不同含水量土壤偏振光谱特征定量分析

水分含量是影响土壤偏振光谱特征的重要参数之一,研究土壤不同水分含量的偏振光谱特征在土壤偏振遥感波段的选择和图像解释上具有重要意义,为应用偏振遥感信息进行土壤调查及理化性质的分析提供依据。本文通过在350-2500 nm波段范围内对不同水分含量的土壤进行偏振光谱测试与分析,研究土壤偏振光谱数据与水分含量之间的关系,建模并检验其精度,对影响土壤偏振光谱特征的各个因素设计科学的正交实验,研究含水量、偏振角度、探测角和方位角等各个因素及其交互作用对土壤偏振光谱特征的影响。结果表明,含水量与偏振角度的交互作用和含水量本身对土壤偏振光谱的影响最大,显著性最强,其次是探测角与含水量的交互作用的影响,而偏振角对土壤偏振光谱有一定的影响,其他因素对土壤光谱偏振反射的影响不大。     

基于地貌类型的土壤有机质多光谱遥感反演

基于地貌类型分析土壤有机质含量与多光谱遥感影像光谱波段之间相关关系,构建不同地貌类型区有机质含量反演模型。结果表明,各波段光谱反射率与土壤有机质含量均呈负相关关系。利用SPSS软件对所有波段进行剔除变量(remove)线性回归分析,当全部波段参与构建反演模型时,一次反演模型拟合效果较好。分地貌类型区构建土壤有机质反演模型精度高于整个区域反演模型精度,与实际值对比,当允许误差为7%时,土壤有机质含量识别度为91.65%。基于地貌类型构建土壤有机质含量反演模型提取研究区土壤有机质含量切实可行,且精度较高。     

绿洲农田典型盐碱斑光谱特征与盐分反演建模

通过野外测定盐碱斑光谱反射率,对光谱数据进行去包络线处理,并分析盐碱斑光谱特征。结合盐碱斑表层土壤电导率,进行盐碱斑电导率和反射光谱相关性分析。结果表明,绿洲农田盐碱斑光谱反射曲线650~700 nm波段为盐碱斑电导率最敏感波段,可通过该波段数据与盐碱斑的电导率建立盐碱斑电导率光谱预测模型。     

基于决策树分类的滨海滩涂围垦区土地利用变化研究

本文以浙江省上虞市海涂围垦区为研究区,运用基于C4.5算法的决策树分类方法对研究区2001年和2010年两期遥感影像数据进行解译,获取土地利用类型及其变化数据。结果表明,两期影像的总体分类精度和kappa系数均达到了90%以上,分类结果理想。该研究区是典型的农林养殖区,土地利用类型以农用地和养殖场为主,共占研究区总面积的74%以上,其中耕地(水稻田和旱地)、果园、菜地是主要的农用地类型。10年间,各地类间存在耕地转变为建设用地、水稻田转变为养殖场、养殖场转变为水稻田、水体转变为滩涂和滩涂转变为水体等多种转变关系。围垦较早的区域内耕地和建设用地比例较大,土地利用变化主要是由耕地转变为建设用地;围垦较晚的子区内水体较多,养殖场与耕地、滩涂与水体存在相互转变的关系。土壤盐分是造成海涂围垦区土地利用变化的重要驱动力。     

基于PLSR的陕北土壤盐分高光谱反演

选取陕北盐渍土为研究对象,通过采集高光谱数据及土壤样品测定,研究土壤盐分含量与反射率之间相关性,遴选盐分特征波段,利用常规回归分析及偏最小二乘回归分析建立土壤盐分的定量反演模型,并利用检验样点进行对比分析和精度检验。研究结果表明,482 nm,1 365 nm,1 384 nm,2 202 nm及2 353 nm为土壤盐分含量的特征波段,利用高光谱数据进行盐分定量反演具有良好的精度;精度检验结果表明,通过Matlab进行偏最小二乘回归计算的反演模型,实测值与预测值相关性更好,精度较高。     

辽河口海涂资源卫星像片判读

海涂是一种独特的湿地环境和重要的自然资源,一般由潮上带、潮间带和潮下带构成。辽河口海涂资源丰富,类型多样,开发潜力较大。本文以遥感技术为主要手段,结合典型地段野外工作,采用多时段、多波段TM,MSS假彩色合成卫星像片和黑白卫星像片,依据地物的影像特征进行判读和制图,判读出了辽河口海涂资源的类型、空间格局和土地利用状况,并对海涂的成因、动态变化及分布规律进行了分析。为本区域海涂资源的开发、生态环境的保护、经济的规划和建设,提供了科学依据。     

