SPATIAL EVALUATION OF URBAN FORM WITH RESPECT TO BUILDING DENSITY IN ÝSTANBUL Burçin Yazgý*, Vedia Dökmeci* *Urban Planning Department, Ýstanbul Technical University, [email protected], [email protected]

ABSTRACT Ýstanbul is the most important socio-economic, cultural and tourist center of the country and being the capital of three empires, it has a very prosperous cultural and social life which reflects in its urban structure. Thus, Ýstanbul has a very rich urban form due to its historical and socio-economic background, technological development and natural ameni- ties. Since urban design has a long tradition of emanating from the past, it can be very stimulating to investigate the urban structure of Ýstanbul in different locations with differ- ent locational, historical and socio-economic background. Thus, the purpose of this paper is to investigate building density with respect to population density, distance to central business district, to the sub-centers and to the sea shores in Ýstanbul. Multiple regression analysis is used as the research method. Building density is taken the dependent variable and the others independent variables of the analysis. According to the results of the analy- sis, building density of the neighborhoods is a function of distance to central business dis- trict and distance to the sea shore. There is no relationship between the building density and distance to sub-centers. Thus, the distribution of density has a wide spectrum within the city depending on the changing life styles, income, history and amenities. The results of the study can be useful for urban planners, urban designers, policy makers, and transportation planners. For further research, it is suggested to investigate the relation- ship between building density and environmental control. Keywords: Urban Form, Multiple Analysis, Building Density, Ýstanbul

INTRODUCTION Throughout the history, Ýstanbul, which has growth under the influence of sever- al different cultures, has very rich urban structure examples and its structure con- tinue to change as the city grows. Structural growth occurs when an aggregate

297 Burçin Yazgý, Vedia Dökmeci

consists of interrelated parts and the process of growth involves changes in the relations between the parts. Boulding (1956) described structural growth as a complicated process but suggested several principles for its analysis. Two of these are important here. The first principle states “at any moment, the form of any object, organism, or organization is a result of its laws of growth up to that moment” (Boulding, 1956). This principle also states: “growth creates form, but form limits growth” (Boulding, 1956). The second principle is called the princi- ple of equal advantage and governs the distribution of “substance” among the parts of the structure (Winsborough, 1962). Theories of the structural form relied on the first one and then the other of these principles. Classical ecological theo- ry emphasized the importance of the interrelationship between growth and form. More modern theory seems to stress the principle of equal advantage. Density is an important factor to affect the form in the city. Much has fairly well demonstrated a considerable synchronic relationship between the degree of con- centration of the city and the availability of transportation and much finds a sig- nificant correlation between the decline in density and the proportion of central city dwellings classified as substandard. So that, the issue density is analysed by different kinds of disciplines in different types such as building density or population density, or with a different point of view density of movement or density of information. The purpose of this paper is to investigate the relationship between the building density and distance to central business district, distance to sub-centers and dis- tance to sea shore, in different urban forms taken from different districts of Ýstan- bul. There are several studies which investigate the spatial analysis of urban form in different countries. For instance, Lee and Ahn (2003), compare the morphologi- cal aspects of the two developments, Radburn and Kentlands, including form, land use, housing, street pattern, accessibility, block plan, and open space. More- over, Tzakou (1995), after giving a brief picture of a Greek island by way of intro- duction, she presents the traditional system as it was preserved until the middle of 1970s and finally she presented her observations on the changes that have taken place in recent years and are continuing to take place. Asami, Kubat and Istek (2001), deal with the characterization of the street networks in the tradition- al Turkish urban form by discriminate functions of several space-syntax-related indices, as well as by image-analysis and graph-theoretical indices. Also White- hand (2001) deals with the physical form of cities by using a historic-geographi- cal approach. Organization of the paper is as follows. The second section explains the background information about Ýstanbul; spatial analysis of urban form from different districts is given in the third section. Final section is devoted a conclu- sion and further research.

298 1st International CIB Endorsed METU Postgraduate Conference Built Environment & Information Technologies, Ankara, 2006

