How much green is needed for a vital neighbourhood

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/authorsrights

http://www.elsevier.com/authorsrights

Author's personal copy

Land Use Policy 38 (2014) 330–345

Contents lists available at ScienceDirect

Land Use Policy

journa l homepage: www.e lsev ier .com/ locate / landusepol

How much green is needed for a vital neighbourhood? In search forempirical evidence

Barbara Szulczewskaa,∗, Renata Giedycha, Jacek Borowskib, Magdalena Kuchcikc,Piotr Sikorskib, Anna Mazurkiewiczd, Tomasz Stanczyke

a Warsaw University of Life of Life Sciences, Faculty of Horticulture, Biotechnology and Landscape Architecture, Department of Landscape Architecture,Nowoursynowska 159, 02-776 Warsaw, Polandb Warsaw University of Life of Life Sciences, Faculty of Horticulture, Biotechnology and Landscape Architecture, Department of Environmental Protection,Nowoursynowska 159, 02-776 Warsaw, Polandc Institute of Geography and Spatial Organization of Polish Academy of Sciences, Department of Geoecology and Climatology, Twarda 51/55, 00-818Warsaw, Polandd Warsaw University of Life of Life Sciences, Faculty of Animal Science, Department of Animal Environment, Nowoursynowska 159, 02-776 Warsaw, Polande Warsaw University of Life of Life Sciences, Faculty of Civil and Environmental Engineering, Department of Environmental Improvement, Nowoursynowska159, 02-776 Warsaw, Poland

a r t i c l e i n f o

Article history:Received 18 December 2012Received in revised form 6 November 2013Accepted 9 November 2013

Keywords:Environmental performanceEco-spatial indexLand-use planning

a b s t r a c t

In this paper, we attempt to find empirical evidence for the proper size of the Polish eco-spatial index,known as the Ratio of Biologically Vital Areas (RBVA). The objective is to establish the minimal proportionof green space required for good environmental performance in neighbourhoods.

Eighteen neighbourhoods (representing the most popular type of residential areas consisting of multi-storey buildings) located in Warsaw and characterised by different RBVA values (varying from ca. 20%to ca. 70%) were chosen as the study area. Different types of measurements and calculations were per-formed to verify the relationships between the size of the RBVA and selected environmental features(e.g., air temperature and humidity, floristic diversity, butterfly species richness, surface outflow, etc.).Based on these values, a threshold of 45% RBVA was recommended as the minimum, which guaranteesenvironmental performance in the neighbourhoods to certain extent. Eco-spatial indices can be recom-mended as a useful planning tool for new projects and for evaluation and enhancement of existing urbanstructures, including residential areas. It should be stated that these indices are not the only measuresfor green space planning, because they do not refer to the major residents’ needs (e.g., social interaction,recreation).

© 2013 Published by Elsevier Ltd.

Introduction

Urban green spaces provide many environmental and socialservices that contribute to the quality of life in cities, i.e., air filter-ing, microclimate regulation, noise reduction, rainwater drainage,recreation and social interaction (Gill et al., 2007; James et al.,2009; Thompson, 2002). Because more than half of the Earth’spopulation resides in cities, the need exists to develop urban struc-tures that are more resilient, sustainable and liveable. From thebirth of urban planning practices, the challenge of protecting and

∗ Corresponding author. Tel.: +48 22 59 321 91.E-mail addresses: barbara [email protected], [email protected] (B. Szul-

czewska), renata [email protected] (R. Giedych), jacek [email protected] (J. Borowski),[email protected] (M. Kuchcik), piotr [email protected] (P. Sikorski),anna [email protected] (A. Mazurkiewicz), tomasz [email protected](T. Stanczyk).

developing green spaces has played an important role. Many dif-ferent concepts that aim to safeguard and enhance urban greenareas have been applied in urban planning history, i.e., green belts(Amati, 2008; Self, 2002), green wedges (Caspersen and Olafsson,2010; Gordon et al., 2009), greenways (Fábos, 2004; Turner, 1995),open space area percentages and catchment area standards (Turner,1992; Byrne and Sipe, 2010), ecological networks (Jongman et al.,2004; Opdam et al., 2006) and green infrastructures (Benedict andMcMahon, 2006; Hostler et al., 2011).

In recent decades, traditional urban planning approaches havemoved from separating different functions into different spaces tomore diverse land uses located in the same space, including morebiodiverse landscapes (Puppim de Oliveira et al., 2011).

Land-use integration principles are closely related to densi-fication of urban areas and development of high-density urbanstructures. However, this situation represents one of the paradoxesof sustainable development principles. Densification is essentialfor protection of the landscape/environment outside of cities and

0264-8377/$ – see front matter © 2013 Published by Elsevier Ltd.http://dx.doi.org/10.1016/j.landusepol.2013.11.006


B. Szulczewska et al. / Land Use Policy 38 (2014) 330–345 331

sustainable use of resources. However, the question arises as towhat extent of densification should be permitted in urban areas,taking into account the built environment conditions, environmen-tal processes and quality of life for urban citizens. Tratalos et al.(2007) asked whether it is possible to build dense and compactcities that maintain areas of natural habitat and provide useful lev-els of ecosystem services. This problem has also been consideredfrom the point of view of climate change and the role of green spacein the mitigation of urban heat islands, improvement of air qual-ity, reduction of storm water runoff, creation of habitats for animalsand enhancement of biodiversity (Alcoforado et al., 2009; Gill et al.,2007; Makhelouf, 2009; Pauleit and Duhme, 2000). Gill et al. (2007)noted that little is known at present with respect to the quantity andquality of green spaces required to adapt cities to climate change.

History emphasises different approaches intended to establishrelationships (a sort of balance) between green and built-up areas. Areview of approaches and methods of planning for open space, alsounderstood as areas dominated by a “natural environment”, waspresented by Maruani and Amit-Cohen (2007). These researchersidentified nine approaches, but for urban areas, only a subset couldbe considered as relevant, i.e., opportunistic, space standard, parksystem, garden city and shape-related models. All these modelsplay different roles and contain certain limitations for establish-ing relationships between green and built-up areas in cities. Spacestandards are particularly important when densification discourseis taken into consideration. The aims for implementation of thesemodels have varied and evolved over the history of urban planning.Maruani and Amit-Cohen (2007) noted their chief limitation, whichis a lack of reference to site features, resulting in underestimationof the nature and heritage values of the plan site. However, thesemodels are easy to put into practice and are therefore popular inmany countries. A creation of places for recreation, particularly forchildren has been the most important aim for their implementation(Byrne and Sipe, 2010).

In recent decades, due to the development of ecologi-cal/biodiversity discourse in spatial planning and problems createdin cities by climate change, a new type of space standard hasappeared and is known as ‘eco-spatial indices’. This concept con-sists of indicating a proportion between the built-up and greenareas and/or establishing development rules related to the typeof greenery and occurrence of open water, permeable paving,vegetated walls and roofs, and storm water infiltration facili-ties over existing vegetation, etc. on the site. The rationale forthese types of indices the retention of as much space as possi-ble for the sake of environmental performance within the builtenvironment.

Eco-spatial indices have been introduced as a planning measure(usually as a component of planning provisions) in certain cities, i.e.,Berlin (Biotope Area Factor), Malmo (Malmo Green Factor), Seattle(Seattle Green Factor) and Singapore (Green Plot Ratio).

Most eco-spatial indices, i.e., the Biotope Area Factor (Hagenand Stiles, 2010), the Malmo Green Factor (GF) (Hagen and Stiles,2010), and the Seattle Green Factor (SGF) (Hirst et al., 2008),express a ratio of the area covered by greenery, open water, perme-able paving, storm water infiltration facilities, etc. to the total sitearea.

Scoring systems also include other green solutions, such as theprotection of existing trees, use of harvested water for irrigationor the presence of drought-tolerant plants. Each of the elementsmentioned above is weighted according to its environmental value.The most valuable sites are areas covered by plants and openwater.

A different approach is represented by the greenery provision(GnP), which was introduced in Singapore. The GnP is weightedaccording to the Green Plot Ratio (GnPR) value as defined by Ongin 2003. The GnPR is based on a biological parameter known as the

Leaf Area Index (LAI), which is defined as the single-side leaf areaper site area (Ong, 2003).

In Poland, an eco-spatial index known as the Ratio of Biologi-cally Vital Area (RBVA) was first introduced in 1997 by Warsaw’sVoiwoda in an ordinance aimed to establish Areas of LandscapeProtection in Warsaw’s Voiwodship (not in force at the moment).The RBVA reported the ratio between the areas covered by vege-tation or open water (not sealed areas) to the plot size and wasintroduced as a standard for built-up area development within theAreas of Landscape Protection. The minimal size of the RBVA wasset as 70% of the plot.

The debate that occurred among planners on the legal basisand precise definition of the ratio led to a new regulation, theEnvironmental Protection Act of 2001. This act introduced theobligation of establishing the RBVA (more precisely, the proper pro-portion between green and built-up areas) in planning documentsto enhance environmental performance and living conditions. Thelegal definition and minimum size of the RBVA (set as 25%, but onlyfor housing and health services) was established in the Ordinanceof the Minister of Infrastructure in 2002.

