High Resolution Hyperspectral Imagery Research Articles

Retrieval of Salt Marsh Above-Ground Biomass from High-Spatial Resolution Hyperspectral Imagery Using PROSAIL

Salt marsh vegetation density varies considerably on short spatial scales, complicating attempts to evaluate plant characteristics using airborne remote sensing approaches. In this study, we used a mast-mounted hyperspectral imaging system to obtain cm-scale imagery of a salt marsh chronosequence on Hog Island, VA, where the morphology and biomass of the dominant plant species, Spartina alterniflora, varies widely. The high-resolution hyperspectral imagery allowed the detailed delineation of variations in above-ground biomass, which we retrieved from the imagery using the PROSAIL radiative transfer model. The retrieved biomass estimates correlated well with contemporaneously collected in situ biomass ground truth data ( R 2 = 0.73 ). In this study, we also rescaled our hyperspectral imagery and retrieved PROSAIL salt marsh biomass to determine the applicability of the method across spatial scales. Histograms of retrieved biomass changed considerably in characteristic marsh regions as the spatial scale of the imagery was progressively degraded. This rescaling revealed a loss of spatial detail and a shift in the mean retrieved biomass. This shift is indicative of the loss of accuracy that may occur when scaling up through a simple averaging approach that does not account for the detail found in the landscape at the natural scale of variation of the salt marsh system. This illustrated the importance of developing methodologies to appropriately scale results from very fine scale resolution up to the more coarse-scale resolutions commonly obtained in airborne and satellite remote sensing.

Using hyperspectral imaging to vector towards mineralization at the Canadian Malartic gold deposit, Québec, Canada

Canadian Malartic is a large-tonnage, low-grade Archean gold deposit (16.3 Moz, 1.08 g/t Au) located in the Abitibi region of Québec, Canada. A large part of the mineralization is hosted in the Pontiac Group metasedimentary rocks, which consist of mudstones to greywackes at upper greenschist to amphibolite facies. In exploration and production environments, these lithologies are challenging to characterize by conventional core logging methods, while in a research setting thin section-sized samples (2 cm × 4 cm) do not capture the full extent of mineralogical variability, which can extend from centimeters to meters away from mineralized zones. Here, high-resolution hyperspectral imagery (0.2–1.0 mm/pixel) in both shortwave infrared (SWIR, 1000–2500 nm) and longwave infrared (LWIR, 8000–12000 nm) is acquired for over two thousand meters of drill core, and is used to visualize changes in mineralogy and mineral chemistry related to metamorphism and hydrothermal alteration. Unaltered metasedimentary rocks contain metamorphic white mica with AlVI contents varying between 1.90 and 1.75 apfu (2195 to 2203 nm), depending on metamorphic grade. Hydrothermal alteration is characterized by white mica which becomes progressively more phengitic with increasing alteration intensity, with AlVI contents ranging from 1.70 to 1.50 apfu (2204 to 2212 nm). Phengitic white mica extends from meters to tens of meters away from major mineralized zones, and can be used as a vector towards mineralization in an exploration setting. White mica composition is correlated to Au content, and can be used to discriminate between unmineralized (<0.3 g/t Au, 2195–2204 nm), weakly mineralized (>0.3 g/t Au, 2205–2208 nm), and highly mineralized (>1.0 g/t Au, >2208 nm) samples, which is a simple metric that can directly be applied to sort ore in a production environment. The Mg# (molar Mg/[Mg + Fe]) of biotite, on the other hand, is unaffected by metamorphic grade, and consistently is Mg# 55–60 (2251–2250 nm) in unaltered samples. In mineralized samples, biotite is Mg-rich (Mg# >65, <2249 nm), but rapidly grades into background compositions outside of the ore zones. LWIR hyperspectral data is used in drill core to estimate silicification abundance from the strength of the 9200 nm quartz reflectance peak. Peak strength was generally in good agreement with the silicification abundance estimated by core logging, and is therefore a metric that can directly be used for characterization of this parameter in exploration and production environments. Over 1500 SWIR point measurements were also acquired in a 50 km by 15 km area around the deposit, and were used to delineate metamorphic isograds on regional scales, and a large 12 km by 3 km hydrothermal alteration halo around the Canadian Malartic deposit. Similar measurements, if performed along the over 200 km long Cadillac-Larder Lake Deformation zone, could be used to rapidly delineate other potentially mineralized areas.

Variational Pansharpening for Hyperspectral Imagery Constrained by Spectral Shape and Gram⁻Schmidt Transformation.

