Schwaab, Meier, Mussetti, Seneviratne, Bürgi & Davin · Nature Communications 12:6763 (2021) · CC BY 4.0
Inventory written page by page from the PDF text layer; figures read from rendered pages. Machine check: every quote found on its stated page, coverage in both directions, on-screen numbers matched to the inventory, reference list 1-84 continuous.
Reviewer: Build agent, page by page against the PDF
Keyboard: Space play/pause · ← → 5 s · N P scene · M sound · C captions
Pause any time and ask a question. Answers are grounded in this paper and labelled by where they come from; citations open the original PDF.
224 inventoried passages, figures, formulas, footnotes and references. Every one has a place; click any entry to see it in the PDF.
Nothing here.
Content the video only touches on, in full and with its place in the PDF.
Context from the introduction that the video only touches on.
Earlier energy-balance modelling showed that more urban vegetation has a limited cooling effect in tropical cities and differs between wet and dry climates.
Studies of land-use change look at vegetation types but not at cities, so they miss regional differences for urban trees and treeless urban green spaces.
Shading can strongly reduce daytime surface and air temperatures; the effect is larger over asphalt than grass and grows with leaf area index.
In dry regions, including parts of Southern Europe, summertime cooling by vegetation can be reduced because evapotranspiration is limited by soil moisture.
During hot extremes, higher surface resistance and a higher vapour pressure deficit pull in opposite directions, so vegetation can get relatively warmer or cooler.
City conditions can raise or lower tree cooling (CO2, nutrients, irrigation versus water stress, compacted soil, pollution), so it is unclear whether rural forests tell us about urban trees.
Which cities and regions are covered and how maps and country means are built.
Tree and land-cover data come from Copernicus for cities in the European Union and some additional countries, including Turkey.
Regional differences use a simplified division into European sub-areas, adding Turkey as a region and merging the Alps and Mid-Europe.
The text points to 'Fig. 3b' for the regions; the region map is panel b of Fig. 2. Fig. 3b shows rural vegetation boxplots by region.
Smooth spatial trends are computed with GAMs as an interpolation between cities; per country, differences are summarised by mean and standard error of the mean.
Predictors next to land cover, and how treeless green is separated.
The digital elevation model EU-DEM v1.0 provides elevation as a predictor and is used to compute slope orientation (aspect).
Aspect is reclassified into south-facing (90-270 degrees) and north-facing slopes; the fraction of north-facing slopes is computed per grid cell.
Because tree cover is a separate layer, the models can estimate the signal of treeless green spaces even though the Urban Atlas does not distinguish tree-covered from treeless green spaces.
How emissivity is set and which second satellite data set was used for validation.
Emissivity is derived from NDVI in three steps: NDVI per cell, fraction of vegetation cover from NDVI, then emissivity as a linear combination between 0.97 (no vegetation) and 0.99 (full vegetation).
0.97 to 0.99 emissivity (dimensionless) · Surface emissivity used in the LST retrieval · Linear in the fraction of vegetation cover, which is estimated from NDVI
Given in words; the main article has no numbered equations.
ASTER surface temperatures (90 m, on the Terra satellite, about 30 minutes later than Landsat, on average 194 scenes per city) were used mainly for comparison and validation.
194 scenes per city (average) · ASTER scenes available per city · 90 m resolution; used for comparison and validation
MODIS products, their resolution and their gaps.
Albedo (MODIS MCD43A3, 500 m) and evapotranspiration (MODIS MYD16A2, 1 km) were averaged over 2006-2018 and attributed to land-cover types with multiple linear regression.
500 m (albedo), 1 km (ET) resolution · Spatial resolution of the MODIS products · Multi-year averages 2006-2018, aggregated per month and city
For small cities there is sometimes too little data; in three cities the models predicted negative evapotranspiration, which was discarded. MODIS products are validated but show potential biases.
The albedo of urban areas is highest in southern European regions and lowest over Scandinavia; forest albedo varies comparatively little.
Fitted line printed in Fig. 4b: y = 1.28 - 1.68 x, R2 = 0.54 (y: green-space-minus-urban-fabric difference in K, x: evapotranspiration of rural pastures in mm/day).
y = 1.28 - 1.68 x; R2 = 0.54 K per mm/day (slope) · Mean summertime LST difference treeless urban green spaces minus continuous urban fabric versus evapotranspiration of rural pastures · One point per city; mean summertime values
Printed inside the figure image, not in the text layer; read from the rendered page.
