Ensemble Modeling Approach Research Articles

Predicting COD and TN in A2O+AO Process Considering Influent and Reactor Variability: A Dynamic Ensemble Model Approach

The prediction of the chemical oxygen demand (COD) and total nitrogen (TN) in integrated anaerobic–anoxic–oxic (A2O) and anoxic–oxic (AO) processes (i.e., A2O+AO process) was achieved using a dynamic ensemble model that reflects the dynamics of wastewater treatment plants (WWTPs). This model effectively captures the variability in the influent characteristics and fluctuations within each reactor of the A2O+AO process. By employing a time-lag approach based on the hydraulic retention time (HRT), artificial intelligence (AI) selects suitable input (i.e., pH, temperature, total dissolved solid (TDS), NH3-N, and NO3-N) and output (COD and TN) data pairs for training, minimizing the error between predicted and observed values. Data collected over two years from the actual A2O+AO process were utilized. The ensemble model adopted machine learning-based XGBoost for COD and TN predictions. The dynamic ensemble model outperformed the static ensemble model, with the mean absolute percentage error (MAPE) for the COD ranging from 9.5% to 15.2%, compared to the static ensemble model’s range of 11.4% to 16.9%. For the TN, the dynamic model’s errors ranged from 9.4% to 15.5%, while the static model showed lower errors in specific reactors, particularly in the anoxic and oxic stages due to their stable characteristics. These results indicate that the dynamic ensemble model is suitable for predicting water quality in WWTPs, especially as variability may increase due to external environmental factors in the future.

Robust Decision Support System for Stress Prediction Using Ensemble Techniques

The prevalence of stress among students has become a significant concern, impacting academic performance and overall well-being. Low-level stress can be beneficial as it helps us maintain focus on our daily routines, enabling us to perform tasks efficiently. However, severe stress can lead to heart disease, diabetes, aches, pains, and numerous other serious health issues. To address this issue, this research presents a comprehensive Decision Support System (DSS) tailored for predicting student stress levels using ensemble techniques. The goal of this project is to develop an ensemble model method of voting to classify the student dataset. An ensemble model is a combination of individual machine-learning algorithms that enhance the accuracy of the model by taking different approaches to the problem and their summary output through base classifiers. Extensive experimentation and validation are conducted using real-world student data to assess the effectiveness of the proposed DSS. In this research, we employed an ensemble modeling voting approach to classify the student dataset. Ensemble models integrate individual machine learning algorithms to enhance accuracy by leveraging the combined output of the base classifiers. Gradient Boosting Classifier (GBC), Random Forest (RF), Bagging Classifier (BC), and Extra Trees Classifier (ETC) served as base learners to construct a voting ensemble model for student dataset classification. Our findings revealed that the voting classifier consisting of Support Vector Machine (SVM), Random Forest (RF), Extra Trees Classifier (ETC), and Bagging Classifier (BC) yielded the highest accuracy at 94.78%, while GBC achieved 84.39% accuracy, RF attained 94.26% accuracy, ETC achieved 94.02% accuracy, and BC obtained 93.65% accuracy. The ensemble modeling approach significantly outperformed individual machine learning algorithms, demonstrating superior accuracy by mitigating variance and classification errors. Implementing such a system not only enhances the student assessment process in the education sector but also fosters interdisciplinary research collaboration across diverse fields within the region.

