arxiv: 2604.24635 · v1 · submitted 2026-04-27 · ⚛️ physics.soc-ph

Recognition: unknown

No one likes it hot, but hotter cities adjust by staying active later

Andrew Renninger , Till Koebe , Ingmar Weber

Authors on Pith no claims yet

Pith reviewed 2026-05-07 17:46 UTC · model grok-4.3

classification ⚛️ physics.soc-ph

keywords urban heat adaptationactivity timing shiftsbimodality indexevening activityclimate resiliencepoints of interest datadaily activity profilesheat stress

0 comments

The pith

Hot days reduce urban activity overall but shift it later into the evening, with stronger effects in historically hotter cities.

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

The paper uses activity data from points of interest across 20 cities in temperate, tropical, and arid climates to examine how extreme heat changes daily patterns. Hot days suppress total activity but also move it away from midday toward later hours. This rescheduling proves stronger in cities with hotter histories, producing smaller net losses and more evening substitution. The authors introduce a Bactrian index to track whether activity profiles show one daytime peak or two peaks including evening activity. Arid cities already show evening peaks while others adopt them only on hot days, indicating adaptation occurs mainly through timing changes rather than outright avoidance.

Core claim

Hot days reduce activity overall while shifting it away from midday and toward later hours. This rescheduling is substantially stronger in historically hotter cities, which exhibit smaller losses and larger evening substitution. Arid desert cities like Doha, Amman, and Kuwait City are more Bactrian in level, but cities like Milan become Bactrian on hot days. Adaptation to heat operates less through avoiding activity altogether than through moving it to cooler hours.

What carries the argument

Bactrian index of bimodality, which quantifies how much a city's daily activity profile consists of one daytime hump versus two humps with a distinct evening peak.

If this is right

Cooler cities gain usable adaptation channels by moving activities to evening hours during heat.
Warmer cities face limits because rising evening temperatures may eventually make those hours intolerable for activity.
Cities can reduce total activity losses by encouraging or enabling evening substitution rather than only midday avoidance.
Activity profiles in non-arid cities shift toward bimodality specifically on hot days.

Where Pith is reading between the lines

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

Urban planners could prioritize evening public spaces, lighting, and transport options as heat-adaptation measures.
The pattern suggests testing whether similar rescheduling appears in non-urban settings or with different data types such as mobility traces.
As global temperatures rise, the effectiveness of evening substitution may decline in more cities, requiring new cooling strategies.

Load-bearing premise

Observed shifts in activity timing are caused by heat adaptation rather than other correlated factors such as work schedules, cultural differences, or data collection biases across the 20 cities.

What would settle it

Data showing no difference in evening rescheduling between cities with similar current temperatures but different historical heat exposure, or activity shifts that disappear after controlling for work hours and cultural variables, would falsify the adaptation claim.

Figures

Figures reproduced from arXiv: 2604.24635 by Andrew Renninger, Ingmar Weber, Till Koebe.

**Figure 1.** Figure 1: Hotter cities stay active later. A Grouped activity profiles on locally hot and mild days, showing that cooler climates have a rise through the day and fall in the evening, while hotter ones have a second rise in the evening when temperatures fall; though subtle, we also see that hotter days show depressed activity in the afternoon and expanded activity in the evenings—especially in arid cities. B Loss or … view at source ↗

**Figure 2.** Figure 2: Hotter cities have multiple active periods. A Our “Bactrian” index on locally hot days, ordered from highest to lowest, showing that hot, arid cities are also the most Bactrian—with multiple active periods; this index is defined as the amplitude of the second daily harmonic relative to the first. B Bactrian index on mild days and hot days for each city: hot and arid cities are more Bactrian on all days but… view at source ↗

read the original abstract

Extreme heat suppresses urban activity, but its effects need not be uniform across climates or across the day. Using data on activity at points of interest in 20 cities spanning temperate, tropical, and arid environments, we show that hot days reduce activity overall while shifting it away from midday and toward later hours. This rescheduling is substantially stronger in historically hotter cities, which exhibit smaller losses and larger evening substitution. To understand these changes, we introduce a Bactrian index of bimodality, which measures the degree to which a city's daily activity profile has one hump or two - one during the day and another during the evening. Arid desert cities like Doha, Amman, and Kuwait City are more Bactrian in level, but cities like Milan become Bactrian on hot days. Together, our results suggest that adaptation to heat in cities operates less through avoiding activity altogether than through moving it to cooler hours. This provides channels for adaptation in cooler cities, but it also suggests limits to adaptation in warmer ones: as evenings become warmer, these too may become intolerable.

