number of successful assessments. The selected approach is then used to reprocess GOES-16 ABI CCR data to show reductions in the annual and diurnal measurement error oscillations.
INR↗
H31G-1596: DeepSAT's CloudCNN: A Deep Neural Network for Rapid Cloud Detection from Geostationary Satellites
Cloud and cloud shadow detection has important applications in weather and climate studies. It is even more crucial when we introduce geostationary satellites into the field of terrestrial remote sensing. With the challenges associated with data acquired in very high frequency (10-15 mins per scan), the ability to derive an accurate cloud shadow mask from geostationary satellite data is critical. The key to the success for most of the existing algorithms depends on spatially and temporally varying thresholds,which better capture local atmospheric and surface effects.However, the selection of proper threshold is difficult and may lead to erroneous results. In this work, we propose a deep neural network based approach called CloudCNN to classify cloudshadow from Himawari-8 AHI and GOES-16 ABI multispectral data. DeepSAT's CloudCNN consists of an encoderdecoder based architecture for binary-class pixel wise segmentation. We train CloudCNN on multi-GPU Nvidia Devbox cluster, and deploy the prediction pipeline on NASA Earth Exchange (NEX) Pleiades supercomputer. We achieved an overall accuracy of 93.29% on test samples. Since, the predictions take only a few seconds to segment a full multispectral GOES-16 or Himawari-8 Full Disk image, the developed framework can be used for real-time cloud detection, cyclone detection, or extreme weather event predictions.
GOES-1↗
The use of Landsat-3 thermal data to help differentiate land covers
Landsat-3 Multispectral Scanner Subsystem (MSS) digital data of the Baltimore, Maryland area gathered on May 24, 1978, are examined to show the usefulness of thermal data in providing better discrimination between agricultural and residential areas, certain types of urban/industrial areas and water, cloud shadows and water, and bare-extractive areas and bright urban cover types. High altitude aircraft imagery taken on May 3, 1978, provides ground truth and training site verification. Two classifications are made for each training site: the initial one using bands 4, 5, and 7 and a second in which the thermal data are included with the visible and near infrared data. This permits a direct comparison of areas spectrally similar with and without the inclusion of the thermal data. Commission errors determined from selected subsets of the data show reductions of 95% for the urban/industrial versus water themes, 84% for the residential versus agriculture themes, 64.0% for the bare-extractive versus bright urban themes, and 24% for the cloud shadow versus water themes when the thermal data are included in the signature.
Ormsby, J. P.↗
Automatic photointerpretation for land use management in Minnesota
The author has identified the following significant results. Primary conclusions from the lake acreage study are: (1) The ERTS-1 band 7 density range of 0-5 reliably indicates open water down to 2 acre size. (2) The density range 6-9 identifies swamps. (3) The depth of the water could not be determined. (4) Cloud shadows can be misread as lakes unless the clouds are detected. (5) ERTS-1 data would provide the information for classifying lakes and for monitoring fluctuations in lake area.
Swanlund, G. D.↗
Photogrammetry and photo interpretation applied to analyses of cloud cover, cloud type, and cloud motion
A determination was made of the areal extent of terrain obscured by clouds and cloud shadows on a portion of an Apollo 9 photograph at the instant of exposure. This photogrammetrically determined area was then compared to the cloud coverage reported by surface weather observers at approximately the same time and location, as a check on result quality. Stereograms prepared from Apollo 9 vertical photographs, illustrating various percentages of cloud coverage, are presented to help provide a quantitative appreciation of the degradation of terrain photography by clouds and their attendant shadows. A scheme, developed for the U.S. Navy, utilizing pattern recognition techniques for determining cloud motion from sequences of satellite photographs, is summarized. Clouds, turbulence, haze, and solar altitude, four elements of our natural environment which affect aerial photographic missions, are each discussed in terms of their effects on imagery obtained by aerial photography. Data of a type useful to aerial photographic mission planners, expressing photographic ground coverage in terms of flying height above terrain and camera focal length, for a standard aerial photograph format, are provided. Two oblique orbital photographs taken during the Apollo 9 flight are shown, and photo-interpretations, discussing the cloud types imaged and certain visible geographical features, are provided.
