the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 Aug 2024
“Pochva”: a new hydro-thermal process model in soil, snow, vegetation for application in atmosphere numerical models
Abstract. This work presents the land model “Pochva”. “Pochva” is a model of hydro-thermal processes at the Earth surface and in the underlying medium. The model simulates the main hydro-thermal parameters of the surface, soil layer, vegetation and snow layer. Its soil process scheme allows to use physical parameters having vertical variations along the soil profile. Its snow processes scheme is a multiple layer scheme and has a numerical algorithm allowing to solve both cases of extremely thin and extremely thick layer. The model is marked by a particular accuracy in simulating the water phase transitions in soil and snow, and by the autonomy in the determination of the lower boundary condition in the soil column. The model can be used as a stand-alone land-surface model driven by observed or analytical forcing data, or coupled to an atmospheric model, either global or limited-area, either in forecast regime or climatic (hindcast) regime. The results of coupling “Pochva” to the numerical weather prediction limited-area model “Bolam” are presented in this article.
- Preprint
(2610 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on gmd-2024-138', Anonymous Referee #1, 19 Sep 2024
This article presents a comprehensive model of hydro-thermal processes in vegetated soil and snow cover, emphasizing the vertical variations of physical parameters along the soil profile. The model features a sophisticated snow processes scheme capable of handling extremely thin and thick snow layers, ensuring accurate simulation of water phase transitions in soil and snow. The interaction between the atmosphere and land surface is simulated, incorporating vegetation and snow cover as independent thermodynamic entities. The model's verification in the NWP model Bolam demonstrates its performance across various climatic zones, revealing biases and RMSEs for temperature and humidity, with particular attention to seasonal variations and geographical differences. The model is based on the idea of describing the thermal state of soil environment using entropy, and pays attention to the soil water phase transition, making it novel and potentially applicable in atmospheric and climate research. However, the article still needs some revisions before it is suitable for publication.
1. The colors in Figure 1 seem to be inconsistent with the colors in the legend (color for "Liquid water on the plant leaf surface"). Please check if there is a problem.
2. The meaning of β veg in equation 11 does not seem to be given in the article.
3. In Figure 2, it mentions k=nlev_soil, where nlev_soil refers to the maximum number of soil layers？Can this maximum number of layers vary, or is it a fixed value? Also, is the vertical stratification of soil layers in the model uniform, or does it vary across different spatial locations?
4. Why were the simulation results from 2013 to 2015 over the European region selected for validating the Pochva model, instead of selecting periods from the routine operational weather prediction results of CNR-ISAC, where Pochva has already been applied?
5. In the evaluation of land surface models, offline mode is commonly used. Can the Pochva model perform offline simulations? If so, it is suggested to use offline simulations for evaluation to more effectively avoid the uncertainty in atmospheric models from influencing the evaluation results.
Minor modifications:
1. It is recommended to remove the phrase 'of National Center of Atmospheric Research (USA)' after the Common Land model in lines 26-27, as the later development and application of this model were mostly carried out in China.
2. line 316 then shuld be than?
3. line 385 does the equation in this line need a number?
4. line 559 snow us should be snow was?Finally, the reviewer would like to express his respect for the author of the article. As we all know, developing a new land surface model all by oneself is extremely difficult and may face various challenges.
Citation: https://doi.org/10.5194/gmd-2024-138-RC1 -
AC1: 'Reply on RC1', Oxana Drofa, 30 Sep 2024
Thank you very much for your appreciation of my work, you are very kind, your evaluation is very valuable to me! Really, this is the result of 15 years of my work.
I will answer you point by point.
1) The color of the water zone of “Liquid water on the plant leaf surface” in the figure and in the legend is the same, but it is semi-transparent, in the figure the color of the vegetation zone comes through it. I aimed to make the effect that one zone is superimposed on another, i.e. it is a part of it, not a separate zone. If you think that this effect is not successful and only creates misunderstanding, then, I redrawn the figure.
2) There was a mistake after formula (6), a and b instead of α and β, formula (11) is the explanation of β coefficient from (6), I hope now it is clear.
3) Yes, you are right, k=nlev_soil is the index of the bottom level in soil column, the boundary condition for solving the differential equation is defined at this level.
Your questions about the distribution of vertical levels and the depth of the bottom level are correct. I did not put the answers to these questions in the text because it is a special question that requires an extensive discussion. Vertical levels in soil for numerical solving the equations of heat and moisture exchange in soil do not have an uniform distribution, the soil layers thickness increases with depth according to the exponential law. But, of course, this distribution can be changed. The definition of the bottom level depth depends on how the model will be applied for specific purposes. The presented numerical scheme assumes that the depth of the bottom level for solving the heat exchange equation may not coincide with the depth of the bottom level for solving the moisture exchange equation. In both cases, this depth is determined by the boundary conditions to be specified. Also, the scheme assumes that both of these depths may be different at different domain numerical grid point depending on the geographical conditions that define the boundary conditions for the heat and moisture exchange equations. I plan to publish my study with a proposal on how these conditions can be set in different geographical conditions, as it is quite a big special study.
