• Like
Land urbanization in Central Italy: 50 years of evolution
Upcoming SlideShare
Loading in...5

Thanks for flagging this SlideShare!

Oops! An error has occurred.

Land urbanization in Central Italy: 50 years of evolution


17/1/2013 - Articolo di Bernardino Romano and Francesco Zullo …

17/1/2013 - Articolo di Bernardino Romano and Francesco Zullo
The purpose of this study is to elaborate data on the quantitative and qualitative evolution of land urbanization in Central Italy from the post-war period (1949–1956) to the first decade of the twenty-first century.

Published in News & Politics
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads


Total Views
On SlideShare
From Embeds
Number of Embeds



Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

    No notes for slide


  • 1. Journal of Land Use ScienceVol. 00, No. 00, Xxxx 2012, 1–22 Land urbanization in Central Italy: 50 years of evolution Bernardino Romano* and Francesco ZulloDepartment of Civil Engineering, Architecture and Environment (DICEAA), University of L’Aquila, Campo di Pile, L’Aquila, Italy (Received 29 June 2012; final version received 28 November 2012) 5 The increasingly fast pace in urban conversion of land over the past 50 years in Italy is a phenomenon that is still difficult to quantify reliably owing to the chronic lack of knowledge at every territorial level, from national to municipal. This article describes the results of a study on the features of urbanization in the 1950s in the peninsular regions of Central Italy, based on uniform historical maps of the entire country. The 10 historical data were compared from a qualitative and quantitative viewpoint with the present-day geography of settlements. Interesting information has emerged on possi- ble significant thresholds in the relationship between demography and urban use of land, in addition to data on landscape effects to be construed as signs of specific trends underway today and scarcely taken into account by land management tools. 15 Keywords: urbanization impact; land-use change; land-use planning; land uptake1. IntroductionThe purpose of this study is to elaborate data on the quantitative and qualitative evolutionof land urbanization in Central Italy from the post-war period (1949–1956) to the firstdecade of the twenty-first century. 20 According to consolidated scientific opinions, land use caused by urbanization is one ofthe main causes of political and social conflicts (Plotkin, 1987) and altered environmentalquality of land (Ellis & Ramankutty, 2008; Sala et al., 2000). The aspects involved, either directly or indirectly, in urban conversion of land includethe following: 25Economic and energetic field: • diseconomies in transport • waste of energy • reduced agricultural produceHydro-geo-pedologic field: 30 • geological destabilization • irreversible use of land • alteration of underground and surface watercourses.*Corresponding author. Email: bernardino.romano@univaq.itISSN 1747-423X print/ISSN 1747-4248 online© 2012 Taylor & Francishttp://dx.doi.org/10.1080/1747423X.2012.754963http://www.tandfonline.com
  • 2. 2 B. Romano and F. ZulloPhysico-climatic field: • increased thermal reflection and climate change 35 • reduced emission absorption capacity • effects on carbon sequestration • spatial propagation of physicochemical disturbancesEcobiological field: • physical erosion and destruction of habitats 40 • ecosystem fragmentation • dystrophy of ecological and biological processes • penalization of the ecosystem services of the environment • reduction in overall ecological ‘resilience’Some European countries that are more sensitive to the foregoing consequences have 45already adopted regulations to curb the phenomenon and one of the most significant expe-riences is the German one (Bundesregierung, 1985; Illy, Hornych, Schwartz, & Rosenfeld,2009). There are not many studies in the international literature on the Italian case (Bonifazi& Capello, 2001; Heins, 2001) and very few regions (only 3 out of 20) have vector infor- 50mation on land use over the past 50 years. There are also very few cases in which datacovering a shorter time span are available (1970s–2000) regarding both overall magnitudeof urbanization and statistically significant historical series and there are no coordinatedsurvey activities among local governments (regions, provinces, and municipalities). The negative aspects of this phenomenon are still considered only marginally by sci- 55entific, communication, and land governance agencies (Emiliani, 2007; Mercalli, 2009;Pileri, 2007). Some contributions on ecological networks and the threats on the biodiversity dueto urbanization are in Battisti (2011). Some Universities (Turin, Florence, Pavia, andL’Aquila) have studied the relationship between land-use change and decreasing of 60ecosystem quality (Gambino & Ramano, 2004, Malcevschi, 2010; www.planeco.org). In other countries, information and publications are far more numerous (Bockstael,2007; Garcia-Call, 2011; Hall, Gracey, Drewett, & Thomas, 1973; Hauri, Steiner, &Vinzens, 2006; Irwin & Zaninetti, 2006; Mellor, 1983; Yanitsky, 1986). Only recently hasthe