• Albers, F., 1989: Aircraft measurements of ice crystal concentrations and size distributions with optical imaging probes in cirrus clouds (in German). Ph.D. thesis, University of Cologne, 85 pp.

  • Allen, J. R., 1971: Measurements of cloud emissivity in the 8–13 μ waveband. J. Appl. Meteor.,10, 260–265.

  • Brogniez, G., F. Parol, J. C. Buriez, and Y. Fouquart, 1992: Bidirectional reflectances of cirrus clouds modelized from observations during the International Cirrus Experiment ’89. Proceedings of the International Radiation Symposium, S. Keevallik and O. Karner, Eds., A. Deepak, 133–136.

  • Chepfer, H., G. Brogniez, L. Sauvage, P. Flamant, V. Trouillet, and J. Pelon, 1999: Remote sensing of cirrus radiative parameters during EUCREX’94. Case study of 17 April 1994. Part II: Microphysical models. Mon. Wea. Rev.,127, 504–519.

  • De Haan, J. F., P. B. Bosma, and J. W. Hovenier, 1986: The adding method for multiple scattering calculations of polarized light. Astron. Astrophys.,183, 371–391.

  • Deschamps, P. Y., M. Herman, A. Podaire, M. Leroy, M. Laporte, and P. Vermande, 1990: A spatial instrument for the observation of polarization and bidirectionality of the Earth’s reflectances: POLDER. Proc. Remote Sensing for the Nineties, IGARS 90, Washington, DC, 1769–1774.

  • ——, F. M. Bréon, M. Leroy, A. Podaire, A. Bricaud, J. C. Buriez, and G. Sèze, 1994: The POLDER mission: Instrument characteristics and scientific objectives. IEEE Trans. Geosci. Remote Sens.,32, 598–615.

  • Foot, J. S., 1988: Some observations of the optical properties of clouds. II: Cirrus. Quart. J. Roy. Meteor. Soc.,114, 145–164.

  • Gayet, J. F., P. A. Brown, and F. Albers, 1993: A comparison of in-cloud measurements obtained with six PMS 2D-C probes. J. Atmos. Oceanic Technol.,10, 180–194.

  • ——, G. Febvre, and H. Larsen, 1996a: The reliability of the PMS FSSP in the presence of small ice crystals. J. Atmos. Oceanic Technol.,13, 1300–1310.

  • ——, ——, G. Brogniez, H. Chepfer, W. Renger, and P. Wendling, 1996b: Microphysical properties of cirrus and contrails: Cloud field study on 13 October 1989. J. Atmos. Sci.,53, 126–138.

  • Goloub, P., J. L. Deuzé, M. Herman, and Y. Fouquart, 1994: Analysis of the POLDER polarization measurements performed over cloud covers. IEEE Trans. Geosci. Remote Sens.,32, 78–88.

  • Heymsfield, A. J., and D. Baumgardner, 1985: Summary of a workshop on processing 2-D probe data. Bull. Amer. Meteor. Soc.,66, 437–440.

  • ——, K. M. Miller, and J. D. Spinhirne, 1990: The 27–28 October 1986 FIRE IFO cirrus case study: Cloud microstructure. Mon. Wea. Rev.,118, 2313–2328.

  • Kneizys, F. X., E. P. Shettle, L. W. Abreu, J. H. Chetwynd, G. P. Anderson, W. O. Gallery, J. E. A. Selby, and S. A. Clough, 1988:Atmospheric transmittance/radiance: Computer code LOWTRAN 7. AFGL-TR-88-0177, Environmental Research Paper 1010, Air Force Geophysics Laboratory, Hanscom AFB, MA, 137 pp.

  • Knollenberg, R. G., 1970: The optical array: An alternative to scattering or extinction for airborne particle size determination. J. Appl. Meteor.,9, 86–103.

  • Krupp, C., 1991: Holographic measurements of ice crystals in cirrus clouds during the International Cirrus Experiment ICE 89. Rep. 4th ICE/EUCREX workshop, Laboratoire d’Optique Atmosphérique, Université des Sciences et Technologies de Lille, 214 pp. [Available from Laboratoire d’Optique Atmosphérique, USTL, 59655 Villeneuve d’Ascq, France.].

  • Liou, K. N., 1986: Review. Influence of cirrus clouds on weather and climate processes: A global perspective. Mon. Wea. Rev.,114, 1167–1199.

  • McClatchey, R. A., R. W. Fenn, J. E. A. Selby, F. E. Volz, and J. S. Garing, 1971: Optical properties of the atmosphere (revised). AFCRL-71-0279, Environmental Research Paper 354, Air Force Geophysics Laboratory, Hanscom AFB, MA, 85 pp. [Available from Air Force Geophysics Laboratory, LG Hanscom, FIEL, Hanscom AFB, MA 01731.].

  • Miloshevich, L. M., and A. J. Heymsfield, 1997: A balloon-borne continuous cloud particle replicator for measuring vertical profiles of cloud microphysical properties: Instrument design, performance, and collection efficiency analysis. J. Atmos. Oceanic Technol.,14, 753–768.

  • Morcrette, J. J., and Y. Fouquart, 1985: On systematic errors in parameterized calculations of longwave radiation transfer. Quart. J. Roy. Meteor. Soc.,111, 691–708.

  • Nicolas, F., L. Bissonnette, and P. H. Flamant, 1997: Lidar effective multiple-scattering coefficients in cirrus clouds. Appl. Opt.,36, 3458–3468.

  • Paltridge, C. W., and C. M. R. Platt, 1981: Aircraft measurements of solar and infrared radiation and the microphysics of cirrus clouds. Quart. J. Roy. Meteor. Soc.,107, 367–380.

  • Pelon, J., P. H. Flamant, and M. Meissonnier, 1990: The French airborne backscatter lidar LEANDRE-1: Concept and operation. 15th Int. Laser Radar Conf., Tomsk, USSR, Institute of Atmospheric Optics, 36–39.

  • Platt, C. M. R., 1972: Airborne infrared radiance measurements (10 to 12 microns wavelength) off tropical east-coast Australia. J. Geophys. Res.,77, 1597–1609.

  • ——, 1979: Remote sounding of high clouds: I. Calculation of visible and infrared optical properties from lidar and radiometer measurements. J. Appl. Meteor.,18, 1130–1143.

  • ——, and G. L. Stephens, 1980: The interpretation of remotely sensed high cloud emittances. J. Atmos. Sci.,37, 2314–2322.

  • Raschke, E., J. Schmetz, J. Heintzenberg, R. Kandel, and R. W. Saunders, 1990: The International Cirrus Experiment (ICE)—A joint European effort. ESA J.,14, 193–199.

  • ——, and Coauthors, 1996: European Cloud and Radiation Experiment (EUCREX). Final Report on the Project, EV5V-CT 92-0130 EUCREX-2, 154 pp. [Available from GKSS Research Center, D-2054, Geesthacht, Germany.].

  • Sassen, K., 1991: The polarization lidar technique for cloud research:A review and current assessment. Bull. Amer. Meteor. Soc.,72, 1848–1866.

  • Saunders, R. W., G. Brogniez, J. C. Buriez, R. Meerkotter, and P. Wendling, 1992: A comparison of measured and modeled broad band fluxes from aircraft data during the ICE’89 field experiment. J. Atmos. Oceanic Technol.,9, 391–406.

  • Spinhirne, J. D., and W. D. Hart, 1990: Cirrus structure and radiative parameters from airborne lidar and spectral radiometer observations: The 28 October 1986 FIRE study. Mon. Wea. Rev.,118, 2329–2343.

  • ——, ——, and D. L. Hlavka, 1996: Cirrus infrared parameters and shortwave reflectance relations from observations. J. Atmos. Sci.,53, 1438–1458.

  • Takano, Y., and K. N. Liou, 1989: Solar radiative transfer in cirrus clouds. Part I: Single-scattering and optical properties of hexagonal ice crystals. J. Atmos. Sci.,46, 3–19.

  • Warren, S. G., J. Hahn, J. London, R. M. Chervin, and R. L. Jenne, 1988: Global distribution of total cloud cover and cloud type amounts over the ocean. NCAR Tech. Note TN-317+STR, Boulder, CO, 210 pp.

  • WCP, 1986: A preliminary cloudless standard atmosphere for radiation computation. WCP-112 WMO, Geneva, Switzerland, 60 pp.

  • WCRP, 1990: Scientific plan for the Global Energy and Water Cycle Experiment. WCRP-40, WMO, Geneva, Switzerland, 50 pp.

  • Wielicki, B. A., and Coauthors, 1990: The 27–28 October 1986 FIRE IFO cirrus case study: Comparison of radiative transfer theory with observations by satellite and aircraft. Mon. Wea. Rev.,118, 2356–2376.

  • Young, A. T., 1980: Revised depolarization corrections for atmospheric extinction. Appl. Opt.,19, 3427–3428.

  • Young, S., 1995: Analysis of lidar backscatter profiles in optically thin clouds. Appl. Opt.,34, 7019–7031.

