Jasper Ridge Research Articles

Graph feature fusion driven by deep autoencoder for advanced hyperspectral image unmixing

In this paper, we propose a pioneering approach for blind Hyperspectral Image (HSI) unmixing named Multi-Features Graph Deep Fusion Learning Networks for HSI Unmixing (MF-GDL). Our method leverages the power of multi-feature graph deep fusion learning networks. MF-GDL integrates spectral HSI graph features with complementary spatial graph features, overcoming the fixed window size limitation observed in prior deep learning approaches in capturing spatial relationships effectively. A key contribution lies in the introduction of multi-view graph learning, a novel aspect in HSI unmixing. By incorporating graph similarity-based spatial information, we address an important gap in the existing literature, marking the first attempt, to our knowledge, to tackle this challenge using a graph deep learning method. Furthermore, we introduce auxiliary information based on Extended Morphological Profiles (EMPs) to enrich spatial context beyond high-level spatial relationships between neighboring pixels. This multi-view fusion learning not only improves accuracy in material abundance estimation but also demonstrates the significance of considering multiple perspectives in HSI unmixing. We validate the efficacy of MF-GDL on four real datasets: Jasper Ridge, Samson, Urban, and Apex showcasing its robust performance and highlighting the substantial contribution of the multi-view aspect in advancing HSI unmixing techniques.

Extreme drought alters methane uptake but not methane sink in semi-arid steppes of Inner Mongolia

Global climate change, particularly drought, is expected to alter grassland methane (CH4) oxidation, a key natural process against atmospheric greenhouse gas accumulation, yet the extent of this effect and its interaction with future atmospheric CH4 concentrations increases remains uncertain. To address this research gap, we measured CH4 flux during an imposed three-month rain-free period corresponding to a 100-year recurrence drought in soil mesocosms collected from 16 different Eurasian steppe sites. We also investigated the abundance and composition of methanotrophs. Additionally, we conducted a laboratory experiment to explore the impact of elevated CH4 concentration on the CH4 uptake capacity of grassland soil under drought conditions. We found that regardless of the type of grassland, CH4 flux was still being absorbed at its peak, meaning that all grasslands functioned as persistent CH4 sinks even when the soil water content (SWC) was <5 %. A bell-shaped relationship between SWC and CH4 uptake was observed in the soils. The average maximum CH4 oxidation rate in the meadow steppe was higher than that in the typical and desert steppe soils during extreme drought. The experimental elevation of atmospheric CH4 concentration counteracted the anticipated reduction in CH4 uptake related to physiological water stress on methanotrophic soil microbes under the drought stress. On the contrary, we found that across the regional scale, nitrogen, phosphorous, and total soil organic content played a crucial role in moderating the duration and magnitude of CH4 uptake with respect to SWC. USC-γ (Upland Soil Cluster γ) and JR-3 (Jasper Ridge Cluster) were the dominant group of soil methanotrophic bacteria in three types of grassland. However, the methanotrophic abundance, rather than the methanotrophic community composition, was the dominant microbiological factor governing CH4 uptake during the drought.

Mixed population trends inside a California protected area: Evidence from long‐term community science monitoring

Protected areas are one of the most widespread and accepted conservation interventions, yet population trends of species within protected areas are rarely compared with regional trends to gain insight into their effectiveness. Here, we leverage two long‐term community science datasets, finding mixed effects of protected areas on long‐term bird population trends. We analysed 31 years of bird transect data recorded by community volunteers across all major habitats of Stanford University's Jasper Ridge Biological Preserve to determine the population trends for a sample of 66 species. We found that nearly a third of species experienced long‐term declines, and on average, species declined by 12%. Further, we averaged species trends by conservation status and key life history attributes to identify correlates and possible drivers of these trends. Observed increases in some cavity‐nesters and declines of scrub‐associated species suggest that long‐term fire suppression may be a key driver, reshaping bird communities through changes in forest and chaparral structure and composition. Additionally, we compared our results with those of the North American Breeding Bird Survey's Central California Coast region (n = 55 species) to place Jasper Ridge in a broader context. Most species experienced similar directional population trends inside and outside of the preserve and only eight species (14.5%) did better inside this small protected area. Therefore, we must identify relevant management strategies for declining populations and explicitly consider how existing protected areas target and manage each species. Further, this analysis underlines the importance of local and national community science for revealing long‐term bird population trends.

