eChapters

ABSTRACT

Leaf N and chlorophyll (Chl) concentration of crop are important indicators of plant nitrogen status. Laboratory determination of plant tissue N is time consuming and costly. Also remote sensing is quickly becoming recognized as a valuable tool that has the potential to asses a wide variety of physiological properties over a large area in a short amount of time. This study examined the relationship of sorghum and bajra canopy reflectance spectral response to its nitrogen and chlorophyll content. The study consisted of two nitrogen level (N₀ : no nitrogen and N₁: recommended dose of nitrogen for each crop) for two crops (sorghum and bajra) . Chlorophyll and N content were measured throughout the growing season for both crops. Reflectance at increased significantly as N fertilizer rate decreased. Chl and N were strongly related to canopy reflectance for either individual wavelength or for spectral vegetative indices or ratios. Sorghum nitrogen fertilizer rate mainly affected leaf reflectance at 575 and 623 nm, while bajra at 565 and 657nm. The reflectance values at 581 and 603 nm had the highest correlation with sorghum leaf N concentration. Similarly, bajra leaf N concentration was highly correlated with 619 and 696 nm. Sorghum and bajra leaf N concentrations were highly correlated with reflectance ratio of R₅₈₁/ R₃₈₁ (r²= 0.44*) and R₆₁₉/R₄₆₂ (r²= 0.78***) respectively with greatest r² value. Similarly, Sorghum and bajra leaf Chl concentrations were highly correlated with R677/R527 (r²= 0.63**) andR688/R409 (r²= 0.71***) or R615/R411 (r2=0.71***),

respectively. Thus our results suggest that spectral reflectance measurements hold promise for the assessment of some physiological parameter at the leaf level real time monitoring of sorghum and bajra N status and N fertilizer management.

ABSTRACT

Hyperspectral data are generally noisier compared to broadband multispectral data. Therefore, it is desirable to smoothen the reflectance spectra. Smoothing needs to preserve an absorption feature while simultaneously removing noise. It is very essential to optimize the smoothing filter to minimize disturbances to the statistical properties of the original data. This study was carried out to see the effect of Fast-Fourier Transform (FFT) based smoothing technique on the statistical properties of vegetation spectral response and to see the effect of the filter size. The data used for the present study is the reflectance spectra data obtained from space borne EO-1 onboard Hyperion sensor. Filter sizes were varied between 3 and 15 with the increment of 2, i.e. 3, 5, 7..........15. Paired t-test was carried out between the original and the smoothed data (for the selective bands) for all the filter sizes in order to see the extent of distortion with changing filter sizes. Our study revealed that the distortion in the spectra increases with the increase in the filter size. Beyond filter size 11, the number of locations within the spectra where the smooth spectra were statistically different from its original counterpart increased to 14 and reaches 21 at the filter size 15.

ABSTRACT

Satellite-based hyperspectral imaging became areality in November 2000 with the successful launch and operation of the Hyperion system on board the EO-1 platform. Hyperion is a push broom imager with 220 spectral bands in the 400-2500 nm wavelength range, a 30 meter pixel size and a 7.5 km swath. Prelaunch characterization of Hyperion measured low signal to noise (SNR<40:1) for the geologically significant shortwave infrared (SWIR) wavelength region (2000- 2500 nm). The impact of this low SNR on Hyperion's capacity to resolve spectral detail is studied for the Musiri and Tattayyangarpettai regions of part of Namakkal and Tiruchirapalli districts of Tamil Nadu state, India which comprises a diverse range of minerals with narrow, diagnostic absorption bands in the SWIR. Following radiative transfer correction of the Hyperion radiance at sensor data to surface radiance (apparent reflectance), diagnostic spectral signatures were clearly apparent including: green vegetation; talc; dolomite; crystalline limestone, chlorite; white mica and sericite. Though the derived surface composition maps generated from these image endmembers were noisy (both random and column), the spatially coherent and correlated well with the known geology. The interpreted and processed mineral map is given below. In addition, the Hyperion data were used to measure and map spectral shifts of <10 nm in the SWIR related to white mica chemical variations.