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.

Desk Editor's Note private letter to a colleague

Hotter cities show bigger evening activity shifts on hot days via a new bimodality index, but the adaptation claim rests on weak identification.

read the letter

The main thing to know is that this paper documents stronger rescheduling of daily activity toward evenings on hot days in cities with hotter historical climates, using POI data across 20 places and a simple new index for profile bimodality. The pattern is that overall activity drops but the drop is smaller and the evening bump larger where heat is more common, with arid cities already bimodal and some temperate ones shifting that way only on hot days. That comparative result and the index are the clearest additions over prior single-city heat-mobility work. The data choice lets them span climate zones, which is useful for spotting the variation. The abstract frames it as adaptation through timing rather than avoidance, which is a reasonable reading of the descriptive patterns. The soft spot is exactly the one in the stress-test note. Hotter cities differ systematically in work schedules, cultural norms, economic mix, and likely in how the POI data samples or biases toward certain activities. The central claim needs the interaction between baseline climate and same-day heat response to reflect adaptation, not those other factors. The abstract supplies no regression details, fixed effects, covariates, or robustness checks, so it is impossible to judge whether the result survives that. If the full paper has city-level controls or matching, the finding strengthens; otherwise it stays observational. This is the sort of paper that would interest researchers on urban climate adaptation or mobility data. A reader focused on empirical patterns in activity timing would take something useful from the index and the cross-city comparison. Anyone needing clean identification of adaptation mechanisms would want more. I would send it to peer review so the methods and potential confounders get proper scrutiny rather than desk-rejecting it outright.

Referee Report

3 major / 2 minor

Summary. The paper uses activity data at points of interest across 20 cities in temperate, tropical, and arid climates to claim that hot days suppress overall urban activity while shifting it away from midday toward evening hours; this rescheduling is stronger in historically hotter cities, which show smaller net losses. The authors introduce a 'Bactrian index' of daily activity bimodality to quantify the shift from unimodal to bimodal profiles on hot days and interpret the pattern as evidence of behavioral adaptation via timing rather than avoidance.

Significance. If the central patterns survive proper identification checks, the work would provide concrete, cross-climate evidence that cities adapt to heat by rescheduling rather than curtailing activity, with direct implications for projecting activity under future warming and for designing heat-resilient urban schedules.

major comments (3)

[Results and methods] Results and methods sections: The manuscript reports consistent patterns across 20 cities but supplies no information on data sources, how activity is measured, sample sizes, definition of 'hot days,' statistical models, error estimation, or robustness checks. Without these, the claim that hotter cities exhibit larger evening substitution cannot be evaluated for confounding or replicability.
[Results] Results section (cross-city comparison): The attribution of smaller losses and larger evening shifts in historically hotter cities to heat adaptation is not supported by any controls for city-level confounders such as work schedules (e.g., siestas), cultural norms, economic composition, or systematic differences in POI data coverage and sampling bias. The interaction between historical temperature and same-day response therefore risks omitted-variable bias.
[Bactrian index section] Section introducing the Bactrian index: The index is described as measuring bimodality but is not given an explicit mathematical definition, normalization, or validation against alternative bimodality metrics; its use to support the adaptation interpretation therefore lacks a clear quantitative foundation.

minor comments (2)

[Abstract] Abstract: The term 'Bactrian index' is introduced without a brief definition or reference to its later formalization.
[Figures] Figure captions: Captions should specify the exact activity metric plotted, the temperature threshold used for 'hot days,' and the number of observations per city.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the thoughtful and constructive report. We address each major comment below and will revise the manuscript to incorporate additional details, clarifications, and robustness material where the comments correctly identify gaps in the current version.

read point-by-point responses

Referee: [Results and methods] Results and methods sections: The manuscript reports consistent patterns across 20 cities but supplies no information on data sources, how activity is measured, sample sizes, definition of 'hot days,' statistical models, error estimation, or robustness checks. Without these, the claim that hotter cities exhibit larger evening substitution cannot be evaluated for confounding or replicability.