  • View in gallery

    Synoptic scheme of flight: the Falcon overflew the cirrus, with the POLDER downward-looking radiometer and PMS FSSP and PMS OAP-2D2-C probes, while the ARAT underflew the same cirrus bank with lidar and upward-looking PRT-5. The two aircraft were equipped with upward- and downward-looking pyranometers and pyrgeometers. The large-scale cloudiness was documented by NOAA-12/AVHRR data.

  • View in gallery

    (a) NOAA-12/AVHRR image obtained in channel 4 (10.3–11.3 μm) at 0855 UTC 17 April 1994 over the area of aicraft measurements. The track of the Falcon overflying the cirrus bank during mission 4 is superimposed on the satellite image (straight solid line). (b) Same as (a) but for image obtained at 1620 UTC. The track of the Falcon overflying the cirrus bank during mission 5 is superimposed on the satellite image (dashed line).

  • View in gallery

    Atmospheric temperature profile obtained for the ascents of the Falcon during missions 4 and 5.

  • View in gallery

    Fig. 4a. Mission 4 proceeded in three legs for the ARAT at an altitude of 4.6 km, three legs over the cirrus at altitudes of 11.2 and 11.5 km, and three legs inside the cirrus at different altitudes for the Falcon. All legs were flown along the same flight track. Legs of the ARAT are indicated by letter A, while the legs of the Falcon are indicated by letter F. Diamonds indicate the beginning of each leg. Cirrus boundaries as measured by lidar on board the ARAT are also indicated (dots), as well as the location of the POLDER data analysis (Δ).

  • View in gallery

    Fig. 4b. Same as Fig. 4a but for mission 5: three legs for the ARAT at a constant altitude of 4.6 km, two legs over the cirrus at 10.8 km, and six legs at different altitudes inside the cirrus for the Falcon.

  • View in gallery

    Comparison between effective beam emittances derived from AVHRR measurements (channel 4) and PRT-5 on board the ARAT during leg A2.

  • View in gallery

    Plane albedo of the earth–atmosphere–cirrus system obtained from pyranometers on board the Falcon during leg F3 of mission 4 and leg F1 of mission 5 as a function of the distance from CROSS Corsen.

  • View in gallery

    Fig. 7a. Bidimensional values of the backscatter coefficient β(x, z) (in km−1 sr−1) for the three legs of mission 4 as measured by the lidar. The space variables x and z are the distance from CROSS Corsen and the altitude, respectively. Legs A1, A2, and A3 are plotted from top to bottom (see text). Color scales concern the values of β(x, z) between 0 and 0.056 km−1 sr−1. White parts of these color scales indicate the maximum values reached by the backscatter coefficient during the leg.

  • View in gallery

    Fig. 7b. Same as Fig. 7a but for leg A2 of mission 5.

  • View in gallery

    Scatterplot of infrared optical thickness δIR derived from the PRT-5 radiometer as a function of the lidar apparent optical thickness δa532. Only data corresponding to mission 4 are reported here. The slope of the least squares fit of data to a straight line is equal to unity.

  • View in gallery

    Fig. 9a. Bidimensional values of the backscattering depolarization ratio Δ(x, z) (in %) obtained from the airborne lidar for the three legs of mission 4. The space variables x and z are the distance from CROSS Corsen and the altitude, respectively. From top to bottom are plotted legs A1, A2, and A3 (see text). Color scales concern the values of Δ(x, z) between 0% and 70%. The cirrus boundaries are indicated by a red dashed line; the black solid line indicates the altitude of the maximum backscattering coefficient.

  • View in gallery

    Fig. 9b. Same as Fig. 9a but for leg A2 of mission 5.

  • View in gallery

    Synoptic scheme of POLDER measurements principle. The solar zenith angle is θs. The particular viewing direction, characterized by the zenith viewing angle θυ and by the relative viewing azimuth angle φυφs, illuminates a particular pixel of the CCD matrix.

  • View in gallery

    Total reflectance image [(a) and (b)] and polarized reflectance image [(c) and (d)] at 864 nm of wavelength of the cirrus bank overflown during leg F3 of mission 4 [(a) and (c)] and during leg F1 of mission 5 [(b) and (d)]. For (a) and (c), the solar zenith angle is θs = 55.3°, whereas for images (b) and (d) θs = 38.0°. The color scale is 35%–60% for total reflectances and 0%–2.5% for polarized reflectances. Each concentric circle represents a constant value of zenithal viewing angle θυ. The solar principal plane is indicated by the 0°–180° axis (0° corresponding to the solar direction). The azimuthal viewing angle φυ is set from this axis.

  • View in gallery

    Images of cirrus banks, at 864 nm of wavelength, overflown during (a) leg F3 of mission 4 and (b) leg F1 of mission 5. These images are constructed using the central cross line of the CCD matrix [242 detectors corresponding to cross tracks of about (a) 3.6 km and (b) 3 km, respectively]. Locations within the legs of the two entire POLDER images of cirrus cloud total and polarized reflectances are indicated by arrows.

  • View in gallery

    Global particle area distribution as determined by FSSP-100 and 2D-C probes during legs F4, F5, and F6 of mission 4 and legs F3, F4, and F5 of mission 5.

  • View in gallery

    Fig. 14a. Histogram of the probability density of cloud boundaries altitude from lidar measurements corresponding to mission 4: (a1) cloud-top altitude probability density, and (a2) cloud-base altitude probability density.

  • View in gallery

    Fig. 14b. Same as Fig. 14a but for mission 5.

  • View in gallery

    Histogram of the probability density of backscatter-to-extinction ratio k obtained during the legs A1, A2, and A3 of mission 4.

  • View in gallery

    Histogram of the probability density of backscattering depolarization ratio obtained during the legs A1, A2, and A3 of mission 4.

  • View in gallery

    Fig. 17a. Histogram of the probability density of optical thickness δ532 given by the lidar for the three legs of mission 4.

  • View in gallery

    Fig. 17b. Histogram of the probability density of infrared optical thickness δIR given by the infrared radiometer for the three legs of mission 4.

  • View in gallery

    Nadir reflectance of cirrus cloud measured by POLDER on board the Falcon (leg F3) as a function of infrared emissivity obtained from PRT-5 on board ARAT (leg A2). Radiative transfer simulation using Henyey–Greenstein scattering phase function with asymmetry factor ranging between 0.5 and 0.9 is reported on this figure (see text).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 24 21 4
PDF Downloads 4 3 0

Remote Sensing of Cirrus Radiative Parameters during EUCREX’94. Case Study of 17 April 1994. Part I: Observations

View More View Less
  • 1 Laboratoire de Météorologie Dynamique, Ecole Polytechnique, Palaiseau, France
  • | 2 Laboratoire d’Optique Atmosphérique, Université des Sciences et Technologies de Lille, Villeneuve d’Ascq, France
  • | 3 Service d’Aéronomie du CNRS, Université Pierre et Marie Curie, Paris, France
  • | 4 GKSS, Institute of Atmospheric Physics, Geesthacht, Germany
© Get Permissions
Full access

Abstract

During the intensive European Cloud and Radiation Experiment 1994 (EUCREX’94) conducted off the coast of Brittany (France) over the Atlantic Ocean during April 1994, natural cirrus have been analyzed from in situ and remote sensing measurements. The authors have particularly studied the case of 17 April 1994. For this day a cirrus bank is described by a complete dataset, that is, classic airborne thermodynamical measurements, microphysical (forward scattering spectrometer probe) and OAP-2D2-C (optical array probe-cloud) probes manufactured by Particle Measuring System, and radiative (Barnes Precision Radiation Thermometer, Eppley pyranometers, and upward- and downward-looking pyrgeometers) measurements above and below the cloud. More specific airborne instruments were used such as upward backscatter lidar with polarization capabilities (LEANDRE) on board the Avion de Recherches Atmosphériques et Télédétection and the Polarization and Directionality of the Earth’s Reflectances (POLDER) radiometer on board the Falcon for measurement of bidirectional and polarized reflectances. The scene was also documented by NOAA-12/Advanced Very High Resolution Radiometer data. However, the nonsphericity of cirrus ice crystals is clearly demonstrated by the lidar backscattering depolarization ratio measurements (Δp = 24%) and by the absence of any rainbow in POLDER bidirectional reflectances. A specular reflection of the solar light observed on POLDER images indicates the presence of horizontally oriented ice particles in the cloud. All these optical properties will be studied in a companion paper (Part II) and compared with optical properties derived from microphysical models in order to evaluate the radiative impact of natural cirrus clouds.

Corresponding author address: Dr. Gérard Brogniez, Laboratoire d’Optique Atmosphérique, Université des Sciences et Technologies de Lille, U.F.R. de Physique, 59655 Villeneuve d’Ascq, Cedex, France.

