Land Use and Landscape Pattern Changes in Chongqing Liangjiang New District

Taking Chongqing Liangjiang new district within second-ring road as researching area, analyzed its land use and landscape pattern changes during 1988, 2001 and 2010. After utilizing ENVI to do the remote sensing images interpretation, land use changes were described in conversion matrix. Method of FRAGSTATS was utilized to calculate the special landscape indices, which included patch density (PD), largest patch index (LPI), land shape index (LSI), contagion (CONTAG), landscape division index (DIVISION), aggregation index (AI). Results show Land use types change from farmland to urban mostly during 2001 to 2010; PD, LPI and LSI drop down, the other three have increasing tendency. Land use and landscape pattern associate with each other, Urbanization featured with human activities, causes landscape changes, besides, preferential policies of Conversion of cropland to forest, urban planning and the typical “Three Reforms” in Chongqing boosts land cover changes too.


Introduction
Recently, Chinese cities have experienced rapid population growth and continuous expansion, resulting in considerable and sustained demand for land resources.Meanwhile, urbanization has imposed significant pressure on the land-use structure, terrestrial and aquatic ecosystems, of these areas (Liu et al, 2007).The changes in land use caused by urbanization greatly affect the structure and function of urban ecosystems (Zhou & Cadenasso, 2011).Land-use structure was associated well with landscape pattern in structural elements, although the quantification of landscape structure is not a new topic (Forman & Godron, 1986;O'Neill, 1988), this field is still evolving due to the ongoing developments in geo information technology (Gardner, Lookingbill, Townsend & Epprecht, 2009;Riitters, Vogt, Soille & Estreguil, 2009;Saura & Castro, 2007).With development of modern cities, urban landscape becomes one of the fastest variable patterns, from the aspect of landscape patterns to review urbanization attracts more and more attention.However, most theories of urban landscape patterns changes focus on plaint cities, different terrains have huge impacts on landscape structures (Ispikoudis, Lyrintzis & Kyriakakis, 1993;Hobbs, 1993;Pignatti, 1993), which means hilly cities show uniqueness in land use and landscape patterns changes, its features need special analysis.
Chongqing experienced rapid economic development since direct jurisdiction of the Central Government.Especially, after Three Gorges being set up, the Urban and Rural Harmonious Experimental District was founded in Chongqing, along with which, the first inland-bonded-port area established in northern main urban district, until 2009, the completed high-way took Chongqing towards the second-ring new era.Following Shanghai Pudong New Area and Tianjin New Coastal Region, Liangjiang New District of Chongqing becomes the third national Special Economic Zones.Founded in June.18 th , 2010, it is given certain of special policies, especially the tax policy with appropriate special discount (15%), which attracts some international factories like HP, Foxconn and Ford ( Su, Yang & Liu, 2011).
Liangjiang New District (within the second-ring road) is chosen to be our researching area.The fast development results in rapid land use changes as well as the landscape pattern changes.Extracted the needed information of Liangjiang New District in 1988District in , 2001District in and 2010 to demonstrate the effects of urbanization on land use and land pattern changes, in addition, discuss the driving forces.Landscape pattern indices are concentration of landscape information, which can reflect features of construction and spatial allocation.Since landscape pattern can be divided into three gradients according to different matrices: class, patch and landscape (Wu, 2007).Numerous metrics have been developed to quantify the composition and conFigureuration of land cover features (Gustafson, 1998;McGarigal, Cushman & Neel, 2002), they have been used to measure a region's landscape change through time (Dunn, Sharpe & Guntenspergen, 1990;Delong & Tanner, 1996;Reed, 1996).In order to investigate and compare the before-and-after landscape patterns, we choose the following landscape pattern indices: patch density (PD), largest patch index (LPI), land shape index (LSI), contagion (CONTAG), landscape division index (DIVISION), aggregation index (AI).With 8 cells as patch neighbors, Fragstats 3.3 were applied to calculate the landscape pattern indices.