Authors: We agree that the submitted version lacks sufficient methodological detail. In the revised manuscript we will expand the Methods section to specify: the POI activity data source and its coverage; the precise construction of the activity measure (normalized visit counts); total sample sizes (POI-day observations per city); the definition of hot days (maximum temperature above the city-specific 90th percentile of the 2000–2020 distribution); the regression specification (city and date fixed effects with an interaction between daily temperature deviation and historical mean temperature); standard-error clustering; and the full set of robustness checks (alternative thresholds, weather controls, and subsample analyses). These additions will allow direct evaluation of replicability and potential confounding. revision: yes
Referee: [Results] Results section (cross-city comparison): The attribution of smaller losses and larger evening shifts in historically hotter cities to heat adaptation is not supported by any controls for city-level confounders such as work schedules (e.g., siestas), cultural norms, economic composition, or systematic differences in POI data coverage and sampling bias. The interaction between historical temperature and same-day response therefore risks omitted-variable bias.

Authors: We acknowledge the risk of omitted-variable bias in the cross-city moderator analysis. Our core identification uses within-city temperature variation, but we will add a dedicated subsection that (i) discusses city-level confounders including siestas and other cultural scheduling practices (which we view as endogenous to long-run adaptation), (ii) reports robustness checks that control for available city characteristics such as GDP per capita and sectoral composition, and (iii) examines POI sampling density and coverage rates across cities to assess differential bias. While exhaustive controls for all unmeasured cultural factors are not feasible, the consistency of the evening-shift pattern across three distinct climate zones provides supporting evidence for behavioral adaptation. We will present these checks and a balanced discussion of limitations. revision: partial
Referee: [Bactrian index section] Section introducing the Bactrian index: The index is described as measuring bimodality but is not given an explicit mathematical definition, normalization, or validation against alternative bimodality metrics; its use to support the adaptation interpretation therefore lacks a clear quantitative foundation.

Authors: We agree that the Bactrian index requires a formal definition. In the revision we will insert the explicit formula: B = (A_evening / A_midday) / (A_evening + A_midday), where A_peak denotes the height of the smoothed activity density at the respective time window, yielding a normalized index in [0,1]. We will also report validation exercises showing that the index correlates strongly with the Hartigan dip test statistic and with the excess kurtosis of the daily activity distribution. These additions will supply the quantitative grounding for interpreting increases in the index as evidence of heat-induced rescheduling. revision: yes

Circularity Check

0 steps flagged

No significant circularity in observational analysis

full rationale

The paper is entirely observational and data-driven, presenting empirical comparisons of POI activity levels and timing across 20 cities on hot versus normal days. It introduces the Bactrian index purely as a descriptive statistic quantifying bimodality in daily activity profiles (one daytime hump versus two humps including evening), without any fitted parameters, predictions, equations, or derivations. No self-citations, ansatzes, or uniqueness claims are invoked as load-bearing steps, and the central findings on adaptation via rescheduling rest on direct data contrasts rather than any reduction to inputs by construction. This is a standard non-circular empirical study.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 1 invented entities

The abstract introduces one new descriptive tool (Bactrian index) but states no explicit free parameters, mathematical axioms, or postulated physical entities; analysis rests on standard assumptions that POI activity data proxy human behavior and that city climate history is an exogenous grouping variable.

invented entities (1)

Bactrian index no independent evidence
purpose: Quantifies the degree to which a city's daily activity profile is unimodal or bimodal (day hump plus evening hump)
Newly defined in the paper to capture the observed shift from single to double peaks on hot days.

pith-pipeline@v0.9.0 · 5488 in / 1264 out tokens · 81341 ms · 2026-05-07T17:46:54.987512+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

29 extracted references · 26 canonical work pages

[1]

& Begley, C

Zhang, K., Chen, Y . & Begley, C. E. Impact of the 2011 heat wave on mortality and emergency department visits in houston, texas.Environ. Heal.14, 11, 10.1186/1476-069X-14-11 (2015)

work page doi:10.1186/1476-069x-14-11 2011
[2]

& Smith, W

Schaffer, A., Muscatello, D., Broome, R., Corbett, S. & Smith, W. Emergency department visits, ambulance calls, and mortality associated with an exceptional heat wave in sydney, australia, 2011: A time-series analysis.Environ. Heal.11, 3, 10.1186/1476-069X-11-3 (2012). 3.Perkins, S. E., Alexander, L. V . & Nairn, J. R. Increasing frequency, intensity and ...

work page doi:10.1186/1476-069x-11-3 2011
[3]

