Abstract:A method is described to characterize the scale dependence of cloud chord length using cloud-type classification reported with the 94-GHz CloudSat radar. The cloud length along the CloudSat track is quantified using horizontal and vertical structures of cloud classification separately for each cloud type and for all clouds independent of cloud type. While the individual cloud types do not follow a clear power-law behavior as a function of horizontal or vertical scale, a robust power-law scaling of cloud chord … Show more
“…values reported in Table 1. Note that these values are similar to but not exactly the same as those calculated in Guillaume et al (2018), for which cloud length was derived from a 2-D curtain of cloud features. The main characteristic shared by all cloud types in Fig.…”
Section: Generalizing To All Scalessupporting
confidence: 69%
“…The cloud chord length was obtained at the cloud top (see Fig. 5) unlike that obtained in Guillaume et al (2018). frequency decrease for larger scales.…”
Section: Generalizing To All Scalesmentioning
confidence: 79%
“…The reasons for the maximum number of observed cloud scenes (210) at a particular horizontal scale (105 km) are not immediately clear. The scale preference depends on the physical characteristics of cloud regimes and the degree to which cloud types are mixed together by region and furthermore depend on cloud length distributions (Guillaume et al, 2018). A simple model is described below that is able to approximate the results of Fig.…”
Section: Cloud Scenes At 1 To 1000 Km Scalesmentioning
confidence: 99%
“…The preponderance of undetermined phase for Ac may indicate frequent supercooled liquid cloud tops (Zhang et al, 2010). Ham et al (2013) showed that Ac are typically 2-3 km lower in altitude than As, and this probably explains some of the difference in liquid and ice phase, as lower clouds are usually warmer. The Ns cloud scene histogram is dominated by ice detection with occasional liquid and undetermined cloud tops.…”
Abstract. A method is described to classify cloud mixtures of cloud
top types, termed cloud scenes, using cloud type classification derived from the CloudSat
radar (2B-CLDCLASS). The scale dependence of the cloud scenes is quantified.
For spatial scales at 45 km (15 km), only 18 (10) out of 256 possible cloud
scenes account for 90 % of all observations and contain one, two,
or three cloud types. The number of possible cloud scenes is shown to depend
on spatial scale with a maximum number of 210 out of 256 possible scenes at
a scale of 105 km and fewer cloud scenes at smaller and larger scales. The
cloud scenes are used to assess the characteristics of spatially collocated
Atmospheric Infrared Sounder (AIRS) thermodynamic-phase and ice cloud
property retrievals within scenes of varying cloud type complexity. The
likelihood of ice and liquid-phase detection strongly depends on the
CloudSat-identified cloud scene type collocated with the AIRS footprint.
Cloud scenes primarily consisting of cirrus, nimbostratus, altostratus, and
deep convection are dominated by ice-phase detection, while stratocumulus,
cumulus, and altocumulus are dominated by liquid- and undetermined-phase
detection. Ice cloud particle size and optical thickness are largest for
cloud scenes containing deep convection and cumulus and are smallest for
cirrus. Cloud scenes with multiple cloud types have small reductions in
information content and slightly higher residuals of observed and modeled
radiance compared to cloud scenes with single cloud types. These results
will help advance the development of temperature, specific humidity, and
cloud property retrievals from hyperspectral infrared sounders that include
cloud microphysics in forward radiative transfer models.
“…values reported in Table 1. Note that these values are similar to but not exactly the same as those calculated in Guillaume et al (2018), for which cloud length was derived from a 2-D curtain of cloud features. The main characteristic shared by all cloud types in Fig.…”
Section: Generalizing To All Scalessupporting
confidence: 69%
“…The cloud chord length was obtained at the cloud top (see Fig. 5) unlike that obtained in Guillaume et al (2018). frequency decrease for larger scales.…”
Section: Generalizing To All Scalesmentioning
confidence: 79%
“…The reasons for the maximum number of observed cloud scenes (210) at a particular horizontal scale (105 km) are not immediately clear. The scale preference depends on the physical characteristics of cloud regimes and the degree to which cloud types are mixed together by region and furthermore depend on cloud length distributions (Guillaume et al, 2018). A simple model is described below that is able to approximate the results of Fig.…”
Section: Cloud Scenes At 1 To 1000 Km Scalesmentioning
confidence: 99%
“…The preponderance of undetermined phase for Ac may indicate frequent supercooled liquid cloud tops (Zhang et al, 2010). Ham et al (2013) showed that Ac are typically 2-3 km lower in altitude than As, and this probably explains some of the difference in liquid and ice phase, as lower clouds are usually warmer. The Ns cloud scene histogram is dominated by ice detection with occasional liquid and undetermined cloud tops.…”
Abstract. A method is described to classify cloud mixtures of cloud
top types, termed cloud scenes, using cloud type classification derived from the CloudSat
radar (2B-CLDCLASS). The scale dependence of the cloud scenes is quantified.
For spatial scales at 45 km (15 km), only 18 (10) out of 256 possible cloud
scenes account for 90 % of all observations and contain one, two,
or three cloud types. The number of possible cloud scenes is shown to depend
on spatial scale with a maximum number of 210 out of 256 possible scenes at
a scale of 105 km and fewer cloud scenes at smaller and larger scales. The
cloud scenes are used to assess the characteristics of spatially collocated
Atmospheric Infrared Sounder (AIRS) thermodynamic-phase and ice cloud
property retrievals within scenes of varying cloud type complexity. The
likelihood of ice and liquid-phase detection strongly depends on the
CloudSat-identified cloud scene type collocated with the AIRS footprint.
Cloud scenes primarily consisting of cirrus, nimbostratus, altostratus, and
deep convection are dominated by ice-phase detection, while stratocumulus,
cumulus, and altocumulus are dominated by liquid- and undetermined-phase
detection. Ice cloud particle size and optical thickness are largest for
cloud scenes containing deep convection and cumulus and are smallest for
cirrus. Cloud scenes with multiple cloud types have small reductions in
information content and slightly higher residuals of observed and modeled
radiance compared to cloud scenes with single cloud types. These results
will help advance the development of temperature, specific humidity, and
cloud property retrievals from hyperspectral infrared sounders that include
cloud microphysics in forward radiative transfer models.
“…Guillaume et al (2018) have shown that the distribution of horizontal cloud chord length evaluated from CloudSat data was heavily skewed toward short scales, so that clouds at the CloudSat horizontal resolution of 1.1 km are vastly more frequent than clouds at scales of about 2,000 km, which are very rare. From these data, we extracted nonoverlapping rectangular patches of radar reflectivity, 64 × 64 radar bins in size.…”
We demonstrate the feasibility of solving atmospheric remote sensing problems with machine learning using conditional generative adversarial networks (CGANs), implemented using convolutional neural networks. We apply the CGAN to generating two‐dimensional cloud vertical structures that would be observed by the CloudSat satellite‐based radar, using only the collocated Moderate‐Resolution Imaging Spectrometer measurements as input. The CGAN is usually able to generate reasonable guesses of the cloud structure and can infer complex structures such as multilayer clouds from the Moderate‐Resolution Imaging Spectrometer data. This network, which is formulated probabilistically, also estimates the uncertainty of its own predictions. We examine the statistics of the generated data and analyze the response of the network to each input parameter. The success of the CGAN in solving this problem suggests that generative adversarial networks are applicable to a wide range of problems in atmospheric science, a field characterized by complex spatial structures and observational uncertainties.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.