Root Mean Research Articles

Non-Destructive Detection of pH Value During Secondary Fermentation of Maize Silage Using Colorimetric Sensor Array Combined with Hyperspectral Imaging Technology

The pH value of maize silage can accurately reflect its quality. In this study, a colorimetric sensor array (CSA) combined with hyperspectral imaging (HSI) was used to predict the pH value of maize silage during secondary fermentation. Seventeen color-sensitive dyes were used to construct the CSA, which was subsequently applied to capture the volatile odor profiles of maize silage samples. Hyperspectral images of the color-sensitive dyes on the CSA were acquired using the HSI technique. Different algorithms were used to preprocess the raw spectral data of each dye, and a partial least squares regression (PLSR) model was built for each dye separately. Subsequently, the adaptive bacterial foraging optimization (ABFO) algorithm was employed to identify three color-sensitive dyes that demonstrated heightened sensitivity to pH variations in maize silage. This study further compared the capabilities of individual dyes, as well as their combinations, in predicting the pH value of maize silage. Additionally, a novel feature wavelength extraction method based on the ABFO algorithm was proposed, which was then compared with two traditional feature extraction algorithms. These methods were combined with PLSR and backpropagation neural network (BPNN) algorithms to construct a quantitative prediction model for the pH value of maize silage. The results show that the quantitative prediction model constructed based on three dyes was more accurate than that constructed based on an individual dye. Among them, the ABFO-BPNN model constructed on the basis of combined dyes had the best prediction performance, with prediction correlation coefficient (RP2), root mean square error of the prediction set (RMSEP), and ratio of performance deviation (RPD) values of 0.9348, 0.3976, and 3.9695, respectively. The aim of this study was to develop a reliable evaluation model to achieve fast and accurate predictions of silage pH.

Effects of supplementation of macular pigment carotenoids on ocular health: a Raman spectroscopic study of human blood serum of glaucoma patients.

Carotenoids are known for their antioxidant and vision protection roles, with dietary supplements often promoted for eye health. An initial trial, the European Nutrition in Glaucoma Management (ENIGMA), assessed macular pigment optical density (MPOD) and other ocular parameters before and after supplementing glaucoma patients with macular pigment (MP) carotenoids. The trial confirmed significant improvements in clinical ocular health. Blood, containing all major dietary carotenoids, serves as an efficient medium for in vivo analysis of carotenoids. Raman spectroscopy, an effective analytical tool, was used to measure the impact of supplementation on serum carotenoid levels and their correlation with MPOD and other ocular responses. Serum samples from baseline and 18-month supplemented participants were analysed. An inverse relationship was found between the percentage change in Raman intensity over the supplementation period and baseline Raman serum measurements, indicating greater relative benefits for people with low MPOD/serum carotenoids pre-supplementation. Partial least squares regression (PLSR) was employed to analyse the spectra after pre-processing, and the loadings reflected the carotenoid content and structural profile. MPOD results correlated at all eccentricities, with a coefficient of determination (R2) of 0.62-0.92 and %Root mean squared error of <44%. Structural, functional, and perceptual parameters also showed good correlation with serum Raman measurements. The results support the ENIGMA trial conclusions, and suggest strategies for optimizing patient responses to supplementation based on baseline carotenoid levels. Additionally, Raman spectroscopy of serum carotenoids shows significant potential as a simple and reliable method for investigating macular pigment carotenoids and assessing patient health.

Machine learning gold price forecasting

ABSTRACT Price projections for main metal commodities have been highly valued by many market players for a substantial period of time. Our research examines the daily reported price of gold in order to address the issue. The sample that is being investigated covers a period of ten years, beginning on 22 April 2014 and ending on 17 April 2024, and the price series that is being examined has significant business repercussions. When it comes to this particular scenario, Gaussian process regression models are constructed by using cross-validation approaches and Bayesian optimization methodologies. The strategies that are produced as a consequence are then used in order to supply price predictions. With a relative root mean square error of 0.8706%, our empirical prediction technique generates price projections that are relatively accurate for the out-of-sample measurement period that spans from 4 May 2022 to 17 April 2024. Investors and governments are provided with the knowledge they need to make informed decisions on the gold market due to the availability of models that anticipate prices.

