Atmospheric Feedback of Urban Boundary Layer with Implications for

Article pubs.acs.org/est

Atmospheric Feedback of Urban Boundary Layer with Implications for Climate Adaptation Marissa S. Liang and Timothy C. Keener* Department of Biomedical, Chemical, and Environmental Engineering, University of Cincinnati, Cincinnati, Ohio 45221, United States S Supporting Information *

ABSTRACT: Atmospheric structure changes in response to the urban form, land use, and the type of land cover (LULC). This interaction controls thermal and air pollutant transport and distribution. The interrelationships among LULC, ambient temperature, and air quality were analyzed and found to be significant in a case study in Cincinnati, Ohio, U.S.A. Within the urban canopy layer (UCL), traffic-origin PM2.5 and black carbon followed Gaussian dispersion in the near road area in the daytime, while higher concentrations, over 1 order of magnitude, were correlated to the lapse rate under nocturnal inversions. In the overlying urban boundary layer (UBL), ambient temperature and PM2.5 variations were correlated among urban-wide locations indicating effective thermal and mass communications. Beyond the spatial correlation, LULC-related local urban heat island effects are noteworthy. The high-density urbanized zone along a narrow highway-following corridor is marked by higher nighttime temperature by ∼1.6 °C with a long-term increase by 2.0 °C/decade, and by a higher PM2.5 concentration, than in the low-density residential LULC. These results indicate that the urban LULC may have contributed to the nocturnal thermal inversion affecting urban air circulation and air quality in UCL and UBL. Such relationships point to the potentials of climate adaptation through urban planning.

■

INTRODUCTION Urban settlement is associated with unique atmospheric structure compared to surrounding rural areas. Paved land surface, latent heat sources, urban activities such as transportation, and radiative forcing of particulate matter (PM) aerosols, all reinforce the role of land use and land cover (LULC) in affecting atmospheric circulation of the urban boundary layer (UBL) and hence pollutant transports.1−5 Due to their distinct thermal properties, urban construction materials (e.g., impervious asphalt pavements, roofs, vegetated greenspace) absorb daytime insolation and release sensible heat by short-wave radiation during the night in the urban canopy layer (UCL), causing higher surface temperature in urban centers.6−8 This thermal effect of LULC may lead to urban heat island (UHI) formation, stagnant air circulation under thermal inversion, and deteriorated air qualities9−12 and perhaps initiate changes to mesoscale atmospheric structure,6,11,13−15. Figure 1 schematically shows the interrelationships for the generalized urban LULC types by Liang et al.16 for Cincinnati, Ohio, U.S.A., and the attributing atmospheric processes after Fernando.3 The extent to which UHI occurrence and atmospheric circulation respond to urban LULC types has remained an interest of research since the early work of Oke and others.13,17−20 Because of thermal and mass communication between the UCL and the overlying UBL,18 LULC feedback to the UCL process may affect the nature of and the caustic association with UHI, atmospheric thermal inversion, and pollutant transport relevant © 2015 American Chemical Society

to urban inhabitants. Recent studies using remote sensing and mesoscale modeling indicated close feedback of land use patterns with implications on controlling mechanisms.15,21,22 One leading UHI formation hypothesis calls for vertical thermal transport of latent heat in the urban core, inducing cooler and heavier near-ground air flow from exurban and thus the development of nocturnal thermal inversion in UBL.9,23−26 In principle, thermal inversion and UHI effect reduce atmospheric circulation thus affecting microclimatic conditions at the urban community scale. Furthermore, urban form and the built environment may enhance or reduce the thermal inversion and thus the UHI occurrence. This looped feedback, as indicated in Figure 1, may regulate the transport of traffic-generated PMs and black carbon aerosols relevant to urban inhabitants. It is hypothesized that these three interacted mechanisms control the interrelationships among LULC, UHI, urban atmospheric structure, and air quality. Previously published studies produced time-snapshots of and provided evidence of the interrelationships from different angles by using various methods.3,10,14,23,27−31 However, questions remain on 1) whether these mechanisms are persistent in time and space and 2) how the atmospheric feedback depends on diverse urban LULC types. Received: Revised: Accepted: Published: 10598

May 18, 2015 July 27, 2015 August 3, 2015 August 3, 2015 DOI: 10.1021/acs.est.5b02444 Environ. Sci. Technol. 2015, 49, 10598−10606

Article

Environmental Science & Technology

Figure 1. Schematic diagram of urban atmospheric structure in relation to LULC types, air pollutant transport, and temperature variability in the case study of Liang et al.16 G − Gaussian; N-G − non-Gaussian; NAAQS − National Ambient Air Quality Station. Solid and open arrows indicate thermal and pollutant flux.

perimeter, and to agricultural lands of native soil in the exurban area. In addition, one regional background station is located in a corn field 83.4 km NNW of Cincinnati. PM2.5 concentration and daily temperature, including the maximum (Tmax), average (Tavg), minimum (Tmin), and diurnal temperature range (δT), were analyzed and correlated to LULC types. The temperature or concentration at the reference station 39-061-0040 (Cref) inside of the high-density urbanized zone (Figure 2) are related to that of other stations (Ci) by a linear regression:

In this paper, these two questions are examined using a case study in the Cincinnati metropolitan area. It describes the correlation among 1) LULC types with UHI, thermal inversion, and PM2.5 transport in UBL and 2) urban climate and atmospheric structure with PM2.5 transport in UCL. Further extended from the results are discussions of their implication on the design of urban adaptation to the urban climate changes.

