Abstract
Turfy soil is a kind of special soil accumulated by undecomposed plants which is detrimental to the engineering. In this paper, particular identified patterns for the turfy soil in the northeast of China was raised including the method of extracting threshold value, block analysis and fuzzy evaluation. And field investigations were undertaken to verify the accuracy of identification by remote sensing, and the correlation of field result and remote sensing result was summarized so as to analyze the regularities of distribution and evolutionary mechanism of turfy soil. The result shows that the combination of extracting threshold value, block analysis and fuzzy evaluation are effective methods to predict the distribution of the turfy soil; with the correlation of membership degree and field result, we can analyze the evolutionary mechanism of turfy soil affected by both nature factors and human activities, which is beneficial for the preservation of the turfy soil and also shows significant environmental ecological benefit.
Similar content being viewed by others
1 Introduction
Turfy soil is a special kind of humus soil formed by the plant remains in the swamp area [29]. Similar to other normal peat soil in China [12, 33], while turfy soil has more notable characteristics of high void ratio [31], high compression [18, 30], high moisture [6], low shear strength [44] and high organic content [37], which cause it more detrimental to the engineering construction [1, 8, 36], while the project of road transportation though the turfy soil area is inevitably associated with the development of economy.
Therefore, identifying the turfy soil by remote sensing is of great value not only for the route selection and design but also for the engineering construction of turfy soil area. On the other hand, the turfy soil swamp is a kind of swamp while the swamp is an important part of wetland so the essence of preserving turfy soil swamp is to preserve the wetlands as well as the diversity of ecosystem [13, 35], which is of great ecological effect [9].
Recognition program for wetland, especially for turfy soil, remain largely uncertain [24]. Remote sensing driven model was proposed in analyzing effect of land use on soil moisture in the Weihe river basin by Wang Y.J. et al. [45], and C-Band SAR was applied to wetland monitoring at high latitudes by Reschke, J., A. et al. [40]. Satellite data have been shown suitable for moisture retrieval of near-surface soil [5]. Turfy soil, formed in the low-lying basin, could be preliminary recognized by remote sensing for the special landscape [30]. TerraSAR-X data was used for soil moisture retrieval by Bai X. et al. [2], and remote sensing was also applied in estimation of soil surface sealing effect on soil koisture by Sela S. et al. [42]. Estimating the surface water and analyzing the change of the moisture in the soil [45] by remote sensing, can be a basis for the analyzing the evolution of turfy soil [7, 46]. And distribution of the gray value of image is Gaussian fitted [22] so as to be the basis of further study.
Support vector machine, artificial neural network, and spectral angle mapper algorithms were selected for crop classification by Kumar et al. [27], and terrain variables and object-based machine learning classification was used in differentiating similar land cover by Maxwell et al. [34]. Method such as object-based classification and machine learning classification [15, 39] was introduced into mapping land cover. Jenicka, S. et al. make land cover classification by fuzzy texture model and support vector machine with remotely sensed images [24], and high resolution satellite images was used in classification based on fuzzy clustering by Singh et al. [43]. And for turfy soil, fuzzy evaluation is an effective method applied to the doline susceptibility mapping [23] and the classification of land-cover [11], such as prediction of soil by membership degree [44], which can be a quantitative expression of evolution of turfy soil, and drilling, an effective method for strata revealment, was adopted to verify the result of remote sensing. Geophysical prospecting is always an auxiliary method for detecting strata and groundwater [17], especially high density electrical technique, which was apply to the investigation for groundwater potential evaluation by Osinowo Olawale et al. [38], and as for turfy soil, a special soil with high moisture content, can be identified by electrical resistivity.