Email: gerard.brogniez@univ-lillel.fr

Abstract

During the intensive European Cloud and Radiation Experiment 1994 (EUCREX’94) conducted off the coast of Brittany (France) over the Atlantic Ocean during April 1994, natural cirrus have been analyzed from in situ and remote sensing measurements. The authors have particularly studied the case of 17 April 1994. For this day a cirrus bank is described by a complete dataset, that is, classic airborne thermodynamical measurements, microphysical (forward scattering spectrometer probe) and OAP-2D2-C (optical array probe-cloud) probes manufactured by Particle Measuring System, and radiative (Barnes Precision Radiation Thermometer, Eppley pyranometers, and upward- and downward-looking pyrgeometers) measurements above and below the cloud. More specific airborne instruments were used such as upward backscatter lidar with polarization capabilities (LEANDRE) on board the Avion de Recherches Atmosphériques et Télédétection and the Polarization and Directionality of the Earth’s Reflectances (POLDER) radiometer on board the Falcon for measurement of bidirectional and polarized reflectances. The scene was also documented by NOAA-12/Advanced Very High Resolution Radiometer data. However, the nonsphericity of cirrus ice crystals is clearly demonstrated by the lidar backscattering depolarization ratio measurements (Δp = 24%) and by the absence of any rainbow in POLDER bidirectional reflectances. A specular reflection of the solar light observed on POLDER images indicates the presence of horizontally oriented ice particles in the cloud. All these optical properties will be studied in a companion paper (Part II) and compared with optical properties derived from microphysical models in order to evaluate the radiative impact of natural cirrus clouds.

Corresponding author address: Dr. Gérard Brogniez, Laboratoire d’Optique Atmosphérique, Université des Sciences et Technologies de Lille, U.F.R. de Physique, 59655 Villeneuve d’Ascq, Cedex, France.

Email: gerard.brogniez@univ-lillel.fr

1. Introduction

Cirrus and high-level clouds permanently cover nearly 20% of the globe (Warren at al. 1988), and their impact on climate has been recognized for more than a decade (Liou 1986). An objective of the World Climate Programme is to improve our present knowledge of cirrus clouds, in order to perform reliable prediction of their impact on the earth radiation budget. This goal requires an appropriate description—in completeness, representativity, and accuracy—of their physical and optical properties to be used in models. In order to illustrate the need in this area, it has been shown that the shape and size distribution of crystals and their orientation in space modify to a large extent the scattering properties of cirrus clouds (Takano and Liou 1989; Brogniez et al. 1992) and, consequently, their radiative properties.

Recently, it has become an issue to document cirrus clouds in order to derive their physical and optical properties that are needed to conduct process studies and parametrization and validation activities in the framework of atmospheric model applications. Until now, few field experiments combining in situ and radiative measurements have been conducted on natural cirrus. Paltridge and Platt (1981) derived empirical relations between ice water content and radiative properties whereas Foot (1988) investigated the optical properties of cirrus cloud from aircraft-based radiative and microphysical measurements. The experimental problem is due to several factors such as a semitransparency of this type of cloud, their high-altitude and large spatial inhomogeneity, and a highly variable composition (i.e., ice crystals of different shapes and sizes, and possibly orientation). Part of the issue has been adressed during the course of projects conducted in the United States and in Europe: the two First International Satellite and Cloud Climatology Project Regional Experiments in 1986 and 1991, respectively (see special issues of Monthly Weather Review, Vol. 118, No. 11, 1990 and the Journal of the Atmospheric Sciences, Vol. 52, No. 23, 1995), and the International Cirrus Experiment (ICE’89) in 1989 (Raschke et al. 1990). During the field campaigns in situ probes, passive (i.e., radiometers) and active (i.e., lidars, radars) remote sensors, either airborne or ground based, have been deployed to document different cirrus cloud layers at the mesoscale while satellite data were provided with synoptic-scale observations.

Actually, a synergism of remote sensors is foreseen as a key issue for the retrieval of cirrus cloud characteristics that are relevant to radiative transfer and parametrization studies. This prompted us to address the potential synergism of a new radiometer, the Polarization and Directionality of the Earth’s Reflectances (POLDER), with a backscatter lidar during EUCREX’94, a field campaign that was held in Brittany, France, in April 1994, in the framework of the European Cloud Regional Experiment (EUCREX) project. POLDER is a multispectral and polarization radiometer. Two POLDER radiometers and the Lidar Embarqué pour l’Etude des Aérosols, Nuages, Dynamique, Rayonnement et Espèces minoritaires (LEANDRE) were installed on board two aircraft.

In this paper (Part I) we report here on a well-documented cirrus cloud case (17 April 1994) over the Atlantic Ocean off the Brittany coast. The cloud properties are derived from airborne instruments (radiometers, lidar, in situ probes) and satellite data. Some of the findings using the remote sensors are validated by in situ measurements. Section 2 presents an overview of EUCREX’94. Section 3 deals with the measurement strategy and presents the synoptic conditions prevailing on 17 April 1994. Section 4 presents the instruments and the methodologies used to retrieve the cirrus cloud radiative parameters. Section 5 presents the whole dataset collected on 17 April 1994. In a companion paper (Chepfer et al. 1999, hereafter Part II) this dataset is used to constrain both microphysical models and radiative transfer codes in order to derive the cirrus cloud optical and radiative properties.

2. Overview of EUCREX’94

The European Cloud Radiation Experiment, a program focused on the earth radiation budget and climate change, has been promoted by the European Community starting in 1991 through 1995 (Raschke et al. 1996). This activity took place in the framework of the World Climate Research Programme and Global Energy and Water Cycle Experiment (WCRP 1990). EUCREX is a follow on of the ICE experiment.

The cooperative field campaigns conducted in the framework of EUCREX are presented in Table 1. During EUCREX’94 different types of clouds, that is, cirrus, stratiform midlevel clouds, and stratocumulus, have been documented. Three instrumented research aircraft and a ground-based site (48°25′N, 4°47′W) were involved. The three aircraft were (i) a turboprop Fokker-27 called ARAT (Avion de Recherches Atmosphérique et Télédétection), operated by the French Institut National des Sciences de l’Univers (INSU) and the Institut Geographique National; (ii) a jet Falcon operated by the German Deutsche forschungsanstalt Luftund Raumfart (DLR); and (iii) a turboprop Merlin IV operated by Météo-France. Table 2 presents the aircraft performance and instrumental payloads. The three aircraft were based at the Brest-Guipavas airfield (48°27′N, 4°25′W). The instrumented site was located at CROSS Corsen (Centre de Recherches, d’Observation, de Secours et de Sauvetage or CC hereafter), 1 km away from the seashore. Several ground-based remote sensors, among them two lidars, have been deployed at CC.

The ARAT did participate in all measurements, that is, low-, mid- and high-level clouds; the Falcon was not involved in measurements of low-level clouds; while the Merlin IV flew for measurements of midlevel stratocumulus clouds. A summary of missions undertaken during EUCREX’94, with date, time (UTC), duration, location over the Atlantic Ocean, and objectives is presented in Table 3. The time duration is the maximum flight duration among the three aircraft. All flight tracks started or ended at the ground-based site (CC) for the purpose of intercomparison and validation. The letters A, F, and M in Table 3 stand for ARAT, Falcon, and Merlin IV, respectively.

In the following sections (3–6) we report on the retrieval of cirrus radiative properties using the results obtained during missions 4 and 5 on 17 April 1994.

3. Measurement strategy for the cirrus case on 17 April 1994

On 17 April 1994, the meteorological charts show the presence of high pressure centered over the northern Atlantic, extending to Ireland and partly over England. A low was located over southern France, extending from the east of Spain to northern Italy. On 16 April, the low was stronger over northern France. This system, in connection with another low over northern Iceland and Sweden, resulted in a strong jet stream bringing polar air over northern Europe. The entrance of the jet stream was located over southern England, northeast of Brittany. It may explain the formation of an extended cirrus layer over the Atlantic Ocean near the Brittany coast, as a result of the presence of an ageostrophic circulation. The wind speed at 300 hPa decreased from 40 m s−1 to 15 m s−1 on two consecutive days (16 and 17 April) over the English Channel, as the low moved southward.

Two aircraft, the ARAT and the Falcon (see section 2 and Table 3), were involved in the measurements conducted on 17 April 1994 to document a single cirrus layer observed on satellite data [Meteosat and National Oceanic and Atmospheric Administration-12/Advanced Very High Resolution Radiometer (NOAA-12/AVHRR)]. Two back-to-back flights were conducted on that day, mission 4 in the morning over northern Brittany and mission 5 in the afternoon over southern Brittany. The two flights took place in the cold sector of the front. The ARAT carrying the lidar and radiometers flew underneath the cirrus layer at a constant altitude of 4.5 km all the time, whereas the Falcon overflew the cirrus layer first for POLDER measurements and then made several legs at different levels inside the cirrus cloud for microphysics measurements using in situ probes (see Fig. 1).

The large-scale cloudiness as observed by NOAA-12/AVHRR (channel 4, 10.3–11.3 μm) at 0855 and 1620 UTC is presented in Figs. 2a and 2b. The flight tracks for missions 4 and 5 are displayed as a solid line and a dashed line, respectively. The parameters to be retrieved are presented in Table 4.

