In the second of a two part article, Drs Sheeba Valsson and Alka Bharat conclude their fascinating study of the variations in intra-urban temperatures in the city of Nagpur, India. What follows is a highly detailed analysis, where elements such as seasonal temperatures and land characteristics are recorded, to arrive at mathematical formulae to explain variations in microclimate.
While further details of the opening stages of this research are available in June’s edition of AWE, readers can familiarise themselves with the nine areas of study on the map and key below.
1. Itwari 2. Jafer Nagar 3. Ravi Nagar 4. Sadar 5. Mahal 6. Wardhaman Nagar 7. Friends’ Colony 8. Ajni (medical college staff quarters) 9. Government quarters – civil lines
Maximum and minimum air temperatures
To understand the behaviour of the intra-urban air temperature variations, the difference between the maximum and minimum air temperatures was analysed in each study area.
The graph in Figure 15 clearly depicts that areas A3, A8 and A9 had the highest maximum and lowest minimum air temperatures in summer.
Figures 16-17 clearly show that areas A3, A8 and A9 had the maximum difference between the highest and lowest values of air temperature in summer and winter.
Correlating air temperature variation
After understanding the behaviour of climatic parameters, the next important stage was to relate the phenomenon observed to the specific land characteristics of the study areas. In the pilot study it was observed that the outdoor air temperature when measured at the centre of the street section was directly influenced by the street section characteristic, hence the height of structure:width of road (H:W) ratio value overrules all the other land characteristics. For further graphical analysis only the H:W ratio and the built up percentage in each of the nine study areas were considered.
From Table 13 it is observed that we can divide the study areas into three distinct groups as per the value of the H:W ratio and the percentage of built up areas: group 1 – A1, A4, A5; group 2 – A2, A6, A7; and group 3 – A3, A8, A9. Group 1 has the highest value of H:W ratio (2.34 – 2.68) and Group 3 has the lowest value of H:W ratio (0.20-0.26).
From the measured temperature data it is inferred that as the value of the H:W ratio increases, the maximum air temperature decreases and the minimum air temperature increases. The old congested areas of A1, A4 and A5 are characterised by narrow streets and high structures. This results in shading of the streets by the adjoining structures, thus the amount of sun penetration into these streets is much less, resulting in lower air temperature at 14:00 hours.
At the same time, it is observed that minimum air temperatures at 05:00 hours are higher in these areas. As the atmospheric temperature drops in the late night and early morning, the heat emitted by the structures and the anthropogenic heat is not allowed to move out of the narrow urban canyons, explaining the cause of a higher minimum temperature as compared to the other study areas. This relationship can be further explained by Figure 18, which shows that the lower the value of the H:W ratio, the greater the difference between maximum and minimum air temperatures.
The relationship between the air temperature and the percentage of the area that is built up is explained through Figure 19. It is observed that as the value of the percentage of built up areas increases there is an increase in the minimum air temperature.
Areas A1, A4 and A5 have a high percentage of built up areas and the value of minimum summer air temperature is also high.
Statistical analysis
After understanding the intra-urban air temperature variation, statistical analysis is conducted to finally correlate the climatic variables with the land characteristics and develop mathematical models for the same.
To study the statistical significance of the average air temperature variations among the nine land use areas during the specified days in summer and winter, the analysis of variance (ANOVA) technique was used and significant variation was proven. After this study, a functional relationship was developed between the climatic parameters and the land characteristics.
Relationship between climatic parameters and land characteristics
Through the use of regression analysis, the following sections will look at the relationships between the average maximum temperature in summer and H:W ratio; the average minimum temperature in summer and H:W ratio; differences (as determined by area) between average maximum and minimum temperature in summer and average H:W ratio; average minimum temperature in summer and percentage of built up areas; and average maximum and minimum temperature of each area and its percentage of built up areas.
Relationship between average maximum temperature in summer and H:W ratio The regression line in Figure 20 depicts that the average maximum temperature in summer decreases as the value of the H:W ratio increases. This means that in the case of areas with a low H:W ratio, the street width is more and the abutting structures are low in height, hence these streets are highly exposed to solar radiation.
The regression statistics (available online) indicate that the proportion of variation in the response data (average maximum temperature) explained by the predictor (H:W) in the regression model is 0.915(R2). This explains high correlation between the two. The regression equation can be written as: Y1 = -2.609X5+48.84 where Y1 is the average maximum temperature (0 C) of summer (measured at 14:00 hours) and X5 is the average ratio of height of structures to width of the road.
