THE MEASUREMENT OF URBAN LAND CONSUMPTION
AS A SOURCE OF INDICATORS OF ECONOMIC
PERFORMANCE AND SUSTAINABILITY
Rodrigo Bastías Castillo[1]. Ministry of Housing and Urban Development. Housing and Urban Studies Commission. Urban Observatory.
Sergio León Balza[2] Ministry of Housing and Urban Development. Housing and Urban Studies Commission. Urban Observatory.
ABSTRACT
Once context is adequately ascertained, urban land consumption analysis may serve to simultaneously measure economic performance and sustainability of cities.
The Urban Observatory of the Ministry of Housing and Urban Development carries out a continuous measurement of urban land consumption in Chile, combining the use of Geographical Information Systems (GIS) and statistical examination. The results of the analysis are published by means of indicators of urban land occupation, urban densities, non-occupied urban land and construction permit trends for different land uses.
The sources of information have been satellite images, used to measure the situation in specific moments in time (so far, 1993, 2003 and 2008), complemented by the analysis of construction permits. The latter provide a monthly source of information that is used to analyze short term (three monthly) as well as longer term dynamics, made public by means of three monthly bulletins.
This paper comments the methodologies used, the main results obtained, estimated errors as well as the shortcomings of the process of analysis carried out since 2005.
Doubts have been resolved validating the findings with qualified actors, both public and private sector. The dissemination of results is complemented by the extensive publication of the method used and the open discussion of uncertainties, leading to the propagation of information in a manner that invites new forms of interpretation and further analysis.
Keywords: Land Use, Urban Development, Land Indicators, GIS, Construction Permits, Urban Sustainability, Urban Land Demand
INTRODUCTION
The importance of measuring land demand
This measurement is used to monitor city growth and the distribution of different types of land use, as well as for observing the economic dynamics associated with different land uses and, at the same time, analyzing the relationship between unoccupied and available land, with the possible impacts on valuable natural or agricultural production[3] sites.
This implies not only conducting statistical analyses, but also geographic analyses of the urban and potentially urban areas established in the Land Use Plans. This paper shall refer only to the statistical analysis of construction permits in cities and urban areas across different regions and municipalities.
How may land demand be measured by type of use?
In 2005, the Urban Observatory began a geographical analysis of urban growth, using satellite images from 1993, 2003 and 2008. This allowed the institution to estimate the amount of occupied and unoccupied land without distinguishing the dynamic contributions of different land uses. Thus, Construction Statistics represent an interesting field of analysis.
The calculation of land demand by reviewing construction permits refers to what may has been calledpermissed land in other publications on urban indicators[4], is only a preliminary approximation to this analysis. This paper includes information from the January 2003 - December 2009 period, at the country’s most basic administrative division: the municipality.
Methodological considerations for estimating permissed land
The figures in this paper may refer to the net surface of urban land associated to construction permits, that is, they do not necessarily consider urban land that has been occupied through legal cessions authorized by the Municipal Constructions Department (DOM) such as streets and new open space. These figures do not include parks and geographical accidents (bodies of water and streams, hills, etc.) or their surrounding areas, which may be incorporated as urban land in the process of occupation.
It must also be considered that new apartment buildings tend to be built on previously occupied land, whereas land consumption associated to houses may imply new consumptionwhen built in municipalities on the outskirts of cities with more than one municipality, or on the surroundings of municipal areas occupied by a city within the municipality[5].
Furthermore, the construction permits database has registries with values that appear too high in relation to the square meters of land authorized for construction, originated by mistakes in filling out the Construction Forms provided by the DOM. The high figures may also be caused bydigitizing mistakes when using this form to add information to the database built by the National Institute of Statistics.
For all these reasons, the information listed below was obtained by applying a series of filters, using the criteria that shall be described in the methodology[6].
It is important to note that these figures were taken from the updated information from the first two months of each quarter, as well as provisional information from the third month[7].
1.ANALYSIS OF THE 2003-2009 PERIOD
1.1 National situation
The following table shows the land distribution, measured in hectares (ha), for twelve types of land uses.