Image pansharpening can generate a high-resolution hyperspectral (HS) image by combining a high-resolution panchromatic image and a HS image. In this paper, we propose a variational pansharpening method for HS imagery constrained by spectral shape and Gram–Schmidt (GS) transformation. The main novelties of the proposed method are the additional spectral and correlation fidelity terms. First, we design the spectral fidelity term, which utilizes the spectral shape feature of the neighboring pixels with a new weight distribution strategy to reduce spectral distortion caused by the change in spatial resolution. Second, we consider that the correlation fidelity term uses the result of GS adaptive (GSA) to constrain the correlation, thereby preventing the low correlation between the pansharpened image and the reference image. Then, the pansharpening is formulized as the minimization of a new energy function, whose solution is the pansharpened image. In comparative trials, the proposed method outperforms GSA, guided filter principal component analysis, modulation transfer function, smoothing filter-based intensity modulation, the classic and the band-decoupled variational methods. Compared with the classic variation pansharpening, our method decreases the spectral angle from 3.9795 to 3.2789, decreases the root-mean-square error from 309.6987 to 228.6753, and also increases the correlation coefficient from 0.9040 to 0.9367.

Mapping and classifying green infrastructure typologies for climate-related studies based on remote sensing data

Despite the current evidence on the thermal benefits of vegetation and water bodies, further research is needed to investigate how cooling capacities are influenced by particular types, amounts, and spatial arrangements of green infrastructure (GI). However, there are no commonly agreed typologies that can be confidently used to compare and report the existing climatological effects of GI. Two previous studies were conducted to respond to this gap, and a conceptual GI typology matrix was developed according to functional, structural and configurational attributes. The present research presents a streamlined version and tests its applicability for the automated mapping, classification and thermal evaluation of GI using remote sensing data. A combination of parameters is introduced, including surface cover fractions and FRAGSTATS metrics estimated from very high-resolution hyperspectral imagery, LiDAR and cadastral data. The proposed framework can be applied at different spatial scales to analyse large urban areas rapidly and with high spatial accuracy. This paper also proposes a replicable workflow that can be implemented by researchers and practitioners to map existing vegetation conditions, prioritise greening interventions and assess thermal conditions with greater confidence. In this study, this workflow was successfully applied at local scale to classify green open spaces, tree canopies and water bodies in the city of Sydney. Evidence presented here demonstrates the applicability of the proposed system to evaluate and compare the intra- and inter-typology variability of land surface temperatures (LSTs), which can be potentially applied for performance assessment across other ecosystem service categories. Despite satisfactory results, the proposed typology is contingent on further developments and tests on a larger spatial extent and with a greater number of observations. Further statistical analysis is also required to determine the cooling capacity of typologies and quantify the influence of measured parameters on LSTs.

Effects of Heterogeneity within Tree Crowns on Airborne-Quantified SIF and the CWSI as Indicators of Water Stress in the Context of Precision Agriculture

This research focused on understanding the effects of structural heterogeneity within tree crowns on the airborne retrieval of solar-induced chlorophyll fluorescence (SIF) and the Crop Water Stress Index (CWSI). We explored the SIF and CWSI variability observed within crowns of trees subjected to different water stress regimes and its effect on the relationships with leaf physiological measurements. High-resolution (20 cm) hyperspectral imagery was acquired to assess fluorescence retrieval from sunlit portions of the tree crowns using the Fraunhofer line depth method, and from entire crowns using automatic object-based tree crown detection methods. We also measured the canopy temperature distribution within tree crowns using segmentation algorithms based on temperature percentiles applied to high-resolution (25 cm) thermal imagery. The study was conducted in an almond orchard cultivated under three watering regimes in Cordoba, in southern Spain. Three airborne campaigns took place during the summer of 2015 using high-resolution hyperspectral and thermal cameras on board a manned aircraft. Relationships between SIF and the assimilation rate improved significantly when the sunlit tree crown pixels extracted through segmentation were used for all flight dates. By contrast, the SIF signal extracted from the entire tree crowns was highly degraded due to the canopy heterogeneity observed within tree crowns. The quartile crown segmentations applied to the thermal images showed that the CWSI values obtained were within the theoretically expected CWSI range only when the pixels were extracted from the 50th percentile class. However, the CWSI values were biased in the upper quartile (Q75) for all watering regimes due to the soil background effects on the calculated mean crown temperature. The relationship between the CWSI and Gs was heavily affected by the crown segmentation levels applied and improved remarkably when the CWSI values were calculated from the middle quartile crown segmentation (Q50), corresponding to the coldest and purest vegetation pixels (r2 = 0.78 in pure vegetation pixels vs. r2 = 0.52 with the warmer pixels included in the upper quartile). This study highlights the importance of using high-resolution hyperspectral and thermal imagery for pure-object segmentation extractions from tree crowns in the context of precision agriculture and water stress detection.