ABSTRACT

Greenseeker is a handheld computer mounted sensor for observing normalized difference vegetation Index [NDVI= (NIR “R)/(NIR+R)], which may be used to estimate site specific amount of fertilizer nitrogen depending up on the crop condition. The sensor can be used for yield estimation and developing remote sensing based growth models. It is an active sensor with a built in light source for illuminating the target and records reflected light from vegetation at wavelength of 670 nm (red region) and 780 nm (near infrared region) of electromagnetic spectrum. The swath covered by the Sensor is 0.75 m when sensor is placed about 1.0 m above crop canopy.

The sensor was used to record NDVI in wheat- sugarcane intercropping system wherein wheat was grown on raised beds of 1.04 m width with six rows of wheat and sugarcane in furrows at row spacing of 1.38 m. The wheat variety PBW 343 was sown with nitrogen levels @ 0, 30, 60, 90, 120, 150, 180 and 210 kg N ha^-1 (all basal) and sugarcane variety CoSe 92423 was planted in furrows on 7^th November, 2006 in a medium textured loam soil.

The NDVI was recorded in wheat at 17, 41, 59, 73, 88, 113 and 129 DAS using Greenseeker and Spectroradiometer (UNISPEC ^TM of PP system, USA) simulating normal satellite bands i.e.

NIR: 770-860 nm, Red: 620-680 nm, Green: 520-580 nm. The maximum NDVI (recorded at 88 DAS) was used to develop INSEY “ yield relationship in wheat. The NDVI recorded by Greenseeker was integrated with NDVI recorded by Spectroradiometer.

The Greenseeker was also used to monitor crop growth in sugarcane. The NDVI was correlated with shoot population, plant height, LAI and total biomass during formative and crop elongation phase of sugarcane till end of November 2007. Highly significant (at 1%) correlation was observed between GS- NDVI and shoot population till mid August and LAI in sugarcane for entire crop cycle. An exponential model (R² = 0.9017) was developed to estimate LAI of sugarcane from NDVI. The NDVI was significantly (at 5 %) was correlated with plant height till mid August. However, the correlation between total crop biomass in sugarcane and NDVI remained non-significant.

ABSTRACT

In the present paper, study of spectral reflectance measurements has been carried out for different type of snow and other ambient objects in snowbound areas of Indian Himalaya using Spectroradiometer (350-2500nm). The observations suggest that there are significant changes in the spectral radiances and reflectance values of snow and other ambient objects, which can be discriminated using satellite multi-spectral bands, however for characterizing the snow using satellite sensor, spectral bands having maximum radiances values should be greater than snow radiances and suitable spectral bandwidth are to be identified. The detail analysis of field collected reflectance suggest that, due to aging and grain size variation in snow, the maximum variations in reflectance was observed in the NIR region i.e. around 1040-1050nm. For contamination and snow depth, the maximum variation was observed in the visible region i.e. around 470nm and 590nm respectively. For the moisture changes, the maximum variations observed around 980 and 1160nm. The study also suggests, normalized difference snow index (NDSI) increases with the increase of aging, grain size and moisture content. This study can be helpful in selection of suitable spectral bands for snow cover applications using satellite

remote sensing and developing the algorithms for seasonal snow cover characteristics using multi-spectral and hyper-spectral data.

ABSTRACT

The Mumbai region of Maharashtra has been witnessing very rapid urbanization development in terms of industrial, commercial and residential establishments. This has resulted in reclamation of tidal flats and estuarine regions. The ‘marine clay', characteristic of this environment has high swelling potential and consequently affects the performance of the structures. This will have a serious negative impact on the urbanization expansion. In this study, an attempt is made to map the spatial distribution of these expansive clays using the state of art Hyperion data. Reflectance spectra of representative samples collected from the field is used to map marine clays using Spectral Angle Mapper (SAM) technique. Laboratory analyses of these foundation soils (reflectance spectroscopy, mineralogy and Atterberg limits) indicate that they have significant amount of montmorrilonite and has good swelling potential (A_C>1.4). The derived expansive clay abundance map will be of use while planning appropriate remedial measure and future development.

