Fusion of Airborne Hyperspectral and LiDAR Remote Sensing Data to Study the Thermal Characteristics of Urban Environments - Computational Approaches for Urban Environments - page 266

11.2 Related Work

Remote sensing has become an increasingly important technology to gain a better
understanding of the urban climate (Arnfield 2003 ; Voogt and Oke 2003 ; Heldens
et al. 2011 ). However, while a lot of studies have considered the urban climate at the
macro- or mesoscale, only little research has been directed toward the use of Earth
observation data for urban microclimate analyses. This is partly because these kind
of analyses usually focus on spatial scales smaller than 2 km (Helbig et al. 1999 )
and, thus, require high spatial resolution input data to properly resolve all relevant
urban land cover elements (Welch 1982 ; Woodcock and Strahler 1987 ;Jensenand
Cowen 1999 ). Thanks to recent technological advancements, this requirement is
fulfilled by a growing number of airborne and satellite-based sensors (Ehlers 2009 ).
Quattrochi and Ridd ( 1994 ) were among the first to employ remote sensing
imagery for the study of urban microclimates. They investigated the thermal day and
night responses of 25 urban surface materials in Salt Lake City, Utah, using airborne
Thermal Infrared Multispectral Scanner (TIMS) data. Ben-Dor and Saaroni ( 1997 )
examined the microscale structures of the UHI in Tel-Aviv, Israel, based on data
provided by a thermal video radiometer mounted on a helicopter. Stone and Norman
( 2006 ) combined Advanced Thermal and Land Applications Sensor (ATLAS) data
with property tax records of Atlanta, Georgia, to assess the influence of the size
and material composition of single-family residential land use parcels on surface
UHI formation. Jung et al. ( 2007 ) performed a joint analysis of Digital Airborne
Imaging Spectrometer (DAIS) and additional thermal data acquired over Gyöngyös,
Hungary, to define the relationship between the abundance of urban vegetation and
land surface temperature (LST). Rigo and Parlow ( 2007 ) utilized satellite images
from different platforms, a digital elevation model (DEM), a digital surface model
(DSM), and in situ measurements to calculate and model the ground (or storage)
heat flux density in Basel, Germany, with three different approaches. Xu et al.
( 2008 ) exploited hyperspectral imagery collected by the Operative Modular Imaging
Spectrometer (OMIS) as well as topographic and meteorological information to map
the spatial variations of turbulent sensible heat flux in Shanghai, China. Sobrino
et al. ( 2012 ) interpreted LST measurements of the Airborne Hyperspectral Scanner
(AHS) along with in situ data of air temperature to define the minimum spatial
resolution required to properly estimate the surface UHI effect at the district level
of Madrid, Spain.
Most relevant to this study is the research conducted by Heldens ( 2010 )
and Heldens et al. ( 2010 , 2012 ). Their work explored the potential of airborne
hyperspectral data and object height information for urban microclimate modeling
in the City of Munich, Germany. To this end, hyperspectral data collected by
the Hyperspectral Mapper (HyMap) sensor and object heights derived from High
Resolution Stereo Camera (HRSC) imagery were utilized to infer an urban surface
material map. This material map was later employed to drive ENVI-met 4 (beta),
a 3D coupled flow-energy balance model to predict the urban microclimate (Bruse