BACKGROUND Since Ýstanbul has a very rich urban structure, it has chosen as a case study in this study, as it is already stated by Dökmeci and Berkoz. Also Dökmeci and Berkoz (1994) stated that “Ýstanbul was a vigorous, core-dominated metropolis until well into the 1950s, with a very limited suburban development in the periphery. With this expanded use of the automobile and the construction of the bridges over the Bosphorus, however, the suburbs, in typical fashion, were pushed further out. At the same time, peripheral growth accelerated as squatter settlements increased due to rural migration”. Various explanations for multi-nucleation in Ýstanbul can be made: 1. Firms established themselves in the first ring, in search of lower land and/or transport costs. These firms tend to have international and national relationships, and their ties to the old centre weakened. 2. After World War II, construction of national highways with Marshall Aid gradually created a new centre in counter balance to the older centre, whose transport links were completely dependent on the city’s port. This new centre developed at the intersection of the radial and peripheral high- ways, which has easy access to both airport and to national highways. 3. As one of the results of the economic restructuring of Ýstanbul after the 1970s, firms required larger spaces. New firms not find sufficient room to locate in the old centre since land parcels were too small and the height of buildings was restricted due to density regulations in the conservation area of the old central business district. The new centre provided large parcels for the modern large office building and parking lots. 4. After the 1980s, as a result of increased international relationships, demand for office space for new office buildings increased and the new centre continued to develop along the highways.” (Dökmeci & Berkoz, 1994) Aru (1997) describes similar features in Traditional Turkish cities: “The old Turk- ish city, which has free and organic fabric, has a rhythmic geometry. The basic forms of these cities have evolved out of the needs their inhabitants. New developments areas which resulted in the expansion of the city, the rapid changes in the population, the multi-central urban structure have affected the dis- tribution of the densities in different residential areas. Lynch (1981) considers five basic dimensions and two meta-criteria for the per- formance dimensions of the spatial form of a city. These are; how settlement form affects vitality, how settlement form affects human sense, the degree to which the

299 Burçin Yazgý, Vedia Dökmeci

settlement form fits the requirements of people, how able people are to access activities, services etc, and how much control people have over services/ activi- ties/ spaces etc. The two meta-criteria are efficiency and justice. The other subject that should be investigated is the elements of the urban form. Jones and MacDonald state that there are four main elements that make up urban form. These are land use, density, position/transport infrastructure and character- istics of the built environment. And additional element is layout. According to Lynch, there is a concept of place legibility, which is essentially the ease with which people understand the layout of a place.

METHODOLOGY This study aimed to investigate the urban structure of Ýstanbul by comparing building densities in different districts with different historical backgrounds and different development strategies and also to state the significant relations between the building density and the other variables. The method of the analysis is the multiple regression analysis. This method is chosen because it gives opportunity to compare and to find the relations between different kinds of variables. For comparability purposes, eight different districts of Ýstanbul are chosen. One of these areas is in Kadýköy (Figure 1) which is in the Asian part of Ýstanbul and has a label as a center of the Asian side developed under the European influence. As it is a district developed under the European influence in the Asian part and a district that has the central identities it is chosen as one of the study areas. The second area is in Uskudar (Figure 2) which was developed under Islamic charac- teristics and always has the locational importance. Different than Kadýköy, being the district developed under Islamic characteristics make the district to be chosen for the study. The third area is in a historical Bosphorus district called (Figure 3) that is expanding with the squatter settlements. Beykoz being a differ- ent and interesting side within the historical background and new squatters was the main aim of this area to be chosen as another district. Sarýyer (Figure 4) which is another historical Bosphorus district in the European side is chosen as a fourth area to study. Being in the European side as one of the oldest districts, cause Sarýyer to be an interesting district to be evaluated. The fifth area, Beyoðlu (Fig- ure 5), is an important historical settlement area which is developed under the European influence. With Tower, this area gains another kind of impor- tance and becomes the centre for different kinds of activities. Having the first development movements and being the long term center for different cultures caused Beyoðlu to have an important role in the metropolitan structure of Ýstan- bul was the reason to be chosen. The sixth area is from the historical peninsula,

300 1st International CIB Endorsed METU Postgraduate Conference Built Environment & Information Technologies, Ankara, 2006

Fatih (Figure 6) which is a district settled as an expansion of the earliest settle- ment area Eminönü. As it is an extension of the oldest central business district of Ýstanbul, it was important to evaluate the spatial analysis of urban form in that is why it is chosen as the sixth study area. The seventh area is in Beþiktaþ (Figure 7) district that is located near to an important transportation intersection. Different than the other study areas its locational importance for transportation made this district to be chosen. The last area is Eminönü (Figure 8), in historical peninsula, is the earliest settlement area of Ýstanbul and this district contains Greek, Roman, Byzantine and Ottoman urban structures. Being the earliest set- tlement area and also the oldest central business district in Ýstanbul, Eminönü has to be analysed by its urban form. Within all these districts the spatial urban struc- ture was investigated by the use of multiple regression analysis.

