基于高光谱数据的土壤电导率估算模型——以河套灌区沙壕渠灌域沙壤土为例

文中采用FieldSpec3便携式地物光谱仪,基于在内蒙古河套灌区沙壕渠灌域获取的含盐土壤高光谱反射率数据构建了两种土壤电导率的估算模型。第一种模型利用全波段(350nm-2500nm)光谱反射率数据作为自变量,采用偏最小二乘回归法拟合回归系数,第二种模型选取光谱吸收特征指数作为预测因子,采用稳健回归法构建估算模型。通过与人工测量值相比,两个模型模拟结果的相对误差平均值在2%以内,相对误差的绝对值在17%以内,模拟值与实测值相关系数在0.97以上,表明构建的模型可以用于该区域土壤电导率的估算。     

基于室内高光谱数据的多种类型土壤有机质估算模型比较

基于高光谱数据的土壤有机质反演是土壤遥感及精准农业的重要研究内容,然而不同的光谱处理及建模方法使得模型的估算能力及精度差异明显,限制了模型之间的通用性。为了构建陕西省土壤有机质含量估算的最优模型,以陕西省9种主要土壤类型的216个土样的光谱反射曲线和土壤有机质含量为数据基础,将光谱反射曲线进行一阶微分d(R)、倒数对数log(1/R)、倒数对数一阶微分d[log(1/R)]和包络线去除N(R)4种变换,结合一元线性回归(SLR)、偏最小二乘回归(PLSR)和支持向量机回归(SVR)3种建模方法构建了不同的土壤有机质含量估算模型。结果显示:不同类型土壤的反射光谱曲线总体态势基本一致,吸收特征位置基本相同,且土壤有机质含量与光谱反射率呈负相关态势;基于d [log(1/R)]光谱变换构建的SVR估算模型精度最高,建模集和验证集的判断系数(R~2)分别为0.9210、0.8874,验证均方根误差(RMSE)为2.18,相对分析误差(RPD)达到2.8751,是估算陕西省土壤有机质含量的最优模型,PLSR次之,SLR最差。     

新疆北部不同类型土壤光谱特征及对有机质含量的预测

对北疆地区淡栗钙土、冷钙土、石灰性黑钙土、石膏灰棕漠土等4种土壤类型的反射光谱进行分析,研究土壤有机质含量与光谱反射率之间的关系。结果表明,石灰性黑钙土的反射率明显低于其它土壤类型。有机质含量高于14 g·kg^-1时光谱反射率与有机质含量呈负相关,有机质含量很低（<8 g·kg^-1）时,土壤的光谱反射率与有机质含量之间呈正相关。分别采用593.6 nm波段的原始光谱反射率、661 nm波段的反射率去除包络线和547.4 nm波段反射率的一阶微分与土壤有机质含量建立回归模型,经检验三种模型均能较好地预测有机质的含量,其中光谱的一阶微分预测精度相对较高,可较好地预测北疆主要类型土壤的有机质含量。     

谷子叶绿素含量高光谱特征分析及其反演模型构建

基于高光谱数据综合分析不同施肥条件下谷子各生长期冠层叶绿素含量的高光谱特征,在分析各光谱特征参数与叶绿素相关性的基础上,基于偏最小二乘法和人工神经网络构建叶绿素含量的遥感反演模型.结果表明:NDVI(归一化植被指数)、GNDVI(绿色归一化植被指数)、PSNDa(特殊色素归一化指数a)、PSSRc(特征色素简单比值指数...     

基于PLSR-BP复合模型的绿洲土壤pH高光谱反演

以土壤pH、野外实测光谱以及多元散射校正(MSC)预处理后的光谱数据为基础,利用数学方法（主成分回归PCA、偏最小二成回归PLSR、BP神经网络模型）分别建立了土壤pH的预测模型。结果表明：土壤实测光谱和经过MSC方法预处理的光谱数据均与pH存在良好的相关性,并呈极显著水平,后者的相关性更高。PCA和PLSR两种土壤pH估测模型均具有良好的预测能力。BP神经网络模型则因输入变量多,预测精度较低。但利用PCA和PLSR模型所获得主成分,作为BP神经网络的输入变量所建立的复合模型,可明显提高模型稳定性和预测能力。     

准噶尔盆地人工林地土壤全盐的高光谱反演

随着高光谱遥感技术的快速发展,通过其定量估测土壤化学成分具有很好的可行性。使用ASD Pro FieldSpec3便携式光谱仪,测量准噶尔盆地人工林地风干土壤样品的可见光-近红外光谱,利用土壤反射光谱值预测全盐的含量。首先,通过皮尔森相关系数分析方法,计算土壤全盐与土壤反射光谱之间的相关性,其中土壤光谱值的二阶导数与土壤全盐的相关系数最高为0.806,均方根误差最小为1.508。其次,在基于光谱反射率的基础上,通过多元统计回归分析,表明土壤光谱在1 130 nm、1 430 nm和1 930 nm波段的全盐反演模型预测的效果较好,可以利用这3个波段建立回归方程,对土壤全盐进行反演估算。     