Perkins-Kirkpatrick, S. E. & Lewis, S. C. Increasing trends in regional heatwaves.Nat. Commun.11, 3357, 10.1038/ s41467-020-16970-7 (2020)

2020
[4]

L.et al.Climate change adaptation to extreme heat: A global systematic review of implemented action

Turek-Hankins, L. L.et al.Climate change adaptation to extreme heat: A global systematic review of implemented action. Oxf. Open Clim. Chang.1, kgab005, 10.1093/oxfclm/kgab005 (2021)

work page doi:10.1093/oxfclm/kgab005 2021
[5]

Á.et al.Human adaptation to heat in the context of climate change: A conceptual framework.Environ

Navas-Martín, M. Á.et al.Human adaptation to heat in the context of climate change: A conceptual framework.Environ. Res.252, 118803, 10.1016/j.envres.2024.118803 (2024)

work page doi:10.1016/j.envres.2024.118803 2024
[6]

Dell, M., Jones, B. F. & Olken, B. A. Temperature shocks and economic growth: Evidence from the last half century.Am. Econ. Journal: Macroecon.4, 66–95, 10.1257/mac.4.3.66 (2012)

work page doi:10.1257/mac.4.3.66 2012
[7]

Burke, M., Hsiang, S. M. & Miguel, E. Global non-linear effect of temperature on economic production.Nature527, 235–239, 10.1038/nature15725 (2015)

work page doi:10.1038/nature15725 2015
[8]

& Tewari, M

Somanathan, E., Somanathan, R., Sudarshan, A. & Tewari, M. The impact of temperature on productivity and labor supply: Evidence from indian manufacturing.J. Polit. Econ.129, 1797–1827, 10.1086/713733 (2021)

work page doi:10.1086/713733 2021
[9]

Zhang, P., Deschenes, O., Meng, K. C. & Zhang, J. Temperature effects on productivity and factor reallocation: Evidence from a half million chinese manufacturing plants.J. Environ. Econ. Manag.88, 1–17, 10.1016/j.jeem.2017.11.001 (2018)

work page doi:10.1016/j.jeem.2017.11.001 2017
[10]

J., Goodman, J., Hurwitz, M

Park, R. J., Goodman, J., Hurwitz, M. & Smith, J. Heat and learning.Am. Econ. Journal: Econ. Policy12, 306–339, 10.1257/pol.20180612 (2020)

work page doi:10.1257/pol.20180612 2020
[11]

& Nyshadham, A

Adhvaryu, A., Kala, N. & Nyshadham, A. The light and the heat: Productivity co-benefits of energy-saving technology. Rev. Econ. Stat.102, 779–792, 10.1162/rest_a_00886 (2020)

work page doi:10.1162/rest_a_00886 2020
[12]

M., Scovronick, N., Sera, F.et al.The burden of heat-related mortality attributable to recent human-induced climate change.Nat

Vicedo-Cabrera, A. M., Scovronick, N., Sera, F.et al.The burden of heat-related mortality attributable to recent human-induced climate change.Nat. Clim. Chang.11, 492–500, 10.1038/s41558-021-01058-x (2021)

work page doi:10.1038/s41558-021-01058-x 2021
[13]

Heal.5, e415–e425, 10.1016/S2542-5196(21)00081-4 (2021)

Zhao, Q., Guo, Y ., Ye, T., Gasparrini, A.et al.Global, regional, and national burden of mortality associated with non- optimal ambient temperatures from 2000 to 2019: A three-stage modelling study.The Lancet Planet. Heal.5, e415–e425, 10.1016/S2542-5196(21)00081-4 (2021). 15.Mazdiyasni, O.et al.Increasing probability of mortality during indian heat waves...

work page doi:10.1016/s2542-5196(21)00081-4 2000
[14]

& Guo, Y

Yin, Q., Wang, J., Ren, Z., Li, J. & Guo, Y . Mapping the increased minimum mortality temperatures in the context of global climate change.Nat. Commun.10, 4640, 10.1038/s41467-019-12663-y (2019)

work page doi:10.1038/s41467-019-12663-y 2019
[15]

& Neidell, M

Graff Zivin, J. & Neidell, M. Temperature and the allocation of time: Implications for climate change.J. Labor Econ.32, 1–26, 10.1086/671766 (2014)

work page doi:10.1086/671766 2014
[16]

Tian, H.et al.Unveiling community adaptations to extreme heat events using mobile phone location data.J. Environ. Manag.366, 121665, 10.1016/j.jenvman.2024.121665 (2024)

work page doi:10.1016/j.jenvman.2024.121665 2024
[17]

