44 research outputs found
Integration of Spaceborne LiDAR and Imaging Spectroscopy in Vegetation Classification
No full textIn the face of a dramatically changing climate, the need to model, monitor, and respond to our environment has never been greater or so nearly within our grasps. Advances in remote sensing have made possible the rise of automated methods to study vegetation at a fine detail over previously unimaginable scales. The 2018 launch of the DLR Earth Sensing Imaging Spectrometer (DESIS) coincided with the beginning of NASA's Global Ecosystem Dynamic Investigation (GEDI) mission. For the first time, high resolution spaceborne LiDAR (GEDI) was onboard the International Space Station (ISS) in tandem with hyperspectral imaging instrumentation (DESIS). This occasion presents a unique opportunity in remote sensing to obtain temporally-proximal spectral and structural information from spaceborne sources. Through the integration of these two data sources, we constructed a random forest classification model to perform successional classification on three classes over a study site in upper Michigan. The classifier was trained over distinct datasets from each instrument, then over a combined dataset utilizing data from both instruments. Over this combined dataset, the model achieved 91.7% classification accuracy, greater than the 80.2% and 88.6% accuracies achieved from either instrument in isolation. These results suggest predictive variation between spectral imaging and structural information from LiDAR can be determined algorithmically as in a random forest classifier.No embargoAcademic Major: Computer Science and Engineerin
Image-Based Plant Leaf Disease Recognition with InceptionV3 Network
No full textEngineering: Physical Sciences (The Ohio State University Denman Undergraduate Research Forum)Most traditional plant disease diagnosis strategies depend on human visual observation and inspection. However, this approach is time-consuming and requires significant human effort and expert knowledge. The recent advances in computer vision and deep learning provide a potential pathway to developing a plant disease diagnosis system that allows rapid detection of disease across large spatial areas with minimal human intervention. In this study, we developed a deep learning approach for plant leaf disease classification problems and conducted a range of experiments to quantify the performance of several state-of-the-art neural network architectures, including ResNet50, InceptionV3, and NASNet. All of the experiments were trained on the PlantVillage dataset with 54305 images in total, spanning over 38 plant disease classes. We evaluated four different performance metrics to assess each architecture: accuracy, precision, recall, and area under the curve (AUC). Our results showed that the InceptionV3 neural network architecture outperformed all other Convolutional Neural Network (CNN) architectures (ResNet50, NASNet-Large, NASNet-Mobile, MobileNet-v3-small, and MobileNet-v3-large) and produced a training accuracy of 94.14% and 97.94% over 6 epochs and 40 epochs of training, respectively. These results suggest that CNN architectures broadly, and the InceptionV3 model specifically, is capable of remote and automated plant disease detection. These results point to exciting future applications in lightweight mobile phone applications or backend workstation developments for plant leaf disease recognition problems.No embargoAcademic Major: Computer Science and Engineerin
Explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimation
No full textProximal remote sensing has the potential to provide critical information on vegetation biophysical factors that can predict land-atmosphere exchange of water and energy. Latent energy (LE) flux is traditionally estimated using process-based models which rely on vegetation parameters that change during the growing season. Data-driven models have the potential to address these issues by offering flexible predictor selection and more efficient utilization of the information in predictor sets. These models require careful choice of predictors to avoid redundancy and allow robust cross-validation. In this study we present a systematic and comprehensive evaluation of machine learning (ML) models to assess the capability of meteorological and proximal sensing data for predicting LE at a half-hourly temporal resolution across multiple growing seasons for an agricultural system. The results presented here demonstrate that a model using four environmental predictors in combination with two proximal sensing variables can capture 88 % of the variability in LE. ML models using only three predictors (one meteorological and two proximal remote sensing) captured 81 % of LE variability, offering the best trade-off between performance and complexity. An ML model utilizing only two predictors, one proximal remote sensing variable and downwelling radiation, captured 77 % of LE variability. These results demonstrate the power of proximal remote sensing and meteorological observations to estimate land-atmosphere water vapor exchange, providing a solution where more direct methods such as eddy covariance are not available and for evaluations of agronomic management and genotypic variations.This article is published as Gaur, Srishti, Guler Aslan-Sungur, Andy VanLoocke, and Darren T. Drewry. "Explainable machine learning to quantify the value of proximal remote sensing in latent energy flux estimation." Agricultural Water Management 317 (2025): 109643. https://doi.org/10.1016/j.agwat.2025.109643DTD and SG acknowledge support from the National Science Foundation Cyber-Physical Systems Program (Award 1954556). DTD also acknowledges support from the National Aeronautics and Space Administration (Award #80NSSC20K1789), the College of Food, Agricultural and Environmental Sciences and the Translational Data Analytics Institute at Ohio State University. This work was supported, in part, by Hatch funds from the USDA National Institute of Food and Agriculture (Hatch Project OHO01509) and by the DOE Center for Advanced Bioenergy and Bioproducts Innovation, U.S. Department of Energy, Office of Science, Biological and Environmental Research Program under Award Number DE-SC0018420
Chlorophyll Can Be Reduced in Crop Canopies with Little Penalty to Photosynthesis
