the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
17 Mar 2021
17 Mar 2021
A differential emissivity imaging technique for measuring hydrometeor mass and type
- 1Department of Mechanical Engineering, University of Utah, Salt Lake City, UT, USA
- 2Department of Atmospheric Sciences, University of Utah, Salt Lake City, UT, USA
- 1Department of Mechanical Engineering, University of Utah, Salt Lake City, UT, USA
- 2Department of Atmospheric Sciences, University of Utah, Salt Lake City, UT, USA
Abstract. The Differential Emissivity Imaging Disdrometer (DEID) is a new evaporation-based optical and thermal instrument designed to measure the mass, size, density, and type of individual hydrometeors and their bulk properties. Hydrometeor spatial dimensions are measured on a heated metal plate using an infrared camera by exploiting the much higher thermal emissivity of water compared with metal. As a melted hydrometeor evaporates, its mass can be directly related to the loss of heat from the hotplate assuming energy conservation across the hydrometeor. The heat-loss required to evaporate a hydrometeor is found to be independent of environmental conditions including ambient wind velocity, moisture level, and temperature. The difference in heat loss for snow versus rain for a given mass offers a method for discriminating precipitation phase. The DEID measures hydrometeors at sampling frequencies up to 1 Hz with masses and effective diameters greater than 1 µg and 200 µm, respectively, determined by the size of the hotplate and the thermal camera specifications. Measurable snow water equivalent (SWE) precipitation rates range from 0.001 to 200 mm h−1, as validated against a standard weighing bucket. Preliminary field-experiment measurements of snow and rain from the winters of 2019 and 2020 provided continuous automated measurements of precipitation rate, snow density, and visibility. Measured hydrometeor size distributions agree well with canonical results described in the literature.
Dhiraj K. Singh et al.
Status: final response (author comments only)
-
RC1: 'Comment on amt-2021-44', Anonymous Referee #1, 12 Apr 2021
As far as I know, the authors are correct in stating that the novel approach for extracting mass and density information from individual snowflakes has never been tried before, at least not published in the open literature. The technique that is used is simple in its concept, although there are issues that need to be addressed before it is ready to be used by the broader scientific community.
I recommend publication after a number of major issues are addressed, not least the haphazard placement of the figures that many times are found in the text well before they are discussed. Whether this was an error by the copy editor or carelessness by the authors, it was a major source of frustration as I tried to review the manuscript and a source of uncessary distraction . Had the topic not been so compelling, I would have rejected this paper early on. I was not one of the first-cut reviewers, but I would have not allowed this to go into the discusson phase in the current form.
1) There are a number of shortcoming that need to be addressed before this paper can be published. The most significant being the lack of a comprehensive error analysis the documents the source of systematic and random erros and then propogates these into the derived quantities that are being highlighted, i.e., equvalent diameter, particle complexity, density, mass, visibility, SWE, etc. There are many potential sources of uncertainty that were mentioed but no quantitative estimates given. This is unacceptable for an instrumentation paper. One of the uncertainties that is given very short shrift concerns the probability that two more more snowflakes will be imaged together, not because they are aggregating when they fall but because one fell one top of the other. A very brief comment is made that under one condition, out of a 1000 images, ony 5 wewre touching. Figure 7 belies that statement since there are many fewer than 1000 particles and I count more than 10 that are touching. Given the long times needed to evaporate ice crystals (see my next ennumerated issue), 30-60 seconds, under even modest precipitation rates the probability must be moderately high that as one crystal melts/evaporates, another will fall on top of it. This situation is not addressed but a very simple calculation needs to be made, similar to what is done with other optical spectrometers, to estimate the coincidence probability for different size distributions and precipitation rates.
2) One of the most critical parameters in all of the equations to predict density and mass, is the time to completely evaporate a crystal; and yet only a single figure (Fig. 5) shows this parameter for a single water droplet. I would like to see some actual Size vs time for ice crystals in field experiments so as to illutrates the variability with size, mass and density. These times also help determine the frame rates and probability of coincidence, so a lot more needs to be discussed about their importance for deriving the parameters that are being advertised as available from this instrument.