A blind convolutional deep autoencoder for spectral unmixing of hyperspectral images over waterbodies

Harmful algal blooms have dangerous repercussions for biodiversity, the ecosystem, and public health. Automatic identification based on remote sensing hyperspectral image analysis provides a valuable mechanism for extracting the spectral signatures of harmful algal blooms and their respective percentage in a region of interest. This paper proposes a new model called a non-symmetrical autoencoder for spectral unmixing to perform endmember extraction and fractional abundance estimation. The model is assessed in benchmark datasets, such as Jasper Ridge and Samson. Additionally, a case study of the HSI2 image acquired by NASA over Lake Erie in 2017 is conducted for extracting optical water types. The results using the proposed model for the benchmark datasets improve unmixing performance, as indicated by the spectral angle distance compared to five baseline algorithms. Improved results were obtained for various metrics. In the Samson dataset, the proposed model outperformed other methods for water (0.060) and soil (0.025) endmember extraction. Moreover, the proposed method exhibited superior performance in terms of mean spectral angle distance compared to the other five baseline algorithms. The non-symmetrical autoencoder for the spectral unmixing approach achieved better results for abundance map estimation, with a root mean square error of 0.091 for water and 0.187 for soil, compared to the ground truth. For the Jasper Ridge dataset, the non-symmetrical autoencoder for the spectral unmixing model excelled in the tree (0.039) and road (0.068) endmember extraction and also demonstrated improved results for water abundance maps (0.1121). The proposed model can identify the presence of chlorophyll-a in waterbodies. Chlorophyll-a is an essential indicator of the presence of the different concentrations of macrophytes and cyanobacteria. The non-symmetrical autoencoder for spectral unmixing achieves a value of 0.307 for the spectral angle distance metric compared to a reference ground truth spectral signature of chlorophyll-a. The source code for the proposed model, as implemented in this manuscript, can be found at https://github.com/EstefaniaAlfaro/autoencoder_owt_spectral.git.

Soil geochemistry of hydrogen and other gases along the San Andreas fault

Natural hydrogen has generated great interest as a potential clean and renewable energy source. To understand the occurrence of natural hydrogen, 103 1-m deep soil gas samples were acquired near the San Andreas Fault at Jasper Ridge and Portola Valley, California, USA. The gas samples were analyzed for hydrogen, helium, carbon dioxide, light hydrocarbons, and fixed gas concentrations. Statistical data analysis was carried out to group samples, reveal their spatial distribution, and understand possible sources of the gases.High concentrations of hydrogen up to 20.3 ppmv and 17.3 ppmv occur in Jasper Ridge and Portola Valley, respectively, ∼30–35 times greater than the atmospheric concentration. Most samples with high hydrogen concentrations fall on or near faults, suggesting an origin by serpentinization or geomechanical activation of catalytic sites in minerals, although a deep-seated primordial origin cannot be excluded. Elevated concentrations of carbon dioxide resulted from aerobic microbial degradation of organic matter and elevated concentrations of light hydrocarbons likely resulted from thermal cracking of organic matter.

SSANet: An Adaptive Spectral–Spatial Attention Autoencoder Network for Hyperspectral Unmixing

