J. Cent. South Univ. Technol. (2011) 18: 1465-1472

DOI: 10.1007/s11771-011-0862-8

Assessment on soil fertility of Dongting Lake wetland area (China) based on GIS and fuzzy evaluation

LI Zhong-wu(李忠武)^{1, 2}, HUANG Jin-quan(黄金权)^{1, 2}, LI Yu-yuan(李裕元)³,

GUO Wang(郭旺)^{1, 2}, ZHU Jian-feng(朱剑峰)^{1, 2}

1. College of Environmental Science and Engineering, Hunan University, Changsha 410082, China;

2. Key Laboratory of Environmental Biology and Pollution Control, Ministry of Education, Hunan University, Changsha 410082, China;

3. Institute of Subtropical Agriculture, Chinese Academy of Sciences, Changsha 410125, China

? Central South University Press and Springer-Verlag Berlin Heidelberg 2011

Key words:

Dongting Lake; wetland; ecosystem; soil fertility; fuzzy evaluation; geographic information system (GIS)；

1 Introduction

Wetland is an especially transitional ecosystem type between land and water, and also is one of ecosystems with the highest biomass and the most abundant biodiversity on the earth surface. The wetland ecosystem has a close relationship with human’s survival and development. At the same time, the wetland ecosystem has many incomparable functions that other ecosystems do not have, such as riverhead protection, flood resisting, pollutant degradation and climate adjustment. As for the researches related with wetland, many researchers focused much importance on wetland resource protection [1], wetland function for human society [2], wetland ecosystem properties and health [3-4], wetland ecological protection and management mode [1, 5] and wetland environment impact assessment, wetland ecological value evaluation and wetland ecological risk appraisement [6-8]. Nowadays, researches on treating environmental problem (waste water) by constructed wetland are of more importance [9]. However, there are only a few studies about the wetland soil fertility. In China and monsoon Asia, the wetland area is the main rice growing region, and the wetland rice production systems in Asia play an important role in the global rice supply [10]. Hence, it is very important to evaluate the wetland soil fertility status.

The soil fertility assessment is an approach of using mathematic method to appraise the soil production ability. The evaluation can represent complex soil fertility by simple numerical value, which can be easily understood by user and will be valuable for researchers to understand the long-term effect of soil current status and tillage measures on the soil fertility [11]. As for the researches about soil fertility assessment, selecting reasonable evaluation means is the emphasis. Especially, quantitative mathematics method is increasingly important for the soil fertility assessment. A lot of researchers have done some studies related with quantitative mathematics method of soil fertility evaluation, and put forward many valuable evaluation methods, including AHP method [12], integrated fertility index method [13], integrated grading evaluation method [14], integrated fuzzy evaluation method [15], cluster analyses method [16] and geo-statistics method [17], of which the integrated fertility index method, cluster analyses method and integrated fuzzy evaluation method are widely used to evaluate the soil fertility.

The integrated fuzzy evaluation method, used to evaluate the soil fertility, has the following advantages.  1) It can take much information, including personal experience and knowledge, into account [18]; 2) It can treat uncertainty factors of soil system during the course of research; 3) It can integrate qualitative date with quantitative data [19]; 4) It can translate the descriptive language of experts into precise numerical value by mathematics method [20]. As a result, the integrated fuzzy evaluation method is more widely applied. However, there are also some problems as follows. 1) There is no systematic theory for the selection of soil fertility index, and the selection depends on personally subjective judgment. In all researches related with soil fertility assessment, the selections of soil fertility index are different, and the number of soil fertility index changes from 4 to 20. 2) The soil fertility assessment does not focus on detailed crop. Many researches about the soil fertility evaluation do not consider the effect of planted crop, which makes that the evaluation result can not reflect the regional actuality. 3) There is few research on integrated fuzzy evaluation method used in soil fertility assessment. Dongting Lake wetland area is an important base of many crops, including rice, wheat, cotton and corn. As a result, the research on soil fertility assessment is of great significance. The research on the soil fertility evaluation of Dongting Lake wetlands is a try in this field.

