There has been significant progress in the studies focusing on LST–vegetation rela- tionship, UHI modeling with remotely sensed TIR data, and estimation of urban sur- face heat fluxes� However, urban climate and environmental studies will be difficult, if not impossible, without TIR sensors having a global imaging capacity� At present, there are few sensors that have such TIR capabilities� The TM sensor aboard Landsat 5 has been acquiring images of the Earth nearly continuously from July 1982 to the present, with a TIR band of 120-m resolution, and is thus long overdue� On April 2, 2007, updates to the radiometric calibration of Landsat 5 TM data processed and dis- tributed by the U�S� Geological Survey (USGS) Earth Resources Observation System (EROS) created an improved Landsat 5 TM data product that is now more compa- rable radiometrically to Landsat 7 ETM+ and provides the basis for continued long- term studies of the Earth’s land surfaces� Another TIR sensor that has global imaging capacity is with Landsat 7 ETM+� On May 31, 2003, the ETM+ scan-line corrector (SLC) failed permanently� Although it is still capable of acquiring useful image data with the SLC turned off, particularly within the central part of any given scene, the National Aeronautics and Space Administration (NASA) has teamed up with USGS to focus on the Landsat Data Continuity Mission (LDCM), which is most likely not to have a TIR imager� In addition, Terra’s ASTER TIR bands of 90-m resolution have been increasingly used in urban climate and environmental studies in recent years�
The ASTER is an on-demand instrument, which means that data are acquired only over requested locations� The Terra satellite, launched in December 1999 as part of NASA’s Earth Observing System, has a life expectancy of 6 years, and is now also overdue� The scientific and user community is looking forward to a Landsat ETM–like TIR sensor� The draft requirements for the LDCM thermal imager indicate that two thermal bands (10�3–11�3 μm and 11�5–12�5 μm) of 90 m or better spatial resolution are preferred (for details, readers are referred to the LDCM Web site, http://ldcm�nasa�gov/
procurement/TIRimagereqs051006�pdf)� The National Research Council Decadal Survey indicates the need for such a TIR sensor� The Hyperspectral Infrared Imager (HyspIRI) is defined as a mission with tier-2 priority to be launched in the next 8–10 years� Because of its hyperspectral visible and shortwave infrared bandwidths and its multispectral TIR capabilities, HyspIRI will be well suited for deriving land-cover and other biophysi- cal attributes for urban climate and environmental studies (for more information, the readers are referred to the HyspIRI Web site, http://hyspiri�jpl�nasa�gov/)� Its TIR imager is expected to provide seven bands between 7�5 and 12 μm and one band at 4 μm, all with 60-m spatial resolution� This TIR sensor is intended for the imaging of global land and shallow water (less than 50 m) with a 5-day revisit at the equator (1 day and 1 night imaging)� These improved capabilities would allow for a more accurate estimation of
LST and emissivity, and for deriving unprecedented information on biophysical char- acteristics and even socioeconomic information such as population, quality of life indi- cators, and human settlements� Such information cannot be obtained from the current generation of satellites devices in orbit, such as MODIS, Landsat, or ASTER� Two major areas of application identified by the HyspIRI science team are urbanization and human health through the combined use of visible to shortwave infrared (VSWIR) and TIR data�
Until then, we may have to bear with Landsat and ASTER for medium-resolution TIR data, and MODIS and AVHRR for coarse-resolution data� It is from this perspective that international collaborations on Earth resources satellites become very important�
Acknowledgments
This research is supported by the National Science Foundation (BCS-0521734) for a proj- ect entitled “Role of Urban Canopy Composition and Structure in Determining Heat Islands: A Synthesis of Remote Sensing and Landscape Ecology Approach�” The author would also like to acknowledge the following individuals for their contributions in vari- ous capacities to this chapter: Dale Quattrochi, Xuefei Hu, Rongbo Xiao, Reza Amiri, and Umamaheshwaren Rajasekar�
References
Amiri, R�, Q� Weng, A� Alimohammadi, and S� K� Alavipanah� 2009� The spatial-temporal dynamics of land surface temperatures in relation to fractional vegetation cover and land use/cover in the Tabriz urban area, Iran� Remote Sens Environ 113(12):2606–17�
Asrar, G�, M� Fuchs, E� T� Kanemasu, and J� L� Hatfield� 1984� Estimating absorbed photosynthetic radiation and leaf area index from spectral reflectance in wheat� Agron J 76(22):300–6�
Balling, R� C�, and S� W� Brazell� 1988� High resolution surface temperature patterns in a complex urban terrain� Photogramm Eng Remote Sens 54(9):1289–93�
Betts, A�, J� Ball, A� Beljaars, M� Miller, and P� Viterbo� 1996� The land surface-atmosphere interaction: A review based on observational and global modeling perspectives� J Geophys Res 101(D3):7209–25�
Boegh, E�, H� Soegaard, N� Hanan, P� Kabat, and L� Lesch� 1998� A remote sensing study of the NDVI-Ts relationship and the transpiration from sparse vegetation in the Sahel based on high resolution satellite data� Remote Sens Environ 69(3):224–40�
Bottyán, Z�, and J� Unger� 2003� A multiple linear statistical model for estimating the mean maximum urban heat island� Theor Appl Climatol 75(3–4):233–43�
Carlson, T� N�, and S� T� Arthur� 2000� The impact of land use–land cover changes due to urban- ization on surface microclimate and hydrology: A satellite perspective� Global Planet Change 25(1–2):49–65�
Carlson, T� N�, J� K� Dodd, S� G� Benjamin, and J� N� Cooper� 1981� Satellite estimation of the surface energy balance, moisture availability and thermal inertia� J Appl Meteorol 20(1):67–87�
Carlson, T� N�, R� R� Gillies, and E� M� Perry� 1994� A method to make use of thermal infrared tempera- ture and NDVI measurements to infer surface water content and fractional vegetation cover�
Remote Sens Rev 9(1–2):161–73�
Carlson, T� N�, and G� A� Sanchez-Azofeifa� 1999� Satellite remote sensing of land use changes in and around San José, Costa Rica� Remote Sens Environ 70(3):247–56�
Carnahan, W� H�, and R� C� Larson� 1990� An analysis of an urban heat sink� Remote Sens Environ 33(1):65–71�
Cendese, A�, and P� Monti� 2003� Interaction between an inland urban heat island and a sea-breeze flow: A laboratory study� J Appl Meteorol 42(11):1569–83�
Chen, X� -L�, H� -M� Zhao, P� -X� Li, and Z� -Y� Yin� 2006� Remote sensing image-based analysis of the relationship between urban heat island and land use/cover changes� Remote Sens Environ 104(2):133–46�
Chrysoulakis, N� 2003� Estimation of the all-wave urban surface radiation balance by use of ASTER multispectral imagery and in situ spatial data� J Geophys Res D Atmos 108(18):ACL 8-1–10�
Chudnovsky, A�, E� Ben-Dor, and H� Saaroni� 2004� Diurnal thermal behavior of selected urban objects using remote sensing measurements� Energy Build 36(11):1063–74�
Dousset, B�, and F� Gourmelon� 2003� Satellite multi-sensor data analysis of urban surface temperature and land cover� ISPRS J Photogramm Remote Sens 58(1–2):43–54�
Eliasson, I� 1996� Urban nocturnal temperatures, street geometry and land use� Atmos Environ 30(3):379–92�
Elvidge, C� D�, K� E� Baugh, E� A� Kihn, H� W� Kroehl, E� R� Davis, and C� W� Davis� 1997� Relation between satellite-observed visible-near infrared emissions, population, economic activity and electric power consumption� Int J Remote Sens 18(6):1373–9�
Fan, H�, and D� J� Sailor� 2005� Modeling the impacts of anthropogenic heating on the urban climate of Philadelphia: A comparison of implementations in two PBL schemes� Atmos Environ 39(1):73–84�
Friedl, M� A�, and F� W� Davis� 1994� Sources of variation in radiometric surface temperature over a tallgrass prairie� Remote Sens Environ 48(1):1–17�
Frohn, R� C� 1998� Remote Sensing for Landscape Ecology� Boca Raton, FL: Lewis Publishers�
Gallo, K� P�, A� L� McNab, T� R� Karl, J� F� Brown, J� J� Hood, and J� D� Tarpley� 1993� The use of NOAA AVHRR data for assessment of the urban heat island effect� J Appl Meteorol 32(5):899–908�
Gallo, K� P�, and T� W� Owen� 1998� Assessment of urban heat island: A multi-sensor perspective for the Dallas-Ft� Worth, USA region� Geocarto Int 13(4):35–41�
Gillies, R� R�, and T� N� Carlson� 1995� Thermal remote sensing of surface soil water content with partial vegetation cover for incorporation into climate models� J Appl Meteorol 34(4):745–56�
Gillies, R� R�, T� N� Carlson, J� Cui, W� P� Kustas, and K� S� Humes� 1997� A verification of the “triangle”
method for obtaining surface soil water content and energy fluxes from remote measurements of the Normalized Difference Vegetation index (NDVI) and surface radiant temperature�
Int J Remote Sens 18(5):3145–66�
Goetz, S� J� 1997� Multisensor analysis of NDVI, surface temperature and biophysical variables at a mixed grassland site� Int J Remote Sens 18(1):71–94�
Goward, S� N�, Y� Xue, and K� P� Czajkowski� 2002� Evaluating land surface moisture conditions from the remotely sensed temperature/vegetation index measurements: An exploration with the simplified biosphere model� Remote Sens Environ 79(2–3):225–42�
Hafner, J�, and S� Q� Kidder� 1999� Urban heat island modeling in conjunction with satellite-derived surface/soil parameters� J Appl Meteorol 38(4):448–65�
Higdon, D� 2002� Space and space-time modeling using process convolution� In: Anderson, C� W�, Barnett, V�, Chatwin, P� C�, El-Shaarawi, A� H� (Eds�), Quantitative Methods for Current Environ- mental Issues� Springer-Verlag, London, pp� 37–56�
Jasinski, M� F� 1990� Sensitivity of the normalized difference vegetation index to subpixel canopy cover, soil albedo, and pixel scale� Remote Sens Environ 32(2–3):169–87�
Jiang, L�, and S� Islam� 2001� Estimation of surface evaporation map over southern Great Plains using remote sensing data� Water Resour Res 37(2):329–40�
Kato, S�, and Y� Yamaguchi� 2005� Analysis of urban heat-island effect using ASTER and ETM+ data:
Separation of anthropogenic heat discharge and natural heat radiation from sensible heat flux�
Remote Sens Environ 99(1–2):44–54�
Kidder, S� Q�, and H� T� Wu� 1987� A multispectral study of the St� Louis area under snow covered conditions using NOAA-7 AVHRR data� Remote Sens Environ 22(2):159–72�
Kim, H� H� 1992� Urban heat island� Int J Remote Sens 13(12):2319–36�
Kustas, W� P�, J� M� Norman, M� C� Anderson, and A� N� French� 2003� Estimating subpixel surface temperatures and energy fluxes from the vegetation index-radiometric temperature relation- ship� Remote Sens Environ 85(4):429–40�
Lambin, F� F�, and D� Ehrlich� 1996� The surface temperature-vegetation index space for land use and land cover change analysis� Int J Remote Sens 17(3):463–87�
Liu, H�, and Q� Weng� 2008� Seasonal variations in the relationship between landscape pattern and land surface temperature in Indianapolis, USA� Environ Monit Assess 144(1–3):199–219�
Lo, C� P�, D� A� Quattrochi, and J� C� Luvall� 1997� Application of high-resolution thermal infrared remote sensing and GIS to assess the urban heat island effect� Int J Remote Sens 18(2):287–304�
Lu, D�, and Q� Weng� 2006� Spectral mixture analysis of ASTER imagery for examining the relation- ship between thermal features and biophysical descriptors in Indianapolis, Indiana� Remote Sens Environ 104(2):157–67�
Moran, M� S�, T� R� Clarke, Y� Inoue, and A� Vidal� 1994� Estimating crop water deficit using the relation between surface-air temperature and spectral vegetation index� Remote Sens Environ 49(3):246–63�
Nemani, R� R�, and S� W� Running� 1989� Estimation of regional surface resistance to evapotranspira- tion from NDVI and thermal-IR AVHRR data� J Appl Meteorol 28(4):276–84�
Nichol, J� E� 1994� A GIS-based approach to microclimate monitoring in Singapore’s high-rise housing estates� Photogramm Eng Remote Sens 60(10):1225–32�
Nichol, J� E� 1996� High-resolution surface temperature patterns related to urban morphology in a tropical city: A satellite-based study� J Appl Meteorol 35(1):135–46�
Nichol, J� E� 1998� Visualization of urban surface temperature derived from satellite images� Int J Remote Sens 19(9):1639–49�
Nichol, J� E� 2005� Remote sensing of urban heat islands by day and night� Photogramm Eng Remote Sens 71(5):613–23�
Nishida, K�, R� R� Nemani, J� M� Glassy, and S� W� Running� 2003� Development of an evapotranspira- tion index from Aqua/MODIS for monitoring surface moisture status� IEEE Trans Geosci Remote Sens 41(2):493–501�
Oke, T� R� 1982� The energetic basis of the urban heat island� Q J R Meteorolog Soc 108(455):1–24�
Oke, T� R� 1988� The urban energy balance� Prog Phys Geog 12(4):471–508�
Oke, T� R�, R� A� Spronken-Smith, E� Jauregui, and C� S� B� Grimmond� 1999� The energy balance of central Mexico City during the dry season� Atmos Environ 33(24–25):3919–30�
Owen, T� W�, T� N� Carlson, and R� R� Gillies� 1998� An assessment of satellite remotely-sensed land cover parameters in quantitatively describing the climatic effect of urbanization� Int J Remote Sens 19(9):1663–81�
Price, J� C� 1990� Using spatial context in satellite data to infer regional scale evapotranspiration�
IEEE Trans Geosci Remote Sens 28(5):940–8�
Quattrochi, D� A�, and N� S� Goel� 1995� Spatial and temporal scaling of thermal remote sensing data�
Remote Sens Rev 12(3–4):255–86�
Quattrochi, D� A�, and J� C� Luvall� 1999� Thermal infrared remote sensing for analysis of landscape ecological processes: Methods and applications� Landsc Ecol 14(6):577–98�
Quattrochi, D� A�, and M� K� Ridd� 1994� Measurement and analysis of thermal energy responses from discrete urban surface using remote sensing data� Int J Remote Sens 15(10):1999–2022�
Rajasekar, U�, and Q� Weng� 2009� Urban heat island monitoring and analysis by data mining of MODIS imageries� ISPRS J Photogramm Remote Sens 64(1):86–96�
Rao, P� K� 1972� Remote sensing of urban heat islands from an environmental satellite� Bull Am Meteorol Soc 53:647–8�
Ridd, M� K� 1995� Exploring a V-I-S (Vegetation-Impervious Surface-Soil) model for urban ecosystem analysis through remote sensing: comparative anatomy for cities� Int J Remote Sens 16(12):2165–85�
Roth, M�, T� R� Oke, and W� J� Emery� 1989� Satellite derived urban heat islands from three coastal cities and the utilisation of such data in urban climatology� Int J Remote Sens 10:1699–720�
Saitoh, T� S�, T� Shimada, and H� Hoshi� 1996� Modeling and simulation of the Tokyo urban heat island� Atmos Environ 30:3431–42�
Sandholt, I�, K� Rasmussen, and J� Andersen� 2002� A simple interpretation of the surface temperature/
vegetation index space for assessment of surface moisture status� Remote Sens Environ 79:213–24�
Schmugge, T�, S� J� Hook, and C� Coll� 1998� Recovering surface temperature and emissivity from thermal infrared multispectral data� Remote Sens Environ 65:121–31�
Small, C� 2001� Estimation of urban vegetation abundance by spectral mixture analysis� Int J Remote Sens 22:1305–34�
Sobrino, J� A�, and N� Raissouni� 2000� Toward remote sensing methods for land cover dynamic monitoring: Application to Morocco� Int J Remote Sens 21:353–66�
Steininger, M� K� 1996� Tropical secondary forest regrowth in the Amazon: Age, area and change estimation with Thematic Mapper data� Int J Remote Sens 17(1):9–27�
Streutker, D� R� 2002� A remote sensing study of the urban heat island of Houston, Texas� Int J Remote Sens 23:2595–608�
Streutker, D� R� 2003� Satellite-measured growth of the urban heat island of Houston, Texas� Remote Sens Environ 85:282–9�
Tong, H�, A� Walton, J� Sang, and J� C� L� Chan� 2005� Numerical simulation of the urban boundary layer over the complex terrain of Hong Kong� Atmos Environ 39:3549–63�
Voogt, J� A�, and T� R� Oke� 1998� Effects of urban surface geometry on remotely sensed surface temperature� Int J Remote Sens 19:895–920�
Weng, Q� 2001� A remote sensing-GIS evaluation of urban expansion and its impact on surface temperature in the Zhujiang Delta, China� Int J Remote Sens 22:1999–2014�
Weng, Q� 2003� Fractal analysis of satellite-detected urban heat island effect� Photogramm Eng Remote Sens 69:555–66�
Weng, Q� 2009� Thermal infrared remote sensing for urban climate and environmental studies:
Methods, applications, and trends� ISPRS J Photogramm Remote Sens 64(4):335–44�
Weng, Q�, H� Liu, and D� Lu� 2007� Assessing the effects of land use and land cover patterns on thermal conditions using landscape metrics in city of Indianapolis, United States� Urban Ecosyst 10:203–19�
Weng, Q�, and D� Lu� 2008� A sub-pixel analysis of urbanization effect on land surface temperature and its interplay with impervious surface and vegetation coverage in Indianapolis, United States� Int J Appl Earth Obs Geoinf 10:68–83�
Weng, Q�, D� Lu, and B� Liang� 2006� Urban surface biophysical descriptors and land surface temperature variations� Photogramm Eng Remote Sensing 72(11):1275–86�
Weng, Q�, D� Lu, and J� Schubring� 2004� Estimation of land surface temperature-vegetation abun- dance relationship for urban heat island studies� Remote Sens Environ 89:467–83�
Xian, G�, and M� Crane� 2006� An analysis of urban thermal characteristics and associated land cover in Tampa Bay and Las Vegas using Landsat satellite data� Remote Sens Environ 104(2):147–56�
Xiao, R�, Q� Weng, Z� Ouyang, W� Li, E� W� Schienke, and W� Zhang� 2008� Land surface temperature variation and major factors in Beijing, China� Photogramm Eng Remote Sensing 74(4):451–61�
Yang, W�, L� Yang, and J� W� Merchnat� 1997� An analysis of AVHRR/NDVI-ecoclimatological relations in Nebraska, USA� Int J Remote Sens 18:2161–80�
Yuan, F�, and M� E� Bauer� 2007� Comparison of impervious surface area and normalized difference vegetation index as indicators of surface urban heat island effects in Landsat imagery� Remote Sens Environ 106(3):375–86�
Zhang, X�, Y� Aono, and N� Monji� 1998� Spatial variability of urban surface het fluxes estimated from Landsat TM data under summer and winter conditions� J Agric Meteorol 54(1):1–11�
Algorithms and Techniques
161
7
Atmospheric Correction Methods for Optical Remote Sensing Imagery of Land
Rudolf Richter