干旱区土壤盐渍化信息遥感建模

以新疆塔里木盆地北缘的渭干河-库车河三角洲绿洲为研究区,利用GF~(-1)与Landsat8 OLI影像数据作为基本数据源,从影像上提取15个盐分指数和5个光谱植被指数,通过灰度关联分析法,对0~10 cm表层土壤含盐量与影像光谱指数进行分析和筛选,确定出与土壤含盐量相关性较高的综合光谱指数,采用多元线性回归,偏最小二乘法回归,支持向量机回归三种方法分别对GF~(-1)与Landsat8 OLI影像构建基于实测数据和影像数据的综合指数土壤含盐量估算模型,并选出最优模型。结果表明:(1)在20个光谱指数中,相关性较好的光谱指数是SR、CSRI、SI、BI、S6、ARVI、SAVI、NDSI,关联系数均达到0.7以上,并基于这8个光谱指数构建综合光谱指数。(2)3种估算模型:基于GF~(-1)多元线性回归模型决定系数R~2为0.6856,高于决定系数R2为0.5142的Landsat8 OLI;偏最小二乘回归模型1~8个主成分,GF~(-1)决定系数2个3个1个,其中2个主成分最高可达0.6104,Landsat决定系数4个3个2个,其中4个主成分最高可达0.549;支持向量机模型3种函数,GF~(-1)决定系数RBFPolynomialLinear,其中RBF函数最高可达0.7969,Landsat决定系数PolynomialRBFLinear,其中Polynomial函数最高可达0.7154。对比3种模型可知,支持向量机回归模型的R2最高,因此该模型相对于多元线性回归和偏最小二乘回归更适于土壤盐渍化估算。     

开都河流域下游绿洲盐渍化土壤高光谱特征

土壤光谱反射特性是土壤遥感的物理基础.通过野外调查采样、土壤盐分实验分析与土壤高光谱数据采集,对土壤高光谱数据一阶和二阶导数微分变换处理,分析土壤样品的光谱特征,建立土壤光谱和土壤盐分含量间的相关关系,对研究区盐渍化土壤含盐量进行定量反演.研究结果表明:1)从土壤光谱反射率的形态特征来看,土壤的光谱反射率曲线总体上变化较为平缓,光谱特征形态较为相似,且基本平行.2)研究区土壤光谱反射率曲线的形状大致可由300 ～ 600nm、600 ～ 800nm、800～1000nm、1000 ～ 1400nm、1400 ～1900nm、1900 ～ 2100nm、2100～ 2500nm七个折线段和560nm、900nm、1400nm、1900nm、2200nm五个特征吸收点来控制.3)利用光谱反射率一阶导数微分的盐渍化土壤含盐量多元线性回归预测模型的预测效果均优于利用反射率原型和反射率二阶导数微分,其中氯化物-硫酸盐型RMSE=0.33,硫酸盐型RMSE=0.31,硫酸盐-氯化物型RMSE=0.22.     

新疆农田土壤有机质含量及组成特征

目前新疆农田土壤有机质含量在5．30～24．21g／kg之间，平均14．65g／kg，与第二次土壤普查相比，新疆农田土壤有机质含量呈上升趋势；胡敏素碳含量占到有机质碳量的53％～70％，平均61％，腐殖酸碳量只占39％。     

基于HJ1A-HSI反演松嫩平原土壤盐分含量

以HSI高光谱遥感影像为数据源,利用地理信息系统和偏最小二乘回归（PLSR）分析方法,结合实地采样的盐分离子、EC值、pH的化验数据,分析盐分离子在HSI数据中的光谱特征,建立土壤盐分与高光谱数据的偏最小二乘回归模型,实现对松嫩平原土壤中主要盐分参数的反演实验。结果表明：偏最小二乘回归分析方法在保证信息量最大的前提下,降低了光谱数据维数,提高了分析的效率。利用偏最小二乘回归建立的预测模型,对全盐、[WTBX]EC[WTBZ]值、Na⁺+K⁺、Cl^-、HCO^-₃有较好的反演精度,模型的判定系数分别为0.799、0.879、0.772、0.791和0.694。在土壤含盐量的定量反演方面,探索了使用HSI影像作为新的数据源,为松嫩平原土壤盐分含量的精确、定量、快速获取及盐碱化防治等提供参考。     

