Depth-resolved estimation of soil organic carbon (SOC) remains challenging because optical measurements originate at the surface while carbon dynamics vary vertically. We propose a physics-aware UAV framework that integrates multispectral (MSI) and hyperspectral (HSI) imagery to estimate SOC concentration (%) across five depths. The experiment was conducted at Plantheaven Farms, Missouri, with 10 sorghum genotypes across three replicates. Feature construction combined spectral derivatives from HSI with texture features from MSI, compressed via principal component analysis. Physics-based regularization was implemented through (i) a second-difference penalty to enforce vertical smoothness and (ii) a profile-integral consistency constraint to preserve whole-profile balance. Four model configurations evaluated on local data showed progressive improvements: MSI-only, MSI+HSI, MSI+HSI with smoothness, and MSI+HSI with full physics constraints. Additionally, transfer learning from the Open Soil Spectral Library (OSSL) was tested to address data limitations. Model fitting on the available data achieved R² = 0.72 at 0-30 cm, with physics-aware constraints notably improving vertical coherence. The physics-aware model reduced variance and improved plausibility. In-sample, transfer learning achieved R² = 0.60 at 0-30 cm, with conservative interpretation below 90 cm due to reduced optical sensitivity. Exploratory genotype patterns suggested higher surface SOC percent for PI 656029 and PI 656057, and lower values for PI 276837 and PI 656044.

Geospatial artificial intelligence (GeoAI) and data processing techniques have significantly advanced object detection, prediction, and classification tasks. However, the availability of machine learning-ready, labeled data for specific applications such as plant disease detection remains the major challenge for the broader adoption of GeoAI. For instance, collecting temporal unmanned aerial vehicle (UAV) imagery of agricultural crops to track disease emergence and progress requires substantial human labor and resources, which is often limited to a small spatial scale. Recognizing the pivotal role of temporal data in pattern recognition, object detection, and scene reconstruction, we introduce an innovative approach to augment multispectral temporal datasets: the geospatial time machine (GTM). Our proposed methodology combines graph neural network (GNN) and generative adversarial network (GAN) architectures to generate comprehensive synthetic temporal data encompassing multivariate time series. The results demonstrate that imagery generated through backcasting can enhance the accuracy of downstream classification tasks by up to 53% in plant disease detection, particularly in the initial stages of analyzing a crop growth using multispectral and multitemporal datasets.

Advancements in hyperspectral imaging technology open new avenues for Soil Organic Carbon (SOC) estimation, a critical factor in enhancing agricultural productivity and understanding carbon sequestration. This study analyzed 96 soil samples using the HySpex sensor system, covering visible and near-infrared (VNIR; 400-1000 nm over 700 spectral bands) and short-wave infrared (SWIR; 960-2500 nm over 362 spectral bands) spectrums. We integrated Discrete and Continuous Wavelet Transforms (DWT and CWT), applying Biorthogonal, Daubechies, Complex Morlet, Morlet, and Mexican hat wavelets for enhanced SOC detection. Our results indicate a marked improvement in SOC estimation, particularly with Complex Morlet and Mexican hat wavelets in CWT. Complementary chemical analysis validated the SOC levels. Depth-wise, the 15-30 cm range in the SWIR region showed the highest SOC correlation. We identified key spectral ranges, including 400-580 nm and 680-910 nm in VNIR, and 1150-2500 nm in SWIR, as strongly correlated with SOC, substantiated by lab analysis.

Monitoring plantations is crucial for crop management and producing healthy harvests. Unmanned Aerial Vehicles (UAVs) have been used to collect multispectral images that aid in this monitoring. However, given the number of hectares to be monitored and the limitations of flight, plant disease signals become visually clear only in the later stages of plant growth and only if the disease has spread throughout a significant portion of the plantation. This limited amount of relevant data hampers the prediction models, as the algorithms struggle to generalize patterns with unbalanced or unrealistic augmented datasets effectively. To address this issue, we propose PlantPlotGAN, a physics-informed generative model capable of creating synthetic multispectral plot images with realistic vegetation indices. These indices served as a proxy for disease detection and were used to evaluate if our model could help increase the accuracy of prediction models. The results demonstrate that the synthetic imagery generated from PlantPlotGAN outperforms state-of-the-art methods regarding the Frichet inception distance. Moreover, prediction models achieve higher accuracy metrics when trained with synthetic and original imagery for earlier plant disease detection compared to the training processes based solely on real imagery.