Understanding the temporal dimension of the red-edge spectral region for forest decline detection using high-resolution hyperspectral and Sentinel-2a imagery

The operational monitoring of forest decline requires the development of remote sensing methods that are sensitive to the spatiotemporal variations of pigment degradation and canopy defoliation. In this context, the red-edge spectral region (RESR) was proposed in the past due to its combined sensitivity to chlorophyll content and leaf area variation. In this study, the temporal dimension of the RESR was evaluated as a function of forest decline using a radiative transfer method with the PROSPECT and 3D FLIGHT models. These models were used to generate synthetic pine stands simulating decline and recovery processes over time and explore the temporal rate of change of the red-edge chlorophyll index (CI) as compared to the trajectories obtained for the structure-related Normalized Difference Vegetation Index (NDVI). The temporal trend method proposed here consisted of using synthetic spectra to calculate the theoretical boundaries of the subspace for healthy and declining pine trees in the temporal domain, defined by CItime=n/CItime=n+1 vs. NDVItime=n/NDVItime=n+1. Within these boundaries, trees undergoing decline and recovery processes showed different trajectories through this subspace. The method was then validated using three high-resolution airborne hyperspectral images acquired at 40 cm resolution and 260 spectral bands of 6.5 nm full-width half-maximum (FWHM) over a forest with widespread tree decline, along with field-based monitoring of chlorosis and defoliation (i.e., ‘decline’ status) in 663 trees between the years 2015 and 2016. The temporal rate of change of chlorophyll vs. structural indices, based on reflectance spectra extracted from the hyperspectral images, was different for trees undergoing decline, and aligned towards the decline baseline established using the radiative transfer models. By contrast, healthy trees over time aligned towards the theoretically obtained healthy baseline. The applicability of this temporal trend method to the red-edge bands of the MultiSpectral Imager (MSI) instrument on board Sentinel-2a for operational forest status monitoring was also explored by comparing the temporal rate of change of the Sentinel-2-derived CI over areas with declining and healthy trees. Results demonstrated that the Sentinel-2a red-edge region was sensitive to the temporal dimension of forest condition, as the relationships obtained for pixels in healthy condition deviated from those of pixels undergoing decline.

Multi-Temporal and Spectral Analysis of High-Resolution Hyperspectral Airborne Imagery for Precision Agriculture: Assessment of Wheat Grain Yield and Grain Protein Content.

This study evaluates the potential of high resolution hyperspectral airborne imagery to capture within-field variability of durum wheat grain yield (GY) and grain protein content (GPC) in two commercial fields in the Yaqui Valley (northwestern Mexico). Through a weekly/biweekly airborne flight campaign, we acquired 10 mosaics with a micro-hyperspectral Vis-NIR imaging sensor ranging from 400–850 nanometres (nm). Just before harvest, 114 georeferenced grain samples were obtained manually. Using spectral exploratory analysis, we calculated narrow-band physiological spectral indices—normalized difference spectral index (NDSI) and ratio spectral index (RSI)—from every single hyperspectral mosaic using complete two by two combinations of wavelengths. We applied two methods for the multi-temporal hyperspectral exploratory analysis: (a) Temporal Principal Component Analysis (tPCA) on wavelengths across all images and (b) the integration of vegetation indices over time based on area under the curve (AUC) calculations. For GY, the best R2 (0.32) were found using both the spectral (NDSI—Ri, 750 to 840 nm and Rj, ±720–736 nm) and the multi-temporal AUC exploratory analysis (EVI and OSAVI through AUC) methods. For GPC, all exploratory analysis methods tested revealed (a) a low to very low coefficient of determination (R2 ≤ 0.21), (b) a relatively low overall prediction error (RMSE: 0.45–0.49%), compared to results from other literature studies, and (c) that the spectral exploratory analysis approach is slightly better than the multi-temporal approaches, with early season NDSI of 700 with 574 nm and late season NDSI of 707 with 523 nm as the best indicators. Using residual maps from the regression analyses of NDSIs and GPC, we visualized GPC within-field variability and showed that up to 75% of the field area could be mapped with relatively good predictability (residual class: −0.25 to 0.25%), therefore showing the potential of remote sensing imagery to capture the within-field variation of GPC under conventional agricultural practices.

