Digital Urban: Comparison of four blocks

My idea for the project for Digital Urban -course was to continue with the theme I started in the the Urban Challenge Studio 1 -course: comparing different building efficiencies in a certain land area. In Esri’s City Engine, I was modifying the plot and building rules such as plot division, building height and setbacks to force the software produce four different type of block structures resulting in a similar efficiency (and floor area m2’s).

I ended up in two different type of closed blocks and two different type of blocks with separate buildings. Blocks with different structures differs in terms of building height, footprint area and how uniform or fragmented the footprint of the built area is. Four models and their key figures are presented in figure 1:

Figure 1. Four different block structures created with City Engine. (* Something unexpected happened to the parcel area of the block C in City Engine, since the size in square meters is smaller than in other model. This adds inaccuracy to the comparisons, but the dot-type housing is still valid example of a common type of Finnish suburban block structure.)

First two models A and B are presenting exactly the same shape of a closed block, but the street setback setting is varied. Street setback is the distance from building footprint to the street. It is often left unbuilt to serve for e.g. maintenance purposes such as snow removal, but most of the time it is unused – which raises the question do we really need them around every building? As the comparison of the key figures of A and B point out, the effect of the 3-meter setback is significant on the building efficiency (1.8 vs. 2.1). Same block without the street setback produces 1440 m2 (+13%) more total floor area, which equals for example to 28 apartments in size of 50 m2. The footprint of the block B is bigger than in A, but since it is utilizing the buffer zone outside the building, it enables also a bigger inner yard area for the inhabitants.

The surprising large effect of the buffer zone is due to the geometric principle, that size of an area isn’t growing linearly with the distance of its radius, but exponentially with the square of that distance. If the floor area of a single block varies this much depending on the setback rule, it is clear how large effect it can have in a neighborhood or city level. The importance of understanding the geometry and the effects of buffers for the efficient use of land area have been discussed e.g. by Unwin (1912) and Vaattovaara & Joutsiniemi (2016). One of the benefits of rule-based approach and softwares is the ability to apply this type of rules, and instantly evaluate the impacts to the whole model whether it is a single block, neighborhood or a complete city.

Despite the problem of parcel size in model C, key figures of the two blocks with separate buildings (C & D) points out, that to reach the same building efficiency than the closed blocks, the number of floors has to be higher. The unbuilt areas of a parcel in dot-type building block C are fragmented into several yards between the houses. In the lamel-type building block D two yards are connected and are forming a wider park-type of yard between the buildings. In comparison to the closed blocks, separate buildings and yards might provide easier administration for the housing condominiums, since they have their own buildings and yards to maintain.

Light conditions

In addition I did experiment with the City Engines sunlight settings. In figure 2 is presented the sunlight conditions of the four blocks in winter- and summertime:

Figure 2. Sunlight conditions of the four blocks in winter- and summertime.

The blocks are pointing towards north-west, since the geographic location in the City Engine model is along Vihdintie in Haaga. Left part presents the sunlight in the end of October at 12:00 and 15:00. Right part presents the sunlight in the middle of June at 12:00 and 18:00. The wintertime afternoon light conditions were captured earlier time of day because at 18:00 in October the sun is already gone down, and it would have demonstrated mainly the darkness without the shades.

Comparison of the closed blocks and separate building blocks reveal, that most of the separate buildings are getting more light in the wintertime than the buildings in the closed blocks. I didn’t pay attention towards placement and angles of buildings in order to optimize the sunlight, but as the dark wintertime pictures show, very little light is entering the inner yards. To enable sunlight to get into the buildings even a little in winter, it is important to take aspects of placement angles and building heights into account. Especially the closed blocks are difficult in terms of sunlight, and for example an entrance into the inner yard could be positioned in the southern end to increase the amount of light getting in to the block.



Unwin, R 1912. Nothing gained by overcrowding!, P.S. Jung & Son, Westminister.

Vaattovaara, M and Joutsiniemi, A 2016. Vääristynyttä tiiviyttä uusia tutkimuskysymyksiä. Terra 128: 1/2016.


