共查询到18条相似文献，搜索用时 156 毫秒
1.
利用AMSR-E 36GHz、89 GHz亮度温度计算出的极化比(PR)与热力学冰厚比较,建立了一种针对南极罗斯冰架冰间湖的AMSR-E薄冰反演算法。热力学冰厚由用MODIS晴空下的表面温度数据、ERA-Interim气象数据,根据表面热通量平衡公式反演得到。在排除混合像元以及水汽对89 GHz数据的干扰后,分别对36、89 GHz的AMSR-E亮温极化比拟合指数模型,反演罗斯冰架冰间湖中的0~0.2 m的薄冰冰厚,均方根误差分别为0.003 5 m和0.005 3 m。 相似文献
2.
基于风云-3B（FY-3B）卫星的微波成像仪（MWRI）数据对HUT模型（Helsinki university of technology snow emission model）进行验证,结果表明,无论是18.7 GHz还是36.5 GHz水平极化亮温,HUT模型模拟亮温都与MWRI亮温存在较大的偏差。因此,本文对消光系数进行了本地化改进,得到了改进的HUT模型（IMPHUT模型）。IMPHUT模型在18.7 GHz水平极化和36.5 GHz水平极化时的模拟亮温偏差分别为-0.91 K和-4.19 K,较原始的HUT模型模拟精度（偏差分别为14.03 K和-16.33 K）有很大提高。最后,利用遗传算法进行雪深反演,基于IMPHUT模型的雪深反演（偏差为-6.79 cm）优于HUT模型和Chang算法,反演与实测雪深具有较好的一致性。 相似文献
3.
《武汉大学学报(信息科学版)》2016,(6)
在积雪深度研究中,地面资料插值产生的平滑效应以及遥感空间分辨率不足的问题,在很大程度上影响着积雪深度的估计精度。本文采用中高分辨率成像光谱仪(moderate resolution imaging spectro-radiometer,MODIS)和微波扫描辐射计(advanced microwave scanning radiometer-EOS,AMSR-E)融合后的无云积雪面积产品构建虚拟站点,弥补了气象站点少且不均匀的不足,修正雪深克里金插值产生的平滑效应。同时,提出了基于数据同化算法融合以地面观测资料为基础的克里金空间插值雪深、MODIS积雪面积产品和AMSRE微波反演雪深产品的雪深估计方法。以新疆北疆地区为研究区域进行了算法应用及验证,并选取不同海拔的站点观测资料对融合结果进行验证分析,通过均方根、偏差和相关性系数指标检证了该方法能够有效地提高雪深估计精度。 相似文献
4.
基于被动微波遥感的青藏高原雪深反演及其结果评价 总被引：21，自引：0，他引：21
采用修正的张氏雪深反演算法，用ＳＳＭ／Ｉ３７ＧＨz和19GHz水平极化亮温值计算了青藏高原及其毗邻地区的积雪深度，对其精度进行了评价，并对误差来源进行了分析，结果显示，此算法能够较好地反映研究区的雪深分布，但局部地区误差较大，总体上雪深被高估，其误差主要来源于冻土，深霜层，植被以及雪层中液态水含量，雪粒的形状和粒径的变化带来的影响，SSM／I数据较低的分辨率和研究区复杂的地形使反演的雪深与观测的雪深缺少可比性，给精度的评价带来影响。 相似文献
5.
被动微波遥感积雪参数反演方法进展 总被引：1，自引：0，他引：1
雪深(snow depth,SD)和雪水当量(snow water equivalent,SWE)是气候水文研究中的重要参数,在雪灾监测中尤为重要。首先,简要介绍了被动微波遥感SD和SWE反演算法的物理基础——积雪微波辐射传输模型,分析了不同微波频段、不同特点的积雪微波辐射和散射特性。然后,根据前人的研究从数学角度将反演算法分为线性亮温梯度法和基于先验知识法,总结了2类算法的优势和局限性:线性亮温梯度法相对简单、速度快,一般只适用于特定的研究区;先验知识法需要获取研究区的样本数据,并反复训练才能达到较好的精度,但对样本的独立性及其均值差异显著性的要求较高。最后,重点介绍了我国风云三号微波成像仪(FY-3 MWRI)的全球SD和SWE反演算法和针对中国区域的改进算法,并对未来的研究热点进行了展望。 相似文献
6.
