GD1: Surface UHI from satellite
Cities as urban clusters: an empirical and large sample study of urban heat island intensity
1Potsdam Institute for Climate Impact Research (PIK), Germany; 2University of Potsdam, Germany
Defining cities as connected urban space from CORINE land cover data and their surroundings as equal-sized boundary, we calculate the urban heat island (UHI) intensity from MODIS land surface temperature data, i.e. the difference between the averages in the city and the boundary. The automatized method allows a systematic study of all urban agglomerations in Europe. We find a cross-sectional and time-variant dependence of the UHI intensity on the size. In addition, plotting the UHI intensity versus the background temperature, we identify a hystersis-like seasonality comprising higher intensities in spring than in fall. We explore the influence of parameters involved in the methodology and discuss the implications.
Directional analyses of UHI intensity over Delhi with respect to variations in vegetation cover in the National Capital Region of India
Jawaharlal Nehru University, New Delhi, India, India
This study examines the variations in day time surface urban heat island (SUHI) over Delhi with respect to the regions situated north, south, east and west of Delhi in terms of the changing vegetation cover dynamics in these regions during the year. LISS III satellite data obtained at different times of the year, reveals that regions surrounding Delhi witness significant variations in vegetation cover during the year. This is primarily due to the fact that agriculture is the predominant land use in the regions surrounding Delhi. In the months of February and March, all the regions surrounding Delhi are covered with vegetation and show NDVI values >0.5. In contrast, it is mainly the region situated north-east of Delhi that has noticeable vegetation cover in the summer months of May and June. NDVI values in this region are higher than those observed for the regions situated south and west of Delhi. Further, day-time surface temperatures obtained from MODIS satellite data have been used to compute SUHI intensity over Delhi with respect to the regions surrounding Delhi. It is seen that SUHI intensity with respect to the regions situated north and east of Delhi (3-5 K), is higher than UHI with respect to the regions situated south and west (1-3K) in the month of March. In contrast, Delhi is found to have greater negative SUHI intensity (-5 to -8 K) with respect to the regions lying south and west as compared to SUHI (-1 to -4 K) with respect to the regions situated north and east in the month of May. Similarly, SUHI intensity of -2 to -5 K is observed with respect to the regions lying south and west as compared to SUHI intensity of 0 to -2 K with respect to the regions situated north and east in the month of November.
The Urban ‘Oasis’: High Resolution Landsat 5TM and ASTER Thermal Imagery Shows the Influence of Water Usage on City-Wide Temperatures in Dubbo, Australia
1School of Earth, Atmosphere and Environment, Monash University, Australia; 2CRC for Water Sensitive Cities, Australia
Vegetation and water availability in a hot-arid urban landscape plays a fundamental role in moderating temperature during periods of extremely hot weather, thereby also mitigating heat-stress related human health impacts. Vegetation provided with an almost unlimited supply of water through irrigation may also promote cooling beyond vegetation boundaries to form an urban ‘oasis’ at the city scale, via facilitated evaporative mechanisms. This study examines increased surface moisture availability, land surface temperature (LST) and landscape oasis effects in and around the extensively-irrigated rural town of Dubbo Australia, a city that is typically exposed to hot-arid summer climate. High-resolution thermal imagery from ASTER and Landsat 5 TM with spatial resolutions of 90 m and 120 m respectively, was used to retrieve LST for hot days (mostly >30°C). Bands in the red and near infrared range of the electromagnetic spectrum were also used to derive Normalized Difference Vegetation Index (NDVI) maps to classify the health and moisture status of various irrigated and non-irrigated vegetated land covers. A strong negative correlation between NDVI and LST was evident across areas of irrigated urban and rural vegetation on relatively hot-dry days, in comparison to non-irrigated rural vegetation. Additionally, city-scale ‘oasis’ effects were evident on 15 of the 20 days examined, whereby the surface cooling was up to 4°C compared to the ‘untreated’ rural landscape. Overall the effects were clearly a result of increased moisture availability, with the cooling disappearing when the entire (urban plus rural) landscape region was moistened by heavy rainfall.
Analysis of the impact of different temporal aggregation techniques of land surface temperature on SUHI indicators and the relationship of surface temperature with population density and night lighting.
Warsaw University of Technology, Poland
Numerous researches in the field of urban climate prove that anthropogenic heat flux (AHF) it is one of the most important components of urban heat island (Sailor & Lu, 2004). AHF is connected with activity of cities' inhabitants, which spatial distribution can be indirectly described by, among others, population density and remotely-sensed night lights (Makar, 2006; Yang, 2014).
Since land surface temperature (LST) is influenced by synoptic conditions, it is widely practiced to use composite datasets for long-term analyses. Surface Urban Heat Island (SUHI - Voogt & Oke, 2003) studies commonly adopt temporally composited remote sensing data, what directly increases the clear sky coverage across urban and rural regions, which are beneficial to SUHI studies. However, most of SUHI studies did not consider the possible errors caused by composite processes (Hu et al., 2013).
Using MODIS (Moderate Resolution Imaging Spectroradiometer) land surface temperature data, we will present analyses of the influence of different temporal aggregation techniques on the urban LST patterns in the city of Warsaw, Poland. The study will be conducted for different seasons, for day and night cases. Also, we will discuss impact of different temporal aggregation techniques on values of several SUHI indicators (reviewed by Schwartz et al., 2011), and relationship between LST, night lights retrieved by DMSP OLI and NPP SUOMI VIIRS satellite instruments and population density.