Spatial Characteristics And Changes Of Soil Nutrients In Cultivated Land Of Guanzhong Region In Shaanxi Province Based On GIS

The study of the spatial characteristics and changing patterns of soil nutrients is the foundation of regional soil quality evaluation and ecological environment assessment. Guanzhong Region locaed in the central part of Shaanxi Province, is the important producing area for wheat and maize in Northern China, the major grain producing and key construction area of Shaanxi province. Recent years, the food production, the main crop yield per unit and agricultural GDP proportion of this region show a downward trend. The blind, unreasonable fertilization and low fertilizer efficiency have severely restricted agricultural capacity, and have increased the surface waters eutrophication trend.The study takes the whole Guanzhong Region as the study area, based on the regional high density soil sampling data(20 ~ 60 thousand) of the Cultivated Land Soil Quality Survey and Assessment Project and Soil Testing and Formulated Fertilization Project of Shaanxi Province(2009-2012), using the tranditional statistical, mathematical statistics, geostatistics and GIS technology to study the spatial variability, spatial distribution pattern and influencing factors of soil macroelements nutrients(organic matter, total nitrogen(N), available N, available phosphorus(P), available potassium(K)) and soil microelements nutrients(available ferrum(Fe), available manganese(Mn), available zinc(Zn) and available copper(Cu)) in regional scale. Secondly, fully collected the soil property data of the second soil census of Shaanxi Province in 1980 s, constructed and integrated the two databases(1980s and 2010s), to analyze the temporal and spatial changing patterns of soil nutrients for nearly 30 years, and discuss the agricultural problems existed in. Thirdly, based on the above study of soil nutrients, further combined the theory of random sampling method to study the influence of the regional soil sampling number(spacing) to the spatial variability and the reasonable regional soil sampling number of soil nutrients. Finally, the study combined the theory of fuzzy mathematics and expert experience method to establish soil nutrients fertility evaluation system to evaluate the regional soil nutrients comprehensive fertility level. The study results can provide scientific basis for regional fertilization, agriculture partition management, comprehensive evaluation of the soil quality and potential production of cultivated land, and soil sampling design. Future more, the study also can provide the theory and practice guidance for the macroscopic decision-making of regional precision agriculture, farmland ecological environment protection, agricultural structure adjustment, etc.The main results and achievements of this study listed as follows:(1) According to the analysis results of traditional statistics with the soil nutrients classification of the second soil census of Shaanxi Province in 1980 s, soil nutrients content in cultivated land of Guanzhong Region is as follows. In soil macroelements nutrients, organic matter is concentrate in the level of 12 ~ 20 g/kg, total N of 0.58 ~ 1.25 g/kg, available N of 45 ~ 90 mg/kg, available P of 10 ~ 30 mg/kg and available K of 120 ~ 200 mg/kg. The average contents of the above soil nutrients is 14.91 g/kg,0.84 g/kg,69.36 mg/kg,21.43 mg/kg and 169.66 mg/kg,being in the 5th, 5th, 4th, 3th and 2nd of the soil nutrients classification standard respectively. In soil microelements nutrients, available Fe is concentrate in the level of 5.0 ~ 10 mg/kg, available Mn of 5 ~ 20 mg/kg, available Zn of 0.5 ~ 2.0 mg/kg and available Cu of 1.0 ~ 1.5 mg/kg. The average contents of these microelements is 7.67 mg/kg,11.10 mg/kg,1.30 mg/kg and 1.39 mg/kg, being in the 3rd, 3rd, 4th and 4th of the soil nutrients classification respectively. And, the average levels of these soil microelements nutrients are higher than their deficiency thresholds of Shaanxi Province and China. In the ratios of soil nutrients, soil carbon / nitrogen is concentrate in 8 ~ 12, with an average value of 10.87, a little higher than the average level of Chinese farmland(10). Soil available N/ P mainly is 1 ~ 6, with an average ratio of 4.31, higher than the appropriate level of crops(2 ~ 3). There is obvious imbalance phenomenon between soil available N and P. According to the coefficient of variation, both of soil nutrients and their ratios belonged to medium variation intensity(10 ~ 100%) with the largest value(74.87%) of soil available N/P. The coefficient of variation of available P(55.46%) is the maximum in the soil macroelements and that of available Fe(64.94%) is in the first place of soil microelements. Correlations are generally existed between soil nutrients. Organic matter has significantly positive correlated with the other nutrients(p<0.01). In comparison, organic matter and total N, available N, and available Cu and available Zn show more strong correlations.