Evaluation of Farming Cultivated Land Quality Based on Extension Method in Reclaimed Region Wang Zhimin, Lu Shibao, Bao Haijun, Zhu Jinxia and Zhao Jiaqian School of Urban-Rural Planning and Management, Zhejiang University of Finance and Economics, Hangzhou 310018, China

This study proposed a new model of cultivated land by the use of extension method, which was based on the index and its evaluation criteria. Results of Yu Huan county, Zhejiang province demonstrated that the proposed models was effective for the evaluation of cultivated land quality and it would a new useful and objective method to evaluate the cultivated land quality.


INTRODUCTION
A large population with relatively little arable land is basic national conditions of China, especially the eastern coastal areas.The per capita cultivated land is less than 1 acre and land carrying capacity has been reached or even exceeded the limitation in the eastern coastal areas.Land provides an important resource security and support for the rapid economic growth of eastern coastal areas.However, the rapid progress of industrialization and urbanization makes the growing shortage of land resources.Tideland resources (Mudflat) is important, which can reserve land resources of eastern coastal areas.The economic value of tideland resources is the most reasonable in the six reserve land resources: barren hills and land, grassland, alkali land, wasteland and barren sand land.Tideland resources are the most viable investment and the greatest potential for development.The arable land that enclosed tideland for cultivation alleviated the contradictions to help supplement the amount of cultivated land.
Agricultural land is the basic material of agricultural production.It is the indispensable condition for the survival of mankind.Arable farmland is the essence of agricultural land and the lifeline of human survival and development.The size of arable land and the level of quality directly related to China's economic construction and social development especially in area of food security.The challenge brought by the decline of the quality of cultivated land will be large.It would be more potential for the socio-economic development after compared with the amount of cultivated land, which had been reduced in a period of rapid urbanization.Therefore, quantity management in farmland management should be detected and more attention should be paid to the quantity and quality of the farmland.
The quality assessment is the basic work of the Quality management of arable land.Cultivated Land Quality Evaluation contains the natural evaluation and economic evaluation and the coordination degree of arable systems (Xu, 2004).Natural quality evaluation of cultivated land mainly takes these two calculation methods by the use of Regulations of Farmland Grading: the geometric mean method and the weighted average methods, then use GIS classification (Liu et al., 2005;Dong et al., 2007;Yu and Zhao, 2006).The geometric mean method will enlarge the role of a limiting factor and the weighted average method will cover up the role of a limiting factor (Zhang et al., 2005).Both of them are difficult to completely reflect the differences in the natural quality of arable land.Thus, academic circles based on gray correlation method (Lu et al., 2006) and the cluster analysis method of fuzzy matter element proximity of Cultivated Land Quality Evaluation (Nie, 2005) was proposed to provide a certain amount of technical supporting for quality evaluation of cultivated land.However, these fuzzy setting based methods rarely focused on the thing itself in the domain and the variability of the nature of the things.The evaluation results are not reliable.
Extension theory can determine the extent of things belonging to a collection based on the magnitude of things about characteristics.The correlation function can make the evaluation robust and quantitative, which provides a new way to solve the identification problem.This article will introduce the extension theory based on the extension set, by setting up the extension model of the natural quality to evaluate the reclamation area and demonstrated.

Extension theory:
The Extension Theory was founded by Tsai et al. (1997) Chinese scholars.It focuses on the rules and methods by the formal tools to solve the problem of contradictions from the point view of the qualitative and quantitative research (Tsai et al., 1997).Extension theory mainly includes matter-element theory and extension set theory and its logic cell is matter-element.
Matter element theory: In order to describe the process of changing of the objective things and making the process to solve the contradiction problem of formalization, extenics introduced the concept of matter-element as the basic elements to describe things.
Definition 1 defines the things name N, the magnitude of the characteristic c is v, we use an ordered triad R = (N, c, v) as the basic element for describing things, called matter-element, where N represents the matter; c, the characteristics; v is the N's.
Definition 2 A thing can have multiple characteristics, if things can be described by n features c 1 , c 2 , ..., c n and the magnitude corresponding to v 1 , v 2 ,..., v n then: The above formula is n-dimensional matterelement.Where in R is the magnitude corresponding to feature vector.

Extension set theory:
To quantify the problem-solving process, extension theory is used to establish the corresponding mathematical tool.
Definition 3 Let U be the domain of discourse, if U defined in any one element u, have a corresponding real number (u) (-,+ ) the domain of discourse Where in y = K (u) is the correlation function of A, K (u) is the degree of association for extension set A.

THE NATURAL QUALITY EVALUATION OF RECLAMATION AREA ON EXTENSION METHOD
Factor index system and evaluation criteria: The natural quality evaluation of reclamation area involves many influencing factors, the prerequisite and basis of the quality evaluation was setting up the index system that can reflect and measure the block comprehensively.The evaluation factors were selected according to the Delphi method, which was in accordance with the system of comprehensive, scientific, comparability and operability requirements.The index system was composed by 7 indicators.These seven indicators are foundation soil fertility, soil texture, organic matter, plow layer thickness, irrigation, elevation and drainage (Table 1).The rating criteria of indicators determined by the Delphi method are shown in Table 2.The minimum polygon formed by few maps (Soil maps, land use map and administrative maps) overlap as evaluation unit.The evaluation unit thus formed had clear space boundaries and administrative affiliation, accurate area and landform types, soil types, land use patterns and farming methods are basically the same.Then score of each evaluation unit was calculated based on the collected data and field surveys.The extension model of natural quality evaluation of cultivated land of reclamation area: Construction of the same intrinsic matter element model: N-dimensional with intrinsic material element model of the cultivated land evaluation unit according to the matter-element theory is as follows: In the above formula N i represents the i th evaluation units of arable land; C 1 , C 2 , ..., C n , represents the main features of the natural quality of the evaluation unit (the evaluation index), such as the basis of fertility, soil texture, organic matter, the plow layer thickness, irrigation, elevation and drainage; V li , V 2i , ..., V nI represents a magnitude which evaluation unit i correspond to C j (j = 1, 2, ..., n).