Insights into the contribution of multiple factors on Ixodes ricinus abundance across Europe spanning 20 years using different machine learning algorithms.

The interplay of biotic and abiotic factors driving Ixodes ricinus abundance trends are not fully understood. Machine learning (ML) approaches are being increasingly used to explore this and predict future abundance patterns of this species, however, the studies focusing on this to date have had limitations (including short study duration, limited sample size, narrow geographical range and use of a single ML model). This study was undertaken to address these limitations by applying 11 predictive ML models (across three data clustering techniques) to a large I. ricinus occurrence dataset (27,150 records) containing geographical and temporal data from a 20-year period across 30 European countries, coupled with data covering a range of climatic and habitat features (temperature, rainfall, Normalised Difference Vegetation Index (NDVI), percentage of discontinuous urban fabric and land use category). To assess which ML model was most suited to prediction of I. ricinus abundance, four performance metric values were calculated per model: Normalised Root Mean Square Error (NRMSE), Scatter Index (SI), Mean Absolute Percentage Error (MAPE) and R2, all of which describe the statistical relationship between predicted and actual I. ricinus abundance values. Furthermore, using a Random Forest (RF) model across three clustering methods, we determined which features most significantly impacted upon I. ricinus abundance. The study demonstrated that Agglomerative Hierarchical Clustering (AC) methods and Linear Regression (LR) modelling performed best with this dataset. Our findings revealed that land use and rainfall were the primary contributors to I. ricinus abundance, with temperature playing a lesser role. This was measured according to the extent of prediction error increase following exclusion of that factor from the analysis. We provide a summary of the factors most strongly linked to I. ricinus abundance, which can be used to guide interventions to aid the control of ticks and tick-borne disease across Europe.

A Novel Evolutionary Deep Learning Approach for PM2.5 Prediction Using Remote Sensing and Spatial–Temporal Data: A Case Study of Tehran

Forecasting particulate matter with a diameter of 2.5 μm (PM2.5) is critical due to its significant effects on both human health and the environment. While ground-based pollution measurement stations provide highly accurate PM2.5 data, their limited number and geographic coverage present significant challenges. Recently, the use of aerosol optical depth (AOD) has emerged as a viable alternative for estimating PM2.5 levels, offering a broader spatial coverage and higher resolution. Concurrently, long short-term memory (LSTM) models have shown considerable promise in enhancing air quality predictions, often outperforming other prediction techniques. To address these challenges, this study leverages geographic information systems (GIS), remote sensing (RS), and a hybrid LSTM architecture to predict PM2.5 concentrations. Training LSTM models, however, is an NP-hard problem, with gradient-based methods facing limitations such as getting trapped in local minima, high computational costs, and the need for continuous objective functions. To overcome these issues, we propose integrating the novel orchard algorithm (OA) with LSTM to optimize air pollution forecasting. This paper utilizes meteorological data, topographical features, PM2.5 pollution levels, and satellite imagery from the city of Tehran. Data preparation processes include noise reduction, spatial interpolation, and addressing missing data. The performance of the proposed OA-LSTM model is compared to five advanced machine learning (ML) algorithms. The proposed OA-LSTM model achieved the lowest root mean square error (RMSE) value of 3.01 µg/m3 and the highest coefficient of determination (R2) value of 0.88, underscoring its effectiveness compared to other models. This paper employs a binary OA method for sensitivity analysis, optimizing feature selection by minimizing prediction error while retaining critical predictors through a penalty-based objective function. The generated maps reveal higher PM2.5 concentrations in autumn and winter compared to spring and summer, with northern and central areas showing the highest pollution levels.