Climate change drives fish communities: Changing multiple facets of fish biodiversity in the Northwest Pacific Ocean

Global marine biodiversity is experiencing significant alterations due to climate change. Incorporating phylogenetic and functional diversity may provide novel insights into these impacts. This study used an ensemble model approach (random forest and boosted regression tree), to predict the habitat distribution of 47 fish species in the Northwestern Pacific under contemporary (2000–2014) and future scenarios (2040–2050, 2090–2100). We first examined the relationship between eleven functional traits and habitat changes, predicting the spatial distribution of functional traits within fish communities. A significant correlation was observed between temperature preference and habitat changes, highlighting the vulnerability of cold-water species and potential advantages for warm-water species in the future. Moreover, fish communities exhibited a spatial gradient distribution with southern regions characterized by shorter-lived and earlier maturity, contrasting with longer-lived and later maturity species in the north. Secondly, to assess the impact of climate change on marine biodiversity, we explored the taxonomic, phylogenetic, and functional diversity under contemporary and future scenarios, revealing higher indices in the East China Sea (ECS) and the coastal sea of Japan, with the Taiwan Strait emerging as a contemporary biodiversity hotspot. In future scenarios, the three biodiversity indices would decline in the Yellow Sea and ECS, but increase in the sea beyond the continental shelf, coastal sea of Hokkaido, and Sea of Okhotsk. Lastly, we explored processes and mechanisms in the change of community composition. By quantifying β-diversity, we identified species loss (nestedness) as the primary driver of fish community change by 2040–2050, with species replacement (turnover) predicted to become dominant in the far future. Our results explore the potential changes in multiple facets of fish biodiversity, providing crucial insights for policymakers aiming to protect fish resources and biodiversity.

Hydro-geomorphological assessment of culvert vulnerability to flood-induced soil erosion using an ensemble modeling approach

Intense precipitation events pose growing threats to forest infrastructure causing flooding, and soil erosion and deposition, creating bottlenecks at road-stream crossing structures (RSCS). We describe a hillslope-scale ensemble hydro-geomorphological vulnerability assessment integrating geospatial Streambank Erosion Vulnerability Assessment (SBEVA), Modified Revised Soil Loss Equation (MRUSLE), and process-based Water Erosion Prediction Project (WEPP) model into an ensemble hydro-geomorphologic vulnerability index (EHVI) for USDA Forest Service (USFS) managed 194 road-culverts at the Hubbard Brook Experimental Forest (HBR-EF) in New Hampshire, USA. The results revealed that five and one culvert with diameters of 0.46m and 0.61m, respectively, have extreme EHVI values between 4 and 5, and fifteen and three culverts with diameters of 0.46m and 0.61m, respectively, have severe EHVI values between 3 and 4, some of which were previously identified as hydrologically vulnerable (undersized) to floods. This knowledge will inform USFS efforts to improve the resilience of the RSCS and protect aquatic habitats.

Current and potential distribution of the invasive apple snail, Pomacea canaliculata in Eastern Africa: evidence from delimiting surveys and modelling studies

The invasive apple snail Pomacea canaliculata has become a significant concern in invaded habitats beyond its native range. It was reported in Kenya in 2020 invading one of the largest rice-producing schemes, the Mwea irrigation scheme. Delimiting surveys were conducted across five key rice-producing schemes (Mwea, Bura, Hola, Ahero and West Kano) in Kenya to establish the extent of the invasion and develop effective quarantine and management strategies within the Mwea scheme and other risk areas. Additionally, the ensemble model approach was used to model the potential distribution of P. canaliculata in Eastern Africa (as defined by the United Nations Geoscheme). Over 80% of the Mwea scheme was infested with P. canaliculata, an expansion from the initial infestation point (Ndekia). The mean number of adults/m2 and egg clutches/m2 were 8.4 ± 0.9 (SEM) and 7.7 ± 1.4 (SEM), respectively, with varying densities across sections. No adults or eggs of P. canaliculata were found in the four schemes outside the Mwea scheme. The model predicted high suitability for P. canaliculata in the southwest of Kenya,and in coastal areas, with all surveyed areas marked as highly suitable.. Regionally, high-risk areas include Malawi, Madagascar, and Uganda. Mozambique, Tanzania, and Ethiopia showed localised areas of high suitability. Conversely, Sudan, South Sudan, Eritrea, Djibouti and Somalia were largely unsuitable for P. canaliculata. Given the potential for further spread, strict quarantine measures are essential to prevent the spread of P. canaliculata in Kenya and its introduction to uninvaded regions of Eastern Africa. Alongside this, implementing IPM e strategies is crucial for effective pest management and the protection of agricultural ecosystems.

