Model Calibration Research Articles

Estimation of LAI of tobacco plant using selected spectral subsets of visible and near-infrared reflectance spectroscopy

Flue-cured tobacco is a main economic crop, and leaves are the direct product of tobacco plant. Monitoring leaf area index (LAI) of tobacco plant is important to field management, yield prediction, and industry regulation. Hyperspectral data have hundreds of narrow spectral bands in continuous spectral ranges, significantly advancing monitoring of LAI. However, considering spectral responses of LAI vary with wavelength, the use of the full spectral range in estimation of LAI is redundant. To reduce spectral redundancy and improve estimation of LAI, spectral subsets were selected based on importance of spectral bands. Variable importance in the projection (VIP) score obtained by partial least squares regression (PLSR) was adopted to measure the importance. The study was conducted in Yunnan Province, China. Canopy reflectance spectra of tobacco plant were measured in two consecutive growth seasons. Genetic algorithm (GA) and PLSR were used for model calibration. The identified important spectral regions for estimation of LAI were red edge, near-infrared (NIR), and green regions. In estimation of LAI of tobacco plant, compared with the estimation using the full spectral range of VNIR reflectance spectra, normalized root mean square error (NRMSE) and coefficient of determination (R2) values were improved from 10.30% and 0.83 to 7.68% and 0.90 and from 18.78% and 0.43 to 7.65% and 0.90 by using the reflectance spectra in identified spectral regions in the growth seasons in 2021 and 2022 separately. In addition to the identified continuous spectral regions, central bands of the identified spectral regions also achieved estimation of LAI, with the highest R2 value reaching 0.72. The selected spectral subsets improved estimation accuracy and reduced model complexity. The results indicate that selected spectral subsets are effective and promising in estimation of LAI, providing an alternative for estimation of LAI using hyperspectral data.

Comparison of simulation tools for optimizing borehole heat exchanger field operation

Model predictive control (MPC) is a promising approach for optimizing the performance of borehole heat exchangers (BHEs) in ground-source heat pump systems. The central element of MPC is the forward model that predicts the thermal dynamics in the ground. In this work, we validate the prediction accuracy of four BHE modeling approaches against real-world measurement data across various operational events and timescales. We simulate the fluid temperature leaving a BHE using a fully discretized 3-D numerical model, a resistance–capacitance model, a g-function model, and a hybrid model. The simulated temperatures are compared to measured temperatures using three validation metrics that quantify temperature offset, noise, and accuracy. The main reason for a mismatch between measured and modeled temperatures is a temperature offset of the simulated temperature. To remove this effect, the models were calibrated for their most sensitive parameter, the ground temperature, and their prediction accuracy over 4 years was evaluated. Thereby, model calibration seems to be a viable solution to account for an unknown load history. The results show that the resistance–capacitance model provides decent predictions in the short term and the g-function model in the long term. However, both models are strongly dependent on accurate calibration. The hybrid model provides the most accurate short and long-term predictions and is less dependent on calibration. Still, its integration into optimization syntax poses challenges compared to the other models. Although not yet applied in model predictive control, the hybrid model stands out as a promising choice for optimizing BHE field operations across various timescales.

Predictive Factors for Delayed Recovery From Anesthesia in Patients Receiving Free Vascularized Flap Reconstruction for Head and Neck Defects: A Retrospective Cohort Study.

