Urban heat reduction

Main goal

The planning objective is to lower day- and night time surface and air temperatures and improve outdoor thermal comfort locally and to keep doing so through dry periods. Green and blue NbS achieve this through three mechanisms: shading (mainly trees), evaporative cooling from moist soil, vegetation and open water and the modification of airflow. For stormwater-related NbS the critical link is water: evaporative cooling only continues into a heatwave if the asset has retained enough water to evapotranspire, which is why soil moisture, substrate moisture and pond levels are both a condition parameter and a performance parameter in this section. [3], [5], [15], [17]

The Baltic Sea Region has historically had a moderate heat problem, but this is changing quickly. Warming is faster at high latitudes; Helsinki summers in 2018, 2021 and 2022 were 1.3–2.1 °C above the 1991–2020 mean and heatwaves are becoming more frequent, longer and more intense. The July 2025 heatwave in Norway, Sweden and Finland was made around 2 °C hotter and at least ten times more likely by climate change, and put hospitals and municipal services under strain. Night-time heat (“tropical nights”) is the main health concern in the region, so monitoring should cover nights as well as afternoon peaks. [1], [2]

Global syntheses report an average local cooling from trees of about 1.5 °C, roughly two to three times the effect of grass (0.7 °C) or green roofs and walls (0.6 °C); European modelling suggests urban green infrastructure lowers city temperatures by about 1 °C on average, and reaching a 1 °C reduction requires roughly 16 % tree cover. A single rain garden will not cool a street; a shaded street or a park will. [4], [6]

Typical NBS types

Cooling NbS work best as a network of shaded, well-watered green and blue spaces rather than as isolated features. The types below are ordered roughly by the size of the local cooling effect they can deliver. [3], [4], [5]

  • Street and park trees: Provide shade and transpiration; the largest and best-evidenced effect at pedestrian level. Cooling depends on canopy size, leaf area and species, so newly planted trees give little cooling for 10–20 years. [4], [5], [6]
  • Parks, green corridors and tree-lined swales: Create “park cool islands” that extend some distance into surrounding streets; the effect grows with area and tree cover. [3], [5]
  • Ponds, wetlands and daylighted streams: Cool through evaporation and high emissivity; air-temperature reductions of up to about 2.5 K near water bodies are reported in meta-analyses. Humidity trade-offs and evaporation losses need to be watched in dry summers. [3]
  • Green roofs and green walls: Reduce surface temperatures on buildings and give a small air-temperature benefit at street level; their cooling drops sharply when the substrate dries out. [4], [15], [17]
  • Rain gardens, bioswales and de-sealed surfaces: Small and very localised direct cooling effect, and their soil can be dry again by the time a heatwave arrives; their main heat value is in retaining water in the landscape so that surrounding trees and vegetation stay healthy and keep transpiring. [22], [23]

Monitoring urban heat reduction

Monitoring of heat mitigation should answer two linked questions: Is the NbS still cooling its surroundings on hot days and warm nights and is it still in a condition (water, canopy, vegetation cover) that allows that cooling to continue? Two features distinguish heat monitoring from flood or water-quality monitoring. First, the “event” is a heatwave, not a storm, so the monitoring calendar concentrates on June–August and on calm, clear, hot days when the urban heat island is strongest. Second, a temperature reading is meaningless on its own: every method needs a paired reference site (a nearby unshaded or paved location) or a before-and-after baseline, because the cooling effect is defined as a difference, not a value. [3], [7]

  • Functional performance monitoring checks whether the NbS is actually lowering temperatures and improving thermal comfort compared with a reference and whether that effect holds through dry spells. The most useful evidence is collected during hot, calm periods; annual averages hide it. Where possible, results should be compared with the design intent (for example, the modelled cooling used to justify the measure) or with an earlier summer’s data. [3], [7]