The study area is located in the northeast of China, with annual precipitation of 500-900mm [50], enough for the formation of the turfy soil in terms of the water condition; This area is also located in the south of Dunmi fault zone, and the tectonic movement in Mesozoic results in the imbalanced subsidence, showing mosaic distribution of plateau and basin [19]. And long-term flooding in the basin is a key point for the formation of turfy soil, different aquifers formed with different landform [14] and different soil layer [28], which make the hydraulic connection in the area complicated [10, 52]. The study area is located in seasonal frozen areas, with the average frost line of 1.5-2.1m, preventing water from infiltrating [16], which reduces the oxidation of the vegetation and promotes the formation of turfy soil. All of these conditions make the turfy soil distribute in this area possible, and the regularities of distribution have been further studied in the test, and the specific location related is marked as Fig. 1.
2 Related work
For turfy soil, Lei Nie et al. studied the influence of organic content and degree of decomposition on the engineering properties [37], and mineral distribution of turfy soil was studied by Zhandong. S. [29], showing the importance of turfy soil in engineering, while there is little study on the distribution of turfy soil.
Vegetation coverage index is an common ways in vegetation extraction [25]. Plants providing enough organic matter is the premise of the formation of turfy soil. While vegetation coverage index is not so fit for the identification of turfy soil. On the one hand, there is much similar between the turfy soil with other landscapes such as forests and farmlands, which can hardly distinguished by vegetation coverage index as well as threshold method [26]. On the other hand, threshold method have been widely applied in cloud detection, road recognition and vegetation identification [4, 32, 41, 47]. Long-term flooding is another premise of turfy soil formation. Turfy soil area should be seen as a mixture of water, plants and soils, and as a result, a special recognition model should be established.
Classification method should be introduced in the recognizing of turfy soil. Support vector machine(SVM), neural net algorithm and fuzzy widely applied in classification [11]. Support vector machine (SVM) is a method which change non-linearly separable area into linearly separable by mapped it to high-dimensional space. He T. et al. Have applied it to the land use/cover classification [20]. By repeatedly learning of neural network to obtain the optimal convergence result, is also an commonly used methods in land cover classification. Bocco et al. Applied it into estimating crop residue cover by crop residue index multiband models [3]. And Zakeri et al. proposed adaptive method by curvelet transform and neural network in synthetic aperture radar images [51]. Fuzzy mathematics describe the classification result by degree of membership, Zhu A.X et al. have applied it into prediction of soil properties membership values [53].
It can be seen that for the specificity of turfy soil, the existing identification mode is not fitted for it, and for the significant environmental ecological benefit, it urge to establish a special identification mode for turfy soil. The framework of the manuscript has been list as Fig. 2. Distribution and evolutionary of turfy soil is an important issue because of both environmental and ecological benefit. On the one hand, the turfy soil always exits with swamps, which is difficult for instance validation, on the other hand, for regional survey, it is impossible to do the survey for every turfy soil swamp, So it is effective to recognize it by remote sensing. In order to analyze the evolutionary mechanism of turfy soil, membership degree of fuzzy evaluation was introduced, and the method of qualitative identification, threshold method, block analysis and fuzzy evaluation were developed for identification of turfy soil. To recognize every pixel, the programs was written. And finally, turfy soil swamps corresponding to the natural evolution of turfy soil and the turfy soil affected by human activities separately, were selected for validation of recognition method and analyzing the evolutionary of turfy soil. Xu et al. [21, 48, 49] try to process surveillance images based on the video structural description technology.
3 Method
3.1 Qualitative identification of turfy soil
Qualitative identification is the basis of automatic identification. Topography and landforms likely to forming the turfy soil, enough vegetation accumulating organic matter as well as long-term flooding inhibiting decompose of organic mater, was all considered in this part. As shown on Fig. 3a, the distribution area of the turfy soil was circled in red, while the roads are more bright and the surrounding mountains are dark green. Taking into account of the topography, the turfy soil is mainly distributed in low-lying area with mountains surrounded, and this is because water can easily remain for a long time, which would inhibit oxidative decomposition of plants, and finally the turfy soil was formed. As shown on Fig. 3b, low-lying terrain, water retention and enough vegetation make the formation of the turfy soil in this area possible.
