Evaluating the Effects of Urbanization Evolution on Air Temperature Trends Using Nightlight Satellite Data
Round 1
Reviewer 1 Report
Comments and suggestions are in the attached file.
Comments for author File: 
Comments.pdf
Author Response
REVIEWER #1
Comments
Section 4.1: I do not agree with the statistical analysis done by the authors to show the link between temperature variation and urbanization trends. I need more explanation and motivation in order to be convinced.
One way to show the relation between the temperature and the urbanization
would be to perform a linear regression using the either models:
T(t) = a1 + b1 x t + b2 x DN(t) + error;
or
T(t) = a3 + b3 x DN(t) + error:
I used the data given in Figure 4 and for both models and the regression coefficients b2 and b3 are not significative (the p-values are respectively 0.6835 and 0.277). However, using the criterion proposed by the authors, this station is in the positive significant trends class.
I also have a problem with the use of p-values proposed in this section. The authors do not perform any test. They rather use the « p-values »to partition the support of a Student distribution into four classes, that is :
More details on the use of pT and pDN are needed.
Thanks for this comment. We partly disagree with the Reviewer: he/she suggests to study the relation between temperature and nightlights by means of a multivariate regression analysis, but this is not the objective of our manuscript. Our purpose is not to correlate temperature and nightlights values, but rather to correlate trends i.e., to relate the temperature variations in recent years to the corresponding variations in nightlight data.
Moreover, our aim is to perform a significance test in order to allocate trends is temperature and nightlight data in four different classes and to measure the degree of concordance between the variations in time of temperature and nightlights.
Anyway, the Reviewer’s comment clearly highlighted that we had to improve the methodological communication, thus we modified the text trying to make the procedure easier to understand and making an effort to better state our aim.
Section 4.2.: There exist several way to see if two variables are related (e.g. contingency table for categorical variables (c1; : : : c4 for T and DN) or regression coefficient (pt and pDN). The proposed method should be compared to some standard methods.
We thank the reviewer for his/her suggestion. We are actually looking at the dynamic effect of urbanization on air temperature trends; that is the reason why we are relating T and DN (L for lights in the RM) trends and not merely values as they are. In this context, our purpose is to introduce a new way of quantifying the relationship between growing cities (as detected by nighttime lights trends) and temperature changes, whereas standard methods just allow us to capture the static relation between variables.
As stated in the previous point, we made efforts to clarify the statistical analysis, we thus invite the Reviewer to see how we modified the text (L 279 on).
Typo
Lines 116 and 119 : In line 116, the authors mention that DMSP manages four satellites while in line 119, the data from six satellites are used. Where do the two extra satellite come from?
We thank the reviewer for the comment. The sentences are misleading, so the text has been modified accordingly. Now the sentence in the RM reads “Annual time series of nightlights are freely provided from the NOAA National Geophysical Data Center (NGDC) as satellite images [30], collected under the Defense Meteorological Satellite Program (DMSP), Operational Linescan System (OLS). OLS consists of two sensors …Six satellites are used, with a total of 34 composite images, generating a product called stable light [2]” (L 130 on).
Figure 1: The Yes and No are missing for the diamond in the first column.
Yes and No have been added as suggested. Maybe it should of help to add reference to the scenario we incur as well, based on the numbered list used in the in the Geo-localization of air temperature stations section. The Figure and caption have been modified accordingly in the RM.
Equations (3) and (4): It should be mentioned that t correspond to the year - 1991.
In Equations (3) and (4) we fit T and DN values (L for lights in the RM) versus time t, being the analysed period from 1992 to 2013. As stated in the previous point, the Methods section has been modified in the RM to make it more clear (L 279 on).
Figure 4 : Looking at the graphic of the temperatures for the Torino Caselle station, the simple linear model does not seem appropriate. Using one or two change points improve significantly the model.
We thank the Reviewer for the comment, although we partly disagree. We are aware of the goodness of change point methods but this is not the purpose of our paper. Change points allow the identification of abrupt variation in the process behaviour due to distributional or structural changes, but we are rather performing a trend analysis i.e., attempting to predict future changes based on observed past data. As raised in the previous points, we modified the Methods section, trying to clarify the statistical analysis, we thus invite the Reviewer to see how the RM was changed (L 279 on).
Equations (5) and (6) : According to Table 1, the values of nTOT are different for both variables (nTOT = 1219 for T(c) and nTOT = 1144 for DN(c).)
We gratefully thank the Reviewer for this comment. Apologies for this. Table 1 and related references have been fixed in the RM (L 332, 336-339, Table 1).
Table 1: nDN should be replaced by WT(c) in the first line. I do not obtain the same results for WT(c) in the classes C3 and C.
Thanks for this comment. This issue has now been solved in the RM (L 336-340).
Line 438 : « several U.S. countries »does not make any sense.
Based on National Conference of State Legislatures [3],at least 18 U.S. states have laws in place to reduce light pollution. This reference has been added in the RM (L 467).
Reference [44 :] I did not find any reference to this reference in the text.
Apologies for this. Appropriate reference has been added in the RM and the entire bibliography section has been settled as well (L 544 on).
References
1. NOAA Earth Observation Group - Defense Meteorological Satellite Progam, Boulder | ngdc.noaa.gov Available online: https://ngdc.noaa.gov/eog/dmsp/downloadV4composites.html (accessed on Dec 13, 2018).
2. Ceola, S.; Laio, F.; Montanari, A. Human-impacted waters: New perspectives from global high-resolution monitoring. Water Resour. Res. 2015, 51, 7064–7079.
3. National Conference of State Legislatures States Shut Out Light Pollution Available online: http://www.ncsl.org/research/environment-and-natural-resources/states-shut-out-light-pollution.aspx (accessed on Jan 11, 2019).
4. Hansen, J.; Ruedy, R.; Sato, M.; Lo, K. Global surface temperature change. Rev. Geophys. 2010, 48, RG4004.
5. Yang, X.; Ruby Leung, L.; Zhao, N.; Zhao, C.; Qian, Y.; Hu, K.; Liu, X.; Chen, B. Contribution of urbanization to the increase of extreme heat events in an urban agglomeration in east China. Geophys. Res. Lett. 2017, 44, 6940–6950.
6. Dodman, D. Environment and Urbanization. Int. Encycl. Geogr. People, Earth, Environ. Technol. 2017, 1–9.
7. IPCC Part A: Global and Sectoral Aspects. (Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change). Clim. Chang. 2014 Impacts, Adapt. Vulnerability. 2014, 1132.
8. Zhou, Y.; Smith, S.J.; Zhao, K.; Imhoff, M.; Thomson, A.; Bond-Lamberty, B.; Asrar, G.R.; Zhang, X.; He, C.; Elvidge, C.D. A global map of urban extent from nightlights. Environ. Res. Lett. 2015, 10, 054011.
9. Santamouris, M.; Cartalis, C.; Synnefa, A.; Kolokotsa, D. On the impact of urban heat island and global warming on the power demand and electricity consumption of buildings—A review. Energy Build. 2015, 98, 119–124.
10. Peng, S.; Piao, S.; Ciais, P.; Friedlingstein, P.; Ottle, C.; Bréon, F.M.; Nan, H.; Zhou, L.; Myneni, R.B. Surface urban heat island across 419 global big cities. Environ. Sci. Technol. 2012, 46, 696–703.
11. Vörösmarty, C.J.; Green, P.; Salisbury, J.; Lammers, R.B. Global water resources: vulnerability from climate change and population growth. Science 2000, 289, 284–8.
12. Jones, G.A.; Warner, K.J. The 21st century population-energy-climate nexus. Energy Policy 2016, 93, 206–212.
13. UN DESA World Urbanization Prospects: The 2018 Revision;
14. Kovats, S.; Akhtar, R. Climate, climate change and human health in Asian cities. IIED) 2008, 20, 165–175.
15. Kalnay, E.; Cai, M. Impact of urbanization and land-use change on climate. Nature 2003, 423, 528–531.
16. McCarthy, M.P.; Best, M.J.; Betts, R.A. Climate change in cities due to global warming and urban effects. Geophys. Res. Lett. 2010, 37, 1–5.
17. Parker, D.E. Urban heat island effects on estimates of observed climate change. Wiley Interdiscip. Rev. Clim. Chang. 2010, 1, 123–133.
18. Hausfather, Z.; Menne, M.J.; Williams, C.N.; Masters, T.; Broberg, R.; Jones, D. Quantifying the effect of urbanization on u.s. Historical climatology network temperature records. J. Geophys. Res. Atmos. 2013, 118, 481–494.
19. Wickham, C.; Rohde, R.; Muller, R.A.; Wurtele, J.; Curry, J.; Groom, D.; Jacobsen, R.; Perlmutter, S.; Rosenfeld, A.; Mosher, S. Influence of Urban Heating on the Global Temperature Land Average using Rural Sites Identified from MODIS Classifications. Geoinformatics Geostatistics An Overv. 2013, 1, 1–6.
20. Arnfield, A.J. Two decades of urban climate research: A review of turbulence, exchanges of energy and water, and the urban heat island. Int. J. Climatol. 2003, 23, 1–26.
21. Jones, P.D.; Groisman, P.Y.; Coughlan, M.; Plummer, N.; Wang, W.C.; Karl, T.R. Assessment of urbanization effects in time series of surface air temperature over land. Nature 1990, 347, 169–172.
22. Oke, T.R. The energetic basis of the urban heat island. Q. J. R. Meteorol. Soc. 1982, 108, 1–24.
23. Creutzig, F. Towards typologies of urban climate and global environmental change. Environ. Res. Lett. 2015, 10.
24. Li, X.; Zhou, Y.; Asrar, G.R.; Imhoff, M.; Li, X. The surface urban heat island response to urban expansion: A panel analysis for the conterminous United States. Sci. Total Environ. 2017, 605–606, 426–435.
25. Environmental Protection Agency, U. Reducing Urban Heat Islands: Compendium of Strategies - Urban Heat Island Basics. 2008, 1–22.