    Practical monitoring approaches: Paired fixed air-temperature and humidity loggers (NbS site vs. reference) with a proper radiation shield; hand-held infrared thermometers or thermal cameras for surface temperature; thermal-camera flights with UAVs for larger sites; satellite land-surface temperature for portfolio- or district-scale trends; crowdsourced personal weather stations where a dense network exists, after quality control; comfort measurements (globe thermometer, wind) or comfort indices such as UTCI or PET at high-value public spaces. In Smart Water, LoRaWAN temperature and humidity sensors (with sunshield) were used in an urban environment to monitor heat at different locations in the city [2], [3], [7], [10], [11]

    Suggested parameters: Air temperature at 1.5–2 m, day and night (difference to reference); surface temperature of pavement, canopy, roof or water; relative humidity; mean radiant temperature or a thermal comfort index at key sites; shade coverage of the pedestrian area at midday; canopy cover and leaf area index; soil or substrate moisture and pond level as water-availability proxies; number of days on which cooling is measurable. [3], [7], [15]

    Realistic monitoring frequency: Continuous logging from May to September where sensors are installed; targeted campaigns on heatwave days and the following nights; a short summer summary every year comparing the cooling difference with previous years; canopy or shade survey every 3–5 years or after major pruning or tree loss.

    Example action triggers: The day-time or night-time difference to the reference site shrinks over successive summers; cooling disappears during dry spells (a water-limitation signal); shade coverage of the intended area drops below the design target; surface temperature of a green roof approaches that of a bare roof; comfort index at a public site remains in “strong heat stress” on days when the design predicted relief.

  • Condition and asset-health monitoring checks whether the physical state of the asset is undermining its cooling capacity. For heat-related NbS this is mostly about water and vegetation: drought stress, canopy loss, dead or thinning vegetation, dry substrate and falling water levels all become visible before the temperature data show a loss. Establishment losses are the biggest single risk: the first one to five years after planting have by far the highest tree mortality, and water is the most limiting factor in that period. [12], [13], [14]

    Practical monitoring approaches: Routine tree inspections and crown-condition scoring (see inspection list at “Learn more”); fixed photo points; canopy-cover mapping from aerial imagery or UAV surveys; UAV multispectral (NDVI) flights for vegetation vigour, as tested in City Blues; soil-moisture probes or LoRaWAN sensors in tree pits, rain gardens and green-roof substrates; green-roof walkovers for cover and substrate depth; pond water-level readings; digital manual inspection forms and citizen science reporting of wilting or dead trees, as tested in City Blues.

    Suggested proxies: Canopy cover and crown dieback; leaf wilting, early leaf drop or scorch; vegetation cover of green roofs (bare patches); substrate or soil moisture; mulch cover and soil compaction in tree pits; irrigation system function; pond level, algal blooms and evaporation losses; mechanical damage to trunks and roots; winter salt and frost damage. [12], [15], [16]

    Realistic monitoring frequency: Weekly to fortnightly during dry periods in the first three growing seasons for new plantings; monthly walkovers in the growing season thereafter; a detailed annual condition survey; additional checks during and after heatwaves and after storms that damage canopy. [13], [14]

    Example action triggers: Wilting or leaf scorch during a heatwave; soil or substrate moisture below the wilting threshold set for the species or product; loss of canopy above the agreed percentage; dead or dying young trees; green-roof cover below the manufacturer’s or design minimum; pond level dropping faster than expected; irrigation system not delivering.

  • Maintenance needs and response monitoring turns inspection findings into asset management. For heat-related NbS the time-critical items are watering during drought and quick replacement of failed plantings, because every lost growing season delays the shade benefit. Portfolio-level tracking shows backlog, recurring failures (for example the same tree pits drying out every summer) and resourcing gaps. [19], [20]

    Practical maintenance reporting options: Maintenance logs; work orders; inspection-to-action checklists; watering-round records; contractor close-out records; quarterly review of repeat issues and overdue actions.

    Suggested parameters: Open issues; repeat defects; watering response time during drought alerts; tree replacement backlog and replacement rate; time from failure to replanting; overdue pruning; recurrence after repair.

    Realistic monitoring frequency: Ongoing logging; weekly review during heat or drought warnings; monthly operational review; annual lessons-learned summary before the next growing season.

