Measurement errors in cirrus cloud microphysical properties H. Larsen, Jean-François Gayet, Guy Febvre, H. Chepfer, G. Brogniez To cite this version: H. Larsen, Jean-François Gayet, Guy Febvre, H. Chepfer, G. Brogniez. Measurement errors in cirrus cloud microphysical properties. Annales Geophysicae, European Geosciences Union, 1998, 16 (2), pp.266-276. hal-00329078 HAL Id: hal-00329078 https://hal.archives-ouvertes.fr/hal-00329078 Submitted on 1 Jan 1998 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Ann. Geophysicae 16, 266±276 (1998) Ó EGS ± Springer-Verlag 1998 Measurement errors in cirrus cloud microphysical properties H. Larsen1, J.-F. Gayet2, G. Febvre2, H. Chepfer3 and G. Brogniez3 1 National Institute for Water and Atmosphere, Wellington, New Zealand 2 Laboratoire de Me te rologie Physique, UPRESA/CNRS n°6016, Universite Blaise Pascal, Clermont-Ferrand, France 3 Laboratoire d'Optique Atmosphe rique, URA CNRS 713, Universite des Sciences et Technologies de Lille, France Received: 12 April 1996 / Revised: 13 June 1997 / Accepted: 21 July 1997 Abstract. The limited accuracy of current cloud micro- cloud microphysical properties (Stephens et al., 1990; physics sensors used in cirrus cloud studies imposes Heyms®eld, 1993). It is determined by how many cloud limitations on the use of the data to examine the cloud's particles there are, how much radiant energy they broadband radiative behaviour, an important element of intercept and emit, and how this energy is distributed the global energy balance. We review the limitations of in space and in wavelength. Modelling the cloud the instruments, PMS probes, most widely used for radiative behaviour requires a knowledge of the total measuring the microphysical structure of cirrus clouds water substance in the cloud, the number concentration and show the eect of these limitations on descriptions of cloud particles and their size distribution, their cross of the cloud radiative properties. The analysis is applied section for the interception of radiation (related to their to measurements made as part of the European projected geometric cross section and hence their Cloud and Radiation Experiment (EUCREX) to deter- shape and size) and their scattering phase functions mine mid-latitude cirrus microphysical and radiative (also determined by the particle shape, and its phase, properties. water or ice). However there have been only a limited number of Key words Atmospheric composition and structure detailed observations of the cirrus cloud microphysical (cloud physics and chemistry) á Meteorology and atmo- properties needed in modelling the interaction of the spheric dynamics á Radiative processes á Instruments cirrus with radiation. What measurements there are and techniques indicate there can be a wide range of ice water content, cloud depth, and ice crystal shapes, sizes and concen- tration, making these clouds dicult to parametrise (Stephens et al., 1990; Ramanathan et al., 1995). They are thus poorly represented in current climate models. 1 Introduction The measurements have tended to be in mid-latitude non-convective clouds formed by large-scale ascent. The understanding and modelling of climate processes is Additional in situ measurements of cirrus cloud micro- a critical global issue, and problems involving cirrus physical properties are needed, particularly of contrail- clouds are amongst the most immediate. Cirrus clouds induced cirrus, tropical cirrus, Southern Hemisphere are important determinants of the climate, modulating cirrus and cirrus formed by convective processes, to both the input of energy to the climate system and the improve the modelling of cirrus in climate processes. At output of energy from it (e.g. Ramanathan et al., 1983, the same time there are limitations on the measurements 1989) over a large area of the globe at any one time that are currently available or that can be made with (Bretherton and Suomi, 1983). Their impact on the currently available instruments. We will show here that global radiation balance and their consequent involve- this means these measurements could mislead the ment in feedback processes will be determined by their modelling work, if not applied with great care. An areal coverage, their temperature and by the way they additional concern is that these measurements, being interact with both incoming short wave (solar) and made in situ, may be treated as `absolute' measurements outgoing long wave (terrestrial) radiation. This broad- and used to calibrate remote sensing instruments with- band radiative behaviour is in turn determined by the out regard for their limitations. We ®rst describe the measurement probes and their limitations, then consider the radiative implications. Correspondence to: J.