■

INVESTIGATION METHODS Urban Form Delineation. The Cincinnati metropolitan area of ∼1.8 million people covers 7350 km2 of land in ten counties on the banks of the Ohio River. Urban growth is monocentric sprawling outward from its downtown area. A major feature is the narrow north−southward high-density urbanized zone of heavy surface pavement and roof cover (Figure 2). It is delineated using a U.S. Geological Survey urban land use map and by the interpretation of a 2013 Google satellite map. The determined high-density zone extends along Interstate Highway 75 (I-75) alongside south-north trending Mill Creek. Its eastern side is bounded by Interstate Highway 71 (I-71). At the eastern and western side of the I-75 highway, two small areas of high-density development are isolated by low-density residential LULC and forest reserves, respectively, in the Western Hill and in the Blue Ash-Mason region (Figure 2). Rolling hills with small topographic relief are typical of the study area. Topographic valleys follow the east−west trending Ohio River and the north−south tributary Mill Creek. They are surrounded by flat to gently hilly farm lands and forest to the east, northeast, and south. Elevations are about 100 m lower in the valleys. Air Quality and Temperature Monitoring and LongTerm Variations. Ambient temperature and PM2.5 concentration were measured from 1999 to 2013 at 15 National Ambient Air Quality Standards (NAAQS) monitoring stations (http://www.epa.gov/airdata/). Measurement height is 162.6− 266.6m above sea level, slightly above the UCL. All 15 stations in Cincinnati are located at roadside or in residential neighborhoods (Figure 2). LULC types range from high-density pavement of commercial and industrial districts in the urban core, to low-density housing and mixed land use in the suburban

Ci = αiCref + ϵi

(1)

For ambient temperature, obtained slope (αi) and intercept (ϵi) are statistically significant with R2 ∼ 0.99. PM2.5 measurements are also strongly correlated; R2 ∼ 0.97−0.99. Atmospheric Temperature Profiling. Thermal inversion plays an important role in controlling air pollution transport in UCL, causing high concentrations of air pollutants in the UCL of morning hours.16 This atmospheric condition is shown by a reversed thermal gradient in the atmospheric temperature profile. Based on the method of Ma et al.,32 atmospheric radiosonde data collected and provided by NOAA/NESDIS were used to reconstruct the atmospheric temperature profiles. Four locations S1 to S4 at a 10-km spatial grid were analyzed and further related to urban formation for comparison (Figure 2). At each location, the hourly temperature data were retrieved for a 10-day period of October 6−15, 2010, which covers the field dispersion experiments of Liang et al.16 Pressures (P) in mbar at each location are converted to relative elevation (Z) for calculation of the temperature lapse rate (Lh) Zi = 67.38

Lh = −

⎡ P ⎤ Ti − 1 + Ti log⎢ i ⎥ 2 ⎣ Pi − 1 ⎦

(2)

Ti − Tref Zi

(3)

where T and Z are, respectively, temperature and elevation above the ground. Lapse rate remains nearly a constant in a short elevation range but changes its sign at the thermal inversion point, for which inversion height (Zinv) and temperature (Tinv) 10599

DOI: 10.1021/acs.est.5b02444 Environ. Sci. Technol. 2015, 49, 10598−10606

Article


Figure 2. Major traffic routes and the urban form in Cincinnati, Ohio. Also shown are S-1 to S-4, the four radiosonde locations (triangles) and EPA NAAQS air quality monitoring stations (circles).

were determined. Below Zinv, the thermal gradient is reversed with colder air at near-ground surface resulting in a lack of thermal buoyance and air circulation.3,33 Near-Ground Black Carbon and PM2.5 Experiments. During the same period, PM2.5 and black carbon transport within the UCL were investigated at the I-75 north site (Figure 2). Air samples collected at ∼1.7 m above ground within the UCL were analyzed for PM2.5, black carbon concentration, and the composition of elemental carbon (EC) and organic carbon (OC). Details of the field study and chemical analysis are in Liang et al.16

technique often used for trend analysis of noisy measurement data sets (e.g., ref 34). At a noise threshold of db = 0.80, the wavelet-denoised daily Tmax, Tavg, Tmin, and ΔT (see the Supporting Information) have the maximum values in AprilAugust and the minima in winter each year. For annual maxima and minima of denoised daily Tmin, linear regression shows a decade-long increase of ∼1.6 and ∼2.1 °C per 10 years, respectively (Figure 3). Tmax yields smaller changes with time.

