J. Cent. South Univ. (2012) 19: 1364-1369
DOI: 10.1007/s11771-012-1151-x
Ecological suitability evaluation for urban growth boundary in red soil hilly areas based on fuzzy theory
JIAO Sheng(焦胜)1, GAO Qing(高青)1, WEI Chun-yu(魏春雨)1, LIU Bei(刘贝)1, LI Xiao-dong(李晓东)2,
ZENG Guang-ming(曾光明)2, YUAN Zhong-xing(袁中兴)2, LIANG Jie(梁婕)2
1. College of Architecture, Hunan University, Changsha 410082, China;
2. College of Environmental Science and Engineering, Hunan University, Changsha 410082, China
? Central South University Press and Springer-Verlag Berlin Heidelberg 2012
Abstract: The fuzziness exists in spatial distribution of geographic data of land suitability evaluation processes, which makes it difficult to quantify land boundaries by using traditional binary logic-based overlay model. Aiming at this limitation, an ecological suitability evaluation analysis model was presented based on fuzzy theory and a research on urban growth boundary (UGB) of the Great-Hexi Leading District (GHLD) of Changsha was conducted. With the support of GIS, RS and MATLAB, slope, elevation, vegetation, soil productivity, soil permeability, water body and land use are selected as the input of model according to the characteristic properties of soil and terrain in red soil hilly areas. The running result of this model indicates that the ratios of highly suitable land, suitable land, moderately suitable land and unsuitable land in GHLD are 18.75%, 10.31%, 64.16%, 6.78%, respectively. This result accords with spatial structure worked out by Space Development Strategy Planning of GHLD. Based on this result, several suggestions are made to guide UGB developments in future.
Key words: land ecological suitability evaluation; fuzzy theory; urban growth boundary (UGB); GIS; MATLAB
1 Introduction
Urban growth boundaries (UGBs) are common tools employed by planners to constrain urban expansion in order to increase density of urban services and protect surrounding rural landscapes [1]. Red soil hilly areas, which have complex terrains and fragile soil characteristics, are susceptible to urbanization and natural disaster. Rapid urbanization area needs the controlling of UGB that gives more priority in ecology. Therefore, evaluation of UGB is often conducted using land suitability models [2]. The implementing results of these models determine the land boundaries. However, traditional land suitability overlay models are based on classical mathematics, which are unable to precisely quantify the fuzzy objects with distinctive boundaries due to their limitations. This shortcoming motivates the requirement of new model that shows better accuracy in land boundary classification.
In classical mathematics, such as Cantor Set Theory, there are only two situations to define how an element is associated with a set: “in” or “not in” [3]. Since Cantor Set Theory derives from the abstract concept ignoring the fuzziness in objects, it is tough to objectively handle a large number of fuzziness phenomena by adopting classical mathematics theory, which exist in real life or sophisticated system [4]. In complicated land suitability evaluation, there is a problem that the assumption of identifying the input data as being accurate is unrealistic, which can be solved by applications of fuzzy theory and fuzzy logic technology [5]. In 1965, ZADEH [6] generalized the crisp membership value “0” and “1” into the fuzzy membership range [0, 1]. The implement of fuzzy membership can quantitatively describe the extent how an element in a domain meets the requirements of set, which realizes the extension to crisp membership. Fuzzy Set Theory is a data analysis method which provides a powerful tool for interpretation and analysis of date uncertainty in GIS. Fuzzy theory is used for optimizing the model in this work. As a novel mathematical theory, it has received wide attentions of researchers in many subjects. With the support of GIS, this theory can be perfectly integrated with land suitability evaluation. For instance, BOROUSHAKIN and MALCZEWSKI [7] conducted an application of a GIS-integrated fuzzy rule-based inference system for land suitability evaluation in agricultural watersheds, presenting that the model is useful in the land suitability assessment in agricultural watersheds. NISAR AHAMED et al [8] examined a GIS-based fuzzy membership model for crop-land suitability analysis, revealing that the maximum area is potentially suitable for growing ground nut, not the crop (?nger millet) being grown. MOUTON et al [9] developed a fuzzy habitat suitability model for river managers. SICAT et al proposed a fuzzy modeling of farmers’ knowledge for agricultural land suitability classi?cation, indicating the usefulness of fuzzy modelling in FK-based classi?cation of agricultural land suitability [10]. BROEKHOVEN et al [11] used a fuzzy rule-based approach in a macroinvertebrate habitat suitability modelling problem. NISAR AHAMED et al [12] introduced fuzzy class membership approach to soil erosion modelling. BOROUSHAKIN and MALCZEWSKI [13] developed a framework for GIS-based multicriteria group decision-making using the fuzzy majority approach. These researches indicated that fuzzy-based model is superior to the traditional land suitability model in land use classification.