Digital urban: effect of the street setback -rule

My idea is to continue with Haaga where I left in the Urban Challenge Studio 1: Vihdintie boulevard. What I did in the studio was calculating possible housing amounts with different settings of housing stock in Excel spreadsheets. I don’t know if it makes any sense to try to model the housing stock (in the building mass) with CityEngine, but at least what I have managed to figure out by now how to tweak different rules in order to achieve different amounts of floor area. Floor areas together with the parcel land areas can be set on the CityEngines dashboard to show the building efficiency (the right panel):

Picture 1. Two scenarios of the same parcel with different street setback -rules.

In picture 1 is shown the same parcel in two different scenarios with otherwise similar housing block, but the rule of “street setback” (buffer zone around the block) is varied. In scenario 1 setback is 0 meters and in scenario 2 setback is 3 meters, which obviously is reducing the volume of the building mass. The 3-meter setback is causing a reduction of 1439 m2 of floor area, which decreases the building efficiency from 2.24 to 2.

In table 1 is presented the calculated number of apartments and inhabitants for the two scenarios based on housing stock assumptions from my studio 1 work:

Table 1. The calculated numbers of apartments and inhabitants in the two scenarios.

To conclude, the difference of two scenarios with the described housing stock assumptions are in the total number of apartments 29 and in the number of inhabitants 45.

I am not sure how will I future develop this study project of Urban Digital course, but this is what I have done and learnt by now. I am still having great difficulties with the user interface and the logic of CityEngine, especially the topography is causing a headache when created streets, blocks and buildings are diving underground…

GIS-practical 6: Accessibility

In this exercise, I was analysing the accessibility of Tikkurila train station (Mika did the Jumbo part), which is a significant hub for passenger train traffic. To evaluate the accessibility, Helsinki Region Travel Time Matrix 2018 developed at the University of Helsinki was used. The matrix contains travel time and distance information between all the 250 m x 250 m grid cells in the Helsinki Capital Region by walking, cycling, public transportation and private car.

Map 1. Accessibility of Tikkurila railway-station by walking.

Tikkurila railway-station is accessible in under 30 minutes by walking from about 1800 meter radius (map 1). In the north-east corner of the radius of walking accessibility drops suddenly, which is explained by the golf course in Hiekkaharju that creates a barrier for pedestrians in the city structure. Compared to the car or public transport accessibility where the infrastructure conditions are in the key role (e.g. speed limits, placements of main roads and railways), the radiuses of travel time classes for walking are almost symmetrical because they are mostly determined by the distance of walking in a situation where walkable street network is dense.

Map2. Rush-hour accessibility of Tikkurila railway-station by public transportation.

Accessibility of Tikkurila railway-station is, not that surprising for one of the biggest railway hubs in Finland, great by public transportation even in the rush-hour (map 2). High accessibility zones of Tikkurila are concentrated clearly along the rail connections and stations. Also, the main roads where there are bus connections and especially those that connect to the railway stations, are decreasing travel time to Tikkurila. Areas relatively far away from Tikkurila that are connected by the railway and metro connections can reach it in significantly by public transportation than by car. E.g. from Lauttasaari which was connected by the Länsimetro recently, can Tikkurila be accessed in under 30 minutes.

Map 3. Rush-hour accessibility of Tikkurila railway-station by private car.

Besides being an important node for railways, Tikkurila is located also close to major highways making it quite well accessible from at least the eastern parts of the metropolitan area by private car (map 3). Tikkurila is located just next to the ring road 3 and horizontally between Tuusulanväylä and Lahdenväylä. By joining the population data from Statistic Finland’s grid database (2016) to the grid cells where accessibility by car in rush-hours is 30 minutes or under, it could be calculated that within this area is living 410625 people, which equals to the 28 % of the population in the Helsinki Metropolitan Area (2017: 1475095,


Data sources:

Tenkanen, H., J.L. Espinosa, E. Willberg, V. Heikinheimo, A. Tarnanen, T. Jaakkola, J. Järvi, M. Salonen, T. Toivonen (2018). Helsinki Region Travel Time Matrix 2018. DOI: 10.13140/RG.2.2.20858.39362

Grid database 2016, Statistics Finland

Helsinki Metropolitan Area population in 2017:


GIS-practical 5: Segregation

In GIS-practical 5 we did a simplified segregation analysis, where clustering of low income and low level of education were used as indicators of the phenomena:

Map 1. Concentration of low income and low level of education.

