Being in an arid zone that is frequently submitted to high winds, south-central Arizona regularly gets impacted by several blowing dust events or dust storms every year. Major consequences of these events are visibility impairment and ensuing road traffic accidents, and a variety of health issues induced by inhalation of polluted air loaded with fine particulate matter produced by wind erosion. Despite such problems, and thus a need for guidance on mitigation efforts, studies dealing with dust source attribution for the region are largely missing. Furthermore, existing dust models exhibit large uncertainties and deficiencies in simulating dust events, rendering them of limited use in attribution studies or early warning systems. Therefore, to address some of these model issues, we have developed a high-resolution (1 km) dust modeling system by building upon an existing modeling framework consisting of Weather Research and Forecasting (WRF), FENGSHA (a dust emission model), and Community Multiscale Air Quality (CMAQ) models. In addition to incorporating new representations in the dust emission scheme, including roughness correction factor, sandblasting efficiency, and dust source mask, we implemented, in the dust model, up-to-date and very high-resolution data on land use, soil texture, and vegetation index. We used the revised dust modeling system to simulate a springtime dust storm (08–09 April 2013) of relatively long duration that caused a regional traffic incident involving minor injuries. The model simulations compared reasonably well against observations of concentration of particulate matter with a diameter of 10 μm and smaller (PM₁₀) and satellite-derived dust optical depth and vertical profile of aerosol subtypes. Interestingly, simulation results revealed that the anthropogenic (cropland) dust sources contributed more than half (~53 % or 260 µg/m³) of total PM₁₀, during the dust storm, over the region including Phoenix and western Pinal County. Contrary to the conventional wisdom that desert is the main dust source, our findings for this region challenge such belief and suggest that the regional air quality modeling over dryland regions should emphasize an improved representation of dust from agricultural lands as well, especially during high wind episodes. Such representations have the potential to inform decision-making in order to reduce windblown dust-related hazards on public health and safety.

From preventing scurvy to being part of religious rituals, citrus are intrinsically connected to human health and perception. From tiny mandarins to head-sized pummelos, citrus capability of hybridization provides a vastly diverse array of fruit sizes and shapes, which in turn corresponds to a diversity of flavors and aromas. These sensory qualities are tightly linked to oil glands in the citrus skin. The oil glands are also key to understanding fruit development, and the essential oils contained by them are fundamental in the food and perfume industries. We study the shape of citrus based on 3D X-ray CT scan reconstruction of 163 different citrus samples comprising 58 different species and cultivars, including samples of all fundamental citrus species. First, using the power of X-rays and image processing, we are able to compare and contrast size ratios between different tissues, such as the size of the skin compared to the rind or the flesh. Second, we model the fruit shape as an ellipsoidal surface, and later we study and infer possible oil gland distributions on this surface using principles of directional statistics. We finally compare and contrast these overall fruit shape models along their gland distributions across different citrus species. This morphological modeling will allow us later to link genotype with phenotype, furthering our insight on how the physical shape is genetically specified in DNA.

Unmanned aerial vehicle (UAV)-based imagery has become widely used in collecting agronomic traits, enabling a much greater volume of data to be generated in a time-series manner. As one of the cutting-edge imagery analysis tools, machine learning-based object detection provides automated techniques to analyze these imagery data. In our previous study, UAVs have been used to collect aerial photography for field trials of 233 diverse inbred lines, grown under different nitrogen treatments. Images were collected during different plant developmental stages throughout the growing season. This dataset of images has here been used in developing machine learning techniques to obtain automated tassel counts at the plot level through the season. To improve detection accuracy, we have developed an image segmentation method to remove non-tassel pixels and then feed these filtered images into machine learning algorithms. As a result, our method showed a significant improvement in the accuracy of maize tassel detection. This method can be used in future research to produce time-series counts of tassels at the plot level, and will allow for accurate estimates of flowering-related traits, such as the earliest detected flowering date and the duration of each plot's flowering period. This phenotypic data and the trait-associated genes provide new opportunities for crop improvement and to facilitate future plant breeding.

