6 Heat Vulnerability Index Development and Application in Medan City, Indonesia
97
6.13 Spatial Data Processing
To produce a spatial risk assessment index and map, the data set was specially
prepared in the ArcMap 10.3, through unifying the projection coordination system
into Universe Transfer Mercator (UTM) 47 North of World Geodetic System
(WGS) 1984, and converting vector/polygon data into raster data, maintaining 30 m
spatial resolution of the vector/polygon data. The Heat Vulnerability Index has been
computed and projected spatially indicating areas of different levels of vulnerability.
6.14 Results
6.14.1 Heat Stress Level in Medan City
In 1982, 1997, and 2014 the spatial distribution of UTCI has drastically increased,
with maximum 4
° C rise between the year 1982 and 2014. In 2014, there were two
levels of UTCI stress level in Medan City, namely strong heat stress (32–37 °C) and
very strong heat stress (38–46 °C) as presented in Fig. 6.4. The increase is highly
related to urban area sprawl. The frequency of days with UTCI at very strong heat
stress level has been rising especially in the urban area, whereas the rural areas the
frequency has been decreasing (Fig. 6.5).
6.15 Land Surface Temperature and Urban Heat Island
in Medan City
The Land Surface Temperature in Medan City varied from 21 to 35 °C at the pixel
level (Fig. 6.6). In general, higher LST was located in the central part of Medan
City and generally lower in the northern and southern part. The distribution of the
high LST was related to building cover (Fig. 6.6), such as northern part of Medan
with dense building coverage indicated high LST, and areas of lower LST were
concentrated in higher vegetation areas (i.e., vegetation, river, ponds) (Fig. 6.6).
6.16 Heat Vulnerability Map
Principal Component Analysis was described in Table 6.2. These explained 77.5%
of the variance. The percentage variance value of each PCA was used as the weights
in the calculation of HVI value. Consideration of the component loading matrix indicates that component 1 related all variables except transportation emission. Furthermore, component 2 only related to transportation emission. The negative sign of
97
6.13 Spatial Data Processing
To produce a spatial risk assessment index and map, the data set was specially
prepared in the ArcMap 10.3, through unifying the projection coordination system
into Universe Transfer Mercator (UTM) 47 North of World Geodetic System
(WGS) 1984, and converting vector/polygon data into raster data, maintaining 30 m
spatial resolution of the vector/polygon data. The Heat Vulnerability Index has been
computed and projected spatially indicating areas of different levels of vulnerability.
6.14 Results
6.14.1 Heat Stress Level in Medan City
In 1982, 1997, and 2014 the spatial distribution of UTCI has drastically increased,
with maximum 4
° C rise between the year 1982 and 2014. In 2014, there were two
levels of UTCI stress level in Medan City, namely strong heat stress (32–37 °C) and
very strong heat stress (38–46 °C) as presented in Fig. 6.4. The increase is highly
related to urban area sprawl. The frequency of days with UTCI at very strong heat
stress level has been rising especially in the urban area, whereas the rural areas the
frequency has been decreasing (Fig. 6.5).
6.15 Land Surface Temperature and Urban Heat Island
in Medan City
The Land Surface Temperature in Medan City varied from 21 to 35 °C at the pixel
level (Fig. 6.6). In general, higher LST was located in the central part of Medan
City and generally lower in the northern and southern part. The distribution of the
high LST was related to building cover (Fig. 6.6), such as northern part of Medan
with dense building coverage indicated high LST, and areas of lower LST were
concentrated in higher vegetation areas (i.e., vegetation, river, ponds) (Fig. 6.6).
6.16 Heat Vulnerability Map
Principal Component Analysis was described in Table 6.2. These explained 77.5%
of the variance. The percentage variance value of each PCA was used as the weights
in the calculation of HVI value. Consideration of the component loading matrix indicates that component 1 related all variables except transportation emission. Furthermore, component 2 only related to transportation emission. The negative sign of