Convolutional neural-network-based autoencoders, which can integrate the spatial correlation between pixels well, have been broadly used for hyperspectral unmixing and obtained excellent performance. Nevertheless, these methods are hindered in their performance by the fact that they treat all spectral bands and spatial information equally in the unmixing procedure. In this article, we propose an adaptive spectral–spatial attention autoencoder network, called SSANet, to solve the mixing pixel problem of the hyperspectral image. First, we design an adaptive spectral–spatial attention module, which refines spectral–spatial features by sequentially superimposing the spectral attention module and spatial attention module. The spectral attention module is built to select useful spectral bands, and the spatial attention module is designed to filter spatial information. Second, SSANet exploits the geometric properties of endmembers in the hyperspectral image while considering abundance sparsity. We significantly improve the endmember and abundance results by introducing minimum volume and sparsity regularization terms into the loss function. We evaluate the proposed SSANet on one synthetic dataset and four real hyperspectral scenes, i.e., Samson, Jasper Ridge, Houston, and Urban. The results indicate that the proposed SSANet achieved competitive unmixing results compared with several conventional and advanced unmixing approaches with respect to the root mean square error and spectral angle distance.

Small Biological Preserves and Smaller Mammals: Can Protected Areas in Our Most Populous Regions Conserve Historic Community Composition?

The multi-faceted impacts of the Anthropocene are increasingly modifying natural ecosystems and threatening biodiversity. Protected spaces in and adjacent to urban landscapes may be critical in protecting species in human-modified systems. Can small, protected spaces act as reservoirs for biodiversity across dynamic spatial and temporal gradients of human impact? To address this question, we identified small mammal remains from modern raptor pellets and Holocene archaeological sites along a human modification gradient in the San Francisco Bay Area, CA. We assessed small mammal alpha and beta diversity for both modern and Holocene sites. We tested for significant differences between sites and time bins using permutational multivariate analysis of variance (PERMANOVA) and visualized these differences using non-metric multidimensional scaling (NMDS). We found that alpha diversity decreased with increasing human modification in Anthropocene sites, with no corresponding change between Holocene sites. Additionally, the alpha diversity of modern sites with moderate/high levels of human modification was significantly lower than that of protected modern sites and all Holocene sites, driven by the dominance of human-commensal and disturbance-tolerant species. On the other hand, the small mammal communities from a small, protected area (Jasper Ridge Biological Preserve) retained Holocene levels of alpha diversity. Jasper Ridge has also changed less over time in terms of overall community composition than more modified sites. Despite this, both PERMANOVA and NMDS show that Holocene and Anthropocene communities are significantly distinct regardless of collection site and level of anthropogenic modification. Our results suggest that even relatively small, protected spaces within an urbanized matrix conserve native faunal communities, highlighting their important role in urban conservation.

DHCAE: Deep Hybrid Convolutional Autoencoder Approach for Robust Supervised Hyperspectral Unmixing

Hyperspectral unmixing (HSU) is a crucial method to determine the fractional abundance of the material (endmembers) in each pixel. Most spectral unmixing methods are affected by low signal-to-noise ratios because of noisy pixels and bands simultaneously, requiring robust HSU techniques that exploit both 3D (spectral–spatial dimension) and 2D (spatial dimension) domains. In this paper, we present a new method for robust supervised HSU based on a deep hybrid (3D and 2D) convolutional autoencoder (DHCAE) network. Most HSU methods adopt the 2D model for simplicity, whereas the performance of HSU depends on spectral and spatial information. The DHCAE network exploits spectral and spatial information of the remote sensing images for abundance map estimation. In addition, DHCAE uses dropout to regularize the network for smooth learning and to avoid overfitting. Quantitative and qualitative results confirm that our proposed DHCAE network achieved better hyperspectral unmixing performance on synthetic and three real hyperspectral images, i.e., Jasper Ridge, urban and Washington DC Mall datasets.

Spatiotemporal impacts of the Anthropocene on small mammal communities, and the role of small biological preserves in maintaining biodiversity