26. Imhoff, M.L.; Zhang, P.; Wolfe, R.E.; Bounoua, L. Remote sensing of the urban heat island effect across biomes in the continental USA. Remote Sens. Environ. 2010, 114, 504–513.
27. Cao, Q.; Yu, D.; Georgescu, M.; Wu, J. Impacts of urbanization on summer climate in China: An assessment with coupled land-atmospheric modeling. J. Geophys. Res. 2016, 121, 10,505-10,521.
28. Jr, B.S. Short Communication Urban and rural temperature trends in proximity to large US cities : 1951 – 2000. 2007, 1807, 1801–1807.
29. Dupont, W.D.; Plummer, W.D. Power and Sample Size Calculations for Studies Involving Linear Regression. Control. Clin. Trials 1998, 19, 589–601.
30. Berkeley Earth Berkeley Earth Available online: http://berkeleyearth.org/about-data-set/ (accessed on Oct 4, 2017).
Reviewer 2 Report
The methodology presented is weak and therefore, the reviewer believes that it is not suitable for a scientific publication. In addition to the limits stated by the authors in the conclusions, another limit reseeds in the fact that global warming increased not uniformly over time. Therefore, this contribution should be subtracted in the temperature analysis. Similarly, also cases of local cooling are not included. As result, the analyses conducted and the methodology developed have too serious flaws to be published.
In addition, the manuscript is generally badly written and would benefit of a substantial increase in the style and language. Some examples are presented below:
Abstract, line 11: "Confounding" does not seem to be the right word.The reviewer suggests to pick another term that might better fit the sentence
Abstract lines 16-18: the sentence is not well structured. The reviewer suggests to reword it
Introduction, line 27: "..climate, along with several environmental problems [1,2]". Examples should be provided here
Introduction lines 27-28: "Rapid urbanization growth is occurring in developing countries,..". Figures or percentages and references should be provided in the text
Introduction, line 29: It should be explained why the urbanization affect human health
Introduction, line 29: It should be explained how urbanization affects the local energy balance and one or more references should be added to the text
Introduction, lines 29-30: It is unclear the meaning of the following sentence "in order to
30 consistently attribute the causes of observed warming at the global scale". The reviewer suggests to reword the sentence
Introduction, lines 32-34: the two sentences are not connected. The reviewer suggests to structure the paragraph in a different manner
Introduction, line 36: "expansion of the built environment plays a key role in urbanization" the sentence seems to be weird since expansion of the built environment is a synonym of urbanization. The reviewer suggests to reword the sentence
Introduction lines 32-37: the paragraph is confused and would benefit of some rewording. The reviewer suggests to restructure the paragraph
Introduction line 40: "particularly during nighttime" references are needed here
Introduction line 40-43: The paragraph is too generic. It would benefit of more details and examples
Introduction lines 43-44: the example in the manuscript does not clarify the sentences before the the example. Therefore, it seems that it is out of the contest
Introduction lines 44-45: the sentence is totally disconnected by the previous sentence. The reviewer would suggest to improve the style
Introduction, line The meaning of "control urbanization impacts" is unclear. The reviewer would suggest to increase clarity throughout the whole manuscript
Introduction, line 51: How? it should be explained
Introduction, lines 51-52: How? It should be explained
Introduction, lines 55-56: How? It should be explained
Introduction, line 56-57: Which effects? It should be stated
Introduction, line 63: Reference 26 is as reference 16. The reviewer suggests to correct it
Introduction, line 64: As the authors refer to an already concluded study, they should use "were" rather than "are"
Introduction, line 70: What are "urban processes"? It should be clearly explained what the author mean with those terms
Introduction, lines 73-75: The sentence is badly written, the reviewer suggests to reword it
Introduction, line 95: Issue is a too generic word, please the clarity of the text using a more fitting word
Introduction, lines 103-104: it is not clear why this study can be considered innovative since in literature can be found already many studies that correlate nightlight data with temperature (e.g., Characterizing Urban Heat Island Effect at Global Settlements Using MODIS and Nightlight Products). The authors should clearly state if and how their study differs from the studies previously published
Figure 1a: Figure 1a should be better described since it is not clear. For example, it is not clear why to a specific year corresponds different numbers of available monthly records
Discussion and conclusions are weak, not well motivated and very limited. Furthermore, the conclusions are often contrasting. The results seem to confirm that the use of the methodology presented in the manuscript cannot be widely applied and that its application might be misleading as also shown in lines 436-444, 450-453, 457-460.
Author Response
REVIEWER #2
Comments
The methodology presented is weak and therefore, the reviewer believes that it is not suitable for a scientific publication. In addition to the limits stated by the authors in the conclusions, another limit reseeds in the fact that global warming increased not uniformly over time. Therefore, this contribution should be subtracted in the temperature analysis. Similarly, also cases of local cooling are not included. As result, the analyses conducted and the methodology developed have too serious flaws to be published.
In addition, the manuscript is generally badly written and would benefit of a substantial increase in the style and language. Some examples are presented below.
We feel that the purpose of the paper has not been completely understood the Reviewer. This is due to a lack of methodological communication by our part, apologies for this. Literature provides lots of global warming trend analyses accounting for the effect of urban warming conducted for example using adjusted urban temperature data, removing sites with suspected urban warming from global [4] and regional warming analyses [5]. Our manuscript does not deal with a simple temporal trend analysis, rather with a strategy to detect possible correlations between variations in time of temperature and urbanization data, the latter represented through nightlights. We thus tried to improve the explanation of our findings and how they build on prior work to highlight the particular contribution of our research. We believe we improved the methodological communication throughout the text, to better introduce the rationale and purposes of the work. Conclusions were widened, comparing and linking our findings to previous studies.
As suggested, we carefully revised the manuscript style and language to comply with this requirement.
We invite the Reviewer to see how we modified the manuscript, taking into account his/her suggestions.
Typo
Abstract, line 11: "Confounding" does not seem to be the right word. The reviewer suggests to pick another term that might better fit the sentence
We partly disagree with the Reviewer on that point: a confounder (or confounding factor) is an extraneous variable or factor whose presence affects the factors being studied so that the results do not reflect the actual relationship between the variables under study. We think that, in this context, “confounding” is the most appropriate word, but we are obviously open to any suggestion the Reviewer could provide us.
Abstract lines 16-18: the sentence is not well structured. The reviewer suggests to reword it
As suggested by the Reviewer, the sentence has been rephrased, now reading as: “By means of a range of statistical methods, our results quantify and outline that the temporal evolution of urbanization affects temperature trends at multiple spatial scales, with significant differences at regional and continental scales.” (L 16-18).
Introduction, line 27: "..climate, along with several environmental problems [1,2]". Examples should be provided here
Examples have been provided along with related references. The text at this point has been rephrased and now reads as: “Urban transition leads to alterations in landscape conditions and to important modifications in urban climate, along with several environmental problems e.g., on water use and quality, on the generation of air pollution, and on the production of solid waste and sewage [6,7]. Because of rapid urbanization growth, even more considerable impacts are expected at broader scale and especially in developing countries, like higher consumption of energy, goods, services and resources demand, which have the potential for greater negative impacts on global environments and ecosystems [8–12]”. (L 27-32)
Introduction lines 27-28: "Rapid urbanization growth is occurring in developing countries,..". Figures or percentages and references should be provided in the text
As suggested by the Reviewer, percentage and references have been added. Now the text reads as: “Because of rapid urbanization growth, even more considerable impacts are expected at broader scale and especially in developing countries, like higher consumption of energy, goods, services and resources demand, which have the potential for greater negative impacts on global environments and ecosystems [8–12]…In this regard, we must also consider that 55% of the world population is residing in urban areas in 2018, which is projected to reach 68% by 2050 [13].” (L 29-37).
Introduction, line 29: It should be explained why the urbanization affect human health
The text has been modified at this point, providing explanation and references as suggested by the reviewer. Now it reads as: “Changes in food supplies, freshwater resources and increase in extreme weather events (e.g., heatwaves and droughts) are expected to lead to several consequences on human health as well, in terms of e.g., heat stress, cardio-respiratory and infectious diseases [7,14].” (L 33-35).
Introduction, line 29: It should be explained how urbanization affects the local energy balance and one or more references should be added to the text
We invite the Reviewer to see how we modified the Introduction at this point (L 27-39).
Introduction, lines 29-30: It is unclear the meaning of the following sentence "in order to consistently attribute the causes of observed warming at the global scale". The reviewer suggests to reword the sentence
The sentence has been rephrased as suggested, now reading as: “In the context of global climate change, it is crucial to better investigate how urban growth affects temperature record trends to consistently attribute the causes of observed warming at wider scales [4,15–20].” (L 37-39).
Introduction, lines 32-34: the two sentences are not connected. The reviewer suggests to structure the paragraph in a different manner
Thanks for the comment. The paragraph has been modified, now reading as : “The impact of urbanization on near surface temperatures has been investigated since the 1980s [21,22]. These studies suggested that a proportion of global warming observed on last century timescale could be related to local warming induced by urbanization. Urban development could thus have a great impact on temperature measurements, particularly for nations and regions experiencing demographic expansion.” (L 40-44).
Introduction, line 36: "expansion of the built environment plays a key role in urbanization" the sentence seems to be weird since expansion of the built environment is a synonym of urbanization. The reviewer suggests to reword the sentence
We thank the Reviewer for the comment. The sentence sounds ambiguous and thus has been rephrased as suggested, now reading as: “Indeed, the rapid expansion of the built environment induced by the urbanization process is known to drive local climate changes due to the relative increase of temperature within the urban area, thus playing a crucial role in near-surface warming [17,23,24].” (L 44-46).
Introduction lines 32-37: the paragraph is confused and would benefit of some rewording. The reviewer suggests to restructure the paragraph
Thanks for the comment. The entire paragraph has been rephrased as suggested, now reading as: “The impact of urbanization on near surface temperatures has been investigated since the 1980s [21,22]. These studies suggested that a proportion of global warming observed on last century timescale could be related to local warming induced by urbanization. Urban development could thus have a great impact on temperature measurements, particularly for nations and regions experiencing demographic expansion. Indeed, the rapid expansion of the built environment induced by the urbanization process is known to drive local climate changes due to the relative increase of temperature within the urban area, thus playing a crucial role in near-surface warming [17,23,24].” (L 40-46).