The main problem in applying this ratio is the lack of an empir-ical basis to inform decisions as to what should be considered ‘theproper’ share of BVAs for different land uses. Decisions on establish-ing the size of the ratio (70% in Warsaw’s Voiwoda Ordinance, 25%in the Minister of Infrastructure Ordinance) were made withoutinput from any previous research.

Analysis of planning documents indicates that planners appliedthe RBVA concept according to their own views and beliefs ratherthan by set principles. In many cases, it is difficult to explain whysuch different RBVA sizes appear under comparable conditions.

In this paper, we present an attempt to find empirical evidencefor the proper size of the Polish eco-spatial index known as the Ratioof Biologically Vital Area (RBVA). Our objective is to establish a min-imum proportion of green space required for good environmentalperformance in neighbourhoods. In this paper, we define a neigh-bourhood as a housing estate consisting of multi-storey blocks offlats. This type of neighbourhood is the most frequent type in Polishcities, accounts for the largest amount of city space and is knownto be important to a large portion of the population.

Methodology

Research concept

To investigate how the proportion between “vital” and “sealed”areas may influence the environmental processes at a site, wedecided to study selected processes in existing neighbourhoods.

We assumed that the results of this survey would allow us tofind the minimum ratio that should be recommended as ‘proper’for planning practices.

The indicator variables that were chosen to inform the environ-mental performance of an area are presented in Table 1.

The Polish Environment Protection Act emphasises ecologicalbalance as a primary reason for introducing the RBVA. However,the term “ecological balance” is a bit misleading because it canbe used in many different contexts. In ecosystem theory, ecolog-ical balance refers to ecological stability (e.g., Grimm and Wissel(1997) described 163 definitions of 70 different concepts of eco-logical stability), but in sustainable development theory, it refers tothe ‘ecological footprint’ used to analyse the supply-demand bal-ance (e.g., Dong et al., 2011). In the Polish Environmental ProtectionAct, this term is used as a metaphor to indicate the sustainment ofenvironmental performance in an area. Although this performanceis closely linked with quality of life, the RBVA concept is based onquantity and not quality, of green spaces. Therefore, in this study,we were interested in if and how the quantity of biologically vital


332 B. Szulczewska et al. / Land Use Policy 38 (2014) 330–345

Table 1Environmental performance variables.

Topic of the survey Purpose of the survey Investigated variables Methods of investigation

Climate To observe how a difference in RBVAinfluenced the thermal, humidity and windconditions; to observe differences in thethermal, humidity and air flow conditionsover different types of land cover

Air temperature Standard methods applied in local climateresearch. Hobo Pro loggers used as stationaryweather stations and in route measurements.

Air humidityWind speed

Hydrology To observe how a difference in RBVAinfluenced the hydrological process intensityand verify the possibility of excessive surfacerunoff retention

Surface outflow Indirect assessment with the use of a runoffcoefficient for different land use types;estimation of evapo-transpiration based onthe leaf area index (LAI).

Evapo-transpiration Results are compared with values typical fornatural, undisturbed land

Infiltration (conditions typical of the periodafter intensive rainfall).Volume of storm water outflow after 30 mmof rain

Ecology To observe how a difference in RBVAinfluenced the biological diversity andnaturalness of green spaces

Floristic diversity Floristic diversity according to number ofplant species and Shannon’s and Simpson’sdiversity indices; naturalness indicesaccording to the ratio of the number of alienspecies to all species within a sample;percentage of native non-synanthropic plantspecies in sample

NaturalnessTo observe how a difference in RBVA andtype of vegetation influenced the number ofbutterfly species

Butterfly species richness Observations and catches of specimenswithin the green spaces of eachneighbourhood

To evaluate the intensity of ecologicalprocesses

Green Plot Ratio LAI (leaf area index) measurements and thetotal GPR coefficient calculated

areas influences the neighbourhood inhabitants’ level of satisfac-tion with green spaces.

Taking into account the fact that planners most frequently spec-ified the size of an RBVA as 40–50% in neighbourhoods that wererecently constructed in Warsaw (one hundred neighbourhoodsconstructed between 2007 and 2009 were surveyed), we decidedto pay particular attention to this size of RBVA.

Study areas

The study areas were chosen according to the following criteria.

(1) The variation in their ratios of biologically vital areas rangedfrom less than 20% to almost 70%; we were particularly inter-ested in those with a ratio between 40% and 50% BVA (7neighbourhoods).

(2) The neighbourhood’s location in relation to the green openspaces as well as its location within the city (Fig. 1). Generally,the surveyed neighbourhoods are situated in densely built-up areas and are located in two zones defined in the SpatialPolicy of Warsaw as the ‘downtown functional zone’ and the‘urban zone’. We aim to measure whether there is a significantdifference between the environmental performance of neigh-bourhoods adjacent to green open spaces and those with noproximity to such areas.

(3) The age of the neighbourhoods. We considered neighbourhoodsconstructed at least 10 years ago such that vegetation wouldhave had time to develop.

(4) The geographical sizes of the neighbourhoods. We chose neigh-bourhoods that had similar areas.

Each selected area of study was described using the Geo-graphic Information System (GIS), and the general characteristicsare shown in Table 2. Biologically vital areas, sealed surfaces andlayouts of trees of the neighbourhoods are shown in Fig. 2.

Methods

ClimateThe city structure changes constantly, especially based on air

temperature, humidity and wind conditions, which is why thesespecific elements were studied.

The air temperature and relative humidity data were obtainedfrom 18 stationary loggers (Hobo Pro) located 1.5 m above theground over lawns in a manner that represented the climate ofthe entire neighbourhood. This method is often used in local cli-mate analyses (Hawkins et al., 2004; Joly et al., 2003; Matzarakiset al., 2007). The loggers operated from autumn of 2008 to August2010 and provided 10-minute averaged data (in addition to dailyor monthly averages). These data were used as the base data for thestudy.

To show the differentiation of bio-thermal conditions (i.e., airtemperature, humidity and wind speed) inside each neighbour-hood, route measurements were conducted three times each day(morning, afternoon, and evening) on two days during the sum-mer of 2009. The measurements were performed with the sameHobo loggers in 8–10 locations over different surfaces and in dif-ferent urban structures in each neighbourhood for 5 min on eachsite (at 1-min intervals). The average of the data from the 3rd and4th minutes of this measurement was compared with the base datafrom the same time period. This methodology has been successfullyemployed in past city climate studies conducted by the authors.

HydrologyAnalyses of the hydrological process intensity were based on

data that characterised the type and area of land cover forms,including buildings, paved surfaces that drain to a storm watersewer system or to pervious areas, and vegetated areas (i.e., lawns,trees and shrubs). An indirect collection method was used due tothe lack of ability to obtain exact measurements of the hydrologicalprocess intensity in the chosen locations. This method was based

Author's personal copyB.Szulczew

skaet

al./LandU

sePolicy

38(2014)

330–345333

Table 2General characteristics of surveyed neighbourhoods.

No. Neighbourhood Totalarea (ha)

Number ofinhabitants

Average heightof buildings instories

Ratio ofBiologicallyVital Areas (%)

Biologicallyvital areas perInhabitant (m2)

Floor AreaRatio

Canopyarea (ha)

Shrubarea (ha)

Lawnarea (ha)

Greenery characteristics Neighbourhood (landuse within the rangeof 0.5 km)

1 Hoza 7.38 5044 4.7 16.41 2.20 2.23 0.30 0.19 1.20 Small patches of intensivelytrampled lawns, very small singlepatches of flowerbeds, numerousruderal communities, single youngtrees

Downtowncommercial andresidential areas,industrial areas

2 Panska 5.13 4273 6.6 17.67 2.12 2.50 0.28 0.09 0.83 Small patches of intensivelytrampled lawns, very small singlepatches of flowerbeds, smallpatches of scattered trees

Downtowncommercial andresidential areas

3 Sandomierska 6.28 3775 4.6 18.24 3.04 1.80 0.55 0.12 1.09 Small patches of intensivelytrampled lawns, very small singlepatches of flowerbeds, numerousreferral communities, single youngtrees, sporadic occurrence ofmature trees

Downtowncommercial andresidential areas

4 Wlodarzewska 5.07 2555 4.4 40.66 9.78 1.25 0.29 0.28 1.99 Predominantly small patches ofmanicured lawns, very smallpatches of flowerbeds, local largepatches of intensively trampledlawns, single young trees

Multifamilyresidential areas,urban green areas

5 ZgrupowaniaZmija

7.57 2944 4.6 41.73 12.25 1.02 0.63 0.59 2.91 Predominantly patches ofmanicured lawns, very small singlepatches of flowerbeds, patches ofscattered trees with grassyundergrowth