The radiosonde temperature profiles at Brest-Guipavas at 0000 and 1200 UTC on 17 April 1994, and those obtained by the Falcon during ascent and descent for the two missions, are in good agreement. The data recorded by the Falcon are displayed in Fig. 3. The tropopause height is at 10.5 km (with an uncertainty of 0.3 km), the temperature is −62°C, and the mean lapse rate in the upper troposphere is −6.9°C/km.

Humidity soundings were performed with three different humidity sensors on board the Falcon: a cryogenic-dewpoint sensor, a Lyman-α sensor, and a Vaisala humicap. The cryogenic-dewpoint sensor provided the most reliable humidity measurements. The Lyman-α sensor gave unrealistically high values of humidity due to an additional absorption by ozone, which has proven difficult to correct. The humicap sensor was unreliable with erratic fluctuations of the measurements. Even so, the cryogenic-dewpoint sensor showed significant disagreement between humidity profiles recorded during ascent and descent, with a 3-h time difference on the same flight, and 5 h between the two missions. The measurements show three moist layers in the vertical at 3–5 km with 30%–50% maximum relative humidity with respect to liquid water, 7–9 km with 40%–60% maximum relative humidity with respect to liquid water, and 10–11.5 km with low relative humidity (5%–20% with respect to liquid water, 10%–35% with respect to ice).

Mission 4 (Fig. 4a) proceeded in three legs for the ARAT—A1, A2, A3, along the same flight track at an altitude of about 4.6 km, and six legs—F1–F6, at different altitudes along the same track for the Falcon. During mission 5 (Fig. 4b), the ARAT performed three legs: A1–A3, at the same altitude along the same track while the Falcon proceeded in eight legs: F1–F8. The cloud boundaries retrieved by lidar (see section 4b) are outlined in Figs. 4a and 4b. The characteristics of each leg are presented in Table 5.

4. Instrumentation and retrieval of radiative parameters

The two aircraft payloads (see Table 4) are complementary for radiation and microphysics measurements. In this section we present the remote sensors and in situ probes, and the methodologies used to retrieve the cirrus cloud parameters.

a. Radiometers

The ARAT and the Falcon were equipped with two sets of Eppley cleardome pyranometers and an Eppley pyrgeometer looking upward and downward (see Tables 2 and 4). A Barnes precision radiation thermometer (PRT-5) looking upward was installed on board the ARAT to measure infrared radiance (L) in the 8–14-μm spectral region within a 35-mrad field of view. It corresponds to a footprint of 100-m diameter at a 3-km range. All the radiometer data were recorded using a 1-Hz sampling frequency.

Upward and downward effective beam emittances ε↑↓ are calculated using IR radiances L↑↓ (Allen 1971):
i1520-0493-127-4-486-e1
where L is the downward IR radiance below the cirrus cloud, which is measured by the PRT-5 on board the ARAT. Here, L is derived from NOAA-12/AVHRR data; L↑↓0 and L↑↓1 are the IR radiances that would be measured if the cirrus cloud emittance is zero (i.e., clear sky) or unity (i.e., blackbody), respectively. The actual effective beam emittances computed using Eq. (1) are slightly different than the true absorption beam emittances as they account for the infrared scattering contribution (Platt and Stephens 1980).

The downward beam emittance ε requires the knowledge of the clear-sky L0 and blackbody radiances L1. Here, L0 is measured in clear air area during the flight, in the absence of cirrus cloud as indicated by the lidar. For L1, a theoretical value for L0 is computed first using the LOWTRAN-7 code (Kneizys et al. 1988). Then, L1 is computed by adding a hypothetical blackbody at an altitude corresponding to the altitude of the maximum of the backscatter coefficient, which is determined by lidar (see section 4b). The temperature at this level is taken from the Falcon sounding. To compute L0 the input parameters are (i) the transmission and spectral characteristics of the interference filter, (ii) the temperature and humidity profiles recorded by the Falcon, and (iii) a vertical profile of ozone density corresponding to a midlatitude atmospheric model (McClatchey et al. 1971). Because the humidity measurements are less reliable, they are adjusted in order to make the computed values for L0 equal to the measurement. The adjustment bears on ±5% or less of the humidity value.

In order to derive ε, the upward IR radiance in clear-sky L0 corresponds to the maximum temperature of a histogram plot of the NOAA-12/AVHRR channel 4 data.

The effective infrared absorption optical thickness δIR is linked to the effective beam emittance (Platt 1972):
δIR↑↓
The intrinsic radiometric noise corresponds to fluctuations of the order of 0.02 in δIR. The main uncertainty in deriving δIR is the cirrus cloud temperature. The variability of the cloud geometrical thickness corresponds to a 5-K uncertainty, which in turn corresponds to an uncertainty of 0.03 in δIR. In comparison, the error resulting from the nonscattering cloud approximation is a few percent or less (Platt and Stephens 1980; Brogniez et al. 1992). The overall uncertainty of the effective beam emittance is about 15%. Figure 5 displays ε retrieved from PRT-5 and AVHRR data as a function of latitude for mission 4. The AVHRR data at 0855 UTC are located along the ARAT flight track. The relative change with latitude for the two datasets agree, while the absolute values show a significant disagreement. It is not due to the difference in footprint (i.e., 100 m for PRT5 and 1 km for AVHRR) but to the temperature used in the computation.

A proper calibration of radiometers is needed to derive the radiative fluxes. In response to this need an intercomparison flight involving the three aircraft was conducted over the Atlantic Ocean on a clear day (mission 1; see Table 3). The aircraft flew in close formation within tens of meters at a constant altitude of 1.6 km. The upward and downward shortwave fluxes F↑↓SW recorded by the pyranometers have been corrected for a change in solar zenith angle (θs = 44.42°–45.81°) during the flight. The results are presented in Table 6. The three datasets are consistent in that the measurements fall within a common range.

The downward longwave fluxes FLW recorded by the pyrgeometer on board the ARAT for two flight altitudes of 1.5 and 5.2 km are compared to theoretical values in Table 7. The experimental data are corrected according to Saunders et al. (1992). The theoretical values are computed using a narrowband radiative transfer code (Morcrette and Fouquart 1985) and a maritime aerosol model (WCP 1986). An aerosol optical thickness at 550 nm of 0.138 between 0 and 5.2 km of altitude (and 0.013 between 0 and 1.5 km) is used to fit the measurements with theoretical values.

A value of the plane albedo (a) of the sea–cirrus cloud–atmosphere system is derived from the pyrgeometer data recorded by the Falcon (F) overflying the cirrus layer:
i1520-0493-127-4-486-e3
Figure 6 shows the plane albedo as a function of distance for leg F3 of mission 4 and leg F1 of mission 5. The results are corrected for a mean solar zenith angle of θs = 55° for mission 4 and θs = 38° for mission 5, respectively.

b. Backscatter lidar

The backscatter lidar provides direct information on cirrus cloud structure (height, geometrical thickness), optical properties (profiles of extinction and backscatter coefficients), and parameters linked to the microphysical characteristics of the cloud particle (backscatter-to-extinction lidar ratio, depolarization ratio).

The lidar LEANDRE was installed on board the ARAT (Pelon et al. 1990). The transmitter is an Nd-YAG laser emitting 60 mJ per pulse at 0.53 μm and 100 mJ per pulse at 1.06 μm with a 10-Hz pulse repetition frequency. The two laser emissions are linearly polarized. The laser beam divergence of 2.5 mrad gives a footprint of 7 m at 3 km. The optical power backscattered off the atmosphere is collected by a receiver telescope (30-cm diameter or 0.08 m2 collecting area, 3.5 mrad field-of-view full angle). Three detection channels record two signals in parallel, at 0.53 μm with parallel (S) and perpendicular (S) polarizations and at 1.06 μm. The lidar signals at 0.53 μm can be calibrated in absolute values in clean air corresponding to a low aerosols burden (see the appendix). The vertical resolution along the line of sight is 15 m; it is set by the transient digitizer sampling frequency (10 MHz). Twelve lidar shots in a row are averaged to increase the signal-to-noise ratio (SNR).

The backscatter lidar can operate either in a near zenith or nadir viewing mode. It takes 1 min to switch the looking configuration, so up- and downlooking measurements can be implemented on successive legs during the same flight. The uplooking measurements are made at a fixed viewing angle of 20° toward the right of the fly track and 10° toward the rear of the aircraft.

For EUCREX’94 the cirrus optical properties are derived using the two 0.53-μm channels (Spinhirne and Hart 1990; Young 1995). Figures 7a and 7b display 2D plots of the backscattering coefficient as a function of distance from CROSS Corsen for legs A1, A2, and A3 of mission 4 and leg A2 of mission 5. The maximum value of the backscattering coefficient reaches 0.06 km−1 sr−1 and 0.10 km−1 sr−1 during mission 4 and 5, respectively. The cirrus boundaries are indicated by a red dashed line; the altitude of the maximum backscattering coefficient is indicated by a black solid line.