3.2 Threshold Gaussian fitted
For quantitative identification, a remote sensing image can be described as an three-dimensional matrice, while a great deal of calculations was needed, taking into consideration that the remote sensing image of each landforms, turfy soil for example, should be seen as a whole, it is necessary to simplify the matrice. Three spectra of red, green and blue was selected in consideration of the recognition of both water and plant. The remote-sensing image of turfy soil of more than 10,000 pixels was obtained from Google Earth for statistics. For each spectrum, set pixel value as X axis, and accumulated the number of each pixel value and set the normalized value of number as Y axis. The correlativity of X and Y was Gaussian fitted, and the distribution of the pixel value of a remote sensing image, an area of turfy soil for example, could be described by only 2 parameters of the means and the standard deviations. For each spectrum, it is described with a formula as follow:
In order to exclude the influence of special points, set 90 % of confidence intervals for extracting threshold value, that is:
So the threshold value of turfy soil can be obtained as follow:
3.3 Block analysis
Taking into account of the formation of the turfy soil that the turfy soil is generally accumulated in an area abundant with water and organic matter rather than dispersion point, it is better to recognized it by a small pixel matrix rather than a single pixel. In the pixel matrix of remote sensing image, numerous adjacent pixels are combined into many small matrices, which unite together to form a big matrix. Described as:
Where p = m/d, q = n/d. Because m,n> > d, indivisible part can be ignored.
3.4 Fuzzy evaluation
On the one hand, turfy soil can not be distinguished when it is surrounded by farmlands and lakes through threshold method because of the similar spectral features, and the fuzzy evaluation is an effective method to solve the problem. On the other hand, the turfy soil is forming and disappearing constantly, so utilizing the membership degree of fuzzy evaluation to study the distribution and evolution is more significative than just to recognize turfy soil area.
All of possible results of landscapes recognized by fuzzy evaluation can be described as follow set:{turfy soil; farmlands; lakes; mountains}, and the influencing factors can be replaced by different spectra, determined as {R; G; B}. As shown on Fig. 4, threshold value of different landscape corresponds to a certain spectrum, taking G for instance, is normal fitted to determine the membership function. Membership functions was identified as linear and piecewise one, described as:
In order to make a better recognition of the turfy soil, certain weight for certain spectrum should be determined. A certain spectrum correspongding to a certain landscape can be described with only 2 parameters: the means and the standard deviations. We can see from the Fig. 4 that it is better for identification with less standard deviations and great difference between each means. So a parameter called Rs was raised for the weight of each spectrum, defined as:
The results of Rs are different with different spectrum, and the normalization of three Rs could serve as the weights of the three spectra (R, G and B) respectively. The formula of normalization is listed as follow:
The membership degree of predicting area depend on the membership function and weights of the three spectra. By multiply-add, we can get the formula for the membership degree:
4 Programs written
For a lot of computation, programs were needed. Program of turfy soil is a pattern of pre-deduction of unknown area by known data, So known remote-sensing image of turfy soil should be prepared, and at least two of confusing area was also needed. For more accurate, remote-sensing image should be in same scale and taking time. As well, pixel numbers should be consistent (at least consistent of rows for each class of remote-sensing image) so as to make matrix superposition for Gaussian fitted.
The identification programs of turfy soil considered the following several aspects mainly:
-
1.
As shown in Fig. 5a, three spectra of R, G and B were chosen for Gaussian fitted, image was described by the two parameters of the means and the standard deviations. And remote sensing matrix of turfy soil in different area could be moved into a vicious cycle.
-
2.
As shown in Fig. 5b, formula was replaced by programming language, and so as the segmented membership functions (Fig. 4), As shown in Fig. 5c.
-
3.
Means of reflectivity of different landscapes were uncertain (Fig. 4), after sorting for membership functions fitted, we need to restore the original order, and statements in Fig. 5d were list to solve the problem.
-
4.
