- Email: [email protected]
Anthropogenic signals in Iranian extreme temperature indices
Atmospheric Research 169 (2016) 96–101
Contents lists available at ScienceDirect
Atmospheric Research journal homepage: www.elsevier.com/locate/atmos
Anthropogenic signals in Iranian extreme temperature indices Robert C. Balling Jr. a,⁎, Mohammad Sadegh Keikhosravi Kiany b, Shouraseni Sen Roy c a b c
School of Geographical Sciences and Urban Planning, Arizona State University, Tempe, AZ 85287, USA Faculty of Geographical Sciences and Planning, University of Esfahan, Esfahan, Iran Department of Geography and Regional Studies, University of Miami, Coral Gables, FL, USA
a r t i c l e
i n f o
Article history: Received 10 April 2015 Received in revised form 21 September 2015 Accepted 27 September 2015 Available online 13 October 2015 Keywords: Temperature extremes Iran Population Weekly cycle
a b s t r a c t We analyzed spatial and temporal patterns in temperature extremes from 31 stations located throughout Iran for the period 1961 to 2010. As with many other parts of the globe, we found that the number of days (a) with high maximum temperatures was rising, (b) with high minimum temperatures was rising, and (c) with low minimum temperatures was declining; all of these trends were statistically significant at the 0.05 level of confidence. Population records from 1956 to 2011 at the station locations allowed us to reveal that the rate of human population growth was positively related to the increase in the number of days with high maximum temperatures and negatively related to days with low maximum temperatures. Our research shows a number of identifiable anthropogenic signals in the temperature records from Iran, but unlike most other studies, the signals are stronger with indices related to maximum, not minimum, temperatures. © 2015 Elsevier B.V. All rights reserved.
1. Introduction There is general consensus in the climate science community regarding the general increase in global mean surface temperatures since the later part of the 19th century. Specifically, the observed temperatures recorded over the last three decades show a successive increase, with the decade of the 2000s as the warmest on record (Stocker et al., 2013). The analysis on long term surface level observations at the global scale indicate a warming of 0.85 [0.65 to 1.06]°C during 1880–2012, ≈0.89 [0.69 to 1.08]°C during 1901–2012, and ≈0.72 [0.49 to 0.89] °C during 1951–2012 based on three independentlyproduced data sets (Stocker et al., 2013). However, there are substantial temporal variations in these trends that are caused by local level teleconnections such as El Niño Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO), as well as the recently observed global warming hiatus since 1998 (Kosaka and Xie, 2013). Other than temporal variations, there are substantial regional variations in the trends in temperatures driven by land use land cover changes including the role of urban heat island (UHI), deforestation, agriculture-related irrigation effects, as well as other anthropogenic activities (Stocker et al., 2013). One of the important aspects of long term temperature trends is the occurrence of extreme temperatures. The results from a recently published report on extreme events by Seneviratne et al. (2012) indicated an overall decrease in cold extremes and increase in warm extremes since the middle of the 20th century. For instance, Donat et al. (2013a) found a significant warming of both maximum (TX) and minimum (TN) temperatures over global land areas since 1950.
⁎ Corresponding author.
http://dx.doi.org/10.1016/j.atmosres.2015.09.030 0169-8095/© 2015 Elsevier B.V. All rights reserved.
Additionally, a greater shift in the distribution of nighttime temperatures compared to the daytime temperatures was observed in several studies (Ballester et al., 2010; Simolo et al., 2011; Donat and Alexander, 2012; Hansen et al., 2012). However, there are substantial variations in the overall trends and the confidence levels associated with those trends. For instance, in the Middle East, there is medium confidence in the increased occurrence of daytime TX above the 90th percentile and nighttime TN above the 90th percentile (Donat et al., 2013b; Zhang et al., 2005). There is also medium confidence in the increasing trends observed in the incidences of heat waves and warm spells in the Middle East (Perkins et al., 2012; Donat et al., 2013a). Many papers have appeared in Atmospheric Research in the past few years on the topic of trends in extreme temperatures. Several have reported overall warming in Portugal (de Lima et al., 2013), Rajasthan, India (Pingale et al., 2014), and Iran (Araghi et al., 2015a,2015b). However, many others have revealed evidence that the warming of TN have exceeded the warming of TX as reflected in a variety of extreme temperature indices. These findings have come from analyses in Greece (Nastos and Kapsomenakis, 2015), the Philippines (Cinco et al., 2014), northwest China (Deng et al., 2014), Iberia (Fernández-Montes et al., 2012), South America (Rusticucci, 2012), and Modena, Italy (Boccolari and Malmusi, 2013). In this investigation, we focus on trends in extreme temperatures throughout Iran for the period 1961 to 2010. Over the past decade, a number of important studies have documented the overall warming of the country during the last half century. Ghahraman (2006) investigated the long term trend of mean annual temperature at 34 synoptic stations for the period from 1968 to 1998 and found that at the 0.05 level of significance 44%, 15%, and 41% of the stations had a positive, negative, and zero trend respectively. Rahimzadeh et al. (2009)
R.C. Balling Jr. et al. / Atmospheric Research 169 (2016) 96–101
Fig 1. General location and topography of Iran.
Fig. 2. Location of 31 stations in Iran.