Species-specific responses to heat and drought overlay environmental effects in ways not directly captured by the MODIS evapotranspiration product used.
Model terms, fit, sensitivity tests and what the model does not include.
Coordinates enter as two-dimensional tensor product smooths, which reduces spatial autocorrelation; all other predictors are thin plate regression splines.
Sensitivity experiments test the model set-up; 38 cities with unstable or non-significant tree effects were removed from that analysis. The fitted models average R2 = 0.64; it is lower in Turkey.
0.64 R2 (average) · Coefficient of determination of the fitted GAMs · Average over all cities; lower in Turkey (Supplementary Fig. 10)
E-OBS (v20.0e) gridded air temperature is the indicator of background temperature; where it is unavailable (some cities in Turkey) the spatial average of each satellite observation is used.
The approach tries to reduce the effect of how land-cover types are arranged in space, but it is unlikely that these effects were fully eliminated.
Analysing Local Climate Zones is an important approach for mixed urban land cover and may complement the separation approach used here.
Urban morphology and viewing-angle effects can bias observed temperatures and are not directly included; adding building height did not substantially change results in selected cities.
Finer regional findings, including urban versus rural vegetation.
Differences between rural forests and urban fabric show an east-west gradient: lower absolute values in Eastern than in Western Europe.
Absolute differences between rural pastures and urban fabric are much smaller than those between rural forests and urban fabric (Supplementary Fig. 1b).
The tree-minus-green-space difference is slightly higher in Central than in Southern European regions; the forest-minus-pasture difference is highest in the Mediterranean and Turkey.
In Scandinavia and over the British Isles, treeless urban green spaces and urban trees both provide substantial cooling.
In Central European regions and particularly Scandinavia, urban trees cool more than rural forests, possibly because factors that raise transpiration in cities outweigh those that reduce it.
In Turkey, cooling by urban trees is generally much lower than by rural forests, while treeless green spaces cool more than rural pastures - possibly an irrigation effect.
The authors say it is still very difficult to derive a clear picture of urban versus rural vegetation differences; high-resolution data on heat fluxes within cities would be needed.
Stomata, roots and surface roughness.
High temperatures could raise evapotranspiration through a high vapour pressure deficit, but transpirational cooling often drops considerably because trees reduce stomatal conductance; some species keep stomata open.
Trees have a larger root depth, which lets them use deeper soil moisture and sustain evapotranspiration when upper soil layers are dry.
Rural forests are rough surfaces that convect heat efficiently; in cities, trees in street canyons can decrease or increase roughness depending on their height relative to buildings.
The overall amount of urban vegetation varies between cities in different regions, which may influence transpiration and the observed surface temperatures.
What the authors expect and what remains to be studied.
Tree cooling decreases during hot extremes in many cities, particularly in Southern and Southeastern Europe; projected summer drying is likely to reduce vegetation benefits further, possibly also in regions with the highest cooling today.
Irrigation could help maintain cooling but may be limited by future water scarcity; this points to additional measures such as higher albedo of roofs and pavements.
There are clear correlations between surface and air temperature, and surface data can be used to estimate air temperature, though with known inaccuracies, e.g. in complex terrain and in daytime.
To increase relevance, it will be important to understand how the surface temperature patterns translate into air temperature and other variables that affect wellbeing and energy use.
Figures, formulas, tables, references, data and supplements.
Every figure of the main article. Figures are also explained in the chapters.

Map of smoothed temperature differences between 100% urban trees and 100% continuous urban fabric during hot extremes; dots are cities, the outer ring shows country means with standard errors and the number of cities per country.

(a) All analysed cities and nine example cities, (b) the European regions used, (c) boxplots of tree-minus-urban-fabric differences per season for nine cities, with the hottest observation as an orange dot.

Boxplots per region of temperature differences between (a) urban vegetation and urban fabric and (b) rural vegetation and urban fabric.

Scatterplots, one dot per city: (a) tree-minus-urban-fabric difference against evapotranspiration of rural forests, (b) green-space-minus-urban-fabric difference against evapotranspiration of rural pastures.

Surface temperature observations from 2006-2018 are the response in GAMs with several predictors; predicted differences are plotted against background temperature, and the smooth's value at the highest temperature (orange dot) is used to compare cities.
The main article has no numbered equations and no tables. These are the formula-like statements it contains.
