the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Aug 2025
Estimating near-surface specific humidity over the ocean
Abstract. The surface latent heat flux is a large term in the surface energy balance and difficult to estimate remotely. The main difficulty for its estimation remotely is a poor ability to measure near-surface humidity. Current methods to retrieve near-surface specific humidity approach the problem statistically and have errors of approximately 1 g kg-1 even in global-annual averages. Using extensive measurements from the EUREC4A field campaign (ElUcidating the RolE of Clouds, Circulation Coupling in Climate), we demonstrate that remote-sensing measurements of cloud base height can provide useful estimates of near-surface humidity. Applying the method to 171 coincident radiosonde and ceilometer pairings collected from a research vessel yields skillful predictions of near-surface specific humidity regarding the mean (mean bias 0.33 g kg-1 compared to observed) and its variability (r = 0.76). We next apply this method using an airborne lidar to estimate cloud base height from above. In two case studies, we find similar skill in the predicted humidity, with low mean biases (-0.06 and -0.03 g kg-1 compared to observed) with substantial variability captured (r=0.61 and r=0.57, respectively). Two main error sources, (i) the relative humidity lapse rate below cloud base and (ii) the temperature difference between the sea surface and near-surface air, are identified and quantified. Our proposed approach allows for estimates of the near-surface specific humidity using downward-staring space-borne lidar. This proof of concept raises the potential for its global application and for improved observational constraints on the surface energy budget.
- Preprint
(8399 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 27 Sep 2025)
-
RC1: 'Comment on egusphere-2025-3551', Anonymous Referee #1, 28 Aug 2025
reply
This paper outlines a relatively simple method of retrieving near-surface humidity as a function of cloud-base height assuming an adiabatic and well-mixed layer below the clouds. The theorized relationship makes sense and a decent amount of data supports the authors’ points. However, there are some fundamental issues here which need to be addressed before this paper can be published. Therefore, I am recommending major revisions as well as a transfer to a different journal.
Fundamentally, this paper assumes that the observed cloud base height can be used to infer the surface and near-surface relative humidity through simple adiabatic thermodynamics. This works if the boundary layer is well-mixed and the clouds are convective in nature as the clouds are coupled to the PBL. For the EUREC4A campaign location of Barbados, those conditions may very well dominate. However, this paper casts a much larger net than the tropics. The title and abstract make no restrictions on the time and place where this approach can be applied. Given that stratocumulus are by far the most common marine cloud type, and the formation of these clouds is largely independent of PBL mixing, the utility of the proposed method is much more limited than what the paper implies. It still may have utility (notwithstanding further critiques below) but the locations where its use is appropriate need to be identified from the very start. The authors partially mitigate this with a statement around Line 278, but even then they are focused on areas of convection and not the more prevalent stratiform clouds. In Line 259 they note that error increases in shallow, poorly mixed layers. That’s much of the globe.
The thesis of the entire paper rests on a single sentence, found in line 63: “The height at which clouds begin to form – the cloud base – is a reliable indicator of the near-surface relative humidity.” This is a rather foundational statement that the authors do not support with any references. The authors note that this is effectively used as a rule of thumb, but many such rules are unsupported by scientific evidence, and it is important to verify the validity of those rules when used in a scientific application. Thus, the authors are obligated to support the crux of their argument with evidence, but it is largely lacking here. The causality between cloud base height (for all clouds, mind you, since there is no effort to specifically segregate convective clouds from others here) must be identified and supported if the authors are going to make such a statement. I realize that Fig. 2 is an attempt to empirically show this relationship, but again this will only be true for convective clouds. Even at that, the relationships shown in Fig. 2 are more tenuous than one might hope: the BCO radiosondes in Fig. 2 exhibit an r2 of approximately 60%. That’s quite a lot of variance that cannot be explained by a simple relationship between the two variables, and this will introduce a substantial amount of uncertainty in the final product. It’s not clear with that much variability that cloud base is truly a “reliable indicator of near-surface relative humidity.”
The mathematics of the paper rely on the claim that the surface air is saturated. This is a statement that left me scratching my head as I read it. Obviously on a bulk level this statement is false: there is no permanent fog layer across each body of water on Earth. But even on a molecular level, is this true? At the ocean skin, is the number of evaporating H2O molecules really always equally balanced by the number of condensing ones? I could be mistaken on this, but it is incumbent upon the authors to justify this statement. Because of this, it calls into question the rest of the analysis that flows from that assumption.
In general, the uncertainty analysis of this paper is underdeveloped. There is instrument error, both from the surface observations of temperature and pressure as well as the lidar observations of cloud base height. There’s uncertainty in the analysis framework, including the assumption that surface q is saturated, that the boundary layer is well mixed, etc. A monte carlo estimation of the uncertainty is probably the most straightforward approach here: by randomly perturbing the h, Ts and other variables; assuming that qs is some fraction of q*; etc., reasonable error bars on the final product can be calculated. As it stands, the value of the product is limited because it is not clear just how trustworthy it is. While the validation relative to some collocated observations is reasonable, that is only for a very specific location and environment. Additional analysis could help greatly inform the utility of this method outside of those regions.
I also found the order of the paper somewhat difficult to follow. Fig 2 shows the bulk of the work (a relationship between RH and h) that much of the rest of the paper is trying to justify. Fig 4 shows some outliners from Fig 3, but how those outliers are influencing the final q analysis isn’t clear. The theory doesn’t appear until Sec. 5 when it really should be part of the methodology. The authors should reconsider the odering of the work and optimizing it for a simple flow for the readers.
Other significant points:
Regardless of the points raised above, the title of the paper must be changed. As it stands, it sounds like a review of all methods of measuring near-surface specific humidity. A more appropriate title would be “Estimating near-surface specific humidity over the ocean in convective environments from cloud base heights” or something along those lines.
I also do not believe this paper falls within the scope of this particular journal. To me, this is much more appropriate as an Atmospheric Measurements Techniques paper as it is directly focused on a measurement technique instead of a more fundamental physical process study. I strongly recommend the authors and editor consider a transfer to this sister journal as its scope is far better suited for this particular paper. Some publishers allow reviewers to select "transfer" as an option alongside the standard ones of "reject," "minor revisions," etc. If that option were enabled here at ACP, I would be clicking that box.
Minor points:
I am not convinced that using the gamma notation over the far more common d/dZ notation improves readability. Since we are dealing with thermodynamic properties, when I see gamma my mind instantly assumes that it is referring to R/cp which made it more challenging to follow the mathematical analysis.
Figure labels have a space between g and kg-1.
Citation: https://doi.org/10.5194/egusphere-2025-3551-RC1
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 597 | 40 | 13 | 650 | 13 | 17 |
- HTML: 597
- PDF: 40
- XML: 13
- Total: 650
- BibTeX: 13
- EndNote: 17
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1