Free flap reconstruction for head and neck defects is currently a common procedure. This study aimed to create and validate a predictive model for identifying patients at risk of delayed recovery from anesthesia after free flap reconstruction for head and neck defect. Electronic medical records from all patients were retrospectively collected. The primary outcome variable was delayed recovery from anesthesia. The least absolute shrinkage and selection operator regression model was employed to identify the most relevant features, followed by the construction of a nomogram model using multivariable logistic regression analysis. The discriminatory power, calibration, and clinical utility of the nomogram model were assessed using receiver operating characteristic curve analysis, calibration curve analysis, and decision curve analysis, respectively. This novel nomogram model incorporated 4 predictors for delayed recovery from anesthesia: preoperative albumin, intraoperative fresh frozen plasma infusion, preoperative platelet-to-lymphocyte ratio, and duration of intraoperative hypotension. The area under the receiver operating characteristic curve (area under the curve) for the nomogram model was determined to be 0.821 (95% CI: 0.803-0.836). After internal validation, the corrected area under the curve was found to be 0.768 (95% CI: 0.639-0.812). In addition, the model exhibited well-fitted calibration curves and demonstrated favorable clinical usability as indicated by the calibration curve and decision curve analysis curve. The authors created and validated a novel predictive model utilizing a limited number of 4 predictors, yet exhibiting commendable predictive performance. This innovative tool holds the potential to mitigate delayed recovery from anesthesia and enhance the efficient allocation of medical resources.

Predictive Factors for Delayed Recovery From Anesthesia in Patients Receiving Free Vascularized Flap Reconstruction for Head and Neck Defects: A Retrospective Cohort Study.

Free flap reconstruction for head and neck defects is currently a common procedure. This study aimed to create and validate a predictive model for identifying patients at risk of delayed recovery from anesthesia after free flap reconstruction for head and neck defect. Electronic medical records from all patients were retrospectively collected. The primary outcome variable was delayed recovery from anesthesia. The least absolute shrinkage and selection operator regression model was employed to identify the most relevant features, followed by the construction of a nomogram model using multivariable logistic regression analysis. The discriminatory power, calibration, and clinical utility of the nomogram model were assessed using receiver operating characteristic curve analysis, calibration curve analysis, and decision curve analysis, respectively. This novel nomogram model incorporated 4 predictors for delayed recovery from anesthesia: preoperative albumin, intraoperative fresh frozen plasma infusion, preoperative platelet-to-lymphocyte ratio, and duration of intraoperative hypotension. The area under the receiver operating characteristic curve (area under the curve) for the nomogram model was determined to be 0.821 (95% CI: 0.803-0.836). After internal validation, the corrected area under the curve was found to be 0.768 (95% CI: 0.639-0.812). In addition, the model exhibited well-fitted calibration curves and demonstrated favorable clinical usability as indicated by the calibration curve and decision curve analysis curve. The authors created and validated a novel predictive model utilizing a limited number of 4 predictors, yet exhibiting commendable predictive performance. This innovative tool holds the potential to mitigate delayed recovery from anesthesia and enhance the efficient allocation of medical resources.

Discrete Element Model of Oil Peony Seeds and the Calibration of Its Parameters

Abstract: Oil peony is an important oil crop that is primarily sown by using artificial methods at present. Its seeder encounters the problems of low efficiency of seeding that significantly limits the highly efficient mechanized production of high-quality peony oil. In this study, Fengdan white oil peony seeds were used as the research object and repose angle as the response value to establish a discrete element model (DEM) and parameter calibration. The range of parameters of oil peony seeds was first obtained through an experiment, and their repose angle was obtained by an inclinometer. A three-dimensional DEM of oil peony seeds was then established. The Plackett–Burman (PB) test was utilized to screen the parameters that had a significant influence on the repose angle, and the steepest ascent (SA) test was applied to determine their optimum range of testing. Following this, based on Box–Behnken (BBD) test results, a second-order regression model between the important parameters and the repose angle was constructed. Finally, the absolute minimum difference between simulated and measured repose angles was utilized as the objective of optimization to obtain the following optimum combination of parameters: The values of the seed–steel collision recovery coefficient (CRC), seed–seed static friction coefficient (SFC), seed–steel SFC, and seed–seed rolling friction coefficient (RFC) were 0.704, 0.324, 0.335, and 0.045, respectively. This optimal combination of parameters was confirmed through simulations, and the error between simulated and measured repose angles was only 0.67%, indicating that the calibrated DEM of oil peony seeds was reliable.