need emerged to set up mechanisms to monitor urban transformation dynamics, but we 65are still far from having systematic and consistent data collection that would make credi-ble comparisons and assessments possible (Batty, 2008; Lowry, 1990; Sharma, Pandey, &Nathawat, 2012). One of the most recent initiatives in this regard is the one taken by the National LandUse Observatory, set up by the Milan Polytechnic and INU (Urban Planning National 70Institute) (http://www.inu.it/attivita_inu/ONCS_2.html) which however makes use of rel-atively recent information bases (end of the 1970s) and has very little data available.Toward the end of the 1980s, other authors highlighted landscape change modalities usinglarge-scale national databases (1:250,000) (Astengo & Nucci, 1990; CNR-IPRA, 1988) orEuropean standards such as Corine Land Cover (CLC) (APAT, 2005; Bossard, Feranec, & 75Otahel, 2000; Comber, 2008) derived from satellite remote sensing on a nominal scale of1:100,000 (Berdini, 2009; Falcucci & Maiorano, 2008). Some international organizations,
  • 3. Journal of Land Use Science 3Table 1. Differential test based on the surveys of urbanized areas taken from CLC and thoseobtained from technical maps or regional land use maps (CUS) after 2000. Urbanized area Urbanized area differenceRegions CLC 2000 (ha) CUS 2000 (ha) %Abruzzo 23,250.13 36,740 0.58Campania 71,145.88 101,163.93 0.42Emilia-Romagna 97,985.00 206,369.06 1.11Friuli-Venezia Giul 49,529.01 69,719.61 0.41Lazio 88,415.77 132,078.31 0.49Liguria 18,900.41 31,047.42 0.64Lombardia 215,942.87 304,765.8 0.41Marche 38,634.30 50,580.37 0.31Molise 4286.21 12,028.05 1.81Puglia 71,797.89 128,190.03 0.79Sardegna 54,130.35 78,061.88 0.44Umbria 23,989.49 30,124.74 0.26Valle d’Aosta 3237.94 4709.3591 0.45Veneto 132,041.36 246,298.58 0.87Total 893,286.63 1,431,877.14 0.60such as the European Environmental Agency (EEA, 2006), have estimated, in the case AQ1of Italy, that approximately 8000 ha/year were artificialized between 1990 and 2000, againbased on CLC satellite remote sensing. Other data have been processed by ISTAT (National 80Institute of Statistics) and the information derives from the census which this Institute con-ducts nationwide (ISTAT, 2009) and, in particular, referred to businesses operating in theagricultural sector. Table 1 shows a differential test on urbanized areas conducted in some Italian regionsbased on CLC satellite remote sensing and data taken from regional land use maps (Carte 85regionali di Uso del Suolo – CUS) drawn up over the past 10 years by all Italian regions bymeans of photo interpretation with levels of detail ranging between 1:10,000 and 1:5000(far more reliable than CLC as they plot scattered and minute built-up areas too). As maybe noted, there is considerable variability and huge differences, sometimes in excess of80% too, which force technical and scientific operators to be extremely cautious in making 90diagnoses and outlining problematic scenarios. Out of the sample of 14 regions shown inTable 1, actual urbanized areas (surveyed by CUS) are 60% greater on average than thesame areas surveyed by CLC remote sensing. In our article, urbanized areas in the 1950s have been extrapolated from uniform his-torical maps of the entire country, using an appropriate GIS technique, in order to obtain AQ2 95comparable values for all Italian regions. Two years of studies were needed to producethese data and we expect to obtain data for the entire country in 3 years time.2. MethodologyAs mentioned previously, in order to develop a reliable national evolutionary picture ofurban conversion of land from the second post-war period to the present day, it is necessary 100to use cartographic bases that are homogenous for the period across the entire countryand with a sufficient scale of detail to highlight even the most scattered and fragmentedurbanized parts.
  • 4. 4 B. Romano and F. Zullo This is why, in the study presented in this article we used Italian maps published on ascale of 1:25,000 by the Italian Military Geographical Institute (IGMI) between 1949 and 1051962, as our reference time source for the 1950s. These maps are part of the 25V Series,plotted on a scale of 1:20,000, organized in 3545 elements (tables) 7 30 longitude and5 latitude in size, in Gauss’s conformal representation and part of the national geodeticsystem (international ellipsoid oriented to Rome Monte Mario – ED40) with a kilometricgrid in the Universal Transverse of Mercator conformal projection (ED50 European data). 110 From these maps, available in raster version, it is possible to extrapolate urbanizedareas in the 1950s, formed by areas covered by buildings plus all ancillary areas (parkingareas, streets within neighborhoods, goods storage and handling areas, and various otherstructures) (Figure 1). Within the framework of this study, we developed a GIS technical procedure capa- 115ble of semi-automating vector extraction of urbanized areas from historical raster maps(Romano & Zullo, 2010). Then, the overall parts corresponding to urban functions (builtFigure 1. Detail of the representation of an urban area on the IGM 1:25,000 map of the 1950s.