Fitted line printed in Fig. 4a: y = 0.187 - 3.13 x, R2 = 0.46 (y: tree-minus-urban-fabric difference in K, x: evapotranspiration of rural forest in mm/day).
y = 0.187 - 3.13 x; R2 = 0.46 K per mm/day (slope) · Mean summertime LST difference urban trees minus continuous urban fabric versus evapotranspiration of rural forest · One point per city; mean summertime values
Printed inside the figure image, not in the text layer; read from the rendered page.
Fitted line printed in Fig. 4b: y = 1.28 - 1.68 x, R2 = 0.54 (y: green-space-minus-urban-fabric difference in K, x: evapotranspiration of rural pastures in mm/day).
y = 1.28 - 1.68 x; R2 = 0.54 K per mm/day (slope) · Mean summertime LST difference treeless urban green spaces minus continuous urban fabric versus evapotranspiration of rural pastures · One point per city; mean summertime values
Printed inside the figure image, not in the text layer; read from the rendered page.
Emissivity is derived from NDVI in three steps: NDVI per cell, fraction of vegetation cover from NDVI, then emissivity as a linear combination between 0.97 (no vegetation) and 0.99 (full vegetation).
0.97 to 0.99 emissivity (dimensionless) · Surface emissivity used in the LST retrieval · Linear in the fraction of vegetation cover, which is estimated from NDVI
Given in words; the main article has no numbered equations.
The main article contains no tables. Land-cover categories and model predictors are listed in Supplementary Tables 1 and 2, which are not part of this PDF.
Checked on all 11 pages: no table in the main article.
Per-city estimates of surface temperature differences, evapotranspiration and albedo are deposited on Zenodo: https://doi.org/10.5281/zenodo.5526674
Landsat LST (rslab.gr), ASTER LST and MODIS (NASA Earthdata), EU-DEM v1.0 and the Urban Atlas with Street Tree Layer (Copernicus), E-OBS (ECA&D).
Code describing how the differences were calculated: https://zenodo.org/record/5526734
Supplements are not part of the PDF. They are listed here with their own status and were not checked.
Cited in the article for, among others, treeless green spaces and pastures (1), rural forests (2, 3), albedo (4, 19), evapotranspiration (5), tree data coverage (6, 7), emissivity tests (8, 21), ASTER comparison (9), model fit (10-12), building height (15, 16), the northward shift (17) and LOESS curves (18).
Not part of the PDF. Available at the article page; not checked in this lesson.
Cited for the study domain (Note 1), model set-up sensitivity (Note 2), albedo approximation (Note 4), building height (Note 5) and selection of example cities (Note 6).
Not part of the PDF; not checked in this lesson.
Model predictors (Table 1) and Urban Atlas land-cover categories (Table 2).
Not part of the PDF; not checked in this lesson.
List of the 293 cities with the number of Landsat scenes per city.
Not part of the PDF; not checked in this lesson.
The role of urban trees in reducing land surface temperatures in European cities, by Schwaab, Meier, Mussetti, Seneviratne, Bürgi and Davin. Nature Communications 12:6763 (2021).
All authors: Institute for Atmospheric and Climate Science, ETH Zurich, Switzerland. Corresponding author: Jonas Schwaab.
Present address of Edouard L. Davin: Wyss Academy for Nature, Climate and Environmental Physics, Oeschger Centre for Climate Change Research, University of Bern.
Nature Communications (2021) 12:6763, DOI 10.1038/s41467-021-26768-w, open access.
Received 27 October 2020; accepted 20 October 2021.
Funded by the Swiss National Science Foundation and the Swiss Federal Office for the Environment through the CLIMPULSE project (grant no. 200021_172715).
J.S. and E.L.D. conceptualised the study; J.S. conducted the analysis and wrote the manuscript with help from all authors.
The authors declare no competing interests.
Peer reviewer reports are available.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open access under a Creative Commons Attribution 4.0 International License: use, sharing, adaptation and reproduction are permitted with credit, a link to the licence and an indication of changes. Figures in this lesson are cropped from the original PDF.
The complete reference list, extracted from the PDF.
Ref. 1 Wang, X. H., Wu, Y., Gong, J., Li, B. & Zhao, J. J. Urban planning design and sustainable development of forest based on heat island effect. Appl. Ecol. Environ. Res. 17, 9121–9129 (2019).