■

RESULTS Long-Term and Seasonal Variations in Urban Core. Temperature and PM2.5 concentrations at reference station 39-061-0040 in the high-density zone display strong daily and seasonal variations during the 10.9-year period. Statistics of all daily maximum (Tmax), daily average (Tavg), daily minimum temperature (Tmin), and diurnal temperature range (ΔT) are listed in Table S1 in the Supporting Information. All temperatures are symmetrically distributed around their annual means, whereas the PM2.5 frequency distribution is asymmetric with a bias toward the decreasing PM concentrations in recent years (kurtosis = 2.21; skewness > 1). For the temperate climate, daily Tavg has a grand mean 14.22 °C (9.20 to 20.29 °C) with the lowest daily Tmin in winter and the highest Tmax in summer. See the Supporting Information for detailed information. Long-term change beyond seasonal variation was analyzed by using the wavelet-denoising method, a robust statistical

Figure 3. Change of ambient nighttime temperatures Tmin and daily temperature range ΔT at station 39-061-0040 in the central urban interior. The regression slopes are statistically significant at p < 0.0001. 10600


Article


Figure 4. Spatial variations of maximum T′avg and PM2.5 A) in sections A-A′ across the high-density zone and B) in section O−O′ along the zone starting at its southern tip. See Figure 2 for location of the cross sections. Vertical bar represents the difference between T′avg and T′0.9 (likewise in PM2.5 concentration differences) for each monitoring station.

average ∼2.9 °C temperature difference (4.3 °C in summer and 1.3 °C in winter) for 38 U.S. urban centers using MODIS satellite imageries21 but comparable to those by others in the U.S. and elsewhere.19,37−40 The source of difference has been discussed in the literature18,19,41 as a result of location-specific UHI-controlling factors (e.g., urban scale, wind fields, soil moisture difference, etc.) as well as a difference in the measurement methods (i.e., in situ monitoring versus remote ensing). The temperature variability is also sensitive to geometry and size of the high-density urbanized zone. No significant UHI effect is observed for station 37-061-0006 located in a small patch of high-density LULC; it is surrounded by the mixed zone characteristic of detached single houses with overgrown trees and large grassed yards dotted in large acres of nature preserves (Figure 2). Long-Term Urban-Scale PM2.5 Variability and LULC Association. The relative concentration PM2.5 ′ (= PM2.5,i − PM2.5,ref) follows similar variations as T′avg (Figure 4). Like ambient temperature, PM2.5 concentrations above UCL are also linearly correlated among all NQAAS stations. All 130 stationyear correlations have R2 from 0.53 to 0.99; average is 0.92. The high degree of correlation is persistent over PM2.5 concentration range of 1.2−52.1 mg/m3 (m̅ = 13.89, N = 1717) at the reference station. These LULC-dependent PM2.5 and Tavg variations are further examined in Figure 5. From eq 1 for each location

The large increase in nighttime temperature largely contributes to a decrease of ΔT by 1.2 °C in a decade. Similar long-term changes in nighttime temperature have been reported elsewhere.4,5,35 Long-Term Urban-Scale Covariations of Temperature Variability and UHI Effects above UCL. Ambient temperatures above the UCL are highly correlated in the 10.9-year period among all 15 monitoring stations. Linear regressions of all 91 to 116 station-year measurements yield an average R2 = 0.993 (0.941−0.999). Martuzevicius et al.36 reached the same conclusion using a smaller data set in the same area. For statistical temperature parameters at reference station 39-0610040 (Tref), their equivalents are calculated for each of the other stations. The obtained temperature equivalents Ti were further converted into temperature difference T′(= Ti − Tref). The statistically characteristic temperature allows comparison among all NAAQS stations in the monitoring period. Obtained T′ varies with LULC types. Higher values are contiguously found in the high-density zone of the urban core. In cross-section A-A′ (see Figure 2), the annual mean and 90% percentile of daily T′avg increases sharply at the boundary (Figure 4). The 90% percentile represents a high daily T′0.9 occurring in a summer month each year. Outside of the zone, the annual mean and the summer T′0.9 are lower by 0.89 ± 0.14 °C and 1.55 ± 0.30 °C, respectively. In the long axis profile O−O′, all parameters display a slight and gradual increase from the edge toward the center. In both cases, the summer UHI in the daily T′avg is more enhanced as shown in cross sections (Figure 4). The observed degree of UHI is much less than predicted using the population-based UHI formula; the UHI equation yields a 4.91 °C increase for the city of ∼1.8 million people. Observed temperature increase is also smaller than the

αΔT =

ΔT i − ϵΔT ΔT ref

R(Tmin/Tmax) = 10601

αTmin αTmax

=

ref i − ϵmin ⎛ Tmax ⎞ Tmin · ⎜ ⎟ i − ϵmax ⎝ Tmin ⎠ Tmax DOI: 10.1021/acs.est.5b02444 Environ. Sci. Technol. 2015, 49, 10598−10606

Article


Figure 5. Decade-long covariation between ΔT and T−PM2.5 ratios for different LULC types. Filled triangle for the reference station; arrows 1 and 2 for trending toward high-density zone (HDZ) and other LULC types, respectively.