(2) Combing semi-variance function, fractal dimension(FD) with global Moran’s Index, soil nutrients in cultivated land of the whole Guanzhong Region and its geomorphic partition(Weibei Plateau and Guanzhong Plain) have moderate spatial correlation and demonstrate significantly spatial autocorrelation(P<0.01). Their nugget to sill ratio is 0.25 ~ 0.75, fractal dimension is 1.866 ~ 1.947 and global Moran’s Index is 0.154 ~ 0.387. In soil macroelement nutrients, soil available N has better spatial structure while organic matter and available P show stronger spatial variability, but available K show complex spatial structure. The microelements nutrients show better spatial correlations than soil macroelements nutrients. Soil available Zn and Mn show better spatial correlation. In terms of geographical distribution, soil macroelements nutrients in Weibei Plateau show weaker spatial autocorrelation and spatial continuity than that in Guanzhong Plain while soil available P and available N in Guanzhong Plain show more spatial inhomogeneity.(3) According to the Kriging maps, soil nutrients contents in cultivated land of Guanzhong Region generally increase from north to south as geomorphic types change. In east-west direction, soil nutrients contents have strong spatial variability, generally high in the west and low in the east. Soil nutrients contents show continuous echelon change in the southern area of Guanzhong Plain to Qinling Mountains. Soil microelements nutrients grades present large of patches in the central part and flaky shapes in the north and south area. According to the plentiful-lack index of soil available nutrients for crops growth, there are about 30% of cultivated land lack of available N, 46.91% for wheat growth and 30.10% for maize growth lack of available P, no more than 5% for them lack of available K. There is about 72.05% of cultivated land with imbalanced ratio of soil available N and P. According to the plentiful-lack index of soil mcroelement nutrients in Shaanxi Province, there are 11.48% of cultivated land lack of available Fe, 5.95% lack of available Mn and 39.67% lack of available Zn.(4) There are significant correlations between soil nutrients and topographic factors such as slope, altitude, terrain index, and climate factor including annual average precipitation, sunshine duration and accumulated temperature. Soil nutrients contents show a decrease with the increase of elevation and slope, an increase with the increase of annual average precipitation and accumulated temperature. At the same time, there are significant negative correlations between soil nutrients and human-made factors such as the distances between the sampling points and their corresponding cities, the main roads and river and ditches. It’s demonstrated that soil nutrients contents is high in the close area of human-made factors whereas low in the long distance area. Meanwhile, there are significant differences aomong soil nutrients in different soils and lanform types(P<0.05). As a whole, the spatial variability of soil nutrients influenced by both natural factors including terrain, climate and soil properties and human activities containing fertilization, irrigation, planting patterns and so on. Correlations between soil nutrients and climate factors are significantly higher than the terrain factors while the largest differences exist in the landform types. Climate factors and landform types are the important natural influence factors in Guanzhong region. Based on the natural factors, human activities and geographic location guiding role is playing more and more important role in the spatial distribution of soil nutrients especially in Guanzhong plain. There is a significant correlation between the original low values of the nutrient index(such as available P, Zn and Mn) and human-made factors. Soil nutrients in the original low level areas changed quickly, whereas the original high level areas changed slowly, and increased fast in the areas close to cities. The effects of human activity on agricultural system on the one hand obscure the natural factors such as soil texture, on the other hand increase the difference of topography which to some extent increase the spatial structure feature of soil nutrients such as available P.