Introduction line 40: "particularly during nighttime" references are needed here
References have been added as suggested (Parker, D.E. Urban heat island effects on estimates of observed climate change. Wiley Interdiscip. Rev. Clim. Chang. 2010, 1, 123–133; Pielke, R.A.; Matsui, T. Should light wind and windy nights have the same temperature trends at individual levels even if the boundary layer averaged heat content change is the same? Geophys. Res. Lett. 2005, 32, L21813). (L 50).
Introduction line 40-43: The paragraph is too generic. It would benefit of more details and examples
Thanks for this suggestion. We invite the Reviewer to refer to our responses reported above and check how the Introduction was reorganized and widened, based on the comments and suggested literature (L 26 on).
Introduction lines 43-44: the example in the manuscript does not clarify the sentences before the example. Therefore, it seems that it is out of the contest
The text at this point has been rephrased and examples have been added as suggested by the Reviewer, now reading as: “UHI effects could be mainly ascribed to solar heat retention by building materials (having low albedo and high heat capacity), obstruction of longwave emission by the built environment, changes in land cover and urban geometries (e.g., reduced vegetation in urban areas) causing a generalized reduced evapotranspiration, anthropogenic heat emissions (e.g., air conditioning, cars and industrial facilities). Additionally, weather and local topographical characteristics contribute to UHI effect as well [17,25,26]. As an example, studies suggest that land-sea interactions induce horizontal thermal advection, whereas mountain landscapes are dominated by vertical advection [20,27].” (L 50-57).
Introduction lines 44-45: the sentence is totally disconnected by the previous sentence. The reviewer would suggest to improve the style
A sentence has been added to better contextualise. Not it reads as: “Two different UHI types are usually considered, i.e. surface and atmospheric UHI. The former refers to the land surface temperature specifically [10], whereas the latter considers the land air temperature as measured by land-based weather stations [19].” (L 58-60).
Introduction, line The meaning of "control urbanization impacts" is unclear. The reviewer would suggest to increase clarity throughout the whole manuscript
The sentence has been rephrased, now reading as: “…and control urban-induced impacts…” (L 62-63).
We invite the Reviewer to see how we modified the Introduction based on his/her suggestions.
Introduction, line 51: How? it should be explained
Examples and explanations are provided in the paragraph and throughout the manuscript, along with related reference in the RM. We invite the Reviewer to see how the Introduction has been changed based on his/her suggestions (L 62 on).
Introduction, lines 51-52: How? It should be explained
Examples and explanations are provided in the paragraph and throughout the manuscript, along with related reference in the RM (L 62 on).
Introduction, lines 55-56: How? It should be explained
As mentioned in the previous point, the Introduction has been widened and modified to address the Reviewer’s concerns (L 62 on).
Introduction, line 56-57: Which effects? It should be stated
We refer to effects of heating effects in this context. The sentence has been modified accordingly in the RM (L 73-74).
Introduction, line 63: Reference 26 is as reference 16. The reviewer suggests to correct it
Thanks for the comment and apologies for this. The reference has been corrected, being now Reference 25 in the RM. (i.e., [27] Cao, Q.; Yu, D.; Georgescu, M.; Wu, J. Impacts of urbanization on summer climate in China: An assessment with coupled land-atmospheric modeling. J. Geophys. Res. 2016, 121, 10,505-10,521.).
Introduction, line 64: As the authors refer to an already concluded study, they should use "were" rather than "are"
The sentence has been corrected as suggested (L 80).
Introduction, line 70: What are "urban processes"? It should be clearly explained what the author mean with those terms
“Urban processes” is misleading and it has been modified accordingly into has been changed, namely into “urbanization” (L 86).
Introduction, lines 73-75: The sentence is badly written, the reviewer suggests to reword it
The sentence has been rephrased as suggested by the reviewer. Namely: “In this regard, it is crucial to quantitatively assess the amplification of global temperature trends by urbanization i.e., to estimate to what extent large urbanized areas may be amplifying warming attributed to climate change at global scale [28].” (L 89-91).
Introduction, line 95: Issue is a too generic word, please the clarity of the text using a more fitting word
As suggested, we tried to clarify this concept and now reads as: “level of poverty estimates” (L 111).
Introduction, lines 103-104: it is not clear why this study can be considered innovative since in literature can be found already many studies that correlate nightlight data with temperature (e.g., Characterizing Urban Heat Island Effect at Global Settlements Using MODIS and Nightlight Products). The authors should clearly state if and how their study differs from the studies previously published
Thanks for this suggestion. We invite the Reviewer to check how the Introduction was reorganized and widened to improve the explanation of our findings and how they build on prior work to highlight the particular contribution of our research, based on the comments and suggested literature (L 101 on). Please also refer to our responses reported above.
Figure 1a: Figure 1a should be better described since it is not clear. For example, it is not clear why to a specific year corresponds different numbers of available monthly records
Figure 1 is the flowchart, probably the Reviewer refers to Figure 3a. In this case, we performed a survey of the available stations from 1992 to 2013 in the Berkeley Earth dataset considering both average monthly temperatures data. We first considered some statistics about the number of times when it is possible to derive the temperature T using the available data records i.e., how many stations have the entire data series and, if they are incomplete, how many months per year are available per each station. Figure 3a is an example of consistency analysis, which allows a rapid assessment of the number of available temperature records. As an example, we can see that, from 1992 to 1999, we have 12 months of recorded data for nearly 12000 stations (the sample size ranges between 12115 and 12613), while from 2001 to 2005 we have 12 months of recorded data for nearly 13000 stations (the sample size ranges between 12935 and 13801).
We thank the Reviewer for this comment, the Figure was not probably enough clear. Additional details have been included in the caption, as suggested by the Reviewer (L 271-273).
Discussion and conclusions are weak, not well motivated and very limited. Furthermore, the conclusions are often contrasting. The results seem to confirm that the use of the methodology presented in the manuscript cannot be widely applied and that its application might be misleading as also shown in lines 436-444, 450-453, 457-460.
As suggested by the Reviewer, the Discussion of the results in the RM was expanded building on previous works as for the Introduction, and appropriate comparison and citations were also included. We invite the Reviewer to see how the RM has been changed in these sections (L .26 on, L 427 on)
References
1. NOAA Earth Observation Group - Defense Meteorological Satellite Progam, Boulder | ngdc.noaa.gov Available online: https://ngdc.noaa.gov/eog/dmsp/downloadV4composites.html (accessed on Dec 13, 2018).
2. Ceola, S.; Laio, F.; Montanari, A. Human-impacted waters: New perspectives from global high-resolution monitoring. Water Resour. Res. 2015, 51, 7064–7079.
3. National Conference of State Legislatures States Shut Out Light Pollution Available online: http://www.ncsl.org/research/environment-and-natural-resources/states-shut-out-light-pollution.aspx (accessed on Jan 11, 2019).
4. Hansen, J.; Ruedy, R.; Sato, M.; Lo, K. Global surface temperature change. Rev. Geophys. 2010, 48, RG4004.
5. Yang, X.; Ruby Leung, L.; Zhao, N.; Zhao, C.; Qian, Y.; Hu, K.; Liu, X.; Chen, B. Contribution of urbanization to the increase of extreme heat events in an urban agglomeration in east China. Geophys. Res. Lett. 2017, 44, 6940–6950.
6. Dodman, D. Environment and Urbanization. Int. Encycl. Geogr. People, Earth, Environ. Technol. 2017, 1–9.
7. IPCC Part A: Global and Sectoral Aspects. (Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change). Clim. Chang. 2014 Impacts, Adapt. Vulnerability. 2014, 1132.
8. Zhou, Y.; Smith, S.J.; Zhao, K.; Imhoff, M.; Thomson, A.; Bond-Lamberty, B.; Asrar, G.R.; Zhang, X.; He, C.; Elvidge, C.D. A global map of urban extent from nightlights. Environ. Res. Lett. 2015, 10, 054011.
9. Santamouris, M.; Cartalis, C.; Synnefa, A.; Kolokotsa, D. On the impact of urban heat island and global warming on the power demand and electricity consumption of buildings—A review. Energy Build. 2015, 98, 119–124.
10. Peng, S.; Piao, S.; Ciais, P.; Friedlingstein, P.; Ottle, C.; Bréon, F.M.; Nan, H.; Zhou, L.; Myneni, R.B. Surface urban heat island across 419 global big cities. Environ. Sci. Technol. 2012, 46, 696–703.
11. Vörösmarty, C.J.; Green, P.; Salisbury, J.; Lammers, R.B. Global water resources: vulnerability from climate change and population growth. Science 2000, 289, 284–8.
12. Jones, G.A.; Warner, K.J. The 21st century population-energy-climate nexus. Energy Policy 2016, 93, 206–212.
13. UN DESA World Urbanization Prospects: The 2018 Revision;
14. Kovats, S.; Akhtar, R. Climate, climate change and human health in Asian cities. IIED) 2008, 20, 165–175.
15. Kalnay, E.; Cai, M. Impact of urbanization and land-use change on climate. Nature 2003, 423, 528–531.
16. McCarthy, M.P.; Best, M.J.; Betts, R.A. Climate change in cities due to global warming and urban effects. Geophys. Res. Lett. 2010, 37, 1–5.
17. Parker, D.E. Urban heat island effects on estimates of observed climate change. Wiley Interdiscip. Rev. Clim. Chang. 2010, 1, 123–133.
18. Hausfather, Z.; Menne, M.J.; Williams, C.N.; Masters, T.; Broberg, R.; Jones, D. Quantifying the effect of urbanization on u.s. Historical climatology network temperature records. J. Geophys. Res. Atmos. 2013, 118, 481–494.