Multifamilyresidential areas,main transit roads

6 Kaminskiego 5.96 1945 4.0 44.48 13.62 0.98 0.77 0.51 2.36 Predominantly patches ofmanicured lawns, very smallpatches of flowerbeds, singleyoung trees with grassyundergrowth

Multifamilyresidential areas,urban green areas

7 Literacka 8.02 1818 4.0 47.12 20.78 0.68 1.52 0.53 3.58 Small patches of intensivelytrampled neglected lawns, verysmall patches of flowerbeds,predominant small young trees,local groups of scattered and densetrees

Multifamilyresidential areas

8 Orzycka 6.09 1934 5.7 48.61 15.31 0.95 0.69 0.18 2.75 Small patches of neglected lawns,very small patches of flowerbeds,groups of scattered trees

Multifamilyresidential areas,urban green areas,main transit roads,industrial areas

9 Rzymowskiego 5.95 1448 4.2 51.18 21.20 0.73 0.53 0.19 2.65 Small patches of manicured lawns,small patches of flowerbeds,groups of scattered trees

Multifamilyresidential areas,urban green areas,main transit roads,industrial areas

10 Duracza 7.61 1788 5.1 51.81 22.04 0.71 1.37 0.39 3.54 Predominantly small patches ofneglected lawns, very small singlepatches of flowerbeds, groups ofscattered trees, local groups ofdense trees


Author's personal copy334

B.Szulczewska

etal./Land

Use

Policy38

(2014)330–345

Table 2 (Continued)

No. Neighbourhood Totalarea (ha)

Number ofinhabitants

Average heightof buildings instories

Ratio ofBiologicallyVital Areas (%)

Biologicallyvital areas perInhabitant (m2)

Floor AreaRatio

Canopyarea (ha)

Shrubarea (ha)

Lawnarea (ha)

Greenery characteristics Neighbourhood (landuse within the rangeof 0.5 km)

11 Olbrachta 7.46 3093 8.7 52.48 12.66 1.24 1.21 0.23 3.76 Small patches of intensivelytrampled lawns, very small singlepatches of flowerbeds, smallpatches of dense trees


12 Kolo 7.26 1930 3.5 54.33 20.43 0.80 1.34 0.22 3.68 Large patches of neglected lawns,small patches of flowerbeds, smallpatches of dense trees, singlemature trees


13 Langego 7.53 2975 9.0 56.91 14.40 1.19 1.47 0.35 4.19 Predominantly large patches ofneglected lawns, very small singlepatches of flowerbeds, smallpatches of dense trees, singlemature trees


14 Domaniewska 5.71 2213 7.3 57.76 15.17 1.14 0.98 0.28 3.13 Small patches of neglected lawns,local intensively trampled, smallsingle patches of flowerbeds, largepatches of scattered trees


15 Bokserska 6.66 1235 3.5 58.61 31.59 0.56 1.13 0.29 3.51 Large patches of neglected lawns,local intensively trampled, small,or very small patches offlowerbeds, large patches of denseand scattered trees


16 Conrada 7.47 3138 9.1 59.58 14.18 1.26 1.34 0.42 4.14 Large patches of neglected lawns,local intensively trampled, verysmall patches of flowerbeds, largepatches of scattered trees, singlemature trees


17 Limanowskiego 7.01 1450 4.4 65.11 31.48 0.62 1.38 0.43 4.18 Predominantly large patches oflawns, very small patches offlowerbeds, large areas of youngtrees with grassy, intensivelytrampled undergrowth


18 Bernardynska 6.82 1635 9.3 67.40 28.11 0.72 1.13 0.23 4.30 Large patches of neglected lawns,local intensively trampled, smallpatches of flowerbeds, smallpatches of scattered trees,abandoned allotment gardens

Multifamilyresidential, urbangreen areas,industrial areas



Fig. 1. Location of the studied neighbourhoods.

on the methods described by Breuer et al. (2003) and Pauleit andDuhme (2000), using data on the influence of land cover formson the intensity of the hydrological processes that indicate theoutcome of the water balance equation. These processes (surfaceoutflow, evapo-transpiration and infiltration) were calculated forevery individual neighbourhood to assess the typical conditionsafter a period of intensive rainfall.

The intensity of evapo-transpiration was assessed using themean value of the Leaf Area Index (LAI) calculated for a neigh-bourhood proportional to the mean typical LAI (approximately7) for natural land with diversified natural vegetation. The studywas conducted with the assumption that in natural terrain, evapo-transpiration constitutes approximately 40% of the total rainwateroutcome (MDE, 2000).

The intensity of surface runoff was calculated by multiply-ing the percentage of paved or built surface area in the locationby the appropriate runoff coefficient according to the literature(MDE, 2000; Pauleit and Duhme, 2000). The intensity of infiltra-tion was quantified as the third component of the outcome ofwater after subtracting the intensity of surface outflow and evapo-transpiration.

The next analysis included the calculation of excessive sur-face runoff after 30 mm of rainfall. This runoff was calculatedin terms of the total runoff minus the runoff characteristic of atheoretical natural terrain with the same area as the neighbour-hood.

We also investigated whether there was sufficient undevelopedarea within the boundary of a neighbourhood for installation of

a retention pond of the required capacity, surface area and safeaverage depth (0.8 m) to detain excessive runoff.

EcologyThe ecological survey consisted of three independent groups of

variables that aimed to evaluate the ecological characteristics of aneighbourhood and their dependencies on the RBVA. These variablegroups were floristic diversity, butterfly species diversity and GreenPlot Ratio.

Floristic diversity. The survey considered areas covered by vegeta-tion, except for the vegetation on roofs and in various containers. Inmost cases, vegetation was associated with moderately maintainedlawns, areas covered by trees with undergrowth and an herba-ceous layer, ruderal vegetation and neglected ornamental gardens.Researchers randomly selected a sample area ranging from 16 m2 innon-wooded units to 100 m2 in wooded areas. In all, 126 vegetationsamples were distinguished according to the Braun-Blanquet phy-tosociological method (Dierschke, 1994). The cover and abundanceof species were also recorded according to the Braun-Blanquetscale. Based on the data collected, we used the number of speciesand Shannon’s and Simpson’s diversity index as universal measuresof diversity and naturalness.

Two naturalness indexes were also calculated: the ratio of alienspecies to all species within a sample and the percentage of nativenon-synanthropic plant species in a sample.



Fig. 2. Layout of green and sealed surfaces in surveyed neighbourhoods. (For interpretation of the references to color in this text, the reader is referred to the web version ofthe article.)



Butterfly (Rhopalocera) species richness. Butterflies are good indi-cators of urbanisation (Blair and Launer, 1997) and provide anevaluation of habitat quality. Many butterfly species are sensitive tochanges in their environment and respond more quickly to changesin habitat quality than other species, such as birds or vascular plants(Erhardt and Thomas, 1991; Thomas et al., 2004). Butterflies usu-ally have well-expressed habitat preferences and are also relativelyeasy to identify in the field (Dennis, 2010).

Field studies were performed from the end of April to the middleof September in 2008 and from the end of April to the end of July in2009. We surveyed each site with similar intensity: approximatelyevery 12 days in the first study year and every 20 days in the sec-ond (except in Kaminskiego, where studies were conducted fromApril to the end of August 2009 every 14 days). Surveys were con-ducted only on cloudless days with minimal breezes and betweenthe hours of 11.00 and 16.00. Field studies were based primarilyon observations and specimen catches using an entomological net(almost all butterfly specimens were identified alive and releasedafter examination).

Green Plot Ratio (two abbreviations are used in publications: GPR andGnPR). According to Ong (2003), the Green Plot Ratio is expressedas the ratio of the total leaf area of all plants growing in the ana-lysed area to the surface of the area. The GPR is based on a biologicalparameter known as the Leaf Area Index (LAI). The intensity of eco-logical processes is primarily dependent on biomass and the areaof active organs that are usually described (among others factorsby the LAI. Breda (2003) emphasises that the LAI values of variousplants differ depending on the species and in particular on the den-sity of the foliage and leaf size. We deemed it necessary to adjustthe LAI to Warsaw conditions.

The total GPR coefficient was calculated for all neighbourhoods,and the total area occupied by plants was taken as a sum of theprojections of crowns, shrubs, hedges, lawns and flowerbeds as wellas the area covered by climbers.

The LAI for lawns, flowerbeds and evergreen plants was takenfrom Ong (2003) and Asner et al. (2003) after verification in thefield.

It is possible to use a coefficient to calculate a tree LAI usingthe LAD (Leaf Area Density) (LI-COR, 1992; Peper and McPherson,2003). The LAI/LAD measurements were performed using LAI 2000(Li-COR) during the mid-vegetation period. Studies of the LAI wereperformed on the most common non-native trees and shrubs grow-ing in the surveyed neighbourhoods. For native species, we useddata from previous studies performed by the authors in the years2003–2005 using the same methodology. We calculated the aver-age LAI for each of the three height categories.

We calculated the average LAD for trees using the assumptionthat the tree crown is an ellipsoid of revolution of the volume. Wetook into account the height of trees and shrubs as well as the lengthof the crown. This crown length was averaged within each treeheight category, assuming that the crown base of urban trees islocated at a height of 2 m.