EVALUATION The method, multiple regression analysis, has been applied to the areas that are described above. According to the analysis, the building density ratios differ from 0.27 which belongs to Beþiktaþ and 0.58 which belongs to Eminönü. As Eminönü is the earliest settlement area it is expected to have the maximum building densi- ty ratio. The other building density ratios are; Beykoz 0.32, Üsküdar 0.42, Sarýy- er 0.48, Kadýköy 0.54, Fatih 0.55, and Beyoðlu 0.57. These changing building density ratios can be explained according to their historical backgrounds. As Bey- oðlu and Eminönü are the first settlement areas of Ýstanbul metropolitan region, they have the maximum building density ratios. The relations between the building density and distance to CBD, to seashore and to the sub-centers are investigated by the use of multiple regression analysis. The results are given below. While the relations between the building density and distance to central business district and to seashore are significant, distance to sub-center is not significant. As a result of this analysis, building density decreases if the distance to the CBD increases and also it is the same for the distance to seashore that if it is increases again the building density decreases. As it was an important issue to be near to the sea and to the centre in the past and also nowadays, the increasing building density comes with the decreasing distance to the centre and to the seashore. It is an expected result that both central areas and seashore areas traditionally densely built areas and they continue to have some situation. In the future, high density development of sub-centers and housing completes might change this trend. This will effect land value development as well as new urban structures in the periphery.

301 Burçin Yazgý, Vedia Dökmeci

Figure 1. Kadýköy District Figure 2. Üsküdar District

Figure 3. Beykoz District Figure 4. Sarýyer District

Figure 5. Beyoðlu District Figure 6. Fatih District

302 1st International CIB Endorsed METU Postgraduate Conference Built Environment & Information Technologies, Ankara, 2006

Figure 7. Beþiktaþ District Figure 8. Eminönü District

Table 1. Result of The Regression Analysis

CONCLUSION In this study the multiple regression analysis is used as a method for different locational districts of Ýstanbul. With different locational and historical properties eight different areas are chosen within the Ýstanbul metropolitan region to evalu- ate the spatial analysis of their urban forms by taking the building density the dependent variable, and the distance to seashore, distance to central business dis- trict and distance to sub-centers independent variables. The results of the evaluation of urban forms in eight different areas by investigat- ing building density illustrate the major difference between the old (Eminönü, Beyoðlu, Fatih, Kadýköy Districts) and modern (Beþiktaþ, Beykoz Districts) urban structures. Studying eight different urban structures of Ýstanbul, and analyzing the different urban forms, it is obvious that, there is a difference in the building density levels of the settlements. As Eminönü and Beyoðlu being the earliest settlement areas they have the highest building density ratios.

303 Burçin Yazgý, Vedia Dökmeci

Within the changing building density ratios, it is investigated that building densi- ty increases while the distance to seashore and distance to central business dis- trict decreases. But the relation between the distance to sub-center and building density is not significant because different districts have different kinds of sub- centers both in function and in form. Besides, the urban pattern: the type of roads, size and form of urban blocks dif- fer a lot. Also the percentage of open areas varies from one district to another. Although the central districts and sea shores have higher building density as a result of traditional urban development than the periphery, this situation might change in the future due to recent high density urban developments in the periph- ery. As a result of this, the living conditions of the districts change. As a conclusion, the urban form of Ýstanbul differs within the different time peri- ods and according to the transformation, related to different topics caused differ- ent urban characteristics. The result of the study can be useful for the urban designers and planners in order to adjust their environment. For further research, it is suggested to increase the number of case studies and to compare the results with other countries with different socio-economic back- grounds.

REFERENCES ARU, K.A. (1997), Conceptions of The General Characteristics of Ottoman Turkish Cities, Prime Ministry Housing Development, Administration, Republic of Turkey. ASAMI Y. et al, (2001), Characterization of The Street Networks in The Traditional Turkish Urban Form, Environment and Planning B: Planning and Design, (vol. 28). BOULDING, K.E. (1956), Toward a General Theory of Growth, chapter in Population Theo- ry and Polic, Spengler, J.J., Duncan, O.D. (eds.), The Free Pres, Glencoe, pp.109-124. DÖKMECI, V. and BERKÖZ L. (1994), Transformation of Ýstanbul from A Monocentric to A Polycentric City, European Planning Studies, (vol. 2). JONES, C., MACDONALD, C. (2004), Sustainable Urban Form and Real Estate Markets, Heriot-Watt University, Edinburgh. LEE C. and AHN K. (2003), Is Kentlands Better than Radburn?, American Planning Associa- tion, (vol. 69). LYNCH K. (1981), A Theory of Good City Form, MIT Press, pp.514 TZAKOU, A. (1994), Sifnos- The Evalution of A Traditional Unity, Ekistics, 368. WHITEHAND, J.W.R. (2001), The Physical Form of Cities: A Historico-Geographical Approach, Sage Publication Ltd., London. WINSBOROUGH, H.H. (1962), City Growth and City Structure, Journal of Regional Science, (4, 2). 304