Key Words : Swelling clay, Spectral characters, Geotechnical properties

ABSTRACT

A study was conducted to compare inter-sensor variability by computing vegetation indices, NDVI & SR. The two sensors compared were Hyperion and ALI (Advanced Land Imager) present on EO-1 satellite. Hyperion is a hyperspectral sensor having a spectral range of 400-2500nm with 10nm bandwidth. ALI is a multispectral sensor with 10 bands of 100nm bandwidth. Hyperion and ALI data were acquired at the same time for two different seasons, dry (April) and wet (October) seasons over Shoolpaneshwar Wildlife Sanctuary, Gujarat, India. Vegetation indices such as SR “ Simple Ratio and NDVI -Normalized Differential Vegetation Index were performed. Seasonal variations were observed in both the data sets. For SR index, Hyperion performed well using October month data. In comparison with ALI, Hyperion showed high NDVI for both deciduous and evergreen species using October month data. The highest NDVI value is assumed to represent the maximum vegetation “greenness” during that period. Results clearly indicated the superiority of Hyperion in vegetation analysis using NDVI & SR indices.

Keywords : Hyperion data (EO-1), ALI (Advance Land Imager),Shoolpaneshwar Wildlife Sanctuary, NDVI, Simple Ratio (SR)

ABSTRACT

Spectral signatures are the specific combination of reflected and absorbed electromagnetic radiation at varying wavelengths which can uniquely identify an object. The “Spectral signature” of an object is a function of 1) incidental EM wavelength and material interaction with that section of the electromagnetic spectrum. The measurements can be made with various instruments, including a task specific spectrometer, although the most common method is separation of the Red, Green, Blue and Near Infrared portion of the EM spectrum as acquired by digital cameras. The spectral signatures serve as diagnostic tools in understanding the composition and structure of the materials. The present paper highlights the usefulness of the spectral signatures in rock and mineral discrimination and mineral exploration and stresses the need for creation of a spectral data base in India for various applications.

1. INTRODUCTION

Objects at the Earth's surface emit and reflect many wavelengths of radiation. In principle, each object has a unique spectral signature, which could be used for identification much like a finger print. Remotely sensed spectral signatures could be utilized for recognizing and mapping for all kinds of features. This can be done to a limited extent with conventional multispectral scanners that operate in a few broad bands of visible and infrared energy. However, broad spectral bands often do not distinguish between similar features. To overcome this limitation, a more sophisticated approach toward spectral signatures has arisen during the past few years. This approach is called image spectroscopy or hyperspectral analysis, which is based on scanning hundreds of closely spaced and very narrow spectral bands. With this technique, it is possible to create continuous spectral response curves, which may be used to identify many objects positively.

ABSTRACT

Mineral composition and their distribution on planetary surfaces provides important clues to the evolutionary processes that the planetary bodies have undergone since their formation. For distant and inaccessible planets, remote sensing, either from Earth-based telescopes or from an orbiter around the planetary body, using reflectance spectroscopy over a wide range of the electromagnetic spectrum (UV-Visible-IR), can provide surface distribution of minerals. With advances in technology, a high-resolution hyperspectral imager onboard an orbiter has developed into a powerful tool for compositional mapping of planetary surfaces. Interpretation of the signals from these distant and relatively unknown destinations is a challenging task due to scarcity of available ground truth. Here, we describe the technique with its applicability to airless bodies, such as Moon, in the context of the Indian Moon mission, Chandrayaan-1, and, the intricacies involved in the data analyses.

Keywords : Reflectance spectroscopy, hyperspectral data, remote sensing, planets, moon

1. INTRODUCTION

The primary goal of planetary exploration is to understand the mechanism by which planetary objects (planets, dwarf planets, satellites and small solar system bodies) have evolved with time. The information gathered from the study of individual bodies is integrated to comprehend the development of our solar system. The present understanding envisages our solar system to have formed from the solar nebula, produced as a consequence of the collapse of a molecular cloud. While the central mass gravitated and formed the sun, the gas and dust surrounding it condensed to form grains and formed the proto-planetary disk. These grains, through a series of processes, finally aggregated to form of the various planetary bodies. Existing Astronomical, chemical and mineralogical evidences from studies of lunar samples, meteorites and, remote sensing observations suggest that the planetary bodies are heterogeneous and possess extreme compositional and morphological diversity. The internal energy of a planetary body, a function of its size, is responsible for the differences in the degree of differentiation a body has undergone leading to the variations in their surface composition. Large planets like the Earth and Venus are still evolving whereas the small asteroids have barely evolved after their formation and are considered primitive. The wide variety of planetary bodies in our solar system is therefore reminiscent of the various stages of evolution that our solar system went through. The degree of melting and cycling through episodes of differentiation determine the mineral inventory on a planetary body. Therefore, distribution and composition of various minerals can be used to understand various stages of chemical evolution of a planetary body. Mineral mapping of planetary surfaces is thus useful in understanding the petrogenetic history.

