pith. sign in

arxiv: 2605.20571 · v1 · pith:CMZRORZAnew · submitted 2026-05-20 · ⚛️ physics.flu-dyn

Multi-scale flow analysis for scale-aware urban-canopy models

Jingzi Huang , Maarten van Reeuwijk This is my paper

Pith reviewed 2026-05-21 02:57 UTC · model grok-4.3

classification ⚛️ physics.flu-dyn
keywords urban canopy modelslarge-eddy simulationscale-aware parameterizationflow heterogeneitynumerical weather predictioncoarse-graininghectometric resolutionmorphology dependence
0
0 comments X

The pith

Urban canopy models in weather forecasts lose accuracy when grid resolution approaches a morphology-dependent heterogeneity scale identified from campus simulations.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper examines what happens when numerical weather models reach resolutions fine enough to partially resolve city blocks. Using a multi-scale filtering method on detailed airflow simulations of a university campus, it tracks how building-induced flow variations change as the effective grid size shrinks. Two layouts are compared: one with large open spaces and one where those spaces are filled with buildings. The analysis reveals a characteristic length where resolved and unresolved effects become comparable, differing sharply between the layouts. Standard drag and turbulence formulas hold only at coarser resolutions where the flow looks uniform, and they fail faster in realistic arrangements than in simple block arrays.

Core claim

Applying multi-scale coarse-graining to building-resolving LES of the Bristol campus in original and infilled morphologies shows that a morphology-dependent heterogeneity scale emerges where resolved and unresolved variability are comparable, at roughly 256 m for the original layout and 64 m for the modified case, and that distributed drag and turbulent-stress parameterisations derived from idealised geometries maintain fidelity only when resolution lies well above this scale.

What carries the argument

Multi-scale coarse-graining framework applied to LES velocity and stress fields to quantify how flow heterogeneity evolves with filter scale and morphology.

If this is right

  • Idealised-geometry parameterisations remain usable only at resolutions substantially coarser than the heterogeneity scale where horizontal transport stays negligible.
  • Parameterisation error grows rapidly once resolution nears the heterogeneity scale because of rising filter-to-filter morphological variability.
  • Realistic urban layouts with open-space contrasts impose stricter resolution limits on existing parameterisations than idealised cuboid arrays.
  • Urban canopy models can be made scale-aware by relating their coefficients directly to the local ratio of grid size to morphology-specific heterogeneity length.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Repeating the coarse-graining analysis on LES of other cities could show whether heterogeneity scales follow predictable patterns tied to building density or street layout.
  • NWP codes could incorporate local morphology statistics at each grid cell to adjust or switch urban parameterisations as resolution changes.
  • The same filtering approach might apply to partially resolved forests or complex terrain to create consistent scale-aware surface schemes across different land types.

Load-bearing premise

The coarse-graining procedure applied to the LES data accurately reproduces the effective filtering that an NWP model grid performs at hectometric resolution.

What would settle it

Running the urban parameterisations on filtered LES fields at resolutions near 256 m and 64 m for both campus layouts and checking whether the error in predicted drag and stresses rises sharply once resolution approaches the identified heterogeneity scale.

Figures

Figures reproduced from arXiv: 2605.20571 by Jingzi Huang, Maarten van Reeuwijk.