R(Tmin/CPM 2.5) =

αTmin αCPM2.5

Figure 6. Trajectories of thermal inversion development in Lh − Zinv diagram for the South End (S2) and the Lunken airport (S3) locations. The phases before and after the peak inversion are shown by filled and empty symbols, respectively. “S” and “W” symbols indicate strong and weak inversion.

ref i − ϵmin ⎛ CPM 2.5 ⎞ Tmin = i ·⎜ ⎟ CPM 2.5 − ϵPM 2.5 ⎝ Tmin ⎠

(4)

followed by a rapid increase in inversion height in the early morning. This two-phased inversion trajectory forms concaved curves in Figure 6. Gaussian and Non-Gaussian Dispersion in UCL. During the inversion episodes, near-road concentrations of black carbon (organic carbon or OC and elemental carbon or EC) and PM2.5 at ground level (Z ∼ 1.7m) were measured at the I-75 highway curbside (Figure 1). The details of this field study can be found in the publication by Liang et al.16 Classic Gaussian dispersion of PM2.5, OC, and EC concentrations were reported in the daytime sampling periods with winds up to 2−3 m/s. The variations were non-Gaussian in early morning with calm winds 1 indicates relatively higher Tmax or higher CPM2.5 for the stations at the inside of the high-density zone than other areas (Figure 5). The division among LULC types is obvious in Figure 5, for which the decade-long, persistent difference between the urban core and the exurban area is inferred. Because of UHI and thermal inversion, the high-density zone has a higher daily Tmax, a larger CPM2.5, with αΔT ≥ 1. For these stations, the slope ratios for both Tmin/Tmax and Tmin/CPM2.5 are close to one. The variation is opposite of those in the exurban areas: smaller Tmax, CPM2.5, and consequently αΔT < 1 with the large slope ratios in Tmin/Tmax and Tmin/CPM2. The end member is represented by the regional background station (Figure 5). Diurnal Inversion Occurrence. Temperature and pollutant transport during each thermal inversion episode was analyzed at hourly intervals for 11 days in October 6−15, 2010. Lapse rate (Lh) and inversion height (Zinv) were determined using eqs 2 and 3 for S1-S4 locations in Figure 2. Reconstructed atmospheric temperature profiles show a sequential daily occurrence of nocturnal thermal inversion (see the Supporting Information). The inversion strength at the S3 location east of the high-density zone was shallower than at other locations. Determined Lh and Zinv form trajectories of thermal inversion progression and destruction in a diurnal cycle (Figure 6). Starting in the evening, both Lh and Zinv increased following either a strong (S) or a weak (W) inversion trajectory. Inversion strength reached a maximum − the largest measured at −30 °C/km, before sunrise in the early morning. Subsequently the rapid inversion destruction occurred at a small increase in Zinv,

ef , EC = 0.0172 − 0.0016Lh ;

R2 = 0.986

ef , OC = 0.0333 − 0.0080Lh ;

R2 = 0.983

COC /CEC = 1.683 − 0.152Lh ;

R2 = 0.992

The black carbon and PM variation at daytime with absent thermal inversion follows the Gaussian dispersion with distance. According to the boundary similarity theory,50 the ef values are negatively correlated with T−1/3 (Figure 7b): ef , OC = 0.195 − 0.449T −1/3;

R2 = 0.792

COC /CEC = 13.659 − 31.761T −1/3; 10602

R2 = 0.779


Article


Figure 7. Linear regression of effective emission factor and OC/EC ratio (R) measured at I-75 highway curbside against (A) lapse rate Lh in morning hours and (B) 1/T1/3 in the afternoon period. Data from Liang et al.16

Figure 8. Near-ground (Z = 1.7 m) minutely average wind speed at the I-75 site covaried with the lapse rate in UBL during thermal inversion periods.

or field experiments.28,29 Despite the limitations in these investigation methods,53 the unique thermal property of urbanization is generally considered as a principle cause for the UHI effect and thermal inversion process. That urban construction materials absorb insulation in daytime and release sensible heat at night is responsible for initiating mesoscale atmospheric structure changes,6,11,13−15. Upward sensible heat flux and air aloof from the warmer urban interior induce the movement of colder air mass from surrounding rural areas. Other major factors in facilitating the circulation include urban-exurban difference in soil moisture, hilly and mountainous perimeters, and urban interior morphology.3,20 The circulation is often in the form of evening urban breezes, thermal inversions of various strength, and hence the concurrence of UHI and weak wind energy.9,10,29 These indicators were abundant for the Cincinnati area, as revealed in this study for the 10.9-yr statistics of air monitoring data. The long-term atmospheric processes and interactions in Figure 9 are corroborated from recent publications.3,9,27,54 Major observations are as follows: •Temperature, PM2.5, and black carbon concentrations above UCL are closely correlated among urban-wide NAAQS stations. This decade-long and statistically persistent correlation, although at coarse spatial resolution, reflects an effective heat flux and air circulation. Within the UCL, air pollutants (e.g., black carbon and PMs) in the absence of thermal inversion were attenuated

