Font Size: a A A

Suitable Interpolation Method And Sampling Density Of Nutrients Based On Spatial-Temporal Variation On Paddy Soil

Posted on:2021-04-19Degree:MasterType:Thesis
Country:ChinaCandidate:T QiaoFull Text:PDF
GTID:2393330614455736Subject:Agriculture
Abstract/Summary:PDF Full Text Request
The point and surface expansion model based on geographic information system?GIS?is an effective method to quantify soil nutrient variability.At present,this type of research was generally limited to a single time scale.And it did not take the effects of differences in soil nutrient attributes at different periods on the appropriate interpolation method and the number of sampling points into account.Therefore,this study focuses on the surface soil of the 2.32 M hm2paddy soil in Tai-lake Region,China.The study analyzes 1 096 topsoil samples of the Second National Soil Survey in 1982 and 1 370 topsoil samples of the"973 Soil Quality Research"Project in 2000 to obtain the influence of different interpolation methods on the prediction accuracy of soil nutrients?organic matter,total nitrogen and available phosphorus?of different periods with deterministic and geostatistical interpolation methods,and determines the optimal interpolation method in two periods.Then,based on the optimal model,the appropriate number of sampling points in the study area is estimated.The results can provide a theoretical basis for the selection of the optimal interpolation model and the determination of reasonable survey samples under the spatial-temporal evolution of the main soil nutrients in paddy fields in southern China.The main conclusions are as follows:1.In 1982,the most accurate interpolation methods for predicting soil organic matter,total nitrogen,and available phosphorus in paddy fields in the Tai-Lake Region were Ordinary Kriging,Disjuncted Kriging,and Inverse Multiquadric Function.However,in 2000,they were Ordinary Kriging,Disjunctive Kriging,and Completely Regularized Spline.From the perspective of different paddy soil subgroup regions,the accuracy of organic matter prediction of each soil subgroup was higher than that of traditional geostatistics under the deterministic interpolation methods of Local Polynomial Interpolation,Global Polynomial Interpolation and Inverse Distance Weighting in 1982 and 2000.Total nitrogen also had higher prediction accuracy under the three deterministic interpolation methods of Local Polynomial Interpolation,Completely Regularized Spline,and Inverse Distance Weighting than traditional geostatistics.And the prediction accuracy of available phosphorus was more accurate than geostatistics under the three deterministic methods of Local Polynomial Interpolation,Completely Regularized Spline,and Inverse multiquadric function.From different administrative regions,the optimal interpolation methods for soil organic matter were geostatistics in most areas in 1982and 2000,and Ordinary Kriging interpolation and Simple Kriging had higher prediction accuracy especially.However,those of total nitrogen and available phosphorus were Inverse Multiquadric Function,Inverse Distance Weighting and Spline With Tension in most areas.In general,the optimal interpolation method for soil organic matter,total nitrogen,and available phosphorus is greatly affected by the distribution of sample points,human activities and environmental factors under the spatial-temporal evolution in Tai-lake paddy fields in different periods.Geo-statistical methods are high accuracy in the prediction of organic matter and total nitrogen throughout the study area.Deterministic interpolation are higher prediction accuracy for available phosphorus in the entire study area,organic matter,total nitrogen and available phosphorus of different subgroup regions,as well as the total nitrogen and available phosphorus in various administrative regions.2.The spatial distribution of the Tai-lake paddy field main soil nutrients are as follows.The organic matter content was suitable(10?20g·kg-1),most suitable(20?30 g·kg-1)and abundant(30?40 g·kg-1)in 1982.The three levels were raised to the most suitable,rich and extremely rich(>40 g·kg-1)level in 2000.Most of the total nitrogen content in the two periods were at the optimal level(1?1.5 g·kg-1),abundant(1.5?2 g·kg-1)and extremely abundant(>2 g·kg-1),and the area occupied by annual richness and extreme richness had increased in 2000.The content of available phosphorus in the two periods was at three levels:lack(3?5mg·kg-1),suitable(5?10 mg·kg-1)and most suitable(10?20mg·kg-1).From the perspective of spatial-temporal variability,organic matter content with 50.11%area had risen by 5-10 g·kg-1from 1982 to2000,and the distribution of accumulation had shown a substantial increase.The total nitrogen content increased more and decreased less.Among them,68.94%area was at the accumulation level of 0-0.5 g·kg-1,while 18.64%area was at-0.5-0 g·kg-1loss level.The available phosphorus increased in the south and decreased in the north,and it's content accumulated in 34.29%of the area was 0-5 mg·kg-1while the area of 32.85%showed a loss level of-5-0 g·kg-1.3.The relationship between the number of sampling and the prediction accuracy under the optimal interpolation methods of last study from paddy soil can be divided into three stages in the Tai-lake Region in1982 and 2000.Organic matter sample points were in the rapid decline period of 5%-15%,the fluctuation period of 15%-65%,and the stationary period of 65%-100%.The soil total nitrogen sample points were at a rapid decline period of 5%-15%,a fluctuation period of 15%-70%,and a stationary period of 70%-100%.The available phosphorus sample points were in the rapid decline period of 5%-15%,the fluctuation period of 15%-80%,and the stable period of 80%-100%.4.Based on the optimal interpolation method,the 20 sampling gradient interpolation results of Tai-lake paddy field soil nutrients shows that the suitable sample density of organic matter was 1.38 samples perhectare in 1982 and 1.59 samples per hectare in 2000;The suitable sampling density for total nitrogen was 1.27 per hectare in 1982 and 1.59per hectare in 2000;the suitable sampling density for available phosphorus was 2.20 per hectare in 1982 and 2.26 per hectare in2000.According to the calculation of the current average cost of about200 yuan and the consumption of 50 g of soil at a certain index for each sample,the sampling strategy with the lowest interpolation requirement can save about 105,200 yuan and 26.3 kg of organic matter,131,600 yuan and 32.9 kg of soil,respectively in 1982 And 2000;Total nitrogen saved96,400 yuan and 24.10 kg of soil,120,600 yuan and 30.15 kg of soil,in1982 and 2000,respectively;Available phosphorus saved 52,600 yuan and13.15 kg of soil in 1982,and 98,600 yuan and 24.65 kg of soil were saved in 2000.
Keywords/Search Tags:paddy field, soil nutrients, spatial-temporal evolution, optimal interpolation method, spatial variation
PDF Full Text Request
Related items