The multi-faceted impacts of the Anthropocene are increasingly modifying natural ecosystems and threatening biodiversity. Can small protected spaces conserve small mammal diversity across spatial and temporal scales of human impact? We identified small mammal remains from modern raptor pellets and Holocene archeological sites along a human modification gradient in the San Francisco Bay Area, CA and evaluated alpha and beta diversity across sites and time periods. We found that Shannon diversity, standardized species richness, and evenness decrease across modern sites based on level of human modification, with no corresponding change between Holocene sites. Additionally, the alpha diversity of modern sites with moderate and high levels of human modification was significantly lower than the diversity of modern sites with low levels of human modification as well as all Holocene sites. On the other hand, the small mammal communities from Jasper Ridge Biological Preserve, a small protected area, retain Holocene levels of alpha diversity. Jasper Ridge has also changed less over time in terms of overall community composition (beta diversity) than more modified sites. Despite this, Holocene and Anthropocene communities are distinct regardless of study area. Our results suggest that small mammal communities today are fundamentally different from even a few centuries ago, but that even relatively small protected spaces can partially conserve native faunal communities, highlighting their important role in urban conservation.

Quadratic Clustering-Based Simplex Volume Maximization for Hyperspectral Endmember Extraction

The existence of intra-class spectral variability caused by differential scene components and illumination conditions limits the improvement of endmember extraction accuracy, as most endmember extraction algorithms directly find pixels in the hyperspectral image as endmembers. This paper develops a quadratic clustering-based simplex volume maximization (CSVM) approach to effectively alleviate spectral variability and extract endmembers. CSVM first adopts spatial clustering based on simple linear iterative clustering to obtain a set of homogeneous partitions and uses spectral purity analysis to choose pure pixels. The average of the chosen pixels in each partition is taken as a representative endmember, which reduces the effect of local-scope spectral variability. Then an improved spectral clustering based on k-means is implemented to merge homologous representative endmembers to further reduce the effect of large-scope spectral variability, and final endmember collection is determined by the simplex with maximum volume. Experimental results show that CSVM reduces the average spectral angle distance on Samson, Jasper Ridge and Cuprite datasets to below 0.02, 0.06 and 0.09, respectively, provides the root mean square errors of abundance maps on Samson and Jasper Ridge datasets below 0.25 and 0.10, and exhibits good noise robustness. By contrast, CSVM provides better results than other state-of-the-art algorithms.

Nonlinear Spectral Unmixing using Semi-Supervised Standard Fuzzy Clustering

Coarse resolution captured in remote sensing causes the combination of different materials in one pixel, called the mixed pixel. Spectral unmixing estimates the combination of endmembers in mixed pixels and their corresponding abundance maps in the Hyper/Multi spectral image. In this paper, a nonlinear spectral unmixing based on semi-supervised fuzzy clustering is proposed. First, pure pixels (endmembers) using Vertex Component Analysis (VCA) are extracted and those pixels are the labelled pixels where the membership value of each is 1 for the corresponding endmember and 0 for the others. Second, the semi-supervised fuzzy clustering is applied to find the membership matrix defining the fraction of the endmember in each mixed pixel and hence extract the abundance maps. The experiments were conducted on both synthetic data such as the Legendre data and real data such as Jasper Ridge data. The non-linearity of the Legendre data was performed by the Fan model on different signal-tonoise ratio values. The results of the new unmixing model show its significant performance when compared with four state-of the art unmixing algorithms

Opti-MSFA: a toolbox for generalized design and optimization of multispectral filter arrays.

Multispectral imaging captures spatial information across a set of discrete spectral channels and is widely utilized across diverse applications such as remote sensing, industrial inspection, and biomedical imaging. Multispectral filter arrays (MSFAs) are filter mosaics integrated atop image sensors that facilitate cost-effective, compact, snapshot multispectral imaging. MSFAs are pre-configured based on application-where filter channels are selected corresponding to targeted absorption spectra-making the design of optimal MSFAs vital for a given application. Despite the availability of many design and optimization approaches for spectral channel selection and spatial arrangement, major limitations remain. There are few robust approaches for joint spectral-spatial optimization, techniques are typically only applicable to limited datasets and most critically, are not available for general use and improvement by the wider community. Here, we reconcile current MSFA design techniques and present Opti-MSFA: a Python-based open-access toolbox for the centralized design and optimization of MSFAs. Opti-MSFA incorporates established spectral-spatial optimization algorithms, such as gradient descent and simulated annealing, multispectral-RGB image reconstruction, and is applicable to user-defined input of spatial-spectral datasets or imagery. We demonstrate the utility of the toolbox by comparing against other published MSFAs using the standard hyperspectral datasets Samson and Jasper Ridge, and further show application on experimentally acquired fluorescence imaging data. In conjunction with end-user input and collaboration, we foresee the continued development of Opti-MSFA for the benefit of the wider research community.

