MODELLING OF THE MAXIMUM URBAN HEAT ISLAND ICB-ICUC'99, Sydney, 8-12 November 1999 J. Unger 1, Z. Sümeghy 1, Á. Gulyás 1, Z. Bottyán 2 and L. Mucsi 3 1. Department of Climatology and Landscape Ecology, University of Szeged, P.O.Box 653, 6701 Szeged, Hungary 2. Department of Natural Sciences, Zrínyi M. University of Defence, P.O.Box 1, 5008 Szolnok, Hungary 3. Department of Physical Geography, University of Szeged, P.O.Box 653, 6701 Szeged, Hungary ABSTRACT This study examines the influence of urban and meteorological factors on the surface air temperature field of the medium-sized city of Szeged, Hungary, on the basis of stationary and mobile measurements under different weather conditions between March and June 1999. This city of about 178,000 is situated in a low, flat flood plain. Our efforts have been concentrated on the investigation of maximum development of the urban heat island (UHI). Tasks include the determination of spatial distribution of mean max. UHI intensity and modelling of existing conditions in the measurement period. Multiple correlation and regression analyses are used to examine the effects of invariable parameters (land-use characteristics, distance from the city centre determined in a grid network) and by variable parameters (wind speed, temperature) on thermal conditions in the study area. The results indicate isotherms increasing in very regular concentric shapes from the suburbs toward the inner urban areas where the mean max. UHI intensity reaches 3.5 C. Strong relationship exists between urban thermal excess and land-use features. In addition, meteorological conditions determine to a great extent the magnitude of the UHI intensity at the time of its maximum development. INTRODUCTION In the field of climatology the investigation of urban heat island (UHI) is one of the most intensively studied environmental modifications caused by urbanisation. The simulation of real factors and physical processes generating this phenomenon is extremely difficult and demands complex and expensive instrumentation, as well as sophisticated numerical and physical models. Despite these difficulties, several models are available for studying small-scale climatic variations within the city, including those based on energy balance (1, 2), radiation (3), heat storage (4) and water balance (5) approaches. One of the less studied characteristics of UHI is the factors that lead to its peak development during the day. As a surmount and attainable solution of the problems mentioned above the utilization of statistical models may provide useful quantitative information on spatial distribution of maximum UHI intensity employing variable (meterological) and invariable (land-use) parameters (6, 7). The main purpose of this study is to investigate the effects and interactions inside the city on the surface air temperature under all weather conditions except rain at the time just a few hours after sunset when the UHI effect is most pronounced. Szeged is located in the southeastern part of Hungary (46 N, 20 E) at 79 m above sea level. The terrain of the city and its surrounding area is a flat flood plain. The settlement is situated along the Tisza River but there are no large water bodies nearby. This environmental situation makes Szeged a good case for the study of a relatively undisturbed
urban climate. The city's population of about 178,000 lives within a larger administration district of 356 km 2, but urban and suburban areas totally occupies only approx. 25-30 km 2. As for the city structure, a number of different land-use types are present including a densely built centre with medium wide streets and large housing estates of tall concrete buildings set in wide green spaces. Szeged also contains zones used for industry and warehousing, areas occupied by detached houses, considerable open spaces along the banks of the river, in parks, and around the city's outskirts (Fig. 1.) a b c d e f g h Tisza River 0 1 2 3 km Figure 1: Main land-use types and road network in Szeged (a: road, b: circle dike, c: border of the study area, d: agricultural and open land, e: industrial, f: 1-2 storey detached houses, g: 5-11 storey apartment buildings, h: historical city core with 3-5 storey buildings) The study area belongs to the climatic region Cf by Köppen's classification (8), which means a temperate warm climate with a rather uniform annual distribution of precipitation (Table 1). The regional climate of Szeged has, however, a certain Mediterranean influence seen mainly on the annual variation of precipitation (9). Table 1: Annual means of meteorological parameters for the region of Szeged (1961-1990) temp. temp. range Jan. temp. July temp. precip. rel. hum. vap. press. suns. dur. wind speed 10.4 C 22.6 C -1.8 C 20.8 C 497 mm 71% 10.1 hpa 2102 h 3.4 ms -1 METHODS The study area is divided into two sectors and subdivided further into 0.5x0.5 km square grids (Fig. 2). This same grid size was employed in a human bioclimatological analysis of Freiburg, Germany, a city of similar size to Szeged (10). One hundred seven grid cells totalling 26.75 km 2 cover the urban and suburban parts of Szeged (mainly inside of the circle dike that protects the city from floods). The outlying parts of the city characterised by village features are not included in the grid. The grid was established by quartering the 1x1 km square network of the Unified National Mapping System (UNMS) that is printed on topographical maps in Hungary. The examination of the max. UHI intensity was based on mobile and stationary observations during the period of March-June 1999. Although these measurements go on continuously, the results of the study focus on this one three- 2
month period. In order to collect data on surface air temperature at every grid cell, mobile measurements were performed on fixed return routes once a week during the study period (altogether 13 times) to accomplish an analysis of air temperature over the entire area. This one-week frequency of car traverses secured information on different weather conditions, except during rain. The division of the study area into two sectors is necessary because of the large number of grids. The Northern and Southern sectors consist of 59 grids (14.75 km 2 ) and 60 grids (15 km 2 ), respectively, with an overlap of twelve grid cells (3 km 2 ). The lengths of the return routes are 68 and 75 km in the Northern and Southern sectors and take about 3 hours to traverse (Fig. 2). Such long and return routes were necessary to gather temperatures in every grid cell and to make time-based corrections. Temperature readings were obtained using a radiation-shielded temperature sensor (resolution of 0.01 C) which was connected with a portable data logger for digital sampling inside the car. The data were collected in every 16 s so that, at an average car speed of 30 kmh -1, the average distance between measuring points was 133 m. The temperature sensor was mounted 0.60 m in front of the car at 1.45 m above ground to avoid engine and exhaust heat. The car speed used was sufficient to secure adequate ventilation for the sensor. R C a b c d e 0 1 2 km Figure 2: Division of the study area into 0.5x0.5 km grids with the locations of rural (R) and central (C) grids, Northern (a), Southern (b) sectors, the area of overlap (c), as well as the measurements routes by sectors (d, e). The fixed measurement site at the University of Szeged is indicated as. After the averaging the measurement values by grid cells, time adjustments to the reference time were applied assuming linear air temperature change with time. This linear change was monitored using the continuous records of the automatic weather station at the University of Szeged. The linear adjusment appears to be correct for data collected a few hours after sunset in urban areas, but only approximately correct for suburban and rural areas because of the different time variations of cooling rates (11). The reference time, namely the likely time of the occurrence of the strongest UHI, was 4 hours after sunset, a value based on earlier measurements. Consequently, grid cells situated in one or another sector can be characterised by one temperature value for every measuring night. The temperature values refer to the centre of each square. The first aim of the investigaton was to construct a horizontal isotherm map to show the average situation of spatial distribution of maximum UHI intensity in the study period. We determined air temperature differences (UHI 3
intensity) by grids referring to the temperature value in that grid cell where the synoptic weather station of the Hungarian Meteorological Service is located. The records of this station were used as rural data in earlier studies on urban climate of Szeged (e.g. 8, 12), so the grid containing this station was regarded as rural (R) (Fig. 2). One hundred seven points (grid cell centerpoints)) covering the urban parts of Szeged secure an appropriate basis to interpolate isolines, which, therefore can show detailed descriptions of thermal field within the city at the time of the strongest effects of urban factors. The second aim was to determine quantitative influences of anthropogenic and natural factors on the UHI intensity. In order to assess the extent of the relationships