Figure 1
Figure 1. Figure 1: Definition sketch of domain and filter A. (a) plan view, (b) elevation view. The domain of Ω is comprised of a fluid subdomain Ωf (in white), a solid subdomain Ωb (in grey), and a fluid-solid interface ∂Ωf (solid black lines). The building-surface 3-D normal vectors N point into the fluid domain, and the decomposition is shown in (c). The 2-D square filter A with averaging length L is shown in red. the val… view at source ↗
Figure 2
Figure 2. Figure 2: (a) A satellite map of the Bristol campus from Google Maps, with the footprints of the buildings, and a dashed circle highlights the circular building configuration occupying the central area. (b) The morphology of the ‘circular’ case (CC) directly extracted from the map. (c) The morphology of the ‘square’ case (SC), filling the corners in (b) with additional buildings. Flow Laboratory (EnFlo) (Bi et al. 2… view at source ↗
Figure 3
Figure 3. Figure 3: Superficially plane-averaged flow statistics for both cases. (a) streamwise velocity, overlaid with logarithmic profiles; (b) kinematic turbulent shear stress and dispersive stress. The height is normalised by the mean building height hm and the dashed horizontal line marks the maximum building height, i.e., the canopy top limit. Well above the canopy region, [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The plane view of the time-averaged wind speed (u, v) overlaid with the velocity vectors (a,b) at the pedestrian level (z = 1.5 m), (c,d) at the mean building height (z = hm), the grey area represents the buildings where there is no fluid. (e, f) The surface pressure of the solid phase. (g, h) The surface shear stress is induced by the streamwise velocity. The left column presents the CC, while the right c… view at source ↗
Figure 5
Figure 5. Figure 5: The vertical profile of plane-averaged (a, b) drag force and its components (unit m/s2), (d, e) cumulative stresses (unit m2/s2), the height is normalised by the maximum building height marked as the horizontal dotted lines. (g, h) The cumulative drag stress τD against the normalised frontal area ζ, overlaid with the parameterisation Eq. (10) at L = ∞; (i, j) The cumulative turbulent stress ⟨τz⟩ against th… view at source ↗
Figure 6
Figure 6. Figure 6: The streamwise velocity field u at various averaging lengths at the pedestrian level z = 1.5m for CC (the first column) and SC (the second column): (a, b) averaging lengths L = 8m, (c, d) L = 64 m, (f, g) L = 512 m; (e, h) are the variances of the streamwise velocity at pedestrian level for CC and SC, respectively. unresolved variance σ 2 U increases, consistent with the coarse￾graining process in [PITH_F… view at source ↗
Figure 7
Figure 7. Figure 7: The contour plot of the interaction covariances of streamwise velocity against the averaging length L and the height z, for CC (a) and SC (b), respectively [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: The coarse-grained kinematic surface stress τ0;L for CC (top panel), and SC (bottom panel), respectively, with the averaging lengths (a, d) L = 8 m, (b, e) L = 64 m and (c, f) L = 512 m. Three high-drag zones are labelled in (b, e). nearly homogeneous across the entire plane, with magnitudes converging to the global average τ0. Since the kinematic surface stress τ0;L represents the drag acting on buildings… view at source ↗
Figure 9
Figure 9. Figure 9: Spatial-averaged kinematic surface stress τ0;L against the plane-area index λp;L and front-area index λf;L for (a) CC, (b) SC, respectively. The data of averaging lengths L = 256, 512 m are used in both figures. transport is considered, and horizontal transport is assumed to be negligible. The broad distribution of τ0;L across λp;L and λf;L confirms that the constants in the parameterisation (Eq. (10)) ind… view at source ↗
Figure 10
Figure 10. Figure 10: The distribution density of normalised local cumulative drag stress τD;L/τ0;L against the local scaled frontal area ζL at different averaging lengths. (a, d) L = 128 m, (b, e) L = 256 m, (c, f) L = 512 m, for CC (top panel) and SC (bottom panel), respectively. The colour represents the statistical probability density function value of the data appearing at each location. The solid line represents the para… view at source ↗
Figure 11
Figure 11. Figure 11: The distribution density of normalised local cumulative turbulent stress ⟨τz;L⟩ / ⟨τz;L⟩max against the local scaled frontal area ζL at different averaging lengths. (a, d) L = 128 m, (b, e) L = 256 m, (c, f) L = 512 m, for CC (top panel) and SC (bottom panel), respectively. The solid line represents the parametrisation Eq. (11) with modified constants A = 0.89, B = 1.82. The coefficient of determination R… view at source ↗
read the original abstract