    Example action triggers: Rising backlog; watering rounds missed during a drought warning; the same site failing in consecutive summers; measurable loss of cooling that is clearly linked to delayed maintenance.

  • Different monitoring questions need different methods. Reliable heat-monitoring programmes combine a small number of well-placed sensors with routine vegetation inspections, occasional targeted field measurements, remote sensing for the overview, and feedback from residents and site users. [3], [7]

    Automated

    Fixed sensor pairs or networks record air temperature, humidity, surface temperature, soil or substrate moisture and (at high-value sites) globe temperature and wind. Low-cost loggers in naturally ventilated radiation shields are workable for intra-urban comparisons, but they can show a daytime bias compared to professional stations, driven by solar heating and low wind and need intercomparison against a reference station. Crowdsourced personal weather stations give dense coverage but need quality control and bias correction before use. [2], [10], [11]

    Strengths: Continuous capture of hot days and nights; consistent event comparison year to year; can trigger watering alerts.

    Limitations: Sensor siting and radiation bias; calibration and power; a minor difference in temperature between sites can be lost in measurement error if shields and heights are not consistent. [10]

    Inspection-based

    Structured tree and vegetation inspections, shade and canopy audits, photo points, green-roof and pond walkovers. Best for drought stress, canopy loss, establishment failure and irrigation problems – the causes of cooling loss rather than the loss itself. [12], [16]

    Strengths: Low cost; direct link to maintenance action; can be done by caretakers or trained volunteers.

    Limitations: Subjective; does not quantify cooling; misses short heat episodes between visits.

    Field testing

    Spot measurements with hand-held infrared thermometers or thermal cameras, mobile microclimate transects, soil-moisture probing and infiltration checks where an asset is meant to keep soil moist. Best for confirming that a suspected loss of performance is real and for diagnosing whether water supply, canopy or surface condition is the cause. [3]

    Strengths: High diagnostic value; quick to deploy on a heatwave day.

    Limitations: Snapshots only; results depend strongly on the weather at the time of measurement.

    Remote sensing and modelling

    Satellite land-surface temperature and vegetation indices (for example from Landsat or Sentinel) give district-wide trends and identify hot spots; UAV thermal and multispectral flights give the same view at site scale; microclimate models such as ENVI-met are used to predict the cooling of a planned measure and to set the performance target that later monitoring is compared with. Models must be calibrated with local measurements, and their limitations (simplified crowns, static wind and cloud) should be kept in mind. [3], [4], [8]

    Strengths: Whole-portfolio coverage; a target to compare measurements against; useful where field cooling is small and hard to attribute.

    Limitations: Satellite surface temperature is not air temperature and is only available on clear days at overpass times; models need calibration. [3], [4]

    User / operator feedback

    Resident and caretaker reports of wilting or dead trees, dry tree pits, uncomfortable squares, algal blooms or unusable areas on hot days; short comfort surveys at public spaces; complaint logs. In City Blues, citizen science and digital manual inspection forms were tested for this purpose.

    Strengths: Fast local intelligence during heatwaves; captures perceived comfort, which sensors cannot.

    Limitations: Inconsistent; needs verification and a standard reporting form.

    Typical use cases in heat-related NbS: Tree-lined streets and squares, schoolyards and playgrounds, green roofs on occupied buildings, park ponds and portfolio-level tracking of young-tree survival.

    A balanced programme usually combines inspection-based monitoring of vegetation and water as the default, a few paired sensor sites for the cooling evidence, field tests on heatwave days for diagnosis, and remote sensing or modelling for the portfolio view and the performance target.