Researchers and end users using climate data face a challenge when they analyze the data they need. Data volumes are increasing very rapidly, and the ability to download all needed data is often no longer possible. Most of the climate analysis tools for research and application needs must use very large datasets, often distributed among several data centres and into a large quantity of files. This is especially true when they are stored in a federated architecture like the ESGF. One of these tools is icclim (https://github.com/cerfacs-globc/icclim ), a flexible python software package to calculate climate indices and indicators. This tool adhere as much as possible to metadata conventions such as CF, implementing also provenance information. It also aims at providing increasing support for all FAIR aspects. It is designed with performance and optimisation in mind, because the goal is to provide on-demand calculations for users. It provides the implementation of most of the international standard climate indices such as ECAD, ETCCDI, ET-SCI, including the correct methodology for calculating percentile indices using the bootstrapping method. It has been validated against R.Climdex as well (https://cran.r-project.org/web/packages/climdex.pcic/index.html ). The new 5.x version of icclim is now based on functions from the xclim python library, which was inspired by earlier versions of icclim, but using xarray and dask for data access and processing. icclim is also a candidate as the software to calculate climate indices for the C3S toolbox (https://cds.climate.copernicus.eu/cdsapp#!/toolbox ). icclim is integrated in the IS-ENES C4I 2.0 platform (https://climate4impact.eu/ ), using a Jupyter notebook collection in a SWIRRL environment (Software for Interactive Reproducible Research Labs https://gitlab.com/KNMI-OSS/swirrl ). Having access to this type of analysis tool is very useful, and seamless integration with front-ends like C4I enable the use of those tools by a larger number of researchers and end users. This project (IS-ENES3) has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N°824084.

The interpretation of ERT data requires inversion, which is challenging due to its nonlinear and ill­posed nature. We aim to solve the inverse problem using a machine learning technique that approximates the complex non-linearity using the data provided without specifying any geophysical or mathematical equations. Specifically, we make use of CNN with U-Net architecture to estimate the true subsurface resistivity model from observed apparent resistivity data (pseudosection). U-Net is selected because it works well when both input and output data are in the form of images, and in the present work, input (pseudosection) and output (subsurface resistivity model) have spatial correspondence and existing local patterns. For ERT inversion, U-Net learns by minimizing pixel mean squared error between the true and predicted subsurface resistivity images. The ERT dataset was generated by simulating a dipole­dipole survey over a 400m x 100m earth containing two anomalous blocks with varying resistivity, size, and position in a variable homogeneous background resistivity. We used a finite element approach with a triangular mesh grid for forward modelling (using pyGIMLi) and added Gaussian noise to the calculated pseudosection so that simulated data resembles the field data. The U-Net model was trained on 4000 samples (20% of training samples used for validation) and tested on 1000 samples. The architecture consisted of an encoder and decoder subnetwork connected by a bridge section. The encoder depth was 7, and each block in the encoder subnetwork consisted of two sets of convolutional and ReLU layers, followed by a max pooling layer. The decoder subnetwork consisted of a transposed convolution layer for upsampling, followed by two sets of convolutional and ReLU layers. The weights of all the convolutional layers were initialized using the He Normal initializer. We used Adam optimizer with a learning rate of 0.0001. The U-Net model was allowed to train for 100 epochs with a batch size of 128. The trained U-Net model gave good predictions of the anomalous blocks and background resistivity with a root mean square misfit of 2.86% on test data and 2% on training data. The results confirm the efficacy and effectiveness of the proposed U-Net architecture in predicting the subsurface resistivity variation from the DC resistivity data.