  • 5. Journal of Land Use Science 5and ancillary areas) were isolated and tested using additional topological devices. Theresults obtained from this study on 1:25,000 maps were then compared with those of urban-ized areas available in vector format from regional maps generally derived from photo 120interpretation on a nominal scale of 1:10,000 or 1:5000. The regions presented in this article are those of Central Italy, namely Umbria, Marche,Abruzzo, Molise, and Lazio. The territories in these regions include the highest mountain-ous areas in the Apennine, with the highest peak in peninsular Italy (the Corno Grande2914 m above sea level), as well as considerable hilly expanses and over 350 km of coastal 125strip. This is why the relationship between morphology and settlements is so close and hasgiven rise to historically highly diversified phenomena, which however have become morehomogenous since the post-war period to the present day, as described later.3. Results3.1. Urbanization in the 1950s 130The regions analyzed have a very diverse territorial history. Prior to the unification of Italyin 1861, Marche, Umbria, and northern Lazio were part of the State of the Church, whilethe Abruzzo and Molise and some areas of southern and eastern Lazio belonged to theKingdom of Naples. Geoclimatic features and political administration have had a varied impact on land 135transformation from both a town planning and a agricultural perspective. For centuries,the prevailing hilly morphology of Umbria and Marche, the abundant watercourses, and aparticularly favorable temperate climate have fostered profitable farming based above allon the specialized production of wine and oil. Farming in these regions was marked bythe sharecropping model (Reid, 1975; Shaban, 1987; Singh, 1989) which was widely used AQ3 140in other Italian regions too, such as Tuscany, Emilia Romagna, and Veneto and stronglydeveloped in the fourteenth and fifteenth centuries. It persisted until the 1960s, when lawno. 756/64 put an end to new sharecropping contracts. This form of agricultural production, through which the landowner and the farmer teamup to farm an estate in order to share produce and profits, has led to very evident results 145in the landscapes of the regions in question. The utmost care of the land, made evident bythe accurate layout of fields and canals, is the consequence of the sharecropper’s need toproduce everything his family needs by tapping the potentialities of the estate as much aspossible. The average expanse of farming estates ranging between 6 and 9 ha (250–300 msquare block), where the house of the sharecropper’s family and outbuildings were situated, 150has led to the significant spread of settlements with highly scattered buildings that clearlydistinguish the regions of Umbria and Marche from all the others considered in the studysample. In the Abruzzo, Molise, and Lazio farming was historically based on an entirely differ-ent model, that of the latifundium (Cardoza, 1994), that is to say large landed estates, 155sometimes even in excess of 1000 ha (the famous record of 14,000 ha being held bythe Torlonia family in Abruzzo) owned by nobles or the ecclesiastic aristocracy used tofarm traditional crops, but more frequently as grazing land, by the farmhands hired by thelandowners. In Western Europe, this system, typical of the Ancien Régime, was overcome only 160between the eighteenth and nineteenth centuries, due to the spread of capitalism in therunning of farms. In these large landed estates, not much attention was paid to innovation and the ownerwas only interested in obtaining a good rent and neglected farming details. This is why
  • 6. 6 B. Romano and F. Zullothe type of farming practiced in large landed estates was often extremely backward. This 165latifundistic model did not require a dense system of settlements on the territory and, as aresult, urbanized centers are historically far more clustered. This phenomenon is very clear when examining Figure 2 and represented analyticallyby the urban dispersion index (UDI), which is calculated as follows: Nuc UDI = Awhere Nuc is the number of urbanized nucleus and A is the regional area (km2 ). 170 Table 2 shows a very high UDI in the 1950s for Umbria and Marche (values rang-ing between 4 and 7), while it is far lower in the regions of Lazio, Abruzzo, and Molise(between 1.6 and 1.3). In the 2000s, the regions of Umbria and Marche are still markedby dispersion, with values significantly on the rise too, while the other regions showed aclustering trend of urbanized areas, mainly linear along the valley lines, with reductions 175even in excess of one-half of the original urban nucleus.3.2. Comparison between the 1950s and the year 2000The regions studied are marked by great differences in urbanization between the 1950s andafter the year 2000 for reasons tied undoubtedly to the aforementioned prevailing territorialpolicies over the years and their climatic, geographical, morphological, and productive 180features. The variations in question are, however, quantitatively significant (Figure 2 andTable 3), with minimum increase rates of about 100% in the case of Umbria, over 200% inMarche, and over 400% in Lazio, Molise, and Abruzzo. In Molise, one of the smallest Italian regions, land has been urbanized at an averagepace of over half a hectare per day, while in Umbria the rate of conversion was slightly lower 185than 1 ha/day. Abruzzo and Marche showed decisively higher rates with over 1.7 ha/day,while Lazio (the region containing Rome) ranked first with over 5 ha/day. Overall in the regions studied, urban areas have increased threefold on average withan increment of almost 200,000 ha in approximately 50 years, that is to say an artificial-ized surface equal to that of a large European national park. Daily land uptake refers to 190varying survey periods, but the average value amounts to about a total of over 10 ha/day.Considering that the territories in the regions analyzed account for 17% of the entire coun-try, if these values were extended to 100% of the Italian territory, we would have an averagedaily conversion surface of about 60 ha/day. If this land transformation trend should per-sist in the future too, we would have a scenario of almost 500,000 ha of built surfaces in 195the next 20 years, which may be roughly represented as a square of over 70 km per side(almost half of the average width of the Italian peninsula between the Tyrrhenian sea andthe Adriatic sea). Table 4 gives us a very effective dynamic image of the actual scope of the processanalyzed. In the last post-war period, the regions studied had very low urbanization density 200values: Molise, Abruzzo, and Marche were below 7‰, while the other regions rangedbetween 1% and 2%. All these values varied their order of magnitude significantly in the50 years observed: Umbria, Abruzzo, and Molise reached 3%, Marche rose and exceeded5%, while Lazio came close to 8%. As far as the per capita value is concerned, in Moliseand Abruzzo it has increased almost fivefold, while the lowest values are found in Umbria 205with an increment factor of less than 3 (even though in the 1950s these two regions had thehighest per capita values).
  • 7. Journal of Land Use Science 7(a) Pesaro Urbino Ancona Macerata Perugia Ascoli piceno km Teramo 0 200 400 Terni Pescara Viterbo Laquila Chieti Rieti Roma Frosinone Isernia Campobasso Latina Urbanized area 1950s 0 20 40 80 120 160 km(b) Pesaro Urbino Ancona Macerata Perugia Ascoli piceno Teramo Terni Pescara Viterbo L’aquila Chieti Rieti Roma Frosinone Isernia Campobasso Latina Urbanized area noughties 0 20 40 80 120 160 km Figure 2. Urbanization maps of the (a) 1950s and of (b) 2000. In the overall territory investigated which, as mentioned earlier, is a sample accounting for over 17% of the entire national surface, urbanization density has increased fivefold on average between the mid-twentieth century and the early twenty-first century, while the per 210 capita values of the areas converted to urban functions have increased threefold.