Figure 8 displays a scatterplot of δIR versus δa532, δIR is derived from data collected by the upward looking PRT-5 radiometer on board the ARAT (see section 4a), and δa532 is the lidar apparent optical thickness not corrected for multiple scattering effects (Nicolas et al. 1997). The best least-squares fit of the data is a straight line set to start at the origin. It shows that δIR/δa532 ≈ 0.99, whereas a factor of 2 is expected from theory for large particles (i.e., geometrical approximation). According to this result and previous studies (Nicolas et al. 1997), we correct the optical thickness by a multiplicative factor of 2. The large scatter of data displayed in Fig. 8 is due to the different pointing direction for the lidar and the PRT-5 (off by ≈20°) and sampling volumes (the footprints are separated by 1 km at the altitude of the cirrus cloud).

A profile of depolarization ratio for particles Δp is derived from a total depolarization ratio for particles and molecules:
i1520-0493-127-4-486-e4
Then Eq. (4) is normalized to the molecular depolarization ratio Δm = 2.8% (Young 1980) assuming clean air above or below the cirrus cloud. Finally, Eq. (4) is corrected for the molecular contribution Δm to derive Δp displayed in Figs. 9a and 9b. Here, Δp is plotted as a function of distance from CROSS Corsen for legs A1, A2, and A3 of mission 4 and leg A2 of mission 5. The maximum value is 50%.

c. POLDER

Two POLDER downward-looking instruments were operated by the Laboratoire d’Optique Atmosphérique (Deschamps et al. 1990) on board the Falcon and the ARAT. POLDER measures the spectral radiances and polarization characteristics of the solar light scattered off the clouds, the atmosphere, and the earth surface. The principle of measurement is presented in Fig. 10. The airborne instruments are a simplified version of the multichannel radiometer on board the Japanese Advanced Earth Observing System platform launched in August 1996 (Deschamps et al. 1994).

The two key components of POLDER are (i) a charge-coupled device (CCD) matrix (see Table 8) set in the focal plane of a wide field-of-view telecentric optics (focal length of 3.565 mm) and (ii) a rotating wheel carrying five narrowband filters, which transmit polarized light (Table 8). For POLDER on board the Falcon the field of view is ±52° along the aircraft track and ±42° cross track. At 1 km the footprint is 2.5 km × 1.8 km and the resolution (for one pixel) is 9 m × 8 m.

The POLDER measurements presented hereafter were taken in the presence of a single cirrus layer only in the absence of low or midclouds underneath. The information provided by POLDER is composed of two-dimensional normalized bidirectional total reflectances ρi and two-dimensional normalized bidirectional polarized reflectances ρip, where i stands for any of the five channels. We have
i1520-0493-127-4-486-e5a
where Li and Lip are the incoming total and polarized radiances, respectively; Eis the solar irradiance at the top of the atmosphere; and θs the solar zenith angle. The uncertainty on reflectances is about 10%.

The present analysis on the cirrus cloud radiative properties is conducted using the 864-nm channel because the molecular contribution to the signals is weak and there is no noticeable absorption. Two sequences of POLDER measurements during mission 4 and 5 are displayed in Fig. 11, which presents the polar projection of POLDER images recorded on homogeneous and thick cirrus clouds during leg F3 of mission 4 (Figs. 11a and 11c) and leg F1 of mission 5 (Figs 11b and 11d). The POLDER measurements are contemporary with lidar measurements. They are correlated to the highest values of the plane albedo (see Fig. 6), that is, at about 90 km and 160 km from CC, for F3 of mission 4, and F1 of mission 5, respectively. Figures 11a and 11c are an average of two images; Figs. 11b and 11d an average of three images. Figures 11a and 11b present the normalized bidirectional total reflectances [Eq. (5a)]. The normalized polarized reflectances [Eq. (5b)] are presented in Figs. 11c and 11d. Each concentric circle deals with a constant zenith angle θυ. The solar principal plane is along the 0°–180° axis, the azimuthal viewing angle φυ is set from this axis, and 0° corresponds to the solar direction. The corresponding solar zenith angles and Falcon headings are θs = 55° and H = +3° (Figs. 11a and 11c), and θs = 38° and H = −20° (Figs. 11b and 11d), respectively.

The total reflectance reaches 65%, while the polarized reflectance reaches only 2.5%. The lowest value for the total reflectance is on the order of 50% for mission 5 (Fig. 11b) and 35% for mission 4 (Fig. 11a). In Figs. 11b and 11d, we notice the presence of a hot spot. This hot spot occurs at a zenithal viewing angle θυ = θs = 38.0°, and it is located in the solar principal plane. This phenomena is persistent for several tens of kilometers;it is even stronger when the cirrus cloud becomes thicker. According to this correlation it is not due to a sea glint effect but to a specular reflection of sunlight on cirrus cloud. Such a phenomenon cannot be observed in Figs. 11a and 11c; it is due to high values of the solar zenith angle in the morning, the POLDER configuration, and geometry of observation.

Figures 12a and 12b present reconstructed POLDER images for the total reflectance. They are made of the only central line perpendicular to the Falcon heading on successive POLDER images and are indicated by two orientations, that is, 93° and 267° in Fig. 11a for leg F3 of mission 4, and 70° and 250° in Fig. 11b for F1 of mission 5. The length of the track is 70 km (i.e., about 700 successive images) for F3 (mission 4) and 214 km (i.e., about 2140 images) for F1 (mission 5). The cross-track dimension is limited by the POLDER field of view and range to the cirrus layer. It is 3.6 km for F3 (mission 4) and 3 km for leg F1 (mission 5) when the Falcon overflew at a height of 1.5 km and 1.3 km, respectively. The arrows in Figs. 12a and 12b indicates the location of the POLDER image presented in Fig. 11; they are correlated to high values of the total reflectance.

d. Microphysics by in situ measurements

Two complementary PMS optical probes (from Particle Measuring System Inc., Boulder, CO) on board the Falcon have been used to determine a size distribution of the particles: (i) a PMS forward scattering spectrometer probe (FSSP-100) assuming spherical particles with equivalent diameter ranging from 2 to 47 μm, and (ii) a PMS optical array probe (OAP-2D2-C), which provides two-dimensional images of particles with mean size from 50 to 800 μm.

Heymsfield et al. (1990) have addressed the problem of using FSSP for a retrieval of the microphysical characteristics of ice cloud particles. Gayet et al. (1996a) showed that FSSP information is pertinent when it agrees with OAP-2D2-C measurements in a common size range, which is the case in the present study. The nominal size resolution of OAP-2D-2C is 25 μm for an airspeed of 100 m s−1; the resolution is decreased to 50 to 70 μm for a true airspeed in the range of 150–180 m s−1.

The size distribution is expressed in cross-section area, which is obtained directly from the OAP-2D2-C probe after correction for the electronic response time (Albers 1989). The cross-section area is computed according to the approaches proposed by Knollenberg (1970), for the depth of field, and Heymsfield and Baumgardner (1985), for the “entire in method.” A discussion on the relative accuracy of 2D-C probes and processing methods is presented in Gayet et al. (1993).

Figure 13 presents the distribution of cross-section area for legs F4, F5, and F6 of mission 4 (panel a), and legs F3, F4, and F5 of mission 5 (panel b). The ice water content (IWC) particle number density (N) and effective cross-section area of particle (Aeff) in Table 9 are calculated according to Gayet et al. (1996b). The results show a high variability of both N and IWC on data taken on successive legs and two missions. During mission 4, N and IWC increased with decreasing height inside the cirrus layer evidencing a settling process, whereas a strong vertical inhomogeneity prevailed during mission 5. The effective cross-section area Aeff is more constant, considering that for the whole dataset the mean value is 〈Aeff〉 = 1852 μm2 with a standard deviation σA = 356 μm2. For an equivalent sphere they correspond to a mean effective radius 〈reff〉 = 24 μm, and standard deviation σr = 2 μm.

5. Cirrus properties

a. Cloud morphology

Histograms of cloud boundaries retrieved by lidar are displayed in Fig. 14. For mission 4 they show a cloud top at a maximum altitude of 10.6 km, that is, for a 95% cumulative probability of the observations and a mean value of 9.9 km with a standard deviation of 0.5 km (Fig. 14a1). The tropopause is located at 10.5 km (see section 3). The cloud base is located at 7.8 km for a 5% cumulative probability, and the mean value is 8.5 km with a standard deviation of 0.6 km (Fig. 14a2). These boundaries agree well with the temperature soundings, which show a lower boundary for the upper moist layer at 7 km (see section 3). The mean geometrical thickness retrieved by lidar is 1.3 km (50% cumulative probability of the observations) with a standard deviation of 0.26 km (i.e., a 20% normalized standard deviation). The layer height is lower to the west end by about 1 km. The cirrus layer fills only one-half of the upper moist layer; even so, the mean geometrical thickness is half the distance between the upper and lower boundaries (2.8 km). A similar analysis for mission 5 undertaken a few hours later shows the cloud top at 10.5 km (Fig. 14b1) and cloud base at 6.8 km (Fig. 14b2). The results suggest the importance of mesoscale process interaction with the background dynamics for the formation of cirrus clouds.

b. Microphysics

Crystal size and shape play a key role in scattering processes in the context of radiative transfer problems. An assumption of spherical particles in cirrus clouds is frequently made despite its limitation for interpretation. The relevant microphysical characteristics of ice crystals for the case study of 17 April 1994, that is, size, shape, and orientation, are presented hereafter.