ABSTRACT

This paper reviews the various application of hyperspectral remote sensing data carried out by the scientists of India emphasizing its advantages.

1. INTRODUCTION

The most significant recent breakthrough in remote sensing has been the development of hyperspectral sensors. The ‘Hyper’ in hyperspectral means ‘too many’ and refers to the large number of measured wavelength bands. Hyperspectral images are spectrally over determined, which means that they provide ample spectral information to identify and distinguish spectrally unique materials. Hyperspectral imagery provides the potential for more accurate and detailed information, extraction than is possible with any other type of conventional remotely sensed data.

Imaging spectroscopy (Goetz et al., 1985) in hyperspectral imaging is concerned with the measurement, analysis, and interpretation of spectra acquired from a given scene (or specific object) at a short, medium or long distance by an airborne or satellite sensor. The concept of imaging spectroscopy originated in the 1980's, when A. F. H. Goetz and his colleagues at NASA's Jet Propulsion Laboratory began a revolution in remote sensing by developing new instruments such as the Airborne Imaging Spectrometer (AIS), then called AVIRIS, for Airborne Visible Infra-Red Imaging Spectrometer (Green et al., 1998). This system is now able to cover the wavelength region from 0.4 to 2.5 μm using more than two hundred spectral channels, at nominal spectral resolution of 10 nm. The special characteristics of hyperspectral datasets pose different processing problems, which must be necessarily tackled under specific mathematical formalisms, such as classification and segmentation (Jia et al., 1999, Green, R 1998) and spectral mixture analysis (Adams et al., 1986; Smith et al., 1990a, b).

ABSTRACT

Over the last decade, hyperspectral remote sensing has become a powerful analysis tool for applications in environment and ecology, aquaculture, forestry, agriculture, and geoscience. A hyperspectral imagers advantage over broadband sensors, such as Landsat Thematic Mapper (TM) and the SPOT Haute Resolution Visible (HRV) sensor series, is its ability to detect molecular absorption and particle scattering signatures of constituents. The finer spectral resolution of a hyperspectral imager allows detection of surface materials and their abundances, as well as inferences of biological and chemical processes. This review highlights the vast extent to which hyperspectral technology and their application with special reference to ocean have been discussed.

ABSTRACT

Remote sensing has undergone tremendous development in past decades. Hyperspectral imaging refers to the imaging of a scene over a large number of discrete, narrow contiguous spectral bands in the visible through thermal infrared portion of the spectrum. Thus it is possible to construct a complete reflectance spectrum of every pixel in the scene.

Classification of hyperspectral image requires thorough understanding of surface material properties, sensor characteristics and atmospheric absorption spectra. The reflectance spectra of most materials on the Earth's surface contain diagnostic absorption features. Remote sensors, capable of acquiring complete reflectance spectra over large areas, offer a powerful tool to study of the Earth and the environment. Methods are being developed to extract information from remote sensing data using image-processing approaches. Specific absorption features of reflectance spectrum can be used to identify a number of important rock forming minerals, and have been used by geologists for geologic mapping and studies of volcanoes.

The present discussion includes an application of thermal characteristics of near- IR bands of Landsat TM to estimate the temperature of over and around the crater of Barren Island volcano in 1991. A method using dual band temperature characteristics related to radiant source of sub-pixel size was developed and utilized to prepare temperature contour map of the Island using unsaturated pixels of band 5 and 7.

ABSTRACT

Multispectral imagery has been used as the data source for natural resource mapping and land observational remote sensing from airborne and satellite systems since the early 1960s. Over the past two decades, advances in sensor technology have made it possible for the collection of several hundred spectral bands. This is commonly refe rred to as hyperspectral imagery. Thus it is possible to construct a complete reflectance spectrum of every pixel in the scene. This review highlights the hyperspectral remote sensing system development, hyper spectral data processing and difference between multispectral and hyperspectral data; spatial and spectral resolutions and focuses on the application of hyperspectral imagery in natural resource studies and, in particular the classification and mapping of land resources.