Assessing the effects of forest health on sun-induced chlorophyll fluorescence using the FluorFLIGHT 3-D radiative transfer model to account for forest structure

Sun-induced fluorescence (SIF) has been proven to serve as a proxy of photosynthesis activity and therefore, as an early indicator of physiological alterations for global monitoring of vegetation. However, the interpretation of SIF over different spatial resolutions is critical to bridge the existing gap between local and global scales. This study provides insight into the influence of scene components, and forest structure and composition on the quantification of the red and far-red fluorescence signal as an early indicator of forest decline. The experiments were conducted over an oak forest (Quercus ilex) affected by water stress and Phytophthora infection in the southwest of Spain. SIF retrievals through the Fraunhofer Line Depth (FLD) principle with three spectral bands F (FLD3) was assessed using high resolution (60cm) hyperspectral imagery extracting sunlit crown, full crown and aggregated pixels. Results showed the link between F (FLD3) extracted from sunlit crown pixels and the tree physiological condition in this context of disease infection, yielding significant relationships (r2=0.57, p<0.01) for midday xylem water potential (ψ), (r2=0.63, p<0.001) for the de-epoxidation state of the xanthophyll cycle (DEPS), and (r2=0.74, p<0.001) for leaf-level measurements of steady-state fluorescence yield (Fs). In contrast, a poor relationship was obtained when using aggregated pixels at 30m spatial resolution, where the relationship between the image-based F (FLD3) and Fs yielded a non-significant relationship (r2=0.25, p>0.05). These results demonstrate the need for methods to accurately retrieve crown SIF from aggregated pixels in heterogeneous forest canopies with large physiological variability among individual trees. This aspect is critical where structural canopy variations and the direct influence of background and shadows affect the SIF amplitude masking the natural variations caused by physiological condition. FluorFLIGHT, a modified version of the three dimensional (3-D) radiative transfer model FLIGHT was developed for this work, enabling the simulation of canopy radiance and reflectance including fluorescence at different spatial resolutions, such as may be derived from proposed satellite missions such as FLEX, and accounting for canopy structure and varying percentage cover. The 3-D modelling approach proposed here significantly improved the relationship between Fs and F (FLD3) extracted from aggregated pixels (r2=0.70, p<0.001), performing better than when aggregation effects were not considered (r2=0.42, p<0.01). The FluorFLIGHT model used in this study improved the retrieval of SIF from aggregated pixels as a function of fractional cover, Leaf Area Index and chlorophyll content yielding significant relationships between Fs ground-data measurements and fluorescence quantum yield estimated with FluorFLIGHT at p<0.01 (r2=0.79). The methodology presented here using FluorFLIGHT also demonstrated its capabilities for mapping SIF at the tree level for single tree assessment of forest physiological condition in the context of early disease detection.

Early Detection and Quantification of Almond Red Leaf Blotch Using High-Resolution Hyperspectral and Thermal Imagery

Red leaf blotch is one of the major fungal foliar diseases affecting almond orchards. High-resolution thermal and hyperspectral airborne imagery was acquired from two flights and compared with concurrent field visual evaluations for disease incidence and severity. Canopy temperature and vegetation indices were calculated from thermal and hyperspectral imagery and analyzed for their ability to detect the disease at early stages. The classification methods linear discriminant analysis and support vector machine, using linear and radial basis kernels, were applied to a combination of these vegetation indices in order to quantify and discriminate between red leaf blotch severity levels. Chlorophyll and carotenoid indices and chlorophyll fluorescence were effective in detecting red leaf blotch at the early stages of disease development. Linear models showed higher power to separate between asymptomatic trees and those affected by advanced stages of disease development while the non-linear model was better in discriminating asymptomatic plants from those at early stages of red leaf blotch development. Leaf-level measurements of stomatal conductance, chlorophyll content, chlorophyll fluorescence, photochemical reflectance index, and spectral reflectance showed no significant differences between healthy leaves and the green areas of symptomatic leaves. This study demonstrated the feasibility of early detecting and quantifying red leaf blotch using high-resolution hyperspectral imagery.