Study S
(1) PD equals the number of patches in the landscapes, divided by total landscape area (m 2 ), multiplied by 10000 and 100 (to convert to 100 hectares).It is a limited, but fundamental, aspect of landscape pattern, which shows fragmentation of the patches.
10,000 100 Where N is the total number of patches in the landscape, A is the total landscape area (m 2 ).PD is ultimately constrained by the grain size of the raster image, because the maximum PD is attained where every cell is separate patch.
(2) LPI equals the area (m 2 ) of the largest patch of the corresponding patch type divided by total landscape area (m 2 ), multiplied by 10000 and 100 (to convert to 100 hectares); in other words, it is the percentage of the landscape comprised by the largest patch.

(2)
Where a ij is the area (m 2 ) of patch ij; A is the total landscape area (m 2 ).LPI approaches 0 when the largest patch of the corresponding patch type is increasingly small.LPI=100 when the entire landscape consists of a single patch of the corresponding patch type; that is, when the largest large patch comprises 100% of the landscape.
(3) LSI equals the total length of edge in the landscape, given in number of cell surfaces, divided by the minimum total length of edge possible, also given in number of cell surfaces, which is achieved when the landscape consists of a single patch. ( Where E is total length of edge in landscape in terms of number of cell surfaces; includes all landscape boundary and background edge segments; min E is minimum total length of edge in landscape in terms of number of cell surfaces.LSI≥1 without limit, it increases as landscape shape becomes more irregular or as the length of edge within the landscape increases; LSI=1 when the landscape consists of a single square patch. (4) CONTAG is the observed contagion over the maximum possible contagion for the given number of patch types.
Where P i is proportion of the landscape occupied by patch type (class) i; g ik is number of adjacencies between pixels of patch types i and k based on the double-count method; m is number of patch types present in the landscape, including the landscape border if present.0＜CONTAG≤100, it approaches 0 when the patch types are maximally disaggregated and interspersed; CONTAG=100 when all patch types are maximally aggregated.
(5) DIVISION is based on the cumulative patch area distribution and is interpreted as the probability that two randomly chosen pixels in the landscape are not situated in the same patch.

∑ ∑ (5)
Where a ij is area (m 2 ) of patch ij; A is total landscape area (m 2 ).0 ≤ DIVISION ＜ 1, DIVISION = 0, when the landscape consists of single patch.DIVISION is achieves its maximum value when the landscape is maximally subdivided; that is, when every cell is a separate patch.

Land Use Conversion Matrix
Land use change detection analysis can be applied directly to satellite imagery (Lu, Morineau & Piron, 2003).
After obtaining post-classification images, via change detection technique (Van Oort, 2007), we got two matrixes of transitions between land categories (Manandhar, Odeh & Pontius, 2010), the first is land use changes between 1988 to 2001, the second is from 2001 to 2010.
Applied ENVI 4.6 to do change detection, exported the information and selected "area" as index to do further analysis.Next, compared the conversion matrixes, and concluded differences in the above two periods.PD shows a constant decrease with a big drop from to 6.60 in 2001, and then a further fall to 5.75 in 2010.

Land Use Changes between 1988 to 2001
Concerning the resolution of the used Lands at TM/ETM+ images, we divided the primary images into four classes: urban, hill, water and farmland.Constrained by natural resources (mountains and river), Urban expansion in this period is around valley along Yangtze River and Jialing River, especially Yangtze River, where plenty of urban area has existed in 2001 (Figure .3).Linear expansion shows the development of transportation, since high-way project was almost completed till 2010, the second-ring road era starts, and urban sprawl in this period is featured with along traffic lines.
Changes of PD, LSI and CONT indicate the effect of urbanization.Firstly, Farmland is much fragmented than urban area, which means more small patches contained in former imagery, value of PD drops when some arable land transfers to urban and becomes part of CBD.Additionally, with less independent patches, urban patches are relatively regular, which results in lower LSI.It is obvious that, urban develops only when population, industries or investments concentrated, consequently, CONT accelerates when urbanization develops with a fast speed.), where the abandonment of agricultural activities has been accompanied by an increase in forests area (Calvo-Iglesias, 2006;Agnoletti, 2007;Geri, 2010).