97
98
R.C. Balling Jr. et al. / Atmospheric Research 169 (2016) 96–101
examined temperature records from 27 stations in Iran and found positive trends in the occurrence of hot days and tropical nights. Tabari et al. (2011) examined trend in annual air temperatures in the west, south, and southwest of Iran for the period 1966–2005. They concluded that an overall warming trend exists and with the trend for TN (0.49 °C dec− 1) slightly exceeding the trend for TX (0.45 °C dec−1). Tabari and Talaee (2011) investigated trends in air temperature time series for 20 stations in the western half of Iran during 1966–2005. The annual TX and TN values showed a positive trend in 85% of the stations and a negative trend in 15% of the stations in the study region. Tabari et al. (2012) analyzed the data of 29 stations in Iran for a period of 40 years from 1966 to 2005 for the existence of monotonic trends and shift changes in the annual, seasonal and monthly mean temperature. It was found that the annual mean air temperature increase at 25 out of the 29 stations showed significant monotonic trends. The spatial analysis of the mean air temperature trends revealed the highest numbers of significant monotonic trends in the larger cities of Iran. Zarenistanak et al. (2014) analyzed trend and change point detection of annual and seasonal precipitation and temperature time series over the period 1950–2007. Their results revealed a significant increase during the summer and spring seasons. TX was more stable than TN, and winter was stable compared to the summer, spring, and autumn seasons. The results of the change point analysis indicated that most of the statistically significant positive mutation points in all temperatures began in the 1990s. Araghi et al. (2015a, b) investigated trends in Iran in the days with TX ≥ 30 °C and TN ≤0 °C from 1961 to 2010 for 30 synoptic stations. They concluded that for 67% of stations, days wtih TN ≤0 °C showed a significant negative trend, while only 40% of stations showed a significant positive trend in days with TX ≥ 30 °C. Ghasemi (2015) studied the variations of Iranian temperature during 1961–2010 using 38
stations across the country. Mann–Kendall tests indicated an overall dominating positive temperature trend over most parts of Iran with a TN increase (0.34 °C dec−1) more than twice the increase for TX (0.15 °C dec−1). The findings of this study also showed that warm climate regions in Iran are warming at a higher rate than cold climate areas. The evidence is overwhelming that near-surface air temperatures in Iran have increased significantly over the past half century, and as with many other parts of the globe, the increase in minimum temperatures has exceeded the increase in maximum temperatures. To significantly add to the literature on temperature trends in Iran, we evaluate the potential role of local population growth on these trends to possibly enlighten a signal in the data that is likely related to anthropogenic activities. 2. Materials and methods Iran is located between 25° and 40°N and 45° and 60°E and is a mountainous country bordering the Gulf of Oman, the Persian Gulf, and the Caspian Sea (Fig. 1). The total area of Iran is 1.648 × 106 km2 which represents 0.32% of the Earth's surface. Overall, sixty percent of Iran is covered by mountains, with the central part of the country consisting of two dry deserts: the Dasht-e-Kavir and the Dasht-e-Lut. The Alborz range in the north, close to the Caspian Sea, extends in an east-west direction with a maximum elevation of approximately 5000 m. The Zagros Mountains are aligned in a northwest to southeast direction and reach a maximum elevation of approximately 3500 m. The country can be divided in at least four different climate zones. The western and southwestern area are hot and dry with precipitation increasing toward the Persian Gulf and the Turkish border. The eastern
Fig. 3. Population changes in Iran from 1956 to 2011 (note that different color scales are used for smaller and larger urban areas).
R.C. Balling Jr. et al. / Atmospheric Research 169 (2016) 96–101
and northern areas of Iran have a Mediterranean climate with dry summers and mild and moderately wet winters. The mountainous regions in the north are Mediterranean-like in terms of precipitation distribution, but the elevation and latitude contribute to substantial snowfall in the winter season. We obtained daily temperature data from the Islamic Republic of Iran Meteorological Organization (IRIMO). IRIMO publishes the quality controlled data with nearly three years of delay. The original data passes WMO procedures before publishing (see WMO (2003) for details). The data are checked for miswritten codes and procedures are used to detect erroneous observations and check for doubtful records. The final quality controlled data are then published in txt files by IRIMO. There are more than 150 synoptic weather stations in Iran but we had to limit our analyses to only 31 stations because of the shortness of the period in many of the stations and gaps in many of the records. Our selected weather stations covered all the climatic regions of the country (Fig. 2) and range from small towns to the largest cities in Iran (Fig. 3). All the weather stations have data for the period from 1961 to 2010. Among the stations used, nearly all of them have more than 99% of available days except for the Saghez, Oroomieh, and Kermanshah stations that have 98.02%, 98.60% and 98.97% of available days, respectively. We calculated the nearest-neighbor statistic, r, for the 31-station network which represents the ratio of the average distance from each station to its nearest neighbor to the average distance between nearest neighbors calculated for a random spatial distribution (Clark and Evans, 1954). The values of r can range from 0 for a completely clustered distribution (all stations have exactly the same latitude and longitude) to 1.0 for a random pattern to 2.15 for a perfectly uniform distribution. The value of r for our 31-station network is 1.20 falling in the desirable random to uniform range. In order to investigate the role of population increase in cities on extreme temperature changes, population data were obtained from the official reports of Statistical Centre of Iran. We collected population data from the first census in 1956 until the last in 2011. During this period the census was conducted approximately every 10 years. Obviously the change in population does not directly impact temperatures in cities, but a multitude of changes to the physical environment related to increases in population certainly can strongly influence spatial and temporal thermal patterns in urban areas. For extreme temperature indices, we selected eight indicators recommended by the Expert Team on Climate Change Detection and Indices (see Peterson (2005)), based on the maximum and minimum daily temperature (Table 1). Two were determined on when the maximum temperature was unusually high based on a threshold of 25 °C or when the value was above the 90th percentile measured at given station. Two others were calculated on unusually low maximum temperatures based on a threshold 0 °C or when the value was below the 10th percentile in the distribution. Two other indices were determined by the number of days with low minimum temperatures based on a threshold of 0 °C or when the value was below the 10th percentile at given station. The other two were associated with relatively warm minimum temperatures being above 20 °C or the 90% percentile of the values at a given station.
99
Table 2 Summary of analysis results. Index
r with year
p with r
p with ln(ΔPop)
TN b 0 TX N 25 TX b 0 TN N 20 TN b 10% TX b 10% TN N 90% TX N 90%
−0.010 +0.023 −0.006 +0.038 −0.015 −0.004 +0.034 +0.018
0.074 0.000 0.283 0.000 0.042 0.602 0.000 0.000
0.111 (−) 0.004 (+) 0.049 (−) 0.586 (+) 0.093 (−) 0.011 (−) 0.426 (+) 0.185 (+)
Bold numbers statistically significant at the 0.05 level of confidence.