Using High-Resolution Hyperspectral and Thermal Airborne Imagery to Assess Physiological Condition in the Context of Wheat Phenotyping

There is a growing need for developing high-throughput tools for crop phenotyping that would increase the rate of genetic improvement. In most cases, the indicators used for this purpose are related with canopy structure (often acquired with RGB cameras and multispectral sensors allowing the calculation of NDVI), but using approaches related with the crop physiology are rare. High-resolution hyperspectral remote sensing imagery provides optical indices related to physiological condition through the quantification of photosynthetic pigment and chlorophyll fluorescence emission. This study demonstrates the use of narrow-band indicators of stress as a potential tool for phenotyping under rainfed conditions using two airborne datasets acquired over a wheat experiment with 150 plots comprising two species and 50 varieties (bread and durum wheat). The flights were performed at the early stem elongation stage and during the milking stage. Physiological measurements made at the time of flights demonstrated that the second flight was made during the terminal stress, known to largely determine final yield under rainfed conditions. The hyperspectral imagery enabled the extraction of thermal, radiance, and reflectance spectra from 260 spectral bands from each plot for the calculation of indices related to photosynthetic pigment absorption in the visible and red-edge regions, the quantification of chlorophyll fluorescence emission, as well as structural indices related to canopy structure. Under the conditions of this study, the structural indices (i.e., NDVI) did not show a good performance at predicting yield, probably because of the large effects of terminal water stress. Thermal indices, indices related to chlorophyll fluorescence (calculated using the FLD method), and carotenoids pigment indices (PRI and CAR) demonstrated to be better suited for screening complex traits such as crop yield. The study concludes that the indicators derived from high-resolution thermal and hyperspectral airborne imagery are efficient tools for field-based phenotyping providing additional information to standard NDVI imagery currently used.

Spatial scaling of reflectance and surface albedo over a mixed-use, temperate forest landscape during snow-covered periods

Albedo products from the MODerate resolution Imaging Spectroradiometer (MODIS) have been validated extensively over spatially homogeneous sites with snow-cover. This study evaluated the spatial scaling of albedo and related reflectance-based quantities over areas of high spatial heterogeneity in temperate mixed forest, deciduous forest, urban and built-up, and cropland/natural mosaic lands under snow covered conditions. Reflectance-based quantities evaluated included spectral radiance, surface directional reflectance, and spectral and broadband albedo derived from ground- and tower-based measurements and high-resolution (5m) hyperspectral imagery (HSI) to coarse resolution (~500m) MODIS satellite data.Our approach first compared ground- and tower-based spectral and broadband reflectance quantities to HSI data, then to evaluate HSI with MODIS reflectance-based products. Over snow-covered pasture, HSI directional reflectance was biased lower than ground-based measurements collected using a spectroradiometer, and greatly underestimated at wavelengths less than 450nm. Tower-based shortwave broadband albedo (280–2800nm) and HSI albedo agreed within ±0.04. HSI directional spectral reflectance agreed well with tower-based measurements of spectral albedo collected using a spectroradiometer and remote cosine receptor above a mixed forest canopy with underlying snow cover. Spectral albedo collected over a dormant deciduous broadleaf canopy increased from 0.10 for snow-free conditions to 0.14–0.18 when snow-cover was present under the canopy. Canopy shortwave broadband albedo was not very sensitive to underlying snowpack depth, indicative of strong vegetation masking.Next, HSI data were spatially aggregated and averaged to 500m MODIS grids and compared to two MODIS albedo products: (1) MODIS/Terra and Aqua Combined Bidirectional Reflectance Distribution Function (BRDF) Albedo V005 (MCD43A, magnitude inversion), and (2) MODIS/Terra Snow Cover Daily L3 Global 500m Grid (MOD10A1). An assessment of surface heterogeneity demonstrated that MODIS products generally perform well for snow-covered landscapes with high spatial heterogeneity, with biases between ±0.04 and RMSEs less than 0.085 for mixed and deciduous broadleaf forested and urban & built-up land cover classes, and less than 0.11 for cropland/natural mosaic. Biases were generally lower for MOD10A1 compared to MCD43A3.Mean shortwave broadband albedo from MODIS and HSI over deciduous broadleaf and mixed forest with underlying snow cover ranged from 0.14 to 0.28. Albedo over MODIS grids characterized as a mosaic of cropland, forests, grasslands, and shrublands (e.g., cropland/natural mosaic) was substantially higher than forested grids under snow-covered conditions, ranging from 0.39 to 0.43.

Deriving Predictive Relationships of Carotenoid Content at the Canopy Level in a Conifer Forest Using Hyperspectral Imagery and Model Simulation

