the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 26 Aug 2026
Global and systematic stratospheric aerosol number and size distributions from the NOAA Balloon Baseline Stratospheric Aerosol Profiles (B2SAP) project
Abstract. Stratospheric aerosol plays critical roles in the global radiative balance and stratospheric chemistry. A robust evaluation of stratospheric aerosol abundance and size distributions over time is required to determine and accurately model aerosol impacts on the Earth system. The Balloon Baseline Stratospheric Aerosol Profiles (B2SAP) project uses weather balloons to collect regular vertical profiles of stratospheric aerosol number and size distributions, at eight latitudinally diverse locations across the globe from the ground to ~30 km altitude. These systematic in situ measurements allow the micro-physical properties of the background stratospheric aerosol and their variability to be examined. These on-going measurements provide constraints for aerosol modules of large coupled models and reference for satellite retrievals relying on previous size distribution assumptions. The lightweight, low-powered Portable Optical Particle Spectrometer (POPS) instrument is deployed on balloon platforms, allowing for high vertical resolution and the detection of stratospheric aerosol down to 140 nm diameter. This paper presents the B2SAP aerosol dataset (DOI: https://doi.org/10.25921/mchy-y552, Smith et al., 2026), including a description of the B2SAP network, POPS measurement principles, sensitivity analysis, calibration procedures, file processing workflow and a comparison with aerosol extinction from the GloSSAC product.
- Preprint
(3446 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on essd-2026-644', Anonymous Referee #1, 06 Oct 2026
-
RC2: 'Comment on essd-2026-644', Anonymous Referee #2, 02 Nov 2026
This paper describes the instrumentation used for NOAA’s program to make regular stratospheric aerosol size distribution profile measurements using in situ optical particle counters. Such a description of this valuable data set is needed. Unfortunately, the presentation here leaves a lot to be desired.
The theoretical basis for the measurements is described, and an instrument counter response function is presented; however, the authors don’t provide any comparisons of that function to the calibration aerosol that is used, nor do they indicate what the signal pulses are. Amplitude is not a quantity. It is a characteristic of a quantity. Considering that each instrument is checked against calibration aerosol there should be plenty of such example comparisons of instrument response to calibration aerosol.
The authors also don’t show comparisons with any other in situ instrument to further establish the validity of their measurements, even though such measurements are available in the literature. Establishing POPS as a baseline measurement for future stratospheric measurements surely suggests some independent confirmation of its accuracy. They do make a comparison to the SAGE based GLoSSAC data base at two locations over a large time window. While this result is favorable, in the > 100 flights from Boulder there should be some that line up with SAGE III overpasses providing a direct head-to-head comparison which could provide more careful statistics, assuming the authors want to use a satellite measurement as a calibration reference. This would be opposite to the usual way of using a ground sourced and calibrated instrument to verify remote measurements.
In addition to these concerns there are a number of questions and confusions that need to be addressed, described in detail below. These relate to confusions about how the bin boundaries are set and illustrated in Figs. 3, 4, and 5, using “ticks” to measure time, a scattered and incomplete description of how particle flow rate is controlled, and the data links are broken.
Text set off with ellipses are quotes from the text, with occasionally a correction included.
168 Mie theory describes the relationship between size and scattering amplitude in all directions, not just the maximum amplitude, which is in the forward direction and is not the direction sampled by POPS, or indeed any OPC.
Fig. 3 What is indicated by the vertical red lines is unclear. The legend indicates these are bin edges, which usually means the minimum and maximum amplitude associated with a size bin, with the maximum forming the minimum of the next larger bin. But here the min/max amplitudes indicated by the red lines overlap with each other significantly for different sized particles. For example, at 250 nm the minimum bin edge has a value that is between the bin edges for the lower six or so size bins, down to particles near 200 nm? Then on Fig. 3b) what do the red lines represent. They span the entire ordinate axis.
This figure would also be significantly improved if it included measurements from the calibration aerosol. That would require another Mie curve at the index of refraction of the calibration aerosol, but would be of high interest for the reader.
189 Why unitless. The signal comes from a PMT so it has units of milli volts or milli amps, or something like that. It is an electrical signal. Such units should also be included in Figs. 3, 4, 5.
Fig.4 What is meant by Ticks? Isn’t the abscissa a time axis measuring the transit of a particle through the light beam? So units of msec, which would be much more interesting than “Ticks”.