天宫一号数据地表温度反演及其在城市热岛效应中的应用

针对天宫一号高光谱成像仪红外波段数据提出了一个单通道地表温度反演算法,算法的输入参数为大气水汽含量和地表发射率.利用模拟数据和黑河流域生态—水文过程综合遥感观测联合试验的地面实测数据对算法进行了精度评价,结果表明算法的均方根误差为2.72 K,能够满足大多数应用研究的需求.以北京市二环以内为研究区域,采用4个时相的天宫一号高光谱红外波段数据进行了城市热岛效应研究,结果表明天宫一号高光谱红外波段数据适合用来进行街区尺度的城市热岛效应研究,具有很大的应用潜力.     

不同类型土壤Cu含量高光谱联合反演建模

为探明不同类型土壤重金属Cu的敏感波段及构建普适性高光谱定量反演模型,该文以湖南省红壤、水稻土和潮土3种主要类型土壤为研究对象,在光谱预处理及组合变换基础上,采用相关性分析和逐步回归筛选重金属Cu敏感波段,并分别构建一元回归和逐步回归联合反演模型。结果表明,相较于原始光谱,组合变换光谱与土壤重金属相关性明显提高;通过逐步回归筛选重金属Cu的敏感波段位于400～850 nm和1 800～2 200 nm区域;相较于一元线性回归,逐步回归模型预测精度显著提升,应用对倒一阶微分光谱中400、590、620、670、790、850、1 790、2 270 nm波段反射率构建逐步回归模型反演精度达到最优,满足重金属Cu含量监测精度需求,同时为发展基于高光谱影像大面积反演不同土壤类型重金属Cu含量提供理论支撑。     

土壤钾含量高光谱定量反演研究

为了更快捷准确地进行土壤钾(K)含量的预测,基于土壤高光谱数据和化学元素分析数据,研究土壤光谱与土壤钾含量之间的定量关系.在对土壤原始光谱进行处理分析基础上,提取反射率(R)、反射率倒数的对数(log(1/R))、反射率一阶微分(R')和波段深度(BD)4种光谱指标,运用偏最小二乘回归方法建立相应的预测模型,并对模型进行检验.结果表明,波段深度是估算土壤钾含量最好的光谱指标,其建模精度超过0.85,均方根误差不超过0.1;全波段高光谱分辨率反射光谱具有快速有效估算土壤钾含量的潜力.     

黑土地硒含量的航空高光谱综合反演模型

针对单纯利用光谱信息建立土壤硒含量反演模型时,通常会产生模型精度受限、模型参数难以解释的问题,该文基于硒的空间分布影响因素开展综合建模.以黑龙江省海伦典型黑土区为研究区,利用CASI/SASI航空高光谱数据、土壤理化性质和地形因子,在筛选模型变量的基础上采用随机森林法建立综合模型.结果 表明:硒与土壤理化性质的关系非常密切,其中与有机质、全氮、全磷、Al2O3、Fe2O3、MgO、CaO、pH呈极显著正相关,与SiO2、Na2O呈极显著负相关;硒与地形因子的关系相对较弱,其中与起伏度和地形粗糙指数呈显著正相关.在光谱特征上,硒与光谱反射率的相关性主要体现了有机质和铁锰氧化物的光谱吸收特征.将筛选的相关指标按照不同组合方式进行建模和对比讨论,结果显示光谱、理化性质和地形因子均对提升模型精度有贡献,整体上光谱和理化性质占主导地位.当模型自变量为全部3类数据时,模型的建模和验证精度均为最高,表明综合模型不仅提升了硒元素的建模精度,而且改善了模型的稳健性.     

