KEYWORDS: Independent component analysis, Signal to noise ratio, Factor analysis, Detection theory, Wavelets, Reflectivity, Signal processing, Foam, Data modeling, Iron
Based on spectral independence of different materials, independent component analysis (ICA), a blind source separation technique, can be applied to separate mixed hyperspectral signals. For the purpose of detecting objects on the sea and improving the precision of target recognition, an original ICA method is applied by analyzing the influence exerted by spectral features of different materials and mixture materials on spectral unmixing results. Due to the complexity of targets on the sea, several measured spectra of different materials have been mixed with water spectra to simulate mixed spectra for mixture spectra decomposition. Synthetic mixed spectra are generated by linear combinations of different materials and water spectra to obtain separated results. We then compared the separated results with the measured spectra of each endmember by coefficient of determination. We conclude that these factors that will change the original spectral characteristics of Gaussian distribution have significant influence on the separated results and selecting a proper initial matrix, and processing spectral data with lower noise can help improve the ICA method for more accurate separated results from hyperspectral data.
Leaf area index (LAI) estimation in a mixed grassland ecosystem is limited by temporal and spatial variations controlled by land surface heterogeneity and ecological parameters. Therefore, simply estimated LAI usually has difficulty in meeting the requirements of the land surface–atmosphere interaction models. We estimated LAI based on the relationship between LAI and normalized difference vegetation index (NDVI) by considering temporal and spatial variations. Spatial variations of both LAI and NDVI were investigated using the Morlet wavelet approach. Based on the ground reflectance data, LAI estimation can be greatly improved by taking temporal and spatial variations into account. The coefficient of determination (r 2 ) values of the LAI-NDVI equations were increased by 0.28, 0.51, and 0.44 in the early, maximum, and late growing seasons, respectively. LAI estimation from SPOT 4/5 and Landsat TM 5 images confirmed the applicability of the proposed estimation approach.
Habitat loss has become one major cause of prairie loggerhead shrike population decline, which is associated with some important grassland biophysical features. However, our understanding of what and how biophysical variables can spatially characterize shrike habitats is poor. The purpose of this study is to investigate the suitability of two vegetation indices (VIs) for spatially characterizing shrike habitats in North American mixed prairies. Our research, conducted in Grasslands National Park of Canada, is based on the normalized difference vegetation index (NDVI) and the adjusted transformed soil-adjusted vegetation index (ATSAVI) as derived from both in situ measurements and SPOT imagery for three types of nesting categories at three spatial scales. Our results demonstrated that shrikes in mixed North American prairies prefer sparsely vegetated areas with a leaf area index less than 2.01 and shrub cover of around 25%. Our results also demonstrated that ATSAVI is superior to NDVI in estimating vegetation abundance and structure. Loggerhead shrikes seems to prefer habitats characterized by NDVI ranging from 0.562 to 0.616 and ATSAVI ranging from 0.319 to 0.372 with the spatial scale varying from 100 to 20 m. ATSAVI also had better performance in detecting the spatial variation of shrike habitats due to its higher sensitivity to background information.
Widespread disturbance has brought a large amount of narrow-linear and small-area disturbance features (e.g., trails, seismic lines, forest roads, well sites, and cut blocks) to forest areas throughout the past decade. This issue has prompted research into finding the appropriate data and methods for mapping these narrow-linear and small-area disturbance features in order to examine their impacts on wildlife habitat. In this paper, we first described the characteristics of small forest disturbances and presented the nature of problem. We then presented a framework for detecting and extracting narrow-linear and small-area forest disturbance features. Using a SPOT 5 high spatial detail image and existing GIS databases, we applied the framework to map narrow-linear and small-area forest disturbance features in a Bear Management area (BMA) in the eastern slopes of the Rocky Mountains in Alberta, Canada. The results indicated that the proposed framework produced accurate disturbance maps for cut blocks, and forest roads & trails. The high errors of omission in the cut lines map were attributed to inconsistent geometric and radiometric patterns in the 'rarely-used' or 'old' cut lines. The study confirmed the feasibility of rapidly updating incomplete GIS data with linear and small-area disturbance features extracted from high spatial detail SPOT imagery. Future work will be directed towards improvement of the framework and the extraction strategy to remove a large amount of spurious features and to increase accuracy for cut lines mapping.
An action plan for recovering species at risk (SAR) depends on an understanding of the plant community distribution, vegetation structure, quality of the food source and the impact of environmental factors such as climate change at large scale and disturbance at small scale, as these are fundamental factors for SAR habitat. Therefore, it is essential to advance our knowledge of understanding the SAR habitat distribution, habitat quality and dynamics, as well as developing an effective tool for measuring and monitoring SAR habitat changes. Using the advantages of non-destructive, low cost, and high efficient land surface vegetation biophysical parameter characterization, remote sensing is a potential tool for helping SAR recovery action. The main objective of this paper is to assess the most suitable techniques for using hyperspectral remote sensing to quantify grassland biophysical characteristics. The challenge of applying remote sensing in semi-arid and arid regions exists simply due to the lower biomass vegetation and high soil exposure. In conservation grasslands, this problem is enhanced because of the presence of senescent vegetation. Results from this study demonstrated that hyperspectral remote sensing could be the solution for semi-arid grassland remote sensing applications. Narrow band raw data and derived spectral vegetation indices showed stronger relationships with biophysical variables compared to the simulated broad band vegetation indices.
Consecutive droughts that have occurred in the Canadian prairies have resulted in significant economic losses, ecological degradation, and environmental deterioration. The purpose of this study was to investigate the efficiency of remotely sensed data on drought assessment combined with climate data. The study area was the Canadian prairie ecozone in the provinces of Alberta, Saskatchewan, and Manitoba. There objectives were five-fold: 1) comparing Kriging and inverse distance weighting (IDW) interpolation methods, 2) comparing four spectral variables, the normalized difference vegetation index (NDVI), the enhanced vegetation index (EVI), the red and the mid infrared (MIR), 3) comparing three moisture indices (P-PET, P/PET and (P-PET/PET), 4) evaluating the relationships between spectral variables and moisture indices, and 5) assessing drought effects on different ecoregions. Results showed that there is no significant difference between Kriging and IDW, the two interpolation methods. MODIS vegetation indices could effectively assess drought conditions, especially EVI. Among the moisture indices compared, P-PET showed a better result. The impacts of droughts vary from year to year and from ecoregion to ecoregion. Aspen Parkland has higher drought resistance because of tree components.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.