In the big picture, widest concentrations of grid cells where low income and level of education is ranking in the highest quartile are found in the eastern parts of Helsinki Metropolitan Area, in Helsinki and Vantaa municipalities. It must be considered, that smaller concentrations and single grid cells where the criteria are met, is found all around the area and there are also some clear clusters in Espoo, too.

In Helsinki the weight is clearly in the east (Laajasalo, Herttoniemi, Roihuvuori, Tammisalo, Merirastila, Myllypuro, Kontula, Vesala, Jakomäki, Vuosaari) and north of the CBD following the railroads (in the west: Pohjois-Haaga, Kannelmäki, Malminkartano and more in the east: Malmi, Pihlajanmäki, Siltamäki, Tapulikaupunki). In Espoo, three most significant clusters are found in Espoonlahti, Espoon keskus and Matinkylä. Most significant concentrations in Vantaa are found in the eastern parts of the city, but minor clustering can be observed also in other parts of the city and in the west. In the north-east clusters are following the railroad (Korso, Koivukylä). Länsimäki in the south-east border of Vantaa is forming a unified concentration with Mellunmäki on Helsinki side of the administrative border, which is in fact part of the largest coherent cluster of in the Metropolitan area.

Also, some interesting irregularities can be observed: grid cells in the so-called ‘high-class neighbourhoods’, where one wouldn’t expect a high share in of low level of education and income (e.g. Kaivopuisto, Eiranranta, Westend, Kuusisaari). An explanation to these can be e.g. tax-avoidance, taxes paid abroad or inherited wealth, that can’t be observed in the statistics maintained by the Finnish authorities. Another interesting observation that could be done in the Hakaniemi-Kallio area: in a former working-class neighbourhood there is only one grid cell where is located a dormitory for homeless (Auroratalo in Helsinginkatu). Instead in neighbouring areas Vallila and Alppila, there can be found several of grid cells, which could be interpreted to indicate that these areas are still in different phases of gentrification, which has already earlier lifted the education and income levels in Hakaniemi-Kallio areas.

Future inhabitants and their services in Haaga

To understand any populated region it is foundational to know something about its inhabitants, so it isn’t surprising at all that also many fellow students have approached the past development of Haaga with population statistics. What I find difficult on developing an interesting and meaningful research question or development project related to Haaga, is the fact that I haven’t yet found almost any alarming or otherwise remarkably abnormal features in Haaga, that would stimulate my imagination to seek solutions for. This supports the interpretation, what I’ve heard especially within this course, that Haaga is lacking identity. Don’t know of that, but to me, Haaga looks like the average of the average normal finnish suburb, whatever that means…

On USP-studies I try to get more familiar with the quantitative methods such as statistics and GIS, and that’s why I would like to focus on the population changes in Haaga. We know very well, or at least we do have comprehensive data on, how the population has changed in the last decades in Haaga, Helsinki and Finland generally. There is also available population forecast as open data from the City of Helsinki, which does not, however, give any answers or speculate on qualities of future population beyond the amounts and age-groups in a specific neighbourhood.

I would suggest a deeper review of socio-demographic statistics and forecasts to understand how has the population structure and social composition of Haaga changed over time, and how the change is predicted to continue in the future. To connect this study to planning and policy level, a useful point of view would be to estimate how the future changes in customer bases will affect or should be taken into account in different services in Haaga. A couple of trends what I have recognized in the data I have examined until now:

  • Total population in Haaga isn’t expected to grow significantly in the near future, but the population structure will change over time. Amount of school-aged population will grow at least with a number of one school unit until 2030 (+664 inhabitants in age-group of 7-16). Potential rest home customers (age-group over 75) will grow (+515) until 2030.
  • It is a known fact that the share of the foreign speaking population will continue growing in the Helsinki Metropolitan Area. The same can be expected to happen also in Haaga, which ranks at the moment under the average in Helsinki (figure 1). How is the increasing share of foreign language speakers affecting the services in Haaga?
Figure 1. The share of foreign language speakers of the total population in Helsinki and in Haaga 1992-2018 (Source:
  • Is there a segregation process going on in Northern-Haaga and how this should be taken into account in services? On this week’s GIS-practice we did analyse segregation in Helsinki Capital Area using the share of low income and low level of education as indicators. When zooming at the Haaga area (figure 2), it seems that there is a clear north-south -distribution pattern of these two indicators within Haaga:
Figure 2. The concentration of low income and low level of education in 250 m cells.

One other perspective to services, which I didn’t notice on anyone’s blog, I would be interested to explore is accessibility: e.g. how many inhabitants of Haaga is able to access a service X within time Y with public transportation. This could be possible to estimate using MetropAccess -tool developed by Digital Geography Lab in the University of Helsinki. I have some experience with the tool since I have done some calculations with it to find out the best routes from schools and kindergartens of Vantaa to libraries in Vantaa.

This ragbag of ideas and thoughts around services could be formulated into several research questions, but as I have understood, this week’s blogpost’s are used more as a base on the discussion for the manuscript on the studio courses Haaga-research entity, instead of ready research plans. I will formulate here a couple of questions that would interest me:

How has the population structure and social composition of Haaga changed over time? How it will be in the future, and how the changes will affect or should be taken into account (location, language, capacity etc.) in different services?

GIS-practical 4: Clusters of workplaces

Our fourth practical was about clusters of different activities in cities and creating heat maps to visualize them. We did this exercise together with Joonas by analysing the employment clusters inside the Helsinki Capital Area (cities of Helsinki, Vantaa, Espoo and Kauniainen) by using the grid data of Finnish Environment Institute SYKE covering employments around the area in 2010. Our target employment fields were the commercial, industry, scientific & technology, and hotel, restaurant & catering beside the overall structure of all jobs. With these we created five heatmaps and one diagram to visualize the results.

Map 1. Concentrations of all jobs in Helsinki Capital area in 2010.

In general view of all jobs, the highest concentration (CBD) is clearly the downtown of Helsinki. The area reaches from the city centre, Katajanokka and Ruoholahti to Kalasatama, Kallio, Pasila, Meilahti and Ruskeasuo. Also, after a small gap, the area continues in Pitäjänmäki and Haaga areas. Outside Helsinki, there are smaller intensive job clusters in Tikkurila, Helsinki-Vantaa Airport, Leppävaara and Tapiola-Otaniemi. It’s obviously not a surprise that these are also the main transport hubs around. Jobs tend to concentrate along good transport connections especially when a significant amount of workers live outside of the metropolis.

Map 2. Concentrations of commercial jobs in Helsinki Capital area in 2010.

In commercial jobs, the distribution of clusters is very much larger than with jobs in total. The Helsinki CBD is already splitted in two smaller “bubbles” in the city centre and Kallio-Kalasatama. Also Itäkeskus, Herttoniemi, Lauttasaari, Munkkiniemi, Kannelmäki and Malmi have intensive commercial hubs. In Vantaa and Espoo the dispersion is even larger. In Vantaa, the “hottest spot” is clearly in Pakkala around the big shopping centre Jumbo. Also Petikko and Pähkinärinne have very red spots. In general, the commercial jobs are concentrated around the Ring Road 3 which punctures the whole city of Vantaa. In Espoo, the two most red spots are Leppävaara (Sello shopping centre) and Kilo (Inex Partners wholesale firm). Also in Matinkylä (Iso Omena) and Tapiola area (Ainoa) have small intensive clusters.

Map 3. Concentrations of scientific and technology jobs in Helsinki Capital Area in 2010.