As Numerical Weather Prediction (NWP) models approach hectometric resolution, they increasingly enter a regime where urban heterogeneity is only partially resolved and the assumptions underlying conventional urban canopy models (UCMs) become questionable. To address this scale gap, we apply a multi-scale coarse-graining framework (van Reeuwijk and Huang 2025, Boundary-Layer Meteorology) to building-resolving Large-Eddy Simulations (LES) of the University of Bristol campus. Two related morphologies are considered: an original layout with large open-space contrasts and a modified configuration with these regions infilled. By systematically filtering the LES fields, we quantify how flow heterogeneity evolves with resolution and identify a characteristic urban length scale at which resolved and unresolved variability are comparable. This scale is strongly morphology-dependent, with values of about 256 m for the original layout and 64 m for the modified case, showing that neighbourhood-scale organisation can remain important at resolutions relevant to next-generation NWP. We then perform an a priori assessment of distributed drag and turbulent-stress parameterisations. Parameterisations derived from idealised geometries perform reasonably well only at sufficiently coarse resolutions, where horizontal transport is negligible and the flow appears approximately homogeneous. At finer resolutions, their fidelity degrades rapidly because of increasing heterogeneity and filter-to-filter variability in morphology, with stronger limitations in realistic layouts than in idealised cuboid arrays. Overall, the results show that the applicability of urban parameterisations depends critically on the relationship between model resolution and a morphology-dependent heterogeneity scale, providing a systematic route for developing scale-aware UCMs for high-resolution NWP.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript applies a multi-scale coarse-graining framework to building-resolving LES of two University of Bristol campus morphologies (original layout with large open-space contrasts and a modified infilled configuration). It identifies morphology-dependent heterogeneity scales (~256 m original, ~64 m infilled) at which resolved and unresolved flow variability become comparable, then performs an a priori assessment showing that distributed drag and turbulent-stress parameterisations derived from idealised geometries perform adequately only at coarse resolutions where the flow appears homogeneous, with rapid degradation at finer scales due to increasing heterogeneity and filter variability. The central claim is that UCM applicability depends critically on the relationship between model resolution and this morphology-dependent scale, providing a systematic route for scale-aware UCMs in hectometric NWP.

Significance. If the results hold, the work supplies a quantitative, morphology-aware criterion for the validity of conventional UCMs as NWP models enter the hectometric regime. The explicit comparison between realistic campus layouts and the demonstration that neighbourhood-scale organisation can persist at resolutions relevant to next-generation NWP models offers a practical framework for future parameterisation development. The a priori tests on real morphologies rather than idealised arrays add concrete evidence of when and why standard approaches fail.

major comments (2)
  1. [Methods (multi-scale coarse-graining) and Results (heterogeneity scale identification)] The central claim that the extracted heterogeneity scales correctly indicate when UCM assumptions break for NWP rests on the multi-scale coarse-graining framework producing an effective filter equivalent to an actual NWP grid at the same nominal resolution. The manuscript applies the van Reeuwijk and Huang (2025) operator to LES fields but provides no direct comparison to the resolved/unresolved partition that would arise inside a finite-volume or finite-difference NWP solver with its specific advection, diffusion, and turbulence closures. This equivalence is load-bearing for the a priori assessment and the claimed route to scale-aware UCMs.
  2. [Results (a priori assessment of distributed drag and turbulent-stress parameterisations)] In the parameterization assessment, the claim that idealised-geometry models perform 'reasonably well only at sufficiently coarse resolutions' is supported by the observed degradation, yet the manuscript does not report quantitative error metrics, confidence intervals, or sensitivity to filter choice across the 256 m / 64 m scales. Without these, the strength of the conclusion that fidelity degrades 'rapidly' because of heterogeneity remains difficult to judge.
minor comments (2)
  1. [Abstract] The abstract introduces 'filter-to-filter variability in morphology' without defining the set of filters employed; a short clarification or reference to the relevant methods subsection would improve readability.
  2. [Methods] Notation for the heterogeneity scale (e.g., how it is extracted from the filtered fields) should be introduced explicitly in the main text rather than relying solely on the 2025 framework citation.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive review and positive assessment of the work's significance. We address each major comment below and indicate the revisions we will make.

read point-by-point responses

  1. Referee: The central claim that the extracted heterogeneity scales correctly indicate when UCM assumptions break for NWP rests on the multi-scale coarse-graining framework producing an effective filter equivalent to an actual NWP grid at the same nominal resolution. The manuscript applies the van Reeuwijk and Huang (2025) operator to LES fields but provides no direct comparison to the resolved/unresolved partition that would arise inside a finite-volume or finite-difference NWP solver with its specific advection, diffusion, and turbulence closures. This equivalence is load-bearing for the a priori assessment and the claimed route to scale-aware UCMs.

    Authors: We agree that the equivalence between the applied coarse-graining operator and the implicit filtering in a specific NWP discretization is an important assumption. The van Reeuwijk and Huang (2025) operator is formulated to represent volume averaging over a grid cell, but we acknowledge that advection schemes, subgrid closures, and numerical dissipation in operational NWP models could alter the resolved/unresolved partition. In the revised manuscript we will add an explicit discussion in the Methods section on the operator's assumptions and limitations, and we will qualify the claims to note that the framework provides a systematic a priori route rather than an exact surrogate for any particular NWP solver. A direct head-to-head comparison with an NWP code would require additional simulations beyond the present scope and is suggested as future work. revision: partial

  2. Referee: In the parameterization assessment, the claim that idealised-geometry models perform 'reasonably well only at sufficiently coarse resolutions' is supported by the observed degradation, yet the manuscript does not report quantitative error metrics, confidence intervals, or sensitivity to filter choice across the 256 m / 64 m scales. Without these, the strength of the conclusion that fidelity degrades 'rapidly' because of heterogeneity remains difficult to judge.