While there have been a few applications to UGB setting, the present fuzzy theory-based land suitability models were mainly constructed for agricultural land suitability evaluation. In addition, few previous overlay models were focused on the accuracy of land use classification. As a result, the goal of this work is to introduce the fuzzy theory into UGB so that the accuracy of land boundary classification can be improved. In terms of fuzzy theory, the modified land ecological suitability model for UGB was built, and the running result could offer scientific evidences for the future setting of UGB.
2 Materials and method
The model was developed for the Great-Hexi Leading District (GHLD), in the City of Changsha. Changsha city, the capital of Hunan Province, is also the core city in economic integration of Changsha-Zhuzhou- Xiangtan. Being laid to the west of Xiangjiang River, the Great-Hexi Leading District (GHLD) covers 1 200 km2, including four counties and fifteen towns in its geographic scope. In addition to sustaining the subtropical monsoon climate type, the GHLD is the typical humid region enjoying a wild summer and winter. It has elevation ranging from 20 to 470 m and complex terrain in which plain, down land, hill area and low mountains cross each other. Moreover, there is chiefly red soil and its subtypes develop in these areas. Consequently, the GHLD pertains to red soil hilly areas.
The model is exhibited in Fig. 1. The inputs of the model (Fig. 2) are identified as slope, elevation, vegetation, soil productivity, soil permeability, surface drainage and land use. Based on RS and GIS technology, the vector layer of land use, water body and vegetation etc are extracted from the interpretation of 2009 Landsat-7 TM RS data by using ERDAS IMAGINE
Fig. 1 Technique routing of urban growth land ecological suitability evaluation in GHLD
Fig. 2 Ecological factors: 1-Slope; 2-Elevation; 3/4-Buffer of water body; 5-Vegetation; 6-Land use; 7-Soil permeability; 8-Soil productivity
ver. 9.2. The soil data in this work are offered by Collage of Environmental Science and Engineering, Hunan University. In ARCGIS 10, the projection coordination systems of all the attribute data are modified in accordance with the national 1:250 000 basic topographic databases. All the attribute maps are systematically classified into different groups and converting discrete to raster format in 100 m resolution. Then, all the attribute raster maps are converted into float data by using the command “Raster to Float” of Conversion Tool in ARCGIS 10. All the float data were imported into MATLAB fuzzy logic toolbox to perform fuzzy classification computation. This calculation includes five steps: fuzzi?cation of attributes, generation of joint membership values, generation of fuzzy rules, aggregation of the fuzzy rules, and defuzzi?cation to find the crisp suitability index [14]. The model running result is classified according to the suitability classification criterion of FAO (Food and Agriculture Organization of the United Nations) [15]. Crisp land suitability value of each grid is processed by clustering analysis, through which land ecological suitability evaluation result and acreage of each classification could be obtained.
3 Results and discussion
3.1 Model application
The classification, data type, selected membership function of each attribute and the classification criteria in this work are given in Tables 1 and 2, respectively.
3.2 Land ecological suitability evaluation result
The land ecological suitability evaluation result of GHLD is shown in Fig. 3:
1) Highly suitable land (224.38 km2) which accounts for 18.75% of total land of GHLD is the area whose slope is less than 5° without natural vegetation, including barren mountain area, low-yield farmland and poor landscape area. These lands facing intensive development need promoting utilization efficiency of resource so as to meet the requirements for population and economic growth at the least cost of land.
2) Suitable land (123.35 km2) which accounts for 10.31% of total land of GHLD is the area which is mostly low-yield farmland whose slope is less than 5°. There are relatively intensive residential settlements in these lands. But after construction measure and ambient compensation arrangement, it can be used as urban growth land. These underdeveloped lands need intensive development and utilization under the direction of planning so as to reserve more land for ecological conservation and restoration.
3) Moderately suitable land (767.74 km2) which accounts for 64.16% of total land of GHLD is the area which is mostly mid-yield farmland whose slope is from 5° to 10°. There are relatively intensive residential settlements in these lands. But after construction measure and ambient compensation arrangement, it can be used as urban growth land. Except for significant road infrastructure, municipal public service, tourist facility, park, ecological agriculture, other human activities are forbidden.
4) Unsuitable land (81.13 km2) which accounts for 6.78% of total land of GHLD is the area whose slope is greater than 10°. From the view of ecology and protection of productive land, this land is not suitable for exploiting which is mostly high-yield farmland or influence area of ravine stream with better vegetation. Any development independent of environmental protection and ecological construction are discouraged.
Table 1 Evaluation attributes and their model parameters
Table 2 Suitability composite classification criteria
Fig. 3 Result of land ecological suitability evaluation in GHLD
Moreover, existing settlement must be reserved and optimized. Meanwhile, protective ecological development is encouraged.
