Soil Fertility Evaluation And Fertility Index Spatial Variation In A Gravel-Sand Mulched Jujube Orchard

Soil fertility level is directly related to crop yield and quality,so effective fertility evaluation can provide theoretical basis for fertility management and fertilization decision-making of gravel-sand mulched.In this paper,field experiments were carried out to study the soil in the area where jujube trees were planted for 5 years.Eight indicators of soil organic matter,total N,total P,total K,available P,available K,salinity and soil pH were selected as fertility evaluation indicators,and combining classical statistics and geostatistics,the fertility status and spatial variability of gravel-sand mulched soil in Northwest China were revealed by fertility evaluation method.The main contents are as follows:?1?Descriptive analys is of various fertility indexes of 0-10 cm soil layer in gravel-sand mulched was carried out using classical statistics.The results showed that the distribution range of organic matter was 4.25-5.75g/kg,and the total N distribution was between 0.17-0.23g/kg.According to the second national soilcensus grading standard,they belonged to the level of barrenness,total P and available P is attributed to poor levels,and available K and salinity are at higher levels.The maximum and minimum values of soil pH were 9.67 and 8.12,which indicated that the soil appear alkaline.The IFI1,IFI2 and P_S were between 0.23-0.35,0.66-0.84 and 0.63-0.81.The coefficient of variation?CV?for organic matter,total N,total K,pH and available K were 6.04,5.69,6.41,4.05 and 6.08%,respectively,which belonged to weak variation.Others were between 17.38 and 38.7%,belonged to moderate variation.The comprehensive fertility index of IFI1,IFI2 and P_S were both less than 10%,which was weak variation.?2?The soil fertility of 0-10 cm soil layer was evaluated by comprehensive index method of soil quality,modified grey relational analysis and Nemerow comprehensive index method,and the optimum evaluation method was obtained through comparative analysis.The results showed that in the radar chart,the mean value of subordination was available K > salinity > total K > available P > total P > pH > organic matter > total N,and the mean fertility index was available K > salinity > total K > available P > pH > total P >organic matter > total N,the order of arrangement is roughly the same.In the spatial distribution map of the soil comprehensive fertility index?IFI1,IFI2 and PS?,the spatial distribution maps drawn are similar and appear plaque-like.Classification of soil fertility was graded by Equal distance method.In IFI1,IFI2 and P_S distribution map,soil fertility is basically occupied by class ?,?,?,etc.,among them,class ? is the most widely distributed,accounting for 53%,57%,and 54% of the total area,respectively.Class ? and below occupied the largest area,accounted for 64%,69%,and 67%,respectively.And it is known that the overall fertility level of the gravel-sand mulched is low.Comparing the three evaluation methods,it was found that the evaluation results showed inconsistency,the modified gray correlation analysis method had a high accuracy rate of 87.50%.?3?The geostatistical method and the optimal evaluation method were combined to accurately evaluate the soil fertility status of the study area and study the spatial variability of vario us fertility indicators.The results showed that the total N,total P and salt accorded with the spherical model,and other fertility ind icators accorded with the exponential model.C₀ /?C₀ +C?of organic matter,total P and pH were 0.139,0.021 and 0.114,respectively,which were all less than 0.25,with strong spatial autocorrelation.Others were between 0.418 and 0.50,indicated moderate spatial autocorrelation.The spatial distribution of each fertility index was not obvious.The optimal semi-variance function model after removing the first-order trend of comprehensive fertility index IFI2 is spherical model.The IFI2 was within the range of 0.66-0.84,soil fertility class ? and below accounted for 69%,all fertility indicators surrounded graphics area was small,which indicated that the fertility level was classified as barren.The spatial distribution of IFI2 is influenced by the distribution of organic matter and total N.?4?Taking organic matter as an example,and based on the classical statistics and geostatistics,spatial variability scale effect of soil organic matter injujube was studied under five sampling scales: 32 × 32,28 × 28,24 × 24,20 × 20 and 16 × 16 m,and three sampling spaces: 4,8,12 m.The result showed that SOM content decreased with the increase of depth in each scale,varying from 2.41 to 5.75 g/kg.The CV of SOM was ranged from 3.90 to 11.89% under five scales,the intensity of variation showed weak variability.The mostly nugget variance was <0.25,indicating strong spatial autocorrelation.The variation trend of Moran's ? coefficient of organic matter in each soil layer decreased with the increase of distance from 0 to 21.33 m,and the total variation range was from-0.548 to 0.548,which indicated that the spatial correlation of soil organic matter was strong.When the sampling space increased,the nugget of soil organic matter increased,but the CV was unaffected.On the spatial distribution map,there were "bulges " and "depressions " in all soil layers,and with the increase of depth,soil organic matter gradually decreased,and the degree of "concave and convex" was roughly the same.With the decrease of sampling scale,the distribution of surface soil organic matter was more uniform,and 8m was a more reasonable sampling interval.By comparing the two interpolation methods,K riging interpolation of soil organic matter had the highest accuracy.