Discrete Element Model of Oil Peony Seeds and the Calibration of Its Parameters

Abstract: Oil peony is an important oil crop that is primarily sown by using artificial methods at present. Its seeder encounters the problems of low efficiency of seeding that significantly limits the highly efficient mechanized production of high-quality peony oil. In this study, Fengdan white oil peony seeds were used as the research object and repose angle as the response value to establish a discrete element model (DEM) and parameter calibration. The range of parameters of oil peony seeds was first obtained through an experiment, and their repose angle was obtained by an inclinometer. A three-dimensional DEM of oil peony seeds was then established. The Plackett–Burman (PB) test was utilized to screen the parameters that had a significant influence on the repose angle, and the steepest ascent (SA) test was applied to determine their optimum range of testing. Following this, based on Box–Behnken (BBD) test results, a second-order regression model between the important parameters and the repose angle was constructed. Finally, the absolute minimum difference between simulated and measured repose angles was utilized as the objective of optimization to obtain the following optimum combination of parameters: The values of the seed–steel collision recovery coefficient (CRC), seed–seed static friction coefficient (SFC), seed–steel SFC, and seed–seed rolling friction coefficient (RFC) were 0.704, 0.324, 0.335, and 0.045, respectively. This optimal combination of parameters was confirmed through simulations, and the error between simulated and measured repose angles was only 0.67%, indicating that the calibrated DEM of oil peony seeds was reliable.

Integrating LiDAR Sensor Data into Microsimulation Model Calibration for Proactive Safety Analysis.

Studies have shown that vehicle trajectory data are effective for calibrating microsimulation models. Light Detection and Ranging (LiDAR) technology offers high-resolution 3D data, allowing for detailed mapping of the surrounding environment, including road geometry, roadside infrastructures, and moving objects such as vehicles, cyclists, and pedestrians. Unlike other traditional methods of trajectory data collection, LiDAR's high-speed data processing, fine angular resolution, high measurement accuracy, and high performance in adverse weather and low-light conditions make it well suited for applications requiring real-time response, such as autonomous vehicles. This research presents a comprehensive framework for integrating LiDAR sensor data into simulation models and their accurate calibration strategies for proactive safety analysis. Vehicle trajectory data were extracted from LiDAR point clouds collected at six urban signalized intersections in Lubbock, Texas, in the USA. Each study intersection was modeled with PTV VISSIM and calibrated to replicate the observed field scenarios. The Directed Brute Force method was used to calibrate two car-following and two lane-change parameters of the Wiedemann 1999 model in VISSIM, resulting in an average accuracy of 92.7%. Rear-end conflicts extracted from the calibrated models combined with a ten-year historical crash dataset were fitted into a Negative Binomial (NB) model to estimate the model's parameters. In all the six intersections, rear-end conflict count is a statistically significant predictor (p-value < 0.05) of observed rear-end crash frequency. The outcome of this study provides a framework for the combined use of LiDAR-based vehicle trajectory data, microsimulation, and surrogate safety assessment tools to transportation professionals. This integration allows for more accurate and proactive safety evaluations, which are essential for designing safer transportation systems, effective traffic control strategies, and predicting future congestion problems.

Modeling of drinking and driving behaviors among adolescents and young adults in the United States: Complexities and Intervention outcomes

Alcohol-impaired driving is a formidable public health problem in the United States, claiming the lives of 37 individuals daily in alcohol-related crashes. Alcohol-impaired driving is affected by a multitude of interconnected factors, coupled with long delays between stakeholders’ actions and their impacts, which not only complicate policy-making but also increase the likelihood of unintended consequences. We developed a system dynamics simulation model of drinking and driving behaviors among adolescents and young adults. This was achieved through group model building sessions with a team of multidisciplinary subject matter experts, and a focused literature review. The model was calibrated with data series from multiple sources and replicated the historical trends for male and female individuals aged 15 to 24 from 1982 to 2020. We simulated the model under different scenarios to examine the impact of a wide range of interventions on alcohol-related crash fatalities. We found that interventions vary in terms of their effectiveness in reducing alcohol-related crash fatalities. In addition, although some interventions reduce alcohol-related crash fatalities, some may increase the number of drinkers who drive after drinking. Based on insights from simulation experiments, we combined three interventions and found that the combined strategy may reduce alcohol-related crash fatalities significantly without increasing the number of alcohol-impaired drivers on US roads. Nevertheless, related fatalities plateau over time despite the combined interventions, underscoring the need for new interventions for a sustained decline in alcohol-related crash deaths beyond a few decades. Finally, through model calibration we estimated time delays between actions and their consequences in the system which provide insights for policymakers and activists when designing strategies to reduce alcohol-related crash fatalities.