Functional Disability and Psychological Impact in Headache Patients: A Comparative Study Using Conventional Statistics and Machine Learning Analysis

Background and Objectives: Recent research has focused on exploring the relationships between various factors associated with headaches and understanding their impact on individuals’ psychological states. Utilizing statistical methods and machine learning models, these studies aim to analyze and predict these relationships to develop effective approaches for headache management and prevention. Materials and Methods: Analyzing data from 398 patients (train set = 318 and test set = 80), we investigated the influence of various features on outcomes such as depression, anxiety, and headache intensity using machine learning and linear regression. The study employed a mixed-methods approach, combining medical records, interviews, and surveys to gather comprehensive data on participants’ experiences with headaches and their associated psychological effects. Results: Machine learning models, including Random Forest (utilized for Headache Impact Test-6, Patient Health Questionnaire-9, and Generalized Anxiety Disorder-7) and Support Vector Regression (applied to Migraine Disability Assessment), revealed key features contributing to each outcome through Shapley values, while linear regression provided additional insights. Frequent analgesic medication emerged as a significant predictor of poorer life quality (Headache Impact Test-6, root mean squared error = 7.656) and increased depression (Patient Health Questionnaire-9, root mean squared error = 5.07) and anxiety (Generalized Anxiety Disorder-7, root mean squared error = 4.899) in the Random Forest model. However, interpreting the importance of features in complex models like supportive vector regression poses challenges, and determining causality between factors such as medication usage and pain severity was not feasible. Conclusions: Our study underscores the importance of considering individual characteristics in optimizing treatment strategies for headache patients.

Boosting the catalytic efficiency of UGT51 for efficient production of rare ginsenoside Rh2.

Ginsenoside Rh2(S) is well-known for its therapeutic potential against diverse conditions, including some cancers, inflammation, and diabetes. The enzymatic activity of uridine diphosphate glycosyltransferase 51 (UGT51) from Saccharomyces cerevisiae plays a pivotal role in the glycosylation process between UDP-glucose (donor) and protopanaxadiol (acceptor), to form ginsenoside Rh2. However, the catalytic efficiency of the UGT51 has remained a challenging task. To this end, we employed site-directed mutagenesis on UGT51 to improve its catalytic efficiency for enhanced production of ginsenoside Rh2. The mutated structure, featuring four key mutations (E805A, S998A, R1031A, and L1032A), exhibited heightened stability, binding affinity, and active site accessibility for protopanaxadiol (PPD) compared to the wild type. Under in vitro conditions, three mutants (E805A, R1031A, and L1032A) demonstrated 10%, 58%, and 65% higher enzymatic activities compared to the wild strain. Notably, the double mutant R1031A/L1032A exhibited an 85% increase in activity. Employing a fed-batch technology with PPD as the substrate yielded a Rh2 production of 4.663g/L. The molecular dynamics (MD) simulations were employed to investigate the movements and dynamic dynamics of UGT51 mutations and PPD complexes. The root mean square deviation (RMSD) analysis revealed substantial alterations in structural conformation, particularly in the R1031A/L1032A mutations, correlating with boosted catalytic efficiency. Furthermore, the root mean square fluctuation (RMSF) simulation study aligned with both the RMSD and the solvent-accessible surface area (SASA) analyses. The computationally guided site-directed mutagenesis approach holds promise for extending its application to the development of commercially significant enzymes.

State of Charge Estimation Based on Improved AILS-CSO Parameter Identification Algorithm

Abstract This paper introduces an adaptive inertial local search-cat-swarm optimization algorithm. The identification of parameters and estimation of charge state for lithium power batteries under diverse operational conditions were undertaken. In this approach, the equivalent circuit model of lithium power batteries is constructed using a second-order RC equivalent circuit model. Then, the cat swarm optimization algorithm is augmented with an adaptive adjustment algorithm for the inertial weight, thereby enhancing the accuracy of the identified battery parameters. Subsequently, the estimated voltage is then compared with the actual voltage in order to validate the feasibility of the algorithm. The root mean square error (RMSE) is less than 0.56%, and the maximum absolute error is 0.116 V. In conclusion, the state of charge of lithium batteries is estimated online under disparate working conditions via the integration of the extended Kalman filter algorithm, and a comparison is conducted. The maximum error is less than 1.74%, and the RMSE is less than 0.5%, indicating that the algorithm exhibits superior stability and estimation accuracy compared to the traditional recursive least squares method.