No full textThe hypothesis that reducing chlorophyll content (Chl) can increase canopy photosynthesis in soybeans was tested using an advanced model of canopy photosynthesis. The relationship between leaf Chl, leaf optical properties, and photosynthetic biochemical capacity were measured in 67 soybean accessions showing large variation in leaf Chl. These relationships were integrated into a biophysical model of canopy-scale photosynthesis to simulate the intercanopy light environment and carbon assimilation capacity of canopies with WT, a Chl-deficient mutant (Y11y11), and 67 other mutants spanning the extremes of Chl to quantify the impact of variation in leaf-level Chl on canopy-scale photosynthetic assimilation and identify possible opportunities for improving canopy photosynthesis through Chl reduction. These simulations demonstrate that canopy photosynthesis should not increase with Chl reduction due to increases in leaf reflectance and non-optimal distribution of canopy nitrogen. However, similar rates of canopy photosynthesis can be maintained with a 9% savings in leaf nitrogen resulting from decreased Chl. Additionally, analysis of these simulations indicate that the inability of Chl reductions to increase photosynthesis arises primarily from the connection between Chl and leaf reflectance and secondarily from the mismatch between the vertical distribution of leaf nitrogen and the light absorption profile. These simulations suggest that future work should explore the possibility of using reduced Chl to improve canopy performance by adapting the distribution of the "saved" nitrogen within the canopy to take greater advantage of the more deeply penetrating light.This is a manuscript of an article published as Walker, Berkley J., Darren T. Drewry, Rebecca A. Slattery, Andy VanLoocke, Young B. Cho, and Donald R. Ort. "Chlorophyll can be reduced in crop canopies with little penalty to photosynthesis." Plant physiology (2017). doi: 10.1104/pp.17.01401.</p
A study of the ability of two groups of freshmen students at Morehouse College to use the Trevor Arnett Library, 1953
No full textFinite-Element Tree Crown Hydrodynamics model (FETCH) using porous media flow within branching elements - a new representation of tree hydrodynamics
No full textEstimating transpiration and water flow in trees remains a major challenge for quantifying water exchange between the biosphere and the atmosphere. We develop a finite element tree crown hydrodynamics (FETCH) model that uses porous media equations for water flow in an explicit three‐dimensional branching fractal tree‐crown system. It also incorporates a first‐order canopy‐air turbulence closure model to generate the external forcing of the system. We use FETCH to conduct sensitivity analysis of transpirational dynamics to changes in canopy structure via two scaling parameters for branch thickness and conductance. We compare our results with the equivalent parameters of the commonly used resistor and resistor‐capacitor representations of tree hydraulics. We show that the apparent temporal and vertical variability in these parameters strongly depends on structure. We suggest that following empirical calibration and validation, FETCH could be used as a platform for calibrating the “scaling laws” between tree structure and hydrodynamics and for surface parameterization in meteorological and hydrological models
Non-invasive diagnosis of wheat stripe rust progression using hyperspectral reflectance
No full textEl autor(es) declara haber recibido apoyo financiero para la investigación, autoría y/o publicación de este artículo. DD y JC reconocen el apoyo de la Fundación Nacional de Ciencias de EE. UU. (Premios n.º 2239877 y 1954556). DD también agradece el apoyo de la Administración Nacional de Aeronáutica y del Espacio (Premio n.° 80NSSC20K1789), el Departamento de Agricultura de los Estados Unidos (Premio USDA n.° 2023-67013-39619), la Facultad de Ciencias Alimentarias, Agrícolas y Ambientales y el Instituto de Análisis de Datos Traslacionales de la Universidad Estatal de Ohio. NC reconoce el financiamiento recibido de la Agencia Nacional de Investigación y Desarrollo de Chile (ANID, Chile), subvenciones PAI 77190085, FONDECYT-11220889 y ATE-220001. Este trabajo fue financiado, en parte, por fondos Hatch del Instituto Nacional de Alimentación y Agricultura del USDA (Proyecto Hatch OHO01509)
Incorporating Plant Phenology Dynamics in a Biophysical Canopy Model
No full textThe Multi-Layer Canopy Model (MLCan) is a vegetation model created to capture plant responses to environmental change. Themodel vertically resolves carbon uptake, water vapor and energy exchange at each canopy level by coupling photosynthesis, stomatal conductance and leaf energy balance. The model is forced by incoming shortwave and longwave radiation, as well as near-surface meteorological conditions. The original formulation of MLCan utilized canopy structural traits derived from observations. This project aims to incorporate a plant phenology scheme within MLCan allowing these structural traits to vary dynamically. In the plant phenology scheme implemented here, plant growth is dependent on environmental conditions such as air temperature and soil moisture. The scheme includes functionality that models plant germination, growth, and senescence. These growth stages dictate the variation in six different vegetative carbon pools: storage, leaves, stem, coarse roots, fine roots, and reproductive. The magnitudes of these carbon pools determine land surface parameters such as leaf area index, canopy height, rooting depth and root water uptake capacity. Coupling this phenology scheme with MLCan allows for a more flexible representation of the structure and function of vegetation as it responds to changing environmental conditions
Diagnosing model error in canopy‐atmosphere exchange using empirical orthogonal function analysis
No full textPhotoSpec - Comprehensive Ground-Based Studies of Solar-Induced Chlorophyll Fluorescence: From the New Methods for Measurements of Photosynthesis from Space Study
No full textThe major goal of the PhotoSpec program was to develop a set of robust ground-based spectrometers that meet the measurement requirements to retrieve solar-induced chlorophyll fluorescence by exploiting solar Fraunhofer lines