Saline degradation of Indian agricultural lands : A case study in Khambhat Taluka,Gujarat state (India), using satellite remote sensing

In certain agricultural fields of Khambhat Taluka in Gujarat State, the salinity has increased considerably rendering the land completely infertile. The occurrence of salinity in this area can be attributed partly to subsurface sea‐water ingress and partly to improper land and water management practices prior to implementation of irrigation. Landsat MSS or TM and IRS IA LISS II data was used to test the feasibility of delineating saline soils by both visual image interpretation and digital analysis. The study of saline soils using multi‐temporal Landsat images of the year 1977, 1983, and 1987, indicated an evident increase in saline areas in past few years. The Soil Brightness Index (SBI) generated from the IRS‐IA data by the application of MSS equivalent coefficients brought out different categories of soil degradation. The supervised classification scheme aided in generating various salinity levels. The analysis of the soil samples of the above area exhibited increasing values of Electrical Conductivity (ECe), and the soluble cations with increasing levels of salinity.     

Remote sensing of soybean stress as an indicator of chemical concentration of biosolid amended surface soils

The accumulation of heavy metals in the biosolid amended soils and the risk of their uptake into different plant parts is a topic of great concern. This study examines the accumulation of several heavy metals and nutrients in soybeans grown on biosolid applied soils and the use of remote sensing to monitor the metal uptake and plant stress. Field and greenhouse studies were conducted with soybeans grown on soils applied with biosolids at varying rates. The plant growth was monitored using Landsat TM imagery and handheld spectroradiometer in field and greenhouse studies, respectively. Soil and plant samples were collected and then analyzed for several elemental concentrations. The chemical concentrations in soils and roots increased significantly with increase in applied biosolid concentrations. Copper (Cu) and Molybdenum (Mo) accumulated significantly in the shoots of the metal-treated plants. Our spectral and Landsat TM image analysis revealed that the Normalized Difference Vegetative Index (NDVI) can be used to distinguish the metal stressed plants. The NDVI showed significant negative correlation with increase in soil Cu concentrations followed by other elements. This study suggests the use of remote sensing to monitor soybean stress patterns and thus indirectly assess soil chemical characteristics.     

Spectral reflectance characteristics of soils and its correlation with soil properties and surface conditions

In-situ spectral reflectance of soils was measured at various test sites of India in four spectral bands within the visible and near-infrared wavelength comparable to Landsat Multispectral Scanner (MSS). Reflectance behaviour of soils under different field conditions was analysed and the spectral reflectance curves for different soil types were obtained. Soil samples pertaining to each test site were analysed for mechanical composition, physioco-chemical properties to identify their relationship with soil reflectance. These spectral reflectance curves were further examined as to their usefulness in discriminating various soil types. Five distinct soil types namely, Black cotton soils (Typic Pellusterts), Marine Soils (Typic Halaquepts), Lateritic Soils (Plinthic Tropohumults), Alluvial Soils (Typic Ustochrepts), Coastal Sandy Soils (Typic Psammaquents), were discriminated on the basis of significant relationships between the spectral reflectance data and soil properties.     

Identification and delineation of saline soils using aerial photographs in Yamuna Alluvial plain,Haryana

Salinity is creating great problems in the Yamuna Alluvial Plain. State Government is taking steps for reclaiming these lands to increase agricultural production. Impeded drainage, excess canal irrigation, topography parent material, shallow depth and bad quality of ground water are the main factors which are responsible for the genesis of saline soils in the alluvial plains of Haryana. On the arial photographs tonal differences on account of salt concentration in the soil are clearly seen. These differences are aslo associated with physiography and soil conditions which can be easily delineated. The information thus; collected can be used for reclamation of saline soils.     

Possibilities of soil spectroscopy for the classification of contaminated areas in river floodplains

During the past decades, large amounts of diffuse contaminated soil material have been deposited in the floodplains of the river Rhine in the Netherlands. The dynamic character of this river causes a large spatial variability in the contamination level of its floodplain soils. Characterisation of the spatial variability exclusively based on soil sampling and analysis is often insufficient and expensive. Hyperspectral images can provide additional spatial information for a proper characterisation of the contamination situation of river floodplains. This paper describes the possible application of soil spectroscopy to estimate metal concentration levels in river floodplains. Soil reflectance spectra in the visible-near infrared region (VNIR) were measured in the laboratory for soil samples taken from two river floodplains along the river Waal, the main tributary of the river Rhine in the Netherlands. A multivariate calibration procedure using partial least squares (PLS) regression was applied to establish a relationship between reflectance spectra in the visible-near infrared (VNIR) region and spectrally active soil characteristics (organic matter and clay content) that are intercorrelated with concentration levels of Cd and Zn. Results of the analysis of two river floodplains are summarised and the influence of scale-level and sub soil material on the prediction capability is discussed.     

