Extraction of impervious features from spectral indices using artificial neural network

An urban area comprises a complex mix of diverse land cover types and materials. Urban ecology and environment is significantly influenced by the proportion of impervious cover that is increasing considerably with time due to the continuous influx of people into urban areas. Therefore, it is of vital importance to determine the spatiotemporal pattern and magnitude of urbanization. In the present study, we have employed a supervised backpropagation neural network in order to extract the impervious features using five spectral indices, such as one vegetation index—Soil-Adjusted Vegetation Index (SAVI), one water index—Modified Normalized Water Index (MNDWI), and three urban indices—Normalized Difference Built-up Index (NDBI), Built-up Index (BUI), and Index-Based Built-up Index (IBI). The study has been performed using Landsat Thematic Mapper data of November, 2011, of the rapidly urbanizing city of Ranchi, capital of Jharkhand state, India. Using different combinations of these spectral indices while keeping SAVI and MNDWI constant, seven composite images were built, and from each of these composites, impervious features were classified and its accuracy assessed with reference to high-resolution images provided by Microsoft Bing Imagery and adequate ground truthing. It was observed that along with SAVI and MNDWI, whenever IBI was used in any combination, it decreased the classification efficiency. On the other hand, NDBI and BUI, individually or when used together, discriminated the impervious features from the others with high accuracy with the combination of SAVI, MNDWI, and BUI achieving the highest accuracy of 90.14 %.     

Monitoring rangeland ground cover vegetation using multitemporal MODIS data

The aim of the present research is to monitor changes in herbage production during the grazing season in the Semirom and Brojen regions, Iran, using multitemporal Moderate Resolution Imaging Spectroradiometer (MODIS) data. At first, various preprocessing steps were applied to a topography map. The atmospheric and topographic corrections were applied using subtraction of the dark object method and the Lambert method. Image processing, including false-color composite, principal component analysis, and vegetation indices were employed to produce land use and pasture production maps. Vegetation sampling was carried out over a period of 4 months during June–September 2008, using a stratified random sampling method. Twenty random sampling points were selected, and herbage production was estimated and verified with the double-checking method. Four MODIS data sets were used in this study. The models for image processing and integrating ground data with satellite images were processed, and the resulting images were categorized into seven classes. Finally, the land covers were verified for accuracy. A postclassification analysis was carried out to verify the seven class change detections. The results confirmed that Normalized Difference Vegetation Index (NDVI) and Soil-Adjusted Vegetation Index (SAVI) maps had a close relationship with the field data. The indices produced with shortwave infrared bands had a close relationship with field data where the ground cover and yields were high. The R ² value observed was 0.85. The changes in the pasture vegetation were high during the growing season in more than 90 % of the pastures. During the growing season, most changes in the pastures belonged to class 5 and 2 in the NDVI and SAVI index maps, respectively.     

Investigating remote sensing indices to monitor drought impacts on a local scale (case study: Fars province,Iran)

As drought occurs in different climates, assessment of drought impacts on parameters such as vegetation cover is of utmost importance. Satellite remote sensing images with various spectral and spatial resolutions represent information about different land covers such as vegetation cover. Hence, the purpose of this study was to investigate the performance of satellite vegetation indices to monitor the agricultural drought on a local scale. In this regard, satellite images including Moderate Resolution Imaging Spectroradiometer (MODIS) and Advanced Very High Resolution Radiometer (AVHRR) data were used to evaluate vegetation cover and their gradual changes effects on agricultural drought. Fars province in Iran with relatively low precipitation values was selected as the study area. Modified Perpendicular Drought Index (MPDI), MPDI1, Vegetation Condition Index (VCI), Normalized Difference Vegetation Index Anomalies (NDVIA), and Standardized Vegetation Index (SVI), were evaluated to select the remote sensing based index with the best performance in drought monitoring. The performance of such indices were investigated during 13 years (2000–2013) for MODIS and 29 years (1985–2013) for AVHRR. To assess the efficiency of the satellite indices in drought investigation, Standardized Precipitation Index (SPI) data of five selected stations were used for 3, 6, and 9 month periods on August. The results showed that NDVI-based vegetation indices had the highest correlation with SPI in cold climate and long-term timescale (6 and 9 month). The highest correlation values between remote sensing based indices and SPI were acquired, respectively, in 9-month and 6-month time-scales, with the values of 43.5% and 40%. Moreover, VCI showed the highest capability for agricultural drought investigating in different climate regions of the study area. Overall, the results proved that NDVI-based indices can be used for drought monitoring and assessment in a long-term timescale on a local time-scale.