The LAI of climbers was calculated based on the coefficient offoliage density to the area of the supporting wall and was depend-ent on the height and width of the wall covered by a plant. Thefoliage area was related to the base (the width of the wall coveredby a plant multiplied by the layer thickness of 0.5 m).

Statistical analysisThe relationships between pairs of examined environmental

variables were analysed using Pearson’s correlation coefficient andsimple linear regression. Multivariate relationships were evaluatedusing principal component analysis (PCA). These analyses were

performed using STATISTICA 10, and the significance of the sta-tistical tests was set at a 0.05 probability level.

Results: environmental performance of the surveyedneighbourhoods

General overview

Multivariate relationships among the examined variables wereevaluated based on the results obtained in the principal compo-nent analysis (Fig. 3). Because the first principal component (PC1)accounts for 56.9% of the total multivariate variability, differencesalong the X-axis are more important than differences along the Y-axis (PC2 accounts for 14.2% of the total variability). Almost allthe variables are strongly correlated with PC1. Four of them, i.e.,CA, TNPS, LA and NBS, are strongly positively correlated (r > 0.75),and three variables are negatively correlated (FAR, ESO and AT-01)(r < −0.79). Only one variable, i.e., AT-07, showed a weak correla-tion with PC1 (r = 0.34) and a strong negative correlation with PC2(r = −0.83).

The PC1 was strongly correlated with RBVA (r = 0.94) and GPR(r = 0.78), which are both important and strongly related to sur-veyed environmental variables.

In contrast, rather weak correlations were observed betweenPC2 and RBVA and GPR (r = 0.12 and r = 0.19).

In addition to multivariate relationships, we can characterise theindividual neighbourhoods based on the PCA results (location ofpoints in Fig. 3) according to the examined variables. For example,neighbourhood No. 18 (Bernardynska) is characterised by a highNBS and Tmax-07 and a low Tmin-01, ESO and FAR.

A regression analysis was performed on the selected environ-mental variables to evaluate their relationships with RBVA and GPR(Fig. 4) together with an analysis of the correlation among RBVA,greenery structure, building coverage and Floor Area Ratio with thesurveyed environmental variables (Table 3).

Hydrological variables were omitted from these analysesbecause of their direct relationships with RBVA as a result of thecalculation method.

The results of the simple regression analysis revealed strongpositive correlations between RBVA and number of butterflies,number of plant species and Simpson’s index of biodiversity butnegative correlation with minimum air temperature in January.Similar relationships were observed for GPR, but the relation-ship with number of butterflies was weaker, and correlations withSimpson’s index of biodiversity and minimum air temperature inJanuary were stronger (the latter relationship is difficult to inter-pret because Polish conditions in winter do not fully describe thesituation in terms of GPR).

Climate

The highest temperatures were identified in the neighbour-hoods with an RBVA of less than 20%, namely, Hoza, Panska, andSandomierska (situated in a ‘downtown functional zone’ within theurban heat island), as well as in Zgrupowania Zmija (located outsidethe main UHI), which has an RBVA of 42% (the high air temperaturein this location could be explained by the high Floor Area Ratio andbuilding layouts).

Among the neighbourhoods with RBVAs of 40–50%, the col-dest were the two with the most external locations in the ‘urbanzone’: Kaminskiego (colder during the summer) and Wlodarzewska(colder during the winter). The warmest was Zgrupowania Zmija,which has already been mentioned.

The air temperatures from the base stations also indicated Bok-serska and Conrada as two other relatively cold neighbourhoods.



Fig. 3. Bi-plot of the first and the second principal components (PC1 and PC2) for the examined traits and neighbourhoods. Acronyms: FAR, Floor Area Ratio; CA, canopy area(%); SA, shrub area (%); LA, lawn area (%); NBS, number of butterfly species; TNPS, total number of plant species; ESO, excessive surface outflow; Tmin-01, mean minimumair temperature in January 2010; Tmax-07, mean maximum air temperature in July 2010. Underlined numbers indicate individual neighbourhoods (numbers are the sameas in Table 2).

These locations are older than the neighbourhoods mentionedabove (30–60 years) and have high RBVAs (at or above 60%). Theseresults imply that the general thermal conditions of neighbour-hoods are strongly correlated with their RBVA but also with theirlocation in the city and the type of area surrounding them (Fig. 1and Table 2).

The measurements also confirmed the fact that the decreasingRBVA, strongly influenced daily air temperature amplitudes in com-bination with increasing sealed surfaces (Fig. 5). The largest dailyair temperature amplitudes (those exceeding 16 ◦C in summer and5–7 ◦C in winter) occurred in Bernardynska, which is situated nextto an open green area (meaning that it warms to a greater extentduring the day and cools to a greater extent at night), and the low-est amplitudes were recorded in Hoza (8 ◦C in summer and 4 ◦C inwinter).

A decrease in RBVA corresponds to additional lawn and shrubarea and greater building coverage and Floor Area Ratio. The signif-icant increases in minimum air temperature in winter are found inTable 3.

The RBVA also shapes the relative air humidity values, whichwere lowest in Hoza (RBVA 16%) and highest in Conrada (RBVA60%), although the differences in humidity between neighbour-hoods were usually in the range of 5–7% (Fig. 5).

In addition to the general thermal conditions of the surveyedareas, the RBVA also contributed to an understanding of the climaticconditions within the neighbourhoods. Those with the lowest RBVA(situated in the ‘downtown functional zone’) are characterised notonly by the highest air temperatures (especially in winter) andthe lowest wind speeds but also by diminished air temperaturedifferences over different surfaces. In Hoza (RBVA 16.7%), the airtemperature difference reached 1.1 ◦C, whereas in Wlodarzewska(RBVA 40.7%), which has a more external location, the differencewas over 2 ◦C and reached 3 ◦C in Bernardynska (RBVA 67.4%).In Bernardynska, vast green patches and building layouts enablegood ventilation but also lead to a wind tunnel effect. Consider-able wind acceleration due to building layout was also observedin Domaniewska, Langego, Zgrupowania Zmija and especially inConrada. However, there were also cases in which no air move-ment was noted in certain parts of surveyed neighbourhoods (Kolo,Sandomierska, Wlodarzewska).

Comparison of the Kaminskiego (RBVA 44.5%) and Wlo-darzewska (RBVA 40.7%) neighbourhoods, which share similarage, RBVA, external location, and urban green spaces, yieldedinteresting results. In Kaminskiego, the building layout andopenwork fencing allow effective ventilation from the outside.Moreover, the gardens located next to neighbourhood blocks,which consist mostly of ornamental plants, ruderal vegetation andbroadleaved trees, mitigate the neighbourhood’s internal climate.Wlodarzewska, situated next to the park, is characterised by lessfavourable climatic conditions than Kaminskiego; it has brick wallsfencing the neighbourhood, which prevent the air flow from thegreen spaces outside them and create areas without ventilation,causing the air temperature to rise as well.

Hydrology

The distribution of water in each of the neighbourhoods dif-fers significantly from the distribution characteristics for naturalterrain. In most cases, superficial outflow was the predominanthydrologic process, and its share in total storm water outcome was3.4–7.8 times greater than in natural terrain (MDE, 2000).

The intensity of infiltration exceeded the intensity of out-flow in only three neighbourhoods (Bernardynska, Bokserska andLimanowskiego), all of which had notably high RBVAs of approxi-mately 60%.

In cases in which the RBVA was lower than 44.5% and the areaof sealed surfaces was quite high, the majority of storm waterran off via sewers. The intensity of the groundwater recharge wasdiminished to no more than 26% of the total water flow in theseneighbourhoods. The maximum intensity of outflow of 73.7% wascalculated for the Hoza neighbourhood with an RBVA of 16.41% andsituated in the city centre.

The vegetation coverage and density expressed by the LAI valuestrongly affects the intensity of evapo-transpiration. In neighbour-hoods with a moderate RVBA (40–44.5%) and high mean LAI values(4.4–5.4), the evapo-transpiration intensity can reach up to twiceas high as the intensity of infiltration.

The calculated volumes of excessive surface outflow after30 mm of rainfall fell within the range of 492–1411 m3. In mostcases, the possibility existed of locating a suitable retention pond



Fig. 4. Relationships (regression functions, coefficients of determination and p-values).

Table 3Correlations among RBVA, greenery structure, building coverage and Floor Area Ratio in neighbourhoods and plants, butterflies, climate, and surface outflow as expressedby different indices: (a) canopy layer, (b) undergrowth layer.