I included a brief explanation of both points in the text.
4) That's a correct question. The verification of the application of the “Pochva” scheme presented in this paper was done before “Pochva” was introduced into the versions of the NWP models used in CNR-ISAC for operational forecasting, moreover the operational setup of the models was not ideal for verifying the Pochva scheme, these are the reasons for which a dedicated long test experiment was performed. The NWP model chosen for this test was the Bolam model, a hydrostatic numerical model that can have an extensive domain. The author aimed to test the Pochva scheme over a wide range of geographical conditions, i.e. a wide domain. Using of the global model “Globo” was not reasonable, since the global domain assumes that 3/4 of the domain is covered by the ocean, i.e. not useful for soil scheme testing, but increases computational resource consumption. The Bolam model was applied on a domain much more extensive than that used for operational forecasting to capture the hot desert regions of Western Asia, the oceanic regions of Europe, whole Mediterranean region, whole Scandinavia (marine polar climate) and the region up to the Northern Urals (continental polar climate).
5) You are right, it would be correct to test the soil scheme in the column version. Of course, the scheme, in its current form, can be used this way. The author presented 6 test datasets obtained from the simulation with the full NWP model for this application for testing the model. Unfortunately, the soil scheme requires many data at the upper boundary (heat and water vapor flows in the surface layer, precipitation flows at the surface, radiation flows at the surface) and the data about vertical profiles of soil physical parameters in the observation points. Observation data about these parameters should be used to obtain with column version of Pochva the results that can be compared with the observational data about air temperature and air humidity in surface layer or/and about soil temperature and soil water content. I was not able to find such observational data for single- column Pochva scheme verification, so the column tests are proposed with data obtained from the full NWP model.
Minor modifications:
Thank you for you attention and corrections. These were my mistakes. I corrected everything in accordance with your indications. For point 3, the formula has been put inline since it is a trivial average.
Citation: https://doi.org/10.5194/gmd-2024-138-AC1 -
RC2: 'Reply on AC1', Anonymous Referee #1, 18 Oct 2024
Thank you for your reply and detailed explanations regarding my feedback. Your further clarification of the model's workings and validation strategy has provided me with a better understanding, and I appreciate your patience.
Regarding Figure 1’s semi-transparent effect, while I understand your intention to show the overlapping areas, this effect may lead to confusion. Therefore, I suggest redrawing the figure to make the different sections clearer and avoid potential misunderstandings.
As for the other issues, I appreciate your detailed clarifications and modifications, which have improved the clarity and readability of the paper. I believe these changes significantly enhance the quality of the manuscript, bringing it closer to publication.
Thank you again for your hard work and dedication!
Citation: https://doi.org/10.5194/gmd-2024-138-RC2
-
RC2: 'Reply on AC1', Anonymous Referee #1, 18 Oct 2024
-
AC1: 'Reply on RC1', Oxana Drofa, 30 Sep 2024
- AC2: 'Reply on RC2', Oxana Drofa, 21 Oct 2024
-
RC3: 'Comment on gmd-2024-138', Anonymous Referee #2, 24 Oct 2024
General Comment
The paper presents “Pochva”, a model of hydro-thermal processes at the Earth surface to be included in meteorological or climate models. The paper is interesting and can provide a valuable contribution to the community. However, in my opinion, some weaknesses should be solved before the paper can be accepted for publication.
In most cases, it is not clear what are the elements of novelty of this scheme with respect to other land surface models. The Author should better emphasize the similarities and the differences of the new parameterization with respect to existing land surface models, also to highlight why users should choose “Pochva” against other land surface models.
The analysis of the performance of the parameterization presented in the paper does not allow a clear evaluation of the strengths and weaknesses of the scheme, since Pochva is online coupled with the NWP model Bolam and errors in an NWP model can be caused by different interacting factors. It would be better to evaluate Pochva offline (directly forced by observations). Otherwise, it would be interesting to compare the results of Bolam online coupled with different land surface models. In this way, it would be easier to evaluate the strengths and weaknesses of Pochva.
Major Comments
- Model description: the formulations adopted to describe the various physical processes should be better justified, pointing out if they have been taken from previous literature, adapted for this new scheme or if they are completely original. As an example, in Section 2.2, it is unclear how and why Equations 3, 4, and 5 were chosen for defining the empirical soil coefficient αsoil. This is valid for most of the equations presented in this paper. The Author should try to emphasize the elements of novelty of Pochva through comparison with the formulations used in other land surface models. In the present version of the manuscript, the comparison with other existing land surface models is completely absent, whereas Pochva should be described in the framework of the literature in this field.