  • 8. 8 B. Romano and F. ZulloTable 2. Territorial dispersion index of urban nucleus. Number urbanized Number urbanized UDI UDIRegions Area (km2 ) centers (1950s) centers (noughties) (1950s) (noughties)Umbria 8461.07 35,109 58,892 4.15 6.96Marche 9749.54 74,487 87,675 7.64 8.99Lazio 17,226.28 28,302 16,886 1.64 0.98Abruzzo 10,830.15 15,366 6057 1.42 0.56Molise 4461.03 5720 7553 1.28 1.69Total and 50,728.07 158,984 177,063 3.13 1.11 averageTable 3. The analytical result of the study for the sample of Italian regions investigated. Urbanized Difference Land uptake area 1950s Urbanized area 1950s- Increasing speedRegions (ha) noughties (ha) noughties rate (m2 /day)Umbria 15,750.51 30,124.74 14,374.23 0.91 8561 (1956–2002)Molise 2316.00 12,028.05 9712.05 4.19 5784 (1956–2002)Abruzzo 7242.98 36,740.00 29,497.02 4.07 17,568 (1956–2002)Marche 16,454.41 50,580.37 34,125.96 2.07 17,641 (1954–2007)Lazio 26,356.00 132,078.31 105,722.31 4.01 53,639 (1956–2004)Sum and 68,119.90 261,551.47 193,431.57 3.05 103,193 averageTable 4. Regional and per capita urbanization density values in the time span considered. Urbanization per Urbanization capita Resident inhabitants density (%) (m2 /inhabitant) RegionalRegions area (km2 ) 1951 2001 1950s Noughties 1950s NoughtiesUmbria 8461.07 803,918 853,676 0.019 0.036 195.92 352.88 (1956–2002)Molise 4461.03 406,823 347,628 0.005 0.027 56.93 346.00 (1956–2002)Abruzzo 10,826.99 1,277,207 1,328,832 0.007 0.034 56.71 276.48 (1956–2002)Marche 9727.7 1,362,863 1,469,826 0.017 0.052 120.73 344.12 (1954–2007)Lazio (1956–2004) 17,206.403 3,340,798 5,089,737 0.015 0.077 78.89 259.50Sum and average 50,683.193 7,191,609 9,089,699 0.013 0.045 101.84 315.80
  • 9. Journal of Land Use Science 9 It is interesting to observe that, although per capita values of urbanization were verydifferent in the 1950s, there was a realignment of all the regions at about 300 m2 /inhabitantin the 2000s, thus marking an urbanized area requirement constant at regional level incontemporary society. In confirmation of this, the per capita quantity in the year 2000 has a 215standard deviation of 13% of the average for the five regions, while this deviation exceeded57% of the average in the 1950s and moreover seems to be independent from demographicchanges and the varying number of industries and services present in the regions. As amatter of fact, in the 50 years considered, there has been substantial demographic stability(Table 4) in Umbria, Marche, Abruzzo, and Molise (the average yearly rate is about 1‰, 220while Molise touched −3%). The only region with a significant rise in population has beenLazio (with an average yearly rate of 8%), but in this regard, it is important to clarify that fordecades the region containing Rome has received a larger number of migrants from Centraland southern Italy, attracted by the many job opportunities offered by the metropolitan areaof the capital. Despite these differences, the present-day per capita urbanization value has 225reached similar levels. The correlation analysis carried out on a municipal basis between urbanization andpopulation variation rates in the 50 years considered (Figure 3) still shows significant dif-ferences between the two groups of regions Marche–Umbria and Abruzzo–Lazio–Molise.In the first group, common aspects include the scarce dependence between the two vari- 230ables (R2 not greater than 0.27) and limited increase in urbanization (between 1% and1.5%) in the case of demographic stability during that time span. In the other three regions, the regression curve is very different: the dependencebetween the variables (R2 between 0.31 and 0.37) is significantly higher, while the increasein urbanization corresponding to a zero-rate demographic rise is of about 5%. In these 235cases, the angular coefficient of the regression curves is much higher, suggesting that evenmodest demographic variations have been followed by significant increases in urbanizedareas. This confirms what we stated earlier regarding the ‘run-up’ phenomenon of percapita levels of urbanization that has occurred in the group of three regions Abruzzo–Lazio–Molise and that in half a century has led them to conform with the ‘standard’ average 240level of 300 m2 /inhabitant of urbanization which Marche and Umbria were far closer toalready back in the 1950s. The regions considered are formed by 1155 municipalities of which the majority, asevidenced by Figure 3, have witnessed a demographic decline in the 1956–2000 period.In the entire territory examined, only two of these municipalities in the 1950s were urban- 245ized beyond 20%, while approximately 700 did not even reach 1%. After the year 2000,municipalities urbanized beyond the 20% threshold were 32, while over 350 were around5%, and only 175 remained below the level of 1% (Figures 4 and 5). Figure 6 shows that the phenomenon is more intense in some inland areas of Lazio andAbruzzo, but above all along the coasts of the Tyrrhenian and Adriatic seas. Considering 250a coastal strip 500 m in width, over the past 50 years, the overall urbanization growthvalues along it amount to almost 350% for the Tyrrhenian side and approximately 300%for the Adriatic side, thus evidencing the important transformations that have occurredalong the coasts. In the last post-war period, urbanization along the coastline of the Lazioregion barely exceeded 1500 ha, but the situation changed drastically after the year 2000, 255when urbanized areas along the coast reached almost 5500 ha. Analyzing the 390 kmcoastline of Molise, Abruzzo and Marche from a historical perspective has shown thaturbanization along the coastal strip has risen from just over 10% in the 1950s to over 30%today, with Marche reaching a peak of 41% (Table 5). It is also interesting to note that thedaily urbanization rate of this strip over the past 50 years, assessed as surface and linear 260
  • 10. 10 B. Romano and F Zullo . (a) Umbria region R2 = 0.2746 Urbanization variation rate –0.3 –0.2 –0.1 0 0.1 0.2 0.3 0.4 Demographic variation rate (b) Molise region R2 = 0.3705 Urbanization variation rate –0.3 –0.2 –0.1 0 0.1 0.2 0.3 Demographic variation rate (c) Lazio region R2 = 0.3447 Urbanization variation rate Area del tracciato –0.3 –0.2 –0.1 0 0.1 0.2 0.3 0.4 Demographic variation rateFigure 3. Regional comparative analysis of the percentage increase in urbanized areas between the AQ151950s and the 2000s and the corresponding demographic variation index on a municipal basis.