RBVA (%) Canopy area (%) Lawns (%) Shrubs (%) Building coverage (%) Floor Area Ratio (%)

Total number of plant species 0.65 0.58 0.66 0.60 -0.69 −0.64Number of plant species in sample (a) 0.30 0.14 0.29 0.06 −0.38 −0.36Number of plant species in sample (c) 0.62 0.58 0.64 0.64 −0.63 −0.59Synanthropisation index (a) 0.04 −0.26 0.03 0.05 −0.07 −0.16Synanthropisation index (c) −0.51 −0.30 −0.49 −0.39 0.53 0.34Simpson’s diversity index (c) 0.63 0.50 0.64 0.65 −0.62 −0.59Total number of butterfly species 0.73 0.41 0.70 0.19 −0.74 −0.70Average min. air temperature – January 2010 −0.64 −0.44 −0.62 −0.54 0.60 0.59Average max. air temperature – July 2010 0.20 0.03 0.22 0.12 −0.31 −0.18

Values significant at p < 0.05 are presented in bold.

inside the boundary of a neighbourhood. With an assumed requiredsafe depth of 0.8 m, a pond should have surface area between615 m2 and 1142 m2 in such a situation. The calculations showedthat the four neighbourhoods with RVBAs of less than 44.5%did not contain sufficient space to build a retention pond. Itcould be possible to apply selected extensive storm water buffersolutions, such as bioretention areas, but they would not be capa-ble of collecting and retaining all the excessive runoff from theseneighbourhoods.

Floristic diversity

A total of 204 plant species were recorded within theinvestigated neighbourhoods in Warsaw. Of these species, 61%were native plants, i.e., Euphorbia dulcis, Euphrasia stricta andRanunculus sceleratus. These plants are rarely found in Warsaw, andonly a few single specimens were recorded. Of all plant species, 58%were found in no more than of the selected neighbourhoods. Thediversity of this vegetation is determined primarily by herbaceous



Fig. 5. Air temperature and humidity characteristics of neighbourhoods with different RBVAs.

plants, whose proportion of the total number of plant species in thesample averaged 93%. The lowest floristic diversity was noted inneighbourhoods with RBVAs of approximately 20% (Sandomierska,Panska and Hoza). Those neighbourhoods are also characterisedby the lowest biodiversity and naturalness indexes. Higher RBVAvalues (ca. 40–60%) have a less clear effect on diversity. An area’sRBVA has a statistically significant effect on the total number ofplant species, Shannon and Simpson’s diversity indices and thepercentage of native plant species have negative correlations withsynanthropisation indices (Table 3). Similarly, the biodiversity andnaturalness indices were correlated with canopy area and lawns.The decline of diversity and plant naturalness is directly related tobuilding coverage (Table 3).

Butterfly species richness

A total of 29 species of butterflies (out of approximately 160living in Poland) representing six families were observed in thestudied neighbourhoods. The number of species found in particularneighbourhoods varied from 4 (Hoza and Panska) to 26 (Bernardyn-ska). Most of the butterflies observed belonged to species that arewidespread in Poland. These species develop on common herba-ceous plants. Five out of all identified species deserve additionalattention, however, because of their limited occurrence in Polandand/or relatively narrow habitat preferences. These species includethe orange tip, Anthocharis cardamines (Pieridae), the silver stud-ded blue, Plebejus argus (Lycaenidae), the short-tailed blue, Cupidoargiades (Lycaenidae), the large copper, Lycaena dispar (Lycaenidae)and the swallowtail, Papilio machaon (Papilionidae). The large cop-per, L. dispar, is listed in Annex II of the EU Habitats Directive.

The abovementioned species were identified in Bernardynska(RBVA 67%), Bokserska (RBVA 58.6%) and Rzymowskiego (RBVA51%).

The number of butterfly species was positively and significantlycorrelated with the RBVA (Table 3), but a high ratio did not explainthe number of species in every case. Only a few butterfly specieswere found in neighbourhoods with RBVAs below 20%.

Butterfly species diversity was positively correlated with theSimpson’s diversity index (r = 0.47) and the percentage of native

ruderal plants (r = 0.43) and negatively correlated with the synan-thropisation index of vegetation (r = −0.38). A negative effect ofbuilding coverage and Floor Area Ratio on the number of butterflyspecies is clearly visible in Table 3.

Green Plot Ratio

Calculation of the GPR was conducted using the LAI calculation.The average LAI of the trees was calculated based on 582 individ-ual measurements in the present study, and 2200 measurementsconducted by the authors in prior studies on native trees (Table 4).

Based on the LAI of trees and shrubs (in each of the height cat-egories) and the area occupied by plants, we calculated the GPR ofeach neighbourhood, and the values are shown in Fig. 6.

Differences in the GPR values for neighbourhoods with notablysimilar RBVAs explain the differences in the environmentalfunctioning of the surveyed neighbourhoods, especially in termsof air temperature and infiltration. Such a situation is observed,for instance, in Kaminskiego (44%) and Zgrupowania Zmija (42%),which are clearly different with respect to their GPRs (2.25 and 1.24,

Table 4LAI and LAD for the studied trees and shrubs.

Species AverageLAD

LAI/height (m)

A > 10 B 5–10 C < 5

Trees Acer negundo 2.51 16 7 2Acer platanoides* 1.59Aesculus hippocastanum 2.85Betula pendula* 1.06Fraxinus excelsior* 1.29Populus sp. 1.21Sorbus aucuparia* 1.62Tilia cordata* 2.52Average 1.8

Shrubs 26species/genera

a > 1.5 b 0.5–1.5 c < 0.53 3.5 2

* Average values follow previous studies performed by the authors in the years2003–2005.



Fig. 6. Comparison of RBVA and GPR values.

respectively). This discrepancy is caused not only by different treecovers (13% and 8%, respectively) but also by differences in the treeheights. In Kaminskiego, large trees over 10 m in height occupy 61%of the tree cover, whereas in Zgrupowania Zmija they account foronly 34% of the cover.

In certain cases, the lower GPR value coinciding with high treecover is caused by the presence of relatively small tree sizes. Thissituation is especially evident in Bokserska, where 77% of the areacovered by trees is occupied by small trees less than 5 m in heightand similarly in Duracza (81%) and Kolo (91%).

Discussion

Environmental performance

The results of our survey revealed that the most thoroughlyinvestigated variables showed a linear correlation with the RBVA.This result means that the greater the RBVA, the better the environ-mental performance conditions in the surveyed neighbourhoods.This phenomenon was particularly visible in terms of floristic andbutterfly diversity.

It was noted that RBVA does not influence plant diver-sity directly but rather reveals the biotic potential of habitats.This potential may be important to the survival of native plantspecies in urban areas (Lososová et al., 2011; Thompson et al.,2003). Of course, the diversity and naturalness of a neighbour-hood’s vegetation strongly depends on the human activity withinthe neighbourhood and the maintenance practices employed(Lundholm and Marlin, 2006; Kattwinkel et al., 2011) The pri-mary source of diversity in more floristically poor and intensivelymaintained lawns (Thompson et al., 2004) and other habitat typescommon in such neighbourhoods are the same rare species thatalways accompany a few dominating species. The numbers ofthese scarce plants increase with decreasing pressure on theenvironment from users. In contrast, wooded areas and lawnswere important to maintaining high diversity in a neighbourhood(Table 3), especially in the case of a mosaic form of vegetation(Goddard et al., 2009).

For butterfly species diversity in cities, several studies havenoted urbanisation’s negative effect on this variable (Blair and

Launer, 1997; Di Mauro et al., 2007; Knapp et al., 2008; Newand Sands, 2002). Our survey confirms that effect: few but-terfly species were found in neighbourhoods with low RBVAs,whereas the number of butterfly species was higher in neigh-bourhoods with higher RBVAs. A highly significant positive effectof area covered by vegetation on butterfly species diversity wasreported by many authors (Brown and Freitas, 2002; Di Mauroet al., 2007; Knapp et al., 2008; Smith et al., 2006). Our researchestablished a clear correlation between RBVA and the numberof butterfly species. In any case, a high RBVA did not translateinto a higher number of butterfly species in all cases. This resultcould stem from the relatively large contribution of areas cov-ered by trees (butterflies in our climate zone are associated withopen areas) and/or from the floristic impoverishment of openareas (indicating a lack of food plants for caterpillars and nectar-producing plants). The second theory seems to be more likelybecause the large proportion of lawns (which are neglected inmany cases) is highly correlated with the number of butterflyspecies, whereas the contribution of trees is of lesser importance(Table 3).

As mentioned previously, most of the butterflies found in theneighbourhoods feed on common herbaceous plants that can growalmost anywhere, even in ruderal places (but not as often in well-tended greenery). The occurrence of these plants is crucial forbutterflies, and these observations are in agreement with those ofÖckinger et al. (2009) and Gutiérrez (2005), who found that ruderalareas displayed higher butterfly species diversity than semi-naturaland traditional parks.

In this situation, it was difficult to decide on a thresholdthat might be recommended as the most appropriate for neigh-bourhood planning. The recommendation of the largest possibleRBVA does not solve the problem of applying the ratio in plan-ning practice. Rather, planners must decide on the minimal sizeof an RBVA that is certain to provide favourable conditions forenvironmental processes within a neighbourhood. To find thissolution, we concentrated on hydrological and climatic condi-tions.

