Search Results

You are looking at 1 - 10 of 28 items for :

  • Author or Editor: C. M. R. Platt x
  • Refine by Access: All Content x
Clear All Modify Search
C. M. R. Platt

Abstract

The technique of using combined lidar (0.694µm) and radiometric (10–12µm) measurements on cirrus clouds was developed in order to determine their infrared emissivity and to examine the experimental relationships between the infrared and lidar data.

Detailed measurements of cirrus infrared emissivity and lidar backscatter were made on several cirrus systems at Adelaide in November 1970. The mean zenith emissivity of cirrus was found to be 0.245. Emissivity was only weakly correlated with thickness. The cirrus extinction coefficient, which is proportional to the particle density, was totally uncorrelated with mid-cloud temperature.

Various infrared and visible optical quantities were calculated from the experimental data and the relationships between some of these quantities were examined. The lidar backscatter amplitude integrated through the cloud was found to be well correlated with the infrared optical thickness. A simple model of absorption and scattering of radiation in a cloud of large (σ50µm) ice spheres was employed to derive a theoretical expression between the integrated lidar backscatter and the infrared optical thickness. The experimental data agreed well with the above expression. The experimentally observed backscatter-to-extinction ratio at 0.694µm was 0.25±0.06. This is considerably lower than values predicted for large ice spheres, in qualitative agreement with the laboratory measurements on ice crystals reported by Huffman and Dugin et al.

A value of the cirrus optical thicknc% al 0.694μ, was calculated from the measured mean infrared optical thickness using the theoretical values of extinction for large ice spheres. Using Fritz's relations between albedo and optical thickness, it was found that at, say, 38S latitude, the cirrus albedo would vary between 0.2 in midwinter and 0.07 in midsummer.

A visible optical thickness can be obtained directly from the lidar data but its value is lower than the true value due to near-forward, multiple-scattered radiation being detected by the laser receiver. It is hoped to remedy this in future experiments by using a narrower receiver aperture.

Full access
C. M. R. Platt

Abstract

Some computations are made of submillimeter wave extinction in clouds and fogs using recent spectrometric results of the optical properties of water. At wavelengths > 1000μ an approximate formula is adequate in which extinction is proportional to cloud water content. At 2000 and 1000μ the extinction is 6.5 and 15.2 dB km−1 per gm m−3, respectively. Between 200 and 1000μ additional extinction occurs due to large droplets of diameters >20μ. Extinction for a typical fog distribution is computed and is found to be 41.1 dB km−1 per gm m−3 at 337 μ and 92.8 dB km−1 per gm m−3 at 200μ. Comparisons with experimental data at 1200 and 337μ shows qualitative agreement, but insufficient data on the composition of the clouds and fogs investigated precludes accurate comparison.

Full access
C. M. R. Platt

Abstract

The design and performance of a narrow-beam radiometer for atmospheric studies are described. It has a beamwidth of 6 mrad and a minimum detectable radiance of 0.0056 mW cm−2 sr−1 in a 1-HZ output bandwidth. The system incorporates a novel method of comparing flux incident on the aperture against a temperature-stabilized blackbody so that effects due to variations in radiation emitted by or reflected from the chopper blades are eliminated. The radiometer is being applied initially to studies in the 10–12 μm spectral band of the atmospheric “window”. Several applications, including the study of water vapor continuum absorption and the emissivity of high layer clouds, are described briefly.

Full access
C. M. R. Platt

Abstract

Using a physical model, bispectral curves of visible albedo versus infrared brightness temperature are derived for various idealized cloud layers. These layers vary between those which are unbroken with variable optical depth (cirrus) to those which are broken, but the clouds have a uniform high optical depth (strato-cumulus). It is shown that the above two cases can be distinguished by the different shapes of the bispectral curves. The effects of variable solar angle and satellite viewing angle on the shapes of the bispectral curves are investigated. The effects of infrared scattering and deep cloud layers are also investigated and found to be important for high clouds with variable optical depth, such as cirrus.