Wind Field and Thermal Inversion Covariations. Furthermore, nocturnal thermal inversion is correlated to nearground wind speed and temperature (Figure 8). The weak SBL in October 14−15 was accompanied by high wind speed at 1.47−2.6 m/s, low OC and EC concentrations, and a low OC/EC ratio around 1.272, all under prevailing Gaussian dispersion from the source. Black carbon concentrations decreased rapidly from road source to an urban background level of a 250 m distance. In the preceding period October 7−13, a very strong SBL (Lh < − 10 °C/km and Lh,min = −29.2 °C/km) occurred in the early morning hours with characteristic weak and oscillating winds. The strong thermal inversion was nonstationary and induced a downward thermal flux due to radiative urban cooling.28,45,51,52 Non-Gaussian transport such as meandering may have occurred under a strong thermal inversion, leading to high concentrations of black carbon (and other pollutants) and weak wind speed in the UCL.16 In agreement, the measured black carbon ef values are quantitatively related to the lapse rate (see Figure 7a).

■

DISCUSSIONS Urban atmospheric structure and diurnal occurrence of thermal inversion modulate UBL height and wind field variations and hence air pollutant concentrations. This interrelationship was implied in the early modeling of nocturnal inversion by Uno et al.45 and later quantitatively examined using model simulation 10603


Article


population size. The use of green space, urban morphology, and infrastructure orientations according to local topography may change UCL composition, thus affecting UHI development and thus the thermal inversion occurrence.9,28,40,61,62 Furthermore, these results arguably indicate a strong potential of reducing the UHI effects and thermal inversion impacts by changing the urban canopy layer and thus modifying its interaction with overlying the UBL. Such adaptation planning necessitates quantitative modeling of UHI and thermal inversion occurrence. For the latter, the use of urban climate zone classification63 and detailed model parametrization of urban canopy and interior heterogeneity28,40,53,55,61 need attention and further studies.

■

ASSOCIATED CONTENT

S Supporting Information *

Figure 9. Schematic illustration of urban formation regulating thermal flux in two phases, thus controlling the UHI formation, stagnant air flow, and air pollutant ventilation.

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.5b02444. Additional details and results on the long-term temperature and PM2.5 variations in urban core, urban-wide covariations in temperature and PM2.5, and thermal inversion determination; Tables S1−S3, and Figures S1 and S2 (PDF)

rapidly by wind-facilitated Gaussian dispersion from traffic routes and local sources. •Above the UCL, temperature spatial variability and UHI effects are a function of LULC types. Contiguous large areas of high-density urban development along the Mill Creek valley are associated with higher daytime maximum Tmax by 1.55 ± 0.30 °C and higher PM2.5 and black carbon concentrations. •Pollutant concentration within UCL increased significantly under nocturnal thermal inversion, particularly with the very stable UBL (Lh < − 10 °C/km), small Zinv, and calm winds with little circulation. After cold air mass occupies the urban core, the upward thermal flux is terminated at the inversion peak around 4 a.m. (Figure 9b), followed by the destruction of thermal inversion and a large increase of Zinv in the morning hours. Notably, the urban thermal flux and air circulation are mechanistically related to the spatial distribution of three LULC types: high-density urbanized area, low-density residential, and exurban.9,55 Above the urban core, thermal perturbation in the form of upward air flux and wind variability are confined to the air column above the urban core.27 These inferences agree well with the large T′avg, especially nighttime T′min, along the narrow high-density zone (see Figure 2). Facilitated by valley topography,3,20,24 the upward thermal and air flows may help form an urban-wide air circulation cell, thus bringing exurban cold air to the urban core under the warmer air leading to thermal inversion (Figure 9a). These mechanistic processes have been demonstrated in both numerical simulations and field experiments previously.9,27,56,57 In the early morning, Lh and Zinv increased rapidly and wind speed increased leading to the recovery of a normal thermal gradient and the disappearance of the diurnal thermal inversion (Figure 8). While the simple mechanisms in Figure 9 appear plausible, other influencing factors can further complicate the occurrence and nature of the UHI and thermal inversion. These include urban-rural soil moisture differences, wind speeds, green space and land cover, local topography, urban form, and albedos of construction materials.15,22,31,58−60 A detailed investigation in the Washington-Baltimore region55 showed the heterogeneous LULC feedbacks of the high-density urbanized area in relation to upwind air flow. This study has also identified low degrees of UHI effect in the predominantly low-density residential areas and the UHI absence for the two isolated small areas of high-density zones. It follows that the aggregate size and location of the high-density zone are the leading factors, rather than

■

AUTHOR INFORMATION

Corresponding Author

*Phone: 513-556-3676. Fax: 513-556-142. E-mail: tim.keener@ uc.edu. Corresponding author address: 472 ERC (ML 0012), Cincinnati, Ohio 45221-0012. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors are grateful to Anna Kelly at the Hamilton County Department of Environmental Services, Dr. Mingming Lu of the University of Cincinnati for data sharing and discussion, Dr. Jeff Yang of the USEPA Office of Research and Development National Risk Management Research Laboratory for his insightful comments and suggestions, and to Dr. Roberta Campbell of EPA for editorial support. Constructive comments from two anonymous reviewers are acknowledged. This study is a cooperative effort of the EPA water resources adaptation program (WRAP) research in support of EPA’s Sustainable and Healthy Community (SHC) and Air, Climate and Energy (ACE) programs. Any opinions expressed herein are those of the authors and do not necessarily reflect the views of the Agency; therefore, no official endorsement should be inferred.