    Authors: We thank the referee for highlighting this gap. The present text describes the degradation qualitatively. We will revise the Results section to report quantitative error metrics (e.g., normalized mean absolute error and root-mean-square error for drag and stress predictions) at each nominal resolution. We will also include sensitivity tests to filter width and location, together with error bars derived from the variability across multiple filter positions to quantify uncertainty. revision: yes

Circularity Check

1 steps flagged

Central heterogeneity scale extraction rests on authors' 2025 self-cited coarse-graining framework

specific steps
  1. self citation load bearing [Abstract]
    "we apply a multi-scale coarse-graining framework (van Reeuwijk and Huang 2025, Boundary-Layer Meteorology) to building-resolving Large-Eddy Simulations (LES) of the University of Bristol campus. ... identify a characteristic urban length scale at which resolved and unresolved variability are comparable. This scale is strongly morphology-dependent, with values of about 256 m for the original layout and 64 m for the modified case"

    The morphology-dependent heterogeneity scale (central to the claim that UCM applicability depends on resolution vs. this scale) is obtained directly by applying the cited multi-scale coarse-graining framework. Because the framework is from the same authors' prior work and is not independently re-derived or validated against NWP numerics in this manuscript, the key result on scale-aware UCMs reduces to load-bearing self-citation.

full rationale

The paper applies the multi-scale coarse-graining framework from van Reeuwijk and Huang (2025) to LES data to extract morphology-dependent heterogeneity scales (~256 m and ~64 m) and assess UCM parameterisation fidelity. This framework is not re-derived here, so the load-bearing step identifying when resolved/unresolved variability becomes comparable depends on the prior self-citation. However, the present work adds independent content through new LES cases (Bristol campus, infilled vs original), quantitative filter-to-filter variability results, and a priori parameterisation tests. No self-definitional, fitted-input, or ansatz-smuggling circularity occurs; the derivation chain remains externally falsifiable via the LES data itself. This yields a low-to-moderate circularity score.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only; free parameters and axioms cannot be audited in detail. The multi-scale framework is imported from the authors' 2025 paper; no new invented entities are stated.

axioms (1)
  • domain assumption The multi-scale coarse-graining framework of van Reeuwijk and Huang 2025 correctly represents sub-grid variability for urban flows.
    Invoked in abstract to justify the filtering procedure.

pith-pipeline@v0.9.0 · 5815 in / 1245 out tokens · 26236 ms · 2026-05-21T02:57:52.971902+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

  • IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    By systematically filtering the LES fields, we quantify how flow heterogeneity evolves with resolution and identify a characteristic urban length scale ℓ at which resolved and unresolved variability are comparable. This scale is strongly morphology-dependent, with values of about 256 m for the original layout and 64 m for the modified case

  • IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    the multi-resolution planar-averaging framework of van Reeuwijk and Huang (2025b). The framework uses convolution filters to derive coarse-grained fields

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

28 extracted references · 28 canonical work pages

  1. [1]

    Auerswald T, Klippel K, Thomas TG,et al.2024

    LES of wind-induced pressures and flow structures around tandem buildings.Journal of Wind Engineering and Industrial Aerodynamics269: 106 304, doi: 10.1016/j.jweia.2025.106304. Auerswald T, Klippel K, Thomas TG,et al.2024. Effect of flow variability on dispersion of continuous and puff releases in a regular street network.Boundary-Layer Meteorology190: 20...