The predicted curves are compared with two-dimensional bispectral histograms obtained from the VISSR radiometer of the GMS-1 satellite. The theoretical curves compare quite well qualitatively. Two detailed quantitative comparisons for clouds which are assumed to be semitransparent high clouds give good agreement.

The limitatins of the model are discussed.

Full access
C. M. R. Platt

Abstract

Some unusual lidar returns from an altostratus cloud are interpreted in terms of reflections from hexagonal ice plates falling with their long axes aligned in the horizontal. Such an explanation is consistent with the observed high backscatter coefficients and low depolarization ratios and also with the temperature range (−12 to −20°C) of the cloud layers, as well as the known fall characteristics of naturally occurring ice platelets. Backscatter efficiencies are calculated for “perfect” ice platelets when illuminated at or near an axis orthogonal to the crystal long axis. It is shown that very high backscatter coefficients can potentially be measured from a cloud of ice plates, depending on the fraction of crystals which are “perfect,&rdquo,; the degree to which the plates' long axes stay horizontal, and the angle of the lidar to the vertical.

It is further shown that reflection from a single crystal gives an appreciable signal-to-noise ratio at the receiver and that only a few crystals will be correctly aligned in the horizontal in a typical laser pulse volume, and for realistic particle number densities.

Full access
C. M. R. Platt

Abstract

Monte Carlo calculations of multiple-scattered contributions to the total energy received in a lidar beam have been made for a representative cirrus ice-cloud scattering phase function. The phase function is varied arbitrarily near the back direction to give three different scattering patterns. The Monte Carlo method of Plass and Kattawar (1971) has been modified and extended to give a greatly increased efficiency for a small decrease in accuracy. This modification has reduced the computation time by ∼80% allowing routine calculations to be made on many different cloud models. When applied to two water cloud models the method gives results which agree well with those obtained by previous workers, The purpose of. the calculations is to investigate the apparent reduction in the cloud optical depth due to multiple scattering when measured by a lidar. This reduction is described here by a multiple scattering factor η It is found that for cirrus clouds the factor η varies considerably with the depth of cloud penetration, with the cloud optical depth and with the cloud extinction coefficient.

The calculations show for the first time the pattern of contribution from each order of scattering separately. It is shown that for cloud optical depths ≳0.1 a model which considers double scattering only is inadequate.

Full access
C. M. R. Platt

Abstract

Measurements by monostatic lidar of linear depolarization ratios and backscatter coefficients in an altostratus cloud revealed a horizontally layered structure. Three different types of layers were observed. The bottom and central layers had depolarization ratios varying from 0.3 to 0.4, which identified them as layers containing mainly ice. The backscatter coefficients were similar to those found in cirrus ice clouds. A central, transient layer had depolarization ratios characteristic of a high-density water cloud, although the total integrated backscatter of 2.3 ± 1.2 was high for this type of cloud. The top layer had a depolarization ratio of 0.2 at the cloud base, decreasing to 0.04 at the cloud center. Backscatter coefficients ranged up to 30 km−1 and the total integrated backscatter was about 7.6 ± 3.8. This value is considerably higher than the range of values predicted for water or cirrus ice clouds and one possible explanation is that specular reflection was occurring from horizontally aligned ice crystal plates.

The variation of backscatter coefficient within each layer was rather regular, with a maximum at the center of the layers. The cloud was situated in a stable air stream and its evolution appeared to be slow.