Vertical Displacements and Sea‐Level Changes in Eastern North America Driven by Glacial Isostatic Adjustment: An Ensemble Modeling Approach

AbstractGlacial isostatic adjustment (GIA) describes the response of the solid Earth, oceans, and gravitational field to the spatio‐temporal evolution of ice sheets during a glacial cycle. Present‐day vertical displacements and sea‐level changes vary throughout eastern North America in response to the melting of the Laurentide Ice Sheet following the Last Glacial Maximum. We use the open‐source software SELEN4.0 (a SealEveL EquatioN solver) to investigate the influence of GIA on vertical land motions and sea‐level changes in eastern North America. Further, we evaluate the uncertainties associated with the lithospheric thickness and viscosity structure using an ensemble modeling approach (129,956 total simulations). We identify the best‐fitting rheological profiles by comparing modeled vertical displacements to vertical velocity rates derived from Global Positioning System (GPS). We find a general pattern of subsidence (causing accelerated relative sea‐level rise) in the eastern United States region and uplift (causing relative sea‐level fall) in the eastern Canada region consistent with previous studies for two tested ice sheet models (ICE‐6G(VM5a) and ICE‐7G(VM7)). Overall, we find lower rates of modeled vertical displacement using ICE‐6G(VM5a) compared with ICE‐7G(VM7), which produces lower residuals when compared with the GPS‐derived vertical velocity rates. Our ensemble analysis identifies adjustments to the nominal VM5a and VM7 viscosity models that improve fits to the GPS‐imaged vertical velocity rates throughout eastern North America and on the North American Atlantic Coast. The differences in our best‐fitting models for inland versus coastal regions highlight the importance of exploring lateral viscosity variations for GIA modeling throughout North America and elsewhere.

Revealing suitable habitats for Juniperus procera and Olea europaea tree species in the remnant dry Afromontane forests of Ethiopia: Insights from ensemble species distribution modeling approach.

Human activities and climate change pose a significant threat to the dry Afromontane forests in Ethiopia, which are essential for millions of people both economically and ecologically. In Ethiopia, trees are planted elsewhere even if they are not likely to be well suited to the area. This study aims to identify the suitable habitat for the most exploited Juniperus procera (J. procera) and Olea europaea (O. europaea) tree species in northern Ethiopia. As inputs, least correlated temperature, moisture, soil, and topographic variables were selected through a stepwise procedure. The study evaluated five individual and ensemble models using the area under the curve (AUC) and true skill statistic (TSS) values. The ensemble model outperformed with mean AUC of 0.95 and TSS of 0.78 for J. procera, while securing the second position for O. europaea with an AUC of 0.88 and TSS of 0.71. Climatic factors emerged as the most influential, followed by soil and topography. Suitable areas for both species were found when Isothermality (Bio3) values range from 52% to 62%, temperature seasonality (Bio4) of 16-29°C. Moreover, well drained soils with soil texture not heavier than sandy clay, and soil organic carbon ranging from 5 to 42 g kg-1 were found suitable. The optimal suitable altitude for J. procera and O. europaea was determined to be 2200-2600 and 2100-2500 m.a.s.l., respectively. The suitable areas for J. procera and O. europaea were estimated to be 3130 and 3946 km2, respectively. Furthermore, potential plantation areas were identified beyond Desa'a and Hugumbirda Grat-Kahsu protected forests, covering 2721 km2 (86.9%) for J. procera and 3576 km2 (90.6%) for O. europaea. These findings hold significance for the conservation and sustainable management of these valuable tree species in northern Ethiopia. We recommend implementing a similar approach for other locally restricted dry Afromontane tree species with wider potential distribution.