Reconstruction of high spectral resolution multispectral image using dictionary-based learning and sparse coding

With the rising development of imaging spectroscopy, there is an increase in demand for images with high spectral, spatial, and temporal resolution. This study proposes a reconstruction from low spectral resolution multispectral imagery to higher spectral resolution hyperspectral imagery to overcome the problem of low spatial resolution and less availability of multi-temporal hyperspectral images. This study proposes a novel approach to dictionary learning based on nonlinear unmixing for dense mixtures and linear unmixing for sparse mixtures. Additionally, least square-based sparse coding has been applied in this framework to reconstruct high-resolution spectra from low-resolution spectra. The proposed method has been implemented on standard as well as real datasets and the performance of reconstruction has been validated through parametric techniques. The proposed method has been compared with states of the art methods like Joint Sparse & Low-Ranking Method and Derivative based Learning Method and it is observed that the proposed method outperforms the other methods. Nonlinear Spectral Unmixing has been performed and classification accuracy has been assessed on regenerated image data and observed to produce satisfactory output. In this study, the regenerated high spectral resolution image has been used for species-level classification and mapping of pure and mixed patches of mangrove species and the output compared with the Hyperion hyperspectral dataset of Henry Island, Sunderbans, India. Similarly, the proposed algorithm has also been applied on the Indian Pines, Salinas, Cuprite and Jasper Ridge datasets and the regenerated high spectral resolution images have been used for subpixel level target detection and classification.

Generation of Sub-Pixel-Level Maps for Mixed Pixels in Hyperspectral Image Data

Hyperspectral data can find wide applications in classification and mapping of pure and mixed pixels in images of different land-cover types. Hyperspectral data of high spectral resolution enhance discrimination of target objects; but the low spatial resolution poses a challenge due to creation of mixed pixels. The cost of acquiring images at high resolution from sensors is high and rarely available. With images of coarser spatial resolution, it is difficult to identify the endmembers and their locations within the mixed pixel. This study utilizes the fractional abundance values of target endmembers obtained from linear spectral unmixing in locating the sub-pixels of a mixed pixel. The study illustrates the preparation of classified maps of finer spatial resolution by locating the sub-pixels through different mapping algorithms. A comparative analysis of these mapping algorithms, viz. attraction model-based sub-pixel mapping, simulated annealing, neighbourhood connectivity, cosine similarity-based mapping and Markov random fieldbased mapping has been made and an output generated. The algorithms have been implemented on standard hyperspectral datasets of Indian Pines having 200 spectral channels, Pavia University of 103 spectral channels and Jasper Ridge of 198 spectral channels. It has been observed that simulated annealing-based mapping produces higher accuracy rate than the other algorithms, whereas in terms of execution time, attraction model takes lesser time. The accuracy has been validated with the ground reference map of available standard hyperspectral datasets on which each algorithm has been tested and analysed.

Incidence of Galls on Sympatric California Oaks: Ecological and Physiological Perspectives

