Spectral Mixture Analysis Approach Research Articles

Characterizing Flood Impact on Swiss Floodplains Using Interannual Time Series of Satellite Imagery

Pressure on the biodiversity of ecosystems along many rivers is growing continuously due to the increasing number of hydropower facilities regulating downstream flow and sediment regimes. Despite a thorough understanding of the short-term processes and interactions at this hydro-biosphere interface, long-term analyses of the impacts on floodplain dynamics are lacking. We used interannual Landsat 4, 5, 7, and 8 time series to analyze the effects of hydrological events on floodplain vegetation in four mountainous floodplains in the Swiss Alps. Using a spectral mixture analysis approach, we demonstrate that the floodplain vegetation dynamics of mountainous rivers can be recovered at a spatial resolution of 30 m. Our results suggest that interactions between floods and floodplain vegetation are complex and not exclusively related to flood magnitude. Of the four reaches analyzed, only data gathered along the submountainous reach with a quasi-natural flow regime show a clear link between remotely sensed vegetation indices and floods. In addition, our 29-year time series shows a continuous upward trend in vegetation indices along the floodplains, strongest in the reaches affected by hydropower facilities. The approach presented in this study can be easily replicated in other mountain ranges by providing available flow data to verify the impact of hydropower on floodplain vegetation dynamics.

Mapping impervious surface fractions using automated Fisher transformed unmixing

Spatial explicit monitoring of impervious surface fractions provides essential information for urban planning, urban disaster prevention and mitigation. Moderate spectral and spatial resolution satellite observation has become a tool to perform such monitoring across different spatial and temporal scales. Yet, one major challenge in urban remote sensing is the high spatial heterogeneity causing the presence of image pixels constituting a mixture of different land cover materials. Here we apply an automated unmixing procedure specifically designed to optimally account for the pronounced spectral variability (i.e. high within-class and low between-class variability) prevailing in urban settings. The approach is based on Fisher Discriminant Analysis (FDA), a data dimensionality reduction technique in which between-class variability is maximized and within-class variability is minimized while transforming spectra from the reflectance to the Fisher feature space. We integrated the FDA transformation in the widely-applied Multiple Endmember Spectral Mixture Analysis (MESMA) approach in an attempt to more effectively address the endmember variability problem. We made use of online spectral libraries to train the Fisher transformation parameters and steer the automated selection of image endmembers. Our Fisher transformed MESMA algorithm (F-MESMA) was tested on medium spatial resolution Landsat scenes and impervious fraction maps were generated for five major cities located in different regions of the world. The output of F-MESMA was compared to the output of state-of-the-art spectral mixture analysis approaches which were specifically designed to reduce endmember variability issues. We could demonstrate that compared to other spectral transformation approaches, the ratio of within- vs between-class variability of the endmembers was most strongly reduced after applying the Fisher transformation. As a direct consequence, across all five cities F-MESMA consistently provided the most accurate impervious surface fraction estimates (RMSEF-MESMA = 0.13 vs. RMSEalternative approaches = [0.16–0.17]).

Detailed urban surface characterization using spectra from enhanced spatial resolution Sentinel-2 imagery and a hierarchical multiple endmember spectral mixture analysis approach

Detailed mapping of urban surfaces is one of the most challenging tasks in remote sensing due to the three-dimensional structure of cities, spatial diversity, and material spectral variability. Satellite urban applications demand better spatial, spectral, and temporal resolution, although there are strict technical constraints among them. Therefore, the development of sophisticated methods that exploit both high spectral and spatial data sources becomes necessary. A hierarchical multiple endmember spectral mixture analysis (MESMA) approach is developed and applied on Sentinel-2 imagery for the detailed quantification of the urban land cover, taking advantage of Worldview-2 high spatial resolution. The case study is the urban and peri-urban area of Heraklion, Greece. The area to point regression kriging (ATPRK) method is applied to downscale Sentinel-2 bands from 10 and 20 m to 2 m (WorldView-2 spatial resolution) and create a spectral library (SL) of urban materials, which contain 180 separate spectra. The urban SL is then used in the developed hierarchical MESMA approach to estimate the abundances of 11 urban land cover classes based on the original Sentinel-2 image. The estimated land cover fractions validate against a very high-resolution (1 m) land cover map of the area. It is proved that the complexity of the urban land cover can be efficiently investigated by the proposed methodology. Error analysis shows good accuracy of the results in all estimated class fractions. Moreover, the good validation results lead to the conclusion that ATPRK fusion algorithm between Sentinel-2 and WorldView-2 bands produced reliable urban material spectra, capable for advanced spectral analysis on Sentinel-2 imagery. The developed methodology is easily transferable to other cities since it is based exclusively on earth observation data and is suitable for multiple urban applications related to urban climate, urban sprawl, and urban regeneration.

