Article
Agronomy
Cheyenne Lei, Jiquan Chen, G. Philip Robertson
Summary: Changes in land surface albedo can impact ecosystem energy balance and climate. This study in southwest Michigan, USA, explored the albedo of six bioenergy cropping systems and found significant variations among different crops, with cellulosic crops being more reflective. Further management interventions are needed to maximize the positive climate benefits of bioenergy crops.
GLOBAL CHANGE BIOLOGY BIOENERGY
(2023)
Article
Engineering, Geological
Aali Pant, G. V. Ramana
Summary: This study proposes the prediction of pullout interaction coefficient of geogrids using data-driven machine learning regression algorithms, primarily focusing on extreme gradient boosting (XGBoost) method. The XGBoost model shows significantly superior and robust prediction compared to the random forest (RF) model, with an accuracy of 85% and 77% respectively. The importance analysis identifies normal stress as the most significant factor influencing the pullout interaction coefficients.
GEOTEXTILES AND GEOMEMBRANES
(2022)
Article
Forestry
Matts Lindbladh, Johan Elmberg, Per -Ola Hedwall, Emma Holmstrom, Adam Felton
Summary: Retention forestry, specifically green tree retention, has long-term positive effects on avian diversity, even several decades after final felling in conifer dominated production stands. The presence of broadleaf green tree retention can benefit bird assemblages by increasing their abundance, species richness, and supporting important guilds such as broadleaf-associated birds and cavity nesters. However, factors like the number and species composition of retained trees, as well as their environmental context within the stand, may affect the effectiveness of green tree retention.
FOREST ECOLOGY AND MANAGEMENT
(2022)
Article
Geosciences, Multidisciplinary
Wengang Zhang, Chongzhi Wu, Haiyi Zhong, Yongqin Li, Lin Wang
Summary: This study applies XGBoost and RF methods to predict the USS of soft clays, showing that these approaches outperform traditional methods. Bayesian optimization is used to determine model hyperparameters, leading to more accurate and robust models.
GEOSCIENCE FRONTIERS
(2021)
Article
Green & Sustainable Science & Technology
Odey Alshboul, Ali Shehadeh, Ghassan Almasabha, Ali Saeed Almuflih
Summary: Accurate prediction of green building costs is crucial for decision-making and management. This study presents machine learning-based algorithms for cost prediction and evaluates their accuracy.
Article
Construction & Building Technology
Yifang Chen, Feng Li, Siqi Zhou, Xiao Zhang, Song Zhang, Qiang Zhang, Yijie Su
Summary: The Ground penetrating radar (GPR) is a non-destructive and continuous electromagnetic (EM) monitoring technology used for density detection of asphalt pavement. This study introduced the random forest (RF) and extreme gradient boosting (XGBoost) approaches combined with Bayesian hyper-parameter optimization (BHPO) and cross-validation for density prediction. The results showed that the ML algorithms significantly improved the accuracy of the GPR system in predicting pavement density.
CONSTRUCTION AND BUILDING MATERIALS
(2023)
Article
Engineering, Civil
Hoang D. Nguyen, Gia Toai Truong, Myoungsu Shin
Summary: This paper presents the application of XGBoost in predicting the punching shear resistance of reinforced concrete interior slabs without shear reinforcement. The XGBoost model showed the best prediction performance compared to other machine learning models, with effective depth identified as the most important input variable in punching shear prediction.
ENGINEERING STRUCTURES
(2021)
Article
Geosciences, Multidisciplinary
Wengang Zhang, Yuwei He, Luqi Wang, Songlin Liu, Xuanyu Meng
Summary: This study aims to evaluate the effects of random forest (RF) and extreme gradient boosting (XGBoost) models on landslide susceptibility and compare their applicability in a landslide-prone area in Fengjie County, Chongqing, China. Field investigation data of 1624 landslides from 1980 to 2020 were collected, and a geospatial database of 16 conditional factors was constructed. RF and XGBoost models were established using a complete dataset of non-landslide points. The results show that the RF model performs better with a higher AUC value, accuracy, and F-score compared to XGBoost.