Recent advancements in proximal remote sensing have increased the spatial and temporal resolution of data collection, as well as the availability of these technologies for applications to precision agriculture. These sensors have allowed the collection of new and large quantities of data, which have been used to successfully determine phenotypes and parametrize crop growth models. So far, these data streams have been mostly used separately, though they contain unique structural, spatial, and spectral information. Thus, this research aims to integrate these disparate data sources to improve estimations of agronomically important crop traits. In this study, we examine two high-throughput and relatively inexpensive remote platforms: unoccupied ground vehicles (UGV) and unoccupied aerial vehicles (UAV). Data were collected on maize hybrids from the Genomes to Fields initiative over 5 years, from 2018 to 2022, in Aurora, NY. We used ground rovers to collect lidar scans, which were converted to point clouds, to construct the three-dimensional sub-canopy architecture of maize plants. Multispectral sensors, covering red, green, blue, red-edge, and near infrared (NIR) were deployed on a UAV platform to characterize maize canopies. Machine learning methods, including autoencoders, will be used to extract latent phenotypes from the lidar point clouds and multispectral images. Ultimately, these will be used to predict manually measured traits, such as yield, in order to compare the prediction accuracies of models using these measurements separately and jointly.  

This study aimed to investigate the physiological and morphological responses of maize seedlings to environmental stressors. High-throughput imaging analysis was used to characterize the stress response phenotypes of 47 distinct maize genotypes exposed to water limitation (40% field capacity), heat (38 °C), and the combination of the two stresses. RGB and NIR images were collected daily, and analyzed with open-source, open-development software PlantCV. Our investigation focused on quantitative measurements of daily area, daily water loss, evaluation of estimated water use efficiency (WUE), and near-infrared (NIR) reflectance as an estimation of water content in tissues. Quantitatively comparing two non-normal distributions, like the NIR histogram data, can be challenging but important, since metrics like median mode values often do not capture variation across a sample. One of the primary obstacles lies in defining an appropriate metric to accurately quantify the variation between two distributions. To analyze NIR reflectance, we evaluated the dissimilarity between pairwise NIR histograms by utilizing the earth mover’s distance (EMD) analysis. The EMD quantifies the difference between two NIR reflectance distributions, and therefore indirectly evaluates the difference in leaf water content for stress vs control conditions. Overall, most genotypes displayed growth reduction under drought and heat. Interestingly, specific lines exhibited heightened WUE under water limitation, suggesting response to water scarcity. EMD results showing dissimilarities in pairwise NIR reflectance of control vs drought, can be used to describe variation in dynamic changes in water content levels among these groups.

Root exudation refers to the processes by which plants release compounds called root exudates into the soil. These exudates are primarily carbon-containing compounds that interact with microbial communities in the rhizosphere. Microbial consumption of exudates reduces the concentration of the exudated compounds in the soil, causing the plant to exude more of those compounds. Currently, there is limited understanding of the interaction between plant-root exudation mechanisms and the surrounding microbial communities. Among the Sorghum Association Panel (SAP), an established and genetically characterized sorghum diversity panel, we observed a spectrum of root colors (tan, yellow, red, purple-brown, black) identical to the range of observed sorghum seed colors. Previous studies examining differentially expressed metabolites between colorful seeds showed that flavonoids and anthocyanins were higher in dark seeds than white seeds. Root color is genotype-dependent and consistent over time. We hypothesized that the observed color diversity of sorghum roots was due to differential metabolite profiles in the root exudates across genotypes. We designed an experiment to collect exudates from 15 genotypes (n=60). After three weeks of growth, sorghum roots were washed and submerged in ultrapure water for 24 hours. The hydroponic solution was filtered and incubated with methanol. The whole root system was also ground after exudation. The root exudate solutions and the ground-up roots underwent either HILIC and RPLC analysis to separate and detect polar and hydrophobic metabolites. Through metabolite profiling of root exudates, we aim to identify sorghum genotypes that more efficiently allocate carbon below ground via their root systems.

The automatic, and accurate plant phenotyping plays important role to improve the crop yield through enabling efficient plant analysis and plant breeding studies. The 3d deep learning allows automatic segmentation of plant parts from point cloud data. However, the network architecture is designed manually and performance is limited to prior experience. The aim of this study is to search for optimal 3d deep networks to perform the plant part segmentation. We perform the 3d neural architecture search by training a super network composed of candidate networks. Using the trained super network, the evolutionary searching is used to search for top performing architecture. The results demonstrate the searched architecture outperforms manually designed architectures by attaining mean IoU and accuracy of more than 90% and 96%, respectively. The searched architecture achieves more than 83% class-wise IoU for all main stem, branches, and boll class. This plant part segmentation method shows promising results and holds potential to be utilized by plant breeders for enhancing the production quality.

