SLUCM+BEM (v2.0): implementing a prognostic indoor temperature scheme for application to global cities

Takane, Yuya; Kikegawa, Yukihiro; Luo, Zhiwen; Kusaka, Hiroyuki; Grimmond, Sue

doi:10.5194/egusphere-2026-1307

Preprints

https://doi.org/10.5194/egusphere-2026-1307

Preprints

01 Apr 2026

| 01 Apr 2026

SLUCM+BEM (v2.0): implementing a prognostic indoor temperature scheme for application to global cities

Yuya Takane, Yukihiro Kikegawa, Zhiwen Luo, Hiroyuki Kusaka, and Sue Grimmond

Abstract. We developed and released the single-layer urban canopy model (SLUCM) coupled with building energy model (BEM) v2.0, a single-layer urban canopy and building energy model capable of accurately simulating urban climates and electricity consumption (EC) across broad areas with substantially lower computational cost than conventional models. The previous version (v1.0) was a simplified model that set boundary conditions for wall and roof temperatures equal to the heating and air conditioning (HAC) setpoint. This prevented the calculation of indoor temperatures (T_in) under natural ventilation conditions (i.e. without HAC), limiting its applicability to wider regions and scenarios. This simplification was also identified as a key factor in the overestimation of EC in office districts of Tokyo. To address these issues, this study introduced a new version of the model in which T_in varies dynamically based on HAC usage, outdoor temperatures, and ventilation conditions. This enables T_in to be calculated during natural ventilation, and was shown to yield results consistent with observations from residential buildings in London under free-running conditions. Additionally, the overestimation of EC in office districts of Tokyo was significantly reduced. This upgrade facilitates the assessment of climate change adaptation measures for both outdoor and indoor environments. It enables an explicit simulation of the interactions between indoor and outdoor climates and human activities, including the consequent increase in outdoor temperatures due to anthropogenic heat emissions. The model is compatible with standard geographical datasets and existing WRF land-surface and urban physics options. SLUCM+BEM v2.0 is released both as an online WRF-coupled implementation and as a standalone version.

Received: 10 Mar 2026 – Discussion started: 01 Apr 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 6232 KB)

Supplement (301 KB)

Download & links

Yuya Takane, Yukihiro Kikegawa, Zhiwen Luo, Hiroyuki Kusaka, and Sue Grimmond

Status: final response (author comments only)

RC1:
'Comment on egusphere-2026-1307', Anonymous Referee #1, 04 May 2026
The manuscript describes an updated version of the Building Energy Model coupled to SLUCM. The new model is tested against a series of indoor temperature data in London and building energy consumption in Tokyo.
This is a needed improvement and a welcome addition to SLUCM. My opinion is certainly positive. I believe the manuscript can be improved, clarifying some aspects of the implementation and adding some details, in particular about the sensitivity of the new model to different input parameters.
I describe below what I think must be addressed, following the order of the lines of the paper (not the importance).
Line 69. “EC was qualitatively overestimated”. Overestimation is a quantitative judgement; I do not understand how it can be qualitative.

There are a few things missing (probably implied) in the equations that describe the new version. Among others:
In equations (9,10,11), Q_Cr and Q_Cw should be multiplied by the roof and wall surfaces. Same for eq(15-17). The floor is missing.

How is the total indoor heat capacity computed? In other words, what is the volume of indoor air? Is it an infinitely long building? Is it a quadrangular prism? With which dimensions?

I am confused about eq.(16,17), which are not in Kikegawa et al. 2003. These represent the exchange of radiation between the indoor surface and the indoor air. But each indoor surface also receives radiation from the other indoor surfaces. For example, the indoor wall receives longwave radiation from the roof, from the floor, and from the opposite walls. This is missing in eq. 7,8,16, and 17. With the formulas of the paper, the indoor surfaces would lose much more energy than in reality. View factors are also missing. To define them, we need to have a clear definition of the indoor volume considered.

It is unclear how T_in is forced to the target temperature. If you have 100% of the buildings with A.C., then (from eq. 3) H_out=H_in. In this case, T_in will stay constant in time, and so the initial temperature must be equal to the target temperature. But what is going to happen if H_in is negative? Will H_out be negative, so heat will be extracted from the atmosphere to keep the indoor temperature constant? In other words, how is the situation treated where there is no need for A.C. to keep the indoor air below the target temperature in summer?

On the same line of point d), if only a fraction of the buildings are with A.C., then H_out will be less than H_in (see eq. 3), and therefore the indoor temperature can change with time. How’s the information on the target temperature used in this case?

Lines 247-248. AHOPTION and SBEMOPTION – these seem to be WRF’s flags, and should be explained.

Section 3.1. Are there differences in the measurements depending on the floor? Is the top floor hotter during the day and cooler during the night than the other floors?

Lines 329. Is vent-10 equal to vent-rate=10 (of Eqn. 19)?

4. I know there are already many lines, but knowing the value of the outdoor air temperature would help in the interpretation.