Assessing drought-induced change in a piñon-juniper woodland with Landsat: a multiple endmember spectral mixture analysis approach

ABSTRACTPiñon-juniper communities exist on mid-elevation mountain ranges throughout the Southwestern USA. These species are drought adapted, and have lived with climatic stochasticity since the end of the Pleistocene. Rising temperatures and drought within the past two decades have stressed much of this community beyond its adaptive limits. With increased drought-induced stress, piñon show greater vulnerability to die-off than juniper. Widespread piñon die-off occurred from extreme drought in northern New Mexico (2002–2004) with minimal juniper die-off. This study assessed the feasibility of quantifying differential piñon and juniper mortality following a recent drought at a site in central New Mexico by performing multiple endmember spectral mixture analysis on six Landsat images from 2009 through 2015 using field-based spectra collected throughout 2015. An ideal spectral separability time of year was identified to maximize separation between constituent land-cover classes by calculating vegetation indices for the five dominant land-cover classes at the site (juniper, piñon, dead piñon, herbaceous, and soil). Ideal separability periods were also determined by evaluating precipitation and temperature data. Peak separabililty between land-cover classes within the study period was determined to occur during the pre-monsoon season between late spring/early summer (May) when minimal spectral overlap occurred between classes (σ = 1). The field-based spectra were then used to unmix each image in the study period. Results indicate a 24.6% decline of piñon across the study period with a comparable 23.8% increase of dead piñon. Accuracy assessment using high spatial (5–8 cm) resolution imagery for 2014 and 2015 showed strong correlation with modelled fractional cover results for 2014: live piñon and juniper (R2 = 0.72), and dead piñon (R2 = 0.79), and 2015: live piñon and juniper (R2 = 0.77), dead piñon (R2 = 0.65). Results demonstrate the potential of MESMA to monitor and accurately quantify the differential die-off of piñon and juniper at a regional scale as climate change-induced drought and higher temperatures are projected to continue in the Southwest.

Detection of Drought-Induced Hickory Disturbances in Western Lin An County, China, Using Multitemporal Landsat Imagery

Hickory plantations play an important role in improving local farmers’ economic conditions, but extreme drought in July–August 2013 seriously influenced hickory nut production. It is necessary to understand the extent and magnitude of this drought-induced hickory disturbance through mapping its spatial distribution using remote sensing data. This paper proposes a new approach to examine hickory disturbance based on multitemporal Landsat imagery. Ratios of green vegetation to soil fractions were calculated, in which the green vegetation and soil fractions were extracted from Landsat multispectral imagery using the linear spectral mixture analysis approach. We used the differences between before-drought and after-drought ratios to detect hickory disturbances. Four disturbance levels—non-disturbance, light, medium, and severe—were grouped according to the field survey data. The spatial distribution of these four levels was developed using the ratio-based approach. The result indicates that this approach is effective to detect drought-induced hickory disturbance and may be transferred to detect other kinds of disturbances, such as forest disease and selective logging. Cautions should be taken to properly select image acquisition dates and the change detection period, in addition to the approach itself.