     

Forest cover change prediction using hybrid methodology of geoinformatics and Markov chain model: A case study on sub-Himalayan town Gangtok,India

In the Himalayan states of India, with increasing population and activities, large areas of forested land are being converted into other land-use features. There is a definite cause and effect relationship between changing practice for development and changes in land use. So, an estimation of land use dynamics and a futuristic trend pattern is essential. A combination of geospatial and statistical techniques were applied to assess the present and future land use/land cover scenario of Gangtok, the subHimalayan capital of Sikkim. Multi-temporal satellite imageries of the Landsat series were used to map the changes in land use of Gangtok from 1990 to 2010. Only three major land use classes (built-up area and bare land, step cultivated area, and forest) were considered as the most dynamic land use practices of Gangtok. The conventional supervised classification, and spectral indices-based thresholding using NDVI (Normalized Difference Vegetation Index) and SAVI (Soil Adjusted Vegetation Index) were applied along with the accuracy assessments. Markov modelling was applied for prediction of land use/land cover change and was validated. SAVI provides the most accurate estimate, i.e., the difference between predicted and actual data is minimal. Finally, a combination of Markov modelling and SAVI was used to predict the probable land-use scenario in Gangtok in 2020 AD, which indicted that more forest areas will be converted for step cultivation by the year 2020.     

基于HJ-CCD影像数据的新疆喀纳斯自然保护区植被叶面积指数估算

以新疆喀纳斯自然保护区为研究区, 评价了HJ-CCD影像数据估算植被叶面积指数(LAI)的能力及其对大气订正方法的敏感性.分别利用6S和FLAASH两种大气订正模型对HJ1B-CCD2影像进行大气订正, 比较了大气订正前后不同植被(针叶林、阔叶林、针阔混交林和草地)反射率及5种植被指数(NDVI、SR、SAVI、MSR、ARVI)的变化, 进而建立了4种植被类型LAI的遥感估算模型, 分析了LAI的空间分布格局.结果表明: 大气订正后可见光波段的反射率降低, 6S模型订正后近红外波段的反射率上升, 而FLAASH模型订正后近红外波段的反射率下降.大气订正后NDVI、SR、SAVI(除针叶林)和MSR上升, 6S模型订正后所有植被类型的ARVI下降, FLAASH模型订正后针叶林和阔叶林的ARVI上升, 而针阔混交林和草地的ARVI下降.大气订正提高了植被指数与LAI之间的相关性, 对于针叶林、阔叶林、针阔混交林而言, 利用6S模型订正后的反射率建立的模型优于FLAASH模型订正后的反射率建立的模型, 而草地却相反.经过大气订正, HJ-CCD影像数据可应用于研究区植被LAI的估算.研究区LAI的高值集中在湖泊和河流附近, 低值分布在海拔较高处.山地森林草原带、亚高山森林带、高山灌丛草甸带、高山冻原、高山冰川带植被LAI的平均值分别为2.6、3.9、2.5、1.7和1.0.     

水平和垂直尺度乔、灌、草覆盖度遥感提取研究进展

植被覆盖及其变化是区域生态系统环境变化的重要指示,而植被覆盖度是植物群落覆盖地表状况的一个综合量化指标,是生态模型、碳循环、水循环模型等的重要特征参量。传统的植被覆盖度是指一定尺度下所有植被(乔、灌、草)覆盖的综合反映值,当考虑植被垂直方向的异质性,垂直尺度的乔、灌、草覆盖度提取为定量化准确衡量生态环境、全球气候变化等领域提供更具有生态意义的植被参量。目前,遥感大面积估算水平尺度乔、灌、草覆盖度已有比较成熟和可靠的算法,主要方法有：植被指数法、回归分析法、分类决策树法、神经网络法、像元分解模型法、物理模型反演法等,其估算精度基本能达到应用要求。植被垂直方向的异质性给垂直尺度乔、灌、草覆盖度遥感提取带来较大挑战,垂直尺度上的乔、灌、草覆盖度遥感提取的研究在欧美等国已经有了一定规模的开展,在国内则处于起步阶段。遥感提取垂直尺度乔、灌、草覆盖度的主要手段有：激光雷达（LIDAR）、多角度遥感以及两层结构冠层反射模型反演。综述了水平尺度和垂直尺度上乔、灌、草覆盖度遥感提取的最新进展,比较和分析主要的遥感提取方法、模型和现存的一些问题,并对未来的研究发展趋势进行了展望。     