针对单一被动微波遥感反演雪深的精度和空间分辨率不足的问题,提出了一种星-地多源数据融合的雪深反演方法。以北疆每日站点观测雪深、AMSR-E遥感反演雪深和SSM/I遥感反演雪深数据为研究对象,首先利用地统计方法结合地面站点观测数据估计北疆区域的雪深,然后采用Triple-Collocation方法分别估计三个雪深数据的误差方差,最后结合最小二乘原理实现星-地雪深观测数据的融合。对融合雪深与地面雪深观测数据进行验证分析,结果显示,与AMSR-E和SSM/I遥感反演雪深相比,融合的雪深与地面观测雪深的相关性更高;融合的雪深的精度有一定程度的提高。实验结果证明了多源数据融合方法的有效性。 相似文献
7.
利用GNSS-MR(Global Navigation Satellite System Multipath Reflectometry)技术反演积雪深度是近年来一种新兴的卫星遥感技术。目前大多数研究仅使用GPS(Global Position System)数据限制了该技术的发展,为了扩展GNSS-MR算法的应用,介绍了基于GNSS-MR算法的雪深反演模型。首先,通过多项式拟合分解GLONASS观测数据获取高精度的信噪比残差序列;然后,利用Lomb-Scargle谱分析法对其进行频谱分析可解算雪深值。选取IGS中心的YEL2站2015年11月到2016年6月共243天的GLONASS卫星L1波段反射信号的SNR数据进行实例分析,并以美国国家气象数据中心提供的加拿大Y-H (Yellowknife Henderson)气象站的实测雪深数据为真值,将反演雪深与实测雪深进行对比验证。所得实验结果如下:(1)与GPS卫星的反演值相比,基于GLONASS-MR(GLONASS Multipath Reflectometry)技术反演积雪深度的精度同样能达到厘米级,RMSE仅3.3 cm,反演值与实测值的空间分布趋势一致且相关性较强,其相关系数R2高达0.969;(2)不同的积雪深度对信噪比的振幅频率与垂直反射距离具有直接影响;(3)对同一卫星而言,信噪比的频谱振幅强度峰值与其对应的反演值存在线性相关;(4)在相同条件下,采用多颗GLONASS卫星数据比单颗GLONASS卫星数据反演雪深的效果明显更优。基于反演的高时间分辨率产品,分析该地区雪深日变化的情况,实验结果表明基于陆基CORS站的GLONASS-MR技术在用于实时、连续的雪深变化监测方面具有良好的潜力和可行性。 相似文献
8.
基于GPS新型L5信号的地表雪深反演研究 总被引：1，自引：0，他引：1
利用GPS多路径反射信号测量地表雪深具有全天候和高时空分辨率的特点，因此其可作为一种代替气象站监测雪深的新手段。然而，先前大多数研究仅使用了GPS L1和L2C波段信噪比数据探测积雪深度。为验证新型的L5信号在雪深反演方面的优越性，本文阐述了GPS-R技术反演雪深的原理，利用Lomb-Scargle周期图法所处理的受积雪表层影响的信噪比数据计算了频谱振幅强度，通过获取频谱特征值与天线高度的关系求解雪深值，最后分别与L1反演结果和实测雪深数据进行了对比。试验结果表明：与现有的GPS-R测量雪深结果相比，利用新型的L5反射信号反演地表雪深的精度更佳；采用GPS-R技术探测雪深对把握测站区域内的雪深变化情况和淡水资源储量具有重要价值。 相似文献
9.
神舟四号飞船微波辐射计定标和检验（Ⅱ）——微波辐射计地物参数反演及其检验 总被引：3，自引：0，他引：3
针对神舟四号飞船微波辐射计(RAD)特殊入射角、频率的特点,开发了海面和大气参数的反演算法。然后用这个算法模拟RAD不同通道的亮温。最后用发射后重新定标的亮温结合国外大洋浮标、小岛上的气象探空数据以及其他星载微波辐射计产品,对RAD反演算法进行改进,最后给出海面温度、风速和大气水汽含量的反演的结果,并且对这些结果进行了检验。 相似文献
10.