19. Wickham, C.; Rohde, R.; Muller, R.A.; Wurtele, J.; Curry, J.; Groom, D.; Jacobsen, R.; Perlmutter, S.; Rosenfeld, A.; Mosher, S. Influence of Urban Heating on the Global Temperature Land Average using Rural Sites Identified from MODIS Classifications. Geoinformatics Geostatistics An Overv. 2013, 1, 1–6.
20. Arnfield, A.J. Two decades of urban climate research: A review of turbulence, exchanges of energy and water, and the urban heat island. Int. J. Climatol. 2003, 23, 1–26.
21. Jones, P.D.; Groisman, P.Y.; Coughlan, M.; Plummer, N.; Wang, W.C.; Karl, T.R. Assessment of urbanization effects in time series of surface air temperature over land. Nature 1990, 347, 169–172.
22. Oke, T.R. The energetic basis of the urban heat island. Q. J. R. Meteorol. Soc. 1982, 108, 1–24.
23. Creutzig, F. Towards typologies of urban climate and global environmental change. Environ. Res. Lett. 2015, 10.
24. Li, X.; Zhou, Y.; Asrar, G.R.; Imhoff, M.; Li, X. The surface urban heat island response to urban expansion: A panel analysis for the conterminous United States. Sci. Total Environ. 2017, 605–606, 426–435.
25. Environmental Protection Agency, U. Reducing Urban Heat Islands: Compendium of Strategies - Urban Heat Island Basics. 2008, 1–22.
26. Imhoff, M.L.; Zhang, P.; Wolfe, R.E.; Bounoua, L. Remote sensing of the urban heat island effect across biomes in the continental USA. Remote Sens. Environ. 2010, 114, 504–513.
27. Cao, Q.; Yu, D.; Georgescu, M.; Wu, J. Impacts of urbanization on summer climate in China: An assessment with coupled land-atmospheric modeling. J. Geophys. Res. 2016, 121, 10,505-10,521.
28. Jr, B.S. Short Communication Urban and rural temperature trends in proximity to large US cities : 1951 – 2000. 2007, 1807, 1801–1807.
29. Dupont, W.D.; Plummer, W.D. Power and Sample Size Calculations for Studies Involving Linear Regression. Control. Clin. Trials 1998, 19, 589–601.
30. Berkeley Earth Berkeley Earth Available online: http://berkeleyearth.org/about-data-set/ (accessed on Oct 4, 2017).
Reviewer 3 Report
I this work, a new approach is proposed to quantify the relationship between urbanisation dynamics and its thermal impact by relating air temperature with satellite derived nightlight data. The paper would be suitable for publication in Atmosphere, once the issues I raise below are addressed.
I have 2 major comments:
If the accuracy of the luminosity tendency can be affected by other factors, like air pollution, is this parameter an appropriate indicator for evaluating the effects of urbanisation on air temperature? Before using the nightlight images, the authors should validate these data against yearly land-cover product extracted from satellite imageries (https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mcd12q1_v006).
Are only 22 subjects enough to perform a robust linear regression analysis? The authors should analyse and comment if the sample size used for each station in the regression analysis is acceptable from the statistical point of view [1].
Comments:
Line 140 The authors should present in the manuscript or as ESM some global maps with satellite nightlights distributions for one year or several years. Also, for the sake of reproducibility of the method applied in the paper, the original raster data could be provided for the analysed years as ESM.
Line 150 The authors must mention if the Berkley Earth dataset could be downloaded for free and if these data are licensed (i.e., subject to some restrictions).
Line 173 The methodology used to determine if the stations are localised with a 30 arcsec resolution should be stated.
Line 180 The authors should be aware that updated metadata regarding all the surface-based observing stations and platforms can be obtained from the Oscar platform, which is the official repository of WIGOS metadata (https://oscar.wmo.int/surface//index.html#/). At least, this should be mentioned in the manuscript.
Line 208 It is not very clear when a buffer of 3 pixels has been applied, and when a buffer of 7 pixels has been applied.
Line 225 Why is necessary to develop and apply a gap filing procedure since an air temperature gridded dataset is also provided by the Berkeley Earth group. The missing data can be extracted from the gridded dataset.
Reference:
Dupont, W.D. and Plummer, W.D. Power and Sample Size Calculations for Studies Involving Linear Regression. New York, 1998. Print.
Author Response
REVIEWER #3
Major comments
In this work, a new approach is proposed to quantify the relationship between urbanisation dynamics and its thermal impact by relating air temperature with satellite derived nightlight data. The paper would be suitable for publication in Atmosphere, once the issues I raise below are addressed.
We acknowledge the Reviewer for his/her valuable and detailed comments. We took into account the suggestions that he/she gave us in order to clarify the purpose of our work and to improve our paper. We invite the Reviewer to see how we modified the RM, taking into account his/her comments.
I have 2 major comments:
If the accuracy of the luminosity tendency can be affected by other factors, like air pollution, is this parameter an appropriate indicator for evaluating the effects of urbanisation on air temperature? Before using the nightlight images, the authors should validate these data against yearly land-cover product extracted from satellite imageries (https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mcd12q1_v006).
Thanks for this comment. Nightlights are meant to be an alternative indicator of human presence at a fine spatial resolution (like land-cover products) which is widely used in many research fields, as outlined throughout the paper. Surely, a combination between satellite imagery and land cover information and/or census data and a validation of nightlights against other data can be of help in creating high resolution and easily-updatable settlement and population distribution maps over large areas, but we feel that it goes beyond the purposes of this work.
In the OM, we made mention to other approaches used for the analysis of relations between air temperature and land-cover/use changes. In this regard, we forgot to add relevant references, apologies for this. This has been fixed in the RM, as suggested by the Reviewer (L 103).
Are only 22 subjects enough to perform a robust linear regression analysis? The authors should analyse and comment if the sample size used for each station in the regression analysis is acceptable from the statistical point of view [1].
We thank the reviewer for this comment. We are aware that a larger sample size should surely increase the robustness of the regression analysis [29] but, for the purposes of this study case we can only rely on yearly values between 1992 and 2013. Trendline values are affected by a certain degree of uncertainty but it is somehow limited by spreading it over 4 classes of p values significance. Anyway, we revised and widened the methodology section in the RM in order to address Reviewer’s concerns (L 279 on).
Comments
Line 140 The authors should present in the manuscript or as ESM some global maps with satellite nightlights distributions for one year or several years. Also, for the sake of reproducibility of the method applied in the paper, the original raster data could be provided for the analysed years as ESM.
The original nightlights raster data are freely downloadable and provided from the NOAA National Geophysical Data Center (NGDC) under the Defense Meteorological Satellite Program (DMSP), Operational Linescan System (OLS) [1]. (L 131-132).
Line 150 The authors must mention if the Berkley Earth dataset could be downloaded for free and if these data are licensed (i.e., subject to some restrictions).
Yes, it can be downloaded for free on Berkeley Earth website [30]. This information has been added in the RM (L 169-170).
Line 173 The methodology used to determine if the stations are localised with a 30 arcsec resolution should be stated.
An iterative procedure was used in Matlab environment, based on comparison between Berkeley and WMO data. Specifically, the precision of the metadata provided by the two datasets has been compared against the minimum required precision i.e., 0.00833° (30 arcsec), and stations with coordinates less precise than this in both datasets have been removed. Some lines have been added as suggested by the Reviewer (L 192-194).
Line 180 The authors should be aware that updated metadata regarding all the surface-based observing stations and platforms can be obtained from the Oscar platform, which is the official repository of WIGOS metadata (https://oscar.wmo.int/surface//index.html#/). At least, this should be mentioned in the manuscript.
We gratefully thank the Reviewer for this comment and reference to the OSCAR/Surface that replaces the WMO Publication No. 9, Volume A, Observing Stations and WMO Catalogue of Radiosondes, that is the volume we referred to. We thus mentioned both sources of data in the RM as rightly suggested by the Reviewer (L 191, L 214).
Line 208 It is not very clear when a buffer of 3 pixels has been applied, and when a buffer of 7 pixels has been applied.
We considered different buffers ranging from 3x3 km to 7x7 km in order to detect eventual variations in the average annual Digital Number value (now denoted with L in the RM). This choice is due to the fact that effects of urban warming could be detected also some kilometres far from the instrument site, even if the major impact is evident in the first ones. For this reason, we decided to show and discuss the outcomes of the analysis in the smaller buffer (3x3 km), and we referred to the Supplementary Material for the results on larger spatial buffers. As the Reviewer may see, analyses on larger spatial buffers lead to similar results compared to the 3x3 km buffer. RM has been widened to clarify the reasons behind the use of multiple spatial buffers (L 229-237).
Line 225 Why is necessary to develop and apply a gap filing procedure since an air temperature gridded dataset is also provided by the Berkeley Earth group. The missing data can be extracted from the gridded dataset.
Thanks for this comment. A very common problem in meteorological and climatological data analyses is the presence of gaps in the time series. Indeed, failures in the measuring instruments as, for example, breaking or the interruption of data transmission, may cause gaps in the record observations. In this work, most part of the considered stations shows lacks in the temperature series in the monitoring period 1992-2013. Berkeley Earth dataset provides monthly data over the period of our interest (1992-2013). Since some stations show gaps in monthly data, we had to apply a statistically-based gap-filling procedure to derive the yearly mean temperature in the presence of limited missing data and then fill the gaps. We widened and modified the Data preparation section in order to better address the issues raised by the Reviewer (L 181 on).
Reference:
Dupont, W.D. and Plummer, W.D. Power and Sample Size Calculations for Studies Involving Linear Regression. New York, 1998. Print.
References
1. NOAA Earth Observation Group - Defense Meteorological Satellite Progam, Boulder | ngdc.noaa.gov Available online: https://ngdc.noaa.gov/eog/dmsp/downloadV4composites.html (accessed on Dec 13, 2018).
2. Ceola, S.; Laio, F.; Montanari, A. Human-impacted waters: New perspectives from global high-resolution monitoring. Water Resour. Res. 2015, 51, 7064–7079.
3. National Conference of State Legislatures States Shut Out Light Pollution Available online: http://www.ncsl.org/research/environment-and-natural-resources/states-shut-out-light-pollution.aspx (accessed on Jan 11, 2019).