The result of membership degree can be obtained, and distribution of turfy soil was determined by both threshold method and fuzzy evaluation, as shown in Fig. 5e.
-
5.
In consideration of the difference of turfy soil swamps and turfy soil forests, the program was little changed with the landscape result of 2 different turfy soil and 2 confusing area, And the Graphical User Interface of identification of turfy soil was finished as shown in Fig. 5f.
Automatic identification of turfy soil by this appletapplet, not only for convenient, but also solve the problem that most of the turfy soil swamp can not be instance validated. This model has simple structure and high operability. Introduction of degree of membership, rather than identify the area of turfy soil, it is more significant to obtain the possibility of turfy soil area, which change the point of “whether it is turfy soil” into “how likely it belong to turfy soil”, and this is of great significance for identification of turfy soil degradation as well as protection of turfy soil swamp. This is a dynamic identification model, identification results can be a new data source after instance validation, and the threshold of turfy soil, the weight of different spectral etc., will be optimized as well.
5 Instance validation
Identification of turfy soil is a process from qualitative to quantitative, manual identification is the premises of automatic identification. In order to verify the accuracy of identification mode, two area of Haerbaling and Jiangyuan was chosen for automatic identification.
Haerbaling is located in the delta in the east of the study area (marked on Fig. 1b) with perennial water, with the special landscape that it is surrounded by highlands on three sides, forming a natural waterlogged depression, meeting the condition of the formation of turfy soil. Haerbaling is far away from the villages and roadways, which show little influence on turfy soil, can be a special instance of the natural evolution of turfy soil. Jiangyuan is located on the southeast of Haerbaling (Fig. 1). Turfy soil is quite likely to form in this area because of the characteristics of the landscape, which could collect the precipitation from the surrounding mountains. Roadway of G201 go through the basin, the construction of engineer and the the operation of the highway will have an impact on the formation and evolution of turfy soil, this is a special instance of the turfy soil affected by human activities.
5.1 Case 1 Haerbaling
The remote sensing image was shown in Fig. 6a. The threshold method and the fuzzy evaluation was used for the recognition of turfy soil, and the result of the distribution of turfy soil was predicted as white areas on Fig. 6b. Membership degree of each pixel belonging to the turfy soil was showed on Fig. 6c. It turns out to be the brighter of the area, the higher for the membership degree to the turfy soil. The scene photo of turfy soil in Haerbaling are shown on Fig. 6d, with perennial water and undecomposed plants in the area, and the actual distribution of turfy soil was circled in red on Fig. 6a.
5.2 Case 2 Jiangyuan
The aerial views of the area was shown on Fig. 7a. Threshold method and block analysis was adopted and the result of recognition was shown on Fig. 7b. Taking into account that the area B (Fig. 7a) may be a farmland judged by the texture, the fuzzy evaluation was introduced and the result was shown on Fig. 7c. In order to make it clearly of the distinction of area A and B (Fig. 7a), drilling was taken and the position of 5 drill holes was marked on Fig. 7a, and the membership degree of this special area was showed on Fig. 7d so as to make comparison of drill-hole data and remote sensing results. The scene photo of turfy soil in Jiangyuan are shown on Fig. 7e.
6 Analysis and discussion
Comparing Fig. 6a with Fig. 6b, we can make a conclusion that the combination of those methods (qualitative identification, threshold method, block analysis and fuzzy evaluation) is an effective means for recognition of turfy soil. Qualitative identification of turfy soil in consideration of the origin mechanism and also whether there are enough given points for fuzzy evaluation and Gaussain fitting, will all determine the accuracy of turfy soil recognition.