Development of the children's primitive reflex integration assessment scale

ObjectiveNon-integrated primitive reflexes (PRs) in children can lead to issues in motor function and psychological wellbeing, while prior studies have shown correlations between PR integration and neurodevelopmental disorders in children. However, measurement methods for PR integration remain unestablished. Therefore, in the present study, we describe the development of a measurement scale for PR integration, a novel assessment tool to evaluate PR integration in children.MethodsCombining a literature review, practical experience, and results of specialized group discussions, a preliminary draft of the Children's Primitive Reflex Integration Measurement Scale (CPRIMS) was formulated. Employing a convenience sampling method, participants were selected from first and second-grade students in three primary schools in Liaoning province, Shenyang city, from May to July 2023. Item Discrimination Method (IDM), Critical Ratio Method (CRM), and Internal Consistency Coefficient Method (ICCM) were used for item analysis of pilot testing data. For formal testing data, Cronbach's α assessed the reliability of the scale, while fit indices such as chi-square value/degrees of freedom (χ2/df), Root Mean Square Error of Approximation (RMSEA), and Comparative Fit Index (CFI), along with tests of construct validity, evaluated the scale's validity.ResultsOverall, 555 participants were selected, 234 children with a mean age of 7.59 ± 0.71 years participated in the pilot testing, while 321 children with a mean age of 7.73 ± 0.71 years participated in the formal testing. CPRIMS comprises seven dimensions and seventeen items, including the Moro reflex (MR), Asymmetrical Tonic Neck Reflex (ATNR), Symmetrical Tonic Neck Reflex (STNR), Tonic Labyrinthine Reflex (TLR), Spinal Galant Reflex (SGR), Spinal Perez Reflex (SPR), and Landau Reflex (LR), explaining 88.2% of the total variance. Confirmatory factor analysis revealed a good model fit (χ2/df = 1.631, RMSEA = 0.044, NFI = 0.950, CFI = 0.980, IFI = 0.932, TLI = 0.972). Cronbach's α coefficients for the seven dimensions ranged from 0.730 to 0.945, demonstrating strong reliability.ConclusionCPRIMS, which includes dimensions such as MR, ATNR, STNR, TLR, SGR, SPR, and LR, demonstrates strong reliability and validity, indicating that this measure could serve as a reliable and effective tool for assessing the integration levels of PRs in children aged 6 to 9 years old.

Redefining social intelligence through self-efficacy and personal autonomy: Development and validation of the efficient social intelligence scale

ABSTRACT Background Over the years, scholars have conceived and operationalized social intelligence in several ways that were mostly inclined toward social compliance. The current research emphasized the role of subjectivity, self-efficacy, and personal autonomy in social intelligence in addition to the traditional aspects of social intelligence such as knowledge about social norms and maintaining healthy social relationships. The objective of the current research was to develop and validate a new scale on social intelligence, focusing more on self-efficacy and personal autonomy. The scale was labeled as Efficient Social Intelligence Scale—ESIS. Methods The ESIS was developed and validated in a series of four phases by involving 744 conveniently selected participants (Mage = 23 years, SD = 6.18; women = 55.5%). The ESIS was evaluated through rigorous statistical procedures involving internal consistency, exploratory & confirmatory factor analyses, and convergent & discriminant validity. Results The ESIS was found to be a highly reliable and valid instrument having nine items and four sub-scales. Indicators such as Cronbach’s alpha (0.830), average item-total correlation (r = 0.614; p < .001), average item-scale correlation (r = 0.913; p < .001), comparative fit index (0.990), Tucker-Lewis index (0.983), root mean square error of approximation (0.045), standardized root mean square residual (0.027), convergent validity with TROMSO social intelligence scale (r = 0.562; p < .001), and discriminant validity with psychosocial illness (r = -0.200; p < .001) established the reliability and validity of the PS. Additionally, men reported significantly higher levels of social intelligence as compared to women (Men = 82.82%; Women = 81.19%; p < .05; Cohen’s d = 0.184). Conclusion The ESIS redefines social intelligence to emphasize personal autonomy and subjective fulfillment within social interactions.

