青藏高原东侧MODIS地表温度产品验证

利用青藏高原东侧理塘大气边界层观测站的热红外测温仪和长波辐射表的实测数据,分析了实测地表温度Ts的精度及其影响因子,对MODIS地表温度(LST)产品进行了评估。结果表明,与用红外测温仪测量的地表辐射温度Tr相比,考虑了地表辐射率和大气下行辐射后的地表温度Ts_R偏低,差值随着地表发射率递减,随着长波净辐射递增,白天偏低可达1.9 K,夜间偏低可达1.1 K。当地表温度285 K时,Ts_R与用长波辐射估算的地表温度Ts_L之间的差值集中在2.5 K以内;而当地表温度285 K时,偏差可达10 K。地表发射率的误差≤0.01时,地表温度计算误差≤0.5 K。Tr、Ts_R和Ts_L与LST之间都存在显著正相关,其中Ts_L与LST的相关系数最大。夜间的相关系数大于白天;Terra卫星的相关系数(尤其是白天)大于Aqua卫星。基于LST的Ts_L回归模型估算标准误差为4.4904 K,达到99%的置信水平。     

风云三号C星微波全球地表温度产品精度评估

地表温度是衡量地表水热平衡的关键参数,微波地表温度因其范围大、全天候等独特的优势,在气候、农业和环境等领域得到广泛应用。基于经质量控制的MODIS地表温度产品对风云三号卫星C星的微波地表温度日产品和月平均产品进行验证评估,结果显示:FY-3C卫星升轨(夜晚)和降轨(白天)微波地表温度日产品平均分别高估8. 7 K、低估13. 2 K,月平均产品平均分别高估7. 9 K、低估12. 0 K,日产品和月平均产品的反演误差都在15K以内。在全球空间分布上,升轨(夜晚)和降轨(白天)月产品误差分别呈现高估和低估,热带雨林区和沙漠、荒漠区域在夜晚分别高估5 K以内和30 K以内,白天则分别低估10 K以内和10～30 K。不同土地覆盖类型间FY-3C微波地表温度反演精度存在差异,总体上升轨(夜晚)比降轨(白天)的精度高,反演精度最高和最低的土地类型分别是常绿阔叶林和荒漠、稀疏植被,不同土地覆盖类型间的地表温度反演精度在季节上存在明显差异。根据分析结果,改进FY-3C微波地表温度的反演算法,可进一步提高微波地表温度的反演精度。     

藏北高原GlobAlbedo地表反照率的精度分析

为了分析欧洲航天局多星观测数据联合反演的全球地表反照率产品Glob Albedo在青藏高原的反演精度,促进其在青藏高原地—气相互作用研究中的应用,利用藏北高原BJ站和西大滩站观测的上行和下行太阳短波辐射资料,对比分析了Glob Albedo的精度,并与MODIS地表反照率产品MCD43B3进行了比较。结果表明:空间分辨率1 km的Glob Albedo短波波段(0.3~5.0μm)的地表反照率与地面观测结果总体上具有较好的一致性,但是精度受积雪覆盖比例的影响较大。积雪覆盖比例0.1时,Glob Albedo短波波段的地表反照率与高质量地面观测结果的均方根误差介于0.0100~0.0218,Glob Albedo的精度完全能够满足气候和陆面模式的精度要求。反之,它们的均方根误差介于0.0252~0.1461,存在较大的不确定性。对比Glob Albedo和MCD43B3,前者的精度略高于后者:Glob Albedo短波波段地表反照率与高质量地面观测结果的均方根误差介于0.0195~0.0959,MCD43B3短波波段地表反照率与高质量地面观测结果的均方根误差介于0.0273~0.1269。     