Evaluating seismic resilience of steel buildings: integrating soil-structure interaction and ensemble modeling approaches

Evaluating seismic resilience of steel buildings: integrating soil-structure interaction and ensemble modeling approaches

Present and future distribution of the deep-sea habitat-forming sponge - Pheronema carpenteri ( ) in a changing ocean

Sponges play vital roles in the ecosystem function of the deep sea. Some species, such as the birds' nest sponge Pheronema carpenteri, can form highly structured and dense habitats (i.e., aggregations), which contribute to the increase of nearby biodiversity. Climate change is expected to have a pronounced impact on the deep sea, particularly on Vulnerable Marine Ecosystems such as those formed by the glass sponge Pheronema carpenteri. These ecosystems are especially vulnerable to climate change and other anthropogenic activities since they are formed by sensitive species with slow growth rates and limited dispersal capability, which can hinder their adaptive capability and recovery after disturbance. The impact that climate change will have on Pheronema carpenteri remains unclear, although it is expected to influence the species' available suitable habitat and distribution range. The aim of this study was to predict the distribution of the glass sponge Pheronema carpenteri both for present day and under several future climate scenarios in the North Atlantic. An ensemble modelling approach was employed, combining Maximum Entropy, Generalized Additive Models and Random Forest techniques. Changes in available suitable habitat were projected to present day and to three future climatic scenarios (RCP 2.6, RCP 4.5 and RCP 8.5). Depth, temperature and dissolved oxygen were identified as the key predictor variables of habitat suitability, which patterns suggest a strong influence of the Mediterranean Outflow Water in shaping the present day distribution of the species, particularly in the eastern North Atlantic. Our results indicate a potential expansion of available suitable habitat in the northernmost region of the study area, with a contraction at lower latitudes, more prominent in the Portuguese archipelago of the Azores. Under the worst-case scenario (RCP 8.5), the area of suitable habitat will likely double compared to present, occupying approximately 6% of the total study area. The management and conservation of areas where Pheronema aggregations can occur should be articulated between different countries, particularly in the Northeast Atlantic since, cumulatively, most of Pheronema's climate refugia occurs within their EEZs. Nonetheless, a significant proportion of the species' climate refugia is located in areas within the High Seas (i.e., Rockall plateau).

High‐Resolution Ensemble Modelling of Coral Distributions in the Northern Gulf of Mexico Based on Geomorphometry: Coral Diversity and Benthic Habitat Fragmentation From Oil and Gas Infrastructure to Inform Spatial Planning

ABSTRACTThe northern Gulf of Mexico is home to several species of corals that provide a wide range of ecosystem services to other organisms. Oil and gas infrastructure, such as platforms and pipelines, form an extensive network throughout the northern Gulf of Mexico. Detrimental impacts associated with oil and gas exploration and extraction have been recorded in this area at depths where corals are found. Due to these ecosystems' vulnerability to long‐term impacts, it is necessary to determine areas of interest that would benefit from further exploration and informed spatial planning. This study aimed to identify potential areas of interest for coral studies in the northern Gulf of Mexico. Ensemble species distribution models for 13 species of corals including scleractinians, black corals, and octocorals were produced based on seafloor characteristics and combined to identify areas with relatively higher coral diversity potential than others. The ensemble modelling approach produced robust outputs, as evaluated by the area under the curve, Cohen's kappa coefficient, sensitivity, specificity and the proportion of correct predictions. The proximity of suitable habitat to active and proposed oil and gas infrastructure was evaluated; this spatial analysis showed that oil and gas infrastructures potentially impact 23.5% of all predicted suitable coral habitat in the study area and contribute to benthic habitat fragmentation. Twelve areas of interest greater than 100 km2 and located outside a 4‐km zone of potential influence from oil and gas infrastructure were delineated and deemed of interest for further exploration and spatial planning, and hypothetical prioritization scenarios for spatial planning are presented. The maps produced can inform discussions among stakeholders to reach the best spatial planning outcomes while considering other ecological, social, economic and governance factors.