1998—2007年新疆植被覆盖变化及驱动因素分析

利用1998-2007年SPOT VGT归一化植被指数(NDVI)数据对新疆植被覆盖的年际和空间变化进行了动态监测,并从气候变化和人类活动双重角度分析了植被覆盖演变的原因.1998-2007年新疆植被覆盖变化经历了2个阶段:1998-2001年植被覆盖严重退化时期;2002-2007年植被覆盖由急剧上升到缓慢下降再到持续升高时期,NDVI明显高于20世纪末期水平.新疆植被覆盖变化存在显著的空间差异,阿尔泰山地森林、巴音布鲁克草原等自然植被NDVI明显退化,农业灌溉区和生态建设地区的植被覆盖明显提高.从不同的土地利用类型来看,沙地和耕地的NDVI上升趋势显著,林地和草地植被的NDVI退化严重.研究表明,新疆植被覆盖变化是气候变化和人类活动共同作用的结果.温度对植被覆盖变化的影响表现为对植被生长年内韵律的控制和春季植被生长期的延长,年降水量的波动式下降是导致新疆植被覆盖变化呈现2个阶段的主导冈素.农业生产水平的提高是新疆农业灌溉区NDVI不断上升的重要原因,同时,近年来大规模实施的生态建设工程所带来的生态效应正在呈现.     

Estimating leaf chlorophyll contents by combining multiple spectral indices with an artificial neural network

Estimating leaf chlorophyll contents through leaf reflectance spectra is efficient and nondestructive, but the actual dataset always based on a single or a few kinds of specific species, has a limitation and instability for a common use. To address this problem, a combination of multiple spectral indices and a model simulated dataset are proposed in this paper. Six spectral indices are selected, including Blue Green Index (BGI), Photochemical Reflectance Index (PRI_5), Triangle Vegetation Index (TVI), Chlorophyll Absorption Ratio Index (CARI), Carotenoid Reflectance Index (CRI) and the green peak reflectance (R₅₂₅). Both stepwise linear regression (SLR) and back-propagation artificial neural network (ANN) are used to combine the six spectral indices for the estimation of chlorophyll content (Cab). In addition, to overcome the limitation of actual dataset, a “big data” is applied by a within-leaf radiation transfer model (PROSPECT) to generate a large number of simulated samples with varying biochemical and biophysical parameters. 30% of the simulated dataset (SIM30) and an experimental dataset are used for validation. Compared with linear regression method, NN yields better result with R² = 0.96 and RMSE = 5.80ug.cm^?2 for Cab if validated by SIM30, while R² = 0.95 and RMSE = 6.39ug.cm^?2 for SLR. NN also gives satisfactory result with R² = 0.80 and RMSE = 5.93ug.cm^?2 for Cab if validated by LOPEX dataset, however, the SLR only gets 0.72 of R² and 12.20ug.cm^?2 of RMSE. The results indicate that integrating multiple spectral indices can improve the Cab estimating accuracy with a better stability in different kind of species and the model simulated dataset can make up the shortfall of actual measured dataset.     