基于MODIS资料的宁夏LST反演方法新探索

为快速、宏观、全面地获取陆面生态重要参数陆面温度(LST),避免分裂窗算法中诸多参数的估计和参数的适用范围限制,加快计算速度,更好地利用中国气象局"三站四网"的建设成果,利用宁夏2005-2007年13个时次过境晴空地表MODIS资料及对应过境时17个自动气象站观测数据,筛选、优化引入对LST影响较大的水汽通道、NDVI和EVI参数,建立基于MODIS遥感和地面自动气象站观测数据反演陆面温度(LST)的统计模式.研究结果表明:引入相关参数后,宁夏各季及全年模式的相关性和精度有较大提高,且水汽通道和EVI的参数组合最优.与分裂窗算法相比,省去了对大气透过率的估算以及对地表比辐射率估计的繁琐计算,与地面自动站观测真实值误差70.1%能够控制在4.0℃以内,计算速度快,能够满足一般业务的需求,易于推广使用.     

FY-3A陆表温度反演及高温天气过程动态监测

采用FY-3A/VIRR数据,利用Becker局地分裂窗改进算法反演得到逐日陆表温度 (LST), 对2009年一次高温天气过程进行动态监测, 并分析不同下垫面的热环境变化。结果显示:此过程中可见光红外扫描辐射计 (VIRR) 陆表温度产品在敦煌辐射校正场地两次验证的误差为-0.17 K和1.77 K,与同时间过境的MODIS产品均方根误差为2.64 K,直方图对比陆表温度的频数分布基本一致;对高温天气过程监测发现,此次出现以华北的石家庄、郑州、北京等地和西北地区东部的西安等地为中心的两个陆表温度高值区, 部分地区达到了320.2 K以上;城市剖面资料证实城市热岛现象存在,并发现工矿用地的热岛效应不容忽视,主要是大面积的工矿用地周围植被破坏严重,地表增温更为显著。     

青藏高原东北部MODIS地表温度重建及其与气温对比分析

通过遥感获取地表温度(地温,LST)可以弥补气象站LST数据局地性的不足.但受某些因素影响,遥感LST影像存在噪音像元,影响了LST数据的应用.本文提出了一种基于高程-温度回归关系的空间重建算法,对2008年青藏高原MODIS LST影像异常低值像元进行了重建,得到空间完整的LST时间序列.分别将原始LST和重建LST...     

基于多源遥感数据的辽宁地表温度反演及空间分布研究

以辽宁地表温度为研究对象,采用普适性单通道算法,利用FY-3A/MERSI数据,并结合MODIS 1000 m分辨率数据,反演了2009年和2010年4-9月间10个时次晴空或局部晴空时的地表温度。结果表明：计算验证了模型的反演精度与同期NASA所发布MODIS地表温度产品的精度相当,其结果与相应的56个气象站点的实际观测数据相一致。多源遥感数据的综合应用,可获得较合理的地表温度反演结果;不同土地覆盖类型间地表温度的高低在相同时间内存在显著差异;研究期内,林地、水田、旱地和建设用地的NDVI与地表温度具有负相关性。综合利用遥感、地理信息系统技术,可以表征地表温度与土地利用类型以及地表温度与归一化植被指数(NDVI)之间的关系。     

基于哨兵2/3卫星高分辨率地表温度的降尺度方法

针对卫星遥感数据提取或生成地表温度（land surface temperature, LST）存在的时空分辨率矛盾,利用哨兵2/3卫星产品数据,其中哨兵3数据提供高时间分辨率影像,哨兵2数据提供高空间分辨率信息,并利用SNAP70及Excel2010软件,建立归一化植被指数与LST的相关关系,利用统计降尺度方法,成功将LST的空间尺度从1 000 m降至10 m,生成高时间分辨率10 m空间分辨率的LST。将原始1 000 m分辨率哨兵3地表温度图与降尺度到10 m空间分辨率的地表温度图对比,可以发现：降尺度地表温度图可以覆盖大部分原始1 000 m分辨率的地表温度信息,说明降尺度结果较好地保留了原始LST影像热特征的分布情况;而且,所生成的高空间分辨率的地表温度产品地物特征清晰,纹理明显。利用地面国家气象自动观测站实测0 cm地温数据验证降尺度结果,可以看出：误差平均值为26 K,误差值较小,说明降尺度结果精度较高。     