4. Hansen, J.; Ruedy, R.; Sato, M.; Lo, K. Global surface temperature change. Rev. Geophys. 2010, 48, RG4004.
5. Yang, X.; Ruby Leung, L.; Zhao, N.; Zhao, C.; Qian, Y.; Hu, K.; Liu, X.; Chen, B. Contribution of urbanization to the increase of extreme heat events in an urban agglomeration in east China. Geophys. Res. Lett. 2017, 44, 6940–6950.
6. Dodman, D. Environment and Urbanization. Int. Encycl. Geogr. People, Earth, Environ. Technol. 2017, 1–9.
7. IPCC Part A: Global and Sectoral Aspects. (Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change). Clim. Chang. 2014 Impacts, Adapt. Vulnerability. 2014, 1132.
8. Zhou, Y.; Smith, S.J.; Zhao, K.; Imhoff, M.; Thomson, A.; Bond-Lamberty, B.; Asrar, G.R.; Zhang, X.; He, C.; Elvidge, C.D. A global map of urban extent from nightlights. Environ. Res. Lett. 2015, 10, 054011.
9. Santamouris, M.; Cartalis, C.; Synnefa, A.; Kolokotsa, D. On the impact of urban heat island and global warming on the power demand and electricity consumption of buildings—A review. Energy Build. 2015, 98, 119–124.
10. Peng, S.; Piao, S.; Ciais, P.; Friedlingstein, P.; Ottle, C.; Bréon, F.M.; Nan, H.; Zhou, L.; Myneni, R.B. Surface urban heat island across 419 global big cities. Environ. Sci. Technol. 2012, 46, 696–703.
11. Vörösmarty, C.J.; Green, P.; Salisbury, J.; Lammers, R.B. Global water resources: vulnerability from climate change and population growth. Science 2000, 289, 284–8.
12. Jones, G.A.; Warner, K.J. The 21st century population-energy-climate nexus. Energy Policy 2016, 93, 206–212.
13. UN DESA World Urbanization Prospects: The 2018 Revision;
14. Kovats, S.; Akhtar, R. Climate, climate change and human health in Asian cities. IIED) 2008, 20, 165–175.
15. Kalnay, E.; Cai, M. Impact of urbanization and land-use change on climate. Nature 2003, 423, 528–531.
16. McCarthy, M.P.; Best, M.J.; Betts, R.A. Climate change in cities due to global warming and urban effects. Geophys. Res. Lett. 2010, 37, 1–5.
17. Parker, D.E. Urban heat island effects on estimates of observed climate change. Wiley Interdiscip. Rev. Clim. Chang. 2010, 1, 123–133.
18. Hausfather, Z.; Menne, M.J.; Williams, C.N.; Masters, T.; Broberg, R.; Jones, D. Quantifying the effect of urbanization on u.s. Historical climatology network temperature records. J. Geophys. Res. Atmos. 2013, 118, 481–494.
19. Wickham, C.; Rohde, R.; Muller, R.A.; Wurtele, J.; Curry, J.; Groom, D.; Jacobsen, R.; Perlmutter, S.; Rosenfeld, A.; Mosher, S. Influence of Urban Heating on the Global Temperature Land Average using Rural Sites Identified from MODIS Classifications. Geoinformatics Geostatistics An Overv. 2013, 1, 1–6.
20. Arnfield, A.J. Two decades of urban climate research: A review of turbulence, exchanges of energy and water, and the urban heat island. Int. J. Climatol. 2003, 23, 1–26.
21. Jones, P.D.; Groisman, P.Y.; Coughlan, M.; Plummer, N.; Wang, W.C.; Karl, T.R. Assessment of urbanization effects in time series of surface air temperature over land. Nature 1990, 347, 169–172.
22. Oke, T.R. The energetic basis of the urban heat island. Q. J. R. Meteorol. Soc. 1982, 108, 1–24.
23. Creutzig, F. Towards typologies of urban climate and global environmental change. Environ. Res. Lett. 2015, 10.
24. Li, X.; Zhou, Y.; Asrar, G.R.; Imhoff, M.; Li, X. The surface urban heat island response to urban expansion: A panel analysis for the conterminous United States. Sci. Total Environ. 2017, 605–606, 426–435.
25. Environmental Protection Agency, U. Reducing Urban Heat Islands: Compendium of Strategies - Urban Heat Island Basics. 2008, 1–22.
26. Imhoff, M.L.; Zhang, P.; Wolfe, R.E.; Bounoua, L. Remote sensing of the urban heat island effect across biomes in the continental USA. Remote Sens. Environ. 2010, 114, 504–513.
27. Cao, Q.; Yu, D.; Georgescu, M.; Wu, J. Impacts of urbanization on summer climate in China: An assessment with coupled land-atmospheric modeling. J. Geophys. Res. 2016, 121, 10,505-10,521.
28. Jr, B.S. Short Communication Urban and rural temperature trends in proximity to large US cities : 1951 – 2000. 2007, 1807, 1801–1807.
29. Dupont, W.D.; Plummer, W.D. Power and Sample Size Calculations for Studies Involving Linear Regression. Control. Clin. Trials 1998, 19, 589–601.
30. Berkeley Earth Berkeley Earth Available online: http://berkeleyearth.org/about-data-set/ (accessed on Oct 4, 2017).
Round 2
Reviewer 1 Report
See the enclosed file.
In your reply, you mentioned that « Our purpose is not to correlate temperature and nightlights values, but rather to correlate trends i.e., to relate the temperature variations in recent
years to the corresponding variations in nightlight data. »
However, you are still taking about statistical independence.
You can do the same analysis and replace the nightlights with the amount of euros spent on e-commerce and you will
obtain the large concordance index. What is the point of your analysis. What am I missing ?
Comments for author File: Comments.pdf
Author Response
REVIEWER #1
This paper propose a new way to investigate the relation between temperature trends and urbanization based on the data from 500 weather stations worldwide and nightlights satellite measurements as proxy for urbanization. I do not think that this paper is ready for publication. The reasons are explained in the Comments section. Consequently, I will recommend that it should be rejected.
Comments
Reply on comment about Section 4.1 and Line 434: In your reply, you mentioned that « Our purpose is not to correlate temperature and nightlights values, but rather to correlate trends i.e., to relate the temperature variations in recent years to the corresponding variations in nightlight data. » However, you are still taking about statistical independence.
You can do the same analysis and replace the nightlights with the amount of euros spent on e-commerce and you will obtain the large concordance index. What is the point of your analysis. What am I missing?
Of course, we are aware that correlation is not causation. However, the mentioned paradoxical example (relating euros spent in e-commerce to increasing temperature) could be applied to falsify any relation between covarying variables: should temperature increase be related to increased C02 concentration in the atmosphere or to the amount spent in e-commerce? The difference stands in in the fact that some relations are hypothesized on a physical ground, while others are not. It is well known that air temperature does increase in urban areas, and it is therefore physically plausible that air temperature is related to urbanization (and to nightlights as a consequence). In this paper we are testing this (physically based) hypothesis using empirical data. If the Reviewer has a possible mechanistic explanation of the linkage between e-commerce and air temperature, we will be happy to consider more seriously his/her comment.
Typo
Lines 251 (and 320, 325, 423): After an equation, there is no need to indent the following line if it is in the same paragraph.
As suggested by the Reviewer, this has been corrected in the RM (L 273, 336, 339, 405).
Legend of Figure 3: On of the two occurrence of « empty dots »should be changed. I suppose you are talking about the empty and the filled dots.
Filled dots refer to the selected stations whereas empty ones to the active ones. This has been corrected in the RM (L 321), apologies for this.
Line 345: n-2 should be in italic (n - 2) as n = 22
Thanks for this comment. This has been corrected in the RM (L 356).
Line 564 : I think you meant to write « states »as in your reply. There is only one country named U.S.
Apologies for this, this has been corrected in the RM (L 24).
Reviewer 2 Report
In general, the reviewer suggests to improve the clarity of the contents throughout the whole manuscript. Furthermore, the reviewer suggests to let a English native speaker revise the use of English
Line 52: "...increase in temperature..." not "increase of temperature...". Please, correct it.
Lines 51-58: the sentences are not well connected and show repetitions. Please, reword those sentences
Lines 58-62: References are missing. Please, add references
Lines 60-61: The authors write: "changes in land cover and urban geometries (e.g., reduced vegetation in urban areas) causing a generalized reduced evapotranspiration" but the change in urban geometries is not connected to urban vegetation. The reviewer believes that the authors refer to urban canyons when they write "urban geometries". If this latter is the case, the reviewer suggests to refer "urban geometries" to the previous example (i.e., "obstruction of longwave emission by the built environment"). Altogether, please reword correctly the sentence.
Lines 67-68: The athors state: " Two different UHI types are usually considered are usually considered, i.e. surface and atmospheric UHI". Actually, UHI is defined as an increase in urban tempearture compared to the rural surroundings. Just a definition of UHI exists. However, I guess that the authors meant that UHI can be measured in different ways (i.e., by means of surface temperature, at the planetary boundary layer, and at the canopy layer) . Although, I can see that it is quite common that authors, in published literature, confuse "types of UHI"" with "types of measurements", I would advice the authors to modify the text
Lines 72-74: The sentence is not well written. The author means that UHI may affect the planetary boundary layer. Whether this is the authors mean, the reviewer suggests to reword the sentence accordingly
Lines 74-78: The sentences are not clear. The reviewer suggest to reword them
Lines 81-86: The sentences are not clear. Please, reword them
Lines 98-99: "Since urban populations will continue to increase in number and size in the future". The sentence is badly written. Please, reword it
Line 125: The reviewer suggests to substitute "urban extent" with "urban extents"
Line 136: The reviewer suggests to substitute "as proxy" with "as a proxy"
Lines 137-139: "Data are first repurposed and prepared for the analysis. Then, we assess the relationship between temperature and nightlights evolution in the last 25 years, by means of trend estimation." Which kind of data? Nightlight data? It should be specified. Which kind of temperature? Surface temperature? Air temperature? Planetary bounday layer temperature? It should be specified
Lines 152-153: What is "a new nightlight product"? Do the authors mean the most recent data? Please, specify
Lines 166-167: "Our analysis is based on calibrated images, which are not saturated at the highest intensities therefore reducing saturation and blooming bias" Do the authors mean: "Our analysis is based on calibrated images, which are not saturated at the highest intensities, in order to reducesaturation and blooming bias" ?