Senecio inaequidens DC. will thrive in future climate: A case study in a Mediterranean biodiversity hotspot

Monitoring the expansion of invasive non-native plants under current and future climatic conditions is crucial for understanding biodiversity threats, addressing the ecological impact, and developing effective management strategies. This study focuses on modelling the expansion and distribution of Senecio inaequidens DC. on the island of Sardinia (Italy) to address these environmental challenges. The objectives were to identify bio-climatically suitable areas under current conditions, project potential future distribution, and evaluate invasion dynamics on the island to localize suitable areas for effective management strategies.Species data were collected from 1991 to the present, supplemented by global databases and analyzed using an ensemble species distribution model approach. This approach utilized presence data, high-resolution current bioclimatic variables (40 m2), developed explicitly for our study area, and two future scenarios derived from the newly Coupled Model Intercomparison Project Phase 6 (CMIP6) under Shared Socio-economic Pathways (SSP2–4.5 and SSP5–8.5 for 2040 and 2060).The ensemble model's findings suggest a close alignment between the currently documented occurrences of S. inaequidens and its bio-climatically suitable habitats in Sardinia. Moreover, predictions indicate high bioclimatic suitability for S. inaequidens in the western and southwestern coastal regions, contrasting with its known occurrences at higher altitudes. Notably, the model also forecasts high bio-climatic suitability across most small islands surrounding the region and in central-east Sardinia, potentially indicating habitats at lower altitudes compared to current records.Under the SSP2–4.5 scenario, suitable areas are expected to nearly double by 2040 and more than double by 2060, compared to current conditions. Under the SSP5–8.5 scenario, the increase in suitable habitats is projected to be about 83.31% by 2040 and more than double by 2060. These results highlight the species' ability to thrive under climate change, with a more pronounced range expansion under the pessimistic management regime (SSP5–8.5) than under the intermediate one (SSP2–4.5), particularly in the central region of the island. This expansion under the more severe management scenario is particularly alarming as it reflects limited implementation of environmental management policies.The study underscores the potential ecological risk posed by S. inaequidens due to its potential range expansion and ability to invade different habitat types, from coastal regions to mountainous areas, under current and future scenarios. Based on these findings, we propose targeted management actions for monitoring and eradicating the species, leveraging prior information and local experiences to mitigate its impact.

Identification and application of the best-suited machine learning algorithm based on thermal comfort data characteristic: A data-driven approach

Integrating machine learning (ML) techniques in thermal comfort analysis is pivotal in contemporary research. This study systematically evaluates and applies ML algorithms for predicting thermal comfort parameters, emphasizing data imputation, parameter-specific behaviour analysis, and the development of an Independent Parameterised Modelling Ensemble (IPME) framework. A data-driven approach addresses missing values in the ASHRAE Global Thermal Comfort Database II using ML algorithms tailored to each thermal comfort (TC) parameter. Performance metrics, such as coefficient of determination (R2) for numerical data and accuracy for categorical data, guide the selection process, resulting in optimised ML-TC pairs for imputation. The analysis reveals parameter-specific behaviours, with numerical parameters exhibiting varying correlations with environmental conditions and parameters related to individual preferences posing challenges. ML models show improved performance with categorical data, potentially because of their ability to capture subjective aspects effectively. The IPME framework was proposed to address the conventional ML limitations by employing a hybrid ensemble strategy that selects tailored TC parameter-specific ML algorithms. This approach aims to enhance prediction accuracy by acknowledging the uniqueness of parameters and harnessing diverse ML capabilities. Future research directions include further exploration of the IPME framework and post-imputation retraining of ML models to capture actual behaviour. This study underscores the potential of ML in enhancing thermal comfort analysis, offering insights into imputation strategies, parameter-specific behaviours, and ensemble modelling approaches for more reliable predictions in building environments.