-F. Gayet There are radiative implications of the limitations on the H. Larsen et al.: Measurement errors in cirrus cloud microphysical properties 267 measurement of particle shape, on the detection of small in cloud physics instrumentation, inherent probe and particles and on the determination of integrated mea- data analysis shortcomings limit the conditions under sures of cloud water. These implications are demon- which the data can be relied on. There are four problems strated through analysis of measurements made as part in particular, related to: (1) the determination of the of the ®eld phase observations within the European absolute particle number concentration, (2) the identi- Cloud and Radiation Experiment (EUCREX). These ®cation of particle shape, (3) the limited range of sizes of observations were made in several areas in Western particles over which the measurements can be made and Europe to determine mid-latitude cirrus microphysical (4) the limited sample volume of the probes. and radiative properties (Raschke, 1996). The ®rst problem, in determining the absolute number concentration, comes from three sources. The major one relates to the data handling in dierent 2 PMS probes probes. Figure 1 shows the size spectra obtained simul- taneously from three 2D-C probes mounted close Most available in situ cirrus microphysics measurements together on the same aircraft during a ¯ight in have been made using aircraft-borne optical imaging or cirrostratus at a temperature of around )35 °C (Gayet scattering probes, commercially available PMS 2D-C et al., 1993). The probes show large dierences in and PMS FSSP-100 respectively (Knollenberg, 1976). particle concentration for the larger sizes, even although There are a limited number of measurements on ice measuring closely spaced volumes of the same cloud. crystals captured on replicators (e.g. Cooper and Vali, The dierences appear to be due to instrumentation 1981; Heyms®eld, 1986) and new instruments are now problems rather than true cloud dierences (Gayet et al., coming into use: the counter¯ow virtual impactor or 1993), and are related particularly to the probe version CVI (StroÈ m and Heintzenberg, 1994), an advanced being used, with the updated 2D2-C version apparently replicator (Strauss et al., 1995), holographic cloud detecting some 50% more images than the original particle imaging systems (Brown, 1989; Lawson, 1995), version 2D-C. The second source of concentration an evaporative technique for total water measurements measurement problems arises through the poor de®ni- (Nicholls et al., 1990), the video ice particle sampler tion of the sampling volume, associated with response (McFarquhar and Heyms®eld, 1996) and a polar time problems. This would be expected to aect nephelometer (Gayet et al., 1997; Cre pel et al., 1997). primarily the smaller size channels (25±100 lm), causing There are also less widely used PMS probes (FSSP-300, a greater degree of undercounting of small particles than 260X-1D-C, 2D-Grey, 2D2-P) which can be used to larger particles, as shown theoretically by Baumgardner extend the data available and so overcome some of the et al. (1986) and Baumgardner (1987). These authors limitations of the other probes. For the moment the proposed airspeed corrections on size and sample alternatives oer a check on the PMS 2D-C and PMS volume which should adequately account for electronic FSSP-100 data rather than a useful database on cirrus response-time problems, but these corrections are often microphysics in their own right. not applied, the investigators being unfamiliar with the During EUCREX some CVI, advanced replicator, probes and their limitations. The third concentration holographic and total water measurements were avail- measurement problem arises in the software used for able but the vast bulk of the data are from the PMS 2D- C and PMS FSSP-100. Thus, users of EUCREX data in particular should realise that unless otherwise indicated the data has the limitations described. 2.1 PMS 2D-C probe The data from these probes are recorded as two dimensional images of individual cloud particles, some- times several hundred per second, passing through the sampling volume of the instrument. Typically the probe covers a particle size range of 25±800 lm with a pixel size of 25 lm. The analysis techniques use speci®c algorithms applied to the image of each particle to identify that particle's shape and size (Heyms®eld and Parrish, 1978; Duroure, 1982; Heyms®eld and Baum- gardner, 1985; Duroure et al., 1994; Fouilloux et al., 1997) and to determine particle concentration. The accuracy and consistency of the 2D-C probes has been examined (Gayet et al., 1993) as part of the ICE/ Fig. 1. Particle-size spectra obtained from simultaneous measure- EUCREX programme to determine cirrus microphysi- ments by three dierent PMS 2D-C probes for a 2-mn segment in a cal properties. Although development and use of the cirrostratus cloud.