3. Results and discussion For each of the eight extreme temperature indices, a matrix was constructed with 50 rows, one for each year from 1961 to 2010, and 31 columns representing the stations; each cell contained an integer representing the number of times in that year the given threshold for an index was exceeded. The data for each column were converted to z-scores, and the z-scores were then averaged across the 31 station network. This procedure produced a new 50 × 8 matrix with the averaged z-scores for each of the eight temperature extreme indices. A simple regression analysis with the year of record as the independent variable generated the linear trend in each of the eight indices. As seen in Table 2 and Fig. 4, there is substantial evidence of overall warming across Iran as reflected in the various extreme temperature indices. The two strongest trends were related to an increase in the number of days with relatively high minimum temperatures (TN N 20, TN N 90%). The next largest upward trends were associated with increases in the number of days in which the maximum temperature was unusually high (TX N 25, TX N 90%). The only other trend significant at the 0.05 level of confidence was the decrease in the number of days with the minimum temperature below the 10th percentile in the frequency distribution (TN b 10%). These findings are consistent with Rahimzadeh et al. (2009) who reported an increase in “tropical nights” and “warm days” over most regions of the country and an overall decrease in “cool nights”. IPCC and many others report similar trends for many land areas of the globe very likely in response to ongoing anthropogenic changes to atmospheric composition (Stocker et al., 2013). One of the goals of this study was to determine if the rate of human population growth near the stations had influenced the trends in extreme temperature indices. We determined the change in population (ΔPop) at the 28 stations for which the data were available from 1956 to 2011; however, a Kolmogorov–Smirnov test for normality showed that this variable had a very significant (p b 0.005) deviation from the Gaussian distribution given the large and dominating increase in population (over 6.6 million people) in Tehran. We used the natural logarithm of each value to produce a new variable, ln(ΔPop), which showed no significant deviation from a normal distribution (p = 0.773). This yielded a new matrix of 28 rows, one for each station with a ln(ΔPop) value, and nine columns including the linear trend for each of the eight temperature extreme indices and the ln(ΔPop) values.
Table 1 Definitions of extreme temperature indices used in this study. No
Index
Definition
Unit
1 2 3 4 5 6 7 8
TN b 0 TX N 25 TX b 0 TN N 20 TN b 10% TX b 10% TN N 90% TX N 90%
Frost days: Annual count when TN (daily minimum) b0 °C Summer days: Annual count when TX (daily maximum) N25 °C Ice days: Annual count when TX (daily maximum) b0 °C Tropical nights: Annual count when TN (daily minimum) N20 °C Cool nights: Percentage of days when TN b10th percentile Cool days: Percentage of days when TX b10th percentile Warm nights: Percentage of days when TN N90th percentile Warm days: Percentage of days when TX N90th percentile
Days Days Days Days Days Days Days Days
100
R.C. Balling Jr. et al. / Atmospheric Research 169 (2016) 96–101
Fig. 4. Trends (r values) for extreme temperature indices.
A simple linear regression analysis linked the trends in the various temperature extreme indices to the ln(ΔPop) values. As seen in Table 2, the change in population significantly (p b 0.05) influenced three of the indices. The strongest influence was found
with the number of days with the maximum temperature above 25 °C (TX N 25); greater population growth enhanced the chances of exceeding this threshold (Fig. 5). The other two significant values were found with a decline in the number of days with relatively low maximum
0.7 0.6 0.5
TX25R
0.4 0.3 0.2 0.1 0.0 -0.1 -0.2 10
11
12
13
14
15
16
ln(dPop)
Fig. 5. Scatterplot showing the relationship between the trend (r-values) in the number of days with maximum temperature above 25 °C and the natural log of the change in population.
R.C. Balling Jr. et al. / Atmospheric Research 169 (2016) 96–101
temperature values (TX b 0, TX b 10%). While many other recent studies continue to show that population increases tend to impact the minimum temperatures more than maximum temperatures (e.g., Wang et al., 2013; Lu and Liu, 2014; Mohan and Kandya, 2015), others show the situation can be more complicated depending on season and setting (Zhou et al., 2013). We found evidence from our study of Iran that population increase impacted extreme temperatures associated with the maximum temperatures, not the minimum temperatures. 4. Conclusions We analyzed a variety of extreme temperature indices over the period 1961 to 2010 for 31 stations located throughout Iran. Trends in each index were consistent with overall warming across the country. Similar to what other authors have found in many other regions, the trends were most significant for indices associated with the daily minimum temperatures as opposed to those associated with maximum temperatures. However, our analysis linking these trends to population growth in Iran indicated that increasing population impacted trends in extreme maximum temperature indices far more that minimum temperature indices. Many other studies reviewed in this paper have documented positive trends in maximum and minimum temperatures in Iran over the past half century. These trends are similar to what other scientists have found in many other parts of the globe, and the overall trends are likely related to the changes in atmospheric composition (most notably the increase of greenhouse gases) over the same time period. Our contribution shows that although small, there remain detectable anthropogenic signals in these temperature trends and variations. References Araghi, A.M., Mousavi-Baygi, M., Adamowski, J., 2015a. Detection of trends in days with extreme temperatures in Iran from 1961 to 2010. Theor. Appl. Climatol. http://dx. doi.org/10.1007/s00704-015-1499-6. Araghi, A.M., et al., 2015b. Using wavelet transforms to estimate surface temperature trends and dominant periodicities in Iran based on gridded reanalysis data. Atmos. Res. 155, 52–72. http://dx.doi.org/10.1016/j.atmosres.2014.11.016 (15 March 2015). Ballester, J., Giorgi, F., Rodo, X., 2010. Changes in European temperature extremes can be predicted from changes in PDF central statistics. Clim. Chang. 98, 277–284. http://dx. doi.org/10.1007/s10584-009-9758-0. Boccolari, M., Malmusi, S., 2013. Changes in temperature and precipitation extremes observed in Modena, Italy. Atmos. Res. 122, 16–31. http://dx.doi.org/10.1016/j. atmosres.2012.10.022. Cinco, T.A., et al., 2014. Long-term trends and extremes in observed daily precipitation and near surface air temperature in the Philippines for the period 1951–2010. Atmos. Res. 145–146, 12–26. http://dx.doi.org/10.1016/j.atmosres.2014.03.025. Clark, P.J., Evans, F.C., 1954. Distance to nearest neighbor as a measure of spatial relationships in populations. Ecology 35, 445–453. de Lima, M.I.P., et al., 2013. Recent changes in daily precipitation and surface air temperature extremes in mainland Portugal, in the period 1941–2007. Atmos. Res. 127, 195–209. http://dx.doi.org/10.1016/j.atmosres.2012.10.001. Deng, H., et al., 2014. Dynamics of temperature and precipitation extremes and their spatial variation in the arid region of northwest China. Atmos. Res. 138, 346–355. http://dx.doi.org/10.1016/j.atmosres.2013.12.001. Donat, M.G., Alexander, L.V., 2012. The shifting probability distribution of global daytime and night-time temperatures. Geophys. Res. Lett. 39, L14707. http://dx.doi.org/10. 1029/2012GL052459. Donat, M.G., et al., 2013a. Updated analyses of temperature and precipitation extreme indices since the beginning of the twentieth century: the HadEX2 dataset. J. Geophys. Res. Atmos. 118, 2098–2118. http://dx.doi.org/10.1002/jgrd.50150. Donat, M.G., et al., 2013b. Changes in extreme temperature and precipitation in the Arab region: long-term trends and variability related to ENSO and NAO. Int. J. Climatol. http://dx.doi.org/10.1002/joc.3707.