Section 3.2. The values of electricity consumption are strongly affected by urban morphology. However, the values used in the simulations (Table 3) are based on only 3 urban classes, with fixed values of urban fraction and building height. How sensitive are the results to these values? Why didn’t the authors use more detailed information (at least the 10 LCZs). Same for the building thermal properties – how realistic are they?

Lines 450-460. “What if” scenario. These results are with the assumption that the ventilation rate stays the same with and without A.C., which is doubtful.

Line 515. The increase in computational time is strongly affected by the number of urban points. Please specify the fraction of urban points present in the domain.

Lines 533-535. I agree. On the relevance of the radiation transmitted through windows, you can refer to Pappaccogli et al. https://www.sciencedirect.com/science/article/pii/S2212095519304213, who made a sensitivity study on this and other parameters.
Citation: https://doi.org/10.5194/egusphere-2026-1307-RC1
- AC1: 'Reply on RC1', Yuya Takane, 01 Jun 2026
  
  Dear reviewer,
  Please see attached our response to your comments.
  Many thanks for your evaluation on our work.
  Best,
  Yuya on behalf of all co-authors
  
  Citation: https://doi.org/10.5194/egusphere-2026-1307-AC1
AC2: 'Revision on egusphere-2026-1307', Yuya Takane, 01 Jun 2026

Dear reviewers,
Please see attached the revision of egusphere-2026-1307 with tracked changes.
Best,
Yuya on behalf of all co-authors

Citation: https://doi.org/10.5194/egusphere-2026-1307-AC2
RC2: 'Comment on egusphere-2026-1307', Anonymous Referee #2, 29 Jul 2026

Review of "SLUCM+BEM (v2.0): implementing a prognostic indoor temperature scheme for application to global cities"
The objective of this study is to enhance the urban canopy model with integreated building energy model SLUCM-BEM v1.0 to SLUCM-BEM v2.0 by adding a prognostic indoor air temperature instead of prescribing fixed values. This allows to better represent situations when there is neither heating nor air conditining, which considerably enhances the model's capability for simulations on global scale and for a variety of meteorological situations. Another improvement is the consideration of air exchanges between indoor and outdoor. The model is evaluated for the simulation of building energy consumption in Tokyo Metropolitan Area and for the simulation of indoor air temperature for buildings situated in London. For the naturally ventilated London buildings, a large variety in the observed indoor temperature is found, which is likely due to different human behaviour. An ensemble of SLUCM-BEM simulations with different ventilation rates is made to represent this variability. It is shown that the average SLUCM-BEM results correspond well with the averaged observations. A reduction of the positive bias in building energy consumption is found for Tokyo in both summer and winter seasons. The sensitivity of building energy consumption to outdoor air temperature is also improved for summer, but not for winter. Similar accuracy for the outdoor air temperature is found for the Tokyo region when using the new model version. A sensitivity study is made to investigate the effect of air conditioning on indoor and outdoor near-surface air temperature for Tokyo. The results are coherent with other previous studies on this topic.
This is a well-written study presenting a relevant urban canopy model development that could be published after some revisions.
Major review points:
- Indoor humidity is not included in the new BEM. This is a drawback since it might influence indoor human thermal comfoort. It might be added and maybe the observation data for London also provide humidity.
- Ventilation seems to be a fixed mechanical ventilation. In reality, there might also be infiltration (air exchange due to tiny holes in the building envelope) or natural ventilation due to window opening. Please discuss that this is not considered. Infiltration might vary depending on the meteorological conditions (e.g. wind speed), which makes the model using a constant ventilation rate different from reality, especially for older, less airtight buildings.
- A potentially questionable method is employed for the internal thermal infrared exchanges (Equations 16 and 17). It is assumed that the infrared radiative exchange takes place between the interior surfaces and the interior air. However, most of the infrared radiation will not be exchanged with the indoor air, but with the other interior surfaces. So it would be more precise to adopt a radiosity approach for the interior infrared exchanges. The uncertainty from the approach adopted in the present study might be low in most cases since the indoor air temperature will be quite close to the interior surface temperatures, but the potential uncertainty should be mentioned or investigated. L193 is also highly questionable. Equation (18) looks like the linearised Stefan Boltzmann equation. So ew should not be the thermal conductivity of air (but the emissivity) and sigma not the dynamic viscosity of air (but the Stefan Boltzmann constant).
- Table 1 seems to provide the state-of-the-art concerning the UCP-BEM models only as they are coupled to WRF. Please research also the developments of