A combined model of shoot phosphorus uptake based on sparse data and active learning algorithm

The soil ecosystem has been severely damaged because of the increasingly severe environmental problems caused by excessive application of phosphorus (P) fertilizer, which seriously hinders soil fertility restoration and sustainable farmland development. Shoot P uptake (SPU) is an important parameter for monitoring crop growth and health and for improving field nutrition management and fertilization strategies. Achieving on-site measurement of large-scale data is difficult, and effective nondestructive prediction methods are lacking. Improving spatiotemporal SPU estimation at the regional scale still poses challenges. In this study, we proposed a combination prediction model based on some representative samples. Furthermore, using the experimental area of Henan Province, as an example, we explored the potential of the hyperspectral prediction of maize SPU at the canopy scale. The combination model comprises predicted P uptake by maize leaves, stems, and grains. Results show that (1) the prediction accuracy of the combined prediction model has been greatly improved compared with simple empirical prediction models, with accuracy test results of R2 = 0.87, root mean square error = 2.39 kg/ha, and relative percentage difference = 2.71. (2) In performance tests with different sample sizes, two-dimensional correlation spectroscopy i.e., first-order differentially enhanced two-dimensional correlation spectroscopy (1Der-2DCOS) and two-trace 2DCOS of enhanced filling and milk stages (filling-milk-2T2DCOS)) can effectively and robustly extract spectral trait relationships, with good robustness, and can achieve efficient prediction based on small samples. (3) The hybrid model constrained by the Newton-Raphson-based optimizer’s active learning method can effectively filter localized simulation data and achieve localization of simulation data in different regions when solving practical problems, improving the hybrid model’s prediction accuracy. The practice has shown that with a small number of representative samples, this method can fully utilize remote sensing technology to predict SPU, providing an evaluation tool for the sustainable use of agricultural P. Therefore, this method has good application prospects and is expected to become an important means of monitoring global soil P surplus, promoting sustainable agricultural development.

Exploring the Psychometric Properties of the Gifted Rating Scale-School Form in the Qatari Context

This study uses a rigorous research process to explore the psychometric properties of the Gifted Rating Scale-School Form (GRS-S) within the Qatari educational context. We employed stratified cluster sampling of 326 students (aged 6–13 years, M = 10.9) from 25 public schools in Doha. Data was collected in the second semester of the 2023–2024 school year and analyzed for internal consistency, factorial validity, and concurrent and predictive validity. Confirmatory factor analysis confirmed the six-factor model (root mean square error of approximation = 0.073, comparative fit index = 0.886). The findings revealed that the GRS-S had high internal consistency (α = 0.926–0.982), moderate to strong correlations with the HOPE scale ( r = 0.651), and solid predictive validity with grade point average ( r = 0.917). These findings validate the GRS-S as a reliable tool for identifying gifted students in Qatar, with significant implications for the region's educational practices and policy development, potentially shaping the future of gifted education in the country.

Online analysis of Amazon's soils through reflectance spectroscopy and cloud computing can support policies and the sustainable development.