Evaluation of the Goss-modified solvation parameter model for the characterization of biphasic systems and descriptor assignments

The solvation parameter model uses six descriptors identified as excess molar refraction, E, dipolarity/polarizability, S, overall hydrogen-bond acidity, A, overall hydrogen-bond basicity, B, McGowan's characteristic volume, V, and the gas-liquid partition constant on hexadecane at 25 °C, L to model the distribution of neutral compounds in biphasic systems. Abraham's version of this model uses all six descriptors with two separate linear free energy relationship models for the transfer of compounds from a gas phase to a condensed phase and between condensed phases. Goss proposed a modification to this model that uses a single calibration model regardless of the physical state for each phase and five of the descriptors employed in Abraham's model (E descriptor is eliminated). The capability of Abraham's model and the Goss-modified model to characterize the contribution of intermolecular interaction to retention for gas and reversed-phase liquid chromatographic systems and distribution in liquid-liquid partition systems is evaluated using the WSU compound descriptor database. These more accurate values for the Abraham descriptors have not been utilized previously for the evaluation of the Goss-modified model and should be more capable of discerning subtle differences in model performance. It is shown that model quality defined by statistical parameters favors Abraham's model over the Goss-modified model with differences in model quality greater for systems in which Abraham's model indicates a significant contribution from electron lone pair interactions and for systems in which one phase is a solvent containing perfluoroalkyl substituents. There is a small systematic difference for the terms describing the combined contributions of cavity formation and dispersion interactions and for interactions of a dipole-type. The contribution of hydrogen-bonding interactions is virtually identical for the two models. The model intercepts are generally different and potentially assigned to a larger contribution from lack-of-fit for the Goss-modified model. Although the Abraham model descriptors have been routinely employed for applications using the Goss-modified model the possibility that Goss-model specific descriptors should be employed was evaluated. Using the Solver method and Goss-model specific calibration models for chromatographic and liquid-liquid partition systems a new set of Goss-specific descriptors was calculated for 28 varied compounds. These descriptors show good statistical agreement with the Abraham descriptor values with an average deviation of 0.009, -0.003, -0.004, and -0.023, respectively, for the S, A, B, and L descriptors, corresponding to a relative absolute deviation in percent of 2.2 %, 3.9 %, 4.3 %, and 1.2 %, respectively.

Assessing the impact of surface water and groundwater interactions for regional-scale simulations of water table elevation

Groundwater flow models are increasingly considered for the regional scale simulation of hydraulic heads and water table elevation. In the most complete configuration, models explicitly simulate two-way interactions between surface water (SW) and groundwater (GW) to reproduce and forecast both SW and GW water levels. In most regional scale groundwater models, however, SW-GW interactions are represented by simplified boundary conditions that only allow one-way interaction from SW to GW, neglecting most of the dynamic exchange fluxes between SW and GW. To evaluate the potential consequences of such simplifications on the simulation of regional GW levels, we compare two models on a 36,900 km2 regional aquifer system in Southern Quebec. One model explicitly simulates both SW and GW flow with two-way SW-GW feedback and the other model only simulates GW flow with a surface boundary flux to represent a one-way interaction with the land surface. Both models are developed with the same numerical code to ensure that the only differences are the representation of SW water flow and SW-GW feedback. The one-way model simulates overall deeper water tables because it removes all exfiltrated groundwater from the system once it exits the subsurface, therefore not allowing exfiltrating groundwater to re-infiltrate. This effect is most pronounced in areas where the water table is close to the surface and for low-flow periods. The inclusion of two-way feedback also reduces the sensitivity of simulated GW levels to the magnitude of the hydraulic conductivity. This result highlights the need for additional data on other system states to improve the calibration of regional scale models that explicitly simulate two-way SW-GW interactions.