6.1 Haerbaling: an instance of the natural evolution of turfy soil
As shown on Fig. 6c, the membership degree of area A and B (Fig. 6a) turns out to be different, these differences indicate the significance to consider the evolutionary mechanism and just recognizing the turfy soil is somewhat thoughtless. A cross section was illustrated on Fig. 6a, the elevation was measured every 6 meters so as to meet the membership degree of pixels (with 6 meters for a pixel on average). The membership degree and the corresponding elevation point were extracted. The normalized elevation and membership degree of the cross section are compared on Fig. 8a, the membership degree of turfy soil shows obvious regulation with the elevation in spite of the dispersion in membership degree, and the cross section was divided into 3 kinds of landscapes taking into account of elevation and membership degree.
It shows that with the water can be easier gathered at a lower elevation which is a requirement for the formation of turfy soil, and that is part of the reason why there is no turfy soil in mountains. Lakes shows a smaller membership degree according to remote sensing because of less vegetation, which leads to no distribution of turfy soil as well. Or we can discuss it from another perspective, as shown on Fig. 8b, the relation of membership degree and normalized elevation was shown on the figure. the membership degree firstly increases and then decreases with the increasing of elevation, which can well reflect the evolution of turfy soil. The main point is the water, an area with tectonic subsidence may accumulate the water, then the plant developed and finally the turfy soil was formed. And relatively, the decreasing of water would also lead to the degradation of turfy soil. This evolution mechanism may be great helpful for the protection of the turfy soil marshes as well as wetlands.
6.2 Jiangyuan: an instance of the turfy soil affected by human activities
the result of recognition in Fig. 7b shows that there are similar spectra between turfy soil and farmland, and it can also explain the evolution between turfy soil and farmlands. Figure 7d shows that the result may be not clear enough because of the accuracy of remote sensing image, but the follow analyses may make the regularity much clearer.
The thickness of turfy soil in each drill hole measured by drilling was shown in table 1, with the corresponding membership degree. And in this part, considering there might be some coincidences as the drill holes are relatively inadequate. So to avoid coincidences, the membership degree adopts the average of a square matrix of 5*5 with the target point in the center instead of just the only one exacted point. There is a village name Hanconggou in the south shown in area C (Fig. 7a), and the nearest distance from each drill hole to the village boundary was measured and listed in table 1.
The relation of membership degree and thickness of turfy soil was plotted on Fig. 9a, and the result shows that the membership degree are positively associated with thickness of turfy soil, a conclusion can be made that membership degree, instead of thickness measured by drill, could be described for the distribution of turfy soil to some extent.
The membership degree was taken instead of measured thickness for the influence of village, and the correlation between membership and the distance to village was shown on Fig. 9b. A curve fitted for these measured point, traveling through the point of (0,0) and (+ ∞,1) theoretically, shows that village has great influence on the existence of turfy soil. That is because the turfy soil is abundant with organic matter, which is conducive to the crops, resulting in turfy soil exploited into farmlands. As shown on Fig. 9b, in terms of the turfy soil in Jiangyuan area, There is little turfy soil existed within 500m surrounds of the village, and the village may hardly have influence on turfy soil with the distance of more than 1800m.
State road of G201 go through the turfy soil area of Jiangyuan, as seen on Fig. 10. In order to ascertain the influence of G201 on the evolution of turfy soil, a profile perpendicular to G201was selected for high density electrical technique tested. High density electrical technique detect stratum by measuring resistivity. For turfy soil, long-term flooding is a main reason for the formation of turfy soil, which is sensitive to resistivity, and this is why the high density electrical technique was selected.
The results of high density electrical technique show that there may be three layers in measure depth (the botten layer in purple is an unmeasured part). It shows low apparent resistivity in surface layer, this is because the long-term flooding leading to saturated soil in surface layer, high water content in turfy soil area of high porosity, result in low apparent resistivity. And as shown on the Fig. 10, A soil layer of higher apparent resistivity below the turfy soil layer, and it may be clay layer, this is deduced by the formation of turfy soil. There is always a water-resisting layer in the bottom of turfy soil to prevent water from infiltration, with out the water-resisting layer, there would never be the turfy soil.