  • 11. Journal of Land Use Science 11 (d) Abruzzo region R2 = 0.3199 Urbanization variation rate –0.3 –0.2 –0.1 0 0.1 0.2 0.3 0.4 Demographic variation rate (e) Marche region R2 = 0.1401 Urbanization variation rate –0.3 –0.2 –0.1 0 0.1 0.2 0.3 0.4 Demographic variation rateFigure 3. (Continued).distance, is very similar (about 2100 m2 /day and around 2 km/year) on both the Adriaticand Tyrrhenian coasts. In addition to the coastal strips in the regions of Central Italy, urban growth has alsohad a significant impact on inland low-hilly areas and basins where the main towns of theprovinces are situated that have always sought economic conditions comparable to those 265of coastal areas (Figure 6). The policies pursued to this end for decades have always ledto an excessive number of public and private services and regular incentives for residentialand productive building activities, regardless of the population load and what the places inquestion were actually suitable for. Regarding demographic trends, Figure 7 shows the distribution, on a municipal basis, 270of the values of the demo-urban increment index (DUI). This parameter has been obtainedas follows: urb(01−51) DUI = m2 /inhabitant pop(01−51)
  • 12. 12 B. Romano and F Zullo . Pesaro <1% Urbino 1–5% Ancona 5–10% Macerata 10–20% >20% Perugia Ascoli piceno Teramo Terni Pescara Viterbo Laquila Chieti Rieti Roma Frosinone Isernia Campobasso Latina 0 20 40 80 120 160 kmFigure 4. Map of percentage of urbanization in municipalities in the 1950s. Pesaro <1% Urbino 1–5% Ancona 5–10% Macerata 10–20% >20% Perugia Ascoli piceno Teramo Terni Pescara Viterbo Laquila Chieti Rieti Frosinone Isernia Campobasso Latina 0 20 40 80 120 160 kmFigure 5. Map of percentage of urbanization in municipalities in the 2000s.
  • 13. Journal of Land Use Science 13 Pesaro Urbino <100% Ancona 100–200% Macerata 200–350% 350–500% Perugia 500–800% Ascoli piceno 800–1200% 1200–2500% Teramo Terni >2500% Pescara Viterbo Laquila Chieti Rieti Roma Frosinone Isernia Campobasso Latina 0 20 40 80 120 160 kmFigure 6. Increased rate of urbanized areas between the 1950s and 2000.where urb(01–51) is the difference between urbanized areas in municipalities between the1950s and the early 2000s and pop(01–51) is the variation in the population residing inmunicipalities between the 1950s and the early 2000s. 275 Figure 7 shows only the positive DUI values, i.e. the values of those municipal ter-ritories where stable or increased urbanized surfaces are matched by a population rise,and a classification based on per capita increased urbanized surfaces (the amount of landconsumed per inhabitant acquired by the municipality in the period considered). The mapshows that average-sized and big cities are mostly in this condition (main cities of provinces 280or regions) and their surrounding hinterland, as well as coastal strips owing to the drivingeffects of agricultural, industrial, and tourist economies. Almost all the areas resulting fromthe aforementioned selection had DUI values at the highest levels of the assessment scale(over 500 m2 /inhabitant in urbanized surface increase). Another index examined, complementary to the DUI, is the demo-urban contradiction 285index (DUC), which is calculated as follows: urb(01−51) DUI = m2 /inhabitant lost − pop(01−51)where urb(01–51) is the difference between urbanized areas in municipalities between the1950s and the 2000s and − pop(01–51) is the demographic drop in municipalities betweenthe 1950s and the 2000s. In this case, we selected only municipalities with a negative demographic balance 290between 1951 and 2001 and calculated the increase in urbanized land between the 1950sand the year 2000, later checking the quantity of the urbanized land corresponding to every
  • 14. 14Table 5. Differences per region in urbanized areas along coastal strips having a width of 500 m. Urban area increasing Coastal belt Urbanized area Urbanized area Urbanized area Urban % Urban % 1950s-noughties Land uptakeRegions area (ha) 1950s (ha) noughties (ha) increasing (ha) 1950s noughties (%) speed (m2 /day)Marche 8721.09 1414.23 3647.57 2233.34 16.22 41.82 257.92 1154.48Abruzzo 6209.94 292.15 1472.18 1180.03 4.07 23.71 503.91 702.82Molise 1782.81 59.87 272.03 212.16 3.36 15.26 454.37 126.36 .Lazio 33,365.71 1583.26 5412.99 3829.73 4.75 16.22 341.89 2185.92Adriatic coast 16,713.84 1766.25 5391.78 3625.53 10.57 32.26 305.27 2069.37 B. Romano and F ZulloTirrenian coast 33,365.71 1583.26 5412.99 3829.73 4.75 16.22 341.89 2185.92