101
Fernández-Montes, S., et al., 2012. Spring and summer extreme temperatures in Iberia during last century in relation to circulation types. Atmos. Res. 127, 154–177. http://dx.doi.org/10.1016/j.atmosres.2012.07.013. Ghahraman, B., 2006. Time trend in the mean annual temperature of Iran. Turk. J. Agric. For. 30, 439–448. Ghasemi, A.R., 2015. Changes and trends in maximum, minimum and mean temperature series in Iran. Atmos. Sci. Lett. 16, 366–372. http://dx.doi.org/10.1002/asl2.569. Hansen, J., Sato, M., Ruedy, R., 2012. Perception of climate change. Proc. Natl. Acad. Sci. U.S.A. 109, E2415–E2423. http://dx.doi.org/10.1073/pnas.1205276109. Kosaka, Y., Xie, S.-P., 2013. Recent global-warming hiatus tied to equatorial Pacific surface cooling. Nature http://dx.doi.org/10.1038/nature12534. Lu, H.L., Liu, G.F., 2014. Recent observations of human-induced asymmetric effects on climate in very high-altitude area. Plos One 9. http://dx.doi.org/10.1371/journal.pone. 0081535. Mohan, M., Kandya, A., 2015. Impact of urbanization and landuse/land-cover change on diurnal temperature range: a case study of tropical urban airshed of India using remote sensing data. Sci. Total Environ. 506, 453–465. http://dx.doi.org/10.1016/j. scitotenv.2014.11.006. Nastos, P.T., Kapsomenakis, J., 2015. Regional climate model simulations of extreme air temperature in Greece. Abnormal or common records in the future climate. Atmos. Res. 152 (15), 43–60. http://dx.doi.org/10.1016/j.atmosres.2014.02.005. Perkins, S.E., Alexander, L.V., Nairn, J.R., 2012. Increasing frequency, intensity and duration of observed global heatwaves and warm spells. Geophys. Res. Lett. 39, L20714. http://dx.doi.org/10.1029/2012GL053361. Peterson, T.C., 2005. Climate change indices. WMO Bull. 54, 83–86. Pingale, S.M., Khare, D., Jat, M.K., Adamowski, J., 2014. Spatial and temporal trends of mean and extreme rainfall and temperature for the 33 urban centers of the arid and semi-arid state of Rajasthan, India. Atmos. Res. 138, 73–90. http://dx.doi.org/ 10.1016/j.atmosres.2013.10.024. Rahimzadeh, F., Asgari, A., Fattahi, E., 2009. Variability of extreme temperature and precipitation in Iran during recent decades. Int. J. Climatol. 29, 329–343. http://dx.doi. org/10.1002/joc.1739. Rusticucci, M., 2012. Observed and simulated variability of extreme temperature events over South America. Atmos. Res. 106, 1–17. http://dx.doi.org/10.1016/j.atmosres. 2011.11.001. Seneviratne, S.I., et al., 2012. Changes in climate extremes and their impacts on the natural physical environment. In: Managing the Risks of Extreme Events, Disasters to Advance Climate Change Adaptation, Field, C.B., Barros, V., Stocker, T.F., Qin, D., Dokken, D.J., Ebi, K.L., Mastrandrea, M.D., Mach, K.J., Plattner, G.-K., Allen, S.K., Tignor, M., Midgley, P.M. (Eds.), A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change (IPCC). Cambridge University Press, Cambridge, UK, and New York, NY, USA, pp. 109–230. Simolo, C., Brunetti, M., Maugeri, M., Nanni, T., 2011. Evolution of extreme temperatures in a warming climate. Geophys. Res. Lett. 38, 6. http://dx.doi.org/10.1029/ 2011GL048437. Stocker, T.F., et al., 2013. Technical Summary. In: Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M. (Eds.), Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Tabari, H., Talaee, P., 2011. Recent trends of mean maximum and minimum air temperatures in the western half of Iran. Meteorog. Atmos. Phys. 111, 121–131. http://dx.doi. org/10.1007/s00703-011-0125-0. Tabari, H., Shifteh Somee, B., Rezaeian Zadeh, M., 2011. Testing for long-term trends in climatic variables in Iran. Atmos. Res. 100, 132–140. http://dx.doi.org/10.1016/j. atmosres.2011.01.005. Tabari, H., Talaee, P., Ezani, A., Shifteh Some'e, B., 2012. Shift changes and monotonic trends in autocorrelated temperature series over Iran. Theor. Appl. Climatol. 109, 95–108. http://dx.doi.org/10.1007/s00704-011-0568-8. Wang, J., et al., 2013. Impact of urbanization on changes in temperature extremes in Beijing during 1978–2008. Chin. Sci. Bull. 58, 4679–4686. http://dx.doi.org/10.1007/ s11434-013-5976-y. WMO, 2003. Guidelines on Climate Metadata and Homogenization. WMO/TD-No. 1186. World Meteorological Organization, Geneva. Zarenistanak, M., Dhorde, A., Kripalani, R.H., 2014. Trend analysis and change point detection of annual and seasonal precipitation and temperature series over southwest Iran. J. Earth Syst. Sci. 123, 281–295. Zhang, X.B., et al., 2005. Trends in Middle East climate extreme indices from 1950 to 2003. J. Geophys. Res. Atmos. 110, D22104. http://dx.doi.org/10.1029/2005JD006181. Zhou, J., et al., 2013. Modelling the diurnal variations of urban heat islands with multisource satellite data. Int. J. Remote Sens. 34, 7568–7588. http://dx.doi.org/10.1080/ 01431161.2013.821576.