Detection of hydrocarbons in clay soils: A laboratory experiment using spectroscopy in the mid- and thermal infrared

Remote sensing has been used for direct and indirect detection of hydrocarbons. Most studies so far focused on indirect detection in vegetated areas. We investigated in this research the possibility of detecting hydrocarbons in bare soil through spectral analysis of laboratory samples in the short wave and thermal infrared regions. Soil/oil mixtures were spectrally measured in the laboratory. Analysis of spectra showed development of hydrocarbon absorption features as soils became progressively more contaminated. The future application of these results airborne seems to be a challenge as present and future sensors only cover the diagnostic regions to a limited extent.     

Spatial Variability of Some Chemical and Physical Soil Properties in Bandipora District of Lesser Himalayas

Soil is a suitable place for vegetation and plant growth. When this valuable resource is not preserved, shortage of food, erosion and damage of natural resources will be respected. Soil is a heterogeneous, diverse and dynamic system and investigation of its temporal and spatial changes is essential. In this paper spatial variability of some chemical and physical soil were investigated. Three hundred fifty eight soil samples were collected by systematic sampling strategy at 20 cm depth on a regular grid spacing of 500 × 500 m² under different vegetation cover and processed for analysis in the laboratory. Soil chemical and physical parameters including pH, electrical conductivity, organic carbon, available phosphorus, available nitrogen, available potassium, sulphur, calcium, magnesium and sodium were measured. After data normalization, classical statistical analysis was used to describe soil properties and geo-statistical analysis was used to illustrate spatial correlation of soil characteristics. By using interpolating techniques, spatial distribution of these properties were prepared. Results indicated that calcium and phosphorus had strong and weak spatial dependence, respectively.     

Spectral reflectance characterisation of different soils of Deccan plateau

Spectral reflectance of different soil subgroups from different soil orders were measured using a field radiometer. The results showed that, on the basis of spectral signatures, different soils can be delineated. The physico-chemical characteristics including moisture content showed a definite bearing on the spectral reflectance. The plant cover, tillage and crop residue content influenced greatly spectral reflectance as compared to bare soil. The spectral reflectance on soil aggregate size showed that reflectance decreased with increase in the aggregate size. The saline and sodic soils also recorded greater spectral reflectance in relation to normal soils.     

Using high-resolution,multispectral imagery to assess the effect of soil properties on vegetation reflectance at an abandoned feedlot

Remotely sensed multispectral imagery, soils and graminoid samples from an abandoned cattle feedlot and adjacent wetlands were used to characterize plant vigour and soil nutrient distribution and evaluate the relationship between soil properties and vegetation reflectance. The feedlot lies on a sandy beach ridge, which likely mitigates the mobility of soil phosphorus. Soil phosphorus remains concentrated directly beneath the feedlot pens, where vegetation indices are low. In contrast, nitrate is transported through preferential pathways into the wetlands, where vegetation indices and plant vigour are high. Although spectral vegetation indices did not show any significant relationship with plant tissue nutrient concentration, the indices showed statistically significant relationships to some soil properties. Results of this study indicate that the abundance of nutrients in the soil does not necessarily enhance plant growth. This can limit the extent that remotely sensed vegetation indices can be used to evaluate soil nutrients concentrations.     

Quasi-quantitative relationship between soil moisture and\npolarization characteristics

Besides amplitude, frequency and phase, the polarization is another basic property of the electromagnetic wave. In the remote sensing field, the polarization is mainly applied in active detection systems of radar and lidar. This paper presents the quantitative relationship between soil moisture and polarization signatures in a certain type of soil in a farm. And this relationship is expected to be introduced on agriculture and hydrology ultimately. The experiments were performed both in the laboratory and the field. Soil samples with different moisture contents were measured at three wavebands on visible spectrum, and at several viewing angles in the plane of incidence. The polarization signature was indicated by the multi-band and multi-angle degree of linear polarization (DOLP) in this paper. The soil moisture were divided into five levels according to the properties of DOLP curves, namely, the quasi-quantitative relationship between soil moisture and its polarization signature were established. The percentages of soil moisture of the five levels are: ≤10%, 10%—20%, 20%—40%, 40%—56% and >56%, respectively. Although this division for soil moisture is on a rather large scale, it will meet the precision of application agricultural and hydrologic remote sensing.