本文選題：土壤養(yǎng)分 + 地統(tǒng)計學與GIS��； 參考：《昆明理工大學》2015年碩士論文

【摘要】：本研究以云南省昌寧縣耕地土壤為研究對象,以研究區(qū)2012年測土配方施肥數(shù)據(jù)為基礎,探討了耕層(0-20cm)土壤pH值、有機質(OM)、全氮(TN)、堿解氮(AHN)、有效磷(AP)、速效鉀(AK)、有效鋅(Zn)、有效錳(Mn)、有效鉬(Mo)等9種現(xiàn)狀及空間分異情況,重點研究和分析了豐缺格局、影響土壤養(yǎng)分變化的因素,并進行了綜合肥力進行評價；在保障精度的條件下,進一步優(yōu)化采樣數(shù)量。主要研究結果為：(1)統(tǒng)計分析表明,各個土壤養(yǎng)分的變異系數(shù)(CV)在13.08%-50.15%之間,PH、有機質、氮、磷、鉀、錳、鋅、鉬的變異系數(shù)均在10%-50%之間,具有中等的空間變異性。pH的塊基比C0/(Co+C)都小于25%,說明具有強烈的空間相關性,有機質、氮、磷、鉀、錳、鋅、鉬的塊基比C0/(Co+C)值在25%-75%之間,表現(xiàn)出中等空間相關性。PH是線性模型,有機質、有效錳是球狀模型,全氮、堿解氮、速效鉀、有效磷、有效鋁是指數(shù)模型。(2)地形因子和其他環(huán)境變量(土壤類型、耕地制度、母質類型等)作為輔助變量,建立了縣域的土壤養(yǎng)分回歸克里格預測模型,結果表明回歸克里格插值精度高于普通克里格和協(xié)同克里格插值的精度。(3)依據(jù)《云南省養(yǎng)分含量等級表》的分級標準,進行空間格局分析。利用主成分分析和BP神經(jīng)網(wǎng)絡算法對土壤綜合肥力進行評價,兩種方法都能有效的表達土壤綜合肥力狀況。(4)采用協(xié)同克里格方法和普通克里格法對土壤養(yǎng)分采樣數(shù)進行優(yōu)化,得出在不同樣點數(shù)目下的協(xié)同克里格插值精度優(yōu)于普通克里格插值精度,并確定研究區(qū)的最大采樣間隔。(5)不同土地利用方式下對耕層土壤8種元素(除有效鋅)差異性顯著；土壤類型對養(yǎng)分元素均產(chǎn)生了顯著水平；不同的成土母質對土壤養(yǎng)分元素含量產(chǎn)生了顯著水平,除有效磷外；海拔對9種養(yǎng)分元素影響顯著；陰坡和半陰坡方向比陽坡更有利于土壤養(yǎng)分的積累,坡向對AP、Mn、Mo影響不顯著；坡度與全氮不顯著。
[Abstract]:In this study, the cultivated land soil in Changning County, Yunnan Province was taken as the research object, and the pH value of the cultivated layer (0-20 cm) was discussed on the basis of the soil formula fertilization data from 2012 in Changning County, Yunnan Province. Organic matter (OMN), total nitrogen (TNN), alkali-hydrolyzed nitrogen (AHN), available phosphorus (APN), available potassium (AK), available zinc (Zn), available manganese (mn), available molybdenum (Mo) and their spatial differentiation were studied and analyzed in detail. The factors affecting soil nutrient change were studied and analyzed. The comprehensive fertility was evaluated and the sampling quantity was optimized under the condition of ensuring precision. The main results were as follows: (1) the coefficient of variation (CV) of soil nutrients was between 13.08% and 50.15%, and the coefficients of variation of organic matter, nitrogen, phosphorus, potassium, manganese, zinc and molybdenum were between 10% and 50%. The ratio of block to base (C0/(Co C) of medium spatial variability. PH is less than 25, indicating that there is a strong spatial correlation. The ratio of organic matter, nitrogen, phosphorus, potassium, manganese, zinc and molybdenum is between 25% and 75%, and the ratio of mass to base of organic matter, nitrogen, phosphorus, potassium, manganese, zinc and molybdenum is between 25% and 75%. The medium spatial correlation. Ph is linear model, organic matter, available manganese is spherical model, total nitrogen, alkali-hydrolyzed nitrogen, available potassium, available phosphorus, available aluminum is exponential model. 2) topographic factors and other environmental variables (soil type, cultivated land system, etc.) As auxiliary variables, the regression Kriging prediction model of soil nutrients in county area was established. The results showed that the precision of regression Kriging interpolation was higher than that of common Kriging and cooperative Kriging interpolation. Principal component analysis (PCA) and BP neural network (BP) algorithm were used to evaluate soil comprehensive fertility. Both methods can effectively express soil comprehensive fertility. (4) Co-Kriging method and common Kriging method were used to optimize soil nutrient sampling number. The results show that the precision of cooperative Kriging interpolation under different points is better than that of common Kriging interpolation, and the maximum sampling interval of the study area is determined. 5) the difference of 8 elements (except available zinc) in topsoil under different land use patterns is significant. Soil types had significant effects on nutrient elements, different parent materials had significant effects on soil nutrient contents, except available phosphorus, elevation had a significant effect on 9 nutrient elements. The direction of shady slope and semi-shady slope is more favorable to the accumulation of soil nutrients than that of sunny slope, and the slope direction has no significant effect on APGMnMn-Mo, and the slope and total nitrogen are not significant.
【學位授予單位】：昆明理工大學
【學位級別】：碩士
【學位授予年份】：2015
【分類號】：S158

【參考文獻】

相關期刊論文 前10條

1 張水清;黃紹敏;郭斗斗;楊尊玉;;基于主成分分析法的土壤肥力評價[J];安徽農業(yè)科學;2011年02期

2 范海榮;常連生;王洪海;李曉;張立娟;;城市草坪土壤肥力綜合評價[J];草業(yè)科學;2010年10期

3 楊建峰;孫燕;鄔華松;鄭維全;;胡椒園土壤養(yǎng)分相關性研究[J];熱帶農業(yè)科學;2009年08期

4 秦松;樊燕;劉洪斌;王正銀;;地形因子與土壤養(yǎng)分空間分布的相關性研究[J];水土保持研究;2008年01期

5 賈樹海;張琦;孟維軍;白靜;王會;王慧強;;基于GIS與地統(tǒng)計學土壤養(yǎng)分空間變異特征研究——以遼寧省凌源市6鄉(xiāng)鎮(zhèn)為例[J];水土保持通報;2009年03期

6 S.J.HERBERT;;Spatial Variability of Nutrient Properties in Black Soil of Northeast China[J];Pedosphere;2007年01期

7 趙倩倩;趙庚星;董超;李文璐;;高密市農田土壤養(yǎng)分空間變異特征研究[J];土壤通報;2012年03期

8 黃紹文,金繼運,楊俐蘋,程明芳;縣級區(qū)域糧田土壤養(yǎng)分空間變異與分區(qū)管理技術研究[J];土壤學報;2003年01期

9 劉付程,史學正,于東升,潘賢章;基于地統(tǒng)計學和GIS的太湖典型地區(qū)土壤屬性制圖研究——以土壤全氮制圖為例[J];土壤學報;2004年01期

10 曹慧,楊浩,孫波,趙其國,臧波;太湖流域丘陵地區(qū)土壤養(yǎng)分的空間變異[J];土壤;2002年04期

相關碩士學位論文 前1條

1 趙莉榮;不同成土母質條件下土壤養(yǎng)分空間變異研究[D];西南大學;2010年

，

本文編號：1990475