the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 Jan 2025
Modelling extensive green roof CO2 exchanges in the TEB urban canopy model
Abstract. Green roofs are promoted to provide ecosystem services and to mitigate climate change in urban areas. This is largely due to their supposed benefits for biodiversity, rainwater management, evaporative cooling, and carbon sequestration. One scientific challenge is quantifying the various contributions of green roofs using reliable methods. In this context, the green roof module already running in the Town Energy Balance urban canopy model for water and energy exchanges was improved by implementing the CO2 fluxes and the carbon sequestration potential. This parametrisation combines the ISBA (Interaction Between Soil Biosphere and Atmosphere) photosynthesis, biomass and soil respiration module with the green roof module in order to quantify the net CO2 amount emitted or fixed by the green roof over a time period. The parametrisation was fully achieved by using data of an extensive Sedum non irrigated green roof located at the Berlin BER airport in Germany from 2016 to 2020. The five years of measurements were used to do a sensitivity analysis of the photosynthesis module parameters in order to classify the parameters according to their influence, followed by a calibration over the most important parameters and evaluation. Results show a good agreement of the simulated leaf area index and CO2 fluxes with in situ observations, with good diurnal, seasonal and inter-annual variability, even if the model tends to be overly responsive on the day to day variability. The model reproduces well the Net Ecosystem Exchange which provides a reliable estimation of the annual carbon sequestration. Those results are encouraging in quantifying the potential of carbon sequestration of green roofs and open up the possibility of applying the new parametrisation on a city-wide scale to evaluate green roof scenarios.
- Preprint
(7866 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on gmd-2024-233', Anonymous Referee #1, 18 Feb 2025
This paper presents the integration of the CO2 fluxes calculations in the TEB green roof model. This manuscript is already in very good conditions regarding its model implementation and testing and should be accepted after the subsequent revisions which are still considered substantial. In particular, more information should be provided on the implications of choices made in the parameterisation of the green roof scheme. The discussion section should be entirely rewritten and more comparison to the existing literature should be perpetrated. More detailed comments can be found below.
1. Regarding the introduction, you should provide more background on the different green roofs models that exist in the urban climate modelling field. You should also provide more background on the variety of applications that these models can have, list the ones that do have a CO2 implementation and comment on how the TEB model compares to other ones previously used for other types of studies. We need to better understand why you implement this CO2 fluxes scheme in TEB more than in another existing model.
2. Please define the two types of upwelling and downwelling radiations for non expert audiences.
3. Could you please detail slightly more what sort of photograph analysis is performed? Does it compare to other remote sensing techniques for flagging green roofs or quantifying LAI? Why is this useful for your moelling and what are the implications of potential uncertainties?
4. Explain the choice for using TN and TX naming. Can you use more easily interpretable anagrams?
Figure 1. The area covered by the green roof seems very large. Could you provide some more information on how that green roof surface compares to other green roofs in the city, in Germany and maybe even in Europe? Why does that make it an interesting case study? If it is just related to accessibility to data records, then say it. An aerial imagery would also be greatly welcomed.
Figure 2. Please prevent from using double Y axes. Separate the plot in two. Why do you use maximum and minimum over average temperatures? Where are these values recorded? Is it at the station on top of the roof or a local nearby AWS? If it is on top of the roof, what are the implications of the height at which records were made for your model parameterisation/evaluation (depending on how you use them)?
Lines 94 to 107 are somewhat of a repetition to the introduction. Can't you provide more details on how these model are distinct from other existing models? What are their pros and cons? You can keep the subsequent model description with equations that is very useful. Maybe you would like to put this in the supplemetary material and only relate certain equations to the calculation of CO2 fluxes which you describe later in section 3.3?
Line 175. Models* Check for typos across the manuscript.
Lines 219 to 220. What are the implications of such reccurent forcing? I would like the authors to test for the influence on meteorological forcing since urban climate models are typically not forced every 30 minutes but more usually at 3-hourly or 6-hourly timesteps -- sometimes hourly for very specific cases.
Line 230. So the sensor is located at 20 meters. How is your model expected to behave in a "2-dimensional" model? Please provide information on how the temperature is interpolated to the roof level within the 6 vertical layers. If it relies on 2m air temperature, then what are the implications?
Section 5.1 and 5.2. Is this sensitivity analysis really necessary for a non-expert audience? Can't you sum it up and provide the details in the appendix?