Recent studies have demonstrated that the R570/R515 index is highly sensitive to carotenoid (Cx + c) content in conifer forest canopies and is scarcely influenced by structural effects. However, validated methods for the prediction of leaf carotenoid content relationships in forest canopies are still needed to date. This paper focuses on the simultaneous retrieval of chlorophyll (Cα + b) and (Cx + c) pigments, which are critical bioindicators of plant physiological status. Radiative transfer theory and modeling assumptions were applied at both laboratory and field scales to develop methods for their concurrent estimation using high-resolution hyperspectral imagery. The proposed methodology was validated based on the biochemical pigment quantification. Canopy modeling methods based on infinite reflectance formulations and the discrete anisotropic radiative transfer (DART) model were evaluated in relation to the PROSPECT-5 leaf model for the scaling-up procedure. Simpler modeling methods yielded comparable results to more complex 3-D approximations due to the high spatial resolution images acquired, which enabled targeting pure crowns and reducing the effects of canopy architecture. The scaling-up methods based on the PROSPECT-5+DART model yielded a root-mean-square error (RMSE) and a relative RMSE of 1.48 μg/cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> (17.45%) and 5.03 μg/cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> (13.25%) for Cx + c and Cα + b, respectively, while the simpler approach based on the PROSPECT-5+Hapke infinite reflectance model yielded 1.37 μg/cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> (17.46%) and 4.71 μg/cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> (14.07%) for Cx + c and Cα+b, respectively. These predictive algorithms proved to be useful to estimate Cα + b and Cx + c from high-resolution hyperspectral imagery, providing a methodology for the monitoring of these photosynthetic pigments in conifer forest canopies.

Evidence of Physiological Decoupling from Grassland Ecosystem Drivers by an Encroaching Woody Shrub

Shrub encroachment of grasslands is a transformative ecological process by which native woody species increase in cover and frequency and replace the herbaceous community. Mechanisms of encroachment are typically assessed using temporal data or experimental manipulations, with few large spatial assessments of shrub physiology. In a mesic grassland in North America, we measured inter- and intra-annual variability in leaf δ13C in Cornus drummondii across a grassland landscape with varying fire frequency, presence of large grazers and topographic variability. This assessment of changes in individual shrub physiology is the largest spatial and temporal assessment recorded to date. Despite a doubling of annual rainfall (in 2008 versus 2011), leaf δ13C was statistically similar among and within years from 2008-11 (range of −28 to −27‰). A topography*grazing interaction was present, with higher leaf δ13C in locations that typically have more bare soil and higher sensible heat in the growing season (upland topographic positions and grazed grasslands). Leaf δ13C from slopes varied among grazing contrasts, with upland and slope leaf δ13C more similar in ungrazed locations, while slopes and lowlands were more similar in grazed locations. In 2011, canopy greenness (normalized difference vegetation index – NDVI) was assessed at the centroid of individual shrubs using high-resolution hyperspectral imagery. Canopy greenness was highest mid-summer, likely reflecting temporal periods when C assimilation rates were highest. Similar to patterns seen in leaf δ13C, NDVI was highest in locations that typically experience lowest sensible heat (lowlands and ungrazed). The ability of Cornus drummondii to decouple leaf physiological responses from climate variability and fire frequency is a likely contributor to the increase in cover and frequency of this shrub species in mesic grassland and may be generalizable to other grasslands undergoing woody encroachment.

Estimating leaf carotenoid content in vineyards using high resolution hyperspectral imagery acquired from an unmanned aerial vehicle (UAV)

Chlorophyll a+b (Ca+b) and carotenoids (Cx+c) are leaf pigments associated with photosynthesis, participation in light harvesting and energy transfer, quenching and photoprotection. This manuscript makes progress on developing methods for leaf carotenoid content estimation, using high resolution hyperspectral imagery acquired from an unmanned aerial vehicle (UAV). Imagery was acquired over 3 years using two different UAV platforms, a 6-band multispectral camera and a micro-hyperspectral imager flown with 260 bands at 1.85nm/pixel and 12-bit radiometric resolution, yielding 40cm pixel size and a FWHM of 6.4nm with a 25-μm slit in the 400–885nm spectral region. Field data collections were conducted in August 2009–2011 in the western area of Ribera del Duero Appellation d’Origine, northern Spain. A total of twelve full production vineyards and two study plots per field were selected to ensure appropriate variability in leaf biochemistry and vine physiological conditions. Leaves were collected for destructive sampling and biochemical determination of chlorophyll a+b and carotenoids conducted in the laboratory. In addition to leaf sampling and biochemical determination, canopy structural parameters, such as grid size, number of vines within each plot, trunk height, plant height and width, and row orientation, were measured on each 10m×10m plot. The R515/R570 index recently proposed for carotenoid estimation in conifer forest canopies was explored for vineyards in this study. The PROSPECT-5 leaf radiative transfer model, which simulates the carotenoid and chlorophyll content effects on leaf reflectance and transmittance, was linked to the SAILH and FLIGHT canopy-level radiative transfer models, as well as to simpler approximations based on infinite reflectance R∞ formulations. The objective was to simulate the pure vine reflectance without soil and shadow effects due to the high resolution hyperspectral imagery acquired from the UAV, which enabled targeting pure vines. The simulation results obtained with synthetic spectra demonstrated the effects due to Ca+b content on leaf Cx+c estimation when the R515/R570 index was used. Therefore, scaling up methods were proposed for leaf carotenoid content estimation based on the combined R515/R570 (sensitive to Cx+c) and TCARI/OSAVI (sensitive to Ca+b) narrow-band indices. Results demonstrated the feasibility of mapping leaf carotenoid concentration at the pure-vine level from high resolution hyperspectral imagery, yielding a root mean square error (RMSE) below 1.3μg/cm2 and a relative RMSE (R-RMSE) of 14.4% (FLIGHT) and 12.9% (SAILH) for the 2 years of hyperspectral imagery. The simpler formulation based on the infinite reflectance model by Yamada and Fujimura yielded lower errors (RMSE=0.87μg/cm2; R-RMSE<9.7%), although the slope deviated more from the 1:1 line. Maps showing the spatial variability of leaf carotenoid content were estimated using this methodology, which targeted pure vines without shadow and background effects.