- Model evaluation: as written in the General Comment, in the present version of the manuscript it is difficult to evaluate the performance of Pochva because (i) Bolam model errors can be due to different reasons, so it is not clear if we are analysing errors related to the land surface model or to other components of the NWP model, (ii) a comparison with the results provided by other land surface models is not present. Therefore, it is not clear how to evaluate strengths and weaknesses of Pochva. Indeed, in Section 7 the Author speculates on the possible causes of model errors, but clear conclusions cannot be drawn.
- Quality of the presentation: The quality of the presentation is not always satisfying. A review by a native English speaker would be beneficial.
Minor and technical comments
Line 30: check the reference style
Line 43: “to the inhomogeneity IN soil parameters along…”
Line 67: “is composed OF a set of fractions…”
Lines 86-87: from this sentence it seems that the fraction of wet low-vegetation leaves is pre-assigned, whereas it is variable and calculated by the parameterization.
Equation 2: T0 should be defined.
Line 108 and 110: asoil should be αsoil
Line 112: “see section 6” should be "see section 7”
Line 137 and line 148: bveg should be βveg
Line 146: W m-2 and not Watt m-2
Line 158: it is not clear how kroot is evaluated.
Line 209: W m-2 and not Watt m-2
Line 210: should be
Line 252: “can either be composed OF soil partially covered…”
Lines 400 and 453: “Let’s” is colloquial.
Equation 68: ficesoil should be fb
Line 458: “temperature before and after…”
Line 500: “and density are conserved.”
Line 516: errors in displaying the symbols.
Lines 517-521: I do not understand the hypothesis that all the liquid water at the beginning of the time step is at level k and at the end of the time step is at level k+1, also considering that the layer thickness is variable in time. This assumption should be better justified.
Line 549: rsnow should be ρsnow.
Line 559: “The heat conductivity of snow IS defined…”
Line 559: check the correct reference style.
Line 567: Is Ssnow the entropy of soil water?
Line 580: soil water entropy or snow entropy?
Lines 593-594: “diagnose the length of the time interval during which snow is exposed to melting”.
Line 610: “at 2 m above SURFACE”
Line 655: “it can thus be noticed that…”
Line 747: “2-m dew-point temperature”
Line 762: “winter and autumn” should be “spring and autumn”.
Lines 782-783: The Author says that the presented model can be useful for modelling conditions over polar or cold climate zones and in the forecast of snow cover. However, model results show the highest errors in winter in continental and mountain areas. In this regard, the Author says that the higher errors can be explained by a poor simulation of the snow cover. Therefore, the conclusions seem inconsistent with the results.
Line 785: “or to OBSERVATIONS of air surface…”
Figure 1: the color of “Liquid water on the plant leaf surface” is not consistent between figure and legend.
Figures 4-7: I would plot the orography in gray shading to better see the station points.
Caption Figure 9: “Average diurnal cycle…"
Caption Figure 10: “Same as figure 9…"
Caption Figure 11: “Average diurnal cycle…"
Caption Figure 10: “Same as figure 11…"
Citation: https://doi.org/10.5194/gmd-2024-138-RC3 -
AC3: 'Reply on RC3', Oxana Drofa, 29 Oct 2024
Thank you for your detailed comments.
I will respond to each point below.
Major Comments
Model description:
I agree with your comment. I will rewrite the explanation to the presented formulas in order to more clearly indicate which formulas are completely original and which are taken from the literature and adapted in this paper so that the results of application of the soil scheme in a full atmospheric model would give the best results.
Model evaluation:
Your comment is fair. The problem is that in the scheme of parameterisation of physical processes is developed and tested in the framework of a specific complete numerical atmospheric model with the purpose of improving the parameterisation of these physical processes in the model. The insertion af a different parameterisation in the same atmospheric model, for comparison, is a separate and big task. which is beyond the scope of the presented work. I agree that it would be useful to verify the “Pochva” scheme on "independent" data. I will look for available observational databases for testing “Pochva”. If I manage to find such data and do an numerical experiment with the column off-line version of “Pochva”, then I will insert the results of this testing into the manuscript.
Quality of the presentation:
I will try to find a service of manuscript review by a native English speaker.
Minor and technical comments
Of course, all your corrections and comments will be included in the manuscript.
Concerning comment at lines 517-521: the author has estimated that the infiltration of water in a snow layer, considering the possible amounts of water involved and the possible layer thicknesses, will take place in a time which is much shorter than a reasonable time step for the integration of the Pochva model or of a full atmospheric model in general, so it is assumed as instantaneous. This assumption will better justified in the text.
Concerning comment at lines 782-783: you are right, there is an apparent inconsistency in the presentation of the results and conclusions; the simulation of surface processes in the polar or cold climate zones is in general more challenging due to the presence of snow and ice cover and to due the poorer simulation of precipitation by atmospheric models in these areas. For this reason, in my view the results in these regions, although worse than the results for other regions, are to be considered good. This will be better justified in the text.
Citation: https://doi.org/10.5194/gmd-2024-138-AC3
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 246 | 119 | 151 | 516 | 8 | 6 |
- HTML: 246
- PDF: 119
- XML: 151
- Total: 516
- BibTeX: 8
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1