Most of the Helsinki peninsula is covered with high concentration for scientific and technology jobs. Outside the CBD, Viikki stands out as a clear hot spot (university campus) and some minor clustering can be observed also in Malmi, Herttoniemi, Käpylä, Pitäjänmäki, Haaga and Malminkartano. In Vantaa, there is one clear hot spot located in Vantaankoski-Martinlaakso-Sanomala area and some minor clustering can be observed in Myyrmäki, Vantaanportti and Tikkurila areas near the Ring Road 3. In Espoo, Otaniemi-Keilaniemi area stands out clearly (hi-tech businesses and Aalto University) and minor clustering in the south in Niittykumpu and Matinkylä and a bit northern in Espoo centre, Nihtisilta and Perkkaa.

Map 4. Concentrations of industrial jobs in Helsinki Capital Area in 2010.

Hot spots of industrial jobs are mainly located, not surprisingly, in industrial areas, which are marked with grey on the map. There are a couple of interesting observations that can be done: hot spots in Kaartinkaupunki-Kauppatori-Hakaniemi area, Ruoholahti and Keilaniemi. Even though these aren’t today described as industrial areas, there are enough jobs categorized in the data as industrial to cause heat in this analysis. Former industrial areas like Pasila, Jätkäsaari and Kalasatama, which are all well-known city building projects today, are marked as industrial areas so it is important to evaluate how up to date the data is.

In Helsinki, the hottest spots can be found in the north-west in Pitäjänmäki, in the east in Herttoniemi and Roihupelto and in the southern part of the city in Alppila-Vallila, Ruoholahti and Kaartinkaupunki-Kauppatori-Hakaniemi area. In the northern part of Helsinki there are no significant industrial clusters, but right across the border on Vantaa side, clusters are found following again the Ring Road 3: Varisto, Vantaankoski-Martinlaakso, Vantaanportti, Tikkurila and Fazerila. All the hot spots in Vantaa are located in the southern part of the city near the ring road, except the one in Helsinki-Vantaa airport. In Espoo concentrations are mostly located in the south-western part of the city in Keilaniemi, Tontunmäki and Karaportti next to Kauniainen.

Map 5. Concentrations of Hotel, restaurant and catering jobs in Helsinki Capital Area in 2010.

The heatmap of hotel, restaurant and catering industry looks quite similar than the first overall job map. The Helsinki CBD stays as the main hub but logically the Helsinki-Vantaa Airport area is the second-most intensive area. Hotels, restaurants and caterings tend to concentrate around the tourist-intensive areas as airports and downtowns. Also in Tapiola, Leppävaara, Haaga, Tikkurila and Itäkeskus there are some clustering of these. Maybe the reason is the main transport hubs and shopping centres, especially with the restaurants.

Overall job structure

Diagram 1. Jobs by industry in Helsinki Capital Area in 2010.

Four industries analysed in heatmaps are highlighted in yellow in the diagram: commercial trade is holding 11%, science & technology 8%, industry 5% and hotels, restaurants & catering 4% shares of all the jobs in Helsinki.. Health & social services is the largest industry in with a share of 13% and commercial trade is the second largest with 11%. Other significant branches are information & communication (9%), the science & technology branch (8%), public governance & defence (7%) and education (6%).

Many of the top industries shares can be explained with typical capital city characteristics: state capital status and concentrated state administration, the most populous city surrounded by two other big cities, high educational level among the population, high amount of universities and research institutes, the busiest international airport, the largest and highly specialized hospitals etc. Together they generate a huge amount of synergy, which attracts even more new firms and employees to move there. In short, it’s very profitable to make business and work in Helsinki, which is the biggest hub of everything, and together with the surrounding cities, the biggest employment area in Finland.

GIS-practical 3: building efficiency ratios

In GIS-course practical 3, we did compare building efficiency ratios in Helsinki neighbourhoods and 250 m grids. We did together with Tuomas and Mika two maps with different scales and discussed the differences between them.

Map 1. Building efficiency ratio in Helsinki neighbourhoods.

The map comparing regional efficiency ratios in Helsinki areas is a nice explanatory tool. Comparing regional differences gives one a good overview in general, but lacks more detailed information. The obvious weakness of the map is that the ratio tells nothing about the building types (aside from a short description in the legend). In the case of Helsinki, this is problematic since the map shows semi-detached houses and rowhouses being the most common type of housing in Helsinki, while in reality that is not the case. A neighbourhood consisting mainly of apartment buildings among large parks might have the average ratio of a “semi-detached house neighbourhood”.