(5) During the past 30 years, soil nutrients in cultivated land of Guanzhong Region increased inordinately. Soil available P increased fastest(about 197%), available Zn took second place(about 130%), available Mn was the third(about 80%), and available K was the last with a slight increase(6.93%), the other soil nutrients increased by 30% ~ 50%. The coefficient of variation of soil nutrients decreased generally. The initial values of soil nutrients significantly affected the changing pattern of soil nutrients. Soil nutrients generally presente high growth in original low level area, while low growth or slight reduction in original high level area. Area close to the downtown promote with high growth. The grade closed up from the poles to the intergrade. Generally, through 30 years of development, soil nutrients content become regional homogenization direction. In the terms of spatial distribution, soil nutrients grades in 1980 s showed strong variability, obvious mosaic structure while showed significant regional differences corresponding with administrative division and geomorphic types in 2010 s. Soil nutrients increased fast in the central part(Guanzhong Plain) than the north and south area. The human activity such as straw returning to field, strength and direction of fertilization, irrigation, planting pattern and policy orientation are the main impact factors influencing the spatial and temporal variation of soil nutrients in cultivated land of Guanzhong Region, and geographical location also plays an important role.(6) The study on the effects of sampling number(spacing) on the statistical parameters, spatial variability and spatial interpolation by Kriging method showed that the spatial variability of soil nutrients exhibited obvious scale effect. In Guanzhong Plain, with the decrease of soil sampling number, the range, standard deviation and coefficient of variation of soil nutrients reducted, and coefficient of variation, ratios of nugget to sill and fractal dimensions of soil nutrients totally had significantly correlation with sampling number and sampling spacing. Spatial autocorrelation became weak with the reduction of sampling number and increase of sampling interval. Standardized Z scores became flat at the average sampling spacing of 3000 m. During the sampling number decreased, the spatial variability of soil nutrients revealed a complex progress with sampling number(spacing), often presented multinomial change with U shape or inversed-U shape, and the smooth effect of Kriging method increased. By Kriging method, in Guanzhong Plain, soil sampling number is demanded by more than 2460 of organic matter, 2271 of available N, 2989 of available P, 2972 of soil K. In order to better meet the requirements of spatial structure stability and high interpolation precision of the original data, the more reasonable sampling number is 4806 of organic matter, 10835 of available N, 14261 of available P, 9076 of available K. The quantity calculation method of Cochran is not suitable in cultivated land of Guanzhong Plain, blindly use of this method may lead to serious shortage of soil sampling number, unable to establish accurate and stable semi-variance function and further cause Kriging interpolation results distortion seriously. The identification of the reasonable sampling number(spacing) is an analytical investigation based on statistical characteristic and spatial structure feature. In addition, the spatial distribution of soil samples may play an important role to effect spatial structure feature and interpolation precision.(7) With the same and abundant sampling numbers, and similar spatial structure feature and spatial autocorrelation distance, Cokriging with organic matter as subsidiary variable is more suitable for spatial interpolation of total N compared with Ordinary Kriging method. For example, in Chang’an District, when the total N samples decreased to 85%(548), Cokriging with original organic matter(n=645) as subsidiary variable can also keep spatial structure and the interpolation precision of the original total N(n=645) by Ordinary Kriging. This method can establish more stable and accurate semi-variance function, provide more local variation information, but unavailable for the raster ratio calculation to obtain soil C/N map. The spatial correlation distance difference of the target and auxiliary variable is an important factor affecting the application of Cokriging method.(8) According to the validation of the food production correlation method, the participation of soil p H can hide up the local variation characteristics of soil nutrients fertility, led to soil fertility level relatively uniform, resulte in the inaccuracy of soil nutrients fertility evaluation. Expert experience method plays more positive effect on weight value of soil nutrients(organic matter, available N, available P and available K) compared with correlation methods, based on the evaluation accuracy of soil nutrients fertility. The soil nutrients fertility in cultivated land of Guanzhong Region belongs to moderate-high level. The area ration of grade I is 10.32%, grade II 22.85%, grade III 24.08%, grade IV 33.14%, and grade V 9.61%. Spatially, soil nutrients fertility is relatively high in the south and west and low in the north and east. The spatial distribution trend is mainly accordance with the topographical features.