3) The camera frame rates that are mentioned vary quite a bit, from 5-240. It appears that the higher frame rates were used just to validate certain aspects about detection and melting rates, but operationally much lower rates are used. Why? This raises a very important issue that is not addressed: "What is the processing time?". With 1.2 Mpixels to process from each frame, how long does it take to identify and accept/reject each particle in a frame, what are the filtering criiteria and has fast can all the derived parameters be output? Is this near-realtime or does this require substantial pot-processing time so that the applications can only be for research and not for operational applications?
4) How do you avoid measuring snow lifted from nearby surfaces, i.e. how do you know that you are measuring freefalling snowflakes?
5) Can you measure graupel or snow pellets that bounce?
6) Snowflakes form on aerosols and scavenge them, as well. These will remain as residue after the crystal melts. What is the impact on the measurements and how does this issue get addressed? How about issues of condensation on optical surface/components of the camera? Turbulent flow around the camera will likely deposit blowing snow on camera surface.
A final note: please review all the references to make sure that they are the original one and not just quoted like you did by using Rogers and Yao to reference parameterization of fall velocity. They published a reference textbook but the equations were developed by others. Please respect the original works and cite accordingly,
- AC1: 'Reply on RC1', Eric Pardyjak, 14 Jun 2021
-
RC2: 'Comment on amt-2021-44', Aaron Kennedy, 18 May 2021
The authors present a novel new instrument for measuring properties of liquid and ice precipitation. Overall, I think this is a worthwhile contribution to the field, but the paper needs clarifications in a few places. Further, the paper would benefit from comparisons to other instrumentation for field components of the paper. It is unclear whether these observations exist or not.
- Is there an operating range for the instrument? What is energy consumption as a function of ambient temperature? This would shed light on the requirements for deployments in different areas.
- What are the impacts of wind flow around the instrument for snow events? Since this is mentioned as a drawback of other instruments, should address. It seems like provided wind tunnel tests were focused on thermal effects of wind (this is good!). This should be pointed out at L295, otherwise reader could interpret this statement too far. The statement at L350+ about minimal interference with the camera seems like a stretch, but that’s coming from someone that lives in a windy environment. I agree that a flat plate implies it is better than other platforms.
- My biggest concern with the paper is the missed opportunity to compare DEID to other common measurements. Take visibility for example… was there a forward scattering sensor on site? Provided that the instrument is sensitive to hydrometeors > 200μm, I presume there could be bias. What about MASC data? Would be great to see PDFs of select variables between the two systems. Statements like Line 310-312 could be backed up with MASC images.
- Can you explain the logic between sampling rates? Why was 12 Hz decided upon for field work? Precipitation rate is mentioned, but is this determined on the fly by input from other instruments? It’s unclear how the range of 2-30Hz is related to the rates quoted in Section 2.2 that mentions tests up to 120 fps / 240 Hz.
- What are the computing requirements like? It was unclear whether the imagery is processed in real-time or not, and if so, what type of resources are needed.
- Figure 12: Is there any significance to the width of the heavy/light snow columns? It seems like these could be broadened.
Minor comments:
Line 25: Might be worth highlighting the challenges of other instruments for snow/wind? See associated references that could be added to this section.
Parsivel:
Battaglia, A., Rustemeier, E., Tokay, A., Blahak, U., & Simmer, C. (2010). PARSIVEL Snow Observations: A Critical Assessment. Journal of Atmospheric and Oceanic Technology, 27(2), 333–344. https://doi.org/10.1175/2009JTECHA1332.1
Loeb, N. and A. Kennedy, 2021: Blowing Snow at McMurdo Station, Antarctica During the AWARE Field Campaign: Surface and Ceilometer Observations. J. Geophys. Res. Atmos., 126, e2020JD033935.
MASC:
Fitch, K. E., Hang, C., Talaei, A., and Garrett, T. J., 2020: Arctic observations and numerical simulations of surface wind effects on Multi-Angle Snowflake Camera measurements, Atmos. Meas. Tech., 14, 1127–1142.
Figure 4 caption: water droplets or water and ice droplets?
L230: Is there an extra – before Collins?
L235: Please clarify- what is a sample referring to if there were 2000 snowflakes or rain drops contained within? I think this is answered at L358… just make sure this is clarified earlier on.
- AC2: 'Reply on RC2', Eric Pardyjak, 14 Jun 2021
Dhiraj K. Singh et al.
Dhiraj K. Singh et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 307 | 109 | 12 | 428 | 5 | 5 |
- HTML: 307
- PDF: 109
- XML: 12
- Total: 428
- BibTeX: 5
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1