Assessing soil organic carbon (SOC) stocks is crucial for understanding the carbon sequestration potential of agroecosystems and for mitigating climate change. This study presents a novel method for assessing SOC and mineral content at various soil depths in sorghum crops using hyperspectral remote sensing. Conducted at Planthaven Farms, MO, the research encompassed ten genotypes across 30 plots, yielding 180 soil samples from six depth intervals (0–150 cm) of bare soil. Chemical analyses determined the SOC and mineral levels, which were then compared to spectral data from HySpex indoor sensors. We utilized time-frequency analysis methods, including discrete wavelet transformation (DWT), continuous wavelet transformation (CWT), and frame transformation along with traditional spectral transformations, specifically fractional derivatives and continuum removal. The analysis revealed the shortwave infrared (SWIR) region, particularly the 1800–2000 nm range, as having the strongest correlations with SOC content (with$R^{2}$exceeding 0.8). The visible near-infrared (VNIR) region also provided valuable insights. Models incorporating CWT achieved high accuracy (test$R^{2}$exceeding 0.9), while frame transformation achieved strong accuracy (test$R^{2}$between 0.7 and 0.8) with fewer features. The random forest regressor (RFR) proved to be most robust, demonstrating superior accuracy and reduced overfitting compared to support vector regression (SVR), partial least squares regression (PLSR), and deep neural network (DNN) models. The models demonstrated the efficacy of hyperspectral data for SOC estimation, suggesting potential for future applications that integrate this data with above-ground biomass to improve SOC mapping across larger scales. This research offers a promising spectral transformation approach for effective carbon management and sustainable agriculture in a changing climate.

Supply chain disruptors such as piracy and navigational obstacles like icebergs, pose a probable risk to economic development and national security. With the warming of the Arctic Sea ice due to intense changes in climatic conditions, the Arctic and the untapped natural resources that reside in the icy waters are becoming more accessible and prone to security risks. Resulting in more traffic throughout the once inaccessible region. This has made the Arctic region a geopolitical hotspot, with nations vying for the superiority of its resources. Icebergs are a navigational risk to vessels as collisions could result in delayed shipments, monetary costs due to vessel damage, and human health risks. Thus, a robust model must be developed to identify and discriminate between icebergs and vessels to aid in navigational efficacy and increase maritime domain awareness. This paper leverages a novel convolutional neural network (CNN) that employs Synthetic Aperture Radar (SAR) to differentiate between icebergs and vessels in Iceberg Alley. The 5,000-image dataset came from the Statoil and the Centre for Cold Ocean Resources Engineering (C-CORE) collaboration. The dataset consists of SAR imagery, in the HH+HV polarizations. The developed model showed promising results for iceberg and vessel classification, achieving near 90% accuracy on the prescribed dataset. Additionally, the developed model is compared to other existing model architectures to compare model efficiency. This work has a direct influence on navigational safety and transferable applicability to other emerging national security threats in the maritime domain.

Near-earth hyperspectral big data present both huge opportunities and challenges for spurring developments in agriculture and high-throughput plant phenotyping and breeding. In this article, we present data-driven approaches to address the calibration challenges for utilizing near-earth hyperspectral data for agriculture. A data-driven, fully automated calibration workflow that includes a suite of robust algorithms for radiometric calibration, bidirectional reflectance distribution function (BRDF) correction and reflectance normalization, soil and shadow masking, and image quality assessments was developed. An empirical method that utilizes predetermined models between camera photon counts (digital numbers) and downwelling irradiance measurements for each spectral band was established to perform radiometric calibration. A kernel-driven semiempirical BRDF correction method based on the Ross Thick-Li Sparse (RTLS) model was used to normalize the data for both changes in solar elevation and sensor view angle differences attributed to pixel location within the field of view. Following rigorous radiometric and BRDF corrections, novel rule-based methods were developed to conduct automatic soil removal; and a newly proposed approach was used for image quality assessment; additionally, shadow masking and plot-level feature extraction were carried out. Our results show that the automated calibration, processing, storage, and analysis pipeline developed in this work can effectively handle massive amounts of hyperspectral data and address the urgent challenges related to the production of sustainable bioenergy and food crops, targeting methods to accelerate plant breeding for improving yield and biomass traits.

The work aims to predict soybean yield using satellite and Unmanned Aerial Vehicle (UAV) synergy and machine learning. UAV RGB imagery and Worldview satellite data was acquired in the summer of 2017 over a soybean field near Columbia, Missouri, USA. Canopy spectral features from satellite data and structural features from UAV imagery were combined and fused to predict soybean grain yield. Commonly used machine learning regression methods including Extreme Learning Regression (ELR), Random Forest Regression (RFR), Support Vector Regression (SVR) and Partial Least Squares Regression (PLSR) were utilized to predict soybean yield using spectral features from satellite data and structural features from UAV imagery. The results showed that canopy structure features such as canopy height and canopy coverage are important indicators for soybean grain yield estimation, and complementarities between Satellite and UAV data reveal a great potential of synergy, and lead to an improved performance for soybean grain yield estimation.