Full access
C. M. R. Platt

Abstract

This article describes a method of determining the visible and infrared properties of high ice clouds using a ground-based lidar and infrared (IR) radiometer. A method of calibrating the lidar is described. This is followed by a method for the correction of the cloud backscatter coefficients for pulse attenuation in the clouds, using an experimentally determined backscatter to extinction ratio k. The IR emissivity is then calculated by assuming a value for the ratio between the visible extinction coefficient and the IR absorption coefficient which is invariant with altitude. This ratio is altered until the computed radiance is equal to the measured cloud radiance. Errors in the calculation of the backscatter coefficient, the visible and the IR optical depths and the IR emissivity are assessed for various errors in the backscatter to extinction ratio. It is found that errors in the cloud optical depth become extremely sensitive to errors in k when the cloud visible optical depth becomes large. However, errors in the IR optical depth and IR emissivity are considerably less because they are constrained to agree with IR radiance measurements. For ”typical“ cirrus, having an emissivity of 0.24, and for a standard error in k of 20%, the error in the visible optical depth is 20–30%, whereas the errors in the IR optical depth and emissivity are only 1 and 2%, respectively. The effects of a variable multiple-scattering factor on the above errors appear to be small. However, the variation of this factor is not known well enough yet in high clouds to assess the errors accurately.

Other sources of error, which will be discussed in later papers of this series, include experimental errors in the measured IR radiance and errors in the determination of the calibration factor S.

Full access
C. M. R. Platt and A. C. Dilley

Abstract

The results of a ruby lidar (0.694 μm wavelength) and infrared radiometer (10–12 μm) study on cirrus clouds are reported for a period covering the autumn and winter months at Aspendale (38°S, 144°E). The lidar and radiometer data have been used to study the temperature dependence of the gross structure and optical properties of cirrus clouds. Well-defined correlations are found between the mid-cloud temperature and cloud depth, infrared absorption coefficient, infrared emittance, backscatter to extinction ratio and ratio of the visible extinction coefficient at 0.693 μm to the infrared absorption coefficient at 10–12 μm. For instance, as the mid-cloud temperature varies from −70 to −30°C, the mean cloud depth increases from 1 to 3.5 km and mean infrared absorption coefficient from 0.04 to 0,25 km−1. These two factors together cause a change in emittance from 0.11 to 0.42. The increase in absorption coefficient with temperature can be attributed to the presence of larger ice particles in the deeper clouds. Over the same temperature range the effective backscatter to extinction ratio has a fairly complex behavior with values of 0.25–0.3 below −45°C, but with a rapid increase to 0.45 at −40°C. The multiple scattering factor is found to increase from 0.54 at −60°C (∼11 km altitude) to 0.76 at −20°C (∼5.5 km altitude). Some cases of very high anomalous lidar backscatter occur for clouds of mean temperature ≳ −35° and emittance ≳0.6. The depolarization ratio of the lidar backscattered radiation also shows complicated variations with temperature.

The observed changes in backscatter to extinction ratio are attributed to a change in the ice crystal habit from simple spatial crystals at temperatures <−40°C to more complex aggregates at greater temperatures. This is based on the fact that supercooled water cannot exist at temperatures below −40°C. The high anomalous backscatter is attributed to specular reflection from horizontally oriented plate crystals or from supercooled water droplets. Changes in depolarization ratio at temperatures greater than −40°C are attributed variously to the presence of mixed-phase clouds, to crystal aggregates and to horizontally oriented hexagonal crystals.

Changes in the multiple-scattering factor with temperature (i.e., altitude) are found to agree qualitatively with theoretical predictions, the main effect being a reduction in the multiple-scattering factor (leading to a more transmitting cloud) as the range or altitude of the clouds increases.

Full access
C. M. R. Platt and A. C. Dilley

Abstract

The results from a series of measurements of the beam emissivity of cirrostratus at 10–12 μm wavelengths are presented, using methods of analysis which were discussed in Part I. A ruby lidar and infrared radiometer were used to gather data remotely from the ground. The various sources and magnitudes of error are discussed. The results for eight large cirrostratus systems which were observed on different days gave a mean beam emissivity of 0.54 (or flux emissivity of 0.70). This compares with a value of 0.245 (0.38 for flux obtained during an earlier period (Platt, 1973). The measurements were obtained at 35°S (Adelaide) and 38°S (Aspendale). The cloud systems at Aspendale all formed in similar synoptic situations.

Full access