Prairie or planted? Using time-series NDVI to determine grassland characteristics in Montana

Temperate grasslands are a highly threatened global biome. Complicating management and conservation strategy development, modern grasslands can be difficult to characterize across landscapes since they range from native and semi-native to completely non-native species compositions such as those found in heavily managed pastures. Similar to methods used to differentiate C3 and C4 grasses, we investigate the ability of using temporal variations in growth characteristics as an alternative pathway to predicting native versus introduced species composition across grassland landscapes. To do this, we conducted an exploratory analysis using a time-series of Normalized Difference Vegetation Index values as a measure of vegetation greenness with Landsat 5 TM imagery across a growing season and performed an unsupervised classification. Results from the classification were compared with field observation to determine if we can differentiate between native and introduced grassland types in the Northwest Glaciated Plains subecoregion of northeastern Montana. Our results indicated that we predicted grassland cover with 81% accuracy within our 200 km² study area and 71% accuracy in our 5000 km² secondary study area. Further extrapolation of our methodology, combined with the refinement of vegetation indices of time-series imagery, classification algorithms and the availability of data from planned Landsat and Sentinel missions, may provide the spatial detail necessary to improve grassland monitoring and rangeland management over large areas.     

An analysis of vegetation indices in relation to tree species diversity using by satellite data in the northern forests of Iran

Vegetation indices have been introduced for analyzing and assessing the status of quantitative and qualitative characteristics of vegetation using satellite images. However, choosing the best indices to be used in forest biodiversity and vegetation is one of the important problems faced by the users. The purpose of this research is to evaluate six vegetation indices in the analysis of tree species diversity in the northern forests of Iran. The present research uses LISS III sensor data from IRS-P6 satellite. Geometric rectification of images was performed using ground control points, and Chavez model was used for atmospheric correction of the data. The six spectral vegetation indices included NDVI, IPVI, Ashburn Vegetation Index (AVI), TVI, TTVI, and RVI. Shannon–Wiener species diversity index was used to analyze diversity, and the value of the index was calculated in each sample plot. Then, the spectral values of each sample plot were extracted from different bands. The best subset regression was used to analyze the relationship between species diversity and the related bands. The results obtained from the regression showed that polynomial equations under scrutiny as independent variables can assess tree and shrub species diversity better than other bands and compounds used (R ²?=?0.47). The obtained results also indicated a higher capacity in the case of the AVI index for estimating tree species diversity in the under study area.     

植被指数研究进展

在遥感应用领域,植被指数已广泛用来定性和定量评价植被覆盖及其生长活力。由于植被光谱表现为植被、土壤亮度、环境影响、阴影、土壤颜色和湿度复杂混合反应,而且受大气空间—时相变化的影响,因此植被指数没有一个普遍的值,其研究经常表明不同的结果。20多年来,已研究发展了40多个植被指数。该文对已有的大部分植被指数进行了归纳分类,评价其各自优势和局限性,并探讨了未来研究的方向,这将有助于遥感在农业、植被和生态环境监测方面进行有效地开发与应用。     

草地植被盖度的多尺度遥感与实地测量方法综述

植被盖度作为一个重要的生态学参数被用在许多气候模型和生态模型中。地表实测和遥感测量是获取植被盖度的两种基本途径。以草地植被盖度的测量为研究对象,综合讨论了目前地表实测和遥感测量常用的方法,分析了它们的优缺点,并对如何提高草地植被盖度的测量精度做出展望。数码相机、高光谱遥感以及多尺度遥感数据的综合使用可能是未来草地植被盖度测量发展的趋势。     

The relationship between land surface temperature and land use/land cover in Guangzhou,China

The integration of remote sensing, geographic information system, landscape ecology and statistical analysis methods was applied to study the urban thermal environment in Guangzhou. Normalized Difference Vegetation Index (NDVI), Normalized Difference Build-up Index (NDBI), Normalized Difference Barren Index (NDBaI) and Modified Normalized Difference Water Index (MNDWI) were used to analyze the relationships between land surface temperature (LST) and land use/land cover (LULC) qualitatively. The result revealed that, most urban built-up lands were located in the middle part, and high LST areas mostly and were in the middle and southern parts. Therefore, the urbanization and thermal environment in the middle and southern parts need to be determined. Land surface temperature increased with the density of urban built-up and barren land, but decreased with vegetation cover. The relationship between MNDWI and LST was found to be negative, which implied that pure water would decrease the surface temperature and the polluted water would increase the surface temperature. A multiple regression between LST and each indices as well as the elevation was created to elevate the urban thermal environment, which showed that NDVI, NDBI, NDBaI, MNDWI were effective indicators for quantifying LULC impacts on LST.     