Determination of the classical domain and section domain:
R 0 is the same matter-element body of the same intrinsic matter-element R 1 , R 2 , ... R m , where G j represents evaluation category j, C i represents indicators i, V ij = <a ij , b ij > is the value range predetermined by N j about C i , that is the classic domain of data range of the corresponding indicators on each of the categories.Let, [ ] where, P = All categories V ip = The magnitude range taken by P about C i P = Section domain and V ij <VIP (i = 1, 2, ... and n; j = 1, 2, ..., m) In this case, the classic domain of each indicators of the evaluation unit is: In the above formula N t represents the evaluation unit level, when t = 1, 2, 3, 4, N t respectively level 1, 2, 3 and 4, respectively.X t1 , X t2 , X t3 , X t4 represents the magnitude of the range specified in the corresponding feature about N t , when t = 1, 2, 3, 4, the scope of its magnitude is <75, 100>, <50, 75>, <25, 50>, <0, 25> respectively.Setting section domain in this example is:

Determination of the weight coefficient of pending matter element and index:
The score results obtained was in accordance with the evaluation index of the pending evaluation unit Q, which was represented by matter element: calling pending matter element of q.V i is the magnitude of q for the evaluation C i , in other words q is the index score.
The weight coefficient of the arable land natural quality was set up using the Delphi method.The evaluation factors indicators C i of Yu Huan reclamation area is a i and T I (# = 1, shown in Table 3.
Calculation of the degree of association of the evaluation: Establish the correlation function of the evaluation unit q on level j as follows: Calculation of matter element based on comprehensive correlation degree: According to the degree of association on level j, which was determined by weight coefficient a i of indicator Ci and each evaluation of evaluation unit q, the comprehensive correlation degree of pending matter element was calculated as K j (q) = I H (˰ ) # .
Rating: Comparing the size of each level associated degrees to determine the results of the assessment.The degree of association of the level j greater, which indicates that the evaluation unit q compliance this level set was good.If , the evaluation unit q is belong to level j 0 .
Calculate Level variable eigenvalue of the evaluation unit q: * m m -j j j=1 j=1 j= jK (q) K (q) ∑ ∑ among them: j j j j j j j j K (q)-minK (q) K (q)= maxK (q) -minK (q) The level variable eigenvalues reflect the deviation degree of the level j 0 of the evaluation unit q tend to the other categories.
Case study: Five evaluation units of Yu Huan were used to test the proposed extensive comprehensive evaluation model.According to Table 1 and 2, scores these five evaluation unit of the various factor index, the results are demonstrated in Table 4.
The correlation about level j of the various factor index of these five evaluation unit was calculated based on the correlation function of the above established evaluation unit q and the evaluation index about j, the results are shown in Table 5.
Five units evaluation results was obtained by the calculation of the weight of each evaluation value, which was based on the comprehensive correlation  function.There are two evaluation units in the level 2 and 3 evaluation units in level 3.In other words, q 1 and q 2 evaluation results is the Level 2 and the remaining three evaluation unit level is 3.According to the level of characteristic values j of the evaluation unit (the smaller value of the characteristics of the present study the higher of the level), its excellent degree sequence is q 2 , q 1 , q 3 , q 5 and q 4 , which is shown in Table 6.

CONCLUSION
This study proposed the extension evaluation model which applied quantitative solutions to the natural quality evaluation of arable land of the reclamation area.This model can reduce the ambiguity and subjectivity of the evaluation and reflect the scientific and objective of the natural quality evaluation of arable land of the reclamation area.The extension evaluation model is further enriched the division method of the natural quality of agricultural land and can provide the basis for farmland grading.The results of the quality evaluation of cultivated land can provide technical support for the specific technical process of the general Land Use Planning (Xiao et al., 2009) (such as structural adjustment and layout of agricultural land, urban development direction, land consolidation area delineation, the reserves of back-up area of arable land, farmland demand forecast and determine the insurance number of cultivated land, etc.).The evaluation results could also be used for the conversion coefficient of farmland grading determination and arable production capacity accounting, etc.In addition, the causing and constraints of different natural quality space differences can be determine by extracting certain attributes like soil condition value, irrigation situation, etc. Optimization of the transformation measures, such as through the strengthening of irrigation and water conservancy facilities construction, changing farming systems of dry farming, optimizing planting structure, promoting water-saving irrigation, afforestation and grass, water saving and sand fixing, improving the production capacity of low-grade farmland, was be possible by the use of this proposed model.

Table 1 :
The evaluation index of cultivated land quality in Yu Huan county

Table 2 :
The criteria of evaluation index of cultivated land quality in Yu Huan county

Table 3 :
The weights of the evaluation index of cultivated land quality in Yu Huan county

Table 4 :
Scores of the evaluation index of each project

Table 5 :
Correlation values of K (v) about the level j for each evaluation index of the evaluated projects

Table 6 :
Results of evaluation Kj