    work page doi:10.1016/j.jweia.2025.106304 2025

  2. [2]

    org/view/journals/mwre/139/12/mwr-d-10-05013.1.xml

    Operational convective-scale numerical weather prediction with the cosmo model: Description and sensitivities.Monthly Weather Review139(12): 3887 – 3905, doi: 10.1175/MWR-D-10-05013.1, URLhttps://journals.ametsoc. org/view/journals/mwre/139/12/mwr-d-10-05013.1.xml. Barlow JF

    work page doi:10.1175/mwr-d-10-05013.1

  3. [3]

    com/science/article/pii/S2212095514000558

    Progress in observing and modelling the urban boundary layer.Urban Climate10: 216–240, doi:https://doi.org/ 10.1016/j.uclim.2014.03.011, URLhttps://www.sciencedirect. com/science/article/pii/S2212095514000558. ICUC8: The 8th International Conference on Urban Climate and the 10th Symposium on the Urban Environment. Bi D, Carpentieri M, Placidi M, Robins A

    work page doi:10.1016/j.uclim.2014.03.011 2014

  4. [4]

    Bryan GH, Wyngaard JC, Fritsch JM

    Turbulence characteristics across a range of idealized urban canopy geometries.Boundary-Layer Meteorology 182: 275–307, doi:10.1007/s10546-021-00658-6, URLhttps:// centaur.reading.ac.uk/99701/. Bryan GH, Wyngaard JC, Fritsch JM

    work page doi:10.1007/s10546-021-00658-6

  5. [5]

    Resolution requirements for the simulation of deep moist convection. Monthly Weather Review131(10): 2394 – 2416, doi: 10.1175/1520-0493(2003)131⟨2394:RRFTSO⟩2.0.CO;2, URL https://journals.ametsoc.org/view/journals/mwre/131/10/ 1520-0493_2003_131_2394_rrftso_2.0.co_2.xml. Cheng WC, Yang Y

    work page doi:10.1175/1520-0493(2003)131 2003

  6. [6]

    Atmosphere15(9), URLhttps://www.mdpi.com/2073-4433/15/ 9/1089

    Comparing large-eddy simulation and gaussian plume model to sensor measurements of an urban smoke plume. Atmosphere15(9), URLhttps://www.mdpi.com/2073-4433/15/ 9/1089. Coburn M, Xie ZT, Herring SJ

    work page 2073

  7. [7]

    Coceal O, Belcher SE

    Numerical simulations of boundary-layer airflow over pitched-roof buildings.Boundary-Layer Meteorology185: 415–442, doi: 10.1007/s10546-022-00738-1, URLhttps://doi.org/10.1007/ s10546-022-00738-1. Coceal O, Belcher SE

    work page doi:10.1007/s10546-022-00738-1

  8. [8]

    03.40, URLhttps://rmets.onlinelibrary.wiley.com/doi/abs/ 10.1256/qj.03.40

    A canopy model of mean winds through urban areas.Quarterly Journal of the Royal Meteorological Society130(599): 1349–1372, doi:https://doi.org/10.1256/qj. 03.40, URLhttps://rmets.onlinelibrary.wiley.com/doi/abs/ 10.1256/qj.03.40. Coceal O, Thomas TG, Castro IP, Belcher SE

    work page doi:10.1256/qj

  9. [9]

    Boundary-Layer Meteorology121(3): 491–519

    Mean flow and turbulence statistics over groups of urban-like cubical obstacles. Boundary-Layer Meteorology121(3): 491–519. Ding SS, Placidi M, Carpentieri M,et al.2023. Neutrally- and stably-stratified boundary layers adjustments to a step change in surface roughness.Experiments in Fluids64: 86, doi: 10.1007/s00348-023-03626-7, URLhttps://doi.org/10.1007...

    work page doi:10.1007/s00348-023-03626-7 2023

  10. [10]

    Giometto M, Christen A, Meneveau C, Fang J, Krafczyk M, Parlange M

    Turbulence in plant canopies.Annual Review of Fluid Mechanics32(Volume 32, 2000): 519– 571, doi:https://doi.org/10.1146/annurev.fluid.32.1.519, URL ©2025 Royal Meteorological SocietyPrepared usingqjrms4.cls 21 https://www.annualreviews.org/content/journals/10.1146/ annurev.fluid.32.1.519. Giometto M, Christen A, Meneveau C, Fang J, Krafczyk M, Parlange M

    work page doi:10.1146/annurev.fluid.32.1.519 2000

  11. [11]

    1007/s10546-004-7956-x

    A simple energy balance model for regular building arrays.Boundary-Layer Meteorology116(3): 423– 443, doi:10.1007/s10546-004-7956-x, URLhttps://doi.org/10. 1007/s10546-004-7956-x. Kusaka H, Ikeda R, Sato T, Iizuka S, Boku T

    work page doi:10.1007/s10546-004-7956-x

  12. [12]