Table No. 1:
Of the total land surface, urban land demand is associated mainly to permits destined to houses (59% of the total demand). This type of land use tends to consume more urban land than the sum of all the others, as is illustrated in the following graph.
Graph No. 1:
1.2. Regional situation
Table No. 2 presents the regional demand of premised land, organized from largest to smallest, in order to identify the participation of each region in decreasing order.
Table No. 2:
Graph No. 2:
Table No. 2 and Graph No. 2 show that five regions concentrate 73.2% of all permissed land: RM, Bio-Bío, Valparaíso, Los Lagos and Araucanía. The first three have metropolitan cities and concentrate 56.3% of this land demand.
1.3.Land surface with construction permits in Regional Capital Cities
An analysis of regional capital cities implies considering the authorizations for constructions in the urban areas of municipalities within those cities. This figure was obtained by applying the definition of the term “city” established by the Ministry of Housing and Urban Development[8].
A comparison of the general demand for permissed land in these cities reveals that Santiago concentrates the largest share (39.2%), followed by Concepción (17.8%) (see Table No. 3). Curiously, Puerto Montt occupies the third place on the list, with a higher demand for permissed land than the city of Valparaíso, as a result of the high demand for houses in 2005.
Table No. 3:
2.ESTIMATED ERRORS AND RESTRICTIONS OF ANALYSIS
2.1 Repeated construction permits: association of several land uses to a single plot of permissed land
Construction permit databases occasionally include single cases that register up to five different land uses. For example, a plot of landmay be granted a permit to build a house, and that permit is registered; later, a new permit is again registered for commercial use, and then another for office use on the same plot of land. This may occur either simultaneously or over a period of time.
On average, each year around 6% of all permits are thus “repeated”.Since no criteria have been established for selecting one of these multiple land uses over another these repeated permits are not accounted when estimating the demand of permissed land.
Separating repeated permits into thedifferent land useswould require a modification of the form used to register construction permit information. However, even if this goal was accomplished, it could not determine easily which land use should be associated with land demand, since this would imply an arbitrary decision that could introduce a bias. Another possibility would be to research when each different land use permit was granted over the same plot of land, identifying the exact day the permit was issued, and associating that land demand to the first permit.In any case the modification of the permit form would obtain greater details from the DOM, but it may still remain imprecise in the association of a given land use to a particular surface of permissed land.
2.2 Repeated entries of permissed land surface for different permits of the same land use
These occur mainly in the “Houses” and “Industry” land uses and, to a lesser degree, in the “Commerce” land use group.
In the housing Real Estate market, it is common to find sites with large surfaces on which groups of houses are built in different stages. In this case, a repeated entry is produced because it is registered over and over again each time a new house permit is granted.
No criteria have been established to identify these redundant housing permits, since they occur in different types of construction activities. It is not enough to subtract the number of times a repetition occurs, since that type of analysis would require sophisticated computer procedures, first to identify all the repeated permits, and subsequently to apply an algorithm capable of ultimately correcting those figures.
“Industry” shows two cases of this phenomenon: in the Municipality of Puchuncaví, in the Valparaíso Region, where a 30.9 hectare site has been listed six times, and in the Municipality of Mulchén, in the Biobío Region, where there has been a similar case in which a 26.6 hectare site was listed five times.
In “Commerce,” this situation occurs only very rarely and the plots of land are smaller.
The “Apartment Building” category does not present this type of irregularity, since building blocks do not experimentgradual construction processes over the same plot of land as it occurs with houses. This has only occurred in one case, in the Municipality of Arica, in the Arica y Parinacota Region, where a 3.81 hectareplot has been listed four times.
3.METHODOLOGY
3.1 Land uses identified in the Construction Permits Statistics Form
This database includes 46 land uses, traditionally organized into three large groups: Housing (which includes 12 different uses of houses and apartment buildings); Industry, Commerce and Financial Establishments (which includes 15 uses); and Services (which includes 19 uses).
In order to generate a more precise analysis of urban construction authorizations –and their diverse impacts on land occupation in each region, city and municipality– the database has been organized into twelve (12) large land use groups. Table No. 4 also considers the filters applied to each group.