基于AVHRR和VIRR数据的改进型Becker“分裂窗”地表温度反演算法

为了将基于NOAA-9/AVHRR数据提出的Becker和Li的“分裂窗”地表温度算法成功地应用于长序列NOAA/AVHRR和FY 3A/VIRR数据的地表温度反演,为气候变化研究提供长序列、高精度、高分辨率的地表温度数据集,从辐射传输方程出发,首先利用MODTRA 4.1模式模拟了多种地表和大气状态下的光谱辐亮度数据,并结合AVHRR和VIRR通道4、5的光谱响应函数建立了温度数据集(TS,T4,T5);然后,基于该数据集采用最小二乘法重新计算了Becker和Li算法中的各参数,提出了一个适用于NOAA/AVHRR和FY-3A/VIRR数据的改进型Becker和Li分裂窗地表温度反演算法;并利用改进型算法对2008年4月27日03时12分(世界时)观测的一景覆盖北京地区的NOAA-17/AVHRR数据进行了地表温度的反演,将反演结果与日本东京大学提供的同地区、同时相的MODIS地表温度产品进行了对比分析.结果表明,两种地表温度产品的相关系数为0.88,均方根偏差(RMSD)为2.1K;在两种地表温度差值图像的频率直方图上有69.6％的像元的值在±2K之内,37％的像元的值在±1K之内.     

基于MODIS数据的金塔绿洲地表温度反演

地表温度(LST)是气象、水文、生态等研究领域中的一个重要参数.本文对MODIS数据的分裂窗算法进行了简要的介绍,并利用MODIS数据计算反演了地表温度所需的关键参数:大气透过率和地表比辐射率,然后运用分裂窗算法反演了金塔绿洲地区的地表温度,并与地面实测数据进行了对比分析.结果表明,这一方法能获得较合理的地表温度,符合金塔绿洲的实际地表状况.     

A Remote Sensing Model to Estimate Sunshine Duration in the Ningxia Hui Autonomous Region, China

Sunshine duration (SD) is strongly correlated with solar radiation, and is most widely used to estimate the latter. This study builds a remote sensing model on a 100 m × 100 m spatial resolution to est...     

基于MODIS遥感数据的混合像元分解技术研究和应用

使用郑州市MODIS（Moderate-Resolution Imaging Spectroradiometer）遥感数据,运用线性混合模型,对MODIS遥感数据进行混合像元分解技术研究。探讨了MODIS遥感数据的预处理、线性光谱分解模型、图像端元组分反射率的求取方法。把结果与分辨率较高的Landsat ETM＋图像分类结果进行对比,并根据得到的均方根误差（RMS;Root Mean Square）进行分析表明,利用这种像元分解方法得到的结果较为理想,MODIS数据可以有效地应用于遥感动态监测和土地覆盖分类研究。     

应用MODIS数据反演青藏高原地区地表反照率

应用RossThick-LiTransit核驱动BRDF（bidirectional reflectance distribution function）模型,选择2004年Terra MODIS（moderate resolution imaging spectraradiometer）500 m分辨率数据,对青藏高原地区的地表反照率进行了反演研究,并以平均气溶胶光学厚度值0.11计算了正午时（北京时间12：00）实际的地表反照率,反演结果与当地的地表覆盖类型和地形具有较好的一致性。此外,藏北高原4个辐射观测站点观测资料与反演结果的比较表明,500 m分辨率反演结果不仅可以满足气候和陆面过程模式的精度要求,而且精度高于美国1 km分辨率反照率反演结果。     

Comparison between MODIS-derived Day and Night Cloud Cover and Surface Observations over the North China Plain