Contents lists available at ScienceDirect
Atmospheric Research journal homepage: www.elsevier.com/locate/atmos
Anthropogenic signals in Iranian extreme temperature indices Robert C. Balling Jr. a,⁎, Mohammad Sadegh Keikhosravi Kiany b, Shouraseni Sen Roy c a b c
School of Geographical Sciences and Urban Planning, Arizona State University, Tempe, AZ 85287, USA Faculty of Geographical Sciences and Planning, University of Esfahan, Esfahan, Iran Department of Geography and Regional Studies, University of Miami, Coral Gables, FL, USA
a r t i c l e
i n f o
Article history: Received 10 April 2015 Received in revised form 21 September 2015 Accepted 27 September 2015 Available online 13 October 2015 Keywords: Temperature extremes Iran Population Weekly cycle
a b s t r a c t We analyzed spatial and temporal patterns in temperature extremes from 31 stations located throughout Iran for the period 1961 to 2010. As with many other parts of the globe, we found that the number of days (a) with high maximum temperatures was rising, (b) with high minimum temperatures was rising, and (c) with low minimum temperatures was declining; all of these trends were statistically significant at the 0.05 level of confidence. Population records from 1956 to 2011 at the station locations allowed us to reveal that the rate of human population growth was positively related to the increase in the number of days with high maximum temperatures and negatively related to days with low maximum temperatures. Our research shows a number of identifiable anthropogenic signals in the temperature records from Iran, but unlike most other studies, the signals are stronger with indices related to maximum, not minimum, temperatures. © 2015 Elsevier B.V. All rights reserved.
1. Introduction There is general consensus in the climate science community regarding the general increase in global mean surface temperatures since the later part of the 19th century. Specifically, the observed temperatures recorded over the last three decades show a successive increase, with the decade of the 2000s as the warmest on record (Stocker et al., 2013). The analysis on long term surface level observations at the global scale indicate a warming of 0.85 [0.65 to 1.06]°C during 1880–2012, ≈0.89 [0.69 to 1.08]°C during 1901–2012, and ≈0.72 [0.49 to 0.89] °C during 1951–2012 based on three independentlyproduced data sets (Stocker et al., 2013). However, there are substantial temporal variations in these trends that are caused by local level teleconnections such as El Niño Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO), as well as the recently observed global warming hiatus since 1998 (Kosaka and Xie, 2013). Other than temporal variations, there are substantial regional variations in the trends in temperatures driven by land use land cover changes including the role of urban heat island (UHI), deforestation, agriculture-related irrigation effects, as well as other anthropogenic activities (Stocker et al., 2013). One of the important aspects of long term temperature trends is the occurrence of extreme temperatures. The results from a recently published report on extreme events by Seneviratne et al. (2012) indicated an overall decrease in cold extremes and increase in warm extremes since the middle of the 20th century. For instance, Donat et al. (2013a) found a significant warming of both maximum (TX) and minimum (TN) temperatures over global land areas since 1950.
⁎ Corresponding author.
http://dx.doi.org/10.1016/j.atmosres.2015.09.030 0169-8095/© 2015 Elsevier B.V. All rights reserved.
Additionally, a greater shift in the distribution of nighttime temperatures compared to the daytime temperatures was observed in several studies (Ballester et al., 2010; Simolo et al., 2011; Donat and Alexander, 2012; Hansen et al., 2012). However, there are substantial variations in the overall trends and the confidence levels associated with those trends. For instance, in the Middle East, there is medium confidence in the increased occurrence of daytime TX above the 90th percentile and nighttime TN above the 90th percentile (Donat et al., 2013b; Zhang et al., 2005). There is also medium confidence in the increasing trends observed in the incidences of heat waves and warm spells in the Middle East (Perkins et al., 2012; Donat et al., 2013a). Many papers have appeared in Atmospheric Research in the past few years on the topic of trends in extreme temperatures. Several have reported overall warming in Portugal (de Lima et al., 2013), Rajasthan, India (Pingale et al., 2014), and Iran (Araghi et al., 2015a,2015b). However, many others have revealed evidence that the warming of TN have exceeded the warming of TX as reflected in a variety of extreme temperature indices. These findings have come from analyses in Greece (Nastos and Kapsomenakis, 2015), the Philippines (Cinco et al., 2014), northwest China (Deng et al., 2014), Iberia (Fernández-Montes et al., 2012), South America (Rusticucci, 2012), and Modena, Italy (Boccolari and Malmusi, 2013). In this investigation, we focus on trends in extreme temperatures throughout Iran for the period 1961 to 2010. Over the past decade, a number of important studies have documented the overall warming of the country during the last half century. Ghahraman (2006) investigated the long term trend of mean annual temperature at 34 synoptic stations for the period from 1968 to 1998 and found that at the 0.05 level of significance 44%, 15%, and 41% of the stations had a positive, negative, and zero trend respectively. Rahimzadeh et al. (2009)
R.C. Balling Jr. et al. / Atmospheric Research 169 (2016) 96–101
Fig 1. General location and topography of Iran.
Fig. 2. Location of 31 stations in Iran.