Lines 167-170: the motivation provided is not robust. The blooming effect might be proportional to the number of the "bright pixels". Therefore, considering that urbanization might not follow a yearly linear trend, the blooming effect might affect the retrieved imagines at different extents. Therefore, the justification provided in the manuscript suffers of conceptual flaws. The reviewer suggests to provide a more solid justification
Line 158: "Berkeley Earth (2018)". Is this a reference? Whether it is a reference it should be shown accordingly to the style of the journal
Line 204: "data includes", maybe the authors meant "data included"
Lines 210-213: Please correct the use of verbs
Lines 225: "In the case" should be substituted with "in case"
Lines 242: Is not clear to the reviewer if the "eventually" in line 242 is used as a synonimous of "finally" or "possibly". Please, check the use of the word "eventually"
Lines 253-256: please check the use of the words "suggestive" and "eventual"
Equations 9 and 10 should be explained
Discussion has not been improved that much. The reviewer already suggested to improve it
Author Response
REVIEWER #2
Comments
In general, the reviewer suggests to improve the clarity of the contents throughout the whole manuscript. Furthermore, the reviewer suggests to let a English native speaker revise the use of English
As suggested by the Reviewer, we tried to improve the methodological communication throughout the text, to better introduce the rationale and purposes of the work. Conclusions were widened, comparing and linking our findings to previous studies. As suggested, we carefully revised the manuscript style and language to comply with this requirement. We invite the Reviewer to see how we modified the manuscript, taking into account his/her suggestions.
Typo
Line 52: "...increase in temperature..." not "increase of temperature...". Please, correct it.
This has been corrected in the RM as suggested (L 48).
Lines 51-58: the sentences are not well connected and show repetitions. Please, reword those sentences
The paragraph has been rephrased as suggested, now reading as: Rapid urban growth has resulted in the expansion of built-up areas in and around cities, particularly for nations and regions experiencing demographic expansion. This plays a crucial role in near-surface warming and on temperature measurements [1–4]. Indeed, this process is known to affect the planetary boundary layer and to drive local climate changes and to lead to the relative increase in temperature within the urban area [1–3] contributing to the so-called Urban Heat Island (UHI) effect within cities. The so-called Urban Heat Island (UHI) effect identifies a sort of microclimate within cities, leading to a difference of temperature between urban and surrounding non-urban areas, characterized by higher and lower temperatures, respectively, particularly during nighttime [1,5].” (L 42-56)
Lines 58-62: References are missing. Please, add references
References [6–8] have been added as suggested (L 60).
Lines 67-68: The authors state: " Two different UHI types are usually considered are usually considered, i.e. surface and atmospheric UHI". Actually, UHI is defined as an increase in urban temperature compared to the rural surroundings. Just a definition of UHI exists. However, I guess that the authors meant that UHI can be measured in different ways (i.e., by means of surface temperature, at the planetary boundary layer, and at the canopy layer). Although, I can see that it is quite common that authors, in published literature, confuse "types of UHI"" with "types of measurements", I would advice the authors to modify the text .
Thanks for this comment. We are aware that the text was misleading in this part, as rightly suggested by the Reviewer, thus it has been rephrased and modified as suggested by the Reviewer, referring to the different type of measurement and mechanisms as well. Now, it read as: “UHI measurement can be defined for different layers of the urban atmosphere and even surfaces. Since their underlying mechanisms and related measurements are different, it is thus important to distinguish between different urban heat islands [9]. Surface UHI refers to the land surface temperature specifically [10], whereas the atmospheric UHI considers the land air temperature as measured by land-based weather stations [11]. Because land air temperature is commonly used in climate warming analyses, in this paper we consider atmospheric UHI only.” (L 64-71)
Lines 72-74: The sentence is not well written. The author means that UHI may affect the planetary boundary layer. Whether this is the authors mean, the reviewer suggests to reword the sentence accordingly.
The sentence has been rephrased as suggested by the Reviewer, now reading as: “Previous works highlight how UHI affects the planetary boundary layer and how urban-atmospheric interactions control urban-induced impacts not only at the local scale (i.e., individual cities), but also over regional scales [12,13].” (L 72-75)
Lines 74-78: The sentences are not clear. The reviewer suggest to reword them
The sentence has been rephrased as suggested by the Reviewer, now reading as: “For instance, Georgescu et al. [13] calculated that urban expansion, separate from greenhouse gas-induced climate change, is projected to increase near-surface temperature up to 2 °C at both local and regional scale. More recent work has advanced this calculation by looking at the dynamic interaction and relative impact of urban to climate change effects on near surface temperature through the 24-hour day/night period [14].” (L 75-80)
Lines 81-86: The sentences are not clear. Please, reword them
The sentence has been rephrased as suggested by the Reviewer, now reading as: “At the local level, the contribution of UHI to global warming during summer is locally important, therefore originating immediate critical situations for human health and well-being [15,16]. At the regional level, urban warming is found to exhibit more relevant effects after some years of urban expansion [17]”. (L 83-88).
Lines 98-99: "Since urban populations will continue to increase in number and size in the future". The sentence is badly written. Please, reword it
The sentence now reads as: “Since urban population is expected to increase in the future [18]…”. (L 101-102)
Line 125: The reviewer suggests to substitute "urban extent" with "urban extents"
The sentence has been corrected as suggested (L 150).
Line 136: The reviewer suggests to substitute "as proxy" with "as a proxy"
The sentence has been corrected as suggested (L 160).
Lines 137-139: "Data are first repurposed and prepared for the analysis. Then, we assess the relationship between temperature and nightlights evolution in the last 25 years, by means of trend estimation." Which kind of data? Nightlight data? It should be specified. Which kind of temperature? Surface temperature? Air temperature? Planetary boundary layer temperature? It should be specified
The sentence has been rephrased as suggested by the Reviewer, now reading as: “Nightlights and air temperature data are first repurposed and prepared for the analysis. Then, we assess the relationship between air temperature and nightlights evolution in the last 25 years, by means of trend estimation.” (L 161-163).
Lines 152-153: What is "a new nightlight product"? Do the authors mean the most recent data? Please, specify
We refer to another product, extensively used in previous papers, which considers the average nightlight value in each pixel. The sentence has been rephrased, now reading as: “In case of years presenting overlapping datasets, the averaged nighttime brightness value, extensively used in recent papers [19], is therefore employed.”(L 176-178)
Lines 166-167: "Our analysis is based on calibrated images, which are not saturated at the highest intensities therefore reducing saturation and blooming bias" Do the authors mean: "Our analysis is based on calibrated images, which are not saturated at the highest intensities, in order to reduce saturation and blooming bias"
The Reviewer is right and we modified the text as suggested (L 192-193).
Lines 167-170: the motivation provided is not robust. The blooming effect might be proportional to the number of the "bright pixels". Therefore, considering that urbanization might not follow a yearly linear trend, the blooming effect might affect the retrieved imagines at different extents. Therefore, the justification provided in the manuscript suffers of conceptual flaws. The reviewer suggests to provide a more solid justification
We are aware that, when dealing with satellite imagery, eventual divergences could be related to saturation and blooming effects. They occur primarily in developed countries and highly populated urban areas, where the intensity of lights is high [15]. Across developing countries, the blooming effect (i.e. the overestimation of lit areas), is particularly evident, because of the presence of broad pitch-dark areas, potentially leading to the inclusion of some rural areas or peripheric in urban areas [20].
As detailed at L in the RM, we based our analysis on calibrated images and adjusted DN values, having the advantage of not being saturated at the highest intensities (see [11] for further details). We also account for this factor by controlling for unlit cells (i.e., pitch dark areas showing a stable DN equal to zero along the whole period of analysis). Furthermore, we are not focusing on nightlights absolute values, rather on trends between 1992 and 2013. Also recent studies [21] assessed the temporal variation of urbanization as we did by analysing the trend over time and by means of a linear trend equation. As in our study, the interceptor b is neglected in this analysis as the authors are interested in the trend rather than the actual level of urbanization. The authors state that since nightlights (and thus urbanization) suffer from annual fluctuation (even after calibration), it is convenient to focus on the slope of the straight line as a straightforward way to summarize the trend line. In our opinion, if a blooming effect is detected, this contamination will equally affect all years along the time series, without any significant influence on the overall trend. The reference has been added in the RM and we modified the text in this part in order to make this concept more clear (L 190-197).
Line 188: "Berkeley Earth (2018)". Is this a reference? Whether it is a reference it should be shown accordingly to the style of the journal
Apologies for this, the reference is [22], this has been corrected in the RM (L 213).
Line 204: "data includes", maybe the authors meant "data included"
The Reviewer is right and we modified the text as suggested (L 229).
Lines 210-213: Please correct the use of verbs
The use of verbs has been checked and corrected where appropriate (L 233-236).
Lines 225: "In the case" should be substituted with "in case"
The sentence has been corrected as suggested (L 250).
Lines 242: Is not clear to the reviewer if the "eventually" in line 242 is used as a synonimous of "finally" or "possibly". Please, check the use of the word "eventually"
We agree with the Reviewer that “eventually” is misleading in this context, thus it has been changed to possibly (L 268). The use of the word has been checked throughout the text (L 277, 283, 195).
Lines 253-256: please check the use of the words "suggestive" and "eventual"
The use of the word has been checked throughout the text (L) and words have been changes if necessary (L 275).