Line 340 to 341. What do you mean here? How was this information integrated in the model and how did that influence the outcome of your study? How are observations consistent with the LAI and Fveg observations? Can you discretise your observations based on these parameters (e.g., through seasonality)? We would need to better understand whether the observed LAI and Fveg served for the parameterisation of the model or simply its evaluation. In both cases we need to be provided with more information on the way in which this data was gathered. Visual inspection is not sufficient. This questions the validity of this data. Also, we need to understand in more details what are the implications of accurately modelling the LAI for the CO2 fluxes calculations. Since you have observations in Figure 7 about CO2 fluxes, why do you need to evaluate this intermediate step? If you can't maybe just say that the model is right but that a limitation is that you don't know whether it is right for the right reasons or not.
Figure 8 and its description should be moved to the results section. An explanation of the reason for this analysis should also be provided in the methods.
Figure 9 should also be provided in the results and not in the discussion. The discussion should discuss the results against other literature outcomes and try to explain the outcomes of this study based on current knowledge.
In general, the whole discussion section should be restructured and rewritten. The comparison to existing literature and models is critically lacking.
Citation: https://doi.org/10.5194/gmd-2024-233-RC1 - AC1: 'Reply on RC1', Aurélien Mirebeau, 11 Apr 2025
-
RC2: 'Comment on gmd-2024-233', Anonymous Referee #2, 04 Mar 2025
Modelling extensive green roof CO2 exchanges in the TEB urban canopy model
Mirebeau et al.
The authors calibrate and evaluate an existing greenroof model and photosynthesis module, coupled with an urban canopy model, against measurements of carbon dioxide fluxes and leaf area index at an extensive green roof. Results are generally satisfying, suggesting the model is capturing key processes adequately.
General comments
Methods are solid and results are interesting. The presentation, primarily the written text, could benefit from improvement throughout. In a number of cases it is difficult to follow or lacking specific definitions or other information.
The Discussion section appears to be another Results/Model Evaluation section, rather than a Discussion section. For example, little to no comparison with other studies, or contextualization of the current results with the literature, is performed. The paper can be strengthened by adding some discussion on the shortcomings of the current modelling/data acquisition process and the possibilities for applying the calibrated variables in city-regional scale simulations of green roofs (vs. the need to calibrate them for other locations, green roof characteristics, etc). The manuscript would be greatly strengthened by these additions.
Specific comments
Lines 7-9: ‘The parametrisation was .. to 2020’ → ‘Measurement data from an extensive... are used to calibrate and evaluate the parametrisation’,
Line 10: ‘according to their influence’: Not sure what this means,
Line 12: ‘even if’ → ‘although’,
Lines 20-24: This is a very long and convoluted sentence,
Lines 25: ‘Firstly ... savings’: This is partially true depending on the energy generation source,
Lines 26-35: It seems unnecessary to list all the scientific names of the mentioned plants given later the authors were grouping them into C3 or C4 types. Is it also possible to put the amount of direct carbon sequestration into perspective, i.e., how much is considered significant or trivial?
Line 45: If the authors are using latex package, CO2 can be written as \chem{CO_2},
Lines 44-51: Reads awkwardly,
Figure 1: In addition to a site photo, can you also add schematics of all the instrument setups? Also: “airport” not “airprot”
Line 81: “contrasting” not “contrasted”
Line 90: ‘(for vegetation and natural soils, lakes, oceans etc.)’ →, ‘such as vegetation and natural soils (cite ISBA) and lakes (cite FLake)’
Line 104: In addition,,
Line 110: ‘+H’ → ’−H’,
Lines 113-115: Can you also include the equations for LEG and LEV?
Lines 116: the following ISBA parametrisation:,
Line 122: ‘allows for the discretisation’ → ‘discretizes’,
Line 141: Explain the Kersten number or provide a reference, since it will not be common knowledge for most readers.
Lines 143-145: ‘Similar to the formulation of radiative, thermal and hydrological exchanges, CO2 fluxes in TEB-GREENROOF are adapted from an existing module in ISBA’,
Lines 149: Suggest moving the formula for Reco given here to previous sentence, and including the formula for NEE here.
Line 153: Suggest adding some description of C3 and C4 plants,
Lines 155-164: This whole paragraph is rather difficult to read and a definition of CAM is also needed.