In terms of hydrological conditions, there were two issues toconsider when deciding on a minimum RBVA size. The first issueis runoff intensity. The hydrological condition alterations stated



Table 5Comparison of eco-spatial indices.

Biotop area factor Berlin(BAF)

Green factor Malmö (GF) Seattle green factor (SGF) Greenery provisionSingapore (GnP)

Biologically Vital AreaRatio Poland (RBVA)

Focal point • Biotop • Biodiversity • Green infrastructure • Green Plot Ratio (GnPR) • Biologically vital area

Scoringsystem

• The ratio betweenecologically effectivesurfaces on site and totalsite area



• The ratio between totalleaf area on site and thetotal site area

• The ratio of thebiologically vital area onsite to the total site area

• Developers can choosefrom 9 elements on the“green menu:”



• Index takes into account3D volume covered byplants using the LeafArea Index (LAI)


◦ Surfaces impermeable towater and air

◦ Surfaces impermeable towater and air

◦ Landscaped area withdifferent depths of soil (2types)

• Developers can choosefrom 9 elements on the“green menu” (forlandscaped areas as wellas for roof gardens):

◦ Surfaces covered byvegetation on theground,

◦ Surfaces permeable towater and air (2 types)

◦ Surfaces permeable towater and air (2 types)

◦ Bioretention facilities ◦ Trees (3 types) ◦ Open water

◦ Surfaces covered byvegetation (3 types)

◦ Surfaces covered byvegetation (5 typesincluding areas coveredby trees and shrubs)

◦ Water features ◦ Palms (2 types) ◦ Greenery on roofs

◦ Storm water infiltrationfacilities over existingvegetation

◦ Storm water infiltrationfacilities over existingvegetation

◦ Area covered by plants inlandscaped areas (7types)

◦ Shrubs (2 types) • Individual landscapecomponents of a site areweighted from 0.5(greenery on roofs) to 1.0(surfaces covered byvegetation on theground)

◦ Vertical greenery ◦ Storm water harvesting ◦ Vertical greenery ◦ Turf Index does not take intoaccount the structure ofvegetation

◦ Greenery on roofs ◦ Open water ◦ Greenery on roofs (2types)

◦ Vertical greenery

• Individual elements areweighted from 0.0(impermeable surfaces)to 1.0 (surfaces coveredby vegetation on theground) per squaremetre

◦ vertical greenery (2types)

◦ Permeable paving (2types)

• Average LAI varies from2.0 (turf) to 4.5 (dicotshrubs)

Index does not consider thestructure of vegetation

◦ Greenery on roofs ◦ Structural soil systems • LAI value can be obtainedfrom the official onlinewebsite

• Individual elements areweighted from 0.0(impermeable surfaces)to 1.0 (surfaces coveredby vegetation on theground, open water) persquare metre

◦ Drought-tolerant plants • Up to 6 points can bescored for greenery

Index considers thestructure of vegetation

◦ Harvested water use inlandscaped areairrigation

Additional points can bereached for restorationof trees on site,conserving or relocatingexisting trees on site andusing compost recycledfrom horticultural waste

◦ Public visibility◦ Food cultivation• Individual landscape

components of a site areweighted from 0.1 (turf,vegetated paving) to 1.0(bioretention facilities)per square foot

Index considers thestructure of vegetation

Areas ofapplication

In the central area of town: Western Harbour Bo01: Commercial andmultifamily residentialzones:

Building developmentswith landscaping areas:

Potentially all areasdedicated todevelopment in land useplans

• Residential areas • All types of development • Developments containingmore than 4 dwellingunits

• Residential buildings



Table 5 (Continued)

Biotop area factor Berlin(BAF)

Green factor Malmö (GF) Seattle green factor (SGF) Greenery provisionSingapore (GnP)

Biologically Vital AreaRatio Poland (RBVA)

• Commercialareas

• Developments of morethan 4000 square feet fornon-residential uses

Non-residential buildings

• Administration andpublic services

Parking lots for more than20 cars

Technical infrastructure

Minimumrequirementsfor index valuein residentialareas

• For new structures: 0.6 0.5 0.6 50 Green Points of which amaximum of 8 are forgreenery provision

25%

For existing structures:0.3–0.6, dependent onthe degree of sealedsurface coverage

Sources Becker and Mohren (1990) Dalman (2002) Conlin (2009) Building and ConstructionAuthority (2010)

Regulation of the Ministerof Economic Planningand Construction (1996)

during the research were typical for urban catchments (Arnold andGibbons, 1996; Mitchell et al., 2003). The highest runoff intensity(73.7%) was obtained from the neighbourhood with the smallestRVBA (Hoza) and was quite similar to the values for storm waterrunoff (75%) calculated for rather densely built-up areas (Pauleitet al., 2005, Nuissl et al., 2009). Even in the neighbourhood withthe highest RVBA (Bernardynska), the runoff intensity was still high(34.1%). Draining of this excessive runoff to sewers without reten-tion may cause harmful results that are not noticeable within theboundaries of a neighbourhood but can result in serious damage ona district scale, such as in the case of receiver overload (Zevenbergenet al., 2010). Retention is recommended to solve this problem. Ouranalysis showed that storm water retention in one or two openretention ponds is possible only in neighbourhoods with RVBAsgreater than 44.5%. Of course, different storm water buffer solu-tions could be applied as well, but a retention pond was consideredto be the most versatile and effective, irrespective of local soil con-ditions.

The second issue is that of evapo-transpiration intensity, whichcan cause serious water deficiencies during dry periods. No suchproblem exists in neighbourhoods with RVBAs greater than 48% inwhich the influence of infiltration in water outcomes is relativelyhigh (above 31.7%) even when the LAI reaches 4.7 (Bernardynska);thus, water management is more sustainable in such neighbour-hoods.

To study climatic conditions, a ‘threshold’ was tracked foranalysing thermal conditions in the surveyed neighbourhoods.From the perspective of human well-being, those conditions weremuch better in Orzycka (RBVA 48.6%) than in Zgrupowania Zmija(RBVA 41.7%) or Wlodarzewska (40.7%). This range also includedKaminskiego, with an RBVA of 44.5%. Good bioclimatic condi-tions in these neighbourhoods could result from their peripherallocations, good ventilation and a considerable number of highleafy trees and not necessarily from the RBVA only. Hence, thethreshold between poor and adequate human thermal condi-tions among those particular 18 neighbourhoods most likely liesbetween the RBVA values of 41.7% and 48.6%. Of course, theseconditions depend on location within the city, building layoutand greenery structure. The survey confirmed that the neigh-bourhood layout and building shapes could create strong tunneleffects and stop air movement (Alcoforado et al., 2009; Eliasson,2000; Oke, 1987). Meteorological measurements in the neigh-bourhoods also proved that the type of fence surrounding theneighbourhood is of great importance (Makhelouf, 2009). Open-work metallic fencing was much more conducive to ventilationthan high brick walls, which did not allow infiltration of air fromthe outside.

Ratio of Biologically Vital Areas on the background of othereco-spatial indices

It is quite difficult to discuss our general findings based onsimilar research. Most publications that tackle the problem ofeco-spatial indexes concentrate on presenting their concept andapplication rules (Ahern, 2007; Cook, 2002; Hagen and Stiles, 2010;Lakes and Kim, 2012). The rationale for recommending an index sizeis not usually a matter of discussion.

Presentations of these concepts and results of our research onRBVA provide us a background for the comparison and assessmentof the role of eco-spatial indexes in the creation of urban structuresthat are desirable from an environmental point of view.

First, we should evaluate the RBVA by comparing it with othereco-spatial indices (Table 5). The RBVA can be described as the sim-plest index, which was created to assure the quantity of green areas(areas that are not ‘sealed’ or biologically vital areas). Our find-ings supported the generally shared opinion that quantity, howeverimportant it may be, should be supported by the quality inherentin the green space structure (trees, shrubs). In this case, the GPRused in Singapore should be considered as especially useful. Theapplication of the GPR led to a diversification in the vegetationvalue in neighbourhoods. The GPR value was clearly dependenton the spatial structure of plants, including the ratio and heightof trees, which are variables that are not recognised at all by theRBVA.

Other indices that use the BAF (Germany) as an initial model alsoconsider quality. These indices promote different elements thatsustain an area’s environmental performance (e.g., areas coveredby vegetation, open water, permeable paving, vegetated walls androofs, and storm water infiltration facilities over existing vegeta-tion) and promote various activities, such as protecting existingtrees, using harvested water for irrigation, or planting drought-tolerant plants.

However, it is worth mentioning that all indices other than theRBVA are implemented locally in certain cities. Only the RBVA isused all over the country.

Limitations and critical considerations of RBVA

The RBVA is easy to implement because it is exclusively basedon a quantitative approach. However, predicting the final resultsof implementation is difficult. First, the layout of buildings deter-mines the conditions for green space development. Next, plannerscan make better or, from an environmental point of view, lessunfavourable use of green space given the same RBVA size. The lay-out of biologically vital areas and the level of their fragmentation



are important for biodiversity. In the concept of ecological land usecomplementation, Colding (2007) stresses the need for clusteringdifferent types of urban green patches to maintain and enhancebiodiversity and ecosystem services. We can apply the same ideato land cover and attempt to cluster areas covered by vegetationlocated in neighbouring plots.