  • 15. Journal of Land Use Science 15 Pesaro Urbino <200% Ancona 200–300 Macerata 300–500 500–800 Perugia >800 Ascoli piceno Teramo Terni Pescara Viterbo Laquila Chieti Rieti Roma Frosinone Isernia Campobasso Latina 0 20 40 80 120 160 kmFigure 7. Map to select the positive values of the DUI.inhabitant lost. The geographical outcome is extremely varied (Figure 8), but shows a sig-nificant trend toward urban growth even in places subject to major population drops, withconcentrations of the highest values (over 800 m2 more urbanized surfaces for every inhabi- 295tant lost) in the mid-hilly strips, but also in inland mountainous and submountainous areas.The tourist models applied to mid-mountainous areas, based on holiday homes, are cer-tainly the main cause of this, but undoubtedly the policies pursued for decades to provideeconomic support to so-called ‘marginal areas’ have contributed to this too. The propensityof municipalities to collect taxes and dues to support public services is hardly negligible, 300but a significant thrust comes from the tendency, over the past 30 years, of privates to investin real property to offset either reduced the economic convenience or high risks of otherforms of investment.4. Effects on landscapeAn effective interpretation of landscape effects of urban sprawl over the past 50 years is 305the one based on the 37 types of physiographic landscape units censused by the NationalInstitute for Environmental Protection and Research (ISPRA, 2004). In the post-war period,the ‘intramountainous plateaus’ had the highest urbanization rates, with almost 6% ofartificialized land, while coastal, open, and valley-line plains were well below 2%. Theurbanization analyzed at the year 2000 shows a clear reversal in the value of this index: 310coastal plains have soared from the previous 2% to over 16%, just like the category ofcoastal elevations which have seen their urbanization rate reach values in excess of 18%,while the other types of plains have increased well beyond 4% and even 5%. However,it is particularly interesting to note that the ranking of hilly areas, scarcely or not at all
  • 16. 16 B. Romano and F Zullo . Pesaro Urbino <100 Ancona 100–200 200–500 Macerata 500–800 Perugia >800% Ascoli piceno Teramo Terni Pescara Laquila Viterbo Chieti Rieti Roma Frosinone Isernia Campobasso Latina 0 20 40 80 120 160 kmFigure 8. Map of the DUC.encroached upon by constructions in the 1950s, has risen. Hilly areas of all types have 315shown a significant increase in their urbanization rate which, from their original level rang-ing between 0% and 5%, now exceeds 2% by far and in some cases even 8%, that is to saycloser to the values of plains. In a country like Italy which lacks plains that may be easily saturated (approximately23% of the national territory), these are clear signs that denote a trend of massive ‘attacks’ 320of hilly areas by settlements, with considerable risks of altering agricultural landscapes,among the finest in Europe, and significantly deteriorating ecosystems, since over 40% ofNature 2000 Sites of the European ecological network (SCIs) are found in hilly areas. While in the 50 years considered, plains were replaced by settlements at an average paceof over 18 ha/day, for hilly areas the rate has been of almost 6 ha/day, but this could merely 325be the sign of the trigger for subsequent and more vigorous development (Figures 9–11). Figure 12 shows a simulation made throughout the national territory by applying theurbanization rate recorded in the past 50 years in the regions studied to the landscapeunits (Figure 9). The results highlight national sensibility concentrated essentially in somesensitive areas which, in addition to having suffered urban events over the past 50 years, 330still have a significant predisposition toward additional increases even today.5. ConclusionsThe research conducted in the regions of Central Italy has highlighted some fundamentalaspects: first, the average regional per capita urbanization value is about 300 m2 /inhabitantfor any type of territory, which suggests that this may be a standard need of present-day 335society. This value, reached spontaneously in half a century in various regions starting from
  • 17. Journal of Land Use Science 17Figure 9. Difference in the urbanization rate of Italian landscape units between the 1950s and afterthe 2000s (data referred to sample of regions studied).Figure 10. Variation of the urbanization rate in Italian landscape types between the 1950s and afterthe 2000s (data referred to sample of regions studied).very diverse bases, may be considered a reference threshold for future planning beyondwhich other land should not be used for urban functions. Second, we have observed that the increase in urbanized areas is scarcely correlatedto demographic trends, either positive or negative. In other words, the development of 340urban areas over the past 50 years has been marked by a huge dispersion in diffused formsscarcely governed by interpretable rules, leading to the systematic reproduction of a citymodel ‘lacking town planning’. The third point concerns the relationship with the landscape, that is to say the signs ofurban sprawl from the plains to the hills with adverse effects on the agro- and ecosystems, 345the most valued agricultural productions and rural landscape.