Equations 9 and 10 should be explained
A line of explanation has been added (L 392-393)
Discussion has not been improved that much. The reviewer already suggested to improve it
As suggested by the Reviewer, the discussion has been widened, now reading e.g., as: “Recent studies at very local scale show that the size of the source radius (i.e., buffer) matters when assessing urban influences on station temperature trends, and the near-station surroundings have been demonstrated to be more influential for temperature values [23]. Nevertheless, analyses with larger spatial buffers tend to confirm…” []. As cities grow, temperatures in pre-existing urban areas in some cases are pretty stable, while temperatures only rise in newly urbanized areas [78]. This is confirmed by recent studies showing that, when temperatures rise in pre-existing urban areas, the air temperature change is smaller than newly urbanized ones [79]. Moreover, based on UN projections [9] and further previous studies, urban population in some cities across Europe is expected to decline by 2050 [9,76,77]…[]. There exist relevant urbanization patterns in Europe that can be captured by nightlights and recent studies confirm that there is a strong variation of urbanization dynamics within single countries and regions, in line with our findings [21]…”
We thus invite the Reviewer to see how we modified the text (L 477 on).
References
1. Parker, D.E. Urban heat island effects on estimates of observed climate change. Wiley Interdiscip. Rev. Clim. Chang. 2010, 1, 123–133.
2. Creutzig, F. Towards typologies of urban climate and global environmental change. Environ. Res. Lett. 2015, 10.
3. Li, X.; Zhou, Y.; Asrar, G.R.; Imhoff, M.; Li, X. The surface urban heat island response to urban expansion: A panel analysis for the conterminous United States. Sci. Total Environ. 2017, 605–606, 426–435.
4. Chapman, S.; Watson, J.E.M.; Salazar, A.; Thatcher, M.; McAlpine, C.A. The impact of urbanization and climate change on urban temperatures: a systematic review. Landsc. Ecol. 2017, 32, 1921–1935.
5. Pielke, R.A.; Matsui, T. Should light wind and windy nights have the same temperature trends at individual levels even if the boundary layer averaged heat content change is the same? Geophys. Res. Lett. 2005, 32, L21813.
6. Environmental Protection Agency, U. Reducing Urban Heat Islands: Compendium of Strategies - Urban Heat Island Basics. 2008, 1–22.
7. Santamouris, M. Energy and Climate in the Urban Built Environment; Routledge, 2013; ISBN 9781315073774.
8. Voogt, J..; Oke, T.. Thermal remote sensing of urban climates. Remote Sens. Environ. 2003, 86, 370–384.
9. Oke, T.R. The energetic basis of the urban heat island. Q. J. R. Meteorol. Soc. 1982, 108, 1–24.
10. Peng, S.; Piao, S.; Ciais, P.; Friedlingstein, P.; Ottle, C.; Bréon, F.M.; Nan, H.; Zhou, L.; Myneni, R.B. Surface urban heat island across 419 global big cities. Environ. Sci. Technol. 2012, 46, 696–703.
11. Wickham, C.; Rohde, R.; Muller, R.A.; Wurtele, J.; Curry, J.; Groom, D.; Jacobsen, R.; Perlmutter, S.; Rosenfeld, A.; Mosher, S. Influence of Urban Heating on the Global Temperature Land Average using Rural Sites Identified from MODIS Classifications. Geoinformatics Geostatistics An Overv. 2013, 1, 1–6.
12. Imhoff, M.L.; Zhang, P.; Wolfe, R.E.; Bounoua, L. Remote sensing of the urban heat island effect across biomes in the continental USA. Remote Sens. Environ. 2010, 114, 504–513.
13. Georgescu, M.; Morefield, P.E.; Bierwagen, B.G.; Weaver, C.P. Urban adaptation can roll back warming of emerging megapolitan regions. Proc. Natl. Acad. Sci. 2014, 111, 2909–2914.
14. Krayenhoff, E.S.; Moustaoui, M.; Broadbent, A.M.; Gupta, V.; Georgescu, M. Diurnal interaction between urban expansion, climate change and adaptation in US cities. Nat. Clim. Chang. 2018, 8, 1097–1103.
15. Chen, X.; Nordhaus, W.D. Using luminosity data as a proxy for economic statistics. 2011, 108.
16. Chand, T.R.K.; Badarinath, K.V.S.; Elvidge, C.D.; Tuttle, B.T. Spatial characterization of electrical power consumption patterns over India using temporal DMSP-OLS night-time satellite data. Int. J. Remote Sens. 2009, 30, 647–661.
17. Cao, Q.; Yu, D.; Georgescu, M.; Wu, J. Impacts of urbanization on summer climate in China: An assessment with coupled land-atmospheric modeling. J. Geophys. Res. 2016, 121, 10,505-10,521.
18. IPCC Part A: Global and Sectoral Aspects. (Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change). Clim. Chang. 2014 Impacts, Adapt. Vulnerability. 2014, 1132.
19. Ceola, S.; Laio, F.; Montanari, A. Satellite nighttime lights reveal increasing human exposure to floods worldwide. Geophys. Res. Lett. 2014, 41, 7184–7190.
20. Huang, Q.; Yang, X.; Gao, B.; Yang, Y.; Zhao, Y. Application of DMSP/OLS nighttime light images: A meta-analysis and a systematic literature review. Remote Sens. 2014, 6, 6844–6866.
21. Stathakis, D.; Tselios, V.; Faraslis, I. Urbanization in European regions based on night lights. Remote Sens. Appl. Soc. Environ. 2015, 2, 26–34.
22. Berkeley Earth Berkeley Earth Available online: http://berkeleyearth.org/about-data-set/ (accessed on Oct 4, 2017).
23. Lindén, J.; Esper, J.; Holmer, B. Using land cover, population, and night light data for assessing local temperature differences in Mainz, Germany. J. Appl. Meteorol. Climatol. 2015, 54, 658–670.
24. NASA EOSDIS Land Processes DAAC MCD12Q1 V006 | LP DAAC :: NASA Land Data Products and Services Available online: https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mcd12q1_v006 (accessed on Jan 16, 2019).
25. Vancutsem, C.; Ceccato, P.; Dinku, T.; Connor, S.J. Evaluation of MODIS land surface temperature data to estimate air temperature in different ecosystems over Africa. Remote Sens. Environ. 2010, 114, 449–465.
26. Schneider, A.; Friedl, M.A.; Potere, D. Mapping global urban areas using MODIS 500-m data: New methods and datasets based on “urban ecoregions.” Remote Sens. Environ. 2010, 114, 1733–1746.
27. Friedl, M..; McIver, D..; Hodges, J.C..; Zhang, X..; Muchoney, D.; Strahler, A..; Woodcock, C..; Gopal, S.; Schneider, A.; Cooper, A.; et al. Global land cover mapping from MODIS: algorithms and early results. Remote Sens. Environ. 2002, 83, 287–302.
28. Fu, P.; Weng, Q. A time series analysis of urbanization induced land use and land cover change and its impact on land surface temperature with Landsat imagery. Remote Sens. Environ. 2016, 175, 205–214.
29. Zhou, Y.; Smith, S.J.; Zhao, K.; Imhoff, M.; Thomson, A.; Bond-Lamberty, B.; Asrar, G.R.; Zhang, X.; He, C.; Elvidge, C.D. A global map of urban extent from nightlights. Environ. Res. Lett. 2015, 10, 054011.
30. Small, C.; Elvidge, C.D.; Balk, D.; Montgomery, M. Spatial scaling of stable night lights. Remote Sens. Environ. 2011, 115, 269–280.
31. Marconcini, M.; Metz, A.; Esch, T.; Zeidler, J. Global Urban Growth Monitoring By Means of Sar Data. 2014, 1477–1480.
32. Yang, X.; Ruby Leung, L.; Zhao, N.; Zhao, C.; Qian, Y.; Hu, K.; Liu, X.; Chen, B. Contribution of urbanization to the increase of extreme heat events in an urban agglomeration in east China. Geophys. Res. Lett. 2017, 44, 6940–6950.
33. Zhang, P.; Imhoff, M.L.; Wolfe, R.E.; Bounoua, L. Canadian Journal of Remote Sensing Journal Canadien de Teledetection Characterizing urban heat islands of global settlements using MODIS and nighttime lights products Characterizing urban heat islands of global settlements using MODIS and nighttime lights. 2014.
34. Arino, O.; Gross, D.; Ranera, F.; Leroy, M.; Bicheron, P.; Brockman, C.; Defourny, P.; Vancutsem, C.; Achard, F.; Durieux, L.; et al. GlobCover: ESA service for global land cover from MERIS. In Proceedings of the 2007 IEEE International Geoscience and Remote Sensing Symposium; IEEE, 2007; pp. 2412–2415.
35. Bartholomé, E.; Belward, A.S. GLC2000: a new approach to global land cover mapping from Earth observation data. Int. J. Remote Sens. 2005, 26, 1959–1977.
36. Bagan, H.; Yamagata, Y. Analysis of urban growth and estimating population density using satellite images of nighttime lights and land-use and population data. GIScience Remote Sens. 2015, 52, 765–780.
Reviewer 3 Report
The authors generally did a good job in answering and accounting for my concerns of the originally submitted manuscript. They clearly invested time in restructuring the manuscript the improve story line and readability. Nevertheless, I think there is on major issue that has not been appropriately addressed.
The authors have not validated the nightlights images against existing landcover products, as indicated in my previous comments. They also mentioned that relevant references related to my asking have been inserted to the manuscript, but in the revised manuscript (L103, document atmosphere-422404-peer-review-v2.pdf) there are no new items added and commented.
I strongly recommend improving this, at least by mentioning the relevant works which validated the nightlights images against landcover products. I don't want to be a pain in the neck, but I really think this would make the paper way better.
Author Response
REVIEWER #3
The authors generally did a good job in answering and accounting for my concerns of the originally submitted manuscript. They clearly invested time in restructuring the manuscript the improve story line and readability. Nevertheless, I think there is on major issue that has not been appropriately addressed.
The authors have not validated the nightlights images against existing landcover products, as indicated in my previous comments. They also mentioned that relevant references related to my asking have been inserted to the manuscript, but in the revised manuscript (L103, document atmosphere-422404-peer-review-v2.pdf) there are no new items added and commented.
I strongly recommend improving this, at least by mentioning the relevant works which validated the nightlights images against landcover products. I don't want to be a pain in the neck, but I really think this would make the paper way better.