Integrating real-time sensor data for improved hydrogeotechnical modelling in landslide early warning in Western Himalaya

This study presents a comprehensive investigation into developing a real-time monitoring framework for a localized landslide early warning system (LEWS), focusing on the hydrological dynamics of an unsaturated slope near the Joshimath Badrinath Highway in Uttarakhand, India. Considering the steepness of the slope and its adjacency to the highway, continuous monitoring and stability assessment are crucial. The framework encompasses primary phases of monitoring, modelling, forecasting, and warning. Monitoring collected hydrological and meteorological data, used for modelling and calibration. Validation using Taylor diagrams ensured accuracy by comparing predicted and monitored data. The calibrated hydrological model guided slope stability modelling to identify instability factors. A machine learning algorithm detected potential instability. Forecasting predicted unstable periods, triggering warnings if the factor of safety (Fs) values drop below 1.5. The study provided important insights into slope stability, highlighting the significance of vegetation parameters in accurately assessing slope stability using the Fs. Calibration of the hydrogeological model, particularly considering rainfall, climate, and vegetation data, improved the alignment between predicted and observed volumetric water content (VWC), especially in shallower depths. However, modelling hydrological dynamics at greater depths remains challenging, emphasizing the need for refined approaches. Machine learning techniques, specifically a Random Forest model, achieve high accuracy in predicting Fs, identifying VWC at 0.3-m and 3-m depths as pivotal variables. Temporal evaluation suggests a 12-month simulation as optimal, showing consistent performance across depths. Overall, the study advances slope stability modelling and offers insights into sustainable slope management practices, highlighting the complex interplay between climate, vegetation, and hydrological dynamics in unsaturated slopes. This approach lays a foundation for effective LEWS in Indian Himalayan states.

Risk of Transfusion in Isolated Coronary Artery Bypass Graft: Models developed from the STS Database

Perioperative blood transfusion is associated with adverse outcomes and higher costs following coronary artery bypass graft surgery (CABG). We developed risk assessments for patients' probability of perioperative transfusion and the expected transfusion volume, to improve clinical management and resource use. Among 1,266,545 consecutive (2008-2016) isolated-CABG operations in STS's Adult Cardiac Surgery Database, 657,821 (51.9%) received perioperative blood transfusions (red blood cell [RBC], fresh frozen plasma [FFP], cryoprecipitate, and/or platelets). We developed "full" models to predict perioperative transfusion of any blood product, and of RBC, FFP, or platelets. Using least absolute shrinkage and selection operator model selection, we built a rapid risk score based on 5 variables (age, body surface area, sex, preoperative hematocrit and use of intra-aortic balloon pump). Full model C-statistics were 0.785, 0.815, 0.707, and 0.699 for any blood product, RBC, FFP, and platelets. Rapid risk assessments' C-statistics were 0.752, 0.785, 0.670, and 0.661 for any blood product, RBC, FFP, and platelets. The observed versus expected risk plots showed strong calibration for full models and risk assessment tools; absolute differences between observed and expected risks of transfusion were <10.8% in each percentile of expected risk. Risk-assessments' predicted probabilities of transfusion were strongly and non-linearly associated (p<.0001) with total units transfused. These robust and well-calibrated risk assessment tools for perioperative transfusion in CABG can inform surgeons regarding patients' risks and number of RBC, FFP, and platelets units they can expect to need. This can aid in optimizing outcomes and increasing efficient use of blood products.