  • 18. 18 B. Romano and F Zullo .Figure 11. Average rate of daily urban conversion of land per type of landscape (m2 /day) for thesample of regions studied. In this regard, it is important to note that in Italy, the contact between city andcountryside has never been clear-cut, but rather nuanced between consolidated fabric, non-consolidated fabric, dispersed and degraded fabric, urbanized countryside, and agriculturalareas. The periurban areas are always marked by forms suggesting their precarious state of 350pre-urbanization. These reasons have led to anarchy in the layout of Italian territories, where residentialsettlements are always scattered and have a chaotic and metastatic organization from adevelopmental point of view (Figure 13), as evidenced by the values of the UDI in thecase of Marche and Umbria. The ensuing urban structure is lacking in public spaces and 355related services, profoundly devastating for the agricultural landscape, the integrity, andecological efficiency of natural habitats and gradually exacerbates all the adverse effects ofuncontrolled urbanization described in the Introduction to this article. Low-density rural settlement, especially when tied to an urban lifestyle, is a source ofmajor organizational difficulties for services and transport. 360 As a matter of fact, especially in Central and southern Italy, ‘hub-and-spoke’ (Babcock,2002) public mobility models are almost inapplicable, owing to the failure to meet thenecessary user and distance thresholds. A territorial organization of this kind inevitablyforces the residing population to depend on private transport means, with all the ensuingconsequences for environmental sustainability and quality of life. 365 The information emerging from this study may serve as a reference for town planningcoordination tools so that they may help stabilize the amount of artificialized areas andreorganize forms and functions of urban spaces. In this respect, some positive signals comefrom Lombardia and Umbria regions that issued laws about regional ecological networks,where the land uptake is controlled, but with tools that should be improved (Malcevschi, 3702010; Romano, 2009).
  • 19. Journal of Land Use Science 19Figure 12. Simulation, extended to landscape units across the national territory, of the percentagevariation of urbanization calculated on the sample regions in the course of the past 50 years. Desirable territorial governance policies should be geared mainly toward the system-atic implementation of actions for the functional recovery and reconversion of abandonedurbanized areas. In order to do this, local governments need to have technical and admin-istrative monitoring tools based on actual data on the evolution of land transformation at 375appropriate details scale.
  • 20. 20 B. Romano and F Zullo . Aggregate growth t1 t2 t3 Time ‘Metastatic like’ growthFigure 13. Differential chart, urban development in seamless aggregates, very frequent in northernEurope, and the generalized, metastatic-like urban development in Italy. In Italy, at present these tools do not exist and every municipality works independentlyas far as the size and distribution of urban areas of any kind are concerned. Moreover, struc-tured communication links between regions and municipalities about land-use changes arenot formalized in current legislation. 380 A so disconnected land planning system makes impossible any quantitative control oftransformations on a regional scale. Therefore, one of the first essential steps to be takentoward regulation involves the setting up of Regional Observatories and Land Registerswhich can be used to monitor the development of the phenomenon at different levels(municipal and regional) and on different scales, using advanced GIS technology. 385AcknowledgmentsWe thank the Region of Umbria (Biodiversity and Landscape Observatory) for economic sup-port of this research. We are also grateful to WWF Italia Onlus and the Environmental EngineersManuele Cargini, Doriana Febo, Consuelas Giuliani, and Cristina Iezzi for their collaboration fordata providing and elaboration. Thanks are due to the editor and the anonymous reviewer for their 390comments.ReferencesAPAT. (2005). La realizzazione in Italia del progetto europeo Corine Land Cover 2000 (Rapporti AQ4 36/2005).Astengo, G., & Nucci, C. (Eds.). (1990). It.Urb.80, Rapporto sullo stato dell’Urbanizzazione in 395 Italia. Quaderni Di Urbanistica Informazioni (Vol. 8). Roma: INU.Babcock, B. A. (2002). Making sense of cities: A geographical survey (pp. 63–94). London: Arnold.Battisti, C. (2011). Ecological network planning, from paradigms to design and back: A cautionary note. Journal of Land Use Science. doi:10.1080/1747423X.2011.639098Batty, M. (2008). The size, scale and shapes of cities. Science, 319, 769–771. 400Berdini, P. (2009). Il Consumo Di Suolo in Italia: 1995–2006. Democrazia E Diritto, 1, 60–73.Bonifazi, C., & Heins, F. (2001). Dynamics of urbanisation in Italy. Proceedings of XXIV general AQ5 population conference, IUSSP, Salvador, Brazil, August 18–24, 2001, 1–26.