Apologies for this. Following Reviewer’s recommendation, we mentioned and commented the relevant works validating nightlights against landcover products e.g., MODIS.: “New techniques and approaches could be helpful in detecting relationships and feedbacks between air temperature and land-use changes. These efforts result to independent and relatively accurate estimates of urban shape and extent. Remote-sensing products, primarily MODIS-500 maps [24–27] and Landsat data [28], satellite nighttime images [29,30] and Synthetic Aperture Radar (SAR) data [31] are widely used to map global urban extent and to identify and geolocalize rural and urban stations [11,32]. More in detail, in recent studies [30] day–night composites combine nightlights and Landsat to show consistencies in land cover and night light brightness. Other relevant works spatially assess UHI signature on land surface temperature amplitude and its mutual relations with development size, intensity and in different biomes for over 3000 cities worldwide combining MODIS and nighttime lights products i.e., nightlight based impervious surface area (ISA) [33]. Nightlight ISA and Landsat ISA for cities in the continental US are first compared to bridge from previous studies. Results highlight significant positive relationships between UHI magnitude, ecological setting and ISA, and that nightlights are good estimator of urban sprawl and more objective than methods based on population density. Zhou et al. [29] develop a method to map urban extent from nightlights at global scale and compare it at the pixel level and regionally with other widely five used global urban products, including MODIS [26], GlobCover [34] and GLC2000 [35]. Results show that nightlight map produced by the author is in high agreement with the other global urban area map products. Bagan and Yamagata [36] also show that the combined use of land cover data and nightlights are both good predictor of population density in Japan. At local scale, recent studies [23] asses the properties of temperature station network and the potential urban influence on temperature records by means of land cover, population density and nightlights at increasing buffer around the weather station. The three methods are found to be highly correlated thus indicating that nightlights and population density are good proxies of urban areas.
Thus, based on comparison with other existing global urban maps, nighttime satellite images are demonstrated to be a good proxy of urban extent and allows for temporal dynamics analyses of urban areas [23,29]. It is shown that satellite nightlights maps…” (L 116 on).
References
1. Parker, D.E. Urban heat island effects on estimates of observed climate change. Wiley Interdiscip. Rev. Clim. Chang. 2010, 1, 123–133.
2. Creutzig, F. Towards typologies of urban climate and global environmental change. Environ. Res. Lett. 2015, 10.
3. Li, X.; Zhou, Y.; Asrar, G.R.; Imhoff, M.; Li, X. The surface urban heat island response to urban expansion: A panel analysis for the conterminous United States. Sci. Total Environ. 2017, 605–606, 426–435.
4. Chapman, S.; Watson, J.E.M.; Salazar, A.; Thatcher, M.; McAlpine, C.A. The impact of urbanization and climate change on urban temperatures: a systematic review. Landsc. Ecol. 2017, 32, 1921–1935.
5. Pielke, R.A.; Matsui, T. Should light wind and windy nights have the same temperature trends at individual levels even if the boundary layer averaged heat content change is the same? Geophys. Res. Lett. 2005, 32, L21813.
6. Environmental Protection Agency, U. Reducing Urban Heat Islands: Compendium of Strategies - Urban Heat Island Basics. 2008, 1–22.
7. Santamouris, M. Energy and Climate in the Urban Built Environment; Routledge, 2013; ISBN 9781315073774.
8. Voogt, J..; Oke, T.. Thermal remote sensing of urban climates. Remote Sens. Environ. 2003, 86, 370–384.
9. Oke, T.R. The energetic basis of the urban heat island. Q. J. R. Meteorol. Soc. 1982, 108, 1–24.
10. Peng, S.; Piao, S.; Ciais, P.; Friedlingstein, P.; Ottle, C.; Bréon, F.M.; Nan, H.; Zhou, L.; Myneni, R.B. Surface urban heat island across 419 global big cities. Environ. Sci. Technol. 2012, 46, 696–703.
11. Wickham, C.; Rohde, R.; Muller, R.A.; Wurtele, J.; Curry, J.; Groom, D.; Jacobsen, R.; Perlmutter, S.; Rosenfeld, A.; Mosher, S. Influence of Urban Heating on the Global Temperature Land Average using Rural Sites Identified from MODIS Classifications. Geoinformatics Geostatistics An Overv. 2013, 1, 1–6.
12. Imhoff, M.L.; Zhang, P.; Wolfe, R.E.; Bounoua, L. Remote sensing of the urban heat island effect across biomes in the continental USA. Remote Sens. Environ. 2010, 114, 504–513.
13. Georgescu, M.; Morefield, P.E.; Bierwagen, B.G.; Weaver, C.P. Urban adaptation can roll back warming of emerging megapolitan regions. Proc. Natl. Acad. Sci. 2014, 111, 2909–2914.
14. Krayenhoff, E.S.; Moustaoui, M.; Broadbent, A.M.; Gupta, V.; Georgescu, M. Diurnal interaction between urban expansion, climate change and adaptation in US cities. Nat. Clim. Chang. 2018, 8, 1097–1103.
15. Chen, X.; Nordhaus, W.D. Using luminosity data as a proxy for economic statistics. 2011, 108.
16. Chand, T.R.K.; Badarinath, K.V.S.; Elvidge, C.D.; Tuttle, B.T. Spatial characterization of electrical power consumption patterns over India using temporal DMSP-OLS night-time satellite data. Int. J. Remote Sens. 2009, 30, 647–661.
17. Cao, Q.; Yu, D.; Georgescu, M.; Wu, J. Impacts of urbanization on summer climate in China: An assessment with coupled land-atmospheric modeling. J. Geophys. Res. 2016, 121, 10,505-10,521.
18. IPCC Part A: Global and Sectoral Aspects. (Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change). Clim. Chang. 2014 Impacts, Adapt. Vulnerability. 2014, 1132.
19. Ceola, S.; Laio, F.; Montanari, A. Satellite nighttime lights reveal increasing human exposure to floods worldwide. Geophys. Res. Lett. 2014, 41, 7184–7190.
20. Huang, Q.; Yang, X.; Gao, B.; Yang, Y.; Zhao, Y. Application of DMSP/OLS nighttime light images: A meta-analysis and a systematic literature review. Remote Sens. 2014, 6, 6844–6866.
21. Stathakis, D.; Tselios, V.; Faraslis, I. Urbanization in European regions based on night lights. Remote Sens. Appl. Soc. Environ. 2015, 2, 26–34.
22. Berkeley Earth Berkeley Earth Available online: http://berkeleyearth.org/about-data-set/ (accessed on Oct 4, 2017).
23. Lindén, J.; Esper, J.; Holmer, B. Using land cover, population, and night light data for assessing local temperature differences in Mainz, Germany. J. Appl. Meteorol. Climatol. 2015, 54, 658–670.
24. NASA EOSDIS Land Processes DAAC MCD12Q1 V006 | LP DAAC :: NASA Land Data Products and Services Available online: https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mcd12q1_v006 (accessed on Jan 16, 2019).
25. Vancutsem, C.; Ceccato, P.; Dinku, T.; Connor, S.J. Evaluation of MODIS land surface temperature data to estimate air temperature in different ecosystems over Africa. Remote Sens. Environ. 2010, 114, 449–465.
26. Schneider, A.; Friedl, M.A.; Potere, D. Mapping global urban areas using MODIS 500-m data: New methods and datasets based on “urban ecoregions.” Remote Sens. Environ. 2010, 114, 1733–1746.
27. Friedl, M..; McIver, D..; Hodges, J.C..; Zhang, X..; Muchoney, D.; Strahler, A..; Woodcock, C..; Gopal, S.; Schneider, A.; Cooper, A.; et al. Global land cover mapping from MODIS: algorithms and early results. Remote Sens. Environ. 2002, 83, 287–302.
28. Fu, P.; Weng, Q. A time series analysis of urbanization induced land use and land cover change and its impact on land surface temperature with Landsat imagery. Remote Sens. Environ. 2016, 175, 205–214.
29. Zhou, Y.; Smith, S.J.; Zhao, K.; Imhoff, M.; Thomson, A.; Bond-Lamberty, B.; Asrar, G.R.; Zhang, X.; He, C.; Elvidge, C.D. A global map of urban extent from nightlights. Environ. Res. Lett. 2015, 10, 054011.
30. Small, C.; Elvidge, C.D.; Balk, D.; Montgomery, M. Spatial scaling of stable night lights. Remote Sens. Environ. 2011, 115, 269–280.
31. Marconcini, M.; Metz, A.; Esch, T.; Zeidler, J. Global Urban Growth Monitoring By Means of Sar Data. 2014, 1477–1480.
32. Yang, X.; Ruby Leung, L.; Zhao, N.; Zhao, C.; Qian, Y.; Hu, K.; Liu, X.; Chen, B. Contribution of urbanization to the increase of extreme heat events in an urban agglomeration in east China. Geophys. Res. Lett. 2017, 44, 6940–6950.
33. Zhang, P.; Imhoff, M.L.; Wolfe, R.E.; Bounoua, L. Canadian Journal of Remote Sensing Journal Canadien de Teledetection Characterizing urban heat islands of global settlements using MODIS and nighttime lights products Characterizing urban heat islands of global settlements using MODIS and nighttime lights. 2014.
34. Arino, O.; Gross, D.; Ranera, F.; Leroy, M.; Bicheron, P.; Brockman, C.; Defourny, P.; Vancutsem, C.; Achard, F.; Durieux, L.; et al. GlobCover: ESA service for global land cover from MERIS. In Proceedings of the 2007 IEEE International Geoscience and Remote Sensing Symposium; IEEE, 2007; pp. 2412–2415.
35. Bartholomé, E.; Belward, A.S. GLC2000: a new approach to global land cover mapping from Earth observation data. Int. J. Remote Sens. 2005, 26, 1959–1977.
36. Bagan, H.; Yamagata, Y. Analysis of urban growth and estimating population density using satellite images of nighttime lights and land-use and population data. GIScience Remote Sens. 2015, 52, 765–780.