GEOLOGICAL JOURNAL
(2023)
Article
Materials Science, Multidisciplinary
Seungro Lee, Joonhee Park, Naksoo Kim, Taeyong Lee, Luca Quagliato
Summary: This paper presents a machine learning methodology that can learn from simulation results, experimental data, or sensor signals, and is capable of predicting and optimizing specific user-defined process and design parameters. The methodology utilizes an enhanced Extreme Gradient Boosting (XGB) algorithm and a metaheuristic search algorithm based on Differential Evolution (DE) architecture for optimization.
MATERIALS & DESIGN
(2023)
Article
Environmental Sciences
Meng Wang, Xue Li, Mei Lei, Lunbo Duan, Huichao Chen
Summary: The petrochemical industry is a key industry in soil pollution, with significant effects on human health and the ecological environment. This study developed an efficient method using the XGBoost algorithm for identifying health risks in the petrochemical industry. The results showed that XGBoost outperformed other models in health risk identification.
ECOTOXICOLOGY AND ENVIRONMENTAL SAFETY
(2022)
Article
Forestry
Benjamin O. Knapp, Calvin J. Maginel, Bradley Graham, John M. Kabrick, Daniel C. Dey
Summary: Frequent prescribed burning is commonly used in the eastern United States to restore and maintain open forest ecosystems. However, this study found that it has multiple effects on tree recruitment, including reducing the source population and increasing mortality or top-kill probability. The abundance of scarlet oak decreased while blackgum increased in the burned area.
FOREST ECOLOGY AND MANAGEMENT
(2022)
Article
Agriculture, Multidisciplinary
Florian Huber, Artem Yushchenko, Benedikt Stratmann, Volker Steinhage
Summary: Accurate prediction of crop yield before harvest is crucial for crop logistics, market planning, and food distribution worldwide. Remote sensing satellite images provide a cheap and reliable way to obtain data for yield prediction, while Deep Learning approaches dominate the field but have limitations. This study proposes using Extreme Gradient Boosting (XGBoost) with processed remote sensing images for yield prediction, showing promising accuracies compared to Deep Learning methods for soybean yield prediction in the United States.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Engineering, Electrical & Electronic
Zhongzong Yan, He Wen
Summary: This article introduces an electricity theft detector based on extreme gradient boosting (XGBoost), which preprocesses metering data and trains a classification model to detect abnormal electricity behavior. Experimental results show that the method can detect electricity theft with higher accuracy or lower false-positive rate.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Ecology
Jessica L. Schedlbauer, Jason Miller
Summary: Anthropogenic disturbance has caused forest fragmentation and modified the soil carbon cycling in forest edges, resulting in increased soil respiration rates, elevated soil temperature, and decreased soil moisture in edge environments. However, the study found that soil carbon storage and decomposition rates remained constant along the edge to interior gradients.
Article
Computer Science, Interdisciplinary Applications
Sajjad Mirzaei, Mehdi Vafakhah, Biswajeet Pradhan, Seyed Jalil Alavi
Summary: This study implemented extreme gradient boosting (EGB) method for flood susceptibility modelling and compared its performance with three advanced benchmark models. Results showed that Random Forest (RF) model and Extreme Gradient Boosting had the best performance. The study found that factors such as distance from rivers have important influence on flood susceptibility mapping, and recommended the application of RF and EGB models for such studies.
EARTH SCIENCE INFORMATICS
(2021)
Article
Ecology
Matthew C. Reeves, Brice B. Hanberry, Hailey Wilmer, Nicole E. Kaplan, William K. Lauenroth
Summary: Through the creation of the RPMS database, trends and variations of ANPP in different types of grasslands in the Great Plains region have been successfully quantified. Significant increases in ANPP have been observed across all major grassland types since 1984, particularly in the northern mixed-grass prairie.
RANGELAND ECOLOGY & MANAGEMENT
(2021)
Article
Ecology
Jacqueline P. Ott, Brice B. Hanberry, Mona Khalil, Mark W. Paschke, Max Post van der Burg, Anthony J. Prenni
Summary: Energy development and production can have significant ecological impacts on grassland ecosystems, affecting soil, vegetation, wildlife, air, and water quality. Careful planning, construction practices, and monitoring during energy production operations can help mitigate these effects and protect grassland health while meeting energy needs. Continued research on mitigation measures is necessary to establish best management practices for sustainable energy development.