研究了GPS干涉反射技术GPS-IR(GPS-Interferometric Reflectometry);在利用GPS卫星SNR信号进行积雪深度探测的基础上构建了支持向量机SVM(Support Vector Machine)辅助的GPS SNR雪深时间序列反演模型;对积雪深度进行时间序列预报和与传统GPS-IR积雪探测模型进行精度对比分析。实验结果显示,相比传统GPS-IR雪深反演模型,SVM辅助的GPS SNR雪深时间序列反演模型的雪深预报结果的精度更高,也更符合实测雪深的变化趋势,可为地面积雪雪深反演提供新方法。 相似文献
11.
中国西部积雪SMMR微波遥感的评价与初步应用 总被引：3，自引：0，他引：3
本文根据1978—1987年OSL的可见光影像及地面台站积雪记录,评价了NASA的用微波亮温反演积雪深度的算式。还根据DEM资料,用GIS技术把中国西部分成高山、高原、低山、丘陵及盆地五个单元,分别求得各区域订正算式,并以此计算了中国西部1978—1987年各季的平均雪盖率与雪量及它们的年际变化,为本区积雪影响东亚气候研究,提供了可靠基础资料。 相似文献
12.
Shui-sen Chen Xiu-zhi Chen Wei-qi Chen Yong-xian Su Da Li 《International Journal of Applied Earth Observation and Geoinformation》2011
The analysis of the passive microwave radiance transfer equation certifies that there is a linear relationship between satellite-generated brightness temperatures (BT) and in situ observation temperature and that land surface temperature (LST) is largely influenced by vegetation cover conditions. Microwave polarization difference index (MPDI) is an effective indicator for characterizing the land surface vegetation cover density. Based on the analysis of LST models from AMSR-E BT with 6.9 GHz MPDI intervals at 0.04, 0.02 and 0.01, respectively, this paper developed a simplified LST regression model with MPDI-based five land cover types, combining observation temperatures from 86 meteorological observation stations. The study shows that smaller MPDI intervals can obtain higher accuracy of AMSR-E LST simulation, and that the combination of HDF Explorer and ArcGIS software was useful for automatically processing the pixel latitude, longitude and BT information from the AMSR-E HDF imagery files. The RMSE of the five LST simulation algorithms is between 1.47 and 1.92 °C, with an average LST retrieval error of 0.91–1.30 °C. Besides, only 7 polarization bands and 5 land surface types are required by the proposed simplified model. The new LST simulation models appears to be more effective for producing LST compared to past most studies, of which the accuracy used to be more than 2 °C. This study is one of the rare applications that combine the meteorological observation temperature with MPDI to produce the LST regression analysis algorithms with less RMSE from AMSR-E data. The results can be referred to similar areas of the world for LST retrieval or land surface process research, in particular under extreme bad weather conditions. 相似文献
13.
Lili Wu Xiaofeng Li Kai Zhao Xingming Zheng 《Journal of the Indian Society of Remote Sensing》2017,45(1):89-100
Snow depth parameter inversion in the farmland using passive microwave remote sensing is of great significance to the agricultural production in Northeast China. Firstly, the Helsinki University of Technology (HUT) snow emission model was validated in the farmland based on microwave radiation imager (MWRI) onboard FengYun-3B satellite (FY-3B). The results showed that there was a big difference between the brightness temperature of HUT model simulation and MWRI for 18.7 GHz horizontal polarization (18.7 H) and 36.5 GHz horizontal polarization (36.5 H). To improve HUT model, the empirical parameter in the model was localized. Then the localized HUT (LHUT) model was built, where the extinction coefficient was calculated by the new extinction coefficient formula. Next, LHUT model was validated based on MWRI data and compared with HUT model. The results showed that LHUT underestimates slightly the brightness temperature with 0.91 and 4.19 K for 18.7 and 36.5 H respectively, and LHUT is superior to HUT model. Finally, the genetic algorithm (GA) was used to invert snow depth based on LHUT. The results showed that snow depth was underestimated with 6.79 cm based on LHUT. The inverted snow depth based on LHUT model is in better agreement with the measured snow depth. 相似文献
14.