A new approach for developing comprehensive agricultural drought index using satellite-derived biophysical parameters and factor analysis method

The accurate assessment of drought and its monitoring is highly depending on the selection of appropriate indices. Despite the availability of countless drought indices, due to variability in environmental properties, a single universally drought index has not been presented yet. In this study, a new approach for developing comprehensive agricultural drought index from satellite-derived biophysical parameters is presented. Therefore, the potential of satellite-derived biophysical parameters for improved understanding of the water status of pistachio (Pistachio vera L.) crop grown in a semiarid area is evaluated. Exploratory factor analysis with principal component extraction method is performed to select the most influential parameters from seven biophysical parameters including surface temperature (T _s), surface albedo (α), leaf area index (LAI), soil heat flux (G _o), soil-adjusted vegetation index (SAVI), normalized difference vegetation index (NDVI), and net radiation (R _n). T _s and G _o were found as the most effective parameters by this method. However, T _s, LAI, α, and SAVI that accounts for 99.6 % of the total variance of seven inputs were selected to model a new biophysical water stress index (BPWSI). The values of BPWSI were stretched independently and compared with the range of actual evapotranspiration estimated through well-known METRIC (mapping evapotranspiration at high resolution with internal calibration) energy balance model. The results showed that BPWSI can be efficiently used for the prediction of the pistachio water status (RMSE of 0.52, 0.31, and 0.48 mm/day on three image dates of April 28, July 17, and August 2, 2010). The study confirmed that crop water status is accounted by several satellite-based biophysical parameters rather than single parameter.     

Assessing the spatiotemporal dynamics of vegetation cover as an indicator of desertification in Egypt using multi-temporal MODIS satellite images

Desertification is the major environmental threat in the arid and semiarid regions. The soil-adjusted vegetation index (SAVI) was used as an indicator to monitor the desertification change in Egypt. A multi-temporal satellite data of moderate-resolution imaging spectroradiometer were used to estimate SAVI and land surface temperature. Also, Global Multi-resolution Terrain Elevation Data 2010 and climatic data were used for the analysis. This research focuses on assessing the trend of the vegetation cover change in the seasons of January, March, June, September, and December for the years 2002, 2005, 2008, and 2011. The magnitude of the vegetation cover change in periods 2002–2005, 2005–2008, and 2008–2011 at ≤100 and >100 m elevation was analyzed. A major increase in the vegetation cover that occurred in the period 2002–2005 was about 3,400 km², as a result of two national megaprojects (Toshka Project and El-Salam Canal). In contrast, vegetation cover decreased by 5,500 km² in March during the period 2005–2008, coinciding with the period when the management of the megaprojects failed. Vegetation cover changed again by 1,500 km² in the period of 2008–2011, and the vegetated areas in the Nile Delta were affected by the sea level rising which was responsible for the soil salinization. Three sites were chosen in this investigation (Kom Ombo, El-Oweinat, and Nile Delta) in order to observe the difference of desertification dynamics and to understand the relationship between the vegetation cover distribution and other environmental variables. Anti-desertification policies and advanced agricultural management are highly required in Egypt to decrease any environmental crises and food shortage.     

两种用于作物冠层叶绿素含量提取的改进光谱指数

在深入探讨目前广泛使用的提取叶绿素含量的植被指数的光谱响应机制基础上,利用PROSPECT+SAIL模型模拟的作物冠层反射率样本数据对比分析了这些植被指数对叶绿素含量变化的敏感性差异,包括PSSRa、PSSRb、PSNDa、PSNDb、NPCI、PRI、MCARI和TVI等.结果表明,上述植被指数或对土壤背景变化敏感,或受高值LAI影响趋于饱和,对作物叶绿素含量反演效果均不理想.提出了4种基于TVI和MCARI的改进植被指数MTVI1、MTVI2、MCARI1和MCARI2,揭示了它们对土壤背景和LAI不敏感,对叶绿素含量变化更为敏感的光谱机制,并根据实验数据对其进行验证.实验表明,改进的植被指数MTVI2和MCARI2是作物冠层叶绿素含量较好的预测器,可据此建立作物冠层叶绿素含量反演模型.     