Maintenance actions for urban heat reduction

  • Establishment-phase maintenance

    During the establishment phase the priority is to keep young trees and vegetation alive and growing, because cooling will only arrive once the canopy has developed. Tree mortality is highest in the first one to five years after planting, water is the most limiting factor, and inadequate post-planting care is a leading cause of loss. Typical actions are a fixed watering schedule through the growing season (for example fortnightly in the first two years, more often in drought), mulching, weed control, stake and guard checks, protection of the root zone from compaction and construction traffic, quick replacement of failed trees and shrubs, and confirming that stormwater actually reaches the tree pits and rain gardens as designed. For green roofs, establishment means securing full vegetation cover before the first dry summer. [12], [13], [14], [16]

    Routine maintenance

    Routine maintenance keeps the water supply and the canopy intact and the site usable. Typical actions include watering during dry spells, checking and topping up mulch, weed control, light formative pruning that preserves crown volume, clearing inlets so stormwater reaches planted areas, checking irrigation and water-storage systems, keeping green-roof drainage clear and cover intact, controlling algae and managing water levels in ponds, and monitoring soil moisture in tree pits and substrates. [15], [17], [20]

    Periodic maintenance

    Periodic maintenance addresses decline that routine visits cannot solve: crown-lifting or structural pruning that keeps shade rather than removes it, soil de-compaction and root-zone improvement, substrate top-up or renewal on green roofs, replacement of drought-damaged plantings with more suitable species, desilting of ponds, renewal of irrigation or water-storage equipment, and recalibration or replacement of sensors before each summer. [15], [20]

    Event-caused maintenance

    Event-caused maintenance is the response to a heatwave, a drought, a storm or a winter event. It includes emergency watering during heat or drought warnings, temporary shading or irrigation of exposed public spaces, removal of storm-damaged limbs and replanting after canopy loss, response to algal blooms or fish kills in ponds, repair of frost and road-salt damage in spring, and documenting how the asset performed during the event so that the next season’s plan can be adjusted. Where the same heat-related failure recurs, the response should move beyond watering to species review or redesign of the water supply. [12], [15], [18]

  • The table below converts common field observations into practical decision rules for street trees, green roofs, ponds and vegetated stormwater assets. [12], [13], [15], [16], [19]

    ObservationLikely issueRequired action
    Young trees wilt, scorch or drop leaves during a dry spellWater deficit in the root ball or tree pit; failed establishmentWater immediately and put the site on a fixed watering round; check that stormwater inlets to the pit are open; replace dead trees before the next season.
    Soil or substrate moisture sensor stays below the set threshold for several days in summerWater supply to the asset has failed or storage is exhaustedVerify in the field, water or restore supply, check for compaction or blocked inlets.
    Green roof shows brown or bare patches in summer or surface temperature close to a bare roofSubstrate dried out; vegetation die-back; drainage or irrigation faultIrrigate if a system exists, inspect drainage and substrate depth, re-seed or re-plant, review irrigation strategy.
    Measured day-time difference to the reference site has shrunkCanopy loss, tree decline, changed reference conditions or sensor driftCheck sensors and siting first, then survey canopy and vegetation; escalate to replanting or design review if the loss is real.
    Night-time temperatures at the site no longer differ from the referenceLoss of evaporative cooling or ventilation; heat storage in surrounding surfacesReview water availability and vegetation cover; consider de-sealing or additional planting around the asset.
    Pond level falls quickly, water turns green or smellsEvaporation exceeding inflow; nutrient loading; stagnationManage water level from stored stormwater where possible, remove algae and litter, review inflow and shading of the pond.
    Crown dieback or thinning in established trees after a hot summerDrought stress, root damage, pests or disease triggered by heatArborist assessment, soil and root-zone improvement, watering plan for the following summer.
    Shade coverage of a square or path is below the design target at middayTrees have not reached design size, pruning removed too much crown, or trees were lostCompare with the planting plan, adjust pruning practice, replant; consider temporary shading for the interim.

Monitoring case studies

Reference project: SmartWater Berlin

SmartWater (November 2022 – September 2026) is a Berlin model project led by Kompetenzzentrum Wasser Berlin with Berliner Wasserbetriebe, Technologiestiftung Berlin, the Berlin Senate departments, the Regenwasseragentur and two district authorities, funded by the Federal Ministry for Housing, Urban Development and Building. Its main product, the Blue-Green Infrastructure Planner (BGI-Planer), is a map-based web tool that shows planners the need for action on heat, heavy rain, sealing, water balance and water-body protection and the effect of blue-green measures; it was launched in September 2026. [8], [9], [21]