  • 21. Journal of Land Use Science 21Bossard, M., Feranec, J., & Otahel, J. (2000). Corine land cover technical guide (EEA Technical AQ6 Report 40), p. 105. 405Bundesregierung. (1985). Unterrichtung durch die Bundesregierung. Bodenschutzkonzepte der Bundesregierung. Deutscher Bundenstag, Drucksache 10.2977, Bonn.Capello, R. (2001). Urban growth in Italy: Economic determinants and socio-environmental conse- AQ7 quences. International symposium in urban design on “Urban systems and public place”, Umeå, Sweden, June 7–8, 2001, CERUM, Report 10, 1–37. 410Cardoza, A. L. (1994). The large landed aristocracy of piedmont, retreat and adaption 1880–1930. In P. P. D’Attorre & A. de Bernardi (Eds.), Studi Sull’agricoltura Italiana (pp. 65–88). Milan: Feltrinelli.CNR-IPRA. (1988). Interaction and competition between urban systems and agriculture for land use purposes (IPRA doc. XXV). 415 AQ8Comber, A. J. (2008). Land use or land cover?. Journal of Land Use Science, 3(4), 199–201.Ellis, C. E., & Ramankutty, N. (2008). Putting people in the map: Anthropogenic biomes of the world. Frontier in Ecology and the Environment, 6. AQ9Emiliani, V (2007, October). Il consumo di suolo in Italia e in Europa. Atti Convegno Nazionale . Comitato per la bellezza, Roma. 420European Commission. (2006). Urban sprawl in Europe: The ignored challenge. Copenaghen: AQ10 Directorate general Joint Research Center.Falcucci, A., & Maiorano, L. (2008). Land use and misuse: Landscape changes in Italy since 1950. In C. Teofili & S. Clarino (Eds.), Riconquistare Il Paesaggio, La Convenzione Europea Del Paesaggio E La Conservazione Della Biodiversità in Italia (pp. 176–187). Roma: MIUR. 425Gambino, R., & Romano, B. (2004). Territorial strategies and environmental continuity in mountain systems, the case of the Apennines (Italy). In D. Harmon & G. Worboys (Eds.), Managing moun- tain protected areas: Challenges and responses for the 21st century (pp. 65–77). Andromeda: IUCN.Garcia-Call, A. (2011). The process of residential Sprawl in Spain: Is it really a problem?. Urban 430 Research and Practice, 4(3), 250–263.Hall, P., Gracey, H., Drewett, R., & Thomas, R. (1973). The containment of urban England (Vol. I, pp. 9–58). London: Beverly Hills.Hauri, E., Steiner, V & Vinzens, M. (2006). Human settlement in Switzerland, spatial development ., and housing. Housing Bulletin, 78, 1–80. 435Illy, A., Hornych, C., Schwartz, M., and Rosenfeld, M. T. W.. (2009). Urban growth in Germany – The impact of localization and urbanization economies. (IWH discussion papers 19:1–53). Halle Institute for Economic Research, Atlanta, GA.Irwin, E. G., & Bockstael, N. E. (2007). The evolution of urban sprawl: Evidence of spatial heterogeneity and increasing land fragmentation. PNAS, 104(52), 20672–20677. 440ISPRA. (2004). Carta dei tipi e delle unità fisiografiche d’Italia scala 1:250.000. Retrieved from http://www.isprambiente.gov.it/site/it-it/Servizi_per_l’Ambiente/Sistema_Carta_della_Natura/ Carta_della_Natura_alla_scala_1_250.000.ISTAT. (2009). Rapporto Annuale La situazione del Paese nel 2008. Roma: ISTAT.Lowry, I. S.. (1990). World urbanization. In Perspective. Population and development review, 16, 445 supplement: Resources, environment, and population: Present knowledge, future options (pp. 148–176). AQ11Malcevschi, S.. (2010). Reti ecologiche polivalenti:Infrastrutture e servizi ecosistemici per il governo del territorio. Milano: Il Verde.Mellor, R. (1983). The urbanization of Britain, a review. IJURR, 7(3), 380–403. 450Mercalli, L. (2009).Che Tempo Che Farà. Breve Storia Del Clima Con Uno Sguardo Al Futuro. Milano: Rizzoli.Pileri, P. (2007).Compensazione Ecologica Preventiva. Milano: Carocci.Plotkin, S. (1987). Property, policy and politics: Towards a theory of urban land-use conflict. IJURR, 11(3), 382–404. 455Reid, Jr., J. D. (1975). Sharecropping in history and theory. Agricultural History, 49(2), 426–440.Romano, B. (2009). Le interferenze insediative. In RERU, Rete Ecologica Regionale dell’Umbria AQ12 (pp. 65–110). Petruzzi: Regione Umbria.Romano, B., & Zullo, F. (2010). Tecniche di rilevamento e valutazione del fenomeno di conver- AQ13 sione urbana dei suoli. Atti della Sesta Conferenza Nazionale INPUT 2010, Potenza (Vol. I, pp. 460 327–337).
  • 22. 22 B. Romano and F Zullo .Sala, E. O., Chapin, F. S., Armesto, J. J., Berlow, E., Bloomfield, J., Dirzo, R., . . . Wall, D. H. (2000). Global biodiversity scenarios for the year 2100. Science, 287(5459), 1770–1774.Sambucini, V (2009, Novembre). Il progetto CORINE Land Cover e la sua evoluzione nell’ambito . AQ14 GMES/INSPIRE. L’uso del suolo delle regioni: confronto nazionale e con esperienze europee, 465 CISIS, RomaShaban, R. A. (1987). Testing between competing models of sharecropping. Journal of Political Economy, 95(5), 893–920.Sharma, L., Pandey, P. C., & Nathawat, M. S. (2012). Assessment of land consumption rate with urban dynamics change using geospatial techniques. Journal of Land Use Science, 7(2), 470 135–148.Singh, N. (1989). Theories of sharecropping. In P. Bardhan (Ed.), The economic theory of agrarian institutions (pp. 33–72). Oxford: Clarendon.Yanitsky, O. (1986). Urbanization in the USSR, theory, tendencies and policy. IJURR, 10(2), 265–287. 475Zaninetti, J. M. (2006). Urban sprawl in France, a regional typology of urbanization trends and its demographics and economy background. Bulletin of Geography, 5, 5–20.