In this contribution, we use field-based hyperspectral imaging (HSI) and partial least squares regression (PLSR) to estimate early indicators of grapevine physiological indicators, and analyze identified significant spectral regions for fast and accurate plant health monitoring. HSI and physiological measurements were carried out at two commercial vineyards in California, USA. The PLSR models were developed between reflectance spectra extracted from hyperspectral images and four vine physiological parameters, including stomatal conductance (Gs) photosynthetic CO2rate (A), intercellular CO2concentration (Ci) and transpiration rate (E). The results demonstrate PLSR models to predict physiological parameters ( R2≥ 0.6), and the best model was found for Gs (R2=0.7). The identified significant spectral regions overlap with most commonly used remote sensing stress indicator, suggesting that HSI coupled with PLSR has great potential for upscaling and broader agricultural applications.

An unmanned aerial system (UAS)-based imaging technology has gained great interests in modern photogrammetry and remote sensing. However, due to the limitations of UAS imaging devices, image enhancement (IE) has become a necessary process for improving the visual appearance of UAS images. Although a great amount of effort has been focused on improving image quality from different aspects, the major obstacles are from computational efficiency and complexity, such as manually adjusting the associated algorithmic parameters that account for various image luminance. To overcome these drawbacks, we propose a new adaptive yet highly efficient luminance enhancement method, namely, adaptive trigonometric transformation function (ATTF), for enhancing the visual quality of digital color images captured by a UAS. The ATTF is derived from a tangent-based transformation function whose characteristics adaptively change with respect to the variation of the image luminance. By combining ATTF with a Laplacian operator and a color restoration process, a well-balanced color image is obtained. The effectiveness of the proposed technique is evaluated on various UAS-based images and compared with other IE techniques.

In this contribution, we examine the potential of using field spectroscopy to discriminate the responses of five soybean cultivars to background ozone concentration. Statistical analysis of hyperspectral data including one-way analysis of variance (ANOVA) and spectral instability analysis (ISI) were used to identify the most effective wavelengths in mapping and differentiating the five cultivars with different tolerance to ozone damage. Our results show several distinctive spectral regions that can be used for effective crop type mapping within species level, and quantifying the effects of ozone damage at leaf and canopy scales. This work demonstrates that hyperspectral remote sensors soon become available from government and private sector satellites offer a new set of high-resolution spectral data that will help to quantify impacts background ozone concentrations due to climate change on food security.

We examine the spectral signatures and foliar biophysical and biogeochemical properties of different soybean cultivars that are known to be sensitive in varying degrees to background concentrations of ozone (O3). Specifically, the potential of plant biophysical variables from leaf reflectance spectra, including chlorophyll index, photochemical reflectance index, and leaf area index, to detect foliar O3damage is explored. The study was conducted at an agricultural test site located in Maryland Heights, Missouri, during the summer of 2014 where five different soybean cultivars were planted. Our results show that the soybean cultivars demonstrated different sensitivity to background O3as demonstrated by spectral indices and plant biophysical measurements. The outcome of this research has potential implications for development of space-based observation of large-scale crop responses to O3damage, as well as for biotechnological breeding efforts to improve O3tolerance under future climate scenarios, as background O3concentrations are expected to increase through the twenty-first century.

Radarsat-2 quad-pol data (C band) and both dual and quad-pol Phased Array L band Synthetic Aperture Radar (PALSAR) data are used to map invasive plant species and forest degradation in Betampona Natural Reserve, Madagascar. Polarimetric feature parameters including the amplitude of the HH-VV correlation coefficient, the polarization ratio HH/VV, the polarimetric phase differences and the linear depolarization ratio (HV/VV) are explored to indentify forest clusters and tree species. A full coherency matrix, a Freeman-Durden model based decomposition and a Cloud-Pottier eigenvalue-eigenvector decomposition are tested following a polarimetric speckle filtering procedure based on the Refined Lee filter. Then, Wishart unsupervised classification is performed, and the results are validated against ground truthing. Our results show that PALSAR polarimetric data holds great promise as a comprehensive survey method for the presence of invasive plant species in tropical rainforest.

Phase noise in an interferogram hinders the accuracy and reliability of interferometric synthetic aperture radar (InSAR) measurements, including deformation estimation and topographic mapping. The Goldstein filter is one of the most commonly used interferogram filters to reduce the effects of phase noise. In this paper, we present a modification to the Goldstein interferogram filter such that the maximum value for the filtering parameter alpha is set to greater than 1 over less coherent areas so that aggressive filtering is implemented on incoherent areas. We also discuss the combined use of estimated coherences from linear and non-linear filters to deal with the coherence saturation due to the strong filtering, which is crucial in generating InSAR deformation products.

Recently, simple methods for real-time surface dryness estimation - perpendicular drought indices based on red and near-infrared wavelengths of satellite data has been developed. In this paper, the normalized form of the method is proposed to assess droughts over U.S. great plains using long term records of the Advanced Very High Resolution Radiometer (AVHRR) from both Land Long Term Data Record (LTDR) and the Global Inventory Modeling and Mapping Studies (GIMMS) dataset.

