Satellites viewing our world: The NASA landsat and the NOAA SMS/GOES

Thousands of individuals throughout the world are now users of satellite data. Hundreds of satellites have been launched—military, navigation, communications, educational, weather, and earth resources. One of the weather satellites (the SMS/GOES) and the NASA earth resources mapping satellite (Landsat) are the subjects of this article. Data from these systems have been highly cost beneficial, not only in the United States, but in developed and developing nations all over the globe as well. There is an increasing demand both for data and for training in data use.Data samples are shown and applications are discussed. Strong reference is made to the value of the digital computer in natural and man-made features mapping and monitoring. Procedures for acquiring NASA data are explained so that the reader may order data for his home region, or for other regions throughout the world which are of interest for their agriculture, forestry, hydrology, marine resources, geology, or land use. The cost of data is incredibly low; some products cost as little as three dollars.Also discussed are the remotely based data-collection platforms that acquire ground or water data daily and relay results to the NASA Landsat or to the NOAA SMS/GOES.     

基于随机森林算法与农户视角的村落级农业面源污染风险分析

当前农业面源污染问题依然严峻,村落级尺度下各地块的污染风险状态精细识别有待进一步研究.本文针对三峡库区典型村落重庆市涪陵区南沱镇睦和村进行研究,结合无人机多光谱技术、农户行为调研、随机森林算法等进行地物识别及泛地块网格划分,通过测算总氮(total nitrogen,TN)、总磷(total phosphorus,TP...     

基于近地多光谱和OLI影像的黄河三角洲冬小麦种植区盐分估算及遥感反演——以山东省垦利县和无棣县为例

黄河三角洲是我国重要的后备土地资源区,而土壤盐渍化对该区的农业生产影响巨大,因此,及时获取该区农作物种植区土壤盐分含量及其分布具有重要意义。论文利用ADC便携式多光谱相机和EC110便携式盐分计,采集该区近地多光谱相片和土壤表层含盐量数据,结合两期遥感影像提取黄河三角洲冬小麦种植区面积分布,构建基于近地多光谱植被指数的土壤含盐量估算模型,进而将模型拟合反演到黄河三角洲OLI影像,得到黄河三角洲冬小麦种植区土壤盐分含量空间分布图,对麦区土壤盐分状况进行了分析。结果显示,土壤含盐量最佳估算模型为以SAVI为因变量的线性模型（Y=-0.754x+0.870,n=80）,估测R²为0.747,精度达到81.44%;研究区冬小麦分布由西南部内陆至东北部沿海呈明显递减的空间趋势;麦田土壤盐分含量主要集中在1.5~3.0 g/kg之间,占种植总面积的76.90%,而含盐大于3.0 g/kg的麦田占14.09%,宜采取针对性栽培管理措施。该研究探索了基于近地多光谱数据和OLI影像的土壤含盐量估算方法,为黄河三角洲麦田管理和土壤盐分含量估算提供了快速有效的技术手段。     

基于微型无人机遥感数据的城市水环境信息提取初探

近年来,黑臭水体整治已经成为各级政府的重要工作内容。为更好地开展黑臭水体筛查和监管工作,提出基于多旋翼无人机搭载多光谱相机的遥感监测方案,尝试采用影像特征变换结合面向对象分类方法提取黑臭水体的遥感信息。结果表明:(1)无人机影像上地物具有更加丰富的色彩和纹理;(2)与主成分变换和纹理分析方法相比,光谱指数对典型地物具有较好的判别能力;(3)采用标准最邻近分类法提取城市水环境信息,能够获得较高分类精度。     

香港近海海域叶绿素a定量反演及时空变化分析

以2009~2019年HJ-1A/B卫星多光谱数据和对应日期的实测数据为数据源,通过预处理提取出各波段组合反射率并与实测叶绿素a浓度数据进行统计相关性分析,选取相关性最高的波段组合作为特征变量与2/3的实测叶绿素a浓度数据进行建模,并用剩下的1/3实测叶绿素a浓度数据进行精度验证以确定最佳遥感反演模型,最后根据最佳反演模型对2009-2019年的香港近海海域叶绿素a浓度进行反演,明晰该海域近10年的叶绿素a浓度时空变化特征.结果表明：利用HJ-1A/B卫星多光谱数据反演香港近海海域叶绿素a浓度的最佳波段组合为第3波段和第2波段比值（B3/B2）,相关系数（r）为0.893;最佳反演模型为利用B3/B2构建的e指数回归模型（Chl=0.004e^6.693(B3/B2)）,决定系数（R²）为0.934,均方根误差（RMSE）为0.255μg/L,平均相对误差（RPD）为25%;近10年香港近海海域的叶绿素a浓度时空变化特征：空间上整体呈现“东高西低,由东向西逐渐减小”的分布特征,西部海域比东部海域平均浓度低5μg/L左右;2017年内呈“春低秋高,夏升冬降”的随季节变化特点,其中秋季最高,夏春两季次之,冬季最低.     