Enhancing Bank Loan Approval Efficiency Using Machine Learning: An Ensemble Model Approach

Enhancing Bank Loan Approval Efficiency Using Machine Learning: An Ensemble Model Approach

Nonbreeding distributions of four declining Nearctic–Neotropical migrants are predicted to contract under future climate and socioeconomic scenarios

ABSTRACT Climate and land use/land cover change are expected to influence the stationary nonbreeding distributions of 4 Nearctic–Neotropical migrant bird species experiencing population declines: Cardellina canadensis (Canada Warbler), Setophaga cerulea (Cerulean Warbler), Vermivora chrysoptera (Golden-winged Warbler), and Hylocichla mustelina (Wood Thrush). Understanding how and where these species’ distributions shift in response to environmental drivers is critical to inform conservation planning in the Neotropics. For each species, we quantified current (2012 to 2021) and projected future (2050) suitable climatic and land use/land cover conditions as components of stationary nonbreeding distributions. Multi-source occurrence data were used in an ensemble modeling approach with covariates from 3 global coupled climate models (CMCC-ESM2, FIO-ESM-2-0, MIROC-ES2L) and 2 shared socioeconomic pathways (SSP2-RCP4.5, SSP5-RCP8.5) to predict distributions in response to varying climatic and land use/land cover conditions. Our findings suggest that distribution contraction, upslope elevational shifts in suitable conditions, and limited shifts in latitude and longitude will occur in 3 of 4 species. Cardellina canadensis and S. cerulea are expected to experience a moderate distribution contraction (7% to 29% and 19% to 43%, respectively), primarily in response to expected temperature changes. The V. chrysoptera distribution was modeled by sex, and females and males were projected to experience a major distribution contraction (56% to 79% loss in suitable conditions for females, 46% to 65% for males), accompanied by shifts in peak densities to higher elevations with minimal changes in the upper elevation limit. Expected changes in precipitation had the greatest effect on V. chrysoptera. Hylocichla mustelina experienced the smallest distribution change, consistent with the species’ flexibility in habitat selection and broader elevational range. We recommend defining priority areas for conservation as those where suitable conditions are expected to remain or arise in the next 25 years. For V. chrysoptera in particular, it is urgent to ensure that mid-elevation forests in Costa Rica and Honduras are adequately managed and protected.

An ensemble model for accurate prediction of key water quality parameters in river based on deep learning methods

Deep learning models provide a more powerful method for accurate and stable prediction of water quality in rivers, which is crucial for the intelligent management and control of the water environment. To increase the accuracy of predicting the water quality parameters and learn more about the impact of complex spatial information based on deep learning models, this study proposes two ensemble models TNX (with temporal attention) and STNX (with spatio-temporal attention) based on seasonal and trend decomposition (STL) method to predict water quality using geo-sensory time series data. Dissolved oxygen, total phosphorus, and ammonia nitrogen were predicted in short-step (1 h, and 2 h) and long-step (12 h, and 24 h) with seven water quality monitoring sites in a river. The ensemble model TNX improved the performance by 2.1%–6.1% and 4.3%–22.0% relative to the best baseline deep learning model for the short-step and long-step water quality prediction, and it can capture the variation pattern of water quality parameters by only predicting the trend component of raw data after STL decomposition. The STNX model, with spatio-temporal attention, obtained 0.5%–2.4% and 2.3%–5.7% higher performance compared to the TNX model for the short-step and long-step water quality prediction, and such improvement was more effective in mitigating the prediction shift patterns of long-step prediction. Moreover, the model interpretation results consistently demonstrated positive relationship patterns across all monitoring sites. However, the significance of seven specific monitoring sites diminished as the distance between the predicted and input monitoring sites increased. This study provides an ensemble modeling approach based on STL decomposition for improving short-step and long-step prediction of river water quality parameter, and understands the impact of complex spatial information on deep learning model.