between temperature differences and other various factors, multiple correlation and regression analyses was used. Variable parameters were the average wind speed and air temperature in the mobile measuring periods recorded by the weather station at the University. Percentage of built-up area, water surface by grid cells, and distance to the city centre (C) were the invariable parameters. We consider this distance as an indicator of the location of a cell within the city. The selection of these parameters is based on their role in determining small-scale climate variations (13) and by the limitations of data available to the present study. The parameters of land-use for the grid cells were determined by GIS methods combined with remote sensing analysis of SPOT XS image (14). Vector and raster-based GIS database were compiled at the University of Szeged in the Applied Geoinformatics Laboratory. The digital satellite image was rectified to the UNMS using 1:10,000 scale maps. The nearest-neighbour method of resampling was employed, resulting in a RMS value of less than 1 pixel. Because the geometric resolution of the image was 20x20 m, small urban units could be assessed independently of their official land-use classification. Normalised Vegetation Index (NDVI) was calculated from the pixel values, according to the following equation: NDVI = (IR-R)/(IR+R) where IR is the pixel value in the infrared band and R is the pixel value in the red band. The NDVI values are ranging from -1 to +1 indicating the effect of green space in the given spatial unit (15). Built-up, water, vegetated and other surfaces were distinguished according to the NDVI value. The spatial distribution of these land-use categories inside each grid element was calculated using crosstabulation. Fig. 3 displays the ratio of the built-up area to the total area of the grid cells in 25% increments. For example, the location of the Tisza River (low built-up ratio) is clearly recognised at a first glance with its east-to-south curve in the southeastern part of the study area (see also Fig. 1). RESULTS AND DISCUSSION It can be seen on the Fig. 3 that the urban effect is present in the spatial pattern of the mean max. UHI intensity (at 4h after sunset). The isotherms show almost regular concentric shapes with values increasing from the suburbs toward the inner urban areas. A deviation from this concentric shape occurs in the northeastern part of the city, where the isotherm of 2 C stretches toward the outskirts. This can be explained by the influence of the large housing estates with tall concrete buildings located mainly in the northeastern part of the city with built-up ratio higher than 75%. The spreading out of the isolines of 2.8 C and 2.4 C to northwestern, the isoline of 2 C to southwestern directions from the centre are also caused by the high built-up ratio of more than 75%. A mean max. UHI intensity of higher than 2 C indicates significant thermal modification and, in Szeged, the extension of the area characterised by this temperature differential is relatively large compared to the size of the study area. It covers about 56 grids (14 km 2 ), about 53% of the total area. 4
As was expected, the highest differences (more than 2.8 C) concentrated mainly in the densely built-up city centre (>75%) covered by about 11 grid cells (2.5-3.0 km 2 ). The highest difference (3.5 C) occurs along the southern edge of the central grid cell (C). a b c d 3.2 2.8 2.4 2.0 1.6 0 1 2 km Figure 3: Built-up characteristics of the study area (ratio of the built-up area to the total area: a - 0-25%, b - 25-50%, c - 50-75%, d - 75-100%) and spatial distribution of the mean max. UHI intensity ( C) between March and June 1999 in Szeged Table 2: (a) Bivariate correlation coefficients between max. UHI intensity ( T) and other parameters, as well as (b) model equation of the max. UHI intensity in Szeged Value of correl. coefficient (r) Standard errorof-estimate ( C) Sig. level n > 700 (a) Bivariate correlation coefficient (r) r T,D 0.526 0.966 0.001 r T,B 0.396 1.043 0.001 r T,T 0.331 1.072 0.001 r T,U 0.091 1.131 0.01 r T,W 0.070 1.133 0.05 (b) Multiple linear regression equation T = -0.547D + 0.007B + 0.089T - 0.104U + 0.009W + 1.252 0.632 0.883 0.001 For the computation of the model equation the variables are: distance from the central grid in km (D), ratio of built-up surface in % (B), ratio of water surface in % (W), mean air temperature in C (T) and mean wind speed in ms -1 (U). The total number of data (one T value in each grid cell and on each measurement night) is 768. Table 2 contains the results of the bivariate correlation