Simultaneous Measurement of Mass, Size and Fallspeed of Individual Solid Hydrometeors using Electronic Balance and Video Camera

The quantitative precipitation estimation (QPE) from the size and fallspeed simultaneous distribution obtained by an optical disdrometer such as 2DVD, PARSIVEL and so on, needs the estimated mass of individual hydrometeor with its observed size and fallspeed. Locatelli and Hobbs (1974) simultaneously measured mass, size and fallspeed of individual particles with detailed microscopic observations for wide range of solid hydrometeor. Their results has been used for many application includes QPE using optical disdrometers. For example, Ishizaka et al. (2013) constructed the parameterization of the mass flux contribution of individual particles as a function of their size and fallspeed with a interpolation using the mass and size relation from Locatelli and Hobbs (1974) and Ishizaka (1995). However, we met some problems when we utilize the results from Locatelli and Hobbs to QPE from our observation using disdrometer made in moderately warm snowy land in Japan. First, the maximum size of particle in their measurements (~3.2mm for graupels, ~12mm for aggregates) were comparatively too small for precipitation events with large graupels or large aggregates. For aggregates, Ishizaka (1995) measured the relationships of mass, size and fallspeed for wide range of aggregates include large particle. Second, their measurements lacked the results for wet hydrometeors which frequently occur in such an area. In this study, to construct the empirical parameterization of mass as a function of size and fallspeed on the basis of individually measured mass applicable to the wide range of hydrometeor types including large or wet hydrometeors, we set up the automatic data collection system for the simultaneous measurements of mass, size and fallspeed of individual solid hydrometeors. By operating this system during two winters, we were able to measure the mass of particles more than 2,000. The observation system were installed in the low temperature room of Falling Snow Observatory at Snow and Ice Research Center, National Institute for Earth Science and Disaster Prevention, Niigata, Japan. The natural falling hydrometeors can be taken in the low temperature room kept at -5 degrees Celsius through the openable roof window. Fig.1 shows the schematic picture of the measurement system. The measurement system is composed of an electronic balance, a video camera and a photography assisting tunnel with light sink in the back for obtaining dark field images of particles. In the middle part of the tunnel, there is a vertical slit for the passage of falling hydrometeors with two strong LED sidelight and an particle intake aperture (4cm x 4cm) upside the tunnel. The hydrometeors passing the slit are successively photographed by the video camera (GC-PX1, JVC) with frame rate of 60(fps) and with shutter speed of 1/4000s and are measured its shape of side view and fall speed by image processing of successive images. Finally, the hydrometeors land onto the shale put on the weighing dish of the electronic balance installed just under the aperture. The electronic balance (AD-4212C, A&D Co., Ltd.) has the minimum weighing value of 0.1mg and fast response time ~1.3sec. The mass of the hydrometeor is able to be measured by the difference of the weight value before and after the land of it. The matching of the weighing data and the particle data were made by manual analysis displaying the time series plot of weighing data and the animation of analyzed particle shape for the same observation time on PC monitor. Fig. 2 shows preliminary result of the measured mass of hydrometeors with simultaneously observed size and fallspeed in 2011/2012 and 2012/2013 winters. In this measurements, we obtained many data not only for dry hydrometeors but also for wet aggregates and wet graupels which were not found in prior researches. Moreover, we had many data in the range of 2mm-10mm in size and 1mg-5mg in mass to construct the empirical parameterization with the estimation of error from its variation. The lower limit of measured mass was about 1mg, because the signal of the particle landing on the weighing time series were indistinct if the mass is less than 1mg. It is the limitation of this measurement system. It also should be noted that we did not made microscopic observation and did not discriminate the type of each hydrometeor for this measurement. It is not so important to discriminate detailed types of hydrometeors as far as we utilize these data to QPE from disdrometer data, though it is not the case for the cloud physics study. For the future issue, we should construct empirical parameterization of mass and estimated density from mass and shape of side view of particle. And the investigation of the relationship of shape and fallspeed with mass is also important issue.