■

REFERENCES

(1) Georgescu, M.; Miguez-Macho, G.; Steyaert, L.; Weaver, C. Climatic effects of 30 years of landscape change over the Greater Phoenix, Arizona, region: 1. Surface energy budget changes. J. Geophys. Res. 2009, 114 (D5); DOI 10.1029/2008JD010745. (2) 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. U. S. A. 2014, 111 (8), 2909−2914. (3) Fernando, H. J. S. Fluid dynamics of urban atmospheres in complex terrain. Annu. Rev. Fluid Mech. 2010, 42, 365−389. (4) Wang, K.; Ye, H.; Chen, F.; Xiong, Y.; Wang, C. Urbanization effect on the diurnal temperature range: different roles under solar dimming and brightening*. J. Clim. 2012, 25 (3), 1022−1027.

10604


Article

Environmental Science & Technology (5) Braganza, K.; Karoly, D. J.; Arblaster, J. Diurnal temperature range as an index of global climate change during the twentieth century. Geophys. Res. Lett. 2004, 31 (13). (6) Erell, E.; Pearlmutter, D.; Boneh, D.; Kutiel, P. B. Effect of highalbedo materials on pedestrian heat stress in urban street canyons. Urban Climate 2014, 10 (2), 367−386. (7) Doulos, L.; Santamouris, M.; Livada, I. Passive cooling of outdoor urban spaces. The role of materials. Sol. Energy 2004, 77 (2), 231−249, http://dx.doi.org/10.1016/j.solener.2004.04.005,. (8) Santamouris, M.; Gaitani, N.; Spanou, A.; Saliari, M.; Giannopoulou, K.; Vasilakopoulou, K.; Kardomateas, T. Using cool paving materials to improve microclimate of urban areas−Design realization and results of the flisvos project. Build. Environ. 2012, 53, 128−136. (9) Hidalgo, J.; Masson, V.; Gimeno, L. Scaling the daytime urban heat island and urban-breeze circulation. Journal of Applied Meteorology and Climatology 2010, 49 (5), 889−901. (10) Rendón, A. M.; Salazar, J. F.; Palacio, C. A.; Wirth, V.; Brötz, B. Effects of Urbanization on the Temperature Inversion Breakup in a Mountain Valley with Implications for Air Quality. Journal of Applied Meteorology and Climatology 2014, 53 (4), 840−858. (11) Oke, T. The heat island of the urban boundary layer: characteristics, causes and effects. NATO ASI Series E Applied SciencesAdvanced Study Institute 1995, 277, 81−108. (12) Rotach, M.; Vogt, R.; Bernhofer, C.; Batchvarova, E.; Christen, A.; Clappier, A.; Feddersen, B.; Gryning, S.; Martucci, G.; Mayer, H. BUBBLE−an urban boundary layer meteorology project. Theoretical and Applied Climatology 2005, 81 (3−4), 231−261. (13) Oke, T. R. The energetic basis of the urban heat island. Q. J. R. Meteorol. Soc. 1982, 108 (455), 1−24. (14) Voogt, J. A.; Oke, T. R. Thermal remote sensing of urban climates. Remote Sens. Environ. 2003, 86 (3), 370−384. (15) Liu, Y.; Shintaro, G.; Zhuang, D.; Kuang, W. Urban surface heat fluxes infrared remote sensing inversion and their relationship with land use types. Journal of Geographical Sciences 2012, 22 (4), 699−715. (16) Liang, M. S.; Keener, T. C.; Birch, M. E.; Baldauf, R.; Neal, J.; Yang, Y. J. Low-wind and other microclimatic factors in near-road black carbon variability: A case study and assessment implications. Atmos. Environ. 2013, 80, 204−215. (17) Oke, T. R. The distinction between canopy and boundary-layer urban heat islands. Atmosphere 1976, 14 (4), 268−277. (18) Roth, M.; Oke, T.; Emery, W. Satellite-derived urban heat islands from three coastal cities and the utilization of such data in urban climatology. Int. J. Remote Sens. 1989, 10 (11), 1699−1720. (19) Bornstein, R. D. Observations of the urban heat island effect in New York City. J. Appl. Meteorol. 1968, 7 (4), 575−582. (20) Clarke, J. F. Nocturnal urban boundary layer over Cincinnati, Ohio. Mon. Weather Rev. 1969, 97 (8), 582−589. (21) 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 (3), 504−513. (22) Buyadi, S. N. A.; Mohd, W. M. N. W.; Misni, A. Green Spaces Growth Impact on the Urban Microclimate. Procedia-Social and Behavioral Sciences 2013, 105, 547−557. (23) Ferrero, L.; Riccio, A.; Perrone, M.; Sangiorgi, G.; Ferrini, B.; Bolzacchini, E. Mixing height determination by tethered balloon-based particle soundings and modeling simulations. Atmos. Res. 2011, 102 (1), 145−156. (24) Anquetin, S.; Guilbaud, C.; Chollet, J. The formation and destruction of inversion layers within a deep valley. J. Appl. Meteorol. 1998, 37 (12), 1547−1560. (25) Sangiorgi, G.; Ferrero, L.; Perrone, M.; Bolzacchini, E.; Duane, M.; Larsen, B. Vertical distribution of hydrocarbons in the low troposphere below and above the mixing height: Tethered balloon measurements in Milan, Italy. Environ. Pollut. 2011, 159 (12), 3545− 3552. (26) Trompetter, W.; Grange, S.; Davy, P.; Ancelet, T. Vertical and temporal variations of black carbon in New Zealand urban areas during winter. Atmos. Environ. 2013, 75, 179−187.

