scanning data based on
f Environment, 104(1),
Systematic designs for
os, 18(3), 283-307.
| asia eagle and hawk.
istruments/specim.asp
D. W. (1998). Criteria
forest management. In
Sustainable forest
older input to advance
ly based indicators.
CIFOR/FAO.
r, M. A., & Teuber, K.
coniferous forest leaf
ar transect protocol.
12, from
+P; ge, accessed 12
irborne and terrestrial
hing, UK, pp. 1-336
ological studies on the
ion in net assimilation
cles and varieties, and
als of Botany, 11(1),
R., Privette, J.L., &
uring fractional cover
ecosystems: Digital
e, and laser altimetry
vironment, 74(1), 45-
m thickness of point
f Botany, 11(2), 178-
[., Coram, J.E., Brady,
(1994). A study of the
psland. Department of
irces, Victoria.
ieving leaf area index
heories, methods and
15.
ENTS
ny PhD. I would like
(DSE), Simon Jones
for their guidance and
Prof. Mark Shortis
(Melbourne Uni) for
o the Vegetation
RO Land and Water,
y Opie, and Glenn
»-kind supporters of
MIT), Mariela Soto-
ndrew Mellor (DSE)
niversity and CRC-SI
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B8, 2012
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia
CLASSIFICATION AND MODELLING OF URBAN MICRO-CLIMATES USING
MULTISENSORAL AND MULTITEMPORAL REMOTE SENSING DATA
B. Bechtel ®*, T. Langkamp*, J. Bóhner “, C. Daneke*, J. OBenbrügge “, S. Schempp*
* KlimaCampus, University of Hamburg, Bundesstraf3e 55, 20146 Hamburg, Germany —
benjamin.bechtel(guni-hamburg.de
Commission VIII/8: Land
KEY WORDS: Urban, Climate, Human Settlement, Classification, Thermal, DEM/DTM, Landsat
ABSTRACT:
Remote sensing has widely been used in urban climatology since it has the advantage of a simultaneous synoptic view of the full
urban surface. Methods include the analysis of surface temperature patterns, spatial (biophysical) indicators for urban heat island
modelling, and flux measurements. Another approach is the automated classification of urban morphologies or structural types.
In this study it was tested, whether Local Climate Zones (a new typology of thermally 'rather' homogenous urban morphologies) can
be automatically classified from multisensor and multitemporal earth observation data. Therefore, a large number of parameters
were derived from different datasets, including multitemporal Landsat data and morphological profiles as well as windowed
multiband signatures from an airborne IFSAR-DHM.
The results for Hamburg, Germany, show that different datasets have high potential for the differentiation of urban morphologies.
Multitemporal thermal data performed very well with up to 96.3 % overall classification accuracy with a neuronal network
classifier. The multispectral data reached 95.1 % and the morphological profiles 83.2 %.The multisensor feature sets reached up to
97.4 % with 100 selected features, but also small multisensoral feature sets reached good results. This shows that microclimatic
meaningful urban structures can be classified from different remote sensing datasets.
Further, the potential of the parameters for spatiotemporal modelling of the mean urban heat island was tested. Therefore, a
comprehensive mobile measurement campaign with GPS loggers and temperature sensors on public buses was conducted in order to
gain in situ data in high spatial and temporal resolution.
1. INTRODUCTION
Urban climatology is an important application for remote
sensing of urban areas. The growing interest in urban climatic
phenomena like the urban heat island (UHI) is motivated by
increasing vulnerability to health risks due to rapid urbanization
in developing countries and climate change.
The urban heat island indicates increased air temperatures in the
urban atmosphere compared to a preurban state (Lowry, 1977)
or (more often) to a rural reference station of identical regional
and topo-climate. It is the most prominent effect of urban
climate (Oke, 1982; Armfield, 2003; Yow, 2007) and has been
studied more intensely than any other effect. The UHI varies
both diurnally and seasonally and depends on the prevailing
synoptic conditions. It is particularly pronounced at low wind
speeds and high air pressure (Oke, 1973) and has been
documented for many towns and cities of different sizes and on
different continents (Oke, 1982). Further, it depends on the size
of the city (Oke, 1973). Within a city the UHI greatly varies
depending on the urban structure and vegetation and is therefore
also referred to as urban heat archipelago.
Remotely sensed data can contribute to descriptive urban
climatic studies and a better understanding of the underlying
climatic processes. The assessment of surface atmosphere
exchanges (Rigo and Parlow, 2007) was fostered by recent
advances in earth observation technologies. Further, remote
sensing has widely been applied to characterise the urban
surface and to determine parameters for model and
experimental studies. Thus, fundamental physical attributes of
* Corresponding author.
measurement sites and key parameters for urban climate models
can be derived from remotely sensed data (Grimmond, 2006).
This includes parameters of the urban energy balance like
albedo and emissivity (Frey et al, 2007; Frey and Parlow,
2009), the urban canyon geometry, and aerodynamic
characteristics like roughness parameters (Bechtel et al., 2011).
Further, automated classification of urban structures in respect
to their microclimatic properties can contribute to urban climate
studies. A typology of such thermally defined urban structures
was recently introduced with the Local Climate Zones (LCZ)
scheme (Stewart and Oke, 2009).
Great attention has been devoted to the application of thermal
infrared data for UHI assessment (Roth et al., 1989; Eliasson,
1992; Gallo and Owen, 1999; Voogt and Oke, 2003; Nichol et
al, 2009; Wong et al, 2009; Fabrizi et al., 2011). Thermal
imagery offers the chance to directly measure the surface
temperature which is crucial for the urban energy balance and
modulates the air temperature of the lowest atmospheric layer.
Therefore, the spatial structure of the surface temperature
patterns is not only directly related to surface characteristics but
also used to study the energy balance and the relation between
atmospheric heat island and surface temperature heat island
(Voogt and Oke, 2003). However, there are various problems
involved in linking thermal imagery to air temperatures which
mostly remain unsolved (Roth et al., 1989; Voogt and Oke,
1997; Voogt and Oke, 2003).
Beside recent advances in the development of surface
parameters for assessing the urban thermal response, "new
methods for estimation of UHI parameters from multi-temporal
