the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 20 Sep 2023
Comparison between ground-based remote sensing observations and NWP model profiles in complex topography: the Meiringen campaign
Abstract. Thermally driven valley winds and near-surface air temperature inversions are common over complex topography and have a significant impact on the mesoscale weather situation. They both affect the dynamics of air masses and pollutant concentration. Valley winds affect it by favoring exchange between the boundary layer and the free troposphere, and temperature inversion by concentrating pollutants in cold stable surface layers. The complex interactions that lead to the observed weather patterns are challenging for Numerical Weather Prediction (NWP) models. To study the performance of the COSMO-1 model anaylsis (KENDA-1), a measurement campaign took place from October 2021 to August 2022 in the 1.5 km wide Swiss Alpine valley called Haslital. A Microwave Radiometer and a DopplerWind Lidar were installed at Meiringen, in addition to a multitude of automatic ground measurement stations observing meteorologic surface variables. Near the measurement’s sites, a low altitude pass, the Brünig Pass, influence the wind dynamic similarly to a tributary. The collected data shows frequent nighttime temperature inversions for all months under study, which persist during daytime in colder months. An extended thermal wind system was also observed during the campaign, except in December and January allowing an extented analysis of along and cross valley winds. The comparison between the observations and the KENDA-1 data provides good model performances for monthly temperature and wind climatologies but frequent and important differences for particular cases, especially in case of foehn events. Modeled nighttime ground temperature overestimations are common due to missed temperature inversions resulting in bias up to 9 °C. Concerning the valley wind system, modeled flows are similar to the observations in their extent and strength, but suffer from a to early morning transition time towards up valley winds. The findings of the present study allow to better understand the temperature distributions, the thermally driven wind system in a medium size valley, the interactions with tributary valley flows, as well as the performances and limitations of a model in such complex topography.
- Preprint
(14380 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2023-1961', Anonymous Referee #1, 11 Oct 2023
General comments
The manuscript describes a 10-month-long experimental campaign in a Swiss Alpine valley based on a complete set of atmospheric remote sensing techniques and meteorological in-situ automatic instruments. The results are discussed with reference to the complex orography of the area and in comparison with a high-resolution NWP model (KENDA-1). The main focus of the measurement-model comparison is to determine if the observed recurring weather patterns (temperature and wind profiles) are well reproduced by KENDA-1, both on average and in selected cases. The study represents an interesting glimpse of the challenges of measurements and simulations over complex orography, of which the chosen site is a great example, and the strengths and weaknesses of remote sensing and NWP models. These aspects make the study relevant to the journal and worth being published.
While the research framework is well-designed and current literature well referenced, the manuscript is apparently written too quickly (missing references, typos, erroneous labels in figures, etc.) and has not yet reached the minimum quality standard for publication. Notably, the manuscript is very long (53 pages, too many considering that it is not a review paper) and more synthesis is required. Indeed, there is a lot of interesting scientific material, but too much detail is to the detriment of the overall focus of the paper and the reader's attention.
I suggest that the authors revise the form of the manuscript and try to shorten it. The (10!) appendices could be transformed into a Supplement, where also a part of the text could be moved. More suggestions are provided hereafter.
Specific comments
- Title: temperature and wind could be cited in the title, as they represent useful keywords
- Introduction: multiple sub-sections in the introduction are a bit unusual. I would suggest that the authors shorten this section and limit the information to the most relevant topics for the paper
- Sect. 1.2: here the authors should better highlight that the analysis of the model "performances" does not focus on the forecast skills, but on the ability by the model to represent the average general patterns. I also suggest that they replace the word "climatology" (which would need a much longer period) with "statistical analysis" or "average weather patterns" or a similar expression
- Sect. 2: the authors could move the description of the site (now Sect. 2.2) and the weather situation (now Sect. 2.4) at the beginning to allow the reader to better follow the description of the instruments. I would suggest that the numerical (now Sect. 2.1) and experimental (Sect. 2.3) tools could be brought closer (as 3rd and 4th subsections).
- Sect. 2.2, l. 134: if these are important geographical features, please include a figure with these references
- l. 149-153: is the altitude bias depicted in Fig. 2 only due to inclusion of the slopes in the model cells (as written at l. 149-150) or is this bias due to different DEMs (KENDA-1 and the 25 m elevation model)? In the first case, I cannot explain why the average bias in some cells is far from zero. As a further note, this paragraph could be moved in the model description
- Fig. 1: a map over a wider area, such as the one in Fig. B1 should be reported in Fig. 1 (e.g., as a subfigure) to help the reader better understand the geography of the valley
- Sect. 2.3.3: can you shortly explain what the MWR "training" implies?