Embedded prompt tuning: Towards enhanced calibration of pretrained models for medical images

Foundation models pre-trained on large-scale data have been widely witnessed to achieve success in various natural imaging downstream tasks. Parameter-efficient fine-tuning (PEFT) methods aim to adapt foundation models to new domains by updating only a small portion of parameters in order to reduce computational overhead. However, the effectiveness of these PEFT methods, especially in cross-domain few-shot scenarios, e.g., medical image analysis, has not been fully explored. In this work, we facilitate the study of the performance of PEFT when adapting foundation models to medical image classification tasks. Furthermore, to alleviate the limitations of prompt introducing ways and approximation capabilities on Transformer architectures of mainstream prompt tuning methods, we propose the Embedded Prompt Tuning (EPT) method by embedding prompt tokens into the expanded channels. We also find that there are anomalies in the feature space distribution of foundation models during pre-training process, and prompt tuning can help mitigate this negative impact. To explain this phenomenon, we also introduce a novel perspective to understand prompt tuning: Prompt tuning is a distribution calibrator. And we support it by analysing patch-wise scaling and feature separation operations contained in EPT. Our experiments show that EPT outperforms several state-of-the-art fine-tuning methods by a significant margin on few-shot medical image classification tasks, and completes the fine-tuning process within highly competitive time, indicating EPT is an effective PEFT method. The source code is available at github.com/zuwenqiang/EPT.

Transferability of machine-learning-based global calibration models for NO2 and NO low-cost sensors

Abstract. It is essential to accurately assess and verify the effects of air pollution on human health and the environment in order to develop effective mitigation strategies. More accurate analysis of air pollution can be achieved by utilizing a higher-density sensor network. In recent studies, the implementation of low-cost sensors has demonstrated their capability to quantify air pollution at a high spatial resolution, alleviating the problem of coarse spatial measurements associated with conventional monitoring stations. However, the reliability of such sensors is in question due to concerns about the quality and accuracy of their data. In response to these concerns, active research efforts have focused on leveraging machine learning (ML) techniques in the calibration process of low-cost sensors. These efforts demonstrate promising results for automatic calibration, which would significantly reduce the efforts and costs of traditional calibration methods and boost the low-cost sensors' performance. As a contribution to this promising research field, this study aims to investigate the calibration transferability between identical low-cost sensor units (SUs) for NO2 and NO using ML-based global models. Global models would further reduce calibration efforts and costs by eliminating the need for individual calibrations, especially when utilizing networks of tens or hundreds of low-cost sensors. This study employed a dataset acquired from four SUs that were located across three distinct locations within Switzerland. We also propose utilizing O3 measurements obtained from available nearby reference stations to address the cross-sensitivity effect. This strategy aims to enhance model accuracy as most electrochemical NO2 and NO sensors are extremely cross-sensitive to O3. The results of this study show excellent calibration transferability between SUs located at the same site (Case A), with the average model performance being R2 = 0.90 ± 0.05 and root mean square error (RMSE) = 3.4 ± 0.9 ppb for NO2 and R2 = 0.97 ± 0.02 and RMSE = 3.1 ± 0.8 ppb for NO. There is also relatively good transferability between SUs deployed at different sites (Case B), with the average performance being R2 = 0.65 ± 0.08 and RMSE = 5.5 ± 0.4 ppb for NO2 and R2 = 0.82 ± 0.05 and RMSE = 5.8 ± 0.8 ppb for NO. Interestingly, the results illustrate a substantial improvement in the calibration models when integrating O3 measurements, which is more pronounced when SUs are situated in regions characterized by elevated O3 concentrations. Although the findings of this study are based on a specific type of sensor and sensor model, the methodology is flexible and can be applied to other low-cost sensors with different target pollutants and sensing technologies. Furthermore, this study highlights the significance of leveraging publicly available data sources to promote the reliability of low-cost air quality sensors.