Filtering high-resolution hyperspectral imagery in a maximum noise fraction transform domain using wavelet-based de-striping

Ever improving technology and computer processing power and decreasing cost have made hyperspectral image acquisition and analysis affordable in many applications. Hyperspectral images, acquired normally using pushbroom sensing systems, are tainted with noise and nonperiodic stripes. Few methods, including wavelet-based ones, have been proposed for reducing nonperiodic stripes from multispectral images; there are even fewer studies dealing with nonperiodic stripes in high-resolution hyperspectral images. Applying de-striping filters directly to individual hyperspectral image bands can be computationally inefficient and complicated considering the large number of bands in this type of image. This article compares the performance of wavelet-based de-striping algorithms as applied on high-resolution hyperspectral imagery. The algorithms are implemented directly on individual bands in the image domain and on selected bands in the image maximum noise fraction (MNF) transform domain. Two wavelet-based de-striping algorithms were tested and compared. The first algorithm eliminates wavelet detail components in the striping direction, while the second algorithm adaptively filters these components. The filtering methods are evaluated through visual and quantitative assessments. Quantitative assessment is performed by analysing the autocorrelation coefficient and signal-to-noise-ratio. The results show that images filtered in the MNF domain are superior in reducing stripes and noise while retaining the image information and without introducing distortions. The technique is computationally effective through filtering fewer bands, which reduces the need for filtering parameter determination and fine tuning. Visual and quantitative assessments also show that adaptive filtering of wavelet components is better than eliminating entire components due to the retention of image content.

Habitat reporting of a heathland site: Classification probabilities as additional information, a case study

Heathlands are man-made habitats and their decline during the last century can be contributed to shifts in both agricultural and management practices as well as to hydrological and atmospheric changes. As a result, many heathland sites, including the Kalmthoutse Heide in Belgium, were included in the European Natura 2000 program, a network of protected areas across the European Union. To assure an accurate mapping of the Kalmthoutse Heide and other Natura 2000 sites in Belgium a classification framework for habitat status reporting with remote sensing data and in particular high resolution hyperspectral imagery was started. In this study we propose a simple and fast context based method for mapping heathland heterogeneity using the intermediate, otherwise redundant, classification probabilities as generated by a hard classification algorithm. Our study proved to be successful in using intermediate classification probabilities as a valuable source of ecological information. The delineated areas have been shown to be statistically sound and robust compared to a neutral model. The technique is not limited to a particular hard classification technique and can easily be adopted into current vegetation monitoring efforts. The resulting maps provided accessible maps which can support management of the protected site and enhance the accuracy of EU reportage as required by the habitat directive.

Hybrid Segmentation – Artificial Neural Network Classification of High Resolution Hyperspectral Imagery for Site-Specific Herbicide Management in Agriculture

Site-Specific Herbicide Management (SSHM) in Precision Agriculture (PA) requires weed detection in crop fields for directed herbicide application instead of spraying entire fields. This has significant economic and environmental advantages for improved agricultural practices that are essential given forecast increases in global population and food production needs. In this study, a new hybrid segmentation - Artificial Neural Network (HS-ANN) method was compared to standard Maximum Likelihood Classification (MLC) for improving crop/weed species discrimination in SSHM/PA. Very high spatial resolution (1.25 mm) ground-based hyperspectral image data were acquired over field plots of canola, pea, and wheat crops seeded with two weed species (redroot pigweed, wild oat) in southern Alberta, Canada. The very high spatial and spectral resolution image data required development of a simple yet efficient vegetation index (Modified Chlorophyll Absorption in Reflectance Index (MCARI)) threshold segmentation to separate vegetation from soil for classification. The HSANN consistently outperformed MLC in both single date and multi-temporal classifications. Higher class accuracies were obtained with multi-temporally trained ANNs (84 to 92 percent overall), with improvements up to 31 percent over MLC. With advancements in imaging technology and computer processing speed, this HS-ANN method may constitute a viable farm option for real-time detection and mapping of weed species for SSHM in agriculture.