Satellite and human visual observation are two of the most important observation approaches for cloud cover. In this study, the total cloud cover(TCC) observed by MODIS onboard the Terra and Aqua satellites was compared with Synop meteorological station observations over the North China Plain and its surrounding regions for 11 years during daytime and7 years during nighttime. The Synop data were recorded eight times a day at 3-h intervals. Linear interpolation was used to interpolate the Synop data to the MODIS overpass time in order to reduce the temporal deviation between the satellite and Synop observations. Results showed that MODIS-derived TCC had good consistency with the Synop observations; the correlation coefficients ranged from 0.56 in winter to 0.73 in summer for Terra MODIS, and from 0.55 in winter to 0.71 in summer for Aqua MODIS. However, they also had certain differences. On average, the MODIS-derived TCC was 15.16%higher than the Synop data, and this value was higher at nighttime(15.58%–16.64%) than daytime(12.74%–14.14%). The deviation between the MODIS and Synop TCC had large seasonal variation, being largest in winter(29.53%–31.07%) and smallest in summer(4.46%–6.07%). Analysis indicated that cloud with low cloud-top height and small cloud optical thickness was more likely to cause observation bias. Besides, an increase in the satellite view zenith angle, aerosol optical depth, or snow cover could lead to positively biased MODIS results, and this affect differed among different cloud types.     

Terra和Aqua MODIS气溶胶光学厚度产品在中国热带沿海地区的适用性评估

This study compares the aerosol optical depth (AOD) Level 2 Collection 5 products from the Terra and Aqua Moderate Resolution Imaging Spectroradiometers (MODIS) with ground-based measurements from a Microtops II sun photometer over Sanya (18.23°N, 109.52°E), a tropical coastal site in China, from July 2005 to June 2006. The results indicate that the Terra and Aqua MODIS AOD retrievals at 550 nm have good correlations with the measurements from the Microtops II sun photometer. The correlation coefficients for the linear regression fits (R²) are 0.83 for Terra and 0.78 for Aqua, and the regressed intercepts are near zero (0.005 for Terra, 0.009 for Aqua). However, the Terra and Aqua MODIS are found to consistently underestimate AOD with respect to the Microtops II sun photometer, with slope values of 0.805 (Terra) and 0.767 (Aqua). The comparison of the monthly mean AOD indicates that for each month, the Terra and Aqua MODIS retrievals are matched with corresponding Microtops measurements but are systematically less than those of the Microtops. This validation study indicates that the Terra and Aqua MODIS AOD retrievals can adequately characterize the AOD distributions over the tropical coastal region of China, but further efforts to eliminate systematic errors are needed.     

STUDY ON CLOUD CLASSIFICATIONS BY USING AVHRR,GMS-5 AND TERRA/MODIS SATELLITE DATA*

This paper presents the automated pixel-scale neural network classification methods being developed at National Satellite Meteorological Center(NSMC) of China to classify clouds by using NOAA/AVHRR and GMS-5 satellite imageries.By using Terra satellite MODIS imageries,an automated pixel-scale threshold technique has been developed to detect and classify clouds.The study focuses on applications of these cloud classification techniques to the Huaihe River and the Changjiang(Yangtze) River drainage basin.The different types of clouds show more clearly on this cloud classification image than single band image.The results of the cloud classifications are the basis of studying cloud amount,cloud top height and cloud top pressure.Cloud mask method sare widely used in SST,LST,and TPW retrieval schemes.Some case studies about cloud mask and cloud classification in satellite imageries,which are related with the study of Global Energy and Water Cycle Experiment(GEWEX) in the Huaihe River and the Changjiang River drainage basin are illustrated.     