Atmospheric corrected Landsat Enhanced Thematic Mapper Plus (ETM+) near-infrared (NIR) and shortwave infrared (SWIR) band reflectances are used to develop a new index to monitor vegetation water content (VWC) in terms of equivalent water thickness (EWT, cm). This paper outlines the first part of a research program to investigate the potential and physical basis of wavelengths in the optical domain to assess the VWC. Then, a method called vegetation water content index (VWCI) were developed using SWIR, and NIR wavelengths of ETM+ data. The relationship between the EWT at canopy level is explored through linking leaf reflectance data obtained from PROSPECT with canopy reflectance from SailH and in-situ measurements. Significant correlations are found between canopy EWT and the developed index for both modeled and ground measured data.

In recent decades, regional and global ecosystems, earth climate, global diversity has been suffering from ever increased negative effects and pressures from rapid population growth and urban expansion. In this paper, global urban growth during the last four decades are analyzed quantitatively using satellite data and GIS. The Digital Chart of the World (DCW) urban layer represents the urban area in 1960s. The GLCNMO (Global Land Cover by National Mapping Organization) urban map by population density, nighttime lights and MODIS NDVI are used to extract the urban expansion in 2003. The result indicates that urban development has been dramatically increasing in the developing countries, expending more noticeably than in the previous decades and urban forms of developing countries are more compact and dense than their counterparts in Europe and North America.

In the agriculture research field, the main object of drought monitoring is to gain the soil moisture, furthermore to confirm the loss of drought. Soil moisture is one of the most important factors affecting the crop especially for the drought area. Besides, the terrain of farmland also affects the soil moisture and the variety of crop. This paper will indicate the effects of the two factors above and the relationship between them. Agriculture drought research group of PKU has used TVDI (Temperature Vegetation Dryness Index) to evaluate the soil moisture in order to establish a model to predict crop yield, which has been used in the real task of meteorology and agriculture department. Using a special software system established by the research group, the TVDI could be conveniently calculated and the result accords with the real situation very well when we use it to compare with the local observation data.

Normalized difference vegetation index (NDVI) plays an important role in the detection of drought and desertification. However, being calculated by direction limited spectral reflectance, it is almost ignorant to anisotropy of target reflection. In this paper, on the basis of atmospheric correction of Landsat-7 Enhanced Thematic Mapper Plus (ETM+) data using 6S (Second Simulation of Satellite Signal in the Solar Spectrum) code, bidirectional reflectance distribution function (BRDF) of surface coverage is obtained to retrieve spectral albedos of certain wavelengths. Narrowband spectral albedo to broadband albedo conversion is accomplished via measured band data and simulation of unknown spectral wavelengths which are not available by Landsat-7 ETM+ imagery. Albedo based ecological safety monitoring index (ESMI) is developed and sensitivity analysis of ESMI in surface application is conducted. Application results show that ESMI has potential use in quantitative monitoring of eco-environmental problems

In the paper, the status of flood disaster in China and the progress of disaster prevention and reduction in the field of property insurance were analyzed. The characteristics and application fields of 3S (GIS, RS and GPS) were also introduced. According to the current need and future development of property insurance company, which were based on the investigation to the work of disaster prevention and reduction in property insurance and casualty company (abbreviated as PICC) China, the authors decided to apply 3S technologies to the field of property insurance and used the successful methods in foreign property insurance companies for reference to develop a decision support system of flood disaster for property insurance. The system linked well with the operational system of insurance company and realized seamless integration between different data sources. The work of disaster prevention and reduction in property insurance company was better organized both in theory ways and in key technologies. As a result, the economic benefit of property insurance company was really improved. The system was applied in Shenzhen, China and the result was satisfying

NDVI is a land surface parameter which plays an important role in ecology. It is also sensitive in the circulation process between land surface and atmosphere. The earlier research indicated that NDVI has closely relationship with some main biophysics parameters such as photosynthetically active radiation, leaf area index, biomass of vegetation, and so on. So NDVI is widely used in the study of global vegetation. In this article, it was presented that ecological indexes have been derived from NDVI using NOAA/AVHRR data by the method of principle component analysis (PCA). After the PCA of NDVI which calculated form the NOAA/AVHRR data, the relationship analysis aimed at the four ecological parameters which derived from NDVI. Then the results show that the PCA can compress the key information into four foremost principal components which would be called the ecological indices (EI). And the first principal component reflects the basic situation of vegetation overlay (EI_all), the second (EI_ss), the third (EI_ws) and the fourth (EI_aw) principal components indicated the vegetation change in seasons respectively. According to the direction curves of seasonal vegetation change, the four principal components seem to be their biological significance