the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 Jan 2026
Insights into the low-level single-layer stratiform cloud optical depth feedback based on two decades of observations at the North Slope of Alaska
Abstract. A 21 year dataset of cloud optical depth (COD) over Utqiagvik, AK is retrieved from longwave (LW) Atmospheric Emitted Radiance Interferometer (AERI) observations. Cloud property changes with temperature, time, and other meteorological variables are investigated. We find that the COD of low-level, single-layer optically thin, stratiform, Arctic clouds at pressures greater than 680 hPa increases with warming for column averaged atmospheric temperatures between -16 °C and -4 °C, while the COD decreases with warming between -4 °C and 0 °C , with little change below -16 °C. This aligns with current literature. COD changes are driven by liquid water content (LWC) variations with temperature, with additional contributions from cloud physical thickness (CPT) changes and bulk cloud phase shifts. The COD of summer and winter clouds decreases with warming at all observed temperatures, while the COD of spring and autumn clouds increases with warming. We extend existing literature with a cloud controlling factor (CCF) analysis, finding that the atmospheric temperature at 850 hPa; the surface windspeed; and sea salt, sulfur dioxide and hydrophilic organic carbon concentrations are important controls on the COD. The length of the dataset allows us to perform novel trend investigation, with statistically significant positive trends found in the total (0.0824 year-1) and liquid (0.0719 year-1) CODs. Evidence is provided for a negative surface cloud feedback. Our results and associated dataset represent an avenue for the evaluation and improvement of model representations of Arctic cloud microphysical processes.
- Preprint
(1066 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-6553', Anonymous Referee #1, 12 Feb 2026
-
RC2: 'Comment on egusphere-2025-6553', Anonymous Referee #2, 21 Feb 2026
Review of egusphere-2025-6553: "Insights into the low-level single-layer stratiform cloud optical depth feedback based on two decades of observations at the North Slope of Alaska" by C. Coulbury et al.
This paper introduces a new data product comprising retrievals of cloud optical depth (COD) and effective radius (reff) of liquid water and ice components in mixed-phase clouds. The retrievals are based on longwave spectral zenith radiance measurements from the Atmospheric Emitted Radiance Interferometer (AERI) which has been continuously deployed at the US Department of Energy Atmospheric Radiation Measurement (ARM) Facility at the North Slope of Alaska (NSA) for 21 years. A rigorous retrieval algorithm based on high resolution radiative transfer simulations called MIXCRA (Mixed-phase Cloud Retrieval Algorithm) uses an optimal-estimation approach to simultaneously retrieve the COD and effective radii and their retrieval uncertainties. The manuscript first describes the AERI instrument and its operational history at the NSA site, and then provides a detailed description of the retrieval algorithm. Subsequent sections provide a brief climatological description of the retrieval results (3.1), determination of the cloud radiative effect (CRE) and an attempt to identify a cloud feedback in the context of Arctic amplification (3.2), a discussion of COD variability with temperature in a climatological context (3.3 and 3.4), variability in COD with non-cloud properties called cloud controlling factors (CCF) (3.5), and finally a detection of linear trends in COD over the entire time series (3.6).
This set of cloud retrievals, for single-layer optically thin clouds in the high Arctic, is a potentially valuable contribution and Atmospheric Chemistry and Physics is an appropriate journal for introducing this data product. The authors may want to consider contributing this data product as an ARM Value Added Product (VAP), though some differentiation would be required from the existing AERIOE VAP (which may only treat liquid water clouds). I believe these results can be published in ACP in some form along the lines of this manuscript's earlier sections.
However, the climate-change-related analysis, beginning in the latter half of Section 3.2 and extending through Section 3.5, has a conceptual flaw. These AERI-based retrievals are a subset of all clouds at the NSA site that is arbitrarily defined by limitations of the observing technique. Specifically, thin clouds with liquid water path in the approximate range 1-40 g m-2, corresponding to COD < ~5, will show spectral variability in their emitted zenith radiance that can be harvested by MIXCRA to retrieve COD and reff. Clouds with larger LWP will radiate as blackbodies in the middle infrared and cannot be explored with AERI except to measure an effective cloud base temperature in the mid-IR window. Furthermore, higher cloud layers must be filtered and these scenes excluded from the data product, because the retrieval algorithm cannot separate out emitted longwave contributions from the higher layers. The result is that this data product represents a small fraction of high Arctic clouds during summer, although possibly a larger fraction during winter and early spring. This limitation is compounded by what the authors describe as more frequent quality control filtering by the algorithm during winter (lines 299-301 and Figure 2). It is not meaningful to draw conclusions related to Arctic amplification or large-scale Arctic change from this limited data product, and such conclusions (e.g., Figures 3-7) might be misleading to readers who are theoreticians and modelers rather than observationalists.
However, this data product pertaining to thin single-layer clouds potentially has considerable value. Mixed-phase cloud microphysics is an immensely complex topic area and the clouds treated by these MIXCRA retrievals can serve as numerous examples, over 21 years and in all seasons, of the simplest and most straightforward Arctic clouds. It may be beneficial for researchers to examine these case studies to have a basis for trying to understand more complicated scenarios.
To that end, I suggest that this manuscript be re-worked to emphasize and expand on Section 3.1. Figure 1 and Table 1 are a good start. Additional questions might include: What are the altitude ranges of these clouds as a function of season? Are there some mid-level or higher layers that are distinct from the presumably more numerous low-level clouds? What is the variability in reff for liquid and ice? Also, what fraction of all NSA stratiform clouds is represented by this data product, in all seasons? It may be appropriate to present some basic linear trends as in Section 3.6, although the limitations of these trend detections should be discussed.
Next, instead of attempting generalizations involving CRE, CCF and cloud feedbacks, perhaps this paper could emphasize smaller but useful timescales. Specifically, the authors imply a high time resolution with a measurement cadence of several minutes. Are there any continuous segments lasting multiple hours or even a few days? If so, are there changes in the COD, reff, and phase partitioning that appear consistent with current conceptions of a "classical" Arctic mixed-phase cloud lifecycle (e.g., Morrison et al., 2012, Nature Geoscience; Sedlar et al., 2021, ACP; Jimenez et al., 2025, ACP)? Are there any summertime segments resembling case studies observed at other well-instrumented high Arctic locations such as ASCOS (e.g., Birch et al., 2012, ACP). Adding some other NSA measurements or VAPs to elucidate these segments would be helpful. An example or two might entice other investigators to further explore this new AERI data product.
Finally, this set of AERI retrievals should be archived and publicly available by the time of the manuscript's publication, if not as an ARM Facility VAP, then in any FAIR repository. Maybe NSF's Arctic Data Center, or any university library digital collection that provides DOIs for researchers' data sets would do. The link given in the manuscript (10.5281/zenodo.17833181) just takes me back to the manuscript.
In summary, if the manuscript is given a major revision to emphasize the retrieval algorithm and a general survey of the data product, with perhaps some time series case studies, it could be a well-cited contribution.
Citation: https://doi.org/10.5194/egusphere-2025-6553-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 200 | 84 | 17 | 301 | 13 | 17 |
- HTML: 200
- PDF: 84
- XML: 17
- Total: 301
- BibTeX: 13
- EndNote: 17
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
see attached PDF