97
98
R.C. Balling Jr. et al. / Atmospheric Research 169 (2016) 96–101
examined temperature records from 27 stations in Iran and found positive trends in the occurrence of hot days and tropical nights. Tabari et al. (2011) examined trend in annual air temperatures in the west, south, and southwest of Iran for the period 1966–2005. They concluded that an overall warming trend exists and with the trend for TN (0.49 °C dec− 1) slightly exceeding the trend for TX (0.45 °C dec−1). Tabari and Talaee (2011) investigated trends in air temperature time series for 20 stations in the western half of Iran during 1966–2005. The annual TX and TN values showed a positive trend in 85% of the stations and a negative trend in 15% of the stations in the study region. Tabari et al. (2012) analyzed the data of 29 stations in Iran for a period of 40 years from 1966 to 2005 for the existence of monotonic trends and shift changes in the annual, seasonal and monthly mean temperature. It was found that the annual mean air temperature increase at 25 out of the 29 stations showed significant monotonic trends. The spatial analysis of the mean air temperature trends revealed the highest numbers of significant monotonic trends in the larger cities of Iran. Zarenistanak et al. (2014) analyzed trend and change point detection of annual and seasonal precipitation and temperature time series over the period 1950–2007. Their results revealed a significant increase during the summer and spring seasons. TX was more stable than TN, and winter was stable compared to the summer, spring, and autumn seasons. The results of the change point analysis indicated that most of the statistically significant positive mutation points in all temperatures began in the 1990s. Araghi et al. (2015a, b) investigated trends in Iran in the days with TX ≥ 30 °C and TN ≤0 °C from 1961 to 2010 for 30 synoptic stations. They concluded that for 67% of stations, days wtih TN ≤0 °C showed a significant negative trend, while only 40% of stations showed a significant positive trend in days with TX ≥ 30 °C. Ghasemi (2015) studied the variations of Iranian temperature during 1961–2010 using 38
stations across the country. Mann–Kendall tests indicated an overall dominating positive temperature trend over most parts of Iran with a TN increase (0.34 °C dec−1) more than twice the increase for TX (0.15 °C dec−1). The findings of this study also showed that warm climate regions in Iran are warming at a higher rate than cold climate areas. The evidence is overwhelming that near-surface air temperatures in Iran have increased significantly over the past half century, and as with many other parts of the globe, the increase in minimum temperatures has exceeded the increase in maximum temperatures. To significantly add to the literature on temperature trends in Iran, we evaluate the potential role of local population growth on these trends to possibly enlighten a signal in the data that is likely related to anthropogenic activities. 2. Materials and methods Iran is located between 25° and 40°N and 45° and 60°E and is a mountainous country bordering the Gulf of Oman, the Persian Gulf, and the Caspian Sea (Fig. 1). The total area of Iran is 1.648 × 106 km2 which represents 0.32% of the Earth's surface. Overall, sixty percent of Iran is covered by mountains, with the central part of the country consisting of two dry deserts: the Dasht-e-Kavir and the Dasht-e-Lut. The Alborz range in the north, close to the Caspian Sea, extends in an east-west direction with a maximum elevation of approximately 5000 m. The Zagros Mountains are aligned in a northwest to southeast direction and reach a maximum elevation of approximately 3500 m. The country can be divided in at least four different climate zones. The western and southwestern area are hot and dry with precipitation increasing toward the Persian Gulf and the Turkish border. The eastern
Fig. 3. Population changes in Iran from 1956 to 2011 (note that different color scales are used for smaller and larger urban areas).
R.C. Balling Jr. et al. / Atmospheric Research 169 (2016) 96–101
and northern areas of Iran have a Mediterranean climate with dry summers and mild and moderately wet winters. The mountainous regions in the north are Mediterranean-like in terms of precipitation distribution, but the elevation and latitude contribute to substantial snowfall in the winter season. We obtained daily temperature data from the Islamic Republic of Iran Meteorological Organization (IRIMO). IRIMO publishes the quality controlled data with nearly three years of delay. The original data passes WMO procedures before publishing (see WMO (2003) for details). The data are checked for miswritten codes and procedures are used to detect erroneous observations and check for doubtful records. The final quality controlled data are then published in txt files by IRIMO. There are more than 150 synoptic weather stations in Iran but we had to limit our analyses to only 31 stations because of the shortness of the period in many of the stations and gaps in many of the records. Our selected weather stations covered all the climatic regions of the country (Fig. 2) and range from small towns to the largest cities in Iran (Fig. 3). All the weather stations have data for the period from 1961 to 2010. Among the stations used, nearly all of them have more than 99% of available days except for the Saghez, Oroomieh, and Kermanshah stations that have 98.02%, 98.60% and 98.97% of available days, respectively. We calculated the nearest-neighbor statistic, r, for the 31-station network which represents the ratio of the average distance from each station to its nearest neighbor to the average distance between nearest neighbors calculated for a random spatial distribution (Clark and Evans, 1954). The values of r can range from 0 for a completely clustered distribution (all stations have exactly the same latitude and longitude) to 1.0 for a random pattern to 2.15 for a perfectly uniform distribution. The value of r for our 31-station network is 1.20 falling in the desirable random to uniform range. In order to investigate the role of population increase in cities on extreme temperature changes, population data were obtained from the official reports of Statistical Centre of Iran. We collected population data from the first census in 1956 until the last in 2011. During this period the census was conducted approximately every 10 years. Obviously the change in population does not directly impact temperatures in cities, but a multitude of changes to the physical environment related to increases in population certainly can strongly influence spatial and temporal thermal patterns in urban areas. For extreme temperature indices, we selected eight indicators recommended by the Expert Team on Climate Change Detection and Indices (see Peterson (2005)), based on the maximum and minimum daily temperature (Table 1). Two were determined on when the maximum temperature was unusually high based on a threshold of 25 °C or when the value was above the 90th percentile measured at given station. Two others were calculated on unusually low maximum temperatures based on a threshold 0 °C or when the value was below the 10th percentile in the distribution. Two other indices were determined by the number of days with low minimum temperatures based on a threshold of 0 °C or when the value was below the 10th percentile at given station. The other two were associated with relatively warm minimum temperatures being above 20 °C or the 90% percentile of the values at a given station.
99
Table 2 Summary of analysis results. Index
r with year
p with r
p with ln(ΔPop)
TN b 0 TX N 25 TX b 0 TN N 20 TN b 10% TX b 10% TN N 90% TX N 90%
−0.010 +0.023 −0.006 +0.038 −0.015 −0.004 +0.034 +0.018
0.074 0.000 0.283 0.000 0.042 0.602 0.000 0.000
0.111 (−) 0.004 (+) 0.049 (−) 0.586 (+) 0.093 (−) 0.011 (−) 0.426 (+) 0.185 (+)
Bold numbers statistically significant at the 0.05 level of confidence.