Capability of Spaceborne Hyperspectral EnMAP Mission for Mapping Fractional Cover for Soil Erosion Modeling

Soil erosion can be linked to relative fractional cover of photosynthetic-active vegetation (PV), non-photosynthetic-active vegetation (NPV) and bare soil (BS), which can be integrated into erosion models as the cover-management C-factor. This study investigates the capability of EnMAP imagery to map fractional cover in a region near San Jose, Costa Rica, characterized by spatially extensive coffee plantations and grazing in a mountainous terrain. Simulated EnMAP imagery is based on airborne hyperspectral HyMap data. Fractional cover estimates are derived in an automated fashion by extracting image endmembers to be used with a Multiple End-member Spectral Mixture Analysis approach. The C-factor is calculated based on the fractional cover estimates determined independently for EnMAP and HyMap. Results demonstrate that with EnMAP imagery it is possible to extract quality endmember classes with important spectral features related to PV, NPV and soil, and be able to estimate relative cover fractions. This spectral information is critical to separate BS and NPV which greatly can impact the C-factor derivation. From a regional perspective, we can use EnMAP to provide good fractional cover estimates that can be integrated into soil erosion modeling.

Assessment of sediment connectivity from vegetation cover and topography using remotely sensed data in a dryland catchment in the Spanish Pyrenees

Many Mediterranean drylands are characterized by strong erosion in headwater catchments, where connectivity processes play an important role in the redistribution of water and sediments. Sediment connectivity describes the ease with which sediment can move through a catchment. The spatial and temporal characterization of connectivity patterns in a catchment enables the estimation of sediment contribution and transfer paths. Apart from topography, vegetation cover is one of the main factors driving sediment connectivity. This is particularly true for the patchy vegetation cover typical of many dryland environments. Several connectivity measures have been developed in the last few years. At the same time, advances in remote sensing have enabled an improved catchment-wide estimation of ground cover at the subpixel level using hyperspectral imagery. The objective of this study was to assess the sediment connectivity for two adjacent subcatchments (~70 km2) of the Isabena River in the Spanish Pyrenees in contrasting seasons using a quantitative connectivity index based on fractional vegetation cover and topography data. The fractional cover of green vegetation, non-photosynthetic vegetation, bare soil and rock were derived by applying a multiple endmember spectral mixture analysis approach to the hyperspectral image data. Sediment connectivity was mapped using the index of connectivity, in which the effect of land cover on runoff and sediment fluxes is expressed by a spatially distributed weighting factor. In this study, the cover and management factor (C factor) of the Revised Universal Soil Loss Equation (RUSLE) was used as a weighting factor. Bi-temporal C factor maps were derived by linking the spatially explicit fractional ground cover and vegetation height obtained from the airborne data to the variables of the RUSLE subfactors. The resulting connectivity maps show that areas behave very differently with regard to connectivity, depending on the land cover and on the spatial distribution of vegetation abundances and topographic barriers. Most parts of the catchment show higher connectivity values in August as compared to April. The two subcatchments show a slightly different connectivity behaviour that reflects the different land cover proportions and their spatial configuration. The connectivity estimation can support a better understanding of processes controlling the redistribution of water and sediments from the hillslopes to the channel network at a scale appropriate for land management. It allows hot spot areas of erosion to be identified and the effects of erosion control measures, as well as different land management scenarios, to be studied.

A comparative analysis of megacity expansions in China and the U.S.: Patterns, rates and driving forces

Research on physical characteristics and land-cover dynamic changes of megacities over time provides valuable insights for effectively regulating urban planning and management. This study conducts a comparative analysis of 30-year urban expansion patterns and rates among three metropolises in China (Beijing, Shanghai, and Guangzhou) and another three in the USA (New York, Los Angeles, and Chicago) based on time-series impervious surface area (ISA) data extracted from multitemporal Landsat images using the linear spectral mixture analysis approach. This research indicates significantly different urbanization patterns and rates between the Chinese and American megacities. The ISA expansion area in Chinese megacities was five times higher than that in American megacities during the past three decades. The Chinese megacities expand outward from the urban core to the periphery in a concentric ring structure, whereas the American megacities increase ISA mainly within the inner cities with patch-filling patterns. The Chinese megacities are in the development stage where population and economic conditions significantly influence urban expansion patterns and rates, but the American megacities are in the developed stage where population and economic conditions are not important forces driving the ISA expansion. The ISA intensity in the American megacities decreases constantly and smoothly, but ISA intensity in Chinese megacities decays abruptly within certain distances, depending on different cities and years. The most obvious urban expansions were between 8 and 20km in Beijing in the 1980s, between 14 and 50km in Shanghai in the 2000s, and between 8 and 18km in Guangzhou in the 1990s.