Diurnal and seasonal variations of wind farm impacts on land surface temperature over western Texas

This paper analyzes seasonal and diurnal variations of MODerate resolution Imaging Spectroradiometer (MODIS) land surface temperature (LST) data at ~1.1 km for the period of 2003–2011 over a region in West-Central Texas, where four of the world’s largest wind farms are located. Seasonal anomalies are created from MODIS Terra (~10:30 a.m. and 10:30 p.m. local solar time) and Aqua (~1:30 a.m. and 1:30 p.m. local solar time) LSTs, and their spatiotemporal variability is analyzed by comparing the LST changes between wind farm pixels (WFPs) and nearby non wind farm pixels (NNWFPs) using different methods under different quality controls. Our analyses show consistently that there is a warming effect of 0.31–0.70 °C at nighttime for the nine-year period during which data was collected over WFPs relative to NNWFPs, in all seasons for both Terra and Aqua measurements, while the changes at daytime are much noisier. The nighttime warming effect is much larger in summer than winter and at ~10:30 p.m. than ~1:30 a.m. and hence the largest warming effect is observed at ~10:30 p.m. in summer. The spatial pattern and magnitude of this warming effect couple very well with the geographic distribution of wind turbines and such coupling is stronger at nighttime than daytime and in summer than winter. Together, these results suggest that the warming effect observed in MODIS over wind farms are very likely attributable to the development of wind farms. This inference is consistent with the increasing number of operational wind turbines with time during the study period, the diurnal and seasonal variations in the frequency of wind speed and direction distribution, and the changes in near-surface atmospheric boundary layer (ABL) conditions due to wind farm operations. The nocturnal ABL is typically stable and much thinner than the daytime ABL and hence the turbine enhanced vertical mixing produces a stronger nighttime effect. The stronger wind speed and the higher frequency of the wind speed within the optimal power generation range in summer than winter and at nighttime than daytime likely drives wind turbines to generate more electricity and turbulence and consequently results in the strongest warming effect at nighttime in summer. Similarly, the stronger wind speed and the higher frequency of optimal wind speed at ~10:30 p.m. than that at ~1:30 a.m. might help explain, to some extent, why the nighttime LST warming effect is slightly larger at ~10:30 p.m. than ~1:30 a.m. The nighttime warming effect seen in spring and fall are smaller than that in summer and can be explained similarly.     

利用MODIS/LST产品分析基准气候站环境代表性

该文基于2001—2007年地表温度遥感反演产品 (MOD11A2),以基准气候站对其周围不同大小窗口内地表温度距平序列的解释方差作为度量,评估了我国142个基准气候站的环境代表性,并将代表性与土地覆盖和高程状况进行相关研究。结果显示:以解释方差大于0.75作为区分是否具有代表性的阈值,约41%的站点代表性较好,代表区域范围可超过51×51 km2,多分布于北方地区;约21%的站点代表性较差,代表区域范围不足7×7 km2,多分布于南方地区;其他代表区域范围居中的站点在南、北方均有分布;站点周围的土地覆盖多样性和地形起伏度与站点代表性存在负相关,且相关性随窗口的增大而加强。文中还评估了基准气候站对所属气候区的代表性,发现在气候特征复杂的西南地区和新疆部分地区,站点对气候区的代表性较差。     

Modeling water and heat balance components for the river basin using remote sensing data on underlying surface characteristics

The method of the AVHRR-3 (NOAA) radiometer measurement data subject processing is produced for the retrieval of underlying surface temperature and several vegetation characteristics under cloud-free conditions. A technology for deriving the values of these parameters from the MODIS (EOS/Terra and Aqua) radiometer data is developed. The estimation of the temperature and vegetation characteristics is carried out for the Seim River basin (Kursk region) with the catchment area of 7460 km² for 2003–2005 vegetation seasons. Practical coincidence of estimations of AVHRR- and MODIS-derived temperatures, as well as the coincidence with ground observation results, is revealed. Statistics of these estimation errors is analyzed. Satellite-derived estimations of land surface temperature (LST) and vegetation characteristics are used for the calibration and verification of the developed model of the vertical heat and water transfer in the soil-vegetation-atmosphere system (SVAT). The model is intended for calculations of evapotranspiration, soil water and heat content, latent and sensible heat fluxes, and other water and heat balance components. The abilities to compute these parameters using the satellite estimations of the leaf area index and projective vegetation cover fraction as the model parameters and LST satellite estimations as the model input variable are investigated.