3. Results and discussion For each of the eight extreme temperature indices, a matrix was constructed with 50 rows, one for each year from 1961 to 2010, and 31 columns representing the stations; each cell contained an integer representing the number of times in that year the given threshold for an index was exceeded. The data for each column were converted to z-scores, and the z-scores were then averaged across the 31 station network. This procedure produced a new 50 × 8 matrix with the averaged z-scores for each of the eight temperature extreme indices. A simple regression analysis with the year of record as the independent variable generated the linear trend in each of the eight indices. As seen in Table 2 and Fig. 4, there is substantial evidence of overall warming across Iran as reflected in the various extreme temperature indices. The two strongest trends were related to an increase in the number of days with relatively high minimum temperatures (TN N 20, TN N 90%). The next largest upward trends were associated with increases in the number of days in which the maximum temperature was unusually high (TX N 25, TX N 90%). The only other trend significant at the 0.05 level of confidence was the decrease in the number of days with the minimum temperature below the 10th percentile in the frequency distribution (TN b 10%). These findings are consistent with Rahimzadeh et al. (2009) who reported an increase in “tropical nights” and “warm days” over most regions of the country and an overall decrease in “cool nights”. IPCC and many others report similar trends for many land areas of the globe very likely in response to ongoing anthropogenic changes to atmospheric composition (Stocker et al., 2013). One of the goals of this study was to determine if the rate of human population growth near the stations had influenced the trends in extreme temperature indices. We determined the change in population (ΔPop) at the 28 stations for which the data were available from 1956 to 2011; however, a Kolmogorov–Smirnov test for normality showed that this variable had a very significant (p b 0.005) deviation from the Gaussian distribution given the large and dominating increase in population (over 6.6 million people) in Tehran. We used the natural logarithm of each value to produce a new variable, ln(ΔPop), which showed no significant deviation from a normal distribution (p = 0.773). This yielded a new matrix of 28 rows, one for each station with a ln(ΔPop) value, and nine columns including the linear trend for each of the eight temperature extreme indices and the ln(ΔPop) values.
Table 1 Definitions of extreme temperature indices used in this study. No
Index
Definition
Unit
1 2 3 4 5 6 7 8
TN b 0 TX N 25 TX b 0 TN N 20 TN b 10% TX b 10% TN N 90% TX N 90%
Frost days: Annual count when TN (daily minimum) b0 °C Summer days: Annual count when TX (daily maximum) N25 °C Ice days: Annual count when TX (daily maximum) b0 °C Tropical nights: Annual count when TN (daily minimum) N20 °C Cool nights: Percentage of days when TN b10th percentile Cool days: Percentage of days when TX b10th percentile Warm nights: Percentage of days when TN N90th percentile Warm days: Percentage of days when TX N90th percentile
Days Days Days Days Days Days Days Days
100
R.C. Balling Jr. et al. / Atmospheric Research 169 (2016) 96–101
Fig. 4. Trends (r values) for extreme temperature indices.
A simple linear regression analysis linked the trends in the various temperature extreme indices to the ln(ΔPop) values. As seen in Table 2, the change in population significantly (p b 0.05) influenced three of the indices. The strongest influence was found
with the number of days with the maximum temperature above 25 °C (TX N 25); greater population growth enhanced the chances of exceeding this threshold (Fig. 5). The other two significant values were found with a decline in the number of days with relatively low maximum
0.7 0.6 0.5
TX25R
0.4 0.3 0.2 0.1 0.0 -0.1 -0.2 10
11
12
13
14
15
16
ln(dPop)
Fig. 5. Scatterplot showing the relationship between the trend (r-values) in the number of days with maximum temperature above 25 °C and the natural log of the change in population.
R.C. Balling Jr. et al. / Atmospheric Research 169 (2016) 96–101
temperature values (TX b 0, TX b 10%). While many other recent studies continue to show that population increases tend to impact the minimum temperatures more than maximum temperatures (e.g., Wang et al., 2013; Lu and Liu, 2014; Mohan and Kandya, 2015), others show the situation can be more complicated depending on season and setting (Zhou et al., 2013). We found evidence from our study of Iran that population increase impacted extreme temperatures associated with the maximum temperatures, not the minimum temperatures. 4. Conclusions We analyzed a variety of extreme temperature indices over the period 1961 to 2010 for 31 stations located throughout Iran. Trends in each index were consistent with overall warming across the country. Similar to what other authors have found in many other regions, the trends were most significant for indices associated with the daily minimum temperatures as opposed to those associated with maximum temperatures. However, our analysis linking these trends to population growth in Iran indicated that increasing population impacted trends in extreme maximum temperature indices far more that minimum temperature indices. Many other studies reviewed in this paper have documented positive trends in maximum and minimum temperatures in Iran over the past half century. These trends are similar to what other scientists have found in many other parts of the globe, and the overall trends are likely related to the changes in atmospheric composition (most notably the increase of greenhouse gases) over the same time period. Our contribution shows that although small, there remain detectable anthropogenic signals in these temperature trends and variations. References Araghi, A.M., Mousavi-Baygi, M., Adamowski, J., 2015a. Detection of trends in days with extreme temperatures in Iran from 1961 to 2010. Theor. Appl. Climatol. http://dx. doi.org/10.1007/s00704-015-1499-6. Araghi, A.M., et al., 2015b. Using wavelet transforms to estimate surface temperature trends and dominant periodicities in Iran based on gridded reanalysis data. Atmos. Res. 155, 52–72. http://dx.doi.org/10.1016/j.atmosres.2014.11.016 (15 March 2015). Ballester, J., Giorgi, F., Rodo, X., 2010. Changes in European temperature extremes can be predicted from changes in PDF central statistics. Clim. Chang. 98, 277–284. http://dx. doi.org/10.1007/s10584-009-9758-0. Boccolari, M., Malmusi, S., 2013. Changes in temperature and precipitation extremes observed in Modena, Italy. Atmos. Res. 122, 16–31. http://dx.doi.org/10.1016/j. atmosres.2012.10.022. Cinco, T.A., et al., 2014. Long-term trends and extremes in observed daily precipitation and near surface air temperature in the Philippines for the period 1951–2010. Atmos. Res. 145–146, 12–26. http://dx.doi.org/10.1016/j.atmosres.2014.03.025. Clark, P.J., Evans, F.C., 1954. Distance to nearest neighbor as a measure of spatial relationships in populations. Ecology 35, 445–453. de Lima, M.I.P., et al., 2013. Recent changes in daily precipitation and surface air temperature extremes in mainland Portugal, in the period 1941–2007. Atmos. Res. 127, 195–209. http://dx.doi.org/10.1016/j.atmosres.2012.10.001. Deng, H., et al., 2014. Dynamics of temperature and precipitation extremes and their spatial variation in the arid region of northwest China. Atmos. Res. 138, 346–355. http://dx.doi.org/10.1016/j.atmosres.2013.12.001. Donat, M.G., Alexander, L.V., 2012. The shifting probability distribution of global daytime and night-time temperatures. Geophys. Res. Lett. 39, L14707. http://dx.doi.org/10. 1029/2012GL052459. Donat, M.G., et al., 2013a. Updated analyses of temperature and precipitation extreme indices since the beginning of the twentieth century: the HadEX2 dataset. J. Geophys. Res. Atmos. 118, 2098–2118. http://dx.doi.org/10.1002/jgrd.50150. Donat, M.G., et al., 2013b. Changes in extreme temperature and precipitation in the Arab region: long-term trends and variability related to ENSO and NAO. Int. J. Climatol. http://dx.doi.org/10.1002/joc.3707.