Spatial distribution of soil moisture,salinity and organic matter in Manas River watershed,Xinjiang,China

With the classical statistical and geostatistical methods,the study of the spatial distribution and its influence factors of soil water,salinity and organic matter was carried out for 0-70 cm soil layers in Manas River watershed.The results showed that the soil moisture data from all soil layers exhibited a normal distribution,with average values of 14.08%-21.55%.Geostatistical analysis revealed that the content of soil moisture had a moderate spatial autocorrelation with the ratios of nugget/sill ranging from 0.500 to 0.718,which implies that the spatial pattern of soil moisture is influenced by the combined effects of structural factors and random factors.Remarkable spatial distributions with stripped and mottled features were found for soil moisture in all different soil layers.The landform and crop planting had a relatively big influence on the spatial distribution of soil moisture;total soil salinity was high in east but low in west,and non-salinized soil and lightly salinized soil appeared at the northwest and southwest of the study area.Under the effect of reservoir leakage,the heavily salinized soils are widely distributed in the middle of the study area.The areas of the non-salinized and lightly salinized soils decreased gradually with soil depth increment,which is contrary to the case for saline soils that reached a maximum of 245.67 km2 at the layer of 50-70 cm.The types of soil salinization in Manas River watershed were classified into four classes:the sulfate,chloride-sulfate,sulfate-chloride and chloride.The sulfate salinized soil is most widely distributed in the surface layer.The areas of chloride-sulfate,sulfate-chloride,and chloride salinized soils increased gradually along with the increment of soil depth;the variation range of the average values of soil organic matter content was between 7.48%-11.33%.The ratios of nugget/sill reduced gradually from 0.698 to 0.299 with soil depth increment,which shows that the content of soil organic matter has a moderate spatial autocorrelation.The soil organic matter in all soil layers met normal distribution after logarithmic transformation.The spatial distribution patterns of soil organic matter and soil moisture were similar;the areas with high organic matter contents were mainly distributed in the south of the study area,with the lowest contents in the middle.     

Identifying the influence of urbanization on soil organic matter content and p H from soil magnetic characteristics

Soil magnetic characteristics are correlated with soil p H and organic matter content. Analyzing soil magnetic characteristics, organic matter content and p H can indirectly evaluate soil pollution caused by human activities. This study analyzed the soil magnetic characteristics, organic matter content and p H in surface soil samples from different land use types in Shihezi city, a newly and rapidly developing oasis city in Xinjiang of China. The aims of this study were to explore the possible relationships among the soil magnetic parameters and thereby improve the understanding of influence of urbanization on soil properties. Eighty surface soil samples at the depth of 0–10 cm were collected from 29 July to 4 August 2013. The results showed that the magnetic minerals in surface soil were dominated by ferromagnetic minerals. Spatially, the magnetic susceptibility(χLF), anhysteretic remanent magnetization susceptibility(χARM), saturation isothermal remanent magnetization(SIRM) and "soft" isothermal remanent magnetization(SOFT) were found to be most dominant in the new northern urban area B(N-B), followed by built-up areas(U), suburban agricultural land(F), and then the new northern urban area A(N-A). The values of χLF, χARM, SIRM and SOFT were higher in the areas with high intensities of human activities and around the main roads. Meanwhile, the property "hard" isothermal remanent magnetization(HIRM) followed the order of UN-BFN-A. Built-up areas had an average p H value of 7.93, which was much higher than that in the new northern urban areas as well as in suburban agricultural land, due to the increased urban pollutant emissions. The average value of soil organic matter content in the whole study area was 34.69 g/kg, and the values in the new northern urban areas were much higher than those in the suburban agricultural land and built-up areas. For suburban agricultural land, soil organic matter content was significantly negatively correlated with χLF, and had no correlation with other magnetic parameters, since the soil was frequently ploughed. In the new northern urban areas(N-A and N-B), there were significant positive correlations of soil organic matter contents with χARM, SIRM, SOFT and HIRM, because natural grasslands were not frequently turned over. For the built-up areas, soil organic matter contents were significantly positively correlated with χLF, χARM, SIRM and SOFT, but not significantly correlated with frequency-dependent susceptibility(χFD, expressed as a percentage) and HIRM, because the soil was not frequently turned over or influenced by human activities. The results showed that soil magnetic characteristics are related to the soil turnover time.     