- Sect. 2.3.4: is the vertical component of the wind velocity used anywhere in the study? Also, there is no mention to a blind (low) zone in the DWL measurements, however the DWL plots start higher than ground level. For the same reason, I guess that the T inversion is not detectable from the wind fields (DWL)? Is it possible to detect any turbulent mixing phenomenon (e.g., development of PBL) at the bottom of the valley from DWL, overlapping to the slope/valley circulation?
- Sect. 3: it may be a matter of taste, but as a reader I would be more comfortable if the results were split between "Climatology" and "Case studies" rather than "Temperature" and "Wind" (each one with analyses on both the long- and short-term). That would first provide an overview on how the model performs, then the focus could be on specific episodes
- Fig. 3: what is the reason for the "erosion" in the temperature field below about 1000 m in May and June at the end of the day?
- l. 259: it is stated that the environmental lapse rate correction is "not precise in specific cases". However, it is not even precise on average. More generally, I would remove the whole paragraph at lines 251-264 and just mention that tests using a fixed lapse rate determined that horizontal/vertical distances between the station and the KENDA-1 cell are not the reason of the observed T discrepancies
- l. 264-265: do these differences present a seasonal cycle? Is there a figure similar to Fig. 5 for each month?
- l. 304: could the difference in the frequencies of T inversion between MWR and ground station be an effect of the surface, i.e. due to the fact that DWL measurements are in the free atmosphere and the station is on the ground?
- l. 309: in March and April, the differences between MWR and KENDA-1 are not "just above" the T inversion
- Sect. 3.1.5: this section is very short and maybe not too relevant. Can it be removed?
- Sects. 3.2.1, l. 341-342: this classification method sounds a bit naive, and some similarity between the w<20 km/h and w>20 km/h diagrams (Fig. 8) are visible, i.e. no clear boundaries are found between synoptic and thermal circulations. Could the authors further elaborate on that?
- l. 396-397: is the difference between 3.5 and 4 hours significant (and relevant)? Also, is the +/-1h offset described at l. 399-400 significant?
- l. 453: the wind is defined as "Up valley", but has E/SE direction
- l. 457: "only observed between 1300 and 2000 m", but it looks like there is a positive along-valley at 800 m in Fig. 10
- l. 464-470: a "3D" figure with the winds depicted as arrows would be very beneficial for the readers not familiar with the Swiss geography
- l. 478: "from NE", this seems to contradict Fig. 8b ("green colour" of the wind direction during the night)
- l. 578-584: is the presence of a lake really discussed in the study? Can you better explain why the model would not be able to deal with a lake?
- l. 596: "equally common", is this the case? The last sentence of Sect. 4.2.1 lets me think the opposite
- l. 602: is the "thickness" of the T inversion analysed here?
- Sect. 4.3: can qualitative concepts such as "accurate results" (l. 675), "large modeling errors" (l. 677), "satisfactory" (l. 686), "very good performance" and "well modeled" (l. 722), etc. be quantified in a more precise way, i.e. based on some performance targets?
- Conclusions should be better synthesised. So many "bullets" and detail are not common in the conclusions and do not help the reader get the overall idea of the outcomes.
Technical remarks
- Please, read the "Manuscript composition" guidelines (https://www.atmospheric-chemistry-and-physics.net/submission.html) and the LaTeX template, if necessary. Correct the bibliographic references, section/table/appendix/figure references, date formats, etc.
- l. 2: why do you cite the "mesoscale" and not the local scale?
- l. 3: "it" or "them"? Vertical exchange is explained here, but horizontal transport should also be mentioned (also at lines 21-22)
- l. 8: "measurement's" --> "measurement"
- l. 9: "influences"
- l. 16: "too"
- l. 22: bibliographic references can be added (e.g., anticipated from the next lines) after "troposphere"
- l. 31: "air T" --> "air temperature (T)"
- l. 51: "slope-flow-induced local subsidence", please rephrase
- l. 121: differences between Schraff et al. (2016) and the "current setup" are not clear
- l. 132: "upper (southern)"
- l. 134: does the valley with 1.5 km wide floor continue towards NE or NW?
- l. 143: "provide from", check syntax
- l. 161: add one sentence explaining why even though some measurements (ground and remote sensing elsewhere) are assimilated into the model, a later comparison of the model to the same type of measurements still makes sense
- l. 177-178: check syntax
- l. 227: can you add the mean ridge height to the plot as horizontal line?
- l. 228: it is not simple to understand where the "T rise shortly at sunset" is visible in Fig. 3
- l. 232: there is no figure D1b (also at l. 234). What are "standard" values?
- l. 233: "(Fig. 3a) near the ground (590-1000 m) for all months..."
- l. 235: "apart" or "onwards"? Please, correct all references
- Fig. 3: explain the dashed vertical lines in the caption. Change the x marks to, e.g., 6h or submultiple of 24h. Is Fig. 3b mentioned anywhere in the text? Maybe it should be moved next to Fig. 6?