Environmental change detection in the central part of Iraq using remote sensing data and GIS

This study aims to assess the potential of several ancillary input data for the improvement of unsupervised land cover change detection in arid environments. The study area is located in Central Iraq where desertification has been observed. We develop a new scheme based on known robust indices. We employ Landsat (multispectral scanner, thematic mapper, and enhanced thematic mapper) satellite data acquired in 1976, 1990, and 2002. We use the Normalized Deferential Vegetation Index, Normalized Differential Water Index (NDWI), Salinity Index (SI), and Eolian Mapping Index. Two new equations were applied for the SI and the NDWI indices. Validation was performed using ground truth data collected in 16 days. We show that such an approach allows a robust and low-cost alternative for preliminary and large-scale assessments. This study shows that desertification has increased in the study area since 1990.     

Effects of the Herbicide Diuron on Cordgrass (<Emphasis Type="Italic">Spartina foliosa</Emphasis>) Reflectance and Photosynthetic Parameters

Early indicators of salt marsh plant stress are needed to detect stress before it is manifested as changes in biomass and coverage. We explored a variety of leaf-level spectral reflectance and fluorescence variables as indicators of stress in response to the herbicide diuron. Diuron, a Photosystem II inhibitor, is heavily used in areas adjacent to estuaries, but its ecological effects are just beginning to be recognized. In a greenhouse experiment, we exposed Spartina foliosa, the native cordgrass in California salt marshes, to two levels of diuron. After plant exposure to diuron for 28 days, all spectral reflectance indices and virtually all fluorescence parameters indicated reduced pigment and photosynthetic function, verified as reduced CO₂ assimilation. Diuron exposure was not evident, however, in plant morphometry, indicating that reflectance and fluorescence were effective indicators of sub-lethal diuron exposure. Several indices (spectral reflectance index ARI and fluorescence parameters EQY, Fo, and maximum rETR) were sensitive to diuron concentration. In field trials, most of the indices as well as biomass, % cover, and canopy height varied predictably and significantly across a pesticide gradient. In the field, ARI and Fo regressed most significantly and strongly with pesticide levels. The responses of ARI and Fo in both the laboratory and the field make these indices promising as sensitive, rapid, non-destructive indicators of responses of S. foliosa to herbicides in the field. These techniques are employed in remote sensing and could potentially provide a link between landscapes of stressed vegetation and the causative stressor(s), which is crucial for effective regulation of pollution.     

人类活动与气候变化对科尔沁沙质草地植被的影响

1992—2006年在科尔沁沙地开展了草地放牧和封育试验,分析研究了人类放牧活动和气候变化对草地植被的影响。研究结果表明：①人类放牧活动对沙质草地植被具有显著影响,其中轻度放牧可使原退化草地植被盖度、高度、物种丰富度和多样性明显提高,中度放牧下虽然草地植被盖度和高度有所下降,但对物种丰富度和多样性无不良影响,持续过度放牧可以导致草地植被迅速破坏;②围栏封育可以促进退化草地植被盖度、高度、物种丰富度和植物多样性得到较快恢复,其恢复速度是草层高度＞植被盖度＞物种丰富度＞多样性;③暖湿气候有利于草地维持较高的植被盖度、高度、物种丰富度和多样性,而持续干旱会导致相应指标的明显下降,多雨时期气温变化对植被的影响较大,干旱时期降水变化对植被的作用较强。     

Semi-automatic shallow landslide detection by the integration of airborne imagery and laser scanning data

Landslide mapping is essential for effective watershed management. In Taiwan, a typhoon or earthquake event can trigger hundreds, even thousands, of shallow landslides in mountainous watersheds. Thus, improving the efficiency of landslide mapping by means of remote sensing techniques is an important issue. This study proposes a new method that uses concurrent aerial laser scanning (ALS) data and color ortho-imagery as input data: the topographic indices of slope, surface roughness, and object height model can be derived from the ALS data and the Green–Red Vegetation Index from the ortho-images. The method first uses these topographic and spectral indices in a global, semi-automatic algorithm to separate landslide from non-landslide pixels. It then offers a region growing tool and a 3D Eraser/Painter to edit detected landslides locally. These global and local operations are designed with a user interface, which is intuitive and user-friendly. Results from four test sites in a mountainous watershed prove that the method is easy, accurate, and suitable for landslide mapping in Taiwan.