Map 2. Building efficiency ratio in 250 x 250 meter grid cells.

250 x 250 m grids or finer area units than districts give more detailed and comparable information of the efficiency because grids are equal sized and built environment that is not buildings, e.g. parks and football fields, are affecting less on the calculated values. The greater area unit we are focusing, the flatter values we get since it is always an average of several features within the area unit. Despite this, both maps tell the same story, highest building efficiencies are found in the downtown, but using the grid level it is possible to examine the variation also within the districts.

When the efficiencies were calculated with the districts, the highest ratio value any district received was 2.10 in Kamppi and the second highest was clearly lower in Punavuori 1.68. When calculating the same ratios with the 250 x 250 meter grids, variation is notably wider and because of that we created an extra class (2.10-3.64 Most dense apartment building blocks): highest grid value 3.64 was found in the Kluuvi district and totally 10 districts included one or more cells where the ratio is above 2.10. This is why many spatial phenomena should be examined in different area units. One example of the importance of using suitable area units can be raised from Vaattovaara & Kortteinen’s segregation studies: the concentrations of underprivileged people in Helsinki region are found in single housing companies or even in stairways, which couldn’t be observed if the area unit used is too large (e.g. blocks or districts).

Building efficiency ratio as a measure of urbanity

Building efficiency ratios are very useful in comparing different districts densities. The ratio gives an idea how densely built area is. Maps showing more detailed efficiency ratios can point where the most urbanized areas of the city are.

However, urban is not all about density and a high building efficiency ratio tells little about the neighbourhood’s character. Sure, a dense neighbourhood offers more encounters between residents, which is essential for an urban environment. But an urban area should offer a wide variety of services. A business district that is dead silent after 5pm in a Friday or a residential area with only high-rise apartment buildings and green areas between them are hardly urban.

Customer density: Understanding where the users are

Density related feature: Customer density for public (why not private too) services to find the optimal locations.

My educational background is in regional and administrative sciences and most of my working history is from the municipal sector, so I’ll approach this task with an example from public service network-point of view. There are many different kinds of municipal services, which are produced towards different age-group based segments (e.g. kindergartens and schools) or other user quality based segments like life situations (e.g. unemployment, illness, disability etc.).

I will use the school network as an example of the need to understand customer density. To plan a well working and resource efficient service network, cities need to understand many factors such as where the customers of a specific service are located (amount of school-aged children in different areas) and how far can this particular customer group travel to reach the service (where to locate the schools). And as always, there is a shortage of resources so the city needs also review if the location is economically wise, or should the schools be concentrated to larger units.

If there is a high amount of primary school-aged children (enough users per service unit) in a specific neighbourhood, it often could be wise to place the school there. If not, the city could arrange a transportation for children to get safely to a nearby school-unit. If the question is about locating a high-school, which students are already older and able to travel more independently, in a situation when there is not enough of students to fill a school-unit in one area, it could reasonable and economically wise to concentrate high-schools to meet the demand of larger areas.

I’ll attach here a map analysis I made as an infill for the portfolio when applying to USP-program year ago:  “Where do the children of Helsinki live and where are their schools located?”. The maps show where the Finnish elementary schools are located, where the relative share of the population aged 0-9 is high and where the absolute amount of population aged 0-9 is high. The latter number is more useful when planning school networks: one school unit can handle X amount of children, despite the percentages.

BUT… for planning the future services, we need to estimate how many users there will be in the future, so the analysis of population in the present isn’t enough. That’s why it is important to study the population forecast data, which clearly shows in the case of Haaga, that it is expected significant growth (+664) in the amount of school-aged population (7-16 years) from today’s amount of 1697 to 2361 in 2030:

School-aged population (7-16) in Haaga

Figure 1. 7-16 year-old population in Haaga within the subdistricts 2010-2030

Data source: Population forecast of Helsinki 2019-2030 (Helsingin väestö ja ruotsinkieliset 1.1.1992-2018 sekä väestöennuste 1.1.2019 – 2030), City of Helsinki 2018. Data reached 11.10.2018:

Beyond Wikipedia

Instead of Wikipedia, I had a look for population and building data, which were combined, selected and visualized in Microsoft Excel and QGIS.