The use of single-date MODIS imagery for estimating large-scale urban impervious surface fraction with spectral mixture analysis and machine learning techniques

Urban impervious surface information is essential for urban and environmental applications at the regional/national scales. As a popular image processing technique, spectral mixture analysis (SMA) has rarely been applied to coarse-resolution imagery due to the difficulty of deriving endmember spectra using traditional endmember selection methods, particularly within heterogeneous urban environments. To address this problem, we derived endmember signatures through a least squares solution (LSS) technique with known abundances of sample pixels, and integrated these endmember signatures into SMA for mapping large-scale impervious surface fraction. In addition, with the same sample set, we carried out objective comparative analyses among SMA (i.e. fully constrained and unconstrained SMA) and machine learning (i.e. Cubist regression tree and Random Forests) techniques. Analysis of results suggests three major conclusions. First, with the extrapolated endmember spectra from stratified random training samples, the SMA approaches performed relatively well, as indicated by small MAE values. Second, Random Forests yields more reliable results than Cubist regression tree, and its accuracy is improved with increased sample sizes. Finally, comparative analyses suggest a tentative guide for selecting an optimal approach for large-scale fractional imperviousness estimation: unconstrained SMA might be a favorable option with a small number of samples, while Random Forests might be preferred if a large number of samples are available.

A spatially adaptive spectral mixture analysis for mapping subpixel urban impervious surface distribution

Spectral mixture analysis (SMA) has been widely employed in analyzing urban environments, especially for estimating urban impervious surface distribution at the subpixel level. When implementing SMA, endmember selection is considered an important step, and an inappropriate endmember set could severely affect the accuracy of fractional land covers. Although different degrees of success have been achieved in existing SMA approaches, the paradox of endmember selection is still unsolved: theoretically “purest” endmembers that can be selected with relative ease do not always yield optimal results, while the selection of the most “representative” endmembers is very difficult with simple SMA. To address this problem, we propose a spatially adaptive SMA (SASMA) technique to automatically extract and synthesize the “most representative” endmembers for SMA through considering both between-class and within-class variations. In particular, we developed a classification tree method to automatically extract endmember candidates through incorporating spectral and spatial information. In addition, to mitigate the effect of within-class variation, we employed synthetic spectra of neighboring endmember candidates in a local search window as the most “representative” endmember signature. Results indicate that SASMA performs well in estimating subpixel impervious surface distribution with relatively high precision (mean absolute error of 8.50%, root mean square error of 15.25%, and R2 of 0.701) and small bias (systematic error of −0.93%).

Evaluating spectral indices and spectral mixture analysis for assessing fire severity, combustion completeness and carbon emissions

We used a Landsat Thematic Mapper (TM) image from the 2011 Wallow fire in Arizona, USA, in combination with field data to assess different methods for determining fire severity. These include the normalised burn ratio (NBR), the differenced NBR (dNBR), the relative dNBR (RdNBR) and the burned fraction (BF) estimated by spectral mixture analysis (SMA). The Geo Composite Burn Index (GeoCBI) and vegetation mortality data were used as ground truth. Of all the remotely sensed measures evaluated the dNBR had the best performance (GeoCBI–dNBR R2=0.84), which supports the operational use of the dNBR for post-fire management. Of the other remotely sensed measures, the SMA-derived BF also had moderately high correlations with the GeoCBI (R2=0.66). Both approaches demonstrated their usefulness for refining modelled CC values, however, the SMA approach has the advantage of providing transferable quantitative estimates without the need for calibration with field data. The carbon emission estimates that included fire severity were more than 50% lower than the estimate derived from modelling alone. These results suggest that for certain fire types, especially mixed-severity fires, current emission estimates are significantly overestimated, which will affect global carbon emission estimates from wildfires.