It was mentioned previously that vegetation structure could bea highly important factor that influences the climatic and hydro-logical conditions of a neighbourhood.

Certain considerations cited and many others important to theenvironmental performance of green spaces are not embraced inthe RBVA. However, currently, this measure is the only legally bind-ing measure in Poland that introduces green spaces for planning ofdifferent types of build-up areas.

Conclusions and recommendations

General conclusions

(1) A nearly linear correlation was observed between the RBVAand environmental performance. This finding indicates that thelarger the RBVA, the better the conditions created for environ-mental performance in the area. This finding made it difficult toestablish a reasonable threshold that could be recommended asa minimum RBVA size to guarantee ‘ecological balance’ in thearea.

(2) A ‘threshold’ between adequate and inadequate human thermalconditions was carefully calculated to fall between the RBVAvalues of 41.7% and 48.6%. The RBVAs between 44.5% and 48.00%were considered favourable for hydrological processes.

(3) In comparison to other eco-spatial indices, the Polish index(RBVA) can be described as the simplest of all indices and isalso the easiest index to implement in planning practices.

Remarks on method and results

The results presented above were achieved using a multidis-ciplinary survey of 18 existing neighbourhoods. It is important tonote that this research did not take place in laboratory conditions.Thus, not all conditions could be fully controlled. In addition to thevariables that were monitored, other factors also may have influ-enced the impact of the RBVA on environmental performance. Forinstance, the performance may depend on the spatial context of theneighbourhood. In addition, the size of the surveyed neighbour-hoods may influence the results obtained.

Additionally, the number of surveyed objects indicates a needfor care in drawing conclusions. For these reasons, the results ofour study represent only prerequisites for establishing standards.

Recommendations

(1) As mentioned previously, the problem with applying the RBVAin planning practice was the arbitrary decisions on the sizeof the RBVA undertaken by planners, often under pressurefrom developers. Additionally, we concluded that the minimumratio, which was established at 25% by an ordinance of the Min-istry of Infrastructure (on technical conditions to be met bybuildings and their locations), is not sufficient to assure envi-ronmental performance in neighbourhoods. Our research andcalculations showed that the recommended size for an RBVAshould most likely be established at a level between 40% and50%. For practical reasons, we recommend a value of 45%. Weare aware that this may be considered an arbitrary decision.However, this decision has been encouraged by empirical evi-dence gathered during our survey.

(2) Eco-spatial indices can be useful as planning tools to facili-tate green urban infrastructure implementation at the parcelscale. Depending on their concept and scope, these indicespreserve green areas and introduce different storm water andexisting greenery management solutions in built-up areas. Theindices also can be quite useful for evaluation and compari-son of attempts at environmental performance enhancementof different elements in existing urban structures, includingresidential areas.

(3) It should be noted that the application of indices offers both prosand cons. On the positive side, it forces planners and developersto respect spatial policy. On the negative side, it is impossibleto consider all the complex and specific conditions of futuredevelopment areas when setting the indices.

(4) It must be understood that eco-spatial indices cannot be theonly measures taken into consideration during the planningof green spaces because they do not consider major residen-tial needs (e.g., social interaction and recreation). These indicesare intended to protect and enhance the ecological functions ofgreen spaces in neighbourhoods and make them more valuableelements of a city’s green infrastructure.

Acknowledgements

The authors would like to thank the reviewers for their helpfulcomments. We are also grateful to Jolanta Latala, Jolanta Pawlakand Witold Pietrusiewicz from the Architecture and City PlanningDepartment of Warsaw’s City Hall for their assistance.

We also thank our colleagues: Czeslaw Wysocki from WarsawUniversity of Life Sciences, Faculty of Horticulture, Biotechnologyand Landscape Architecture for his suggestions; and MałgorzataPstragowska from Warsaw University of Life Sciences, Faculty ofHorticulture, Biotechnology and Landscape Architecture, DariuszGozdowski from Warsaw University of Life Sciences, Faculty ofAgriculture and Biology and Jerzy Kisiel from 24GIS s.c. for theirassistance with field data collection and analysis.

A research grant was provided by the Ministry of Science andHigher Education (project nr N527 0669 33) and was utilised from2007 to 2010 at the Warsaw University of Life Sciences.

References

Ahern, J., 2007. Green infrastructure for cities: the spatial dimension. In: Novotny,V., Brown, P.R. (Eds.), Cities for the Future Towards Integrated Sustainable Waterand Landscape Management. IWA Publishing, London, pp. 267–283.

Alcoforado, M.J., Andrade, H., Lopes, A., Vasconcelos, J., 2009. Application of climateguidelines to urban planning. The example of Lisbon (Portugal). Landscape andUrban Planning 90, 56–65.

Amati, M. (Ed.), 2008. Urban green belts in the 21st Century. Ashgate, London.Arnold, C.L., Gibbons, C.J., 1996. Impervious surface coverage. The emergence of a

key environmental indicator. Journal of American Planning Association 62 (2),243–258.

Asner, G.P., Scurlock, J.M.O., Hicke, J.A., 2003. Global synthesis of leaf area indexobservations: implications for ecological and remote sensing studies. GlobalEcology and Biogeography 12, 191–205.

Benedict, M.A., McMahon, E.T., 2006. Green Infrastructure. Linking Landscapes andCommunities. Island Press, Washington DC.

Building and Construction Authority, 2010. The Code for Environmental Sus-tainability of Buildings, second ed, http://www.bca.gov.sg/EnvSusLegislation/others/Env Sus Code2010.pdf

Becker, G., Mohren, R., 1990. The Biotope Area Factor as an Ecological Parameter,Berlin, http://www.stadtentwicklung.berlin.de/umwelt/landschaftsplanung/bff/download/Auszug BFF Gutachten 1990.pdf (in German).

Blair, R.B., Launer, A.E., 1997. Butterfly diversity and human land use: species assem-blages along an urban gradient. Biological Conservation 80, 113–125.

Breda, N.J., 2003. Ground-based measurements of leaf area index: a review of meth-ods, instruments and current controversies. Journal of Experimental Botany 54,2403–2417.

Breuer, L., Eckhardt, K., Frede, H.G., 2003. Plant parameter values for models intemperate climates. Ecological Modelling 169, 237–293.

Brown, K.S., Freitas, A.V.L., 2002. Butterfly communities of urban forest fragmentsin Campinas, São Paulo, Brazil: structure, instability, environmental correlates,and conservation. Journal of Insect Conservation 6, 217–231.



Byrne, J., Sipe, N., 2010. Green and Open Space Planning for Urban Consolidation –A Review of the Literature and Best Practice, Urban Research Program, IssuesPaper 11. Griffith University, Brisbane.

Caspersen, O.H., Olafsson, A.S., 2010. Recreational mapping and planning for enlarge-ment of the green structure in greater Copenhagen. Urban Forestry & UrbanGreening 9, 101–112.

Colding, J., 2007. ‘Ecological land-use complementation’ for building resilience inurban ecosystems. Landscape and Urban Planning 81, 46–55.

Conlin, R., 2009. Seattle Ordinance 122935, http://clerk.ci.seattle.wa.us/∼archives/Ordinances/Ord 122935.pdf

Cook, E.A., 2002. Landscape structure indices for assessing urban ecologicalnetworks. Landscape and Urban Planning 58, 269–280.

Dalman, E. (Ed.), 2002. Revised Quality program for Bo01 dp 4537. , http://www.malmo.se/download/18.22ccf94712cdcd54b038000909/vh bo01kvalitetsprogram rev sv.pdf (in Swedish).

Dennis, R.L.H., 2010. A Resource-Based Habitat View for Conservation. Butterflies inthe British Landscape. Wiley-Blackwell, Oxford.

Di Mauro, D., Dietz, T., Rockwood, L., 2007. Determining the effect of urbanizationon generalist butterfly species diversity in butterfly gardens. Urban Ecosystems10, 427–439.

Dierschke, H., 1994. Pflanzensoziologie. Grundlagen und Methoden. Ulmer,Stuttgart.

Dong, L., Zhiming, F., Yanzhao, Y., Zhen, Y., 2011. Spatial patterns of ecologicalcarrying capacity supply-demand balance in China at county level. Journal ofGeographical Sciences 21 (5), 833–844.

Eliasson, I., 2000. The use of climate knowledge in urban planning. Landscape andUrban Planning 48, 31–44.

Erhardt, A., Thomas, A., 1991. Lepidoptera as indicators of changes in the semi-natural grasslands of lowland and upland Europe. In: Collins, N., Thomas, J.A.(Eds.), The Conservation of Insects and Their Habitats. Academic Press, London,pp. 213–237.

Fábos, J.G., 2004. Greenway planning in the United States: its origins and recent casestudies. Landscape and Urban Planning 68, 321–342.