Batur, I.et al.Understanding how extreme heat impacts human activity-mobility and time use patterns.Transp. Res. Part D: Transp. Environ.136, 104431, 10.1016/j.trd.2024.104431 (2024)

work page doi:10.1016/j.trd.2024.104431 2024
[18]

Y ., Goggins, W

Ho, J. Y ., Goggins, W. B., Mo, P. K. H. & Chan, E. Y . Y . The effect of temperature on physical activity: An aggregated timeseries analysis of smartphone users in five major chinese cities.Int. J. Behav. Nutr. Phys. Activity19, 68, 10.1186/ s12966-022-01285-1 (2022)

2022
[19]

L., Gasparrini, A

Bell, M. L., Gasparrini, A. & Benjamin, G. C. Climate change, extreme heat, and health.New Engl. J. Medicine390, 1793–1801, 10.1056/NEJMra2210769 (2024)

work page doi:10.1056/nejmra2210769 2024
[20]

Temperature, worker productivity, and adaptation: Evidence from survey data production.Am

LoPalo, M. Temperature, worker productivity, and adaptation: Evidence from survey data production.Am. Econ. Journal: Appl. Econ.15, 192–229, 10.1257/app.20200547 (2023)

work page doi:10.1257/app.20200547 2023
[21]

A., Shindell, D., Tigchelaar, M., Zhang, Y

Parsons, L. A., Shindell, D., Tigchelaar, M., Zhang, Y . & Spector, J. T. Increased labor losses and decreased adaptation potential in a warmer world.Nat. Commun.12, 7286, 10.1038/s41467-021-27328-y (2021)

work page doi:10.1038/s41467-021-27328-y 2021
[22]

& Zheng, S

Fan, Y ., Wang, J., Obradovich, N. & Zheng, S. Intraday adaptation to extreme temperatures in outdoor activity.Sci. Reports 13, 473, 10.1038/s41598-022-26928-y (2023)

work page doi:10.1038/s41598-022-26928-y 2023
[23]

Ly, A., Davenport, F. V . & Diffenbaugh, N. S. Exploring the influence of summer temperature on human mobility during the COVID-19 pandemic in the san francisco bay area.GeoHealth7, e2022GH000772, 10.1029/2022GH000772 (2023)

work page doi:10.1029/2022gh000772 2023
[24]

& Magee, D

Seijas, A., Karunanethy, S. & Magee, D. Adaptive urban economies: Evidence of intra-day temporal behavioural adaptation to extreme heat in australian cities.npj Urban Sustain.6, 1, 10.1038/s42949-025-00297-7 (2026). 7/18

work page doi:10.1038/s42949-025-00297-7 2026
[25]

& Cabrera, C

Renninger, A. & Cabrera, C. Extreme heat reduces and reshapes urban mobility.PNAS Nexus5, pgag078, 10.1093/ pnasnexus/pgag078 (2026)

2026
[26]

& Simorangkir, R

Fesselmeyer, E., Liu, H., Salvo, A. & Simorangkir, R. P. B. Heat and observed economic activity in the rich urban tropics. Econ. J.134, 3445–3460, 10.1093/ej/ueae046 (2024)

work page doi:10.1093/ej/ueae046 2024
[27]

Li, W., Wang, Q., Liu, Y ., Small, M. L. & Gao, J. A spatiotemporal decay model of human mobility when facing large-scale crises.Proc. Natl. Acad. Sci.119, e2203042119, 10.1073/pnas.2203042119 (2022)

work page doi:10.1073/pnas.2203042119 2022
[28]

& Wolff, H

Albouy, D., Graf, W., Kellogg, R. & Wolff, H. Climate amenities, climate change, and american quality of life.J. Assoc. Environ. Resour. Econ.3, 205–246, 10.1086/684573 (2016)

work page doi:10.1086/684573 2016
[29]

& Blanchard, P

Renninger, A., Holubowska, O. & Blanchard, P. Remote sensing and gps mobility reveal heat’s impact on human activity across diverse climates.EPJ Data Sci.10.1140/epjds/s13688-026-00652-3 (2026). Acknowledgments The work is supported by funding from the Alexander von Humboldt Foundation and the Federal Ministry of Education and Research (Bundesministerium ...

work page doi:10.1140/epjds/s13688-026-00652-3 2026