Evaluating the Sediment Yield and Environmental Impact in Mining Expansion Scenarios in Sta. Cruz – Candelaria, Zambales

The study addressed the environmental concerns arising from industrial activities, such as mining, by focusing on the Candelaria-Santa Cruz region. Nickel mining in the region has raised issues related to water quality and sedimentation patterns. Employing the Modified Universal Soil Loss Equation (MUSLE) model, sediment yield computations were conducted, with consideration on factors like bare soil coverage, slope characteristics, and precipitation rates. The results emphasize the need for sustainable land management practices to mitigate adverse effects. The research covers a 10,000-ha area, simulating overburden removal scenarios ranging from 100 ha to 5,847.78 ha, and assesses sediment yield implications under different precipitation rates. The study also determines optimal relative growth rates and plant cover densities for progressive rehabilitation, highlighting the potential for effective erosion control. The findings underscore the significant impact of mining on sediment yield and water quality, prompting the call for comprehensive monitoring, calibration, and validation of the MUSLE model to guide informed decision-making and promote sustainable practices in the region. This research highlights the crucial role of integrated approaches in understanding and mitigating the consequences of resource exploitation on the environment and local communities.

ChatGPT Code Detection: Techniques for Uncovering the Source of Code

In recent times, large language models (LLMs) have made significant strides in generating computer code, blurring the lines between code created by humans and code produced by artificial intelligence (AI). As these technologies evolve rapidly, it is crucial to explore how they influence code generation, especially given the risk of misuse in areas such as higher education. The present paper explores this issue by using advanced classification techniques to differentiate between code written by humans and code generated by ChatGPT, a type of LLM. We employ a new approach that combines powerful embedding features (black-box) with supervised learning algorithms including Deep Neural Networks, Random Forests, and Extreme Gradient Boosting to achieve this differentiation with an impressive accuracy of 98%. For the successful combinations, we also examine their model calibration, showing that some of the models are extremely well calibrated. Additionally, we present white-box features and an interpretable Bayes classifier to elucidate critical differences between the code sources, enhancing the explainability and transparency of our approach. Both approaches work well, but provide at most 85–88% accuracy. Tests on a small sample of untrained humans suggest that humans do not solve the task much better than random guessing. This study is crucial in understanding and mitigating the potential risks associated with using AI in code generation, particularly in the context of higher education, software development, and competitive programming.

SENSITIVITY OF A 1D LAKE MODEL TO THE INPUT METEOROLOGICAL DATA

The paper evaluates how the choice of a source of meteorological data used as boundary conditions and the changes in model calibration parameters affect the adequate performance of a model of lake thermo- and hydrodynamics. The quality of simulation of water temperature dynamics in a small lake by a onedimensional GLM model using data from an automatic meteorological station on the lake shore, a state weather station and two global reanalyses was quantitatively assessed. It is shown that the best modeling result (RMSE of water temperature calculation 0,8-2,0°C, Nash-Sutcliffe efficiency &gt;0,7) can be achieved using data sets that explicitly provide incoming shortwave and longwave radiation fluxes. At the same time, good representation of the bottom water temperature requires correct setting of wind regime, which is only possible if data from instrumental ground-based observations at a stationary weather station are applied. Combining ground-based observations of wind speed, air temperature and humidity with radiation fluxes from global reanalyses makes it possible to achieve a calculation error ≤1°C at all depths.The analysis of model sensitivity to calibration parameters in case when the reanalysis data are applied showed that parameters related to wind mixing, i.e. the scaling factor for wind speed and constants related to the in-model description of wind impact energy, have the greatest influence on the final calculation error. If we change the intensity coefficients of other types of mixing during calibration it does not lead to an obvious trend in the quantitative estimates of modeling quality. It could also be reasonable to introduce correction multipliers to the values of air temperature and solar radiation fluxes, thus neutralizing the effect of unrepresentativeness of the meteorological data applied for a particular lake, as well as to correct the value of solar radiation extinction coefficient relative to the value obtained from the Secchi depth. The results of the study could be used to justify the choice of meteorological data source and to optimize the methods of calibration of lake thermal stratification models

Advances in the process-based models of constructed wetlands and a way forward for integrating emerging organic contaminants.