101
Fernández-Montes, S., et al., 2012. Spring and summer extreme temperatures in Iberia during last century in relation to circulation types. Atmos. Res. 127, 154–177. http://dx.doi.org/10.1016/j.atmosres.2012.07.013. Ghahraman, B., 2006. Time trend in the mean annual temperature of Iran. Turk. J. Agric. For. 30, 439–448. Ghasemi, A.R., 2015. Changes and trends in maximum, minimum and mean temperature series in Iran. Atmos. Sci. Lett. 16, 366–372. http://dx.doi.org/10.1002/asl2.569. Hansen, J., Sato, M., Ruedy, R., 2012. Perception of climate change. Proc. Natl. Acad. Sci. U.S.A. 109, E2415–E2423. http://dx.doi.org/10.1073/pnas.1205276109. Kosaka, Y., Xie, S.-P., 2013. Recent global-warming hiatus tied to equatorial Pacific surface cooling. Nature http://dx.doi.org/10.1038/nature12534. Lu, H.L., Liu, G.F., 2014. Recent observations of human-induced asymmetric effects on climate in very high-altitude area. Plos One 9. http://dx.doi.org/10.1371/journal.pone. 0081535. Mohan, M., Kandya, A., 2015. Impact of urbanization and landuse/land-cover change on diurnal temperature range: a case study of tropical urban airshed of India using remote sensing data. Sci. Total Environ. 506, 453–465. http://dx.doi.org/10.1016/j. scitotenv.2014.11.006. Nastos, P.T., Kapsomenakis, J., 2015. Regional climate model simulations of extreme air temperature in Greece. Abnormal or common records in the future climate. Atmos. Res. 152 (15), 43–60. http://dx.doi.org/10.1016/j.atmosres.2014.02.005. Perkins, S.E., Alexander, L.V., Nairn, J.R., 2012. Increasing frequency, intensity and duration of observed global heatwaves and warm spells. Geophys. Res. Lett. 39, L20714. http://dx.doi.org/10.1029/2012GL053361. Peterson, T.C., 2005. Climate change indices. WMO Bull. 54, 83–86. Pingale, S.M., Khare, D., Jat, M.K., Adamowski, J., 2014. Spatial and temporal trends of mean and extreme rainfall and temperature for the 33 urban centers of the arid and semi-arid state of Rajasthan, India. Atmos. Res. 138, 73–90. http://dx.doi.org/ 10.1016/j.atmosres.2013.10.024. Rahimzadeh, F., Asgari, A., Fattahi, E., 2009. Variability of extreme temperature and precipitation in Iran during recent decades. Int. J. Climatol. 29, 329–343. http://dx.doi. org/10.1002/joc.1739. Rusticucci, M., 2012. Observed and simulated variability of extreme temperature events over South America. Atmos. Res. 106, 1–17. http://dx.doi.org/10.1016/j.atmosres. 2011.11.001. Seneviratne, S.I., et al., 2012. Changes in climate extremes and their impacts on the natural physical environment. In: Managing the Risks of Extreme Events, Disasters to Advance Climate Change Adaptation, Field, C.B., Barros, V., Stocker, T.F., Qin, D., Dokken, D.J., Ebi, K.L., Mastrandrea, M.D., Mach, K.J., Plattner, G.-K., Allen, S.K., Tignor, M., Midgley, P.M. (Eds.), A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change (IPCC). Cambridge University Press, Cambridge, UK, and New York, NY, USA, pp. 109–230. Simolo, C., Brunetti, M., Maugeri, M., Nanni, T., 2011. Evolution of extreme temperatures in a warming climate. Geophys. Res. Lett. 38, 6. http://dx.doi.org/10.1029/ 2011GL048437. Stocker, T.F., et al., 2013. Technical Summary. In: Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M. (Eds.), Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Tabari, H., Talaee, P., 2011. Recent trends of mean maximum and minimum air temperatures in the western half of Iran. Meteorog. Atmos. Phys. 111, 121–131. http://dx.doi. org/10.1007/s00703-011-0125-0. Tabari, H., Shifteh Somee, B., Rezaeian Zadeh, M., 2011. Testing for long-term trends in climatic variables in Iran. Atmos. Res. 100, 132–140. http://dx.doi.org/10.1016/j. atmosres.2011.01.005. Tabari, H., Talaee, P., Ezani, A., Shifteh Some'e, B., 2012. Shift changes and monotonic trends in autocorrelated temperature series over Iran. Theor. Appl. Climatol. 109, 95–108. http://dx.doi.org/10.1007/s00704-011-0568-8. Wang, J., et al., 2013. Impact of urbanization on changes in temperature extremes in Beijing during 1978–2008. Chin. Sci. Bull. 58, 4679–4686. http://dx.doi.org/10.1007/ s11434-013-5976-y. WMO, 2003. Guidelines on Climate Metadata and Homogenization. WMO/TD-No. 1186. World Meteorological Organization, Geneva. Zarenistanak, M., Dhorde, A., Kripalani, R.H., 2014. Trend analysis and change point detection of annual and seasonal precipitation and temperature series over southwest Iran. J. Earth Syst. Sci. 123, 281–295. Zhang, X.B., et al., 2005. Trends in Middle East climate extreme indices from 1950 to 2003. J. Geophys. Res. Atmos. 110, D22104. http://dx.doi.org/10.1029/2005JD006181. Zhou, J., et al., 2013. Modelling the diurnal variations of urban heat islands with multisource satellite data. Int. J. Remote Sens. 34, 7568–7588. http://dx.doi.org/10.1080/ 01431161.2013.821576.