Larsen, P. A.↗
Automated, per pixel Cloud Detection from High-Resolution VNIR Data
CASA is a fully automated software program for the per-pixel detection of clouds and cloud shadows from medium- (e.g., Landsat, SPOT, AWiFS) and high- (e.g., IKONOS, QuickBird, OrbView) resolution imagery without the use of thermal data. CASA is an object-based feature extraction program which utilizes a complex combination of spectral, spatial, and contextual information available in the imagery and the hierarchical self-learning logic for accurate detection of clouds and their shadows.
Varlyguin, Dmitry L.↗
The effect of spatial resolution upon cloud optical property retrievals. I - Optical thickness
High spectral and spatial resolution Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) imagery is used to study the effects of spatial resolution upon fair weather cumulus cloud optical thickness retrievals. As a preprocessing step, a variation of the Gao and Goetz three-band ratio technique is used to discriminate clouds from the background. The combination of the elimination of cloud shadow pixels and using the first derivative of the histogram allows for accurate cloud edge discrimination. The data are progressively degraded from 20 m to 960 m spatial resolution. The results show that retrieved cloud area increases with decreasing spatial resolution. The results also show that there is a monotonic decrease in retrieved cloud optical thickness with decreasing spatial resolution. It is also demonstrated that the use of a single, monospectral reflectance threshold is inadequate for identifying cloud pixels in fair weather cumulus scenes and presumably in any inhomogeneous cloud field. Cloud edges have a distribution of reflectance thresholds. The incorrect identification of cloud edges significantly impacts the accurate retrieval of cloud optical thickness values.
Feind, Rand E.↗
Remote Sensing of Liquid Water and Ice Cloud Optical Thickness and Effective Radius in the Arctic: Application of Airborne Multispectral MAS Data
A multispectral scanning spectrometer was used to obtain measurements of the reflection function and brightness temperature of clouds, sea ice, snow, and tundra surfaces at 50 discrete wavelengths between 0.47 and 14.0 microns. These observations were obtained from the NASA ER-2 aircraft as part of the FIRE Arctic Clouds Experiment, conducted over a 1600 x 500 km region of the north slope of Alaska and surrounding Beaufort and Chukchi Seas between 18 May and 6 June 1998. Multispectral images of the reflection function and brightness temperature in 11 distinct bands of the MODIS Airborne Simulator (MAS) were used to derive a confidence in clear sky (or alternatively the probability of cloud), shadow, and heavy aerosol over five different ecosystems. Based on the results of individual tests run as part of the cloud mask, an algorithm was developed to estimate the phase of the clouds (water, ice, or undetermined phase). Finally, the cloud optical thickness and effective radius were derived for both water and ice clouds that were detected during one flight line on 4 June. This analysis shows that the cloud mask developed for operational use on MODIS, and tested using MAS data in Alaska, is quite capable of distinguishing clouds from bright sea ice surfaces during daytime conditions in the high Arctic. Results of individual tests, however, make it difficult to distinguish ice clouds over snow and sea ice surfaces, so additional tests were added to enhance the confidence in the thermodynamic phase of clouds over the Beaufort Sea. The cloud optical thickness and effective radius retrievals used 3 distinct bands of the MAS, with the newly developed 1.62 and 2.13 micron bands being used quite successfully over snow and sea ice surfaces. These results are contrasted with a MODIS-based algorithm that relies on spectral reflectance at 0.87 and 2.13 micron.
King, Michael D.↗
Analyses of the cloud contents of multispectral imagery from LANDSAT 2: Mesoscale assessments of cloud and rainfall over the British Isles
The author has identified the following significant results. It was demonstrated that satellites with sufficiently high resolution capability in the visible region of the electromagnetic spectrum could be used to check the accuracy of estimates of total cloud amount assessed subjectively from the ground, and to reveal areas of performance in which corrections should be made. It was also demonstrated that, in middle latitude in summer, cloud shadow may obscure at least half as much again of the land surface covered by an individual LANDSAT frame as the cloud itself. That proportion would increase with latitude and/or time of year towards the winter solstice. Analyses of sample multispectral images for six different categories of clouds in summer revealed marked differences between the reflectance characteristics of cloud fields in the visible/near infrared region of the spectrum.
Barrett, E. C.↗