Urban Planning
According to urban master planning, the available urban area in Liangjiang New District is 550 km 2 , the remained land covers are water and ecological districts that cannot be developed.The old cities reconstruction of southern part started in 2002; by 2005, Chongqing Theater was commenced as well as the road construction; since 2009, the commercial street was set up in central Liangjiang New District, with a target of making Chongqing to be the financial center in the upper and middle Yangtze River, Liangjiang New District is increasingly indispensable.As a result, urban expansion accelerates with a high speed and land covers change a lot during 10 years.

"Three Reforms"
Since the party put forward the strictest arable land protection, like the dynamic equilibrium of the whole land, it means urban development should guarantee to offer the same amount of farmland before occupying arable land.With rapid urbanization, Liangjiang New District has a serious of reform to solve this problem, including land quota, household registry reform and construction of public rental housing.
The first land quota was verified in February 2010, which initiates a creative road for obtaining land index (Yang, 2011).It contains four steps: land reclamation, acceptance, transaction and using the land quota, before requisition, equal amounts of arable land in other districts of Chongqing will be reclaimed, hence, it makes up for the deficiency of Requisition-compensation Balance and keeps the dynamic equilibrium of the whole land in Chongqing municipal city.
Development of Liangjiang New District supplies considerable job opportunities for immigrants, most of who comes from countryside.In order to absorb such new employees and make them to be citizens, household registry reform was implemented since August 2010.Before migration, farmers are given compensation (67000 Yuan on average) for releasing their residential land, arable land or forest.It encourages rural residents especially the youths to work and live in Liangjiang New District.
Accompany with urbanization, housing price increase much more rapidly, which greaten pressure to citizens, for new immigrants from rural area, living comes first.In this situation, Chongqing carries out the public renting house in 20 residential areas, 9 of which are founded in Liangjiang New District.Differentiated from other modern cities like Beijing, Shanghai, public renting house in Chongqing has no household registry limits.Once works in Chongqing stably, singles with monthly salary lower than 2000 Yuan or families with 3000 minus monthly can apply for public renting house.Since such rents only accounts for 60% of ordinary ones, it attracts more people to work and live in Chongqing.

Conclusions
The landscapes of Liangjiang New District within second-ring road are more dynamic from 2001 to 2010 than the former period.In details, CONTAG increases significantly, PD, LPI and LSI drop down a lot, DIVISION remains steady.Correspondently, Land use changes mostly occur between farmland and urban area, since urban area rises, several fragmented patches aggregate together and becomes more concentrated, which in turn, can be reflected by landscape patterns.
Urbanization and boost landscape and land use changes.Constrained by natural environments, urban sprawl in Liangjiang New District is almost along the Yangtze River and Jialing River, between Zhongliang Mountain and Tongluo Mountain, which results in higher aggregation degree.
Conversion of cropland to forest policy explains the increases of forest in the latter period.With guidance of urban master planning, urban expansion in Liangjiang New District accelerates since 2001."Three Reforms" in Chongqing ensures urban areas rises: land quota trading makes constructed land-use index legally transforms from rural to urban, after that, moving farmer-immigrants to " Liangjiang New District" with help of residence reform, additionally, public renting house offers low renting fee for their stable lives.
Future research will involve the quantitative analysis on relationship between landscape indices and land use changes; the analysis of land cover maps and socioeconomic data to determine the main factors driving landscape dynamics.

( a )
Figure 2. of landscape indices Conversion matrix was made to investigate changes among the four land use types from 1988 to 2001, 2001 to 2010.
Generally, forest changes the least during 1988 to 2001 (Table.1),most of which (0.63 km 2 ) transfers to farmland, whilst, 0.33 km2 farmland turns to be forest at the same period, hence, the total area of forest keeps almost Forces of Policies 4.2.1 Conversion of Cropland to Forest Since 2000, China The Strategy of Large-Scale Western Development, numbers of western cities especially Chongqing enjoys several preferential policies.Concerning about deteriorated environment in such areas, Conversion of Cropland to Forest is carried out, based on which, the poor land with low output should be improved and then transfer to be forest.Responses of the policy can be reflected by land use changes from farmland to forest.Studies elsewhere confirm this trend, for example in Galicia (Spain), Siena (Italy) or Tuscany (Italy