the UCP-BEM models that are not integrated in WRF.
- There is a mix of symbols rather mathematical (e.g. T_in) and symbols that are rather like model variables (e.g. TBLEND). I suggest unifying the nomenclature to use only mathematical symbols and following the typical nomenclature in the field.
- Sometimes a double naming of variables seems to be employed. For example in L145: is AB_BUILD_RATIO = a in Equation (3)? In this cast it is confusing. Please use the same symbol at all places for a given variable.
- The lengthy Tables 3 and 5 with the parameter setting should be moved to an Appendix.
- A nomenclature could be added in the Appendix.
Minor review points:
- Page 2, L36-37: This sentence seems not correct.
- Page 2, L43-46: It seems not 100% clear what is Qf and what Qfb.
- Page 2, L55: What means "official integration"?
- Page 3, L63: use subscripts for Sup and Lup.
- Page 3, L63: the formula for the urban energy balance also includes the anthropogenic term (qant), which seems here neglected in the formula calculating the storage heat flux as a residual.
- Based on the explanations on Page 3, L67, is not actually clear how SLUCM-BEM v1.0 works. How can the indoor wall and ceiling temperatures be equal to the heating and cooling setpoints. Can these two values be different? How is it decided which one is taken at which time? Or must heating and cooling setpoint be the same value in SLUCM-BEM v1.0?
- Page 3, L75: Maybe better to use a mathematical symbol for "wall surface temperature"
- Page 4, Table 1: What is the difference between "prescribed" and "input"?
- Table 2 is quite difficult to read since it contains a lot of text.
- Page 6, L150: is it EC or Qfb which switches sign between summer and winter?
- Page 7, L155: what is done in the tropical regions which do not have a summer or winter?
- Figure 1: It might be better to redo this figure without explicitly hardcoding the number of layers to 4. Maybe just use a symbol for the number of layers

and say in the caption that the default value is 4. The hardcoded number 4 is also appearing in some equations (e.g. 7 and 8).
- Page 9: There might be an issue in the Equations involving Qcr, Qrr, and so on. Since there is not the same roof area than wall area, must there not be factors converting the m2 of walls/roofs into m2 of built area?
- L185: which value is assumed for the thermal conductivity of air here? Is this a fixed value for laminar conditions?
- L240 (and other instances in the manuscript): the names of the urban land cover classes (C, Rm, Rd) are not understandable.
- L247-250: is it necessary to include these lengthy details on technical model options?
- Figure 2: It is not necessary to use the numbers (31, ...) for the urban types. Please only use one meaningful name in the text and in the plot legend.

In (b), the height label is colliding with the text.
- Figure 3: Why is there this double legend? This needs to be explained.
- Section 2.3 -> metrics
- L317: "it then decreased daily" -> is "daily" needed?
- Figure 5: The panels could be reorganised to make them larger, but still using only one page.
- Figure 6: is it really "type of urban building" or better "type of urban land cover"?

The legend also exceeds the plot.
- Figure 9: it is not necessary to repeat details like colours (given in the legend) in the plot caption.
- Table 4: is "verification" here the right word or should it be "evaluation metrics" or similar.
- Figures 10 and 11: heat_load -> heat load
- L469: the section 4.3 title needs to be reformulated.
- L553: The beginning of the summary might contain a short summary of the main developments and findings. Instead these are spread out in later parts of the summary.

Citation: https://doi.org/10.5194/egusphere-2026-1307-RC2

Yuya Takane, Yukihiro Kikegawa, Zhiwen Luo, Hiroyuki Kusaka, and Sue Grimmond

Supplement

https://doi.org/10.5194/egusphere-2026-1307-supplement

Data sets

WRF-SLUCM+BEM: Input data for the evaluation at Tokyo Metropolitan Area Yuya Takane, Yukihiro Kikegawa, Ko Nakajima, and Hiroyuki Kusaka https://doi.org/10.5281/zenodo.13932603

Model code and software

WRF-SLUCM+BEM (v2.0) source code for GMD submission Yuya Takane and Yukihiro Kikegawa https://doi.org/10.5281/zenodo.18918749

Yuya Takane, Yukihiro Kikegawa, Zhiwen Luo, Hiroyuki Kusaka, and Sue Grimmond

Viewed

Total article views: 650 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
397	220	33	650	81	27	37

HTML: 397
PDF: 220
XML: 33
Total: 650
Supplement: 81
BibTeX: 27
EndNote: 37

Views and downloads (calculated since 01 Apr 2026)

Month	HTML	PDF	XML	Total
Apr 2026	263	120	16	399
May 2026	61	30	5	96
Jun 2026	31	12	6	49
Jul 2026	37	50	5	92
Aug 2026	5	8	1	14

Cumulative views and downloads (calculated since 01 Apr 2026)

Month	HTML	PDF	XML	Total
Apr 2026	263	120	16	399
May 2026	61	30	5	96
Jun 2026	31	12	6	49
Jul 2026	37	50	5	92
Aug 2026	5	8	1	14

Viewed (geographical distribution)

Total article views: 624 (including HTML, PDF, and XML) Thereof 624 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 06 Aug 2026

Short summary

We developed and released the single-layer urban canopy model coupled with building energy model v2. This incorporates a new scheme enabling dynamic changes in indoor temperature. This allows the model to be applied not only to air-conditioned conditions but also to non-air-conditioned scenarios, making it applicable to all seasons and cities worldwide. This upgrade facilitates the assessment of climate change adaptation measures for both outdoor and indoor environments.


Total:	0
HTML:	0
PDF:	0
XML:	0