This research examines advancements in the development of process-based models of constructed wetlands (CWs) tailored for simulating conventional water quality parameters (CWQPs). Despite the promising potential of CWs for emerging organic contaminant (EOC) removal, the available CW models do not yet integrate EOC removal processes. This study explores the need and possibility of integrating EOCs into existing CW models. Nevertheless, a few researchers have developed process-based models of other wastewater treatment systems (e.g., activated sludge systems) to simulate certain EOCs. The EOC removal processes observed in other wastewater treatment systems are analogous to those in CWs. Therefore, the corresponding equations governing these processes can be tailored and integrated into existing CW models, similarly to what was done successfully in the past for CWQPs. This study proposed the next generation of CW models, which outlines 12 areas for future work: integrating EOC removal processes; ensuring data availability for model calibration and validation; considering quantitative and sensitive parameters; quantifying microorganisms in CWs; modifying biofilm dynamics models; including pH, aeration, and redox potential; integrating clogging and plant sub-models; modifying hydraulic sub-model; advancing computer technology and programming; and maintaining a balance between simplicity and complexity. These suggestions provide valuable insights for enhancing the design and operational features of current process-based models of CWs, facilitating improved simulation of CWQPs, and integration of EOCs into the modelling framework.

Insights from utilizing data of different quality levels for simulating barley performance under Nordic conditions: APSIM model evaluation

Abstract Crop model-aided ideotyping can accelerate the breeding of resilient barley cultivars. Yet, the accuracy of process descriptions in the crop models still requires substantial improvement, which is only possible with high-quality experimental data. Despite being demanded frequently, such data is still rarely available, especially for Northern European barley production. This study is one of the first to contribute to closing this existing data gap through the targeted collection of high-quality experimental data in pluri-annual, multi-location spring barley field trials in Denmark. With this data the prediction accuracy of APSIM significantly increased in contrast to commonly utilized lower quality datasets. Using this data for model calibration resulted in more accurate predictions of in-season plant development and important state variables (e.g. final grain yield and biomass). The model’s prediction accuracy can ultimately be further improved by examining remaining model weaknesses that were discoverable with the high quality data. Process descriptions regarding, e.g., early and late leaf development, soil water dynamics and respective plant response appeared to require further improvement. By illustrating the effect of data quality on model performance we reinforce the need for more model-guided field experiments.

Analysis of risk factors related to chronic non-healing wound infection and the construction of a clinical prediction model.

This study is aimed to analyse the risk factors associated with chronic non-healing wound infections, establish a clinical prediction model, and validate its performance. Clinical data were retrospectively collected from 260 patients with chronic non-healing wounds treated in the plastic surgery ward of Shanxi Provincial People's Hospital between January 2022 and December 2023 who met the inclusion criteria. Risk factors were analysed, and a clinical prediction model was constructed using both single and multifactor logistic regression analyses to determine the factors associated with chronic non-healing wound infections. The model's discrimination and calibration were assessed via the concordance index (C-index), receiver operating characteristic (ROC) curve and calibration curve. Multivariate logistic regression analysis identified several independent risk factors for chronic non-healing wound infection: long-term smoking (odds ratio [OR]: 4.122, 95% CI: 3.412-5.312, p < 0.05), history of diabetes (OR: 3.213, 95% CI: 2.867-4.521, p < 0.05), elevated C-reactive protein (OR: 2.981, 95% CI: 2.312-3.579, p < 0.05), elevated procalcitonin (OR: 2.253, 95% CI: 1.893-3.412, p < 0.05) and reduced albumin (OR: 1.892, 95% CI: 1.322-3.112, p < 0.05). The clinical prediction model's C-index was 0.762, with the corrected C-index from internal validation using the bootstrap method being 0.747. The ROC curve indicated an area under the curve (AUC) of 0.762 (95% CI: 0.702-0.822). Both the AUC and C-indexes ranged between 0.7 and 0.9, suggesting moderate-to-good predictive accuracy. The calibration chart demonstrated a good fit between the model's calibration curve and the ideal curve. Long-term smoking, diabetes, elevated C-reactive protein, elevated procalcitonin and reduced albumin are confirmed as independent risk factors for bacterial infection in patients with chronic non-healing wounds. The clinical prediction model based on these factors shows robust performance and substantial predictive value.