Gill, S.E., Handley, J.F., Ennos, R., Pauleit, S., 2007. Adapting cities for climate change:the role of the green infrastructure. Built Environment 33 (1), 115–133.

Goddard, M.A., Dougill, A.J., Benton, T.G., 2009. Scaling up from gardens: biodiversityconservation in urban environments. Trends in Ecology and Evolution 25, 90–98.

Gordon, A., Simondson, D., White, M., Moilanen, A., Bekessy, S.A., 2009. Integratingconservation planning and landuse planning in urban landscapes. Landscapeand Urban Planning 91, 183–194.

Grimm, V., Wissel, C., 1997. Babel, or the ecological stability discussions: an inven-tory and analysis of terminology and a guide for avoiding confusion. Oecologia109, 323–334.

Gutiérrez, D., 2005. Effectiveness of existing reserves in the long-term protection ofa regionally rare butterfly. Conservation Biology 19, 1586–1597.

Hagen, K., Stiles, R., 2010. Contribution of landscape design to changing urban cli-mate conditions. In: Müller, N., Werner, P., Kelcev, J.G. (Eds.), Urban Biodiversityand Design. Wiley-Blackwell, Oxford, pp. 572–592.

Hawkins, T.W., Brazel, A., Stefanov, W., Bigler, W., Saffell, E.M., 2004. The role of ruralvariability in urban heat island determination for Arizona. Journal of AppliedMeteorology 43, 476–486.

Hirst, J., Morley, J., Bang, K., 2008. Functional Landscapes: Assessing elements ofSeattle Green Factor. International Report. The Berger Partnership PS, http://www.seattle.gov/dpd/cms/groups/pan/@pan/@permits/documents/webinformational/dpdp016505.pdf

Hostler, M., Allen, W., Meurk, C., 2011. Conserving urban biodiversity? Creatinggreen infrastructure is only the first step. Landscape and Urban Planning 100,369–371.

James, P., Tzoulas, K., Adams, M.D., Barber, A., Box, J., Breuste, J., Elmqvist, T., Frith,M., Gordon, C., Greening, K.L., Handley, J., Hawoth, S., Kazimierczak, A.E., John-ston, M., Korpela, K., Moretti, M., Niemelä, J., Pauleit, S., Roe, M.H., Sadler, J.P.,Thompson, C.W., 2009. Towards an integrated understanding of green space inthe European built environment. Urban Forestry and Urban Greening 8, 65–75.

Joly, D., Nilsen, L., Fury, R., Elvebakk, A., Brossard, T., 2003. Temperature interpolationat a large scale: test on a small area in Svalbard. International Journal Climatology23, 1637–1654.

Jongman, R.H.G., Külvig, M., Kristiansen, I., 2004. European ecological networks andgreenways. Landscape and Urban Planning 68, 305–319.

Kattwinkel, M., Biedermann, R., Kleyer, M., 2011. Temporary conservation for urbanbiodiversity. Biological Conservation 144, 2335–2343.

Knapp, S., Kühn, I., Mosbrugger, V., Klotz, S., 2008. Do protected areas in urban andrural landscapes differ in species diversity? Biodiversity and Conservation 17,1595–1612.

Lakes, T., Kim, H., 2012. The urban environmental indicator biotope area ratio – anenhanced approach to assess and manage the urban ecosystem services usinghigh resolution remote-sensing. Ecological Indicators 13, 93–103.

LI-COR, 1992. LAI–2000 Plant Canopy Analyzer. Operating Manual. LI–COR Inc, Lin-coln.

Lososová, Z., Horsák, M., Chytry, M., Cejka, T., Danihelka, J., Fajmon, K., Hájek, O.,Juricková, L., Kintrová, K., Láníková, D., Otypková, Z., Rehorek, V., Tichy, L., 2011.Diversity of central European urban biota: effects of human-made habitat typeson plants and land snails. Journal of Biogeography 38, 1152–1163.

Lundholm, J.T., Marlin, A., 2006. Habitat origins and microhabitat preferences ofurban plant species. Urban Ecosystems 9, 139–159.

Makhelouf, A., 2009. The effect of green spaces on urban climate and pollution.Iranian Journal of Environmental Health Science and Engineering 6 (1), 35–40.

Maruani, T., Amit-Cohen, I., 2007. Open space planning models: a review ofapproaches and methods. Landscape and Urban Planning 81, 1–13.

Matzarakis, A., Rutz, F., Mayer, H., 2007. Modelling radiation fluxes in simple andcomplex environments-application of the RayMan model. International Journalof Biometeorology 51, 323–334.

MDE, 2000. Maryland Stormwater Design Manual I & II. Maryland Departmentof the Environment Water Management Administration, Baltimore, http://www.mde.state.md.us/programs/Water/StormwaterManagementProgram/MarylandStormwaterDesignManual/Pages/programs/waterprograms/sedimentandstormwater/stormwater design/index.aspx

Mitchell, V.G., McMahon, T.A., Mein, R.G., 2003. Components of the total water bal-ance of an urban catchment. Environmental Management 32, 735–746.

New, T., Sands, D.P.A., 2002. Conservation concerns for butterflies in urban areas ofAustralia. Journal of Insect Conservation 6, 207–215.

Nuissl, H., Haase, D., Lanzendorf, M., Wittmer, H., 2009. Environmental impactassessment of urban land use transitions – a context-sensitive approach. LandUse Policy 26, 414–424.

Öckinger, E., Dannestam, A., Smith, H.G., 2009. The importance of fragmentation andhabitat quality of urban grasslands for butterfly diversity. Landscape and UrbanPlanning 93, 31–37.

Oke, T.R., 1987. Boundary Layer Climates, second ed. Methuen, London/New York.Ong, B.L., 2003. Green plot ratio: an ecological measure for architecture and urban

planning. Landscape and Urban Planning 63, 197–211.Opdam, P., Steingröver, E., van Rooij, S., 2006. Ecological networks: a spatial con-

cept for multi-actor planning of sustainable landscapes. Landscape and UrbanPlanning 75, 322–332.

Pauleit, S., Duhme, F., 2000. Assessing the environmental performance of land covertypes for urban planning. Landscape and Urban Planning 52, 1–20.

Pauleit, S., Ennos, R., Golding, Y., 2005. Modeling the environmental impacts of urbanland use and land cover change – a study in Merseyside, UK. Landscape andUrban Planning 71, 295–310.

Peper, P.J., McPherson, E.G., 2003. Evaluation of four methods for estimating leafarea of isolated trees. Urban Forestry and Urban Greening 2 (1), 19–29.

Puppim de Oliveira, J.A., Balabana, O., Doll, C.N.H., Moreno-Penaranda, R., Gasparatos,A., Iossifova, D., Suwa, A., 2011. Cities and biodiversity: perspectives and gover-nance challenges for implementing the convention on biological diversity (CBD)at the city level. Biological Conservation 144, 1302–1313.

1996. Regulation of the Minister of Economic Planning and Construction of 4April 1996 amending Regulation on the Technical Conditions to be metby Buildings and their Location, http://isap.sejm.gov.pl/DetailsServlet?id=WDU19960450200 (in polish).

Self, P., 2002. The evolution of the Greater London Plan 1944–1970. Progress inPlanning 57, 145–175.

Smith, R.M., Warren, P.H., Thomson, K., Gaston, K.J., 2006. Urban domestic gardens(VI): environmental correlates of invertebrate species richness. Biodiversity andConservation 15, 2415–2438.

Thomas, J.A., Telfer, M.G., Roy, D.B., Preston, C.D., Greenwood, J.J.D., Asher, J., Fox, R.,Clarke, R.T., Lawton, J.H., 2004. Comparative losses of British butterflies, birds,and plants and the global extinction crisis. Science 303, 1879–1881.

Thompson, C.W., 2002. Urban open space in the 21 century. Landscape and UrbanPlanning 60, 59–72.

Thompson, K., Austin, K.C., Smith, R.M., Warren, P.H., Angold, P.G., Gaston, K.J., 2003.Urban domestic gardens (I): putting small-scale plant diversity in context. Jour-nal of Vegetation Science 14, 71–78.

Thompson, K., Hodgson, J.G., Smith, R.M., Warren, P.H., Gaston, K.J., 2004. Urbandomestic gardens (III): composition and diversity of lawn floras. Journal of Veg-etation Science 15, 373–378.

Tratalos, J., Fuller, R.A., Warren, P.H., Davies, R.G., Gaston, K.J., 2007. Urban form,biodiversity potential and ecosystem services. Landscape and Urban Planning83, 308–317.

Turner, T., 1992. Open space planning in London: from standards per 1000 to greenstrategy. Town Planning Review 63 (4), 365–386.

Turner, T., 1995. Greenways, blueways, skyways and other ways to a better London.Landscape and Urban Planning 33, 269–282.

Zevenbergen, C., Cashman, A., Evelpidou, N., Pasche, E., Garvin, S., Ashley, R., 2010.Urban Flood Management. CRC Press/Balkema, New York.

How much green is needed for a vital neighbourhood

Documents