1) Size

Microphysical in situ measurements (section 4d) show an effective radius of about 24 μm for equivalent spheres.

2) Shape

The measurements indicate the presence of nonspherical particles according to (i) the significant lidar depolarization ratios measured by LEANDRE (Δp = 20%–30%), and (ii) the absence of a rainbow at 150° (scattering angle) in bidirectional reflectances measured by POLDER (see Fig. 11), as a rainbow characterizes the presence of spherical particles. Our results and conclusions are in agreement with previous studies that reported the existence of a large variety of crystal shapes in cirrus clouds (e.g., Krupp 1991; Miloshevich and Heymsfield 1997).

3) Orientation

A possible preferred orientation of ice crystals has been discussed in the past. It has been reported using lidar measurements. This phenomonon has not been observed by in situ measurements. It cannot be observed by LEANDRE in the present case because the lidar is pointing at 20° from the zenith. We take advantage of the POLDER measurements to address this issue. During the course of leg F1 of mission 5 the POLDER data show a peak of reflectance in the solar principal plane when the viewing zenith angle is equal to the solar zenith angle θυ = θs = 38° (see Figs. 11b and 11d). This specular reflection of solar light (cirrus glint) is an indication of the presence of horizontally oriented particles. However, it cannot be concluded that the same phenomenon did not occur during leg F3 of mission 4 when the solar angle was θs ≈ 55° because the POLDER observation angle θυ = ±50° precludes such an observation.

c. Optical properties

1) Backscattering phase function and depolarization ratio

The probability density function of the backscatter-to-extinction lidar ratio k measured during mission 4 is presented in Fig. 15. The mean value is 0.024 sr−1 with a standard deviation of 0.006 sr−1. It corresponds to a backscattering phase function P(π) = 0.30 with a standard deviation of 0.08. Figure 16 displays the probability density function of the lidar depolarization ratio for mission 4. The mean value is Δp = 24% with a standard deviation of 4%. The values obtained for P(π) and Δp, which are characteristics of the size, shape, and orientation of particles, are compared to the results of calculations for nonspherical particles (see Part II).

2) Optical thickness

The probability density function of δ532 during mission 4, legs A1, A2, and A3 (see section 4b), displays a negative exponential distribution (Fig. 17a). The mean optical thickness is 0.49 with a standard deviation of 0.48. The same histogram for the infrared optical thickness δIR (see section 4a) shows the same behavior (Fig. 17b), and the mean value is 0.25 with a standard deviation of 0.24. Similar results are obtained for mission 5.

3) Asymmetry factor

The asymmetry factor g for ice crystals is essential in the radiative transfer problem. A value for g can be derived using shortwave reflectances and infrared emittances according to Wielicki at al. (1990). The theoretical reflectances are computed using an adding–doubling code (De Haan et al. 1986) with a clear-sky albedo of about 5%, an atmosphere made of a cirrus layer, and particles with optical properties that can be described by Henyey–Greenstein scattering functions. The directional reflectances have been computed for g ranging between 0.5 and 0.9, and shortwave optical thickness δSW between 0.1 and 15. The infrared optical thickness δIR is calculated using the approximation of large particles, and the cloud emittance is derived from Eq. (2). Using this method, the cloud directional reflectances can be plotted as a function of cloud emittance.

Following the method presented above, a comparison was made with the thicker and more homogeneous part of the cirrus layer observed during mission 4. In Fig. 18 the normalized reflectances at nadir measured by POLDER on board the Falcon (leg F3) are plotted as a function of the beam emittances derived from the PRT-5 on board the ARAT (leg A2, see section 4a). The mean solar zenith angle is θs = 55°. The results show a weak correlation of the experimental data. A noncoincidence in footprint between the two radiometers (PRT-5 and POLDER) as discussed above may account for the discrepancy.

6. Discussion and conclusions

The various measurements collected on an extended cirrus cloud on 17 April 1994 during EUCREX’94 constitutes an extremely valuable dataset for radiative transfer studies at the mesoscale. The dataset includes different measurements: (i) in situ thermodynamical measurements (pressure, temperature, and humidity profiles); (ii) in situ microphysics measurements; (iii) radiative measurements, for example, upward and downward solar and infrared fluxes collected from above and below the cirrus cloud, and nadir infrared radiances collected from below the cloud; (iv) airborne lidar measurements, that is, backscattered profiles, linear depolarization ratio, backscatter-to-extinction lidar ratio, and optical thickness; and (v) airborne POLDER bidirectional reflectances, and bidirectional polarized reflectances.

This dataset, collected during the case study on 17 April, allows us to present a description as complete as possible of a cirrus cloud at the mesoscale, regarding its radiative properties (i.e., albedo, transmission, optical thickness) and microphysical properties (i.e., crystals size). Additional information has been collected on the way the scattering processes by ice crystals can modify the state of polarization of visible light (i.e., lidar depolarization ratio and POLDER bidirectional polarized reflectances).

An analysis conducted on the dataset leads to the following conclusions.

  1. The cirrus cloud observed on 17 April 1994 is strongly inhomogeneous and its spatial variability seems to be quasiperiodic at a scale of about 10 km.

  2. The in situ microphysical measurements indicate an equivalent radius of 20 μm for the cirrus cloud particles (under the crude assumption of spherical particles), with no information on the shape of particles. The assumption of spherical particles that is currently made is somewhat contradictory with the result below but is not adressed in the present study, as it requires dedicated instrumentation not available during EUCREX’94.

  3. Nonsphericity of cirrus cloud particles is suggested by different observations: significant values of the lidar depolarization ratio (see Sassen 1991), and absence of a rainbow in bidirectional polarized reflectances made by POLDER for scattering angles around 150°. This suggests a significant contribution of nonspherical particles (Goloub et al. 1994).

  4. The presence of particles horizontally oriented is indicated by specular reflection in solar light collected by POLDER (total reflectance and polarized reflectance). This phenomenon persists for some time and becomes stronger when the cloud is optically thicker, indicating clearly that the specular reflection is not due to sea glint but must be attributed to the presence of horizontally oriented crystals in the cirrus cloud.

In addition, a first analysis of the optical properties of ice crystals has been conducted using simulations based on Henyey–Greenstein scattering functions and the present dataset. A comparison of the nadir reflectances measured by POLDER and the effective beam emittances derived from the PRT-5 radiometer, with different simulation results, does not give satisfactory agreement. This is attributed to the fact that the two instruments were not on board the same aircraft and do not have the same spatial resolution.

As a summary, this paper presents an extented dataset collected during the EUCREX’94 campaign, and the manner in which the different data have been processed. We also check the self-consistency of the dataset, which is used in Part II, to derive the optical and microphysical properties of the cirrus cloud documented on 17 April 1994.

Acknowledgments

This work was supported by the European Community under Contract EV5V-CT 92-0130 (EUCREX-2), the Centre National d’Etudes Spatiales, and the European Space Agency. The authors are specially grateful to Y. Fouquart who has managed EUCREX-2, and to the members of the Centre d’Aviation Météorologique, INSU, and DLR who operated the Merlin, the ARAT and the Falcon, respectively. Thanks are also due to the French weather service and to many scientists, engineers, technicians, and students, particularly H. Fimpel (DLR), R. Valentin (LMD), J. Y. Balois, and C. Verwaerde (LOA) who participated in the experiment. The authors are grateful to two anonymous reviewers for their fruitful comments.

REFERENCES

  • Albers, F., 1989: Aircraft measurements of ice crystal concentrations and size distributions with optical imaging probes in cirrus clouds (in German). Ph.D. thesis, University of Cologne, 85 pp.

  • Allen, J. R., 1971: Measurements of cloud emissivity in the 8–13 μ waveband. J. Appl. Meteor.,10, 260–265.

  • Brogniez, G., F. Parol, J. C. Buriez, and Y. Fouquart, 1992: Bidirectional reflectances of cirrus clouds modelized from observations during the International Cirrus Experiment ’89. Proceedings of the International Radiation Symposium, S. Keevallik and O. Karner, Eds., A. Deepak, 133–136.

  • Chepfer, H., G. Brogniez, L. Sauvage, P. Flamant, V. Trouillet, and J. Pelon, 1999: Remote sensing of cirrus radiative parameters during EUCREX’94. Case study of 17 April 1994. Part II: Microphysical models. Mon. Wea. Rev.,127, 504–519.

  • De Haan, J. F., P. B. Bosma, and J. W. Hovenier, 1986: The adding method for multiple scattering calculations of polarized light. Astron. Astrophys.,183, 371–391.

  • Deschamps, P. Y., M. Herman, A. Podaire, M. Leroy, M. Laporte, and P. Vermande, 1990: A spatial instrument for the observation of polarization and bidirectionality of the Earth’s reflectances: POLDER. Proc. Remote Sensing for the Nineties, IGARS 90, Washington, DC, 1769–1774.