Analyzing soil in large and remote areas such as the Amazon River Basin (ARB) is unviable when it is entirely performed by wet labs using traditional methods due to the scarcity of labs and the significant workforce requirements, increasing costs, time, and waste. Remote sensing, combined with cloud computing, enhances soil analysis by modeling soil from spectral data and overcoming the limitations of traditional methods. We verified the potential of soil spectroscopy in conjunction with cloud-based computing to predict soil organic carbon (SOC) and particle size (sand, silt, and clay) content from the Amazon region. To this end, we request physicochemical attribute values determined by wet laboratory analyses of 211 soil samples from the ARB. These samples were submitted to spectroscopy Vis-NIR-SWIR in the laboratory. Two approaches modeled the soil attributes: M-I) cloud-computing-based using the Brazilian Soil Spectral Service (BraSpecS) platform, and M-II) computing-based in an offline environment using R programming language. Both methods used the Cubist machine learning algorithm for modeling. The coefficient of determination (R2), mean absolute error (MAE) and root mean squared error (RMSE) served as criteria for performance assessment. The soil attributes prediction was highly consistent, considering the measured and predicted by both approaches M-I and M-II. The M-II outperformed the M-I in predicting both particle size and SOC. For clay content, the offline model achieved an R2 of 0.85, with an MAE of 86.16gkg⁻1 and RMSE of 111.73gkg⁻1, while the online model had an R2 of 0.70, MAE of 111.73gkg⁻1, and RMSE of 144.19gkg⁻1. For SOC, the offline model also showed better performance, with an R2 of 0.81, MAE of 3.42gkg⁻1, and RMSE of 4.57gkg⁻1, compared to an R2 of 0.72, MAE of 3.66gkg⁻1, and RMSE of 5.53gkg⁻1 for the M-I. Both modeling methods demonstrated the power of reflectance spectroscopy and cloud computing to survey soils in remote and large areas such as ARB. The synergetic use of these techniques can support policies and sustainable development.

Comparative Analysis of Regression Methods for Estimation of Remaining Useful Life of Lithium Ion Battery

Lithium batteries play a critical role in modern technological applications, including electric vehicles and portable electronic devices. Ensuring accurate estimation of their remaining useful life is essential to improve system efficiency and reliability. This study focuses on predicting the remaining useful life of lithium batteries using advanced regression methods. Data were collected from lithium battery charge-discharge cycles, encompassing key operational parameters such as voltage, current, and temperature. The analysis employed several regression models, including linear regression, lasso regression, and Ridge regression, to identify relationships between these parameters and battery life. The models were evaluated based on estimation accuracy, with Root Mean Square Error (RMSE) as the primary performance metric. The findings demonstrate that regression methods can effectively capture non-linear relationships between input variables and the remaining useful life, with lasso and Ridge regression showing superior performance in reducing prediction errors. These results underscore the potential of regression-based approaches in providing robust and reliable estimations of battery life. The conclusions highlight the importance of these models for developing predictive battery management systems, which can optimize battery performance and extend their operational lifespan across various applications. This research establishes a solid foundation for future studies on intelligent battery health monitoring and management.

Quantum chemically calculated abraham parameters for quantifying and predicting polymer hydrophobicity.

The leakage and accumulation of plastic in the environment is a significant and growing problem with numerous detrimental impacts and has led to a push toward the design and development of more environmentally benign materials. To this end we have developed a quantum chemistry (QC) based model for predicting the mobility of polymer materials from molecular structure. Hydrophobicity is used as a surrogate for mobility given that hydrophobic interactions drive much of the partitioning of contaminants in and out of various environmentally relevant compartments. To model polymer hydrophobicity we adjusted a previously developed Quantum Chemically Calculated Abraham Parameter (QCAP) model to calculate Abraham Parameters (AP) of small molecules from molecular structure information. The resulting model predicted the octanol-water partition coefficient (KOW) of polymer repeating units with a root mean square error (RMSE) of 0.48 (log scale). Additionally, the hydrophobicity of high molecular weight polymer materials was captured though solubility parameters and nile red staining experiments from the literature and predicted with RMSEs of 1.21 (J/cc)0.5 and 3.42 nm respectively. Finally, to test the environmental applicability of the model the relative adsorption capacity of three polymers were predicted and used to unify sorption isotherms across multiple sorbates and polymer sorbents.