analysis on the max. UHI intensity ( T) against the independent parameters considered in this study. As the Table 2 shows, among the examined parameters D has the largest and W has the smallest correlation coefficient. The first three coefficients are significant even at the 0.001, the U at the 0.01 and the W at the 0.05 significance levels. The multiple linear regression equation indicated the relative influence of land-use on the UHI intensity. The correlation coefficient of this model equation is 0.632 with a significance level of 0.001, which means, that this type of modelling of T based on the existing conditions in the measurement period is an appropriate process. As the coefficients of parameters show, the short distance from the city centre and the high built-up ratio, 5
which are prevealing in the inner parts of the city, play important roles in increasing of the urban temperature. On the other hand, the max. UHI intensity also strongly depends on mean air temperature (T). It is refer to the possibility, that the max. UHI intensity has a seasonal fluctuation, however, its verification needs longer-term data sets. These preliminary results show that the statistical approach to determine the behaviour of the UHI intensity in Szeged is promising and this fact urge us to make more detailed investigations. We plan to extend this project by modelling urban thermal patterns as they are affected by variable (meteorological) conditions with a time lag. We intend to employ the same parameters used in this study as well as additional parameters to model the magnitude and spatial distribution of the maximum UHI intensity on days characterised by any kind of weather condition (apart from precipitation) at any time of the year without recourse to extra mobile measurements. These tasks require longer-term data sets, so we intend to gather data for a period of more than one year. The results will be of practical use in predicting the pattern of energy consumption inside the city. They can be used forecast and plan for energy demand, particularly in the cold and warm periods of the year when energy consumption for heating and cooling, respectively, is highest. ACKNOWLEDGEMENTS This research and the presentation of the results were supported by the Hungarian Scientific Research Fund (OTKA T/023042), the Fundation for Szeged (Szegedért A.) and the WMO Support Fund. The authors wish to give special thanks to Mr. L. Csikász and the students who took part in the measurement campaigns and in the data pre-processing. REFERENCES 1. Johnson, G.T. et al. 1991. Simulation of surface urban heat islands under 'ideal' conditions at night, I: Theory and tests against field data. Boundary Layer Met. 56:275-294. 2. Myrup, L.O., McGinn, C.E. and Flocchini, R.G. 1993. An analysis of microclimatic variation in a suburban environment. Atm. Environment 27B:129-156. 3. Voogt, J.A. and Oke, T.R. 1991. Validation of an urban canyon radiation model for nocturnal long-wave radiative fluxes. Boundary Layer Met. 54:347-361. 4. Grimmond, C.S.B., Cleugh, H.A. and Oke, T.R. 1991. An objective urban heat storage model and its comparison with other schemes. Atm. Environment 25B:311-326. 5. Grimmond, C.S.B. and Oke, T.R. 1991. An evapotranspiration-interception model for urban areas. Water Resources Res. 27:1739-1755. 6. Park, H-S. 1986. Features of the heat island in Seoul and its surrounding cities. Atm. Environment 20:1859-1866. 7. Kuttler, W., Barlag, A-B. and Roßmann, F. 1996. Study of the thermal structure of a town in a narrow valley. Atm. Environment 30:365-378. 8. Unger, J. 1997. Some features of the development of an urban heat island. Studia Univ. Babes-Bolyai, Geographia 42(1-2):125-131. 9. Koppány, G. and Unger, J. 1992. Mediterranean climatic character in the annual march of precipitation. Acta Climatol. Univ. Szeg. 24-26:59-71. 10. Jendritzky, G. and Nübler, W. 1981. A model analysing the urban thermal environment in physiologically significant terms. Arch. Met. Ge. Bi. Ser.B. 29:313-326. 11. Oke, T.R. and Maxwell, G.B. 1975. Urban heat island dinamics in Montreal and Vancouver. Atm. Environment 9:191-200. 12. Unger, J. 1996. Heat island intensity with different meteorological conditions in a medium-sized town: Szeged, Hungary. Theor. Appl. Climatology 54:147-151. 13. Adebayo, Y.R. 1987. Land-use approach to the spatial analysis of the urban 'heat island' in Ibadan. Weather 42:272-280. 14. Mucsi, L. 1996. Urban land use investigation with GIS and RS methods. Acta Geogr. Univ. Szeg. 25:111-119. 15. Lillesand, T.M. and Kiefer, R.W. 1987. Remote sensing and image interpretation. (New York, J. Wiley & Sons). 6
7