甘肃陇西中药材种植区土壤有机质分布特征

通过采集和分析甘肃省陇西县17个乡镇200个中药材种植区的土壤有机质含量数据,对陇西县中药材主产区的土壤有机质水平及储量空间分布格局进行研究。结果表明：县域各乡镇土壤有机质密度水平在27.03×10³~65.87×10³ kg·hm^-2范围之间,变异系数在19.30%~51.78%之间,属中等变异强度。陇西县土壤有机质含量的分布表现出自西向东逐渐减少的变化规律,并且局部地区具有聚集效应。土壤有机质密度随土壤深度增加呈现先上升后下降的整体趋势,其中表层0~10 cm的土壤有机质密度水平最低,为10.76×10³ kg·hm^-2,10~30 cm土层最高,为18.93×10³ kg·hm^-2。土壤有机质空间分布受到纬度、坡度、海拔、坡向的影响,纬度、海拔的影响程度最大。土壤有机质密度与降雨量、蒸发量、气温、相对湿度4个气象因子之间存在显著相关性,其中与降雨量、相对湿度之间呈显著、极显著负相关,与气温和蒸发量之间呈显著、极显著正相关。     

南疆沙漠腹地地温变化特征

利用塔中气象站三年的气象数据,分析了塔中地温的变化特征。结果显示:塔中0～40cm地温年变化规律显著,温度自1月开始迅速上升,在7月达到最高值,达30℃以上,其后又持续下降,冬季最低;80cm以下地温变化平缓,最高、最低温度出现月份逐渐后延,年振幅渐小;在典型晴天、阴天、沙尘天气时,表层地温都是日出前后最低,日出后迅速上升,至午后16时左右达最大值,其后迅速下降,于日落时降至一低值,并在夜间继续缓慢下降;四季0～80cm各层地温夏季﹥秋季﹥春季﹥冬季;天气对地温的影响主要体现在地表温度和浅层地温,至40cm及以下各层几乎无影响,日变化与典型晴天规律相同;阴天和浮尘对地温的影响较小,扬沙和沙尘暴天气对地温影响较大,其影响随深度增加而渐小;各层地温的平均日变化曲线近于正弦曲线,随深度增加地温特征线趋于平缓,极值出现时间逐渐后延。塔中地温的年际、日内变化的逐层性,与沙漠表面风成沙的物理属性有关。     

新疆阜康盐生植物园植物简介

为了保护盐生植物种质资源,建立长期研究盐生植物耐盐机理的基地和未来的开发利用,中国科学院新疆生态与地理研究所在阜康荒漠生态系统国家野外科学实验研究站内,历经6年,初步建成了以强耐盐盐生植物为主的植物园,占地约7 hm2。现已栽种102种盐生植物,隶属于20科59属,以当地藜科、柽柳科、豆科、禾本科为主。介绍了新疆阜康盐生植物园的植物种类、生活型、生境、耐盐型、功用等。新疆阜康盐生植物园位于扇缘地下水溢出带,是天然盐生植物集中分布区,该处的生态与环境自然选育了丰富的盐生植物种类,在此建园基本不会改变盐生植物的生物学和生态型特征。新疆是盐土的"博物馆",决定了盐生植物种类多样化,持续完善、维护该盐生植物园对盐生植物宝贵抗逆基因的研究和利用具有深远意义。     

天山北坡阜康市土壤侵蚀区域分异特征及防治重点

土壤侵蚀具有显著的区域分异性,研究土壤侵蚀分异特征对指导区域水土流失防治具有重要意义。文中以天山北坡阜康市为对象,通过综合评判法、空间叠置及地统计等方法,分析土壤侵蚀及主要影响因子空间分异特征,开展水土保持区域布局,划定重点防治范围与对象。结果如下:(1)土壤侵蚀及其影响因子空间分异明显,自然因素是产生土壤侵蚀区域分异的先决条件,人为活动的差异进一步加剧了土壤侵蚀区域分异。(2)区域内土壤侵蚀强度以中度-轻度为主,北部沙漠区以中度风蚀为主,中部平原区为轻度风水复合侵蚀,南部山丘区以中度水蚀为主;土壤侵蚀面积占土地总面积75.35%。(3)划分了北部沙漠防风固沙-生态维护区、中部平原农田防护-人居环境维护区、南部山地水源涵养-生态维护区3个水土保持分区;确定了7片重点防治范围、2448.71km~2的重点防治面积。研究成果可为天山北坡水土流失防治提供技术支撑,为新疆区域水土保持生态环境建设提供科学依据。     

阿尔泰山区土壤有机碳组成及分布规律研究

研究了阿尔泰山地土壤有机碳的组成及分布变化规律,研究认为,随着海拔高度的增加,土壤的易氧化有机碳逐渐减少,其氧化稳定系数,即Kos值逐渐增大,具有一定的垂直地带性;虽然土壤有机碳总量随海拔高度的增加而增加,但其有效肥力却逐渐减少,由低海拔到高海拔,土壤有机质逐渐以累积为主。