We investigate the robustness of 3D point-based deep learning for organ segmentation of 3D plant models against varying reconstruction quality of the surface. The reconstruction quality is quantified in two ways: 1) The number of acquisitions for partial 3D scans and 2) the amount of noise. High quality models of real rosebush plants are used to collect point clouds in a controlled simulation environment as a way to degrade surface quality systematically. We show that the well-known 3D point-based neural network PointNet++ is capable of operating effectively on low quality and corrupted data for the task of plant organ segmentation. The results indicate that investing on developing deep learning methods has the potential of advancing applications of automated phenotyping, especially for low-quality 3D point clouds of plants. Keywords: plant phenotyping, organ segmentation, robustness analysis, point-based deep learning (a) (b) Figure 1: A 3D rosebush model from ROSE-X data set: (a) point cloud; (b) triangular mesh model.

The 17th International Conference on Thermochronology (Thermo2021) was held in Santa Fe, New Mexico, on September 12-17, 2021. This bi-annual conference series evolved via the coalescence of the International Workshops on Fission Track Thermochronology, held since 1978, and the European Workshops on Thermochronology. It has become the premier forum for thermochronology practitioners and users to discuss fundamental and methodological topics and opportunities related to their science and its future. Each conference is independently organized, and a Standing Committee consisting of past organizers and other community members helps to ensure their continuation into the future. Thermo2021 was greatly affected by the COVID-19 pandemic. Normally the meeting would have been expected to draw ~250 attendees, but travel restrictions limited in-person attendance to 86, plus 21 remote presenters. Nearly all in-person participants were from the US, and only four were international. Talks and posters were distributed among five themes: (U-Th)/He; fission track; other thermochronometers; frontiers in data handling, statistics, interpretation methods, and modeling; and integration and interpretation. Although COVID-19 presented many challenges, it also allowed the Organizing Committee to adapt creatively and transform adversity into opportunity. In particular, the smaller number of attendees permitted more talks by students and early-career scientists, both within the theme sessions and in the Charles & Nancy Naeser Early Career Session. Discussion time was prioritized: at a Tuesday evening “swap meet” for ideas, in 30-40-minute time slots within each theme session, and in Friday afternoon breakouts for the first four themes and another dedicated to early career and DEI issues. These were used to identify emergent ideas and concerns across a broad range of topics, from the theory and practice of the various thermochronometric techniques, to their interpretation through thermal history modeling and other methods, to anticipated trends in data dissemination and management, to the needs of the next generation of thermochronologists, particularly in the US. Each Friday breakout designated a scribe who recorded the discussion and distributed their notes. Each group then designated one or more writers to transform the notes into text for this White Paper. Notes or early write-up versions were provided to the international thermochronology community, and feedback solicited. In addition, cross-cutting themes that occurred across multiple breakout groups were identified and compiled. This White Paper is the outcome of these efforts. We hope that it will serve as a record for the meeting, and an overview of where the predominantly US-based component of the thermochronology community considers the current state of knowledge to be and where future efforts should be directed, for developing both the science and its human infrastructure.