With the building and working of G201, the high density electrical results show different with the distance to the roadway, this is because the roadway, especially the roadbed, prevent the hydraulic connection of the mountains and the turfy soil swamp. Generally, it will result in reducing moisture content in turfy soil swamp. But the high density electrical technique results show different in the Jiangyuan area, the reason may be that the roadway is much higher than turfy soil area, and the precipitation was collected in the turfy soil area near the G201 because of the special geography, and finally resulting in a partial shallow lagoon facies area, which contains less organic matter than turfy soil, and this is another degradation of turfy soil, as explanation in Fig. 8.
And summarize from the membership degree of turfy soil with different distance to G201, the identification profile can be divided to 3 part. The western part of about 350 m, show lower membership degree because of topographic and geomorphic conditions, resulting in less water accumulating, while the eastern part, not only water accumulating leading to less vegetation growth, pollution with the operation of highway, the influence of permeability coefficient with the construction of G201, especially the roadbed, will all have great impact on turfy soil swamp, which shows different on plants and soil moisture content. As far as Jiangyuan turfy soil swamp is concerned, the influence of highway to turfy soil is about 200 m.
7 Conclusion
-
1.
The result of field investigation verified that the combination of qualitative identification, threshold method, block analysis and fuzzy evaluation is an effective means for recognition of turfy soil.
-
2.
The block analysis based on the formation mechanism of turfy soil and the fuzzy evaluation considering the evolution of turfy soil, are consistent with the physical meaning of the turfy soil.
-
3.
Identification with membership degree shows great value not only for the distribution of turfy soil but also for the analysis of evolution of turfy soil.
-
4.
Through two instances corresponding to turfy soil affected by nature and by human activities respectively, the evolution of turfy soil can be concluded by remote sensing, which may show great value for the preservation of the turfy soil as well as the ecological benefit of environment.
References
Bai X, He B (2015) Potential of Dubois model for soil moisture retrieval in prairie areas using SAR and optical data. Int J Remote Sensing 36 (22)
Bai X, He B, Xing M (2015) Method for soil moisture retrieval in arid prairie using TerraSAR-X data. J Appl Remote Sens 9(1)
Bocco M, Sayago S, Willington E (2015) Neural network and crop residue index multiband models for estimating crop residue cover from Landsat TM and ETM+ images. Int J Remote Sensing 35 (10)
Boughrara H, Chtourou M, Chokri BA (2016) Facial expression recognition based on a mlp neural network using constructive training algorithm. Multimed ToolsAppl 75(2):709–731
Boylan N et al (2008) Peat slope failure in Ireland. Q J Eng Geol Hydrogeol 41:93–108
Bricker SH et al (2014) Controls on the basin-scale distribution of hydraulic conductivity of superficial deposits: a case study from the Thames Basin, UK. Q J Eng Geol Hydrogeol 47(3):223–236
Campbell J (1996) Introduction to remote sensing, 2nd edn. Taylor and Francis, London
Canters F (1997) Evaluating the uncertainty of area estimates derived from fuzzy land-cover classification. Photogramm Eng Remote Sens 63:403–414
Chapin F III, Zavaleta E, Eviner V, Naylor R, Vitousek P, Reynolds H, Hooper D, Lavorel S, Sala O, Hobbie S, Mack M, Diaz S (2000) Consequences of changing biodiversity. Nature 405:234–242
Chen N, Li J, Zhang X (2015) Quantitative evaluation of observation capability of GF-1 wide field of view sensors for soil moisture inversion. J Appl Remote Sens 9(1)