211 Why isn’t Rayleigh scattering mentioned earlier as contributing to the BSD? Here is clear evidence that it contributes to the noise, as is common.
214 Don’t the authors mean the average BSD above 200 hPa is measured? How is it calculated, other than averaging the measurements? Where has “noise level” been defined? Where do these exact numbers come from? What is the relationship between these numbers and BSD at the surface?
Fig. 5 caption has problems. Figure captions should only define/describe the figure, they should not be used to discuss the figure, that is for the text. In this case the whole figure caption is discussing the figure. Such discussion should be merged into the text. Fig. 5 caption should be something like a) POPS amplitude versus diameter for instruments with varying BSDs. b), same as a) but for particles < 250 nm. But is a) necessary. It is just a repeat of Fig. 3 with a fan at the lowest sizes. b) Shows clearly how the fan goes away when diameters are > 200 nm. But the minimum POPS size is 140 nm, where these noise curves have little impact. Thus, what is the point of this figure? How does it inform the reader about measurements of aerosol > 140 nm?
Fig. 5 shows that an amplitude of 10 – 25 might indicate a particle diameter of 100 nm, depending on the noise level of 0 to 15. But this is misleading. The POPS minimum diameter is 140 nm, so why does the figure extend to 100 nm, and what is the equivalent spread at 140 nm? The figure would have to be significantly enlarged to see this spread, but even if one does, the rationale behind the tick marks on the abscissa is not obvious. There are two tick marks between 100 and 150. Are they at 120 and 140 or ?? Why isn’t normal log scaling with appropriate tick marks indicating each decade used? Again there should be units on the ordinate.
234-235 how about … to adjust the instrument gain so that the POPS response to this nebulized solution appears in the correct POPS size bin … What size bin is this? It must be different than the 500 nm size bin because of differences in index of refraction between calibration and stratospheric aerosol?
240 …at five different pump voltages …? Is this the same as five different flow rates? If so then say that, which is much clearer. The reader doesn’t know what five different voltages will do.
240-241 “… calibration curve is used to set the flow at different pressure levels during a POPS flight …” How is the flow adjusted at different pressure levels? Is there a telecommand option, or ?? What does this calibration curve look like?
Eq‘n C6 What is T_pops? It is on both sides of the equation, but is not defined. Where is it measured? Also just indicate that the temperature is in K. There is no need to show the conversion from Celsius to Kelvin. Are two equations really needed to define this simple correction. Why not combine C6/7 into one equation? And then explain why the flow rate needs to be corrected by the ratio of radiosonde to POPS temperature. Or better just include this relationship and the explanation in the description of the pumps and flow rate.
241-244 How is it that, at surface pressure, 180 ml/min is 3 ml/sec, which is obvious by a straight unts conversion, but at 30 hPa, 360 ml/min is also 3 ml/sec? Why isn’t it 6 ml/sec?
255-263 How is time measured in “ticks” and what does it represent? Is the full pulse shape as a function of time recorded or just the maximum pulse height and the time of the measurement?
The “maximum scattering amplitude” is the instruments fundamental measurement, so of course it makes no assumptions about particle composition. The assumption about particle composition only enters when the scattering amplitude is used to infer particle size.
“Flag out peaks” or just flag peaks? What are the units on the amplitude? Isn’t 53.7 the baseline plus one standard deviation so how can the software flag “peaks” below this value, and why should peaks > the maximum calibration range also be discarded? Couldn’t these latter be particles > 2500 nm, which is possible in the atmosphere? 1S – does this mean 1 Hz housekeeping data?
268-269… To ensure accurate time alignment between the radiosonde (GPS time) and POPS datasets (internal clock time), a variable common to both files (POPS flow rate) is used… It seems odd to suggest that POPS flow rate is common to both POPS and the radiosonde.
291-294 Why the redundancy here? … this volumetric sample flow is corrected to the ambient temperature (temperature-corrected flow) to get the correct sample volume (Eqs. C6, C7). The count of particles in each bin is divided by the temperature-corrected flow to calculate a number concentration per bin (cm−3). The volumetric sample flow is corrected using the ambient temperature to obtain the correct sample volume (Eq. C7)…
315-316 Isn’t this a requirement of all instrumentalists? And don’t they all follow such a rule? It just seems so obvious, why state it?