- l. 251: "station altitude" instead of "real topography"
- l. 257: what does "difference in the effect" stand for?
- Fig. 4: the dashed vertical lines should be explained in the figure caption
- Fig. 5: T differences between ... and SMN/MER" (subtraction is non-commutative). Also, is "MERn" a typo?
- l. 279: explain why "705" m
- l. 284: correct missing reference ("??") and similar occurrences
- l. 289: "presents slightly"
- l. 294: "shows"
- Fig. 7b: check sign of deltaT. Is it T_SMN - T_MER (>0) or the opposite (as in the main text)? Use submultiples of 24h on the time axis. Mention why there are data gaps
- l. 319: "or not of the values" please check this sentence
- l. 326: what is a "Bise" wind?
- l. 349: "1500 m", is this altitude limit so clear?
- l. 355: "between 1300 and 1700 m", not clear from the figure
- Fig. 8: how do you deal with direction in calm wind conditions? Why are the white boxes (data gaps) similar for all sources? Clearly write the difference between plots b) and c) even in the figure, not only in the caption
- Fig. 9: explain what positive and negative speeds mean in the caption. Clearly state what interpolation/smoothing technique was used in the contour plots
- l. 403-404: is 800 m "ground level"?
- l. 405: "reduced" compared to what?
- l. 408: "occurred"
- l. 416: "ridge height"
- Fig. 10: y label should be "DWL/MEE" instead of "DWL/MER"
- l. 453: "reported at 200 m", explain why in the instrument description
- Fig. 11: add "FEB"
- l. 467-470: unclear, please rephrase. Maybe clearly state that MER is more sheltered from the N wind than MEE? Can BRZ be influenced by katabatic nighttime winds from the slope?
- Fig. 12: is BRZ really representative of thermal circulation? Also, use correct colours in the legend of Fig. 12b. Add the indication "(orange)" after "sunshine amount" in the caption
- l. 486-487: I cannot fully understand the meaning of this sentence
- l. 490: "up to 1500 m during all other months", not so clear in summer. The N component up to 2500 m is difficult to separate
- l. 503: "Foehn is"
- l. 512: "measured wind speed"
- l. 516: "better agreement ... (not shown)", are your referring to the difference between the two red boxes in Fig. 14b? Is this difference significant?
- Fig. 14: explain the number row at the top of the plots. Correct the intervals on the x axis in Fig. 14b. Correct order of factors in caption (difference should be with reference to SMN/MER)
- l. 521: "(mettre...)", correct typo
- l. 549: why do you mention three different foehn types, when the selected cases are all of deep foehn?
- Fig. 15: correct x labels for 2nd and 3rd columns
- l. 558-560: this sentence is a bit general. Maybe rephrase so that it can be understood that the focus is on "atmosphere over complex topography"
- l. 560-561: isn't it the same over flat terrain?
- l. 569: "As such", please check syntax
- l. 570-576: most of it has already been written. Please, shorten this part
- Sect. 4: use the space here to anticipate the structure of the discussion, especially Sects. 4.2 and 4.3
- l. 585-589: this can be shortened by writing that the forecast skills were not the focus of the paper
- l. 596: "pairs", do you mean at different altitudes?
- l. 607: "months"
- Sect. 4.2.2: too much detail. Do not repeat all results, just mention the most important similarities and differences
- l. 684: "other study" by whom?
- l. 722: are winds "well modeled" despite "the onset is predicted with a larger inaccuracy" (l. 724)? Does the sentence only stands for the wind velocity?
- Table A1: is the last column a % of a %? Write the year the anomaly refers to in the caption
- Figure A1: a y label for the sunshine duration is missing
- Figure B1: Explain the (black and coloured) arrows in the figure. Caption: "downloaded"
- Figure B2: red/blue lines are too thin
- Appendix D, E, F, G, H, I, J: titles should precede the figures
- Figures E1 and E2: y axis label (and measurement unit) is missing. Explain the two lines of plots
- Figure F1: add a legend for the purple, blue and red lines. Where are the dashed lines mentioned in the caption?
- Figure H1: "sunrise"
- Figure I1: the second colour key should report "°", not "°C"
Citation: https://doi.org/10.5194/egusphere-2023-1961-RC1 - AC2: 'Reply on RC1', Martine Collaud Coen, 31 Jan 2024
-
RC2: 'Comment on egusphere-2023-1961', Anonymous Referee #2, 10 Nov 2023
Please find the comments in the attached document.
- AC1: 'Reply on RC2', Martine Collaud Coen, 31 Jan 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 387 | 130 | 72 | 589 | 63 | 37 |
- HTML: 387
- PDF: 130
- XML: 72
- Total: 589
- BibTeX: 63
- EndNote: 37
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1