The optical part of the electromagnetic spectrum covers wavelengths from 100 nm to 1 mm� However, only a small part of the optical spectrum can be used for remote sens- ing from airborne and spaceborne platforms, because of the characteristics of the scat- tering, absorption, and emission of radiation by the terrestrial atmosphere� Figure 7�1 presents a typical atmospheric transmittance curve in those spectral regions that can be exploited with remote sensing techniques� Basically, there exist three large spectral intervals: 0�4–2�5 μm, 3–5 μm (mid-infrared or MIR), and 8–14 μm (thermal infrared or TIR)� For technical reasons, the first region is often split into the visible to near-infrared (VIS to NIR or VNIR; 0�4–1�0 μm; no detector cooling required) and short-wave infrared (SWIR; 1�0–2�5 μm; detector cooling required) regions� The main absorbing gases in the atmosphere are water vapor, ozone, carbon dioxide, and oxygen; the most variable gas in space and time is water vapor�
The 0�4–3�0 μm region is often referred to as a “reflective” or “solar” region� The reflected solar radiation dominates in this region compared to ambient emitted radia- tion, whereas the emitted TIR radiation dominates in the 8–14 μm domain (Figure 7�2)�
The reflected solar radiation is plotted for three surface reflectance (ρ) levels, and the emitted radiation for a 300-K blackbody� The atmospheric influence is neglected in this figure�
In the MIR interval, reflected solar and emitted thermal radiations have the same order of magnitude, and both contributions have to be considered during atmospheric correc- tion (AC)� As the majority of existing high spatial resolution instruments does not possess MIR channels, we will not discuss this case but refer to the works of Hook et al� (2001) and Mushkin, Balick, and Gillespie (2005)�
Table 7�1 contains an overview of some typical multispectral and hyperspectral instru- ments covering the reflective region, the TIR region with one channel, and the TIR region with more than 10 channels� Currently, all high spatial resolution (footprint <100 m) CONTENTS
7�1 Atmospheric Correction for Hyperspectral Instruments (Solar Region) ������������������ 163 7�2 Atmospheric Correction for the Thermal Region ��������������������������������������������������������� 165 7�3 Atmospheric Correction for Multispectral Instruments (Solar Region) �������������������� 166 7�4 Combined Atmospheric and Topographic Correction ������������������������������������������������� 167 7�5 Nonstandard Atmospheric Conditions (Haze, Cirrus, Cloud Shadow) �������������������� 168 7�6 Atmospheric Correction Codes for Land ����������������������������������������������������������������������� 169 7�7 Open Challenges ���������������������������������������������������������������������������������������������������������������� 170 References ������������������������������������������������������������������������������������������������������������������������������������� 171
1.0
0.8 O3
O2 H2O
H2O H2O
H2O H2O H2O
CO2 CO2
CO2
CO2 0.6
Transmittance 0.4
0.2
0.0 0.5 1.0 1.5
Wavelength (μm) 2.0 2.5
4 6 8
Wavelength (μm)10 12 14 O3
H2O CO2
H2O CO2 1.0
0.8
0.6
Transmittance 0.4
0.2
0.0 H2O CO2
FIgure 7.1
Atmospheric transmittance in the 0�4–2�5 μm and 3–14 μm regions�
102 101 100 10−1 10−2 (W.m−2.sr−1. μm−1)
10−3 10−4
1 10
Wavelength (μm) ρ =1.0
ρ =0.1 ρ =0.01 Sun 5800 K
300 K
FIgure 7.2
Solar or reflective region, mid-infrared (3–5 μm) region, and thermal infrared (8–14 μm) region�
hyperspectral instruments in orbit are restricted to the VNIR/SWIR region as they lack thermal channels� Airborne instruments covering the solar region and possessing at least a few thermal channels are still rare, and this is even more true for hyperspectral systems�
AC methods can be grouped into empirical approaches and physical models describ- ing radiative transfer (RT) in the Earth’s atmosphere� Here, we will only discuss RT-based approaches� As AC algorithms necessarily depend on the spectral regions covered by an instrument and also on the available number of channels, we will present the retrieval algorithms beginning with hyperspectral systems and terminating with a few channel multispectral instruments