Based on the above analysis, considering the particularity of Dongting Lake wetland area, the wetland soil fertility was evaluated by means of integrated fuzzy assessment. In the research, paddy was regarded as an important factor affecting the selection of soil fertility factor, and the determination of weight of each factor. The objectives of the research are: 1) to explore the method and index system of soil fertility evaluation fit for Dongting Lake wetland area, 2) to investigate the spatial distribution of soil fertility properties, and 3) to discuss the relationship between soil fertility and paddy yield. The research result will provide a theoretical base for decision makers and agricultural policy.

2 Study site and methods

2.1 Study site

In this study, the research area related with the Dongting Lake wetland area is about 39 107 km², located in North of Hunan Province of China and ranges from N27°54′ to N29°57′ and from E110°52′ to E113°46′ (Fig.1). The climate type belongs to a transitional region of middle-subtropics to north subtropics, where the annual temperature is 17 °C and the annual rainfall is   1 400-1 500 mm. In Dongting Lake wetland area, the main physiognomy type is alluvial plain of river and lake, and paddy field soil is the dominating type. Generally, for paddy field soil, the humus content is abundant, and the soil fertility level is high. As a result, the Dongting Lake wetland area which belongs to a typical double- cropping paddy area of China is a famous and important commodity grain production base, and also one of nine industrial belts of top agricultural product (belonging to top grain industrial belt of double-cropping paddy in Yangtze River), whose farmland area owns 23.3% of the total in Hunan Province. In Dongting Lake wetland area, the area of paddy plantation is about 116×104 hm², and the average yield of paddy achieves 7.2×10⁸ kg. However, as the long-term tillage disturbance and using intensively and maintain slightly, the land ecological balance and soil nutrient cycle are destroyed and the soil fertility descended, which have become an obstacle of agriculture development of Dongting Lake wetland area. Meanwhile, the researches on soil fertility of Dongting Lake wetland area are mainly focused on single factor analysis, including the spatial distribution of soil organic matter, and the relationship between soil animal and soil fertility [21]. However, the research on integrated fertility assessment of this area is infrequent.

Fig.1 Location of study area

2.2 Data sources

The data used in this research comes from the result of the Second National Soil General Survey. By collecting the data of the Second National Soil General Survey, all the data, including soil organic matter, total nitrogen, total phosphorus, total potassium, pH value, available phosphorus, and available potassium, were treated by ArcGIS software to obtain the registration and vectorization of map.

2.3 Assessment methods

2.3.1 Selection of soil fertility index

How to select the representative soil fertility index will, to some extent, influence the result of soil fertility assessment. All the indexes should cover the basic properties of soil, which should also be related with the soil fertility. Theoretically speaking, the more general the evaluation index related with the assessment index system, the more precise the assessment result compared with the actuality, and the more the money and time will be consumed. As a result, the most perfect thing is to select a minimum dataset of typical indexes affecting the soil fertility. Based on the above consideration, in this research, seven typical indexes, including soil organic matter, total nitrogen, total phosphorus, total potassium, available phosphorus, available potassium and pH value, were selected.

1) Soil organic matter

The soil organic matter (SOM) is an important source of soil nutrient elements. The colloid properties made the SOM absorb more cations, which leads to soil with specialties of maintaining soil moisture and soil fertility, cultivation, buffering and aeration. Hence, the SOM can improve the soil physical traits. At the same time, the SOM is the source of soil carbon. As a result, the content of SOM is an important indicator of soil fertility.

2) Total nitrogen

The soil total nitrogen is an important index for assessing the soil fertility level, which is a symbol of supply ability of soil nitrogen to a certain extent. The ebb and flow of total nitrogen depend on the relative intensity of nitrogen accumulation and consuming. The inorganic nitrogen and organic nitrogen reflect the temporal and potential abilities of soil fertility level, and the nitrogen distribution, soil nitrogen fixation and release are indicators of soil fertility level. Some results showed  that, with the increase of fertilization nitrogen, the biomass and the SOM accumulation increased [22]. Meanwhile, the nitrogen is one of the most important indexes affecting rice yield, and the paddy is the widest crop in Dongting Lake wetland area. Hence, the selection of total nitrogen is necessary.

3) Total phosphorus and total potassium

Phosphorus and potassium are main nutrient elements for plant’s growth. Although what the plant absorbs directly is available phosphorus and available potassium, total phosphorus and total potassium are the storages of available phosphorus and available potassium, and the total phosphorus and total potassium are directly related with the plant growth and crop yield. Therefore, the index system should include the total phosphorus and total potassium.