Spaceborne passive microwave data have been available for the past 27 years, and have supported the development of several algorithms for the retrieval of snow water equivalent and snow depth that, in turn, can be used for mapping snow-covered areas. In contrast, only recently has the application of spaceborne active microwave instruments been investigated for remote sensing of snow on a global scale. This raises the question of whether a technique combining active and passive microwave data can improve the mapping of snow parameters with respect to techniques based solely on passive data. In this letter, we report results concerning the mapping of snow-covered area (SCA) in the Northern Hemisphere between the years 2000 and 2004 derived from the combination of the brightness temperatures at 19.35 and 37 GHz measured by the Special Sensor Microwave Imager Radiometer with backscatter coefficients at 13.4 GHz measured by the NASA's QuickSCAT. SCA derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) is used as a reference to evaluate the performance of the microwave-based techniques and their combination. Results show that, generally, the technique using passive data provides better agreement with MODIS SCA than the technique using only scatterometer data. However, the results when both datasets are used show considerable improvement, demonstrating the potential benefits of a multisensor approach 相似文献
15.
16.
V. D. Mishra P. Mathur R. P. Singh 《Journal of the Indian Society of Remote Sensing》2005,33(3):381-393
Snow cover is an important variable for climatic and hydrologic models due to its effect on surface albedo, energy, and mass
balance. Satellite observations successfully provide a global and comprehensive hemispheric-scale record of the short-term,
as well as inter-seasonal variations in snow cover. Passive microwave sensors provide an excellent method to monitor temporal
and spatial variations in large-scale snow cover parameters, overcoming problems of cloud cover. Using microwave remote sensing
data, snow parameters (snow surface temperature, snow water equivalence, scattering index, emissivity, snow depth) have been
retrieved to integrate with the snow cover simulation model developed by SASE for avalanche risk assessment on regional basis.
Multispectral and multitemporal brightness temperature data obtained from the Special Sensor Microwave Imager (SSM/I), flown
onboard the DMSP satellites, for the period November 2000 to April 2001 and from November 2001 to February 2002 have been
analysed. A comparative data set on snow measurements and meteorological observations of a region covering large area of Pir-Panjal
and the Greater Himalayan range, available on near real time basis from SASE field observatories were also used. Model calculations
were carried out to study the effects of atmospheric transmission on the microwave radiation emitted from the snow covered
and snow free ground and atmosphere. The sensitivity of combinations of the SSM/I channels at 19, 37 and 85 GHz, in both horizontal
and vertical polarizations, in respect to snow depth, surface temperature of the snowpack have been carried out. Decision
rule based algorithms are developed to identify snow cover and non-snow area. 相似文献
17.
青海高原春秋季地表土冻融的微波遥感监测 总被引：5，自引：0，他引：5
用青海高原地区 1983年的SMMR18及 37GHz水平极化微波亮温记录与台站地表 5cm深温度实况相比较 ,统计确定部分考虑含水量对区分地表土冻融影响的亮温T37H及亮温谱梯度SG两参数综合模式 ,区分的正确率达 74%。青海高原地处干旱半干旱环境 ,春季地表土含水量低 ,Zuerndorfer提出的负亮温谱梯度 ,在春季区分冻融的效果并不明显。该文作了改进 ,效果明显 相似文献
18.
MPDI在微波辐射计植被覆盖区土壤水分反演中的应用 总被引：5，自引：0，他引：5
大尺度上的土壤水分变化监测对于建立全球的水循环模型意义重大，是实现气候变化预测和洪涝监测的基础。星载辐射计为实现大尺度上土壤水分的监测提供了监测途径。但是在星载辐射计观测时，地表植被层的吸收和散射作用会对土壤向上的微波辐射产生衰减影响，这种影响在反演土壤水分的过程中必须予以计算和消除。原有的反演算法中，在计算这部分影响的时候，需要大量的关于地表植被状况的辅助数据，而这些即时的辅助数据往往不易获得。以AMSR—E数据为例，研究证明了微波极化差异指数（MPDI）能够反映地表植被覆盖状况。以中国华北、华东地区为实验区，选择2004年4月8日的AMSR—E亮温数据和MODIS数据为样本数据，建立起MPDI与NDVI之间的负指数关系方程。基于对NDVI的认识，得到植被覆盖度高、中、低三种状况所对应的MPDI域值，以此域值为依据对中等植被覆盖度地区作出自动判断，并用MPDI计算植被层不透明度。 相似文献