基于无人机多光谱影像的小微水域水质要素反演

总磷(TP)、悬浮物浓度(SS)、浊度(TUB)3种水质参数可以直接通过遥感反演得到,常用于评价区域水环境的污染状况.以浙江农林大学东湖为研究对像,使用无人机携带多光谱传感器(Mica Sense Red Edge)获取多光谱影像,进而提取16个光谱参数,分别构建东湖水域TP、SS、TUB的反演模型.结果表明:光谱参数V5(NIR 0.770～0.890μm)与TP、SS相关性显著(r分别为0.470、-0.537,p0.05),V4(0.670～0.760μm)与TUB相关性显著(r=0.486,p0.05).在建立的TP反演模型中,指数函数模型精度最高,决定系数R~2为0.7829;在建立的SS、TUB反演模型中,多项式函数模型精度最高,决定系数R~2分别为0.7503、0.7334.经检验,TP、SS、TUB模型估测值与实测值线性拟合曲线的决定系数R~2分别为0.7374、0.8978、0.6726,满足水质要素反演的精度要求.最后利用建立的模型,结合多光谱影像数据,建立了东湖水域各参数的空间分布图,实现了水质参数的可视化,可为小微水域的污染防治提供技术支撑.     

HYDROLOGICAL APPLICATIONS OF LANDSAT IMAGERY USED IN THE STUDY OF THE 1973 INDUS RIVER FLOOD,PAKISTAN1

ABSTRACT: During August and September 1973, the Indus River Valley of Pakistan experienced one of the largest floods on record, resulting in damages to homes, businesses, public works, and crops amounting to millions of rupees. Tremendous areas of lowlands were inundated along the Indus River and major tributaries. Landsat data made it possible to easily measure the extent of flooding, totaling about 20,000 km² within an area of about 400,000 km² south from the Punjab to the Arabian Sea. The Indus River data were used to continue experimentation in the development of rapid, accurate, and inexpensive optical techniques of flood mapping by satellite begun in 1973 for the Mississipi River floods. The research work on the Indus River not resulted in the development of more effective procedures for optical processing of flood data and synoptically depicting flooding, but also provided potentially valuable ancillary information concerning the hydrology of much of the Indus River Basin.     

基于无人机多光谱影像和OPT-MPP算法的水质参数反演

无人机多光谱遥感可用于监测多个水质参数,如悬浮物、浊度、总磷和叶绿素等,建立稳定和准确的水质参数反演模型是开展这一工作的前提.matching pixel-by-pixel(MPP)算法是一种针对无人机影像高分辨率特点的反演算法,但其存在运算量过大和过拟合的问题,基于此,提出optimize-MPP(OPT-MPP)算法,以克服运算量过大和过拟合的问题.本研究以浙江省杭州市青山湖作为研究区域,采集45个样本,分别构建悬浮物浓度(SS)与浊度(TU)的OPT-MPP算法反演模型.结果表明,最佳悬浮物反演模型的决定系数R²达到0.787 0,综合误差为0.130 8;最佳浊度反演模型的决定系数R²达到0.804 3,综合误差为0.150 3.最后利用分别建立的两个参数的最优模型,实现青山湖各实验区域的水质参数空间分布信息的反演.     

WATER RESOURCES INVESTIGATIONS IN WEST CENTRAL IRAN USING LANDSAT DATA1

ABSTRACT: Extensive investigations have been undertaken to determine the utility of Landsat data for detecting and analyzing hydrologic characteristics of an interior watershed of Iran that drains to Daryachehye-Namak (salt lake). This interior playa serves as the terminus for surface water discharging to it from the Karaj, Shur, Saveh, and Ghom Rivers and ground water from their underlying aquifers. These drainage systems encompass heavily populated and industrial sectors of west central Iran, including Tehran. The result of this investigation demonstrates the applicability of Landsat data for mapping and monitoring water regimen as an aid in interpreting hydrologic conditions throughout this arid region. Fluctuation of water area in a playa lake, occupying the lowest part of this closed basin, was monitored on repetitive Landsat coverage. As the result of field investigations combined with optical and digital analyses of the Landsat data, fluctuating water depths were determined in order to estimate the volume of water present in this lake during various seasons. A comparison between stream discharge rates and the estimated volume of standing water make it possible to quantitatively evaluate the hydrologic regimen and to detect the significance of ground water discharge.     

SELECTING RECONNAISSANCE STRATEGIES FOR FLOODPLAIN SURVEYS1

ABSTRACT: Multispectral aircraft and satellite data over the West Branch of the Susquehanna River were analyzed to evaluate potential contributions of remote sensing to floodplain surveys. Multispectral digital classifications of land cover features indicative of floodplain areas were used by interpreters to locate various floodprone area boundaries. The boundaries thus obtained were found to be more striking and continuous in the Landsat data than in the low altitude aircraft data. The digital approach permitted satellite results to be displayed at 1:24,000 scale and aircraft results at even larger scales. Results indicate that remote sensing techniques can delineate floodprone areas more easily in agricultural and limited development areas than in areas covered by a heavy forest canopy. At this time it appears that the remote sensing data would be best used as a form of preliminary planning information or as an internal check on previous or ongoing floodplain studies. In addition, the remote sensing techniques can assist in effectively monitoring floodplain activities after a community enters into the National Flood Insurance Program.