Ref. 2 Manoli, G. et al. Magnitude of urban heat islands largely explained by climate and population. Nature 573, 55–60 (2019).
Ref. 3 Chakraborty, T. & Lee, X. A simplified urban-extent algorithm to characterize surface urban heat islands on a global scale and examine vegetation control on their spatiotemporal variability. Int. J. Appl. Earth Observation Geoinf. 74, 269–280 (2019).
Ref. 4 Martilli, A., Krayenhoff, E. S. & Nazarian, N. Is the Urban Heat Island intensity relevant for heat mitigation studies? Urban Clim. 31, 100541 (2020).
Ref. 5 Li, Y. et al. Local cooling and warming effects of forests based on satellite observations. Nat. Commun. 6, 6603 (2015).
Ref. 6 Duveiller, G., Hooker, J. & Cescatti, A. The mark of vegetation change on Earth’s surface energy balance. Nat. Commun. 9, 679 (2018).
Ref. 7 Silva, J. S., Silva, R. M. D. & Santos, C. A. G. Spatiotemporal impact of land use/land cover changes on urban heat islands: a case study of Paço do Lumiar, Brazil. Build. Environ. 136, 279–292 (2018).
Ref. 8 Winbourne, J. B. et al. Tree transpiration and urban temperatures: current understanding, implications, and future research directions. Bioscience 70, 576–588 (2020).
Ref. 9 Wang, C. H., Wang, Z. H. & Yang, J. C. Cooling effect of urban trees on the built environment of contiguous United States. Earth Future 6, 1066–1081 (2018).
Ref. 10 Rahman, M. A., Moser, A., Rotzer, T. & Pauleit, S. Comparing the transpirational and shading effects of two contrasting urban tree species. Urban Ecosyst. 22, 683–697 (2019).
Ref. 11 Rahman, M. A. et al. Traits of trees for cooling urban heat islands: a meta-analysis. Build. Environ. 170, 106606 (2020).
Ref. 12 Coutts, A. M., White, E. C., Tapper, N. J., Beringer, J. & Livesley, S. J. Temperature and human thermal comfort effects of street trees across three contrasting street canyon environments. Theor. Appl. Climatol. 124, 55–68 (2016).
Ref. 13 Smithers, R. J. et al. Comparing the relative abilities of tree species to cool the urban environment. Urban Ecosyst. 21, 851–862 (2018).
Ref. 14 Wang, C. H., Wang, Z. H., Wang, C. Y. & Myint, S. W. Environmental cooling provided by urban trees under extreme heat and cold waves in US cities. Remote Sens. Environ. 227, 28–43 (2019).
Ref. 15 Su, Y. et al. Phenology acts as a primary control of urban vegetation cooling and warming: a synthetic analysis of global site observations. Agric. For. Meteorol. 280, 107765 (2020).
Ref. 16 Meili, N. et al. Tree effects on urban microclimate: diurnal, seasonal, and climatic temperature differences explained by separating radiation, evapotranspiration, and roughness effects. Urban Forestry Urban Green. 58, 126970 (2021).
Ref. 17 Manoli, G., Fatichi, S., Bou-Zeid, E. & Katul, G. G. Seasonal hysteresis of surface urban heat islands. Proc. Natl Acad. Sci. USA 117, 7082 (2020).
Ref. 18 Wang, P., Li, D., Liao, W. L., Rigden, A. & Wang, W. Contrasting evaporative responses of ecosystems to heatwaves traced to the opposing roles of vapor pressure deficit and surface resistance. Water Resour. Res. 55, 4550–4563 (2019).
Ref. 19 Pataki, D. E., McCarthy, H. R., Litvak, E. & Pincetl, S. Transpiration of urban forests in the Los Angeles metropolitan area. Ecol. Appl. 21, 661–677 (2011).
Ref. 20 Mussetti, G. et al. COSMO-BEP-Tree v1.0: a coupled urban climate model with explicit representation of street trees. Geosci. Model Dev. 13, 1685–1710 (2020).
Ref. 21 Brondfield, M. N., Hutyra, L. R., Gately, C. K., Raciti, S. M. & Peterson, S. A. Modeling and validation of on-road CO2 emissions inventories at the urban regional scale. Environ. Pollut. 170, 113–123 (2012).