Relationship between average minimum temperature in summer and H:W ratio The regression line in Figure 21 shows that the average minimum temperature in summer increases as the value of H:W increases. In the case of areas with a high H:W ratio, the width of the street is less and the abutting structures are of greater height, hence these streets are mutually shaded by the structures and are not exposed to direct solar radiation – explaining the lower temperature.
As the atmospheric temperature drops in the late night and early morning, the heat emitted by the structures and the anthropogenic heat is not allowed to move out of the narrow urban canyons, explaining the cause of higher minimum temperature as compared to the other study areas.
The regression statistics indicate that the proportion of variation in the response data (average maximum temperature) explained by the predictor (H:W) in the regression model is 0.729 (R2). This explains high correlation between the two. The regression equation can be written as: Y2=1.323X5 +32.12 where Y2 is the average minimum temperature (0 C) of summer (measured at 05:00 hours) and X5 is the average ratio of height of structures to width of the road.
Relationship between area dependent differences between average maximum and minimum temperature in summer and average H:W ratio The regression line in Figure 22 depicts that the difference between maximum and minimum temperature in summer decreases as the value of H:W ratio increases.
The regression statistics explain that the proportion of variation in the response data is explained by the predictor. The average ratio of height of structure:width of road in the regression model is 0.889 (R2). This explains high correlation between the two.
The regression coefficient of Y3 on X5 is -3.934 and is found to be significant at the 5% level of significance. The regression equation is the average range of summer temperature, which is 16.725 – 3.934 x average H:W ratio. This explains high correlation between the two.
This can be written as Y3 = 16.725 – 3.934 x X5 where Y3 is the difference in the average maximum and minimum temperature of summer in 0 C and X5 is the average ratio of height of structures to width of the road.
Relationship between average minimum temperature in summer and percentage of built up areas Figure 23 depicts that the average minimum temperature increases as the percentage of built up areas increases. This means low density areas are highly exposed to solar radiation, hence the temperature at 14:00 hours is high as compared to high density areas. At the same time because of high density and low sky view factors, the radiative cooling is low and so the minimum temperature is higher in these areas as compared to the low density areas.
The regression statistics indicate that the proportion of variation in the response data (average minimum temperature) explained by the predictor in the regression model is 0.629(R2). This explains medium correlation between the two. As per Table 26, the regression equation can be written as Y2=0.049X3+31.45 when Y2 is the average minimum temperature (0 C) of summer (05:00 hours) and X3 is the percentage of built up areas.
Relationship between differences in average maximum and minimum temperatures of each area and percentage of built up areas As depicted in Figure 24, in less built up areas there is increase in land area exposed to direct sunlight, meaning the temperature taken at the centre of the road is very high during the afternoon hours. After sunset, heat rays absorbed by different surfaces reradiate back and so by the early morning the temperature is very low, thus explaining the increased difference between the maximum and minimum temperature of this area.
In highly built up areas, the land exposed to direct sunlight is low, hence the temperature taken at the centre of the road is very low during the afternoon hours and the heat absorbed by different surfaces is unable to reradiate back after sunset. This explains the cause of less difference between the maximum and minimum temperature of this area.
The regression statistics indicate that the proportion of variation in the response data (difference between average maximum and average minimum temperature) explained by the predictor (% of built up) in the regression model is 0.822(R2). This explains high correlation between the two.
As per Table 29, the regression equation can be written as Y3=-0.153X3+18.98 where Y3 is the difference in the average maximum and minimum temperature of summer (0 C) and X3 is the built up area (%).
Summary
The aim of this study was to assess the variation in microclimate character, which is attributed to physical properties of urban fabric and an ever increasing urbanisation, and to understand the inter-relationships between these factors. The spatial and temporal variability of temperature and relative humidity and their relationship to specific physical characteristics of land has been demonstrated, by choosing nine representative regions in Nagpur city.
Relationships between the various urban and climatic parameters have been investigated, and mathematical predicting models for estimating the climatic variables were developed.
The following are the noteworthy results of the study:
• The average maximum and minimum summer temperatures at the selected nine areas in Nagpur during summer differ significantly
• The nine areas under study were divided into three homogeneous groups on the basis of their average maximum temperatures during summer, which are found to be homogenous with respect to urban characteristics: open space area, green area, road network, average built up area and average H:W ratio
• The smaller the H:W ratio the greater the difference between the maximum and minimum temperature of that area
• Low density areas are highly exposed to solar radiation, hence the temperature at 14:00 hours is high as compared to high density areas. At the same time, because of high density and low sky view factor the radiative cooling is low, hence the minimum temperature is higher in these areas as compared to the low density areas
• The difference between the maximum and minimum temperatures increases with the decrease in the percentage of built up areas
Published: 05th Sep 2013 in AWE International