  • ——, F. M. Bréon, M. Leroy, A. Podaire, A. Bricaud, J. C. Buriez, and G. Sèze, 1994: The POLDER mission: Instrument characteristics and scientific objectives. IEEE Trans. Geosci. Remote Sens.,32, 598–615.

  • Foot, J. S., 1988: Some observations of the optical properties of clouds. II: Cirrus. Quart. J. Roy. Meteor. Soc.,114, 145–164.

  • Gayet, J. F., P. A. Brown, and F. Albers, 1993: A comparison of in-cloud measurements obtained with six PMS 2D-C probes. J. Atmos. Oceanic Technol.,10, 180–194.

  • ——, G. Febvre, and H. Larsen, 1996a: The reliability of the PMS FSSP in the presence of small ice crystals. J. Atmos. Oceanic Technol.,13, 1300–1310.

  • ——, ——, G. Brogniez, H. Chepfer, W. Renger, and P. Wendling, 1996b: Microphysical properties of cirrus and contrails: Cloud field study on 13 October 1989. J. Atmos. Sci.,53, 126–138.

  • Goloub, P., J. L. Deuzé, M. Herman, and Y. Fouquart, 1994: Analysis of the POLDER polarization measurements performed over cloud covers. IEEE Trans. Geosci. Remote Sens.,32, 78–88.

  • Heymsfield, A. J., and D. Baumgardner, 1985: Summary of a workshop on processing 2-D probe data. Bull. Amer. Meteor. Soc.,66, 437–440.

  • ——, K. M. Miller, and J. D. Spinhirne, 1990: The 27–28 October 1986 FIRE IFO cirrus case study: Cloud microstructure. Mon. Wea. Rev.,118, 2313–2328.

  • Kneizys, F. X., E. P. Shettle, L. W. Abreu, J. H. Chetwynd, G. P. Anderson, W. O. Gallery, J. E. A. Selby, and S. A. Clough, 1988:Atmospheric transmittance/radiance: Computer code LOWTRAN 7. AFGL-TR-88-0177, Environmental Research Paper 1010, Air Force Geophysics Laboratory, Hanscom AFB, MA, 137 pp.

  • Knollenberg, R. G., 1970: The optical array: An alternative to scattering or extinction for airborne particle size determination. J. Appl. Meteor.,9, 86–103.

  • Krupp, C., 1991: Holographic measurements of ice crystals in cirrus clouds during the International Cirrus Experiment ICE 89. Rep. 4th ICE/EUCREX workshop, Laboratoire d’Optique Atmosphérique, Université des Sciences et Technologies de Lille, 214 pp. [Available from Laboratoire d’Optique Atmosphérique, USTL, 59655 Villeneuve d’Ascq, France.].

  • Liou, K. N., 1986: Review. Influence of cirrus clouds on weather and climate processes: A global perspective. Mon. Wea. Rev.,114, 1167–1199.

  • McClatchey, R. A., R. W. Fenn, J. E. A. Selby, F. E. Volz, and J. S. Garing, 1971: Optical properties of the atmosphere (revised). AFCRL-71-0279, Environmental Research Paper 354, Air Force Geophysics Laboratory, Hanscom AFB, MA, 85 pp. [Available from Air Force Geophysics Laboratory, LG Hanscom, FIEL, Hanscom AFB, MA 01731.].

  • Miloshevich, L. M., and A. J. Heymsfield, 1997: A balloon-borne continuous cloud particle replicator for measuring vertical profiles of cloud microphysical properties: Instrument design, performance, and collection efficiency analysis. J. Atmos. Oceanic Technol.,14, 753–768.

  • Morcrette, J. J., and Y. Fouquart, 1985: On systematic errors in parameterized calculations of longwave radiation transfer. Quart. J. Roy. Meteor. Soc.,111, 691–708.

  • Nicolas, F., L. Bissonnette, and P. H. Flamant, 1997: Lidar effective multiple-scattering coefficients in cirrus clouds. Appl. Opt.,36, 3458–3468.

  • Paltridge, C. W., and C. M. R. Platt, 1981: Aircraft measurements of solar and infrared radiation and the microphysics of cirrus clouds. Quart. J. Roy. Meteor. Soc.,107, 367–380.

  • Pelon, J., P. H. Flamant, and M. Meissonnier, 1990: The French airborne backscatter lidar LEANDRE-1: Concept and operation. 15th Int. Laser Radar Conf., Tomsk, USSR, Institute of Atmospheric Optics, 36–39.

  • Platt, C. M. R., 1972: Airborne infrared radiance measurements (10 to 12 microns wavelength) off tropical east-coast Australia. J. Geophys. Res.,77, 1597–1609.

  • ——, 1979: Remote sounding of high clouds: I. Calculation of visible and infrared optical properties from lidar and radiometer measurements. J. Appl. Meteor.,18, 1130–1143.

  • ——, and G. L. Stephens, 1980: The interpretation of remotely sensed high cloud emittances. J. Atmos. Sci.,37, 2314–2322.

  • Raschke, E., J. Schmetz, J. Heintzenberg, R. Kandel, and R. W. Saunders, 1990: The International Cirrus Experiment (ICE)—A joint European effort. ESA J.,14, 193–199.

  • ——, and Coauthors, 1996: European Cloud and Radiation Experiment (EUCREX). Final Report on the Project, EV5V-CT 92-0130 EUCREX-2, 154 pp. [Available from GKSS Research Center, D-2054, Geesthacht, Germany.].

  • Sassen, K., 1991: The polarization lidar technique for cloud research:A review and current assessment. Bull. Amer. Meteor. Soc.,72, 1848–1866.

  • Saunders, R. W., G. Brogniez, J. C. Buriez, R. Meerkotter, and P. Wendling, 1992: A comparison of measured and modeled broad band fluxes from aircraft data during the ICE’89 field experiment. J. Atmos. Oceanic Technol.,9, 391–406.

  • Spinhirne, J. D., and W. D. Hart, 1990: Cirrus structure and radiative parameters from airborne lidar and spectral radiometer observations: The 28 October 1986 FIRE study. Mon. Wea. Rev.,118, 2329–2343.

  • ——, ——, and D. L. Hlavka, 1996: Cirrus infrared parameters and shortwave reflectance relations from observations. J. Atmos. Sci.,53, 1438–1458.

  • Takano, Y., and K. N. Liou, 1989: Solar radiative transfer in cirrus clouds. Part I: Single-scattering and optical properties of hexagonal ice crystals. J. Atmos. Sci.,46, 3–19.

  • Warren, S. G., J. Hahn, J. London, R. M. Chervin, and R. L. Jenne, 1988: Global distribution of total cloud cover and cloud type amounts over the ocean. NCAR Tech. Note TN-317+STR, Boulder, CO, 210 pp.

  • WCP, 1986: A preliminary cloudless standard atmosphere for radiation computation. WCP-112 WMO, Geneva, Switzerland, 60 pp.

  • WCRP, 1990: Scientific plan for the Global Energy and Water Cycle Experiment. WCRP-40, WMO, Geneva, Switzerland, 50 pp.

  • Wielicki, B. A., and Coauthors, 1990: The 27–28 October 1986 FIRE IFO cirrus case study: Comparison of radiative transfer theory with observations by satellite and aircraft. Mon. Wea. Rev.,118, 2356–2376.

  • Young, A. T., 1980: Revised depolarization corrections for atmospheric extinction. Appl. Opt.,19, 3427–3428.

  • Young, S., 1995: Analysis of lidar backscatter profiles in optically thin clouds. Appl. Opt.,34, 7019–7031.

APPENDIX

Analysis of Lidar Signal

The lidar signal S(x, z) after digitization can be written as
i1520-0493-127-4-486-ea1
where z′ is a dummy variable, x is the horizontal location of measurements, z is the altitude, and z0 is the aircraft flight altitude. In addition, β(x, z) and α(x, z) are the total backscattering coefficient (in km−1 sr−1) and extinction coefficient (km−1) for particles and molecules, respectively; η is a multiple scattering coefficient (Platt 1979). The instrumental constant K depends on the system characteristics, that is, transmitted laser energy, receiver collecting area, optical efficiency, photodetector quantum yield, and electronic gain. The noise is computed at long ranges after the lidar signal is negligible (SNR ≫ 1).

The cloud boundaries—top, base, transitions between clean air (low aerosol content and no cloud), and cloud layers—are assigned by lidar using a threshold algorithm (Young 1995). The uncertainty is estimated to be 15–30 m for SNR ≥ 3.

In order to solve Eq. (A1) for β (or α) an additional equation linking these two parameters is required:
βx, zx, z
k is the backscatter-to-extinction lidar ratio also written as k = P(π)/4π, where P(π) is the backscattering phase function for cloud particles whose integral value is equal to unity. Inside the cloud k is constant, for example, only the particle number density changes to allow β(x, z) and α(x, z) to vary with range. Equation (A2) holds also for apparent parameters accounting for multiple scattering (x, z) = kaαa(x, z) with ka = k/η and αa = ηα.