318-322 This is confusing and it is unclear why all the discussion/description of pump flow rate, measurement, and performance can’t be consolidated to one section, dealt with and then move on. In this manuscript the topic first appears at line 156 where … a laminar flow element (LFE) to accurately measure the volume of air sampled… is introduced, then line 224 where temperature correction is introduced, then 290-295 where a differential pressure sensor is mentioned, then 318-322 where a discussion of increasing the flow in steps depending on pressure is described. But nowhere is it explained why a constant volume flow rate pump is not used, nor why the pump performance degrades with pressure, and how to compensate for that. What do the higher voltages do? How does that change the pump operation? The authors list the manufacturer but fail to explain which of the manufacturers pumps they use.
Fig. 7 is this the usual way to show stratospheric aerosol concentration measurements, adjusted to STP rather than ambient concentrations or aerosol mixing ratios? Are these the units used in the 1S ICARRT data files? Why the 1S? Does this indicate it is 1 Hz data? The data links provided do not work. When data download is requested, the message returned is “Sorry, the page or resource you requested could not be found.”
340-343 That the index of refraction has a real and imaginary part is well known. An equation is not required.
340-420 Section 8, comprising 20% of a paper titled: Global and systematic stratospheric aerosol number and size distributions from the NOAA Balloon Baseline Stratospheric Aerosol Profiles (B2SAP) project” and submitted to Earth Systems Science Data, is out of place, off topic, and hence not reviewed.
Already the authors have failed to provide a clear and concise description of their valuable data set, now they want to further confuse the reader with speculation about how dependent the measurements are on the small fraction of stratospheric aerosol which are not sulfate. Such speculations are fine, but should be addressed elsewhere, treated much more carefully, with clear examples of what aerosol would lead to the indices of refraction used, and perhaps even some observations in the lower stratosphere, if the authors think they have sampled such aerosol. Such a discussion and figures are out of place In this paper and journal, as is appendix D.
Fig. 11 The GLoSSAC median should be much thicker so the reader can see it. There should be some discussion of the factor of 2 or more difference between POPS and GLoSSAC medians at Boulder. At least an acknowledgment of the difference. The Lauder comparison is much better. The figure caption should state that POPS flights also occurred at Lauder. Presently the caption indicates they only occurred at Boulder.
Appendix C equations. These simple and obvious equations are not necessary to include? The equation involving the temperature correction should be described in the pump/flow rate description, and isn’t the equation for extinction missing an r^2.
Appendix D That the sizing on the POPS bins is accurate to < 0.1 nm and needs 6 significant figures to describe these boundaries is unbelievable. Where is the evidence for that? And if that is not the case, why are they presented so? Just because that is what the calculations give? That is not sufficient reason. Are these the bin boundaries included in the ICARRT data?
…A scattering amplitude of 1000 was used in all scenarios to estimate the optical particle diameter (S1 b, d, f) to emphasize… Perhaps the authors understand what is meant by this statement, but their readers will not.
Aren’t the plots D1 b, d, f just expanded views of D1 a, c, e? What does an amplitude of 1000 have to do with the diagrams. It is still unclear how the vertical lines define the bin boundaries. The vertical lines span a large range of amplitudes from 1/3 to ½ of an amplitude decade scale. Such a range also can’t be ascribed to just one size. Rather it intersects the amplitude theoretical curve at many sizes. Doesn’t that lead to ambiguity about particle size, particularly when the amplitude within a bin boundary also appears in the bin boundaries of other sizes. Still the usefulness of this appendix for this paper to establish the B2SAP project as a baseline, like section 8, should be removed to another paper.
Minor comments
53 Begin a new paragraph with the aircraft measurement description.
63-63 This sentence needs a comma or a “which”
83 new paragraph
95 new paragraph
174 Perhaps include Rosen, 1964.
200 … particle’s peak …
201 jusr say: … and the the maximum scattering amplitude …
202 … noise. This difference is small for small particles with a maximum …
223 What is a PCB?
229 This sentence needs to be fixed, … it meets the baseline is steady …??
267 … this is also merged to the POPS operational and bins…?
278 (e.g. South Pole, AQ) What is AQ?
Citation: https://doi.org/10.5194/essd-2026-644-RC2
Data sets
Aerosol number and size distribution data (140 - 2500 nm) from the Balloon Baseline Stratospheric Aerosol Profiles (B2SAP) network K. Smith et al. https://doi.org/10.25921/mchy-y552
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 414 | 216 | 146 | 776 | 157 | 136 |
- HTML: 414
- PDF: 216
- XML: 146
- Total: 776
- BibTeX: 157
- EndNote: 136
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Please find my review in the PDF file