Galls are abnormal outgrowths on the external tissues of plants caused by a restricted group of organisms. In this study, we surveyed the incidence and diversity of galling structures in sympatric oak species of a biological preserve (Jasper Ridge Biological Preserve, NC, USA). We also measured different physiological parameters (SLA-specific leaf area, chlorophyll, nitrogen, flavonol, anthocyanin, and water content) in galled and ungalled leaves on the same individuals of the most abundant oak species (Quercus agrifolia, Q. lobata, and Q. douglasii). Overall, Q. lobata showed the highest gall incidence, with 64.5% of the sampled leaves affected by galls, followed by Q. douglasii, Q. agrifolia, Q. durata, and Q. kelloggii. The proportion of stems with galls was considerably lower than for leaves in all oak species, ranging from 0% incidence in Q. kelloggii to 27.4% in Q. lobata. The highly schlerophyllous Q. agrifolia supported the most diverse galling community at Jasper Ridge, with ten species, mostly belonging to the Cynipidae family. Our results show that leaf galling had no significant impact on the studied ecophysiological variables. The lack of differences between galled and ungalled leaves under controlled conditions (same tree and position in the tree) suggests that the ecophysiological variables measured are not significantly affected by galling agents or that our data collected for fully-developed galls (end of summer) are not sensitive enough to detect differences. However, there were some trends in plant responses to galling. Changes in galled vs. ungalled leaves were greater in flavonols, followed by chlorophyll, nitrogen, anthocyanin, SLA, and water content, indicating a nutrient deficiency in the plant nutrients. Our findings underscore the complexity of the gall-plant interaction and suggest some promising lines of future research.

A Blind Spectral Unmixing in Wavelet Domain

In this article, a wavelet-based energy minimization framework is developed for joint estimation of endmembers and abundances without assuming pure pixels while considering noisy scenario. Spectrally dense and overlapped hyperspectral data are represented using biorthogonal wavelet bases that yield a compact linear mixing model in the wavelet domain. It acts as the data term and helps to reduce solution space of the unmixed components. Three prior terms are incorporated to better handle the ill-posedness, i.e., the logarithm of determinant volume regularizer enforces minimum endmember simplex, smoothness (spatial) prior to individual abundance maps, and spectral constraint through learning dictionary of abundances. Alternating nonnegative least-squares is employed to optimize the regularized and constrained nonnegative matrix factorization functional in the wavelet domain. We conduct theoretical analysis, discuss convergence and algorithmic details. Experiments are conducted on synthetic and three real benchmark hyperspectral data AVIRIS Cuprite, HYDICE Urban, and AVIRIS Jasper Ridge. The efficacy of the proposed algorithm is evaluated by comparing results with state of the art.

Target spectrum based feature selection (TSFS): A new method based on chain coding for target detection problems

One of the controversial issues in hyperspectral remote sensing methods for target detection is whether the feature selection will be useful. Generally, feature selection methods are divided into variance based and wavelength based methods. Variance based feature selection methods like information-theory-based methods may eliminate the distinctive features because the distinctive features probably are not statistical principal components. Distinctive features are crucial to distinguish target from background and are maintained in wavelength based methods which concentrate on wavelength information. However, beside to wavelength-based information, target spectrum fluctuations are also critical for target detection. In addition, the wavelength based methods are often time consuming iterative methods with high computational cost. This paper introduces a new feature selection method considering hyperspectral target spectrum. The proposed algorithm has been developed based on Chain coding idea. We proposed Chain Filtering, Chain Encoding, and Chain Statistics as filter, embedded, and wrapper feature selection methods. In this paper, Chain filtering, Chain statistics, and Chain encoding approaches have been compared with different types of feature selection methods such as Principle Component Analysis (PCA) and Minimum Noise Fraction (MNF). Numerical tests have been executed using 4 datasets including Cuprite Nevada and Jasper Ridge datasets from AVIRIS, Botswana, and local datasets of Isfahan province from Hyperion sensor and using Constrained Energy Minimization (CEM) target detection. The results show the accuracy of target detection applying the proposed feature selection method increases from 85% to about 92% for Kaolinite and from 77% to 96% for Buddingtonite in comparison with PCA for Cuprite dataset. Furthermore, the increments more than 5% and 17.5% will be achieved in comparison with MNF, respectively. The results have shown that not only the proposed method overcomes the accuracy decrement issue of feature selection in target detection, but also it improves target detection accuracy by eliminating non-informative features for target detection applications. So feature selection will be an efficient tool for target detection if the applied feature selection method picks out the distinctive features well.