    A simple single- layer urban canopy model for atmospheric models: Comparison with multi-layer and slab models.Boundary-Layer Meteorology 101(3): 329–358, doi:10.1023/A:1019207923078, URLhttps:// doi.org/10.1023/A:1019207923078. Lean HW, Theeuwes NE, Baldauf M, Barkmeijer J, Bessardon G, Blunn L, Bojarova J, Boutle IA, Clark PA, Demuzere M, Dueben P, Frogn...

    work page doi:10.1023/a:1019207923078

  13. [13]

    wiley.com/doi/abs/10.1002/qj.4858

    The hectometric modelling challenge: Gaps in the current state of the art and ways forward towards the implementation of 100- m scale weather and climate models.Quarterly Journal of the Royal Meteorological Society150(765): 4671–4708, doi:https:// doi.org/10.1002/qj.4858, URLhttps://rmets.onlinelibrary. wiley.com/doi/abs/10.1002/qj.4858. Lipson MJ, Grimmo...

    work page doi:10.1002/qj.4858

  14. [14]

    wiley.com/doi/abs/10.1002/qj.4589

    Evaluation of 30 urban land surface models in the Urban- PLUMBER project: Phase 1 results.Quarterly Journal of the Royal Meteorological Society150(758): 126–169, doi:https:// doi.org/10.1002/qj.4589, URLhttps://rmets.onlinelibrary. wiley.com/doi/abs/10.1002/qj.4589. Lu J, Nazarian N, Hart MA, Krayenhoff ES, Martilli A

    work page doi:10.1002/qj.4589

  15. [15]

    Ludwig S, Plein M, Barlow JF, Escobar-Ruiz V, Grimmond S, Matthews J, van Reeuwijk M, Zeeman M, Christen A

    A one-dimensional urban flow model with an eddy-diffusivity mass- flux (EDMF) scheme and refined turbulent transport (MLUCM v3.0).Geoscientific Model Development17(7): 2525–2545, doi: 10.5194/gmd-17-2525-2024, URLhttps://gmd.copernicus.org/ articles/17/2525/2024/. Ludwig S, Plein M, Barlow JF, Escobar-Ruiz V, Grimmond S, Matthews J, van Reeuwijk M, Zeeman...

    work page doi:10.5194/gmd-17-2525-2024 2024

  16. [16]

    doi:10.5281/zenodo.15537606, URLhttps: //doi.org/10.5281/zenodo.15537606

    Street-level weather station network in Bristol, UK: Station documentation (1.0). doi:10.5281/zenodo.15537606, URLhttps: //doi.org/10.5281/zenodo.15537606. Macdonald R, Griffiths R, Hall D

    work page doi:10.5281/zenodo.15537606

  17. [17]

    An urban surface exchange parameterisation for mesoscale models.Boundary- Layer Meteorology104: 261–304, doi:10.1023/A:1016099921195. Masson V, Heldens W, Bocher E, Bonhomme M, Bucher B, Burmeister C, de Munck C, Esch T, Hidalgo J, Kanani- S¨ uhring F, Kwok YT, Lemonsu A, L´ evy JP, Maronga B, Pavlik D, Petit G, See L, Schoetter R, Tornay N, Votsis A, Zeidler J

    work page doi:10.1023/a:1016099921195

  18. [18]

    Urban Climate31: 100 536, doi:https://doi.org/10.1016/j.uclim

    City-descriptive input data for urban climate models: Model requirements, data sources and challenges. Urban Climate31: 100 536, doi:https://doi.org/10.1016/j.uclim. 2019.100536, URLhttps://www.sciencedirect.com/science/ article/pii/S2212095518303663. Nazarian N, Krayenhoff ES, Martilli A

    work page doi:10.1016/j.uclim 2019

  19. [19]

    A one- dimensional model of turbulent flow through “urban” canopies (MLUCM v2.0): updates based on large-eddy simulation.Geoscientific Model Development13(3): 937–953, doi:10.5194/gmd-13-937-2020, URLhttps: //gmd.copernicus.org/articles/13/937/2020/. Nepf HM

    work page doi:10.5194/gmd-13-937-2020 2020

  20. [20]