Ref. 22 Decina, S. M., Templer, P. H., Hutyra, L. R., Gately, C. K. & Rao, P. Variability, drivers, and effects of atmospheric nitrogen inputs across an urban area: emerging patterns among human activities, the atmosphere, and soils. Sci. Total Environ. 609, 1524–1534 (2017).
Ref. 23 Zipper, S. C., Schatz, J., Kucharik, C. J. & Loheide, S. P. II Urban heat island-induced increases in evapotranspirative demand. Geophys. Res. Lett. 44, 873–881 (2017).
Ref. 24 Reyes-Paecke, S., Gironas, J., Melo, O., Vicuna, S. & Herrera, J. Irrigation of green spaces and residential gardens in a Mediterranean metropolis: gaps and opportunities for climate change adaptation. Landsc. Urban Plan. 182, 34–43 (2019).
Ref. 25 Melaas, E. K., Wang, J. A., Miller, D. L. & Friedl, M. A. Interactions between urban vegetation and surface urban heat islands: a case study in the Boston metropolitan region. Environ. Res. Lett. 11, 054020 (2016).
Ref. 26 Chen, X. P., Zhou, Z. X., Teng, M. J., Wang, P. C. & Zhou, L. Accumulation of three different sizes of particulate matter on plant leaf surfaces: effect on leaf traits. Arch. Biol. Sci. 67, 1257–1267 (2015).
Ref. 27 Meineke, E., Youngsteadt, E., Dunn, R. R. & Frank, S. D. Urban warming reduces aboveground carbon storage. Proc. R. Soc. B Biol. Sci. 283, 20161574 (2016).
Ref. 28 Jim, C. Y. Soil volume restrictions and urban soil design for trees in confined planting sites. J. Landsc. Architecture 14, 84–91 (2019).
Ref. 29 Manickathan, L., Defraeye, T., Allegrini, J., Derome, D. & Carmeliet, J. Parametric study of the influence of environmental factors and tree properties on the transpirative cooling effect of trees. Agric. For. Meteorol. 248, 259–274 (2018).
Ref. 30 Christidis, N. & Stott, P. A. The influence of anthropogenic climate change on wet and dry summers in Europe. Sci. Bull. 66, 813–823 (2021).
Ref. 31 Pauleit, S. et al. Tree establishment practice in towns and cities – results from a European survey. Urban Forestry Urban Green. 1, 83–96 (2002).
Ref. 32 Tsiros, I. X. Assessment and energy implications of street air temperature cooling by shade tress in Athens (Greece) under extremely hot weather conditions. Renew. Energy 35, 1866–1869 (2010).
Ref. 33 Chrysoulakis, N. et al. Urban energy exchanges monitoring from space. Sci. Rep. 8, 11498 (2018).
Ref. 34 Denissen, J. M. C., Teuling, A. J., Reichstein, M. & Orth, R. Critical soil moisture derived from satellite observations over Europe. J. Geophys. Res. Atmos. 125, e2019JD031672 (2020).
Ref. 35 McAdam, S. A. M. & Brodribb, T. J. The evolution of mechanisms driving the stomatal response to vapor pressure deficit. Plant Physiol. 167, 833–843 (2015).
Ref. 36 Teskey, R. et al. Responses of tree species to heat waves and extreme heat events. Plant Cell Environ. 38, 1699–1712 (2015).
Ref. 37 De Kauwe, M. G. et al. Examining the evidence for decoupling between photosynthesis and transpiration during heat extremes. Biogeosciences 16, 903–916 (2019).
Ref. 38 Roman, D. T. et al. The role of isohydric and anisohydric species in determining ecosystem-scale response to severe drought. Oecologia 179, 641–654 (2015).
Ref. 39 Wang, L., Huang, M. & Li, D. Where are white roofs more effective in cooling the surface? Geophys. Res. Lett. 47, e2020GL087853 (2020).
Ref. 40 Davin, E. L., Seneviratne, S. I., Ciais, P., Olioso, A. & Wang, T. Preferential cooling of hot extremes from cropland albedo management. Proc. Natl Acad. Sci. USA 111, 9757–9761 (2014).
Ref. 41 Iio, A., Hikosaka, K., Anten, N. P. R., Nakagawa, Y. & Ito, A. Global dependence of field-observed leaf area index in woody species on climate: a systematic review. Glob. Ecol. Biogeogr. 23, 274–285 (2014).
Ref. 42 Paschalis, A., Chakraborty, T., Fatichi, S., Meili, N. & Manoli, G. Urban forests as main regulator of the evaporative cooling effect in cities. AGU Adv. 2, e2020AV000303 (2021).