Relevant for heat monitoring: the heat effects of blue-green infrastructure were modelled with ENVI-met for a typical Gründerzeit street in the Ostkreuzkiez pilot area, using several climate scenarios, measured in the pilot area(e.g. hot day 15 August 2025, over 34 °C, followed by a tropical night) as model input. Adding green roofs, façade greening, tree rows and raised beds gave a simulated cooling of up to 1.8 °C at 1.5 m at 14:00. In Smart Water calibration and validation of these model results with its own temperature measurements in the pilot areas was done, the model-plus-measurement approach recommended in this section: a modelled target, verified by paired field measurements. [8], [9]

Learn more

    • Systematic review of methods, indicators and data sources for evaluating the cooling effects of green infrastructure [3]
    • Review of green infrastructure for urban cooling: techniques, design strategies and benefits [4]
    • Empirical review of urban greening to cool towns and cities [5]
    • EU handbook for practitioners on evaluating the impact of nature-based solutions [7]
    • Low-cost and crowdsourced temperature networks: evaluation, biases and quality control [10], [11]
    • Urban tree mortality and establishment: evidence base for watering and replacement rules [12], [13], [14]
    • Green-roof evapotranspiration, substrate moisture and irrigation [15], [16], [17]
    • Urban heat islands and heatwaves in Nordic cities [2], [18]

    • SmartWater Berlin – Blue-Green Infrastructure Planner with ENVI-met heat modelling validated by field measurements [8], [9]
    • cool (Belgium) – crowdsourced weather-station network used for urban heat island monitoring after quality control [11]
    • Bern (Switzerland) – 79 low-cost loggers in custom radiation shields mapping intra-urban heat in summer 2018 [10]

Literature

[1] World Weather Attribution, ‘Intense two-week heatwave in Fennoscandia hotter and more likely due to climate change’, Aug. 2025. [Online]. Available: https://www.preventionweb.net/news/intense-two-week-heatwave-fennoscandia-hotter-and-more-likely-due-climate-change

[2] Taylor, J., Simpson, C.H., Vanhatalo, J., Sohail, H., Brousse, O. and Heaviside, C. (2025) ‘Analysing cold-climate urban heat islands using personal weather station data’, Buildings and Cities, 6(1), p. 182–200.

[3] C. Bartesaghi Koc, P. Osmond, and A. Peters, ‘Evaluating the cooling effects of green infrastructure: A systematic review of methods, indicators and data sources’, Solar Energy, vol. 166, pp. 486–508, 2018.

[4] Azmeer, Amjad & Tahir, Furqan & Al-Ghamdi, Sami G.. (2024). Progress on green infrastructure for urban cooling: Evaluating techniques, design strategies, and benefits. Urban Climate. 56. 102077. 10.1016/j.uclim.2024.102077.

[5] D. E. Bowler, L. Buyung-Ali, T. M. Knight, and A. S. Pullin, ‘Urban greening to cool towns and cities: A systematic review of the empirical evidence’, Landscape and Urban Planning, vol. 97, no. 3, pp. 147–155, 2010.

[6] Li, X., Chen, Y., Sun, G., Young, D., McNulty, S., Williams, M., Wang, C., Sutton, W. B., Chang, A., Byl, T. D., & Johnson, M. (2026). A systematic review of the cooling effects of urban forests. Arboriculture & Urban Forestry, 52(2), 125–141. https://doi.org/10.48044/jauf.2026.003

[7] European Commission: Directorate-General for Research and Innovation, Evaluating the impact of nature-based solutions – A handbook for practitioners. Publications Office of the European Union, 2021. doi: 10.2777/244577.

[8] L. Junghans, F. Knopf, and P. Schütz, ‘SmartWater: Digitale Tools für eine klimaresiliente Stadt’, Neue Landschaft, no. 5/2025, p. 28 ff., May 2025.Available: https://neuelandschaft.de/artikel/smartwater-digitale-tools-fuer-eine-klimaresiliente-stadt-13471

[9] Land Berlin / Gemeinsam Digital: Berlin, ‘SmartWater – Digital Solutions for Climate-Resilient Cities’. Accessed: Sep. 16, 2026. [Online]. Available: https://gemeinsamdigital.berlin.de/en/smart-water/

[10] M. Gubler, A. Christen, J. Remund, and S. Brönnimann, ‘Evaluation and application of a low-cost measurement network to study intra-urban temperature differences during summer 2018 in Bern, Switzerland’, Urban Climate, vol. 37, 100817, 2021.