(27) Banta, R. M.; Senff, C. J.; White, A. B.; Trainer, M.; McNider, R. T.; Valente, R. J.; Mayor, S. D.; Alvarez, R. J.; Hardesty, R. M.; Parrish, D. Daytime buildup and nighttime transport of urban ozone in the boundary layer during a stagnation episode. J. Geophys. Res. 1998, 103 (D17), 22519−22544. (28) Martilli, A. Numerical study of urban impact on boundary layer structure: Sensitivity to wind speed, urban morphology, and rural soil moisture. J. Appl. Meteorol. 2002, 41 (12), 1247−1266. (29) Rotach, M.; Vogt, R.; Bernhofer, C.; Batchvarova, E.; Christen, A.; Clappier, A.; Feddersen, B.; Gryning, S.; Martucci, G.; Mayer, H. BUBBLE−an urban boundary layer meteorology project. Theoretical and Applied Climatology 2005, 81 (3−4), 231−261. (30) Chou, C. C.; Lee, C.; Chen, W.; Chang, S.; Chen, T.; Lin, C.; Chen, J. Lidar observations of the diurnal variations in the depth of urban mixing layer: A case study on the air quality deterioration in Taipei, Taiwan. Sci. Total Environ. 2007, 374 (1), 156−166. (31) Weng, Q.; Lu, D.; Schubring, J. Estimation of land surface temperature−vegetation abundance relationship for urban heat island studies. Remote Sens. Environ. 2004, 89 (4), 467−483. (32) Ma, X. L.; Schmit, T. J.; Smith, W. L. A nonlinear physical retrieval algorithm-its application to the GOES-8/9 sounder. J. Appl. Meteorol. 1999, 38 (5), 501−513. (33) Andronache, C.; Chameides, W.; Rodgers, M.; Martinez, J.; Zimmerman, P.; Greenberg, J. Vertical distribution of isoprene in the lower boundary layer of the rural and urban southern United States. J. Geophys. Res. 1994, 99 (D8), 16989−16999. (34) Torrence, C.; Compo, G. P. A practical guide to wavelet analysis. Bull. Am. Meteorol. Soc. 1998, 79 (1), 61−78. (35) Rebetez, M. Changes in daily and nightly day-to-day temperature variability during the twentieth century for two stations in Switzerland. Theoretical and Applied Climatology 2001, 69 (1−2), 13−21. (36) Martuzevicius, D.; Luo, J.; Reponen, T.; Shukla, R.; Kelley, A. L.; Clair, H. S.; Grinshpun, S. A. Evaluation and optimization of an urban PM 2.5 monitoring network. J. Environ. Monit. 2005, 7 (1), 67−77. (37) Morris, C.; Simmonds, I.; Plummer, N. Quantification of the influences of wind and cloud on the nocturnal urban heat island of a large city. J. Appl. Meteorol. 2001, 40 (2), 169−182. (38) Peng, S.; Piao, S.; Ciais, P.; Friedlingstein, P.; Ottle, C.; Bréon, F.; Nan, H.; Zhou, L.; Myneni, R. B. Surface urban heat island across 419 global big cities. Environ. Sci. Technol. 2012, 46 (2), 696−703. (39) Gaffin, S.; Rosenzweig, C.; Khanbilvardi, R.; Parshall, L.; Mahani, S.; Glickman, H.; Goldberg, R.; Blake, R.; Slosberg, R.; Hillel, D. Variations in New York City’s urban heat island strength over time and space. Theoretical and applied climatology 2008, 94 (1−2), 1−11. (40) Lindén, J. Nocturnal cool island in the Sahelian city of Ouagadougou, Burkina Faso. Int. J. Climatol. 2011, 31 (4), 605−620. (41) Peterson, T. C. Assessment of urban versus rural in situ surface temperatures in the contiguous United States: No difference found. J. Clim. 2003, 16 (18), 2941−2959. (42) Thoma, E. D.; Shores, R. C.; Isakov, V.; Baldauf, R. W. Characterization of near-road pollutant gradients using path-integrated optical remote sensing. J. Air Waste Manage. Assoc. 2008, 58 (7), 879− 890. (43) Chock, D. P. General Motors sulfate dispersion experiment: An analysis of the wind field near a road. Boundary-Layer Meteorology 1980, 18, 431−451. (44) Baldauf, R.; Watkins, N.; Heist, D.; Bailey, C.; Rowley, P.; Shores, R. Near-road air quality monitoring: Factors affecting network design and interpretation of data. Air Qual., Atmos. Health 2009, 2 (1), 1−9. (45) Uno, I.; Wakamatsu, S.; Ueda, H.; Nakamura, A. An observational study of the structure of the nocturnal urban boundary layer. Bound. -Layer Meteorol. 1988, 45 (1−2), 59−82. (46) Guzmán-Torres, D.; Eiguren-Fernández, A.; Cicero-Fernández, P.; Maubert-Franco, M.; Retama-Hernández, A.; Villegas, R. R.; Miguel, A. H. Effects of meteorology on diurnal and nocturnal levels of priority polycyclic aromatic hydrocarbons and elemental and organic carbon in PM 10 at a source and a receptor area in Mexico City. Atmos. Environ. 2009, 43 (17), 2693−2699. 10605