Ref. 43 Teuling, A. J. et al. Contrasting response of European forest and grassland energy exchange to heatwaves. Nat. Geosci. 3, 722–727 (2010).
Ref. 44 Yosef, G. et al. Large-scale semi-arid afforestation can enhance precipitation and carbon sequestration potential. Sci. Rep. 8, 996 (2018).
Ref. 45 Burakowski, E. et al. The role of surface roughness, albedo, and Bowen ratio on ecosystem energy balance in the Eastern United States. Agric. For. Meteorol. 249, 367–376 (2018).
Ref. 46 Schenk, H. J. & Jackson, R. B. The global biogeography of roots. Ecol. Monogr. 72, 311–328 (2002).
Ref. 47 Rotenberg, E. & Yakir, D. Contribution of semi-arid forests to the climate system. Science 327, 451–454 (2010).
Ref. 48 Giometto, M. G. et al. Effects of trees on mean wind, turbulence and momentum exchange within and above a real urban environment. Adv. Water Resour. 106, 154–168 (2017).
Ref. 49 Zhao, L., Lee, X., Smith, R. B. & Oleson, K. Strong contributions of local background climate to urban heat islands. Nature 511, 216–219 (2014).
Ref. 50 Li, D. et al. Urban heat island: aerodynamics or imperviousness? Sci. Adv. 5, eaau4299 (2019).
Ref. 51 Schwaab, J. et al. Increasing the broad-leaved tree fraction in European forests mitigates hot temperature extremes. Sci. Rep. 10, 14153 (2020).
Ref. 52 Parastatidis, D., Mitraka, Z., Chrysoulakis, N. & Abrams, M. Online global land surface temperature estimation from Landsat. Remote Sens. 9, 16 (2017).
Ref. 53 Zhou, W., Huang, G. & Cadenasso, M. L. Does spatial configuration matter? Understanding the effects of land cover pattern on land surface temperature in urban landscapes. Landsc. Urban Plan. 102, 54–63 (2011).
Ref. 54 Schwarz, N. & Manceur, A. M. Analyzing the influence of urban forms on surface urban heat islands in Europe. J. Urban Plan. Dev. 14, A4014003 (2015).
Ref. 55 Bechtel, B. et al. SUHI analysis using Local Climate Zones—a comparison of 50 cities. Urban Clim. 28, 100451 (2019).
Ref. 56 Hu, L., Monaghan, A., Voogt, J. A. & Barlage, M. A first satellite-based observational assessment of urban thermal anisotropy. Remote Sens. Environ. 181, 111–121 (2016).
Ref. 57 Martilli, A. et al. Summer average urban-rural surface temperature differences do not indicate the need for urban heat reduction. OSF Preprint at https:// doi.org/10.31219/osf.io/8gnbf (2020).
Ref. 58 Manoli, G. et al. Reply to Martilli et al. (2020): Summer average urban-rural surface temperature differences do not indicate the need for urban heat reduction. OSF Preprint at https://doi.org/10.31219/osf.io/mwpna (2020).
Ref. 59 Chakraborty, T., Hsu, A., Manya, D. & Sheriff, G. A spatially explicit surface urban heat island database for the United States: characterization, uncertainties, and possible applications. ISPRS J. Photogramm. Remote Sens. 168, 74–88 (2020).
Ref. 60 Zhang, P., Bounoua, L., Imhoff, M. L., Wolfe, R. E. & Thome, K. Comparison of MODIS land surface temperature and air temperature over the continental USA meteorological stations. Can. J. Remote Sens. 40, 110–122 (2014).
Ref. 61 Good, E. J. An in situ-based analysis of the relationship between land surface “skin” and screen-level air temperatures. J. Geophys. Res. Atmos. 121, 8801–8819 (2016).
Ref. 62 Mildrexler, D. J., Zhao, M. & Running, S. W. A global comparison between station air temperatures and MODIS land surface temperatures reveals the cooling role of forests. J. Geophys. Res. Biogeosci. https://doi.org/10.1029/ 2010JG001486 (2011).
Ref. 63 Novick, K. A. & Katul, G. G. The duality of reforestation impacts on surface and air temperature. J. Geophys. Res. Biogeosci. 125, e2019JG005543 (2020).