4) Available phosphorus and available potassium

The plant can directly absorb available phosphorus and available potassium, which has a close relation with photosynthesis. The available phosphorus of soil can reflect the soil phosphorus supply level, and manage the fertilization, and also is one of the indexes judging the soil available fertility. The available potassium is an important fertility index reflecting soil potassium supply ability. As a result, the available phosphorus and available potassium should be considered in the assessment system.

5) pH value

The soil pH value is an indicative index reflecting the soil fertility and affecting the soil cation exchange capacity. The soil pH value is very important for paddy seeding and germination [23]. The endurance range of pH value is from 4.5 to 8.5, and the most suitable pH value changes between 6.0 and 7.0. Hence, the soil pH value should be considered.

2.3.2 Computation of soil fertility index membership grade

The relationship between soil fertility index and crop growth and development presents two types, including S-type curve and parabola type curve (Fig.2). Except that the pH value index belongs to parabola type curve, the other selected indexes take on the S-type curve. According to the curve type related to each index, the primary data are divided into five classes, from class 1 to class 5. The bigger the class type, the worse the nutrient level. And the soil fertility index membership grades are evaluated to be 1, 2, 3, 4 and 5, respectively (Table 1). As the limitation of data source, the pH value index is only classified into three levels, including class 1, class 3 and class 5.

2.3.3 Weight determination of each soil fertility index

The weight of soil fertility index is the effect degree of each index on the soil fertility, which can reflect the difference of each fertility index. The weight determination of each monomial fertility index is a bottleneck problem of soil fertility assessment. At present, there are many kinds of methods of weight determination, including method of Delphi (point ranking by expert), method of factor analysis, method of correlation coefficient, etc [24]. In this research, considering that the method of correlation coefficient can obtain an impersonal weight and avoid the subjectivity of researcher, correlation coefficient method is adopted to determine the weight of each index. The detailed process of weight determination is as follows: 1) calculating the correlation coefficient calculation between each monomial fertility index (Table 2); 2) computing the mean value of correlation coefficient absolute value of any index to other indexes; 3) calculating the percentage of mean value of any index to the sum of all mean values, which will be regarded as the weight of any monomial index in soil fertility system (Table 3).

Fig.2 S-type (a) and parabola type curve (b) representing relationship between soil fertility index and crop growth and development

Table 1 Evaluation system on soil fertility in Dongting Lake wetland area

Table 2 Single correlation coefficient between each index

Table 3 Mean value of correlation coefficient and weight value of each index

2.3.4 Calculation of soil fertility index

In this research, under the support of spatial analysis module of ArcGIS software, the integrated fertility index can be calculated by

                                                                ⁽¹⁾

where I_F is the integrated fertility index value, Q_i is the membership grade of soil fertility index i, and W_i is the weight value of soil fertility index i.

Using the ArcGIS software, under the support of the spatial analysis module, the vector map of seven assessment indexes in the study site is transformed into many grids of 50 m×50 m, which is the soil fertility evaluation cell, and the membership grade of each index is evaluated for each grid. Then, according to Eq.(1), the integrated assessment value of each grid for seven soil fertility indexes can be obtained by raster calculation of spatial analysis module of ArcGIS software.

The distribution extent of calculated integrated fertility index in Dongting Lake wetland area changes from 0 to 5. The higher the value, the better the soil fertility. In this research, the soil fertility of Dongting Lake wetland area is divided into four classes by the standard (Table 4), and Class I is the best, Class II better, Class III less, Class IV the least.

Table 4 Classification standard of soil fertility class

3 Result

Under the support of geographical information system (GIS), using the raster calculation function of spatial analysis module, the soil fertility spatial distribution of Dongting Lake wetland area (Fig.3) can be obtained. Based on Fig.3, it can be seen that main soil class in Dongting Lake wetland area is Class III, whose area achieves 1 6776 km², owning 42.9% of total area of research site. Areas of Class II and Class IV are less, and the area of Class I is the least, only 1 759.82 km². The result shows that, although the Dongting Lake wetland area is an important commodity grain production base of China, the whole soil fertility is not very good, and the soil fertility class of more than 60% area in research site is below Class II. The area of soil fertility Class I only owns 4.5% of the total (Table 5). As a result, further measures should be taken to improve the soil fertility.