3.3 Discussion
The result of evaluation is in accordance with “One modern core service zone, two ecological preservation subzones, and three industry service subzones (The “1+2+3” double-dimension)” consisting of each industry cluster and its surrounding towns, worked out by Space Development Strategy Planning of Changsha GHLD [16]. In addition, by comparing the urban boundary interpreting from 1989, 1999 and 2009 remote sensing data (Fig. 4), there is obvious evidence indicating that result of this work is in step with the growth direction of the GHLD in the past 30 years.
Over the past decade, rapid urbanization occurs in the GHLD, which should be restricted by environment. The setting of UGB can effectively preserve ecology and species [17]. Moreover, UGB has a crucial function of confining geographic position in land intensive utilization of hilly cities [18]. In short, the setting of UGB must be in the orientation of ecology factor, emphasizing on integrality of ecological factor as well as protecting fragile environment of hilly areas [19].
However, UGBs and greenbelts are not intended to be static but are adjusted according to new needs [20]. On one hand, the UGB of hilly cities should be set up under the guide of a “rigid boundary” which can protect the integrality of ecological space [21]. On the other hand, there should be a certain of elastic land allowed for unforeseen land growth [22]. In order to obtain reasonable pilot control, the UGB of GHLD was formulated in Space Development Strategy Planning of Changsha GHLD. This boundary is fundamentally identical with the result of this work although minor conflicting areas exist in locality as demonstrated in Table 3.
Fig. 4 Urban growth conditions of GHLD in 1989 (a), 1999 (b) and 2009 (c)
The elastic land of GHLD covers Gushan neighbor land, the wetland of Maqiao River mouth and Bairuopy cluster. The acreages of these elastic lands are 667 hm2, 173 hm2 and 1 630 hm2, 2 470 hm2, respectively. Referring to conflict analysis, several suggestions are made:
1) Designating the wetland of Maqiao River mouth, which is adjoined to Xiangjiang River, as a building area would have adverse impacts on Xiangjiang River and
Table 3 Conflict area of evaluation result with UGB (GHLD)
Maqiao River. Thus, the land use of Maqiao River mouth is recommended to be unsuitable land so that the environment of Xiangjiang River and Maqiao River and the wetland of Maqiao River mouth can be protected.
2) The south of elastic land in Gushan neighbor land is situated in unsuitable land but the west is not. Due to better environment in Gushan neighbor land, the elastic land in Gushan neighbor land is recommended to be unsuitable land so that an ecological corridor could be generated, linking the middle part of GHLD with eastern Xiangjiang River.
3) The building land which is far away from Jinzhou avenue region could be recommended as an elastic land reserved for key construction site selection and new country construction, because Bai Ruopu cluster is not a key area for construction in the near future.
4 Conclusions
1) In comparison with the traditional numerical value evaluation method, the land ecological suitability evaluation (LESE) is better to realize quantitative analysis in a simple and convenient manner. Based on GIS and RS, LESE organizes the extraction of land information and numerical calculation with the processing of spatial data. By interpreting remote sensing data to obtain the quantitative value of each attribute and introducing fuzzy theory into evaluation model, the growth land ecological suitability is evaluated and several suggestions are proposed to the setting of UGB in GHLD based on the result of evaluation. The model, procedure and result of this evaluation offer effective method and pattern for the scientific evaluation of urban growth land, which has reference value on the urban land use planning and evaluation.
2) The adopting of fuzzy theory can solve the problem caused by fuzziness of single factor attribute in geographic space distribution, enhancing the objectivity and scientificity of evaluation result. By using fuzzy membership function to engage the fuzzification of attribution, two uncertainties in traditional binary logic classification can be avoided. One is the uncertainty that two things which have great disparity are regarded as being in the same class, and the other is that two things which have minor disparity are regarded as being in different class. At present, no uniform standard has been given, so the selection of fuzzy membership function remains to be further studied.
3) This result basically accords with present status of development in GHLD. Meanwhile, it coincides with the control principle and control key point of regional spatial structure (The “1+2+3” double-dimension) worked out by Space Development Strategy Planning of Changsha GHLD and land growth direction of GHLD.
References
[1] TAYYEBI A, PIJANOWSKI B C, TAYYEBI A H. An urban growth boundary model using neural networks, GIS and radial parameterization: An application to Tehran, Iran [J]. Landscape and Urban Planning, 2010, 26(3): 349-357.