Ref. 64 Hu, Y. et al. Comparison of surface and canopy urban heat islands within megacities of eastern China. ISPRS J. Photogramm. Remote Sens. 156, 160–168 (2019).
Ref. 65 Hooker, J., Duveiller, G. & Cescatti, A. A global dataset of air temperature derived from satellite remote sensing and weather stations. Sci. Data 5, 180246 (2018).
Ref. 66 Serra, C. et al. Air temperature in Barcelona metropolitan region from MODIS satellite and GIS data. Theor. Appl. Climatol. 139, 473–492 (2020).
Ref. 67 Benali, A., Carvalho, A. C., Nunes, J. P., Carvalhais, N. & Santos, A. Estimating air surface temperature in Portugal using MODIS LST data. Remote Sens. Environ. 124, 108–121 (2012).
Ref. 68 Kloog, I., Chudnovsky, A., Koutrakis, P. & Schwartz, J. Temporal and spatial assessments of minimum air temperature using satellite surface temperature measurements in Massachusetts, USA. Sci. Total Environ. 432, 85–92 (2012).
Ref. 69 Alonso, L. & Renard, F. A new approach for understanding urban microclimate by integrating complementary predictors at different scales in regression and machine learning models. Remote Sens. 12, 35 (2020).
Ref. 70 Mutiibwa, D., Strachan, S. & Albright, T. Land surface temperature and surface air temperature in complex terrain. IEEE J. Sel. Top. Appl. Earth Observations Remote Sens. 8, 4762–4774 (2015).
Ref. 71 Ho, H. C., Knudby, A., Xu, Y., Hodul, M. & Aminipouri, M. A comparison of urban heat islands mapped using skin temperature, air temperature, and apparent temperature (Humidex), for the greater Vancouver area. Sci. Total Environ. 544, 929–938 (2016).
Ref. 72 Christensen, J. H. & Christensen, O. B. A summary of the PRUDENCE model projections of changes in European climate by the end of this century. Clim. Change 81, 7–30 (2007).
Ref. 73 EEA. EU-DEM Statistical Validation (EEA, 2014).
Ref. 74 European Commission. Mapping Guide v4.7 for a European Urban Atlas. https://land.copernicus.eu/user-corner/technical-library/urban-atlas-2012-mapping-guide-new/ (2012).
Ref. 75 Carlson, T. N. & Ripley, D. A. On the relation between NDVI, fractional vegetation cover, and leaf area index. Remote Sens. Environ. 62, 241–252 (1997).
Ref. 76 Gillespie, A. et al. A temperature and emissivity separation algorithm for Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) images. IEEE Trans. Geosci. Remote Sens. 36, 1113–1126 (1998).
Ref. 77 Schaaf, C. & Wang, Z. MCD43A3 MODIS/Terra+Aqua BRDF/Albedo Daily L3 Global - 500 m V006, distributed by NASA EOSDIS Land Processes DAAC. https://doi.org/10.5067/MODIS/MCD43A3.006 (2015).
Ref. 78 Running, S., Mu, Q. & Zhao, M. MYD16A2 MODIS/Aqua Net evapotranspiration 8-day l4 global 500m SIN grid V006. 2017, distributed by NASA EOSDIS Land Processes DAAC. https://doi.org/10.5067/MODIS/ MYD16A2.006 (2017).
Ref. 79 Cescatti, A. et al. Intercomparison of MODIS albedo retrievals and in situ measurements across the global FLUXNET network. Remote Sens. Environ. 121, 323–334 (2012).
Ref. 80 Wood, S. N. Generalized Additive Models: An Introduction with R 2nd edn. (CRC Press/Taylor & Francis Group, 2017).
Ref. 81 R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2019).
Ref. 82 Beale, C. M., Lennon, J. J., Yearsley, J. M., Brewer, M. J. & Elston, D. A. Regression analysis of spatial data. Ecol. Lett. 13, 246–264 (2010).
Ref. 83 Cornes, R. C., van der Schrier, G., van den Besselaar, E. J. M. & Jones, P. D. An Ensemble version of the E-OBS temperature and precipitation data sets. J. Geophys. Res. Atmos. 123, 9391–9409 (2018).
Ref. 84 Cleveland, W. S., Grosse, E. & Shyu, W. M. Local regression models In: Statistical Models in S (eds Chambers, J. M. & Hastie, T. J.) 309–376 (Wadsworth & Brooks/Cole, 1992).