Detecting and locating microearthquakes from continuous waveform records is the fundamental step in microseismic processing. Dense local networks and arrays have introduced the possibility to detect large numbers of far weaker events, but when viewed on seismic records from individual stations their waveforms are often obscured by noise. Furthermore, areas of interest for microseismic monitoring often feature extremely high event rates, highlighting the limitations of traditional techniques based on phase picking and association. In order to maximise the new insights gained, we require fully automated techniques which can exploit modern recordings to produce highly complete earthquake catalogues containing few artefacts. QuakeMigrate is a new modular, open-source python package providing a framework to efficiently, automatically and robustly detect and locate microseismicity. The user inputs continuous seismic data, a velocity model or pre-calculated look-up table and list of station locations. Instead of reducing the raw waveforms to discrete time picks, they are transformed (by amplitude, frequency and/or polarisation analysis) to continuous functions representing the probability of a particular phase arrival through time. These ‘onset functions’ from stations across the network are then migrated according to a travel-time look-up table and stacked to perform a grid-search for coherent sources of energy in the subsurface. This enables detection of earthquakes at close to or below the signal-to-noise ratio at individual stations, and implicitly associates phase arrivals even at very small inter-event times. We demonstrate the flexibility and power of this approach with examples of basal icequakes detected at the Rutford Ice Stream, Antarctica, dike- and caldera-collapse induced seismicity at Bárðarbunga central volcano, Iceland, and the aftershock sequence from a M5 earthquake at Mt. Kinabalu, North Borneo. The modular nature of the workflow and wide range of automatic plotting options makes parameter choice straightforward, and robust event location uncertainty statistics facilitate filtering to produce a robust catalogue. QuakeMigrate also outputs phase picks and local magnitude estimates, with an architecture designed to promote further community-driven extension in future.

Solar energy development is land intensive and recent studies have demonstrated the negative impacts of large-scale solar deployment on vegetation and soil. Co-locating vegetation with managed grazing on utility scale solar PV sites could provide a sustainable solution to meeting the growing food and energy demands, along with providing several co-benefits. However, the impacts of introducing grazing on soil properties at vegetated solar PV sites are not well understood. To address this knowledge gap, we investigated the impacts of episodic sheep grazing on soil properties (micro and macro nutrients, carbon storage, soil grain size distribution) at six commercial solar PV sites (MN, USA) and compared that to undisturbed control sites. Results indicate that implementing managed sheep grazing significantly increased total carbon storage (10-80%) and available nutrients, and the magnitude of change correlated with the grazing frequency (1-5 years) at the study sites. Furthermore, it was found that sites that experienced consecutive annual grazing treatments benefitted more than intermittently grazed sites. The findings will help in designing resource conserving integrated solar energy and food/fodder systems, along with increasing soil quality and carbon sequestration.

Floods are the most frequent, costliest natural disasters having devastating consequences on people, infrastructure, and the ecosystem. The accurate and rapid mapping of the flooded areas becomes more crucial when floods strike densely populated cities. During flood events near real-time satellite imagery has been proven to be an efficient management tool for disaster management authorities. However one of the challenges is accurate classification and segmentation of flooded water and permanent water. Binary segmentation using the threshold split-based method is commonly used in this regard, however, the generalization ability of this method is limited due to the effects of backscatter, geographical area, and time of image collection. Recent advancements in deep learning algorithms for image segmentation has demonstrated the excellent potential of Convolutional Neural Networks(CNN) for improving flood detection, although there have been limited studies in this domain due to the lack of large scale labeled flood event dataset. In this project, we present a U-net based deep learning approach by leveraging publicly available Sentinel-1 dataset provided jointly by NASA Interagency Implementation and Advanced Concepts Team and IEEE GRSS Earth Science Informatics Technical Committee. Dataset is composed of 66,810 tiles of 256×256 pixels, distributed respectively across the training, validation and test sets and cover flood events from Nebraska, North Alabama, Bangladesh, Red River North and Florence. Specifically we proposed an Unet architecture based convolutional neural network (CNN) with a backbone of EfficientNetb7, trained against the dataset. We then evaluated the performance of the model with multiple training, testing and validation. Two evaluation methods - Intersection over Union (IOU) and F-Score are adopted to evaluate the model performance. During testing, the model achieved the meanIOU score of 75.06% and F-Score of 74.98%. We hope to further improve the performance of the network by performing hyper-parameter tuning and to develop a model which can be used for near-real-time flood mapping.