An apparent backscatter-to-extinction ratio ka, at 532 nm, accounting for multiple scattering, is determined using two different expressions for the integrated backscattering coefficient over the entire cirrus cloud (Spinhirne et al. 1996): γ = ztzbβ(x, y) dy and γ = ka(1 − T2)/2, where zb and zt are the cloud base and top, respectively, and T2 is the cirrus two-way transmission. It results in a linear dependence of T2 as a function of γ. The experimental lidar data are fitted by a straight line using a least squares method. An extrapolation at T2 = 0 gives ka = 2γ.

Fig. 1.
Fig. 1.

Synoptic scheme of flight: the Falcon overflew the cirrus, with the POLDER downward-looking radiometer and PMS FSSP and PMS OAP-2D2-C probes, while the ARAT underflew the same cirrus bank with lidar and upward-looking PRT-5. The two aircraft were equipped with upward- and downward-looking pyranometers and pyrgeometers. The large-scale cloudiness was documented by NOAA-12/AVHRR data.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 2.
Fig. 2.

(a) NOAA-12/AVHRR image obtained in channel 4 (10.3–11.3 μm) at 0855 UTC 17 April 1994 over the area of aicraft measurements. The track of the Falcon overflying the cirrus bank during mission 4 is superimposed on the satellite image (straight solid line). (b) Same as (a) but for image obtained at 1620 UTC. The track of the Falcon overflying the cirrus bank during mission 5 is superimposed on the satellite image (dashed line).

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 3.
Fig. 3.

Atmospheric temperature profile obtained for the ascents of the Falcon during missions 4 and 5.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

i1520-0493-127-4-486-f4a

Fig. 4a. Mission 4 proceeded in three legs for the ARAT at an altitude of 4.6 km, three legs over the cirrus at altitudes of 11.2 and 11.5 km, and three legs inside the cirrus at different altitudes for the Falcon. All legs were flown along the same flight track. Legs of the ARAT are indicated by letter A, while the legs of the Falcon are indicated by letter F. Diamonds indicate the beginning of each leg. Cirrus boundaries as measured by lidar on board the ARAT are also indicated (dots), as well as the location of the POLDER data analysis (Δ).

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

i1520-0493-127-4-486-f4b

Fig. 4b. Same as Fig. 4a but for mission 5: three legs for the ARAT at a constant altitude of 4.6 km, two legs over the cirrus at 10.8 km, and six legs at different altitudes inside the cirrus for the Falcon.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 5.
Fig. 5.

Comparison between effective beam emittances derived from AVHRR measurements (channel 4) and PRT-5 on board the ARAT during leg A2.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 6.
Fig. 6.

Plane albedo of the earth–atmosphere–cirrus system obtained from pyranometers on board the Falcon during leg F3 of mission 4 and leg F1 of mission 5 as a function of the distance from CROSS Corsen.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

i1520-0493-127-4-486-f7a

Fig. 7a. Bidimensional values of the backscatter coefficient β(x, z) (in km−1 sr−1) for the three legs of mission 4 as measured by the lidar. The space variables x and z are the distance from CROSS Corsen and the altitude, respectively. Legs A1, A2, and A3 are plotted from top to bottom (see text). Color scales concern the values of β(x, z) between 0 and 0.056 km−1 sr−1. White parts of these color scales indicate the maximum values reached by the backscatter coefficient during the leg.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

i1520-0493-127-4-486-f7b

Fig. 7b. Same as Fig. 7a but for leg A2 of mission 5.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 8.
Fig. 8.

Scatterplot of infrared optical thickness δIR derived from the PRT-5 radiometer as a function of the lidar apparent optical thickness δa532. Only data corresponding to mission 4 are reported here. The slope of the least squares fit of data to a straight line is equal to unity.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

i1520-0493-127-4-486-f9a

Fig. 9a. Bidimensional values of the backscattering depolarization ratio Δ(x, z) (in %) obtained from the airborne lidar for the three legs of mission 4. The space variables x and z are the distance from CROSS Corsen and the altitude, respectively. From top to bottom are plotted legs A1, A2, and A3 (see text). Color scales concern the values of Δ(x, z) between 0% and 70%. The cirrus boundaries are indicated by a red dashed line; the black solid line indicates the altitude of the maximum backscattering coefficient.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

i1520-0493-127-4-486-f9b

Fig. 9b. Same as Fig. 9a but for leg A2 of mission 5.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 10.
Fig. 10.

Synoptic scheme of POLDER measurements principle. The solar zenith angle is θs. The particular viewing direction, characterized by the zenith viewing angle θυ and by the relative viewing azimuth angle φυφs, illuminates a particular pixel of the CCD matrix.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 11.
Fig. 11.

Total reflectance image [(a) and (b)] and polarized reflectance image [(c) and (d)] at 864 nm of wavelength of the cirrus bank overflown during leg F3 of mission 4 [(a) and (c)] and during leg F1 of mission 5 [(b) and (d)]. For (a) and (c), the solar zenith angle is θs = 55.3°, whereas for images (b) and (d) θs = 38.0°. The color scale is 35%–60% for total reflectances and 0%–2.5% for polarized reflectances. Each concentric circle represents a constant value of zenithal viewing angle θυ. The solar principal plane is indicated by the 0°–180° axis (0° corresponding to the solar direction). The azimuthal viewing angle φυ is set from this axis.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 12.
Fig. 12.

Images of cirrus banks, at 864 nm of wavelength, overflown during (a) leg F3 of mission 4 and (b) leg F1 of mission 5. These images are constructed using the central cross line of the CCD matrix [242 detectors corresponding to cross tracks of about (a) 3.6 km and (b) 3 km, respectively]. Locations within the legs of the two entire POLDER images of cirrus cloud total and polarized reflectances are indicated by arrows.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 13.
Fig. 13.

Global particle area distribution as determined by FSSP-100 and 2D-C probes during legs F4, F5, and F6 of mission 4 and legs F3, F4, and F5 of mission 5.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

i1520-0493-127-4-486-f14a

Fig. 14a. Histogram of the probability density of cloud boundaries altitude from lidar measurements corresponding to mission 4: (a1) cloud-top altitude probability density, and (a2) cloud-base altitude probability density.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

i1520-0493-127-4-486-f14b

Fig. 14b. Same as Fig. 14a but for mission 5.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 15.
Fig. 15.

Histogram of the probability density of backscatter-to-extinction ratio k obtained during the legs A1, A2, and A3 of mission 4.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 16.
Fig. 16.

Histogram of the probability density of backscattering depolarization ratio obtained during the legs A1, A2, and A3 of mission 4.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

i1520-0493-127-4-486-f17a

Fig. 17a. Histogram of the probability density of optical thickness δ532 given by the lidar for the three legs of mission 4.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

i1520-0493-127-4-486-f17b

Fig. 17b. Histogram of the probability density of infrared optical thickness δIR given by the infrared radiometer for the three legs of mission 4.

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Fig. 18.
Fig. 18.

Nadir reflectance of cirrus cloud measured by POLDER on board the Falcon (leg F3) as a function of infrared emissivity obtained from PRT-5 on board ARAT (leg A2). Radiative transfer simulation using Henyey–Greenstein scattering phase function with asymmetry factor ranging between 0.5 and 0.9 is reported on this figure (see text).

Citation: Monthly Weather Review 127, 4; 10.1175/1520-0493(1999)127<0486:RSOCRP>2.0.CO;2

Table 1.

Major field campaigns during the period 1987–95 as part of the two European programs: ICE and EUCREX.

Table 1.
Table 2.

Aircraft performance and instrumental payloads.

Table 2.
Table 3.

Summary of the various missions conducted during EUCREX’94. All flight tracks started or ended at the Cross Corsen site (an instrumented site). A, F, and M stand for ARAT (Fokker F-27, Falcon, and Merlin IV), respectively.

Table 3.
Table 4.

Cirrus radiative parameters and thermodynamical parameters measured by remote sensors and in situ probes on board the two aircraft (ARAT and Falcon).

Table 4.
Table 5.

Description of the legs of Falcon and ARAT for missions 4 and 5. Shown are times of beginning and end of measurements, corresponding distance from CC, and mean altitude.

Table 5.
Table 6.

Intercomparison of pyranometer measurements (mean value and standard deviation) during mission 1 (see Table 3), when the three aircraft flew at an altitude of 1.5 km.

Table 6.
Table 7.

Comparison of pyrgeometer measurements on board the ARAT with model results for clear air and maritime aerosols, at flight altitudes of 1.5 and 5.2 km (see text).

Table 7.
Table 8.

Characteristics of the two airborne POLDER instruments on board the Falcon and ARAT (Fokker 27), respectively.

Table 8.
Table 9.

Mean values of ice water content (IWC), particle concentration (N), and effective cross-section area of the particle size distribution obtained from in situ measurements of PMS FSSP and PMS OAP-2D2-C on board the Falcon during the legs of missions 4 and 5 inside the cirrus cloud.

Table 9.
Save