Du Y, Lu X, Chen L (2016) An interval type-2 T-S fuzzy classification system based on PSO and SVM for gender recognition. Multimed Tools Appl 75(2):987–1007
Dykes A et al (2008) Peat slope failures and other mass movements in western Ireland. Q J Eng Geol Hydrogeol 44:5–16
Friedl M, McIver D, Hodges J, Zhang X, Muchoney D, Strahler A (2002) Global land cover mapping from MODIS: algorithms and early results. Remote Sens Environ 83(1):287–302
Godio A et al (2014) Coupling ground-penetrating radar and flowmeter investigations for the characterization of a fissured aquifer. Q J Eng Geol Hydrogeol 47(4):351–361
Goodin, DG, Anibas KL, Bezymennyi M (2015) Mapping land cover and land use from object-based classification: an example from a complex agricultural landscape. Int J Remote Sensing 36(18)
Gropius M (2010) Numerical groundwater flow and heat transport modelling of open-loop ground source heat systems in the London Chalk. Q J Eng Geol Hydrogeol 43(1):23–32
Guellala R, Abidi M, Balti N (2016) Geophysical prospecting for groundwater exploration in northwestern Tunisia. Hydrol Sci J-J Des Sciences Hydrol 61(3):636–645
Han Y, Shi X, Che G (2012) Experimental study on the frost heaving characteristics of turfy Soil Foundation. Inform Bus Intell 267(599-605)
Hao L, Zhao X, Boorder D, Hu G (2014) Origin of PGE depletion of Triassic magmatic Cu-Ni sulfide deposits in the central-southern area of Jilin province. NE Chin, ORE Geol Rev 63:226–237
He T, Sun Y, Xu J (2015) Enhanced land use/cover classification using support vector machines and fuzzy k-means clustering algorithms. J Appl Remote Sens 8(1)
Hu C, Xu Z et al (2015) Video structured description technology for the new generation video surveillance system. Front Comput Sci 9(6):980–989
Huang Z, Chau K (2007) A new image thresholding method based on Gaussian mixture model. Appl Math Comput 205(2):899–907
Işık Y et al (2015) Rock mass parameters based doline susceptibility mapping in gypsum terrain. Q J Eng Geol Hydrogeol 48:124–134
Jenicka S, Suruliandi A (2014) Fuzzy texture model and support vector machine hybridization for land cover classification of remotely sensed images. J Appl Remote Sens 8(1)
Kalra G, Singh S (2016) Efficient digital image denoising for gray scale images. Multimed Tools Appl 75(8):4467–4484
Krankina O, Pflugmacher D, McGuire HAD, Hansen MC, Häme T, Elsakov V, Nelson P (2011) Vegetation cover in the eurasian arctic: distribution, monitoring, and role in carbon cycling. In: Gutman G, Reissell A (eds) Eurasian Arctic Land Cover and Land Use in a Changing Climate. Springer, New York, pp 79–108
Kumar P, Gupta DK, Mishra, VN (2015) Comparison of support vector machine, artificial neural network, and spectral angle mapper algorithms for crop classification using LISS IV data. Int J Remote Sensing 36 (6)
Lee E, Fooke P (2015) A note on the origins of engineering geomorphology in the UK. Q J Eng Geol Hydrogeol 10(1144):147–156
Lei N, Su Z (2015) Study on mineral distribution of peat soil in Northeast of China. Asian J Chem 25(18):10150–10152
Liao K, Xu S, Wu J (2013) Spatial estimation of surface soil texture using remote sensing data. Soil Sci Plant Nutri 59(4):488–500
Liu Y, Cao G, Meng Y (2014) study on the microstructure feature and strength mechanism of the Tien Lake peat soil. Environ Eng 864-867(2695-2702)
Liu T, Miao Q, Xu P (2016) Color topographical map segmentation Algorithm based on linear element features. Multimed Tools Appl 75(10):5417–5438
Lv Y, Nie L, Xu K (2012) Study on the difference between turfy soil and normal peat soil in China. Vibration, Struct Eng Measurement I, 105–107 ( 1551–1554)
Maxwell AE, Warner TA (2015) Differentiating mine-reclaimed grasslands from spectrally similar land cover using terrain variables and object-based machine learning classification. Int J Remote Sensing 36 (17)