3D scans of real world objects are often represented by point clouds, creating XYZ-coordinates of individual scan points. However, unlike point clouds that are generated from CAD data, points generated from a real world scene lack information about their local context, making segmentation of the structural information contained in the data difficult. Using neural networks (e.g. PointNet) has shown promising results. However, this approach is not well suited for scans of large areas of similar objects, like e.g. a wheat field, because of limitations of the input vector size of the neural network. In addition, point clouds are often unordered, further complicating processing. Since point clouds of biological objects often contain recurring features, we propose to subdivide the point cloud into locally neighboring subsets with a fixed number of points. The collection of subsets can then be used to train neural networks. This approach preserves the original resolution of the point cloud while offering simple data augmentation concepts like creating a number of different subset collections from the same ground truth. There are several advantages to this approach, like significantly simplifying the training phase, because a single, large annotated scan can be sufficient for training, utilizing the similarity of the instances of a plant in the field.

Unmanned aerial vehicles (UAVs) provide growers and researchers with an efficient way to evaluate fields at high resolution. Flying UAVs and collecting imagery are made easily approachable through high-performance sensors that measure a wide spectrum of light and free flight software available on smartphones and tablets. In contrast to the efficiency of collecting imagery, extracting data from this imagery presents a major hurdle for researchers and growers. Current data analysis options require either an expensive subscription service or complex coding packages, effectively preventing many from utilizing remote sensing data. These solutions also are designed as a "black-box", where imagery goes in and data comes out, making customization and adaptability to the user's needs a challenge. To address these shortcomings, I developed an open-source analysis pipeline that is both approachable and robust. Starting with an orthomosaic and combining stock tools in the QGIS graphical user interface, this pipeline follows a simple step-by-step process to mask out soil and apply any user-defined index. From there the user can segment plots using a fast yet highly customizable gridding system, allowing for plot segmentation in unusual field layouts or planting regimes. This feature has been previously unsupported in many subscription and open-source programs alike. Plot-level data can then be exported for statistical analyses. Ultimately, this pipeline is aimed to attract more researchers and growers towards using remote sensing data in their research.

This study aims to enhance the accuracy and reliability of the Global Ensemble Forecast System’s (GEFS) precipitation forecasts over the Indian subcontinent using two post-processing techniques, namely the Analog method (AN) and Logistic Regression method (LR). The post-processing techniques and GEFS Numerical Weather Prediction Model (NWP) outputs were evaluated against the observed dataset using probabilistic and deterministic evaluation metrics. Results found that both the methods considerably improves short range to medium range (1-15 day) precipitation forecasts over India. Overall results showed that both the methods perform poorly during the monsoon seasons compared to other seasons. Basin analysis showed that both the methods underperform in the Western Ghats, while the performance is comparable and decent in other parts of India. Analysis of precipitation at different terciles showed that both the AN and LR methods underperforms at higher terciles compared to the lower ones. This is because the GEFS model itself was performing poorly in detecting the heavy precipitation events. The comparison of logistic regression and analog methods shows that the LR method outperforms the AN method in almost all the locations and lead times.

The tandem rise in satellite-based observations and computing power has changed the way we (can) see rivers across the Earth’s surface. Global datasets of river and river network characteristics at unprecedented resolutions are becoming common enough that the sheer amount of available information presents problems itself. Fully exploiting this new knowledge requires linking these geospatial datasets to each other within the context of a river network. In order to cope with this wealth of information, we are developing Veins of the Earth (VotE), a flexible system designed to synthesize knowledge about rivers and their networks into an adaptable and readily-usable form. VotE is not itself a dataset, but rather a database of relationships linking existing datasets that allows for rapid comparison and exports of river networks at arbitrary resolutions. VotE’s underlying river network (and drainage basins) is extracted from MERIT-Hydro. We link within VotE a newly-compiled dam dataset, streamflow gages from the GRDC, and published global river network datasets characterizing river widths, slopes, and intermittency. We highlight VotE’s utility with a demonstration of how vector-based river networks can be exported at any requested resolution, a global comparison of river widths from three independent datasets, and an example of computing watershed characteristics by coupling VotE to Google Earth Engine. Future efforts will focus on including real-time datasets such as SWOT river discharges and ReaLSAT reservoir areas.