3.2 Obtaining the twelve land use groups and applying site surface filters[9]
Three key steps were followed:
- Extracting non-urban land uses: “Agricultural Industry,” “Fishing” and “Mines and Quarries.” After this third step, the remaining land uses are organized into the 12 groups mentioned in Table No. 4.
- Selecting the permits for the urban area. This step eliminates approximately 50% of all construction permits to be analyzed.
- Selecting permits for “New Constructions” and “Regularization of New Constructions”, but not “Extensions” and “Regularization of extensions,” since these take place on plots that already count with permits.
Table No. 4:
Filters were established by measuring the site surface of diverse land uses in different cities throughout the country, using “Google Earth” software and considering the four city categories established by the Ministry of Housing and Urban Development. The reason for determining filters was the existence of construction permits with very large surfaces in the data base.
It was observed that actual blocks showed a surface of almost 1.2 hectares. Large malls, supermarkets and hospitals were observed to have been built on sites of up to 33 hectares.A slightly higher margin –35 hectares– was established for these cases, as the measurements samples were carried out only in some cities or municipalities.
A similar procedure was applied to each of the remaining groups of land uses.
Table No. 5 shows the percentage of permits obtained for each group of land use after applying the filters.
Table No. 5
4.FINAL COMMENTS
Although it may be necessary to review the methodology for the analysisof plot sizes associated withspecific land uses, the methodological process generated data that has been found acceptable by urban development experts. This in itself is a substantial achievement.
It could be recommended to define a series of indicators, such as those established by the Urban Audit[10], for example, the proportion of different land uses as part of total land; occupied and unoccupied urban land, and occupation rates for different land uses, among others.
We could also take advantage of the available geographical information systems to calibrate the precision of the information extracted from Construction Statistics.
[1] Rodrigo Bastías Castillo is a graduate Statistician, Catholic UniversityChile, who works as analyst/researcher at the Urban Observatory.
[2] Sergio León is B.Sc. Rural Environment Studies, University of London. M.A. Latin American Studies, University of Liverpool. UK. Head of the Urban Observatory. Housing and Urban Studies Commission (CEHU), Ministry of Housing and Urban Development (MINVU) – Chile.
[3]Therefore the classification of urban land indicators may be found in the area of the economy or the environment.
[4] For example, refer to the “Atlas de Indicadores de Desarrollo Urbano de la Ciudad Autónoma de Buenos Aires”. Ministerio de Desarrollo Urbano, Gobierno de la Ciudad de Buenos Aires. Argentina. 2009.It refers to new urban land estimations taken from the land associated to urban construction permits.
[5]“Peripheral municipalities” are those located on the outskirts of cities.
[6] In December 2009, for the first time, the Land and Construction Report published the results of the application of this methodology. .
[7] MINVU has established three types of construction permit figures: definitive, updated and preliminary. The database for each year contains definitive figures only afterINE sends a revised data base in June of the following year. The monthly information used to write the Land and Construction Reports is considered updated two months after the permit date. Before the third month, the figures published are considered preliminary. The quarterly data included in each report considers updated figures for the first two months and temporary figures for the last month.
[8] MINVU has defined 147 cities, divided into 3 Metropolitan Cities (Santiago, Valparaíso and Concepción), 18 Major Intermediate Cities, with populations of 100,000 to 300,000 residents according to the 2002 Census; 32 Minor Intermediate Cities, with 20,000 to 99,999 residents; and 94 Small Cities, with 5,000 to 19,999 residents according to the Census. It has defined urban conglomerates as a group of localities that are inter-connected by urban public transportation systems. Please refer to“Definición y listado de ciudades de Chile” (November2007), published on in the section called “Documentos del Observatorio Urbano.”
[9] The professional team at the Urban Observatory has measured the surface of urban land occupied by Chilean cities. Please refer to the section called “Indicadores por Comuna” on the website, and the document called “Medición de la superficie ocupada por las ciudades de Chile de más de 15.000 habitantes: 1993 – 2003”, from February 2007, in the section called “Documentos del sitio”. These measurements were made with satellite images of the first months of 1993 and 2003.
[10] The Urban Audit has defined land use indicators. Please refer to Urban Audit-Methodological Handbook, pages 41 and 42., 2007.