Meyer W, Turner B II (eds) (1994) Changes in land use and land cover: a global perspective. Cambridge University Press, Cambridge, pp 437–471
Michael L, Boylan N (2013) Predictions of settlement in peat soils. Q J Eng Geol Hydrogeol 46(3):303–322
Nie L et al (2012) Influence of organic content and degree of decomposition on the engineering properties of a peat soil in NE China. Q J Eng Geol Hydrogeol 45(4):435–446
Osinowo Olawale O, Olayinka AI (2012) Very low frequency electromagnetic (VLF-EM) and electrical resistivity (ER) investigation for groundwater potential evaluation in a complex geological terrain around the Ijebu-Ode transition zone, southwestern Nigeria. J Geophys Eng 9(4):374–396
Pont D, Kimberley MO, Brownlie RK (2015) Calibrated tree counting on remotely sensed images of planted forests. Int J Remote Sensing 36 (15)
Reschke J, Bartsch A, Schlaffer S, Schepaschenko D (2012) Capability of C-Band SAR for operational wetland monitoring at high latitudes. Remote Sens 4(10):2923–2943
Saha S, Basu S, Nasipuri M (2015) iLPR: an Indian license plate recognition system. Multimed Tools Appl 74(23):10621–10656
Sela S, Svoray T, Assouline S (2014) Soil surface sealing effect on soil moisture at a semiarid hillslope: implications for remote sensing estimation. Remote Sens 6(8):7469–7490
Singh PP, Garg RD (2014) Classification of high resolution satellite images using spatial constraints-based fuzzy clustering. J Appl Remote Sens 8(1). doi:10.1117/1.JRS.8.083526
Tang S, Nie L, Qiao D (2012) Study on the difference between multipotiential surface model and Duncan-Chang model based on the triaxial test of Heda highway turfy soil. Mater Process Technol II, 538-541(965-970)
Wang Y, Wang S, Yang S (2014) Using a remote sensing driven model to analyze effect of land use on soil moisture in the Weihe River Basin, China. IEEE J Select Topics Appl Earth Observ Remote Sens 7(9):3892–3902
Xu H, Jin Y, Liu X (2006) The climatic effect of Carex. meyeriana mire in the Changbai Mountain valley. Ecol Environ 15(1):120–123
Xu C, Sui H, Li H (2015) An automatic optical and SAR image registration method with iterative level set segmentation and SIFT. Int J Remote Sensing 36 (15)
Xu Z et al (2015) Semantic based representing and organizing surveillance big data using video structural description technology. J Syst Software 102:217–225
Xu Z et al (2016) Semantic enhanced cloud environment for surveillance data management using video structural description. Computing 98(1-2):35–54
Yan Y, Zhu J, Yan Q (2014) Modeling shallow groundwater levels in Horqin Sandy Land, North China, using satellite-based remote sensing images. J Appl Remote Sens 8(1):083647
Zakeri F, Zoej MJ (2015) Adaptive method of speckle reduction based on curvelet transform and thresholding neural network in synthetic aperture radar images. J Appl Remote Sens 9(1)
Zhang T, Jiang L, Zhao T (2014) Soil temperature independent algorithm for estimating bare surface soil moisture. J Appl Remote Sens 8(1):083558
Zhu A, Qi F, Moore A (2010) Prediction of soil properties using fuzzy membership values. Geoderma 158(3-4):199–206
Acknowledgments
This project was financially supported by the National Natural Science Foundation of China (Grant NO. 41502272, Grant NO.41572254), the Basic Research Foundation of Jilin University (Grant NO.450060491447), Science and Technology Development Program of Jilin Province (Grant NO.20150520077JH) and China Postdoctoral Science Foundation (Grant No. 2014M551453). All of them are gratefully acknowledged.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Nie, L., Huang, Y. & Xu, Y. Distribution and evolutionary of turfy soil identified by remote-sensing images based on fuzzy evaluation. Multimed Tools Appl 76, 14635–14651 (2017). https://doi.org/10.1007/s11042-016-3842-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-016-3842-z