[2] LI Wei, XU Jian-chun. Study on spatial structure of Hangzhou city based on UGB [J]. Urban Studies, 2009, 16(4): 106-111. (in Chinese)
[3] WEN Xin, ZHOU Lu, LI Dong-jiang. Matlab fuzzy logic toolbox analysis and application [M]. Beijing: Science Press, 2001: 1-4. (in Chinese)
[4] ZHANG Xiao-xiang, YAO Jing, LI Man-chun. A review of fuzzy sets on spatial data handling [J]. Remote Sensing Information, 2005(2): 47-51. (in Chinese)
[5] HE Ying-bin, CHEN You-qi, YANG Peng, WU wen-bin, YAO yan-min, LI zhi-bin. An overview and perspective of alien land suitability evaluation: study based on GIS technology [J]. Progress in Geography, 2009, 28(6): 898-904. (in Chinese)
[6] ZADEH L H. Fuzzy sets: Information and control [J]. 1965, 8: 338-353.
[7] RESHMIDEVI T V, ELDHO T I, JANA R. A GIS-integrated fuzzy rule-based inference system for land suitability evaluation in agricultural watersheds [J]. Computers & Geosciences, 2009, 101(1/2): 101-109.
[8] NISAR AHAMED T R, GOPAL RAO K, MURTHY J S R. GIS-based fuzzy membership model for crop-land suitability analysis [J]. Agricultural Systems, 2000, 63(2): 75-95.
[9] MOUTON M A, ALCARAZ-HERN?NDEZ J D, BAETS B D, GOETHALS P L M, MART?NEZ-CAPEL F. Data-driven fuzzy habitat suitability models for brown trout in Spanish Mediterranean Rivers [J]. Environmental Modelling & Software, 2011, 26(5): 615-622.
[10] SICAT R S, CARRANZA E J M, NIDUMOLU U B. Fuzzy modeling of farmers’ knowledge for land suitability classi?cation [J]. Agricultural Systems, 2005, 83(1): 49-75.
[11] BROEKHOVEN E V, ADRIAENSSENS V, BAETS B D, VERDONSCHOT P F M. Fuzzy rule-based macroinvertebrate habitat suitability models for running waters [J]. Ecological Modelling, 2006,198(1/2): 71-84.
[12] NISAR AHAMED T R, GOPAL RAO K, MURTHY J S R. Fuzzy class membership approach to soil erosion modeling [J]. Agricultural Systems, 2000, 63(2): 97-110.
[13] BOROUSHAKI S, MALCZEWSKI J. Using the fuzzy majority approach for GIS-based multicriteria group decision-making [J]. Computers & Geosciences, 2010, 36(3): 302-312.
[14] Fuzzy logic toolbox tutorial [EB/OL]. [2011-03-04]. http://www.mathworks.com/help/toolbox/fuzzy/fp754.html.
[15] FAO (Food and Agriculture Organization of the United Nations). A Framework for land evaluation. Soil bulletin, No 32 [R]. Rome: FAO. 1976.
[16] Planning Administrative Bureau in Changsha, Urban Planning Design Institute in Shenzhen, Planning Design Consultant Limited Company in Changsha. Space Development Strategy Planning of Changsha GHLD [R]. 2008. ( in Chinese)
[17] GORDON A, SIMONDSON D, WHITE M, MOILANEN A, BEKESSY S A. Integrating conservation planning and land use planning in urban landscapes [J]. Landscape and Urban Planning, 2009, 91(4): 183-194.
[18] WANG Guo-en, XIAO Rong-bo, PENG Tao. Land intensive use planning and developing strategy in mountain city: Taking Zunyi as an example [J]. Modern Urban Research, 2010(2): 74-79. (in Chinese)
[19] DUAN De-gang, LU Shou-yi, TIAN Tao. Some preliminary exploration of UGB establishment-management system [J]. Planners, 2009, 25(8): 11-14. (in Chinese)
[20] GENNAIO M P, HERSPERGER A M, BURGI M. Containing urban sprawl-Evaluating effectiveness of urban growth boundaries set by the Swiss land use plan [J]. Land Use Policy, 2009, 26(2): 224-232.
[21] ZHU Zhong-wen, MO Bin, XIE Fu-rong. Delimitation of urban growth boundary based on land ecological suitability evaluation: A case of Fangchenggang [J]. Planners, 2009, 25(11): 40-44. (in Chinese)
[22] FENG Ke, WU Ci-fang, WEI Shi-chuan, LIU Yong. Urban expansion management: Review on the theory of UGB and its reference for China [J]. China Land Science, 2008, 22(5): 77-80. (in Chinese)
(Edited by YANG Bing)
Foundation item: Project(2006BAJ04A13) supported by the National Science and Technology Pillar Program during the 11th Five-Year Plan of China; Project(2009FJ4056) supported by the Key Project of Science and Technology Program of Hunan Province, China; Project(20090161120014) supported by the New Teachers Fund of Department of Education, China
Received date: 2011-03-01; Accepted date: 2011-07-28
Corresponding author: GAO Qing; Tel: +86-15802601524; E-mail: gq2213297@hotmail.com