Modeling of the habitat suitability of European sprat (Sprattus sprattus, L.) in the Adriatic Sea under several climate change scenarios

The Mediterranean Sea represents the lower latitudinal limit of the European sprat range, where it is considered a sentinel species favoring temperate–cold temperatures. Sprattus sprattus is a plankton feeder that plays an important ecological role in contributing to the transfer of energy from lower to higher trophic levels, but climate-driven increases in sea temperatures may reduce the suitability of the pelagic habitat and threaten the tropho-dynamic role of sprat in areas such as the Adriatic Sea. The latter is an enclosed basin characterized by shallow waters and high annual temperature variations. Here, to investigate present and future habitat suitability areas for sprat, we applied four species distribution models (SDMs) using fishery-independent data collected from 2004 to 2021, along with remotely sensed and modeled environmental variables. A set of nine environmental predictors was tested, and the resulting best model was averaged in an ensemble model approach. The best ensemble models revealed good to high accuracy (sensitivity and specificity ≥ 0.8). The sea surface temperature and chlorophyll concentration emerged as the main explanatory variables in predicting the potential habitat of sprats, followed by bathymetry. The resulting probability of occurrence maps revealed that the species is bounded in the northern Adriatic Sea, where a longitudinal shift of high-suitability habitats from inshore to deeper and colder waters was detected between early and late summer. Future projections under IPCC representative concentration pathway (RCP) scenarios 4.5 (intermediate emission) and 8.5 (high emission-warm) underline small changes along with a gain of new areas in late summer in the short-term period up to 2050. Conversely, the temperature increase projected for the end of the century is predicted to cause a loss of suitable habitat area for sprats of up to 88% under a high emission-warm scenario relative to current habitat occupancy throughout the basin.

Prediction of Single-Well Production Rate after Hydraulic Fracturing in Unconventional Gas Reservoirs Based on Ensemble Learning Model

To address the significant challenges in determining the single-well production of tight gas and shale gas after hydraulic fracturing, artificial intelligence (AI) methods were implemented. Machine learning (ML) algorithms such as random forest (RF), extremely randomized trees (ET), lightweight gradient boosting machines (LightGBM), gradient boosting regression (GBR), and linear regression (LR) were utilized in conjunction with reservoir geology, engineering parameters, and production data to develop several foundational models for forecasting the production of unconventional gas wells. The accuracy of these models was evaluated. Based on this, improvements in the models’ predictive accuracy and generalizability were achieved through the ensemble of machine learning models. Furthermore, this paper selected two representative tight and shale gas reservoirs to demonstrate the application of the ensemble model for well production forecasting, and a comparative analysis with actual production data was conducted. For tight gas reservoir A, the blending model achieved an MAE of 0.8419 and an MSE of 1.0930, with an R2 score of 0.8812. For shale gas reservoir B, the blending model achieved an MAE of 1.4841 and an MSE of 3.1629, with an R2 score of 0.9524. The results of the case studies indicate that the ensemble model approach employed in this study has a higher predictive accuracy and reliability than a single machine learning algorithm, and is capable of handling high-dimensional, large-scale, and imbalanced data, offering scientific validation and technical support for the assessment of the well productivity in tight and shale gas wells.