Section 3.3: Can you add the equation for Rleaf calculation? It may be better to add more context to the leaf respiration description by moving part of the content in Appendix A here, to provide context on why certain modelling decisions were made. Variables like Γ and ϵ0 used in the sensitivity tests were not even mentioned in the main text.
Line 167: ‘put the inhibition functions to 1’, a little more context is needed,
Eqs. 8, 9: Define “Q10”
Line 175: ‘model’ → ‘models’,
Lines 182-183: “This empirical formulation is simpler than the one proposed by Calvet (2000) for herbaceous plants.” The significance of this sentence is not clear.
Line 186: ‘set to prevent extreme values’: please include threshold values.
Line 191: ‘three different reservoirs’ → ‘three different reservoirs (Bi)’,
Line 195: Can you include some description of the formulation of Di and Ri?
Line 200: ‘SLAI’ → ‘SLA’,
Line 202: You could consider removing (since it is redundant): “…which is the case on the experimental site studied here”
Lines 208-210: ‘For application ... (see Sect. 5).’ → ‘Since it is challenging to find appropriate ecophysiological data to derive these parameter values for Sedum, we chose to calibrate some of them in Section 5.’
Sections 2 and 4: I recommend adding a table with all the measured variables and also indicate if set variables are used as model input/output to help shorten the description in Section 4. It’s probably better to combine Sections 4.1 and 4.2 and focus on adding details on model configuration that are not included in the Tables. The coefficient βcoef has not been mentioned previously and needs a proper introduction. In general, it is not easy to track the various parameters and variables, and how they are set or used in the various components of the paper. Some unified way to improve the communication on this front would be helpful.
Table 1: Can you add the symbols for each parameter? A roughness length of 0.03 m seems quite high (implying vegetation on the order of 30 cm in height). Dry soil heat capacity is substantially higher than any published values I’ve seen – is the soil composed more of clay or sand or…? It would help to have references backing up some of these choices.
Line 230: 1.15 m is very close to the green roof surface for a wind speed measurement – is this for source area considerations?
Lines 238-240: This is a bit vague. It doesn’t seem like the BEM should have significant impacts on rooftop CO2 exchange or plant growth. However, if it does a few more details should be given related to the BEM setup.
Line 254: “… on the specific characteristics of the current green roof to be simulated…”
Line 267: NEE has not been defined,
Line 274: “As for Si, …”
Sections 5.1: I’m not sure if it’s necessary to go into detail on the statistical operation, i.e., the Sobol index method, in the main text. It may be good enough to explain the two indices and move the operation procedure into the appendix.
Line 286: ‘d = 8’?
Line 293: What is the sampling matrix, and why is this the appropriate formula for the number of simulations. Please describe your methodology to be comprehensible to a general urban climate modeler.
Line 299 ‘make possible’ → ‘make it possible’,
Line 314: ‘Tab. 2.’ → ‘(Table 2)’,
Line 319: Can you provide a reference to the two tested τm values?
Line 334: “So a choice needs to be made” -> Delete, or clarify and correct grammar.
Section 5.3.1: Can you state again the rough temporal resolution of the estimated LAI from the photographs?
Line 352: The last sentence seems out of place.
Lines 359-362: Please quantify these discrepancies and report them in the text.
Figure 8: Is it possible to condense this caption description while retaining the essential elements?
Lines 375-384: The panels are discussed starting on 8c) and ending with 8a). Consider switching the order of the panels in Fig. 8 accordingly.
Figure 9: Is it one standard deviation? Also, can you make the zero line thinner?
Line 387: “does not simulate the diurnal CO2 cycle” ? -> unclear, related to my next point?
Lines 386-388: I think model and obs need to be switched in this sentence… e.g. see Fig. 9. The model has a diurnal amplitude during winter, not obs.
Line 388: However
Lines 395-396: This is confusing to read, please rephrase.
Figure 10: Can you add a faded background grid to help assess the monthly time scale?
Table 3: I’m confused by the split between calibration and evaluation period. Can you specify which set of parameters are used for the simulations in Section 6?
Line 410: “this is ... NEE is rather high”, not sure if I understand this statement,
Citation: https://doi.org/10.5194/gmd-2024-233-RC2 - AC2: 'Reply on RC2', Aurélien Mirebeau, 11 Apr 2025
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 200 | 48 | 12 | 260 | 10 | 11 |
- HTML: 200
- PDF: 48
- XML: 12
- Total: 260
- BibTeX: 10
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1