Fig.3 Soil fertility class map of Dongting Lake wetland area

Table 5 Area and percentage of each soil fertility class

3.1 Class I soil

The area of Class I is 1 759.82 km², covering only 4.5% of the whole Dongting Lake wetland area. The Class I soil distributes in the northeast part of Linxiang county, the northeast part and south part of Taoyuan county, central part of Changde city, the northwest part of Ningxiang county, a few area in Linli county and Xiangyin county. In the northwest part of Ningxiang county and Taoyuan county, the soil type belongs to red soil, whose land use type is the forest. The soil type in the south part of Taoyuan county is yellow soil, whose land use type is grassland. The soil type for other Class I soil is paddy field soil.

3.2 Class II soil

The area of class II soil is about 10 011.39 km², owing 25.6% of the whole Dongting Lake wetland area. The class II soil distributes in circumjacent area of Class I soil, including the north part of Linxiang county, the central part of Xiangyin county, the north part of Taoyuan county, Changde City, the central part of Hanshou county, the central and south part of Yuanjiang city, the central part of Linli county, the north part of Anxiang county, the central part of Nanxian county and the west part of Ningxiang county. For the soil type, the Class II soil in Linli county, Taoyuan county and Ningxiang county is red soil, with the forest of land use type, and the other places of Class II soil have the paddy field soil. The area of paddy field soil for Class II soil is larger than that of red soil, which shows the soil fertility level of paddy field is better than that of red soil. For the paddy field, as the cultivation pattern is the precision plantation, the soil management is suitable, which makes the soil fertility level better. For the red soil, although the land use type is the forest, the litter can meliorate soil; with the intensive eluviations of rainfall, the soil fertility is much lower.

3.3 Class III soil

The distributed area of Class III soil is the widest, achieving 16 776.9 km², owing 42.9% of the whole Dongting Lake wetland area. The Class III soil distributes in each county, and mainly in Anxiang, Hanshou, Xiangyin, Linli and Yuanjiang county. The Class III soil along the Dongting Lake wetland area is mainly paddy field soil. As the long-term cultivation and insufficiency of fertilizer, the soil fertility level locates in low level, which is in accordance with the actuality of annually rice yield.

3.4 Class IV soil

The area of Class IV soil is about 1 0519.78 km², owing 26.9% of the whole Dongting Lake wetland area, and distributing in Anxiang county, Yuanjiang city, Miluo city, the south part of Yueyang City, Yueyang county, the central part of Taoyuan county, Taojiang county and Wangcheng county. And Anxiang, Yuanjiang, Miluo and Yueyang counties are the main regions of Class IV soil, whose soil type is paddy field soil. The Class IV soil for Taojiang county, Wangcheng county and the central part of Taoyuan county distribute randomly, and the soil type is red soil.

4 Discussion

The main soil types in Dongting Lake wetland area are red soil and paddy field soil. The paddy field soil locates in the central and west part of Dongting Lake wetland area, whose soil fertility class mainly belongs to Class I, Class II and Class III. The soil type in the east part and south part is red soil, whose soil fertility classes are mainly Class III and Class IV. The soil fertility of paddy field soil is better than that of red soil, which is in accordance with the actuality of study site.

The soil fertility of main soil types of Hunan Province shows that the main soil fertility class is Class III, the contents of soil organic matter, total nitrogen, and total potassium are larger, and the total phosphorus content is less. The result is similar to that obtained in Dongting Lake wetland area. LUAN et al [25] researched the degeneration process of soil fertility of Sanjiang Plain area, and drew a conclusion that the main soil classes in research site are Class III and Class IV, and there is no Class I and Class II as a whole, which is also in accordance with the research result in Dongting Lake wetland area. EDEM and NDAEYO [26] also discussed the soil fertility status of three soil types in Nigeria by collecting soil samples, and obtained similar result that the soil fertility status in study area is low.

At the same time, the achievement comparison of soil fertility research to the annual mean rice yield of Dongting Lake wetland area shows that the soil fertility research result is basically coincident with the spatial distribution of rice yield. If the fertility index is high, the rice yield is also high. For example, in Linli County and Lixian County with high soil fertility index, the annual mean rice yields achieve 8 025 kg/hm² and 6 654 kg/hm², respectively. But in Nanxian County and Yiyang City with low fertility index, the annual mean rice yields are only 6 045 kg/hm² and 5 872 kg/hm², and the maximal gap achieves 2 153 kg/hm². In Changsha City, the soil fertility index is not high, but with good economic condition and huge supply of fertilizer, the rice yield still reaches 7 170 kg/hm² (Table 6).