RANGELAND ECOLOGY & MANAGEMENT
(2021)
Article
Biodiversity Conservation
Brice B. Hanberry
Summary: Land use and fire exclusion have had significant impacts on ecosystems worldwide, leading to alternative ecosystem states. Examples from the southeastern United States demonstrate the decline of fire-dependent open pine and pine-oak forests, with dynamics shifting towards replacement forests.
Article
Forestry
Brice B. Hanberry
Summary: Eastern redcedar is increasing in density and expanding in range to the west in the eastern United States, while western Juniperus species are also increasing but not expanding in range. The intersection of eastern redcedar and western Juniperus species creates a new assemblage. The predicted climate envelopes suggest that major climate conditions have been suitable for centuries. The southern western Juniperus species and Rocky Mountain juniper have the greatest potential for intersection with eastern redcedar.
Article
Ecology
Brice B. Hanberry, Reed F. Noss
Summary: Historically, the eastern United States had grasslands with limited tree presence surrounded by open oak and pine forests. However, these open ecosystems have mostly been lost due to agriculture and fire exclusion. Using random forests and C5.0 classifiers, we identified the potential extent of fire-maintained grasslands in the eastern United States based on historical grassland maps, topography, and wind speed. Potential grasslands covered 27 million hectares, or 14% of the eastern United States, although these locations may have also been savannas or open woodlands historically. Currently, potential grassland locations are primarily used for crops, pasture, forests, and developed land.
Article
Ecology
Brice B. Hanberry
Summary: A common global definition of urban populations is lacking, making it difficult to compare urban status across countries. The use of population models and adjusted population density thresholds can provide a pathway for equal comparison of urban status.
LANDSCAPE AND URBAN PLANNING
(2022)
Article
Geography, Physical
Brice B. B. Hanberry
Summary: This study analyzed pollen reconstructions of tree genera in North America since glaciation and predicted the paleoecological ranges of boreal and temperate tree species from 20,000 years ago to 5,000 years ago. The study found that climate change was the main factor driving the distribution changes of tree species and identified key climate variables that influenced species distribution. The movement speed of species gradually decreased over the past 10,000 years and retreated southward between 7,000 and 5,000 years ago.
JOURNAL OF QUATERNARY SCIENCE
(2023)
Article
Ecology
Brice B. Hanberry
Summary: The study found that there is a positive correlation between human population density and greenhouse density with the number of non-native plant species, while land use disturbance and transportation have a smaller impact on non-native plant species richness.
ECOLOGICAL PROCESSES
(2023)
Article
Forestry
Brice B. Hanberry, Jonathan M. Stober, Don C. Bragg
Summary: Longleaf pine used to dominate the coastal plains of the southeastern United States, but due to overcutting, land clearing, and other silvicultural influences, its population has significantly declined and been replaced by other pine and oak species.
Article
Environmental Sciences
Brice B. Hanberry
Summary: Climate classifications provide inference about general vegetation types based on climate visualization. This study incorporates temperature and aridity information into all climate classes and adds a hypertropical class to better detect warming and drying. By comparing current climate to past climate, the study reveals that the hypertropical class has experienced the greatest change in area and the greatest increase in percentage of arid classes.
Article
Multidisciplinary Sciences
Stephen J. Tulowiecki, Brice B. Hanberry, Marc D. Abrams
Summary: Researchers assessed the proximity of tree-ring fire-scar records (FSRs) to past Native American settlements in the eastern US and found a correlation between fire frequency and distance to Indigenous settlements.
SCIENTIFIC REPORTS
(2023)
Article
Environmental Studies
Brice B. Hanberry
Summary: Tree distributions and densities in North America have been changing since Euro-American settlement. Fire-tolerant species have decreased, while fire-sensitive, successional species have increased. Tree species have also expanded westward into the Great Plains grasslands. These changes may be attributed to climate change.
Article
Environmental Studies
Brice B. B. Hanberry
Summary: The increased abundance of historically rare native tree species is closely associated with land-use change. By modeling the agricultural area and the composition of tree genera and species, the study found that the increase in agricultural area is correlated with the increase in certain tree species.
Article
Environmental Studies
Brice B. Hanberry
Summary: Long-term observations reveal relationships among changes in vegetation, climate, and land use. The study found that increases in tree density in prairie areas coincided with the expansion of agricultural land area to over 25% in the region. Inconsistent changes in moisture availability did not show significant correlations with comprehensive tree increases.