Advancing terrestrial snow depth monitoring with machine learning and L-band InSAR data: a case study using NASA’s SnowEx 2017 data

Current terrestrial snow depth mapping from space faces challenges in spatial coverage, revisit frequency, and cost. Airborne lidar, although precise, incurs high costs and has limited geographical coverage, thereby necessitating the exploration of alternative, cost-effective methodologies for snow depth estimation. The forthcoming NASA-ISRO Synthetic Aperture Radar (NISAR) mission, with its 12-day global revisit cycle and 1.25 GHz L-band frequency, introduces a promising avenue for cost-effective, large-scale snow depth and snow water equivalent (SWE) estimation using L-band Interferometric SAR (InSAR) capabilities. This study demonstrates InSAR’s potential for snow depth estimation via machine learning. Using 3 m resolution L-band InSAR products over Grand Mesa, Colorado, we compared the performance of three machine learning approaches (XGBoost, ExtraTrees, and Neural Networks) across open, vegetated, and the combined (open + vegetated) datasets using Root Mean Square Error (RMSE), Mean Bias Error (MBE), and R2 metrics. XGBoost emerged as the superior model, with RMSE values of 9.85 cm, 10.46 cm, and 9.88 cm for open, vegetated, and combined regions, respectively. Validation against in situ snow depth measurements resulted in an RMSE of approximately 16 cm, similar to in situ validation of the airborne lidar. Our findings indicate that L-band InSAR, with its ability to penetrate clouds and cover extensive areas, coupled with machine learning, holds promise for enhancing snow depth estimation. This approach, especially with the upcoming NISAR launch, may enable high-resolution (∼10 m) snow depth mapping over extensive areas, provided suitable training data are available, offering a cost-effective approach for snow monitoring. The code and data used in this work are available at https://github.com/cryogars/uavsar-lidar-ml-project.

Predicting changes in agricultural yields under climate change scenarios and their implications for global food security

Climate change has direct impacts on current and future agricultural productivity. Statistical meta-analysis models can be used to generate expectations of crop yield responses to climatic factors by pooling data from controlled experiments. However, methodological challenges in performing these meta-analyses, together with combined uncertainty from various sources, make it difficult to validate model results. We present updates to published estimates of crop yield responses to projected temperature, precipitation, and CO2 patterns and show that mixed effects models perform better than pooled OLS models on root mean squared error (RMSE) and explained deviance, despite the common usage of pooled OLS in previous meta-analyses. Based on our analysis, the use of pooled OLS may underestimate yield losses. We also use a block-bootstrapping approach to quantify uncertainty across multiple dimensions, including modeler choices, climate projections from the sixth Coupled Model Intercomparison Project (CMIP6), and emissions scenarios from Shared Socioeconomic Pathways (SSP). Our estimates show projected yield responses of − 22% (maize), − 9% (rice), − 15% (soy), and − 14% (wheat) from 2015 to 2080–2100 under the business-as-usual scenario of SSP5–8.5, which reduce to − 3.8%, − 2.7%, 1.4%, and − 1.5% respectively under the lower emissions scenario of SSP1–2.6. Without mitigation and adaptation, countries in South Asia, sub-Saharan Africa, North America, and Oceania could become at risk of being unable to meet national calorie demand by the end of the century under the most severe emissions scenario.