Table 6 Annual mean rice yields of important region in Dongting Lake wetland area

5 Conclusions

1) By using the fuzzy integrated evaluation method to assess the soil fertility of Dongting Lake wetland area, the result shows that the soil fertility level is not good in the research site. The soil fertility indexes of the central part and west part of Dogting Lake wetland area are better than those of the east part and south part. Among the two main soil types of paddy field soil and red soil, the fertility of paddy field soil is better than that of red soil.

2) The areas of Class III and Class IV own 69.8% of total Dongting Lake wetland area, which shows that it is very necessary to take integrated countermeasures to meliorate the paddy field of middle and low yield.

3) There are many advantages for the fuzzy method to evaluate the soil fertility. Considering its predominance, the fuzzy method is adopted. Integrated with geographical information system (GIS), reasonable index is selected and the weight of each index is determined according to the actuality of Dongting Lake wetland area, and a set of method and index system of assessing soil fertility is put forward. The results show that the fuzzy method can be efficiently used in evaluating the soil fertility. The method and index system can be applied to other similar soil to assess the soil fertility. Therefore, the method could be very interesting to policy makers involved in regional soil fertility assessment, because it can allow decision makers to clearly know the current status of the quality of their regional environment.

References

[1] PANDIT A K. Conservation of wildlife resources in wetland ecosystems of Kashmir, India [J]. Journal of Environmental Management, 1991, 33(2): 143-154.

[2] BAI Xue, MA Ke-ming, YANG Liu, ZHANG Jie-yu, ZHANG Xiao-lei. Differences of ecological functions inside and outside the wetland nature reserves in Sanjiang Plain, China [J]. Acta Ecologica Sinica, 2008, 28(2): 620-626. (in Chinese)

[3] WANG N M, WILLIAM J, MITSCH A. Detailed ecosystem model of phosphorus dynamics in created riparian wetlands [J]. Ecological Modelling, 2000, 126(2/3): 101-130.

[4] XU F L, ZHAO Z Y, ZHAN Wei, ZHAO Shan-shan, DAWSON R W, TAO Shu. An ecosystem health index methodology (EHIM) for lake ecosystem health assessment [J]. Ecological Modelling, 2005, 188(2/3/4): 327-339.

[5] STEVER G D, LLEWELLYN D W. Coastal wetlands planning, protection, and restoration act: A programmatic application of adaptive management [J]. Ecological Engineering, 2000, 15(3/4): 385-395.

[6] PASCOE G A, BLANCHETR R J, LINDER G. Ecological risk assessment of a metals-contaminated wetland: Reducing uncertainty [J]. Science of the Total Environment, 1993, 134(Supplement 2): 1715-1728.

[7] XU Xue-gong, LIN Hui-ping, FU Zai-yi, BU Ren-cang. Regional ecological risk assessment of wetland in the Huanghe river delta [J]. Acta Scicentiarum Naturalum Universitis Pekinesis, 2001, 37(1): 111-120. (in Chinese)

[8] JONES K, LANTHIER Y, VOET P, VALKENGOED E,TAYLOR D, FERNANDEZ P. Monitoring and assessment of wetlands using earth observation: The glob wetland project [J]. Journal of Environmental Management, 2009, 90(7): 2154-2169.

[9] FAULWETTER J L, GAGNONB V, SUNDBERGC C, CHAZARENC F, BURR M D, BRISSON J, CAMPER A K, STEIN O R. Microbial processes influencing performance of treatment wetlands: A review [J]. Ecological Engineering, 2009, 35: 987-1004.

[10] SAHRAWAT K L. Soil fertility advantages of submerged rice cropping systems: A review [J]. Journal of Sustainable Agriculture, 2007, 31(3): 5-24.

[11] KARLEN D L, TOMER M D, NEPPEL J, CAMBARDELLA C A. A preliminary watershed scale soil fertility assessment in north central Iowa, USA [J]. Soil and Tillage Research, 2008, 99(2): 291-299.