A comparison of methods for estimating fractional vegetation cover in arid regions

We compared a set of methods for estimating the fractional vegetation cover (fc) of sparse desert vegetation over an arid region of southern Xinjiang, China. Six kinds of remote sensing inversion models (an NDVI regression, a spectral mixture analysis (SMA), a pixel dichotomy model, a three-band maximal gradient difference (TGDVI) model and two modified TGDVI models) were used to derive fc from remote sensing data, and the results were compared with fc values measured in the field to select an appropriate model to derive the fractional cover of sparse desert vegetation in arid regions. The NDVI regression based on field fc and the NDVI for the sampled pixels in September 2006 showed the highest precision, while the results of 2007 showed that the NDVI regression method is inappropriate for depicting vegetation characteristics in other growing season because the empirical model highly depend on the specified in situ measurement. The SMA approaches yielded higher precision than the other models, indicating that it is applicable for analysing the coverage of sparse desert vegetation. The pixel dichotomy model can yield a high precision based on finely detailed vegetation maps. However, it requires the measurement of many parameters. The TGDVI model is simple and easy to implement, and the values that it predicted for the coverage of high-density vegetation and barren areas were close to those measured in the field, but the fc values of sparsely vegetated areas were underestimated. The predictions of the modified TGDVI models were close to the values measured in the field, indicating that these modified models can reliably and effectively extract information on the fractional cover of sparse vegetation in an arid region. We analyzed the models’ sensitivity with respect to rainfall because the short-wavelength infrared bands used in the two TGDVI models proposed in this study are sensitive to moisture. The results showed that the modified TGDVI models’ accuracy was not affected by increasing soil moisture content caused by rain. However, the NDVI regression, SMA and TGDVI were sensitive to the change of soil moisture content. Moreover, the two modified TGDVI models yielded negative values for water sources, such as reservoirs and rivers, implying that they are effective for characterising water bodies. However, the modified TGDVI models cannot predict fc in snow- and glacier-covered regions, producing abnormally high rather than zero values. Additionally, the predictions before and after snowfall on the top of a mountain show a linear increasing relationship, suggesting that the short-wavelength infrared band may be useful to predict snow depth.

Hyperspectral Satellite Data in Mapping Salt-Affected Soils Using Linear Spectral Unmixing Analysis

Development of salt-affected soils in the irrigated lands of arid and semi-arid region is major cause of land degradation. Hyperion hyperspectral remote sensing data (EO-1) was used in the present study for characterization and mapping of salt-affected soils in a part of irrigation command area of Indo-Gangetic alluvial plains. Linear spectral mixture analysis approach was used to map various categories of salt affected soils represented by spectral endmembers of slightly, moderately and highly salt-affected soils. These endmembers were related to surface expression of various categories of salt-affected soils in the area. The endmembers were selected by performing minimum noise fraction (MNF) transformation and pixel purity index (PPI) on Hyperion (EO-1) data with reference to high resolution LISS IV data and field data. The results showed that various severity classes of salt-affected soils could be reliably mapped using linear spectral unmixing analysis. A low RMSE value (0.0193) over the image was obtained that revealed a good fit of the model in identification and classification of endmembers of various severities of salt affected soils. The overall classification accuracies for slight, moderate and highly salt-affected soils were estimated of 78.57, 79.81 and 84.43% respectively.

An automated waveband selection technique for optimized hyperspectral mixture analysis

Linear spectral mixture analysis (SMA) has been used extensively in remote sensing studies to estimate the sub-pixel composition of spectral mixtures. The lack of ability to account for sufficient temporal and spatial variability between and among ground component or endmember spectra has been acknowledged as a major shortcoming of conventional SMA approaches. In an attempt to overcome this problem, a novel and automated linear spectral mixture protocol, referred to as stable zone unmixing (SZU), is presented and evaluated. Stable spectral features (i.e. least sensitive to spectral variability) are automatically selected for use in the mixture analysis based on a minimum InStability Index (ISI) criterion. ISI is defined as the ratio of the spectral variability within and the spectral variability among the endmember classes that are present within the mixture. The algorithm was tested on a set of scenarios, generated from in situ measured hyperspectral data. The scenarios covered both urban and natural environments under differing conditions. SZU provided reliable endmember cover distribution maps in all scenarios. On average, an absolute gain in R2—the coefficient of determination of the modelled versus the observed sub-pixel cover fractions—of 0.14 over the traditional SMA approaches was observed while the absolute gain in fraction abundance error was 0.06. It was concluded that the SZU protocol has potential to be an effective and efficient SMA algorithm for generating optimal cover fraction estimates regardless of the scenario considered. Moreover, the subset selection protocol, as implemented in SZU, can be regarded as complementary to conventional SMA approaches resulting in a further reduction of spectral variability.