Article

Environmental Science & Technology (47) Kumar, M. S.; Anandan, V.; Rao, T. N.; Reddy, P. N. A climatological study of the nocturnal boundary layer over a complexterrain station. Journal of Applied Meteorology and Climatology 2012, 51 (4), 813−825. (48) Vinod Kumar, A.; Patil, R. S.; Nambi, K. S. V. A composite receptor and dispersion model approach for estimation of effective emission factors for vehicles. Atmos. Environ. 2004, 38 (40), 7065−7072. (49) Kristensson, A.; Johansson, C.; Westerholm, R.; Swietlicki, E.; Gidhagen, L.; Wideqvist, U.; Vesely, V. Real-world traffic emission factors of gases and particles measured in a road tunnel in Stockholm, Sweden. Atmos. Environ. 2004, 38 (5), 657−673. (50) Venkatram, A.; Isakov, V.; Thoma, E.; Baldauf, R. Analysis of air quality data near roadways using a dispersion model. Atmos. Environ. 2007, 41 (40), 9481−9497. (51) Uno, I.; Ueda, H.; Wakamatsu, S. Numerical modeling of the nocturnal urban boundary layer. Bound. -Layer Meteorol. 1989, 49 (1− 2), 77−98. (52) Iziomon, M. G.; Mayer, H.; Matzarakis, A. Downward atmospheric longwave irradiance under clear and cloudy skies: Measurement and parameterization. J. Atmos. Sol.-Terr. Phys. 2003, 65 (10), 1107−1116. (53) Mirzaei, P. A.; Haghighat, F. Approaches to study urban heat island−abilities and limitations. Build. Environ. 2010, 45 (10), 2192− 2201. (54) Mahrt, L. Stratified atmospheric boundary layers. Bound. -Layer Meteorol. 1999, 90 (3), 375−396. (55) Chen, F.; Kusaka, H.; Bornstein, R.; Ching, J.; Grimmond, C.; Grossman-Clarke, S.; Loridan, T.; Manning, K. W.; Martilli, A.; Miao, S. The integrated WRF/urban modelling system: development, evaluation, and applications to urban environmental problems. Int. J. Climatol. 2011, 31 (2), 273−288. (56) Banta, R. M.; Mahrt, L.; Vickers, D.; Sun, J.; Balsley, B. B.; Pichugina, Y. L.; Williams, E. J. The very stable boundary layer on nights with weak low-level jets. J. Atmos. Sci. 2007, 64 (9), 3068−3090. (57) Tennekes, H. A model for the dynamics of the inversion above a convective boundary layer. J. Atmos. Sci. 1973, 30 (4), 558−567. (58) Loridan, T.; Grimmond, C. Characterization of energy flux partitioning in urban environments: links with surface seasonal properties. Journal of Applied Meteorology and Climatology 2012, 51 (2), 219−241. (59) Kotthaus, S.; Grimmond, C. Energy exchange in a dense urban environment−Part II: Impact of spatial heterogeneity of the surface. Urban Climate 2014, 10 (2), 281−307. (60) Avissar, R.; Pielke, R. A. A parameterization of heterogeneous land surfaces for atmospheric numerical models and its impact on regional meteorology. Mon. Weather Rev. 1989, 117 (10), 2113−2136. (61) Salamanca, F.; Martilli, A.; Tewari, M.; Chen, F. A study of the urban boundary layer using different urban parameterizations and highresolution urban canopy parameters with WRF. Journal of Applied Meteorology and Climatology 2011, 50 (5), 1107−1128. (62) Baik, J.; Kim, Y.; Chun, H. Dry and moist convection forced by an urban heat island. J. Appl. Meteorol. 2001, 40 (8), 1462−1475. (63) Oke, T. R. Initial guidance to obtain representative meteorological observations at urban sites; World Meteorological Organization Geneva: 2004.

10606


Atmospheric Feedback of Urban Boundary Layer with Implications for

Recommend Documents