Classifying fine-scale spatial structure of riparian forests using hyperspectral high-resolution remotely sensed imagery at the Cedar River municipal watershed in western Washington, USA

Logging in the Pacific Northwest has often removed conifer trees from riparian forests, which are then colonized and dominated by red alder (Alnus rubra) and deciduous shrubs. As a result, deciduous-dominated sites are over-represented in streamside areas compared to historic conditions, causing degradation of stream habitat. A popular restoration strategy in the region is the reintroduction of conifer trees into riparian areas. An accurate characterization of streamside forest conditions at large spatial scales is required to identify disturbed areas. Traditional remote sensing imagery has not been capable of capturing the fine-scale spatial heterogeneity within riparian zones. We analyzed a MODIS–ASTER (MASTER) image collected at the Cedar River municipal watershed in western Washington to determine if this high-resolution (5 m) hyperspectral imagery (50 bands) would be better able to characterize the high level of heterogeneity common in this environment. A supervised classification procedure using a maximum likelihood algorithm was conducted delineating nine of the dominant land cover types in the study. Subsequent field observations and validation procedures indicated that the classification had an overall accuracy of roughly 80%. The results show that this technique provided a cost-effective means of mapping the fine-scale spatial heterogeneity typical of streamside forest stands in the region.

Hyperspectral data processing for repeat detection of small infestations of leafy spurge

Leafy spurge ( Euphorbia esula L.) is an invasive plant species in the north central and western U.S. and southern Canada. Idaho has established populations in the north and southeastern regions which are spreading into new sites. This study demonstrates the ability of high resolution hyperspectral imagery to provide high quality data and consistent methods to locate small and low percent canopy cover occurrences of leafy spurge. Locating leafy spurge in its early stages of invasion is critical for land managers in order to prioritize treatment, conservation, and restoration activities. Hyperspectral data were collected in 2002 and 2003 for the study area in southeastern Idaho. The imagery was classified with the Mixture Tuned Matched Filtering (MTMF) algorithm. Although classifications from single date images provided discrimination of leafy spurge at approximately 10% cover in one 3.5 m pixel, for repeatability and consistency purposes, the threshold for leafy spurge discrimination is approximately 40% cover. We hypothesize that georegistration errors, small differences in leafy spurge reflectance, training endmember selection, and image processing and field validation biases between years influence multi-date detection limits. Although hyperspectral imagery is costly, in some situations, the advantages of having reliable and repeatable mapping abilities for discrimination of economically damaging invasive species such as leafy spurge outweigh the image and processing costs.

Geostatistical and local cluster analysis of high resolution hyperspectral imagery for detection of anomalies

This paper describes a new methodology to detect small anomalies in high resolution hyperspectral imagery, which involves successively: (1) a multivariate statistical analysis (principal component analysis, PCA) of all spectral bands; (2) a geostatistical filtering of noise and regional background in the first principal components using factorial kriging; and finally (3) the computation of a local indicator of spatial autocorrelation to detect local clusters of high or low reflectance values and anomalies. The approach is illustrated using 1 m resolution data collected in and near northeastern Yellowstone National Park. Ground validation data for tarps and for disturbed soils on mine tailings demonstrate the ability of the filtering procedure to reduce the proportion of false alarms (i.e., pixels wrongly classified as target), and its robustness under low signal to noise ratios. In almost all scenarios, the proposed approach outperforms traditional anomaly detectors (i.e., RX detector which computes the Mahalanobis distance between the vector of spectral values and the vector of global means), and fewer false alarms are obtained when using a novel statistic S 2 (average absolute deviation of p-values from 0.5 through all spectral bands) to summarize information across bands. Image degradation through addition of noise or reduction of spectral resolution tends to blur the detection of anomalies, increasing false alarms, in particular for the identification of the least pure pixels. Results from a mine tailings site demonstrate the approach performs reasonably well for highly complex landscape with multiple targets of various sizes and shapes. By leveraging both spectral and spatial information, the technique requires little or no input from the user, and hence can be readily automated.