Population timeline

Picture 1. Rapid growth (1950-1965) with orange and the population peak with red, forecast (2018-2024) with lighter blue.

As seen in picture 1. Haaga started growing rapidly in the 1950’s (post-war baby boom, great migration from countryside to cities), which ended in a population peak in 1965. In that period of time, the population grow from 3952 to 29 835 inhabitants. After the peak total population declined a bit and since 1970’s the population have stayed quite stable in under 28 000 inhabitants (2018: 27 606). Population forecast until 2024 doesn’t either predict any radical changes in the number of inhabitants in Haaga.

Picture 2. Population growth within the sub-districts of Haaga.

When looking at the population growth regionally within the sub-districts of Haaga, picture 2. shows that especially Etelä-Haaga and also Pohjois-Haaga are the oldest and most populated parts of the area. Kivihaka sub-district has been stable and small since the 1960’s. Lassila subdistrict had significant growth between 1974-1988, but have been relatively stable since then.

Buildings and the importance borders

Building registry data features (buildings) were selected in QGIS by location, which was defined as the borders of Haaga, and visualised by the year of taken into use. When doing this kind of classifying of objects by their locations, it is essential to define the borders which we are using as intersections. So what do we really mean with the concept of Haaga area?

Picture 3. Buildings of Haaga by the year of taken into use. Left map with district border (kaupunginosa), right map with sub-district borders (pienalue).

In the GIS-course we have been using a provided dataset named Helsinki_small_areas.shp, which includes the borders of districts or neighbourhoods in Helsinki (kaupunginosa). I found the name of this dataset confusing because City of Helsinki is using the word pienalue, that translates directly to small area, to refer in sub-districts such as Etelä-Haaga, Pohjois-Haaga, Lassila and Kivihaka, which are generally perceived as parts of Haaga district. In addition to these four sub-districts, the Haaga district (picture 3, left map) is covering also the North-Western part of Pikku-Huopalahti. Right map of picture 3 shows the sub-district borders, note that Pikku-Huopalahti is larger area than the part of it which is included within the Haaga district borders.

I personally don’t perceive Pikku-Huopalahti as a part of Haaga, and because it is in many ways totally different kind of area compared to the other sub-districts of Haaga, this has significant impacts on how things look in statistics. It is also worth to note, that City of Helsinki has also an area classification called statistical areas (used e.g. in the population data), where Haaga is including the four sub-districts but not any part of Pikku-Huopalahti. Pikku-Huopalahti is an area mostly built in the 1990’s which stands out as a cluster of dark blue features on the southern part of the map. The picture 4. is showing a cumulative sum of buildings of Haaga in a timeline, where the effect of Pikku-Huopalahti can also be observed as a steepened growth at the beginning of the 1990’s:

Picture 3. Rapid growth (1950-1965) highlighted with orange. Note also the steepened growth at the beginning of the 1990’s.

Do you perceive North-Western part of Pikku-Huopalahti as a part of Haaga?


Data sources:

Population data 1950-1960: Historialliset tilastot,  Helsingin henkikirjoitettu väestö kaupunginosittain 1.1.1875-1960. City of Helsinki. Downloaded from 29.9.2018.

Population data 1962-2018: Helgingin väestö alueittain 1962-. City of Helsinki. Downloaded from Helsinki Region Infoshare 29.9.2018.

Population forecast 2018-2024: Helsingin ja Helsingin seudun väestöennuste 2015-2050. Downloaded from Helsinki Region Infoshare 29.9.2018.

Border data: Kaupunginosajako (district), pienalueet (sub-district). City of Helsinki. Downloaded from API 29.9.2018:

Building data: SeutuCD 2013, Helsinki Region Environmental Services Authority (HSY). Reached from the Urban GIS -course material.