Machine learning-driven prediction of biochar adsorption capacity for effective removal of Congo red dye

Congo red, a widely utilized dye in the textile industry, presents a significant threat to living organisms due to its carcinogenic properties and non-biodegradable nature. This study proposes a data-driven machine-learning approach to optimize biochar characteristics and environmental conditions to maximize the adsorption capacity of biochar for the removal of Congo red dye. Therefore, six machine learning models were trained and tested on a dataset containing eleven input parameters (related to biochar properties and environmental conditions) and adsorption capacity. The models were evaluated using performance metrics such as R-squared (R2), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE). With the highest R2 (0.9785) and lowest RMSE (0.1357), Random Forest Regression (RF) outperformed other machine learning models. DT and XGB also performed well, achieving slightly lower R2 values of 0.9741 and 0.9577, respectively. The LR model performed the worst, with the lowest R2 (0.4575) and the highest RMSE (0.6821). Moreover, the reliability of these models was validated using a 10-fold cross-validation method. RF once again performed the best with an R2 value of 0.9762. Feature analysis revealed that the initial dye concentration relative to biochar dosage (C0), specific surface area (BET), and pore volume (PV) are the most significant factors affecting the dye adsorption capacity of biochar, while parameters such as carbon content (C), the oxygen and nitrogen to carbon molar ratio [(O + N)/C], and pore diameter (D) had minimal impact. This research demonstrates that machine learning models can accurately predict biochar’s contaminant adsorption capacity, enhancing wastewater treatment and promoting efficient, cost-effective environmental management.Graphical

A Reservoir Dam Monitoring Technology Integrating Improved ABC Algorithm and SVM Algorithm

A reservoir dam is a water conservancy project with large investment and high social and economic benefits, which plays an irreplaceable role in flood control, power generation, water storage, and urban water supply. There is a risk of accidents in the process of reservoir dams, so dam monitoring is an important means to achieve the safe operation of reservoirs. In this paper, taking advantage of the high-dimensional and nonlinear characteristics of dam monitoring data samples, the fusion-improved ABC (artificial bee colony) algorithm is introduced, and the SVM (support vector machine) algorithm is used to optimize the penalty factor and kernel function parameters. The test results of the ABC and SVM algorithm are relatively stable, with small fluctuation amplitude, which can continuously monitor water level, pore water pressure, dam deformation, temperature, humidity, vibration, and other indicators is less than 10%, which is significantly lower than the standard ABC algorithm, the standard ANN algorithm, and the standard SVM algorithm. The independence and characteristics of the ABC–SVM algorithm are significantly higher, and the correlation is 0.03, the RMS (root mean square) is 0.2334, which is lower than that of the standard ABC algorithm of 0.09, and the standard ANN algorithm of 0.8. The stability of the results and performance stability are analyzed, which is greater than 90%. The ABC and SVM is used to predict the displacement and deformation law of the reservoir dam.

Long-Term Trends of Lake Surface Water Temperatures in Lowland Polish Temperate Lakes

In this study, long-term lake surface water temperature (LSWT) data were used to investigate the impact of climate change on thermal conditions in 25 Polish lowland lakes. The results show that the warming rate of the annual mean LSWT ranges from 0.14 °C per decade to 0.69 °C per decade with an average value of 0.44 °C per decade. The annual maximum LSWT presented the strongest warming trend, with the warming rate varying between 0.26 °C per decade and 1.06 °C per decade (average value of 0.65 °C per decade). Warming rates were observed in all seasons but with different intensities, with warming rates increasing from spring to autumn and then to summer. The warming rate of the summer LSWT varied between 0.29 °C per decade and 0.87 °C per decade with an average value of 0.56 °C per decade. Conversely, winter and annual minimum LSWTs did not present clear increasing trends. The increase in the annual maximum, annual average, and seasonal LSWTs correlated well with the inter-annual variability in air temperature. To understand the relationship between LSWT and air temperature, the non-linear regression model (S-curve) was used in this study. The results indicate that the non-linear regression model can help to present the relationship between LSWT and air temperature in the studied lakes (the average values of the root mean squared error (RMSE), the mean absolute error (MAE), and the Nash–Sutcliffe efficiency coefficient (NSE) are 1.68 °C, 1.28 °C, and 0.95, respectively). The warming trends of LSWTs observed for the studied lakes in Poland are coherent and in some cases larger than the data from other lakes worldwide, and should be seriously considered by policy makers.