    Owens SO, Majumdar D, Wilson CE, Bartholomew P, van Reeuwijk M

    Flow and transport in regions with aquatic vegetation.Annual Review of Fluid Mechanics 44(Volume 44, 2012): 123–142, doi:https://doi.org/ 10.1146/annurev-fluid-120710-101048, URLhttps: //www.annualreviews.org/content/journals/10.1146/ annurev-fluid-120710-101048. Owens SO, Majumdar D, Wilson CE, Bartholomew P, van Reeuwijk M

    work page doi:10.1146/annurev-fluid-120710-101048 2012

  21. [21]

    A conservative immersed boundary method for the multi-physics urban large-eddy simulation model uDALES v2.0.EGUsphere2024: 1–33, doi:10.5194/ egusphere-2024-96, URLhttps://egusphere.copernicus.org/ preprints/2024/egusphere-2024-96/. Pope SB. 2000.Turbulent flows. Cambridge University Press, doi: 10.1017/CBO9780511840531. ©2025 Royal Meteorological Society...

    work page doi:10.1017/cbo9780511840531 2024

  22. [22]

    Schlichting H, Gersten K

    A Dynamic Urban Canopy Parameterization for Mesoscale Models Based on Computational Fluid Dynamics Reynolds-Averaged Navier-Stokes Microscale Simulations.Boundary-Layer Meteorology137(3): 417–439, doi: 10.1007/s10546-010-9538-4. Schlichting H, Gersten K. 2017.Boundary-layer theory. Springer Berlin, Heidelberg, ISBN 978-3-662-52917-1, doi: 10.1007/978-3-66...

    work page doi:10.1007/s10546-010-9538-4 2017

  23. [23]

    Suter I, Grylls T, S¨ utzl BS, Owens SO, Wilson CE, van Reeuwijk M

    Local climate zones for urban temperature studies.Bulletin of the American Meteorological Society93(12): 1879 – 1900, doi:10.1175/BAMS-D-11-00019.1, URLhttps://journals.ametsoc.org/view/journals/bams/93/ 12/bams-d-11-00019.1.xml. Suter I, Grylls T, S¨ utzl BS, Owens SO, Wilson CE, van Reeuwijk M

    work page doi:10.1175/bams-d-11-00019.1 1900

  24. [24]

    copernicus.org/articles/15/5309/2022/

    uDALES 1.0: a large-eddy simulation model for urban environments.Geoscientific Model Development15(13): 5309–5335, doi:10.5194/gmd-15-5309-2022, URLhttps://gmd. copernicus.org/articles/15/5309/2022/. S¨ utzl BS, Rooney GG, Finnenkoetter A, Bohnenstengel SI, Grimmond S, van Reeuwijk M

    work page doi:10.5194/gmd-15-5309-2022 2022

  25. [25]

    Meteorological Applications20(4): 417–426, doi:https://doi.org/ 10.1002/met.1300, URLhttps://rmets.onlinelibrary.wiley

    The benefits of the Met Office variable resolution NWP model for forecasting convection. Meteorological Applications20(4): 417–426, doi:https://doi.org/ 10.1002/met.1300, URLhttps://rmets.onlinelibrary.wiley. com/doi/abs/10.1002/met.1300. Uno I, Cai XM, Steyn DG, Emori S

    work page doi:10.1002/met.1300

  26. [26]

    Huennefeld, W

    A simple extension of the louis method for rough surface layer modelling.Bound.-Layer Meteorol.76: 395–409. van Reeuwijk M, Huang J. 2025a. Dataset for ”multi-scale analysis of flow over heterogeneous urban environments” . doi:10. 5281/zenodo.17102845, URLhttps://doi.org/10.5281/zenodo. 17102845. van Reeuwijk M, Huang J. 2025b. Multi-scale Analysis of Flo...

    work page doi:10.5281/zenodo

  27. [27]

    Xie ZT, Coceal O, Castro IP

    An eddy-viscosity subgrid-scale model for turbulent shear flow: Algebraic theory and applications.Physics of Fluids16(10): 3670–3681, doi:10.1063/1.1785131, URLhttps: //doi.org/10.1063/1.1785131. Xie ZT, Coceal O, Castro IP

    work page doi:10.1063/1.1785131

  28. [28]

    ©2025 Royal Meteorological SocietyPrepared usingqjrms4.cls

    Urban neighbourhood classification and multi-scale heterogeneity analysis of Greater London.Environment and Planning B: Urban Analytics and City Science50(6): 1534–1558, doi:10.1177/23998083221140890. ©2025 Royal Meteorological SocietyPrepared usingqjrms4.cls

    work page doi:10.1177/23998083221140890 2025