Nonlinear Hyperspectral Mixture Analysis for tree cover estimates in orchards

Accurate monitoring of spatial and temporal variation in tree cover provides essential information for steering management practices in orchards. In this light, the present study investigates the potential of Hyperspectral Mixture Analysis. Specific focus lies on a thorough study of non-linear mixing effects caused by multiple photon scattering. In a series of experiments the importance of multiple scattering is demonstrated while a novel conceptual Nonlinear Spectral Mixture Analysis approach is presented and successfully tested on in situ measured mixed pixels in Citrus sinensis L. orchards. The rationale behind the approach is the redistribution of nonlinear fractions (i.e., virtual fractions) among the actual physical ground cover entities (e.g., tree, soil). These ‘virtual’ fractions, which account for the extent and nature of multiple photon scattering only have a physical meaning at the spectral level but cannot be interpreted as an actual physical part of the ground cover. Results illustrate that the effect of multiple scattering on Spectral Mixture Analysis is significant as the linear approach provides a mean relative root mean square error (RMSE) for tree cover fraction estimates of 27%. While traditional nonlinear approaches only slightly reduce this error (RMSE = 23%), important improvements are obtained for the novel Nonlinear Spectral Mixture Analysis approach (RMSE = 12%).

A weighted linear spectral mixture analysis approach to address endmember variability in agricultural production systems

The least squares error (LSE) technique is frequently used to estimate abundance fractions in linear spectral mixture analysis (LSMA). The LSE is typically equally weighted for all wavebands, assuming equally important effects. This is, however, not always the case and therefore traditional LSMA often results in suboptimal fraction estimates. This study presents a weighted LSMA approach that prioritises wavebands with minor or no negative effects on fraction estimates. Synthetic mixed pixel spectra compiled from in situ measured spectra of bare soil, citrus tree and weed canopies were used for validation. The results show markedly improved fraction estimates obtained for the weighted approach, with a mean absolute gain of 0.24 in R 2 and a mean absolute reduction in fraction abundance error of 0.06.

Hyperspectral detection of chemical vegetation stress: evaluation for the Canadian HERO satellite mission

The potential for direct detection of chemicals on vegetation and for detecting vegetation stress and damage from exposure was reviewed within the context of the proposed Hyperspectral Environment and Resource Observer (HERO) satellite mission. The direct detection of 10 selected toxic chemicals was deemed unlikely. Detecting vegetation stress resulting from exposure to these chemicals may be feasible, based on several image analysis approaches considered for two levels of damage: (i) severe defoliation, mortality, and effects at the full crop or forest stand scale; and (ii) more subtle stress and damage such as discolouration, spotting, bleaching, and light defoliation that occur at subpixel scales. Image classification, vegetation indices, spectral mixture analysis (SMA), and change detection are expected to be suitable for the former, whereas only SMA and advanced modelling approaches are expected to be suitable for the latter because of the subtle nature of reflectance change at subpixel scales. A key recommendation is the need for a spectral library of toxic substances and the stress and damage caused to various types of vegetation at different growth stages. Current information on this is limited and represents a constraint to effective analysis and evaluation. Based on this application, HERO mission parameters were considered, with recommendations that the current spectral resolution and ground sampling distance (GSD) be improved, with a sensitivity analysis required to determine proper values to provide optimal detection of toxic hazards and vegetation stress. Other sensor and mission parameters such as spectral range, signal-to-noise ratio (SNR), swath width, and orbital repeat period with off-nadir sensing and temporal relook frequency were deemed acceptable, but improvements would be desirable. Extensive testing involving laboratory experiments and airborne hyperspectral imagery for a variety of toxic substances over different areas and vegetation types is required for assessment using different image analysis algorithms to confirm this and other recommendations for HERO prelaunch.