[11] Beele E, Reyniers M, Aerts R, Somers B. Quality control and correction method for air temperature data from a citizen science weather station network in Leuven, Belgium. Earth Syst Sci Data. 2022;14:4681–4717. doi:10.5194/essd-14-4681-2022.

[12] D. R. Hilbert, L. A. Roman, A. K. Koeser, J. Vogt, and N. S. van Doorn, ‘Urban tree mortality: A literature review’, Arboriculture & Urban Forestry, vol. 45, no. 5, pp. 167–200, 2019.

[13] D. J. Wattenhofer and G. R. Johnson, ‘Understanding why young urban trees die can improve future success’, Urban Forestry & Urban Greening, vol. 64, 127247, 2021.

[14] I. A. Smith, V. K. Dearborn, and L. R. Hutyra, ‘Live fast, die young: Accelerated growth, mortality, and turnover in street trees’, PLoS ONE, vol. 14, no. 5, e0215846, 2019.

[15] Wollschläger N, Schlink U, Graß R, Moeller L. Weather forecast driven irrigation of green roofs improves their thermal and hydrological performance. Sci Total Environ. 2026;1028:181690. doi:10.1016/j.scitotenv.2026.181690

[16] J. Lönnqvist, H. M. Hanslin, B. G. Johannessen, T. M. Muthanna, M. Viklander, and G. Blecken, ‘Temperatures and precipitation affect vegetation dynamics on Scandinavian extensive green roofs’, International Journal of Biometeorology, vol. 65, pp. 837–849, 2021

[17] S. Cascone, J. Coma, A. Gagliano, and G. Pérez, ‘The evapotranspiration process in green roofs: A review’, Building and Environment, vol. 147, pp. 337–355, 2019.

[18] F. Wang, A. Aldama-Campino, D. Belušić, J. H. Amorim, I. Ribeiro, L. Wiréhn, D. Segersson, R. Döscher, C. Navarra, T.-S. Neset, and P. Lind, ‘Interactions of urban heat islands and heat waves in Swedish cities under present and future climates’, Urban Climate, vol. 59, 102286, 2025.

[19] K. Kaur et al., ‘Failure modes across the life cycle of nature-based stormwater solutions: a systematic review of planning, design, and maintenance challenges’, Blue-Green Systems, vol. 7, no. 2, pp. 444–467, Oct. 2025, doi: 10.2166/bgs.2025.122.

[20] J. G. Langeveld et al., ‘Asset management for blue-green infrastructures: a scoping review’, Blue-Green Systems, vol. 4, no. 2, pp. 272–290, Nov. 2022, doi: 10.2166/bgs.2022.019.

[21] GFA News, ‘Berlin startet digitales Planungstool für blau-grüne Infrastruktur’, Sep. 2026. [Online]. Available: https://www.gfa-news.de/news/berlin-startet-digitales-planungstool-fuer-blau-gruene-infrastruktur

[22] A. Coutts, N. Tapper, M. Loughnan, M. Demuzere, A. Broadbent, A. Motazedian, E. White, T. Phan, J. Thom, L. Gebert und D. Pankhania, Determine the microclimatic influence of harvesting solutions and WSUD at the micro-scale: Presented as Frequently Asked Questions. Melbourne: Cooperative Research Centre for Water Sensitive Cities, Nov. 2016.

[23] A. M. Coutts, N. J. Tapper, J. Beringer, M. Loughnan und M. Demuzere, ‘Watering our cities: The capacity for Water Sensitive Urban Design to support urban cooling and improve human thermal comfort in the Australian context’, Progress in Physical Geography, Bd. 37, Nr. 1, S. 2–28, 2013, doi: 10.1177/0309133312461032.