Superpixel based compression of hyperspectral image with modified dictionary and sparse representation

ABSTRACT Sparse representation provides an efficient way for the compression of hyperspectral images in the literature. In this work, an improved version of the Spectral-Spatial Adaptive Sparse Representation (SSASR), Modified SSASR (MSSASR), is proposed for hyperspectral image compression. In the first step of the proposed method, superpixel maps are generated for adaptive spatio-spectral representation. Then, the best possible dictionary is computed for the representation of the data. Afterwards, sparse coefficients are determined depending on the dictionary by Simultaneous Orthogonal Matching Pursuit (SOMP) method. In the final step, the dictionary and sparse coefficients are encoded by quantization and entropy encoding. This paper has the following novelties: modified dictionary learning step, new ordering scheme and Differential Pulse Code Modulation (DPCM) usage. Owing to modified dictionary learning, the sparse coefficients can be represented more compact than traditional SSASR. By using of new ordering scheme, it is not needed to send the superpixel map as side information. Moreover, DCPM usage lowers the magnitudes of sparse coefficients. Thanks to these modifications, the proposed method achieves an important improvement on compression performance. In the experimental results, the proposed method is compared with PCA+JPEG2000, DWT+JPEG2000, 3D-SPECK, 3D-TARP and SSASR methods on Indian Pines, Washington DC Mall, Jasper Ridge and Moffett Field scenes. The evaluation is carried out not only using distance and similarity metrics, namely, signal-to-noise ratio, mean spectral angle and mean spectral correlation metrics but also computation times. Additionally, reconstruction quality in anomaly regions is also used for comparison. Experimental results show that the proposed method outperforms the other compression methods in terms of quality metrics and anomaly preserving performance.

Hyperspectral unmixing using deep convolutional autoencoder

ABSTRACTHyperspectral Unmixing (HU) estimates the combination of endmembers and their corresponding fractional abundances in each of the mixed pixels in the hyperspectral remote sensing image. In this paper, we address the linear unmixing problem with an unsupervised Deep Convolutional Autoencoder network (DCAE). The proposed DCAE is an end-to-end model that consists of two parts: encoder and decoder. First, deep convolutional encoder is employed to extract significant, denoised, non-redundant feature vector. Then one-layer decoder maps the aforementioned feature vector to obtain the endmembers and their abundance percentages. Extensive experiments were carried to evaluate the performance of the proposed method using synthetic and real hyperspectral datasets namely Samson, Cuprite, Urban, and Jasper Ridge. Analyses of the results demonstrate that the proposed DCAE significantly outperforms benchmark unmixing methods even in a noisy environment in terms of both Root Mean Square Error (RMSE), and Mean Square Error (MSE). The achieved results of the proposed DCAE in terms of mean absolute error were 0.0097, 0.001, 0.0141, and 0.0145 for Samson, Cuprite, Urban, and Jasper Ridge datasets respectively.

Fire affects the taxonomic and functional composition of soil microbial communities, with cascading effects on grassland ecosystem functioning.

Fire is a crucial event regulating the structure and functioning of many ecosystems. Yet few studies have focused on how fire affects taxonomic and functional diversities of soil microbial communities, along with changes in plant communities and soil carbon (C) and nitrogen (N) dynamics. Here, we analyze these effects in a grassland ecosystem 9months after an experimental fire at the Jasper Ridge Global Change Experiment site in California, USA. Fire altered soil microbial communities considerably, with community assembly process analysis showing that environmental selection pressure was higher in burned sites. However, a small subset of highly connected taxa was able to withstand the disturbance. In addition, fire decreased the relative abundances of most functional genes associated with C degradation and N cycling, implicating a slowdown of microbial processes linked to soil C and N dynamics. In contrast, fire stimulated above- and belowground plant growth, likely enhancing plant-microbe competition for soil inorganic N, which was reduced by a factor of about 2. To synthesize those findings, we performed structural equation modeling, which showed that plants but not microbial communities were responsible for significantly higher soil respiration rates in burned sites. Together, our results demonstrate that fire 'reboots' the grassland ecosystem by differentially regulating plant and soil microbial communities, leading to significant changes in soil C and N dynamics.