Estimating and mapping crop residues cover on agricultural lands using hyperspectral and IKONOS data

Crop residues left on agricultural lands after harvest play an important role in controlling and protecting soil against water and wind erosion. One challenge of remote sensing is to differentiate crop residues from bare soil and crop cover, especially when the residues have been weathered and/or when the crop cover phenology is more advanced. Several techniques for mapping and estimating crop residues exist in the literature. However, these methods are time consuming and not suited for quantitative evaluation. They have the disadvantage of being less rigorous and accurate because they do not consider the spectral mixture of different materials in the same pixel. In this study, the potential of hyperspectral (Probe-1) and multispectral high spatial resolution (IKONOS) data were compared for estimating and mapping crop residues on agricultural lands using the constrained linear spectral mixture analysis approach. Image data were spectrally and radiometrically calibrated, atmospherically corrected, as well as geometrically rectified. Pure spectral signatures of residues, bare soil and crop cover were manually extracted from image data based on prior knowledge of the fields. Percent (fraction) cover for each sampling point was extracted using unmixing and validated against ground reference measurements. The best results were achieved for the crop cover (index of agreement ( D) = 0.92 and root mean square error (RMSE) = 0.09) adjusted for the impurity of the endmembers canola, pea and wheat, followed by the wheat residues ( D = 0.76 and RMSE = 0.12). Considering only the wheat residues in fields with a canola crop, D increases to 0.86. The soil fractions were generally underestimated with D = 0.72, and no significant improvements could be made after adjusting for the shadow effect. The estimations from the IKONOS data were poorer for the same cover types (residues: D = 0.40 and RMSE = 0.24; crop: D = 0.51 and RMSE = 0.38; soil: D = 0.58 and RMSE = 0.29). Relative to the IKONOS data, the better performance of the hyperspectral data is mainly due to the improved spectral band characteristics, especially in the SWIR, which is sensitive to the residues (lignin and cellulose absorption features), soil and crop cover.

Incorporating Remote Sensing Information in Modeling House Values

This paper explores the possibility of incorporating remote sensing information in modeling house values in the City of Milwaukee, Wisconsin, U.S.A. In particular, a Landsat ETMimage was utilized to derive environmental character- istics, including the fractions of vegetation, impervious surface, and soil, with a linear spectral mixture analysis approach. These environmental characteristics, together with house structural attributes, were integrated to house value models. Two modeling techniques, a global OLS regression and a regression tree approach, were employed to build the relationship between house values and house structural and environmental characteristics. Analysis of results indicates that environmental characteristics gener- ated from remote sensing technologies have strong influ- ences on house values, and the addition of them improves house value modeling performance significantly. Moreover, the regression tree model proves as a better alternative to the OLS regression models in terms of predicting accuracy. In particular, based on the testing dataset, the mean average error (MAE) and relative error (RE) dropped from 0.202 and 0.434 for the OLS model to 0.134 and 0.280 for the regression tree model, while the correlation coefficient between the predicted and observed values increased from 0.903 to 0.960. Further, as a nonparametric and local model, the regression tree method alleviates the problems with the OLS techniques and provides a means in delineating urban housing submarkets.

Application of spectral mixture analysis to Amazonian land-use and land-cover classification

Abundant vegetation species and associated complex forest stand structures in moist tropical regions often create difficulties in accurately classifying land-use and land-cover (LULC) features. This paper examines the value of spectral mixture analysis (SMA) using Landsat Thematic Mapper (TM) data for improving LULC classification accuracy in a moist tropical area in Rondônia, Brazil. Different routines, such as constrained and unconstrained least-squares solutions, different numbers of endmembers, and minimum noise fraction transformation, were examined while implementing the SMA approach. A maximum likelihood classifier was also used to classify fraction images into seven LULC classes: mature forest, intermediate secondary succession, initial secondary succession, pasture, agricultural land, water, and bare land. The results of this study indicate that reducing correlation between image bands and using four endmembers improve classification accuracy. The overall classification accuracy was 86.6% for the seven LULC classes using the best SMA processing routine, which represents very good results for such a complex environment. The overall classification accuracy using a maximum likelihood approach was 81.4%. Another finding is that use of constrained or unconstrained solutions for unmixing the atmospherically corrected or raw Landsat TM images does not have significant influence on LULC classification performances when image endmembers are used in a SMA approach.