[12] ZHANG Hai-bo, LUO Yong-ming, ZHAO Qi-guo, ZHANG Gan-lin, HUANG Ming-hong. Hongkong soil researches VI: Integrated evaluation of soil fertility quality based on the improved analytic hierarchy process [J]. Acta Pedologica Sinica, 2006, 43(4): 577-583. (in Chinese)

[13] WANG Jun-yan, ZHANG Feng-rong, WANG Ru, JIA Xiao-hong, ZHANG Cai-yue. Application of integrated fertility index method in evaluating changes in soil fertility [J]. Rural Eco-environment, 2001, 17(3): 16-20. (in Chinese)

[14] XU Yong-mei, WANG Jiang-li, LIU Hua. Evaluation of grey desert soil fertility by grading method [J]. Chinese Journal of Soil Science, 2005, 36(4): 465-468. (in Chinese)

[15] KAUFMANN M, TOBIAS S, SCHULIN R. Quality evaluation of restored soils with a fuzzy logic expert system [J]. Geoderma, 2009, 151(3/4): 290-302.

[16] CHEN Liu-mei, GUI Lin-guo, LV Jia-long, WANG Chong-guang, LI Zheng-zhong, WANG Zeng, SUN Rong. Evaluation on soil fertility quality of newly cultivated light sierozem under different fertilization with methods of principal component and cluster analyses [J]. Soil, 2008, 40(6): 971-975. (in Chinese)

[17] CORWIN D L, KAFFKA S R, HOPMANS J W, MORI Y, GROENIGEN J W, KESSEL C, LESCH S M, OSTER J D. Assessment and field-scale mapping of soil fertility properties of a saline-sodic soil [J]. Geoderma, 2003, 114(3/4): 231-259.

[18] BALAS C E, ERGIN A, WILLIAMS A T, KOC L. Marine litter prediction by artificial intelligence [J]. Marine Pollution Bulletin, 2004, 48: 449-457.

[19] SALSKI A, FRANZLE O, KANDZIA P. Introduction to special issue fuzzy logic in ecological modeling [J]. Ecological Modelling, 1996, 85: 1-2.

[20] BORRI D, CONCILIO G, CONTE E. A fuzzy approach for modelling knowledge in environmental systems evaluation [J]. Computers, Environment and Urban Systems, 1998, 22: 299-313.

[21] FU Xiu-qin, ZHANG Zhi-gang, HU Bo, ZHANG Wei, YAN Heng-mei. Relationship between soil fauna community characteristics and soil fertility in Dongting Lake wetland [J]. Research of Agricultural Modernization, 2007, 28(6): 685-687. (in Chinese)

[22] YANG Rui-ji, YANG Qi-feng, NIU Jun-yi. Research progress on soil fertility major indexes [J]. Journal of Gansu Agricultural University, 2004, 39(1): 86-91. (in Chinese)

[23] QUAN Song-hua, WANG Yong-guo, LUO Shao-qiu, TANG Fang-lan. Effects of the pH value of bed soil on seedling quality in rice [J]. Hybrid Rice, 2004, 19(4): 45-46. (in Chinese)

[24] LI Fang-min, ZHOU Zhi-an, AI Tian-cheng, YUAN Xiong-ren. Integrated evaluation of waterlogged soil fertility: The case of Gaochang farm in Qianjiang city, Hubei Province [J]. Resources Sciences, 2002, 24(1): 25-29. (in Chinese)

[25] LUAN Zhao-qin, SONG Chang-chun, DENG Wei. Study on soil fertility variation during wetland reclamation and utilization in the Naoli River watershed of the Sanjiang Plain [J]. Journal of Jilin Agricultural University, 2003, 25(5): 544-547, 556. (in Chinese)

[26] EDEM S O, NDAEVO N U. Fertility status and management implications of wetland soils for sustainable crop production in Akwa Ibom State, Nigeria [J]. Environment, Development and Sustainability, 2009, 11: 393-406.

(Edited by YANG Bing)

Foundation item: Projects(40971170, 51039001) supported by the National Natural Science Foundation of China; Project(2007AA10Z222) supported by the National High Technology Research and Development Program of China

Received date: 2010-07-27; Accepted date: 2010-11-18

Corresponding author: LI Zhong-wu, Professor, PhD; Tel: +86-731-88640078; E-mail: lizw@hnu.cn