Predicting the potential risk of Caragana shrub encroachment in the Eurasian steppe under anthropogenic climate change

Climate change and human activities drive widespread shrub encroachment in global grassland ecosystems, particularly in the Eurasian steppe. Caragana shrubs, the primary contributors to shrub encroachment in this region, play a crucial role in shaping the ecosystem's structure and function. Future changes in the suitable distribution range of Caragana species will directly affect the ecological security and sustainable socio-economic development of the Eurasian steppe ecosystem. We used an ensemble modeling approach to predict Caragana shrub-dominated plant communities' current and future distribution in three major steppe subregions: the Black Sea-Kazakhstan steppe, the Tibetan Plateau steppe, and the Central Asian steppe. We assessed the potential risk of Caragana shrub encroachment by predicting changes in the suitable distribution area of 19 Caragana shrub species under future climate changes. Our research findings suggest that the expansion of Caragana species in different subregions of the Eurasian steppe is influenced by the effects of climate change in various ways. The distribution of Caragana species is primarily influenced by precipitation and temperature, and the global human modification (ghm) has a significant impact on the Central Asian and Tibetan Plateau subregions. Minimal changes are expected in the Black Sea-Kazakhstan subregion, a slight increase on the Tibetan Plateau, and a substantial rise in the Central Asian subregion, which suggests a higher potential risk of Caragana species shrub encroachment in that area. Our research provides valuable insights into the response of Caragana shrub encroachment to changing climates and human activities. It also has implications for the sustainable management of different areas of the vast Eurasian steppe ecosystem.

Robust fracture intensity estimation from petrophysical logs and mud loss data: a multi-level ensemble modeling approach

This study presents a pioneering machine learning approach to continuously model fracture intensity in hydrocarbon reservoirs using solely conventional well logs and mud loss data. While machine learning has previously been applied to predict discrete fracture properties, this is among the first attempts to leverage well logs for continuous fracture intensity modeling leveraging advanced ensemble techniques. A multi-level stacked ensemble methodology systematically combines the strengths of diverse algorithms like gradient boosting, random forest and XGBoost through a tiered approach, enhancing predictive performance beyond individual models. Nine base machine learning algorithms generate initial fracture intensity predictions which are combined through linear regression meta-models and further stacked using ridge regression into an integrated super-learner model. This approach achieves significant improvements over individual base models, with the super-learner attaining a mean absolute error of 0.083 and R^2 of 0.980 on test data. By quantifying the crucial fracture intensity parameter continuously as a function of depth, this data-driven methodology enables more accurate reservoir characterization compared to traditional methods. The ability to forecast fracture intensity solely from conventional well logs opens new opportunities for rapid, low-cost quantification of this parameter along new wells without requiring advanced logging tools. When incorporated into reservoir simulators, these machine learning fracture intensity models can help optimize production strategies and recovery management. This systematic stacked ensemble framework advances continuous fracture intensity modeling exclusively from well logs, overcoming limitations of prior techniques. Novel insights gained via rigorous model evaluation deepen the understanding of naturally fractured reservoirs.

District-Level Forecast of Achieving Trachoma Elimination as a Public Health Problem By 2030: An Ensemble Modelling Approach.

Assessing the feasibility of 2030 as a target date for global elimination of trachoma, and identification of districts that may require enhanced treatment to meet World Health Organization (WHO) elimination criteria by this date are key challenges in operational planning for trachoma programmes. Here we address these challenges by prospectively evaluating forecasting models of trachomatous inflammation-follicular (TF) prevalence, leveraging ensemble-based approaches. Seven candidate probabilistic models were developed to forecast district-wise TF prevalence in 11 760 districts, trained using district-level data on the population prevalence of TF in children aged 1-9 years from 2004 to 2022. Geographical location, history of mass drug administration treatment, and previously measured prevalence data were included in these models as key predictors. The best-performing models were included in an ensemble, using weights derived from their relative likelihood scores. To incorporate the inherent stochasticity of disease transmission and challenges of population-level surveillance, we forecasted probability distributions for the TF prevalence in each geographic district, rather than predicting a single value. Based on our probabilistic forecasts, 1.46% (95% confidence interval [CI]: 1.43-1.48%) of all districts in trachoma-endemic countries, equivalent to 172 districts, will exceed the 5% TF control threshold in 2030 with the current interventions. Global elimination of trachoma as a public health problem by 2030 may require enhanced intervention and/or surveillance of high-risk districts.